powerpc/64: interrupts and syscalls series
Subsystems (5)
Nicholas Piggin
reviewerv3 cover powerpc/64: interrupts and syscalls series
This is a long overdue update of the series, with fixes from me Michal and Michael. Does not include Michal's syscall compat series.
Patches 1-22 are changes to low level 64s interrupt entry assembly which has been posted before, no change except adding patch 21 and fixing patch 22 to reconcile irq state in the soft-nmi handler to avoid preempt warnings.
Patches 23-26 are to turn system call entry/exit code into C. Bunch of irq and preempt and TM warnings and bugs caught by selftests etc fixed, plus a few peripheral patches added (sstep and zeroing regs).
Patches 27-29 are to turn interrupt exit code into C. This had a bit more change, most significantly a change to how interrupt exit soft irq replay works.
Patches 30-32 are for scv system call support. Lot of changes here to turn it into something a bit better than RFC quality. Discussion about ABI seems to be settling and not very controversial.
Thanks,
Nick
Nicholas Piggin (32):
powerpc/64s/exception: Introduce INT_DEFINE parameter block for code
generation
powerpc/64s/exception: Add GEN_COMMON macro that uses INT_DEFINE
parameters
powerpc/64s/exception: Add GEN_KVM macro that uses INT_DEFINE
parameters
powerpc/64s/exception: Expand EXC_COMMON and EXC_COMMON_ASYNC macros powerpc/64s/exception: Move all interrupt handlers to new style code
gen macros
powerpc/64s/exception: Remove old INT_ENTRY macro powerpc/64s/exception: Remove old INT_COMMON macro powerpc/64s/exception: Remove old INT_KVM_HANDLER powerpc/64s/exception: Add ISIDE option powerpc/64s/exception: move real->virt switch into the common handler powerpc/64s/exception: move soft-mask test to common code powerpc/64s/exception: move KVM test to common code powerpc/64s/exception: remove confusing IEARLY option powerpc/64s/exception: remove the SPR saving patch code macros powerpc/64s/exception: trim unused arguments from KVMTEST macro powerpc/64s/exception: hdecrementer avoid touching the stack powerpc/64s/exception: re-inline some handlers powerpc/64s/exception: Clean up SRR specifiers powerpc/64s/exception: add more comments for interrupt handlers powerpc/64s/exception: only test KVM in SRR interrupts when PR KVM is
supported
powerpc/64s/exception: sreset interrupts reconcile fix powerpc/64s/exception: soft nmi interrupt should not use
ret_from_except
powerpc/64: system call remove non-volatile GPR save optimisation powerpc/64: sstep ifdef the deprecated fast endian switch syscall powerpc/64: system call implement entry/exit logic in C powerpc/64: system call zero volatile registers when returning powerpc/64: implement soft interrupt replay in C powerpc/64s: interrupt implement exit logic in C powerpc/64s/exception: remove lite interrupt return powerpc/64: system call reconcile interrupts powerpc/64s/exception: treat NIA below __end_interrupts as soft-masked powerpc/64s: system call support for scv/rfscv instructions
Documentation/powerpc/syscall64-abi.rst | 42 +- arch/powerpc/include/asm/asm-prototypes.h | 17 +- .../powerpc/include/asm/book3s/64/kup-radix.h | 24 +- arch/powerpc/include/asm/cputime.h | 29 + arch/powerpc/include/asm/exception-64s.h | 10 +- arch/powerpc/include/asm/head-64.h | 2 +- arch/powerpc/include/asm/hw_irq.h | 6 +- arch/powerpc/include/asm/ppc_asm.h | 2 + arch/powerpc/include/asm/processor.h | 2 +- arch/powerpc/include/asm/ptrace.h | 3 + arch/powerpc/include/asm/setup.h | 4 +- arch/powerpc/include/asm/signal.h | 3 + arch/powerpc/include/asm/switch_to.h | 11 + arch/powerpc/include/asm/time.h | 4 +- arch/powerpc/kernel/Makefile | 3 +- arch/powerpc/kernel/cpu_setup_power.S | 2 +- arch/powerpc/kernel/cputable.c | 3 +- arch/powerpc/kernel/dt_cpu_ftrs.c | 1 + arch/powerpc/kernel/entry_64.S | 1017 +++----- arch/powerpc/kernel/exceptions-64e.S | 287 ++- arch/powerpc/kernel/exceptions-64s.S | 2168 ++++++++++++----- arch/powerpc/kernel/irq.c | 183 +- arch/powerpc/kernel/process.c | 89 +- arch/powerpc/kernel/setup_64.c | 5 +- arch/powerpc/kernel/signal.h | 2 - arch/powerpc/kernel/syscall_64.c | 379 +++ arch/powerpc/kernel/syscalls/syscall.tbl | 22 +- arch/powerpc/kernel/systbl.S | 9 +- arch/powerpc/kernel/time.c | 9 - arch/powerpc/kernel/vector.S | 2 +- arch/powerpc/kvm/book3s_hv_rmhandlers.S | 11 - arch/powerpc/kvm/book3s_segment.S | 7 - arch/powerpc/lib/sstep.c | 5 +- arch/powerpc/platforms/pseries/setup.c | 8 +- 34 files changed, 2769 insertions(+), 1602 deletions(-) create mode 100644 arch/powerpc/kernel/syscall_64.c
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 1/32 powerpc/64s/exception: Introduce INT_DEFINE parameter block for code generation
The code generation macro arguments are difficult to read, and defaults can't easily be used.
This introduces a block where parameters can be set for interrupt handler code generation by the subsequent macros, and adds the first generation macro for interrupt entry.
One interrupt handler is converted to the new macros to demonstrate the change, the rest will be coverted all at once.
No generated code change.
--- arch/powerpc/kernel/exceptions-64s.S | 77 ++++++++++++++++++++++++++-- 1 file changed, 73 insertions(+), 4 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +73 -4
@@ -193,6 +193,61 @@ END_FTR_SECTION_NESTED(ftr,ftr,943) mtctr reg; \ bctr +/* + * Interrupt code generation macros + */ +#define IVEC .L_IVEC_\name\() +#define IHSRR .L_IHSRR_\name\() +#define IAREA .L_IAREA_\name\() +#define IDAR .L_IDAR_\name\() +#define IDSISR .L_IDSISR_\name\() +#define ISET_RI .L_ISET_RI_\name\() +#define IEARLY .L_IEARLY_\name\() +#define IMASK .L_IMASK_\name\() +#define IKVM_REAL .L_IKVM_REAL_\name\() +#define IKVM_VIRT .L_IKVM_VIRT_\name\() + +#define INT_DEFINE_BEGIN(n) \ +.macro int_define_ ## n name + +#define INT_DEFINE_END(n) \ +.endm ; \ +int_define_ ## n n ; \ +do_define_int n + +.macro do_define_int name + .ifndef IVEC + .error "IVEC not defined" + .endif + .ifndef IHSRR + IHSRR=EXC_STD + .endif + .ifndef IAREA + IAREA=PACA_EXGEN + .endif + .ifndef IDAR + IDAR=0 + .endif + .ifndef IDSISR + IDSISR=0 + .endif + .ifndef ISET_RI + ISET_RI=1 + .endif + .ifndef IEARLY + IEARLY=0 + .endif + .ifndef IMASK + IMASK=0 + .endif + .ifndef IKVM_REAL + IKVM_REAL=0 + .endif + .ifndef IKVM_VIRT + IKVM_VIRT=0 + .endif +.endm + .macro INT_KVM_HANDLER name, vec, hsrr, area, skip TRAMP_KVM_BEGIN(\name\()_kvm) KVM_HANDLER \vec, \hsrr, \area, \skip
@@ -474,7 +529,7 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) */ GET_SCRATCH0(r10) std r10,\area\()+EX_R13(r13) - .if \dar + .if \dar == 1 .if \hsrr mfspr r10,SPRN_HDAR .else
@@ -482,7 +537,7 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) .endif std r10,\area\()+EX_DAR(r13) .endif - .if \dsisr + .if \dsisr == 1 .if \hsrr mfspr r10,SPRN_HDSISR .else
@@ -506,6 +561,14 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) .endif .endm +.macro GEN_INT_ENTRY name, virt, ool=0 + .if ! \virt + INT_HANDLER \name, IVEC, \ool, IEARLY, \virt, IHSRR, IAREA, ISET_RI, IDAR, IDSISR, IMASK, IKVM_REAL + .else + INT_HANDLER \name, IVEC, \ool, IEARLY, \virt, IHSRR, IAREA, ISET_RI, IDAR, IDSISR, IMASK, IKVM_VIRT + .endif +.endm + /* * On entry r13 points to the paca, r9-r13 are saved in the paca, * r9 contains the saved CR, r11 and r12 contain the saved SRR0 and
@@ -1143,12 +1206,18 @@ END_FTR_SECTION_IFSET(CPU_FTR_HVMODE) bl unrecoverable_exception b . +INT_DEFINE_BEGIN(data_access) + IVEC=0x300 + IDAR=1 + IDSISR=1 + IKVM_REAL=1 +INT_DEFINE_END(data_access) EXC_REAL_BEGIN(data_access, 0x300, 0x80) - INT_HANDLER data_access, 0x300, ool=1, dar=1, dsisr=1, kvm=1 + GEN_INT_ENTRY data_access, virt=0, ool=1 EXC_REAL_END(data_access, 0x300, 0x80) EXC_VIRT_BEGIN(data_access, 0x4300, 0x80) - INT_HANDLER data_access, 0x300, virt=1, dar=1, dsisr=1 + GEN_INT_ENTRY data_access, virt=1 EXC_VIRT_END(data_access, 0x4300, 0x80) INT_KVM_HANDLER data_access, 0x300, EXC_STD, PACA_EXGEN, 1 EXC_COMMON_BEGIN(data_access_common)
Signature · 1 line
2.23.0
Re: [PATCH v3 01/32] powerpc/64s/exception: Introduce INT_DEFINE parameter block for code generation
Quoted 7 lines · Nicholas Piggin
The code generation macro arguments are difficult to read, and defaults can't easily be used.
This introduces a block where parameters can be set for interrupt handler code generation by the subsequent macros, and adds the first generation macro for interrupt entry.
Quoted 6 lines
One interrupt handler is converted to the new macros to demonstrate the change, the rest will be coverted all at once.
No generated code change.
Signed-off-by Nicholas Piggin <[email protected]>
Patches 1-30 applied to powerpc next, thanks.
https://git.kernel.org/powerpc/c/a42a239db3262b8185cb1a07a9350392ef1439ca
cheers
Nicholas Piggin
reviewerPATCH v3 2/32 powerpc/64s/exception: Add GEN_COMMON macro that uses INT_DEFINE parameters
In reply to Nicholas PigginNo generated code change.
--- arch/powerpc/kernel/exceptions-64s.S | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +17 -7
@@ -206,6 +206,9 @@ END_FTR_SECTION_NESTED(ftr,ftr,943) #define IMASK .L_IMASK_\name\() #define IKVM_REAL .L_IKVM_REAL_\name\() #define IKVM_VIRT .L_IKVM_VIRT_\name\() +#define ISTACK .L_ISTACK_\name\() +#define IRECONCILE .L_IRECONCILE_\name\() +#define IKUAP .L_IKUAP_\name\() #define INT_DEFINE_BEGIN(n) \ .macro int_define_ ## n name
@@ -246,6 +249,15 @@ do_define_int n .ifndef IKVM_VIRT IKVM_VIRT=0 .endif + .ifndef ISTACK + ISTACK=1 + .endif + .ifndef IRECONCILE + IRECONCILE=1 + .endif + .ifndef IKUAP + IKUAP=1 + .endif .endm .macro INT_KVM_HANDLER name, vec, hsrr, area, skip
@@ -670,6 +682,10 @@ END_FTR_SECTION_NESTED(CPU_FTR_CFAR, CPU_FTR_CFAR, 66) .endif .endm +.macro GEN_COMMON name + INT_COMMON IVEC, IAREA, ISTACK, IKUAP, IRECONCILE, IDAR, IDSISR +.endm + /* * Restore all registers including H/SRR0/1 saved in a stack frame of a * standard exception.
@@ -1221,13 +1237,7 @@ EXC_VIRT_BEGIN(data_access, 0x4300, 0x80) EXC_VIRT_END(data_access, 0x4300, 0x80) INT_KVM_HANDLER data_access, 0x300, EXC_STD, PACA_EXGEN, 1 EXC_COMMON_BEGIN(data_access_common) - /* - * Here r13 points to the paca, r9 contains the saved CR, - * SRR0 and SRR1 are saved in r11 and r12, - * r9 - r13 are saved in paca->exgen. - * EX_DAR and EX_DSISR have saved DAR/DSISR - */ - INT_COMMON 0x300, PACA_EXGEN, 1, 1, 1, 1, 1 + GEN_COMMON data_access ld r4,_DAR(r1) ld r5,_DSISR(r1) BEGIN_MMU_FTR_SECTION
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 3/32 powerpc/64s/exception: Add GEN_KVM macro that uses INT_DEFINE parameters
In reply to Nicholas PigginNo generated code change.
--- arch/powerpc/kernel/exceptions-64s.S | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-)
arch/powerpc/kernel/exceptions-64s.S +11 -1
@@ -204,6 +204,7 @@ END_FTR_SECTION_NESTED(ftr,ftr,943) #define ISET_RI .L_ISET_RI_\name\() #define IEARLY .L_IEARLY_\name\() #define IMASK .L_IMASK_\name\() +#define IKVM_SKIP .L_IKVM_SKIP_\name\() #define IKVM_REAL .L_IKVM_REAL_\name\() #define IKVM_VIRT .L_IKVM_VIRT_\name\() #define ISTACK .L_ISTACK_\name\()
@@ -243,6 +244,9 @@ do_define_int n .ifndef IMASK IMASK=0 .endif + .ifndef IKVM_SKIP + IKVM_SKIP=0 + .endif .ifndef IKVM_REAL IKVM_REAL=0 .endif
@@ -265,6 +269,10 @@ do_define_int n KVM_HANDLER \vec, \hsrr, \area, \skip .endm +.macro GEN_KVM name + KVM_HANDLER IVEC, IHSRR, IAREA, IKVM_SKIP +.endm + #ifdef CONFIG_KVM_BOOK3S_64_HANDLER #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE /*
@@ -1226,6 +1234,7 @@ INT_DEFINE_BEGIN(data_access) IVEC=0x300 IDAR=1 IDSISR=1 + IKVM_SKIP=1 IKVM_REAL=1 INT_DEFINE_END(data_access)
@@ -1235,7 +1244,8 @@ EXC_REAL_END(data_access, 0x300, 0x80) EXC_VIRT_BEGIN(data_access, 0x4300, 0x80) GEN_INT_ENTRY data_access, virt=1 EXC_VIRT_END(data_access, 0x4300, 0x80) -INT_KVM_HANDLER data_access, 0x300, EXC_STD, PACA_EXGEN, 1 +TRAMP_KVM_BEGIN(data_access_kvm) + GEN_KVM data_access EXC_COMMON_BEGIN(data_access_common) GEN_COMMON data_access ld r4,_DAR(r1)
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 4/32 powerpc/64s/exception: Expand EXC_COMMON and EXC_COMMON_ASYNC macros
In reply to Nicholas PigginThese don't provide a large amount of code sharing. Removing them makes code easier to shuffle around. For example, some of the common instructions will be moved into the common code gen macro.
No generated code change.
--- arch/powerpc/kernel/exceptions-64s.S | 160 ++++++++++++++++++++------- 1 file changed, 117 insertions(+), 43 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +117 -43
@@ -757,28 +757,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_CAN_NAP) #define FINISH_NAP #endif -#define EXC_COMMON(name, realvec, hdlr) \ - EXC_COMMON_BEGIN(name); \ - INT_COMMON realvec, PACA_EXGEN, 1, 1, 1, 0, 0 ; \ - bl save_nvgprs; \ - addi r3,r1,STACK_FRAME_OVERHEAD; \ - bl hdlr; \ - b ret_from_except - -/* - * Like EXC_COMMON, but for exceptions that can occur in the idle task and - * therefore need the special idle handling (finish nap and runlatch) - */ -#define EXC_COMMON_ASYNC(name, realvec, hdlr) \ - EXC_COMMON_BEGIN(name); \ - INT_COMMON realvec, PACA_EXGEN, 1, 1, 1, 0, 0 ; \ - FINISH_NAP; \ - RUNLATCH_ON; \ - addi r3,r1,STACK_FRAME_OVERHEAD; \ - bl hdlr; \ - b ret_from_except_lite - - /* * There are a few constraints to be concerned with. * - Real mode exceptions code/data must be located at their physical location.
@@ -1349,7 +1327,13 @@ EXC_VIRT_BEGIN(hardware_interrupt, 0x4500, 0x100) INT_HANDLER hardware_interrupt, 0x500, virt=1, hsrr=EXC_HV_OR_STD, bitmask=IRQS_DISABLED, kvm=1 EXC_VIRT_END(hardware_interrupt, 0x4500, 0x100) INT_KVM_HANDLER hardware_interrupt, 0x500, EXC_HV_OR_STD, PACA_EXGEN, 0 -EXC_COMMON_ASYNC(hardware_interrupt_common, 0x500, do_IRQ) +EXC_COMMON_BEGIN(hardware_interrupt_common) + INT_COMMON 0x500, PACA_EXGEN, 1, 1, 1, 0, 0 + FINISH_NAP + RUNLATCH_ON + addi r3,r1,STACK_FRAME_OVERHEAD + bl do_IRQ + b ret_from_except_lite EXC_REAL_BEGIN(alignment, 0x600, 0x100)
@@ -1455,7 +1439,13 @@ EXC_VIRT_BEGIN(decrementer, 0x4900, 0x80) INT_HANDLER decrementer, 0x900, virt=1, bitmask=IRQS_DISABLED EXC_VIRT_END(decrementer, 0x4900, 0x80) INT_KVM_HANDLER decrementer, 0x900, EXC_STD, PACA_EXGEN, 0 -EXC_COMMON_ASYNC(decrementer_common, 0x900, timer_interrupt) +EXC_COMMON_BEGIN(decrementer_common) + INT_COMMON 0x900, PACA_EXGEN, 1, 1, 1, 0, 0 + FINISH_NAP + RUNLATCH_ON + addi r3,r1,STACK_FRAME_OVERHEAD + bl timer_interrupt + b ret_from_except_lite EXC_REAL_BEGIN(hdecrementer, 0x980, 0x80)
@@ -1465,7 +1455,12 @@ EXC_VIRT_BEGIN(hdecrementer, 0x4980, 0x80) INT_HANDLER hdecrementer, 0x980, virt=1, hsrr=EXC_HV, kvm=1 EXC_VIRT_END(hdecrementer, 0x4980, 0x80) INT_KVM_HANDLER hdecrementer, 0x980, EXC_HV, PACA_EXGEN, 0 -EXC_COMMON(hdecrementer_common, 0x980, hdec_interrupt) +EXC_COMMON_BEGIN(hdecrementer_common) + INT_COMMON 0x980, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl hdec_interrupt + b ret_from_except EXC_REAL_BEGIN(doorbell_super, 0xa00, 0x100)
@@ -1475,11 +1470,17 @@ EXC_VIRT_BEGIN(doorbell_super, 0x4a00, 0x100) INT_HANDLER doorbell_super, 0xa00, virt=1, bitmask=IRQS_DISABLED EXC_VIRT_END(doorbell_super, 0x4a00, 0x100) INT_KVM_HANDLER doorbell_super, 0xa00, EXC_STD, PACA_EXGEN, 0 +EXC_COMMON_BEGIN(doorbell_super_common) + INT_COMMON 0xa00, PACA_EXGEN, 1, 1, 1, 0, 0 + FINISH_NAP + RUNLATCH_ON + addi r3,r1,STACK_FRAME_OVERHEAD #ifdef CONFIG_PPC_DOORBELL -EXC_COMMON_ASYNC(doorbell_super_common, 0xa00, doorbell_exception) + bl doorbell_exception #else -EXC_COMMON_ASYNC(doorbell_super_common, 0xa00, unknown_exception) + bl unknown_exception #endif + b ret_from_except_lite EXC_REAL_NONE(0xb00, 0x100)
@@ -1610,7 +1611,12 @@ EXC_VIRT_BEGIN(single_step, 0x4d00, 0x100) INT_HANDLER single_step, 0xd00, virt=1 EXC_VIRT_END(single_step, 0x4d00, 0x100) INT_KVM_HANDLER single_step, 0xd00, EXC_STD, PACA_EXGEN, 0 -EXC_COMMON(single_step_common, 0xd00, single_step_exception) +EXC_COMMON_BEGIN(single_step_common) + INT_COMMON 0xd00, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl single_step_exception + b ret_from_except EXC_REAL_BEGIN(h_data_storage, 0xe00, 0x20)
@@ -1641,7 +1647,12 @@ EXC_VIRT_BEGIN(h_instr_storage, 0x4e20, 0x20) INT_HANDLER h_instr_storage, 0xe20, ool=1, virt=1, hsrr=EXC_HV, kvm=1 EXC_VIRT_END(h_instr_storage, 0x4e20, 0x20) INT_KVM_HANDLER h_instr_storage, 0xe20, EXC_HV, PACA_EXGEN, 0 -EXC_COMMON(h_instr_storage_common, 0xe20, unknown_exception) +EXC_COMMON_BEGIN(h_instr_storage_common) + INT_COMMON 0xe20, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl unknown_exception + b ret_from_except EXC_REAL_BEGIN(emulation_assist, 0xe40, 0x20)
@@ -1651,7 +1662,12 @@ EXC_VIRT_BEGIN(emulation_assist, 0x4e40, 0x20) INT_HANDLER emulation_assist, 0xe40, ool=1, virt=1, hsrr=EXC_HV, kvm=1 EXC_VIRT_END(emulation_assist, 0x4e40, 0x20) INT_KVM_HANDLER emulation_assist, 0xe40, EXC_HV, PACA_EXGEN, 0 -EXC_COMMON(emulation_assist_common, 0xe40, emulation_assist_interrupt) +EXC_COMMON_BEGIN(emulation_assist_common) + INT_COMMON 0xe40, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl emulation_assist_interrupt + b ret_from_except /*
@@ -1708,11 +1724,17 @@ EXC_VIRT_BEGIN(h_doorbell, 0x4e80, 0x20) INT_HANDLER h_doorbell, 0xe80, ool=1, virt=1, hsrr=EXC_HV, bitmask=IRQS_DISABLED, kvm=1 EXC_VIRT_END(h_doorbell, 0x4e80, 0x20) INT_KVM_HANDLER h_doorbell, 0xe80, EXC_HV, PACA_EXGEN, 0 +EXC_COMMON_BEGIN(h_doorbell_common) + INT_COMMON 0xe80, PACA_EXGEN, 1, 1, 1, 0, 0 + FINISH_NAP + RUNLATCH_ON + addi r3,r1,STACK_FRAME_OVERHEAD #ifdef CONFIG_PPC_DOORBELL -EXC_COMMON_ASYNC(h_doorbell_common, 0xe80, doorbell_exception) + bl doorbell_exception #else -EXC_COMMON_ASYNC(h_doorbell_common, 0xe80, unknown_exception) + bl unknown_exception #endif + b ret_from_except_lite EXC_REAL_BEGIN(h_virt_irq, 0xea0, 0x20)
@@ -1722,7 +1744,13 @@ EXC_VIRT_BEGIN(h_virt_irq, 0x4ea0, 0x20) INT_HANDLER h_virt_irq, 0xea0, ool=1, virt=1, hsrr=EXC_HV, bitmask=IRQS_DISABLED, kvm=1 EXC_VIRT_END(h_virt_irq, 0x4ea0, 0x20) INT_KVM_HANDLER h_virt_irq, 0xea0, EXC_HV, PACA_EXGEN, 0 -EXC_COMMON_ASYNC(h_virt_irq_common, 0xea0, do_IRQ) +EXC_COMMON_BEGIN(h_virt_irq_common) + INT_COMMON 0xea0, PACA_EXGEN, 1, 1, 1, 0, 0 + FINISH_NAP + RUNLATCH_ON + addi r3,r1,STACK_FRAME_OVERHEAD + bl do_IRQ + b ret_from_except_lite EXC_REAL_NONE(0xec0, 0x20)
@@ -1738,7 +1766,13 @@ EXC_VIRT_BEGIN(performance_monitor, 0x4f00, 0x20) INT_HANDLER performance_monitor, 0xf00, ool=1, virt=1, bitmask=IRQS_PMI_DISABLED EXC_VIRT_END(performance_monitor, 0x4f00, 0x20) INT_KVM_HANDLER performance_monitor, 0xf00, EXC_STD, PACA_EXGEN, 0 -EXC_COMMON_ASYNC(performance_monitor_common, 0xf00, performance_monitor_exception) +EXC_COMMON_BEGIN(performance_monitor_common) + INT_COMMON 0xf00, PACA_EXGEN, 1, 1, 1, 0, 0 + FINISH_NAP + RUNLATCH_ON + addi r3,r1,STACK_FRAME_OVERHEAD + bl performance_monitor_exception + b ret_from_except_lite EXC_REAL_BEGIN(altivec_unavailable, 0xf20, 0x20)
@@ -1829,7 +1863,12 @@ EXC_VIRT_BEGIN(facility_unavailable, 0x4f60, 0x20) INT_HANDLER facility_unavailable, 0xf60, ool=1, virt=1 EXC_VIRT_END(facility_unavailable, 0x4f60, 0x20) INT_KVM_HANDLER facility_unavailable, 0xf60, EXC_STD, PACA_EXGEN, 0 -EXC_COMMON(facility_unavailable_common, 0xf60, facility_unavailable_exception) +EXC_COMMON_BEGIN(facility_unavailable_common) + INT_COMMON 0xf60, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl facility_unavailable_exception + b ret_from_except EXC_REAL_BEGIN(h_facility_unavailable, 0xf80, 0x20)
@@ -1839,7 +1878,12 @@ EXC_VIRT_BEGIN(h_facility_unavailable, 0x4f80, 0x20) INT_HANDLER h_facility_unavailable, 0xf80, ool=1, virt=1, hsrr=EXC_HV, kvm=1 EXC_VIRT_END(h_facility_unavailable, 0x4f80, 0x20) INT_KVM_HANDLER h_facility_unavailable, 0xf80, EXC_HV, PACA_EXGEN, 0 -EXC_COMMON(h_facility_unavailable_common, 0xf80, facility_unavailable_exception) +EXC_COMMON_BEGIN(h_facility_unavailable_common) + INT_COMMON 0xf80, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl facility_unavailable_exception + b ret_from_except EXC_REAL_NONE(0xfa0, 0x20)
@@ -1860,7 +1904,12 @@ EXC_REAL_BEGIN(cbe_system_error, 0x1200, 0x100) EXC_REAL_END(cbe_system_error, 0x1200, 0x100) EXC_VIRT_NONE(0x5200, 0x100) INT_KVM_HANDLER cbe_system_error, 0x1200, EXC_HV, PACA_EXGEN, 1 -EXC_COMMON(cbe_system_error_common, 0x1200, cbe_system_error_exception) +EXC_COMMON_BEGIN(cbe_system_error_common) + INT_COMMON 0x1200, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl cbe_system_error_exception + b ret_from_except #else /* CONFIG_CBE_RAS */ EXC_REAL_NONE(0x1200, 0x100) EXC_VIRT_NONE(0x5200, 0x100)
@@ -1874,7 +1923,12 @@ EXC_VIRT_BEGIN(instruction_breakpoint, 0x5300, 0x100) INT_HANDLER instruction_breakpoint, 0x1300, virt=1 EXC_VIRT_END(instruction_breakpoint, 0x5300, 0x100) INT_KVM_HANDLER instruction_breakpoint, 0x1300, EXC_STD, PACA_EXGEN, 1 -EXC_COMMON(instruction_breakpoint_common, 0x1300, instruction_breakpoint_exception) +EXC_COMMON_BEGIN(instruction_breakpoint_common) + INT_COMMON 0x1300, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl instruction_breakpoint_exception + b ret_from_except EXC_REAL_NONE(0x1400, 0x100)
@@ -1974,7 +2028,12 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR) b . #endif -EXC_COMMON(denorm_common, 0x1500, unknown_exception) +EXC_COMMON_BEGIN(denorm_common) + INT_COMMON 0x1500, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl unknown_exception + b ret_from_except #ifdef CONFIG_CBE_RAS
@@ -1983,7 +2042,12 @@ EXC_REAL_BEGIN(cbe_maintenance, 0x1600, 0x100) EXC_REAL_END(cbe_maintenance, 0x1600, 0x100) EXC_VIRT_NONE(0x5600, 0x100) INT_KVM_HANDLER cbe_maintenance, 0x1600, EXC_HV, PACA_EXGEN, 1 -EXC_COMMON(cbe_maintenance_common, 0x1600, cbe_maintenance_exception) +EXC_COMMON_BEGIN(cbe_maintenance_common) + INT_COMMON 0x1600, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl cbe_maintenance_exception + b ret_from_except #else /* CONFIG_CBE_RAS */ EXC_REAL_NONE(0x1600, 0x100) EXC_VIRT_NONE(0x5600, 0x100)
@@ -1997,11 +2061,16 @@ EXC_VIRT_BEGIN(altivec_assist, 0x5700, 0x100) INT_HANDLER altivec_assist, 0x1700, virt=1 EXC_VIRT_END(altivec_assist, 0x5700, 0x100) INT_KVM_HANDLER altivec_assist, 0x1700, EXC_STD, PACA_EXGEN, 0 +EXC_COMMON_BEGIN(altivec_assist_common) + INT_COMMON 0x1700, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD #ifdef CONFIG_ALTIVEC -EXC_COMMON(altivec_assist_common, 0x1700, altivec_assist_exception) + bl altivec_assist_exception #else -EXC_COMMON(altivec_assist_common, 0x1700, unknown_exception) + bl unknown_exception #endif + b ret_from_except #ifdef CONFIG_CBE_RAS
@@ -2010,7 +2079,12 @@ EXC_REAL_BEGIN(cbe_thermal, 0x1800, 0x100) EXC_REAL_END(cbe_thermal, 0x1800, 0x100) EXC_VIRT_NONE(0x5800, 0x100) INT_KVM_HANDLER cbe_thermal, 0x1800, EXC_HV, PACA_EXGEN, 1 -EXC_COMMON(cbe_thermal_common, 0x1800, cbe_thermal_exception) +EXC_COMMON_BEGIN(cbe_thermal_common) + INT_COMMON 0x1800, PACA_EXGEN, 1, 1, 1, 0, 0 + bl save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl cbe_thermal_exception + b ret_from_except #else /* CONFIG_CBE_RAS */ EXC_REAL_NONE(0x1800, 0x100) EXC_VIRT_NONE(0x5800, 0x100)
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 5/32 powerpc/64s/exception: Move all interrupt handlers to new style code gen macros
In reply to Nicholas PigginAside from label names and BUG line numbers, the generated code change is an additional HMI KVM handler added for the "late" KVM handler, because early and late HMI generation is achieved by defining two different interrupt types.
--- arch/powerpc/kernel/exceptions-64s.S | 556 ++++++++++++++++++++------- 1 file changed, 418 insertions(+), 138 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +418 -138
@@ -206,8 +206,10 @@ END_FTR_SECTION_NESTED(ftr,ftr,943) #define IMASK .L_IMASK_\name\() #define IKVM_SKIP .L_IKVM_SKIP_\name\() #define IKVM_REAL .L_IKVM_REAL_\name\() +#define __IKVM_REAL(name) .L_IKVM_REAL_ ## name #define IKVM_VIRT .L_IKVM_VIRT_\name\() #define ISTACK .L_ISTACK_\name\() +#define __ISTACK(name) .L_ISTACK_ ## name #define IRECONCILE .L_IRECONCILE_\name\() #define IKUAP .L_IKUAP_\name\()
@@ -570,7 +572,7 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) /* nothing more */ .elseif \early mfctr r10 /* save ctr, even for !RELOCATABLE */ - BRANCH_TO_C000(r11, \name\()_early_common) + BRANCH_TO_C000(r11, \name\()_common) .elseif !\virt INT_SAVE_SRR_AND_JUMP \name\()_common, \hsrr, \ri .else
@@ -843,6 +845,19 @@ __start_interrupts: EXC_VIRT_NONE(0x4000, 0x100) +INT_DEFINE_BEGIN(system_reset) + IVEC=0x100 + IAREA=PACA_EXNMI + /* + * MSR_RI is not enabled, because PACA_EXNMI and nmi stack is + * being used, so a nested NMI exception would corrupt it. + */ + ISET_RI=0 + ISTACK=0 + IRECONCILE=0 + IKVM_REAL=1 +INT_DEFINE_END(system_reset) + EXC_REAL_BEGIN(system_reset, 0x100, 0x100) #ifdef CONFIG_PPC_P7_NAP /*
@@ -880,11 +895,8 @@ BEGIN_FTR_SECTION END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) #endif - INT_HANDLER system_reset, 0x100, area=PACA_EXNMI, ri=0, kvm=1 + GEN_INT_ENTRY system_reset, virt=0 /* - * MSR_RI is not enabled, because PACA_EXNMI and nmi stack is - * being used, so a nested NMI exception would corrupt it. - * * In theory, we should not enable relocation here if it was disabled * in SRR1, because the MMU may not be configured to support it (e.g., * SLB may have been cleared). In practice, there should only be a few
@@ -893,7 +905,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) */ EXC_REAL_END(system_reset, 0x100, 0x100) EXC_VIRT_NONE(0x4100, 0x100) -INT_KVM_HANDLER system_reset 0x100, EXC_STD, PACA_EXNMI, 0 +TRAMP_KVM_BEGIN(system_reset_kvm) + GEN_KVM system_reset #ifdef CONFIG_PPC_P7_NAP TRAMP_REAL_BEGIN(system_reset_idle_wake)
@@ -908,8 +921,8 @@ TRAMP_REAL_BEGIN(system_reset_idle_wake) * Vectors for the FWNMI option. Share common code. */ TRAMP_REAL_BEGIN(system_reset_fwnmi) - /* See comment at system_reset exception, don't turn on RI */ - INT_HANDLER system_reset, 0x100, area=PACA_EXNMI, ri=0 + __IKVM_REAL(system_reset)=0 + GEN_INT_ENTRY system_reset, virt=0 #endif /* CONFIG_PPC_PSERIES */
@@ -929,7 +942,7 @@ EXC_COMMON_BEGIN(system_reset_common) mr r10,r1 ld r1,PACA_NMI_EMERG_SP(r13) subi r1,r1,INT_FRAME_SIZE - INT_COMMON 0x100, PACA_EXNMI, 0, 1, 0, 0, 0 + GEN_COMMON system_reset bl save_nvgprs /* * Set IRQS_ALL_DISABLED unconditionally so arch_irqs_disabled does
@@ -971,23 +984,46 @@ EXC_COMMON_BEGIN(system_reset_common) RFI_TO_USER_OR_KERNEL -EXC_REAL_BEGIN(machine_check, 0x200, 0x100) - INT_HANDLER machine_check, 0x200, early=1, area=PACA_EXMC, dar=1, dsisr=1 +INT_DEFINE_BEGIN(machine_check_early) + IVEC=0x200 + IAREA=PACA_EXMC /* * MSR_RI is not enabled, because PACA_EXMC is being used, so a * nested machine check corrupts it. machine_check_common enables * MSR_RI. */ + ISET_RI=0 + ISTACK=0 + IEARLY=1 + IDAR=1 + IDSISR=1 + IRECONCILE=0 + IKUAP=0 /* We don't touch AMR here, we never go to virtual mode */ +INT_DEFINE_END(machine_check_early) + +INT_DEFINE_BEGIN(machine_check) + IVEC=0x200 + IAREA=PACA_EXMC + ISET_RI=0 + IDAR=1 + IDSISR=1 + IKVM_SKIP=1 + IKVM_REAL=1 +INT_DEFINE_END(machine_check) + +EXC_REAL_BEGIN(machine_check, 0x200, 0x100) + GEN_INT_ENTRY machine_check_early, virt=0 EXC_REAL_END(machine_check, 0x200, 0x100) EXC_VIRT_NONE(0x4200, 0x100) #ifdef CONFIG_PPC_PSERIES TRAMP_REAL_BEGIN(machine_check_fwnmi) /* See comment at machine_check exception, don't turn on RI */ - INT_HANDLER machine_check, 0x200, early=1, area=PACA_EXMC, dar=1, dsisr=1 + GEN_INT_ENTRY machine_check_early, virt=0 #endif -INT_KVM_HANDLER machine_check 0x200, EXC_STD, PACA_EXMC, 1 +TRAMP_KVM_BEGIN(machine_check_kvm) + GEN_KVM machine_check #define MACHINE_CHECK_HANDLER_WINDUP \ /* Clear MSR_RI before setting SRR0 and SRR1. */\
@@ -1039,8 +1075,7 @@ EXC_COMMON_BEGIN(machine_check_early_common) bgt cr1,unrecoverable_mce /* Check if we hit limit of 4 */ subi r1,r1,INT_FRAME_SIZE /* alloc stack frame */ - /* We don't touch AMR here, we never go to virtual mode */ - INT_COMMON 0x200, PACA_EXMC, 0, 0, 0, 1, 1 + GEN_COMMON machine_check_early BEGIN_FTR_SECTION bl enable_machine_check
@@ -1128,15 +1163,15 @@ BEGIN_FTR_SECTION mtspr SPRN_CFAR,r10 END_FTR_SECTION_IFSET(CPU_FTR_CFAR) MACHINE_CHECK_HANDLER_WINDUP - /* See comment at machine_check exception, don't turn on RI */ - INT_HANDLER machine_check, 0x200, area=PACA_EXMC, ri=0, dar=1, dsisr=1, kvm=1 + GEN_INT_ENTRY machine_check, virt=0 EXC_COMMON_BEGIN(machine_check_common) /* * Machine check is different because we use a different * save area: PACA_EXMC instead of PACA_EXGEN. */ - INT_COMMON 0x200, PACA_EXMC, 1, 1, 1, 1, 1 + GEN_COMMON machine_check + FINISH_NAP /* Enable MSR_RI when finished with PACA_EXMC */ li r10,MSR_RI
@@ -1208,6 +1243,22 @@ END_FTR_SECTION_IFSET(CPU_FTR_HVMODE) bl unrecoverable_exception b . + +/** + * 0x300 - Data Storage Interrupt (DSI) + * This interrupt is generated due to a data access which does not have a valid + * page table entry with permissions to allow the data access to be performed. + * DAWR matches also fault here, as do RC updates, and minor misc errors e.g., + * copy/paste, AMO, certain invalid CI accesses, etc. + * + * This interrupt is delivered to the guest (HV bit unchanged). + * + * Linux HPT responds by first attempting to refill the hash table from the + * Linux page table, then going to a full page fault if the Linux page table + * entry was insufficient. RPT goes straight to full page fault. + * + * PR KVM ...? + */ INT_DEFINE_BEGIN(data_access) IVEC=0x300 IDAR=1
@@ -1237,15 +1288,25 @@ MMU_FTR_SECTION_ELSE ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) +INT_DEFINE_BEGIN(data_access_slb) + IVEC=0x380 + IAREA=PACA_EXSLB + IRECONCILE=0 + IDAR=1 + IKVM_SKIP=1 + IKVM_REAL=1 +INT_DEFINE_END(data_access_slb) + EXC_REAL_BEGIN(data_access_slb, 0x380, 0x80) - INT_HANDLER data_access_slb, 0x380, ool=1, area=PACA_EXSLB, dar=1, kvm=1 + GEN_INT_ENTRY data_access_slb, virt=0, ool=1 EXC_REAL_END(data_access_slb, 0x380, 0x80) EXC_VIRT_BEGIN(data_access_slb, 0x4380, 0x80) - INT_HANDLER data_access_slb, 0x380, virt=1, area=PACA_EXSLB, dar=1 + GEN_INT_ENTRY data_access_slb, virt=1 EXC_VIRT_END(data_access_slb, 0x4380, 0x80) -INT_KVM_HANDLER data_access_slb, 0x380, EXC_STD, PACA_EXSLB, 1 +TRAMP_KVM_BEGIN(data_access_slb_kvm) + GEN_KVM data_access_slb EXC_COMMON_BEGIN(data_access_slb_common) - INT_COMMON 0x380, PACA_EXSLB, 1, 1, 0, 1, 0 + GEN_COMMON data_access_slb ld r4,_DAR(r1) addi r3,r1,STACK_FRAME_OVERHEAD BEGIN_MMU_FTR_SECTION
@@ -1269,15 +1330,23 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) b ret_from_except +INT_DEFINE_BEGIN(instruction_access) + IVEC=0x400 + IDAR=2 + IDSISR=2 + IKVM_REAL=1 +INT_DEFINE_END(instruction_access) + EXC_REAL_BEGIN(instruction_access, 0x400, 0x80) - INT_HANDLER instruction_access, 0x400, kvm=1 + GEN_INT_ENTRY instruction_access, virt=0 EXC_REAL_END(instruction_access, 0x400, 0x80) EXC_VIRT_BEGIN(instruction_access, 0x4400, 0x80) - INT_HANDLER instruction_access, 0x400, virt=1 + GEN_INT_ENTRY instruction_access, virt=1 EXC_VIRT_END(instruction_access, 0x4400, 0x80) -INT_KVM_HANDLER instruction_access, 0x400, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(instruction_access_kvm) + GEN_KVM instruction_access EXC_COMMON_BEGIN(instruction_access_common) - INT_COMMON 0x400, PACA_EXGEN, 1, 1, 1, 2, 2 + GEN_COMMON instruction_access ld r4,_DAR(r1) ld r5,_DSISR(r1) BEGIN_MMU_FTR_SECTION
@@ -1289,15 +1358,24 @@ MMU_FTR_SECTION_ELSE ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) +INT_DEFINE_BEGIN(instruction_access_slb) + IVEC=0x480 + IAREA=PACA_EXSLB + IRECONCILE=0 + IDAR=2 + IKVM_REAL=1 +INT_DEFINE_END(instruction_access_slb) + EXC_REAL_BEGIN(instruction_access_slb, 0x480, 0x80) - INT_HANDLER instruction_access_slb, 0x480, area=PACA_EXSLB, kvm=1 + GEN_INT_ENTRY instruction_access_slb, virt=0 EXC_REAL_END(instruction_access_slb, 0x480, 0x80) EXC_VIRT_BEGIN(instruction_access_slb, 0x4480, 0x80) - INT_HANDLER instruction_access_slb, 0x480, virt=1, area=PACA_EXSLB + GEN_INT_ENTRY instruction_access_slb, virt=1 EXC_VIRT_END(instruction_access_slb, 0x4480, 0x80) -INT_KVM_HANDLER instruction_access_slb, 0x480, EXC_STD, PACA_EXSLB, 0 +TRAMP_KVM_BEGIN(instruction_access_slb_kvm) + GEN_KVM instruction_access_slb EXC_COMMON_BEGIN(instruction_access_slb_common) - INT_COMMON 0x480, PACA_EXSLB, 1, 1, 0, 2, 0 + GEN_COMMON instruction_access_slb ld r4,_DAR(r1) addi r3,r1,STACK_FRAME_OVERHEAD BEGIN_MMU_FTR_SECTION
@@ -1320,15 +1398,24 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) bl do_bad_slb_fault b ret_from_except +INT_DEFINE_BEGIN(hardware_interrupt) + IVEC=0x500 + IHSRR=EXC_HV_OR_STD + IMASK=IRQS_DISABLED + IKVM_REAL=1 + IKVM_VIRT=1 +INT_DEFINE_END(hardware_interrupt) + EXC_REAL_BEGIN(hardware_interrupt, 0x500, 0x100) - INT_HANDLER hardware_interrupt, 0x500, hsrr=EXC_HV_OR_STD, bitmask=IRQS_DISABLED, kvm=1 + GEN_INT_ENTRY hardware_interrupt, virt=0 EXC_REAL_END(hardware_interrupt, 0x500, 0x100) EXC_VIRT_BEGIN(hardware_interrupt, 0x4500, 0x100) - INT_HANDLER hardware_interrupt, 0x500, virt=1, hsrr=EXC_HV_OR_STD, bitmask=IRQS_DISABLED, kvm=1 + GEN_INT_ENTRY hardware_interrupt, virt=1 EXC_VIRT_END(hardware_interrupt, 0x4500, 0x100) -INT_KVM_HANDLER hardware_interrupt, 0x500, EXC_HV_OR_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(hardware_interrupt_kvm) + GEN_KVM hardware_interrupt EXC_COMMON_BEGIN(hardware_interrupt_common) - INT_COMMON 0x500, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON hardware_interrupt FINISH_NAP RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD
@@ -1336,28 +1423,42 @@ EXC_COMMON_BEGIN(hardware_interrupt_common) b ret_from_except_lite +INT_DEFINE_BEGIN(alignment) + IVEC=0x600 + IDAR=1 + IDSISR=1 + IKVM_REAL=1 +INT_DEFINE_END(alignment) + EXC_REAL_BEGIN(alignment, 0x600, 0x100) - INT_HANDLER alignment, 0x600, dar=1, dsisr=1, kvm=1 + GEN_INT_ENTRY alignment, virt=0 EXC_REAL_END(alignment, 0x600, 0x100) EXC_VIRT_BEGIN(alignment, 0x4600, 0x100) - INT_HANDLER alignment, 0x600, virt=1, dar=1, dsisr=1 + GEN_INT_ENTRY alignment, virt=1 EXC_VIRT_END(alignment, 0x4600, 0x100) -INT_KVM_HANDLER alignment, 0x600, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(alignment_kvm) + GEN_KVM alignment EXC_COMMON_BEGIN(alignment_common) - INT_COMMON 0x600, PACA_EXGEN, 1, 1, 1, 1, 1 + GEN_COMMON alignment bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl alignment_exception b ret_from_except +INT_DEFINE_BEGIN(program_check) + IVEC=0x700 + IKVM_REAL=1 +INT_DEFINE_END(program_check) + EXC_REAL_BEGIN(program_check, 0x700, 0x100) - INT_HANDLER program_check, 0x700, kvm=1 + GEN_INT_ENTRY program_check, virt=0 EXC_REAL_END(program_check, 0x700, 0x100) EXC_VIRT_BEGIN(program_check, 0x4700, 0x100) - INT_HANDLER program_check, 0x700, virt=1 + GEN_INT_ENTRY program_check, virt=1 EXC_VIRT_END(program_check, 0x4700, 0x100) -INT_KVM_HANDLER program_check, 0x700, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(program_check_kvm) + GEN_KVM program_check EXC_COMMON_BEGIN(program_check_common) /* * It's possible to receive a TM Bad Thing type program check with
@@ -1383,10 +1484,12 @@ EXC_COMMON_BEGIN(program_check_common) mr r10,r1 /* Save r1 */ ld r1,PACAEMERGSP(r13) /* Use emergency stack */ subi r1,r1,INT_FRAME_SIZE /* alloc stack frame */ - INT_COMMON 0x700, PACA_EXGEN, 0, 1, 1, 0, 0 + __ISTACK(program_check)=0 + GEN_COMMON program_check b 3f 2: - INT_COMMON 0x700, PACA_EXGEN, 1, 1, 1, 0, 0 + __ISTACK(program_check)=1 + GEN_COMMON program_check 3: bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD
@@ -1394,15 +1497,22 @@ EXC_COMMON_BEGIN(program_check_common) b ret_from_except +INT_DEFINE_BEGIN(fp_unavailable) + IVEC=0x800 + IRECONCILE=0 + IKVM_REAL=1 +INT_DEFINE_END(fp_unavailable) + EXC_REAL_BEGIN(fp_unavailable, 0x800, 0x100) - INT_HANDLER fp_unavailable, 0x800, kvm=1 + GEN_INT_ENTRY fp_unavailable, virt=0 EXC_REAL_END(fp_unavailable, 0x800, 0x100) EXC_VIRT_BEGIN(fp_unavailable, 0x4800, 0x100) - INT_HANDLER fp_unavailable, 0x800, virt=1 + GEN_INT_ENTRY fp_unavailable, virt=1 EXC_VIRT_END(fp_unavailable, 0x4800, 0x100) -INT_KVM_HANDLER fp_unavailable, 0x800, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(fp_unavailable_kvm) + GEN_KVM fp_unavailable EXC_COMMON_BEGIN(fp_unavailable_common) - INT_COMMON 0x800, PACA_EXGEN, 1, 1, 0, 0, 0 + GEN_COMMON fp_unavailable bne 1f /* if from user, just load it up */ bl save_nvgprs RECONCILE_IRQ_STATE(r10, r11)
@@ -1432,15 +1542,22 @@ END_FTR_SECTION_IFSET(CPU_FTR_TM) #endif +INT_DEFINE_BEGIN(decrementer) + IVEC=0x900 + IMASK=IRQS_DISABLED + IKVM_REAL=1 +INT_DEFINE_END(decrementer) + EXC_REAL_BEGIN(decrementer, 0x900, 0x80) - INT_HANDLER decrementer, 0x900, ool=1, bitmask=IRQS_DISABLED, kvm=1 + GEN_INT_ENTRY decrementer, virt=0, ool=1 EXC_REAL_END(decrementer, 0x900, 0x80) EXC_VIRT_BEGIN(decrementer, 0x4900, 0x80) - INT_HANDLER decrementer, 0x900, virt=1, bitmask=IRQS_DISABLED + GEN_INT_ENTRY decrementer, virt=1 EXC_VIRT_END(decrementer, 0x4900, 0x80) -INT_KVM_HANDLER decrementer, 0x900, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(decrementer_kvm) + GEN_KVM decrementer EXC_COMMON_BEGIN(decrementer_common) - INT_COMMON 0x900, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON decrementer FINISH_NAP RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD
@@ -1448,30 +1565,45 @@ EXC_COMMON_BEGIN(decrementer_common) b ret_from_except_lite +INT_DEFINE_BEGIN(hdecrementer) + IVEC=0x980 + IHSRR=EXC_HV + IKVM_REAL=1 + IKVM_VIRT=1 +INT_DEFINE_END(hdecrementer) + EXC_REAL_BEGIN(hdecrementer, 0x980, 0x80) - INT_HANDLER hdecrementer, 0x980, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY hdecrementer, virt=0 EXC_REAL_END(hdecrementer, 0x980, 0x80) EXC_VIRT_BEGIN(hdecrementer, 0x4980, 0x80) - INT_HANDLER hdecrementer, 0x980, virt=1, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY hdecrementer, virt=1 EXC_VIRT_END(hdecrementer, 0x4980, 0x80) -INT_KVM_HANDLER hdecrementer, 0x980, EXC_HV, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(hdecrementer_kvm) + GEN_KVM hdecrementer EXC_COMMON_BEGIN(hdecrementer_common) - INT_COMMON 0x980, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON hdecrementer bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl hdec_interrupt b ret_from_except +INT_DEFINE_BEGIN(doorbell_super) + IVEC=0xa00 + IMASK=IRQS_DISABLED + IKVM_REAL=1 +INT_DEFINE_END(doorbell_super) + EXC_REAL_BEGIN(doorbell_super, 0xa00, 0x100) - INT_HANDLER doorbell_super, 0xa00, bitmask=IRQS_DISABLED, kvm=1 + GEN_INT_ENTRY doorbell_super, virt=0 EXC_REAL_END(doorbell_super, 0xa00, 0x100) EXC_VIRT_BEGIN(doorbell_super, 0x4a00, 0x100) - INT_HANDLER doorbell_super, 0xa00, virt=1, bitmask=IRQS_DISABLED + GEN_INT_ENTRY doorbell_super, virt=1 EXC_VIRT_END(doorbell_super, 0x4a00, 0x100) -INT_KVM_HANDLER doorbell_super, 0xa00, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(doorbell_super_kvm) + GEN_KVM doorbell_super EXC_COMMON_BEGIN(doorbell_super_common) - INT_COMMON 0xa00, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON doorbell_super FINISH_NAP RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD
@@ -1604,30 +1736,47 @@ TRAMP_KVM_BEGIN(system_call_kvm) #endif +INT_DEFINE_BEGIN(single_step) + IVEC=0xd00 + IKVM_REAL=1 +INT_DEFINE_END(single_step) + EXC_REAL_BEGIN(single_step, 0xd00, 0x100) - INT_HANDLER single_step, 0xd00, kvm=1 + GEN_INT_ENTRY single_step, virt=0 EXC_REAL_END(single_step, 0xd00, 0x100) EXC_VIRT_BEGIN(single_step, 0x4d00, 0x100) - INT_HANDLER single_step, 0xd00, virt=1 + GEN_INT_ENTRY single_step, virt=1 EXC_VIRT_END(single_step, 0x4d00, 0x100) -INT_KVM_HANDLER single_step, 0xd00, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(single_step_kvm) + GEN_KVM single_step EXC_COMMON_BEGIN(single_step_common) - INT_COMMON 0xd00, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON single_step bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl single_step_exception b ret_from_except +INT_DEFINE_BEGIN(h_data_storage) + IVEC=0xe00 + IHSRR=EXC_HV + IDAR=1 + IDSISR=1 + IKVM_SKIP=1 + IKVM_REAL=1 + IKVM_VIRT=1 +INT_DEFINE_END(h_data_storage) + EXC_REAL_BEGIN(h_data_storage, 0xe00, 0x20) - INT_HANDLER h_data_storage, 0xe00, ool=1, hsrr=EXC_HV, dar=1, dsisr=1, kvm=1 + GEN_INT_ENTRY h_data_storage, virt=0, ool=1 EXC_REAL_END(h_data_storage, 0xe00, 0x20) EXC_VIRT_BEGIN(h_data_storage, 0x4e00, 0x20) - INT_HANDLER h_data_storage, 0xe00, ool=1, virt=1, hsrr=EXC_HV, dar=1, dsisr=1, kvm=1 + GEN_INT_ENTRY h_data_storage, virt=1, ool=1 EXC_VIRT_END(h_data_storage, 0x4e00, 0x20) -INT_KVM_HANDLER h_data_storage, 0xe00, EXC_HV, PACA_EXGEN, 1 +TRAMP_KVM_BEGIN(h_data_storage_kvm) + GEN_KVM h_data_storage EXC_COMMON_BEGIN(h_data_storage_common) - INT_COMMON 0xe00, PACA_EXGEN, 1, 1, 1, 1, 1 + GEN_COMMON h_data_storage bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD BEGIN_MMU_FTR_SECTION
@@ -1640,30 +1789,46 @@ ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_TYPE_RADIX) b ret_from_except +INT_DEFINE_BEGIN(h_instr_storage) + IVEC=0xe20 + IHSRR=EXC_HV + IKVM_REAL=1 + IKVM_VIRT=1 +INT_DEFINE_END(h_instr_storage) + EXC_REAL_BEGIN(h_instr_storage, 0xe20, 0x20) - INT_HANDLER h_instr_storage, 0xe20, ool=1, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY h_instr_storage, virt=0, ool=1 EXC_REAL_END(h_instr_storage, 0xe20, 0x20) EXC_VIRT_BEGIN(h_instr_storage, 0x4e20, 0x20) - INT_HANDLER h_instr_storage, 0xe20, ool=1, virt=1, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY h_instr_storage, virt=1, ool=1 EXC_VIRT_END(h_instr_storage, 0x4e20, 0x20) -INT_KVM_HANDLER h_instr_storage, 0xe20, EXC_HV, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(h_instr_storage_kvm) + GEN_KVM h_instr_storage EXC_COMMON_BEGIN(h_instr_storage_common) - INT_COMMON 0xe20, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON h_instr_storage bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl unknown_exception b ret_from_except +INT_DEFINE_BEGIN(emulation_assist) + IVEC=0xe40 + IHSRR=EXC_HV + IKVM_REAL=1 + IKVM_VIRT=1 +INT_DEFINE_END(emulation_assist) + EXC_REAL_BEGIN(emulation_assist, 0xe40, 0x20) - INT_HANDLER emulation_assist, 0xe40, ool=1, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY emulation_assist, virt=0, ool=1 EXC_REAL_END(emulation_assist, 0xe40, 0x20) EXC_VIRT_BEGIN(emulation_assist, 0x4e40, 0x20) - INT_HANDLER emulation_assist, 0xe40, ool=1, virt=1, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY emulation_assist, virt=1, ool=1 EXC_VIRT_END(emulation_assist, 0x4e40, 0x20) -INT_KVM_HANDLER emulation_assist, 0xe40, EXC_HV, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(emulation_assist_kvm) + GEN_KVM emulation_assist EXC_COMMON_BEGIN(emulation_assist_common) - INT_COMMON 0xe40, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON emulation_assist bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl emulation_assist_interrupt
@@ -1675,11 +1840,32 @@ EXC_COMMON_BEGIN(emulation_assist_common) * first, and then eventaully from there to the trampoline to get into virtual * mode. */ +INT_DEFINE_BEGIN(hmi_exception_early) + IVEC=0xe60 + IHSRR=EXC_HV + IEARLY=1 + ISTACK=0 + IRECONCILE=0 + IKUAP=0 /* We don't touch AMR here, we never go to virtual mode */ + IKVM_REAL=1 +INT_DEFINE_END(hmi_exception_early) + +INT_DEFINE_BEGIN(hmi_exception) + IVEC=0xe60 + IHSRR=EXC_HV + IMASK=IRQS_DISABLED + IKVM_REAL=1 +INT_DEFINE_END(hmi_exception) + EXC_REAL_BEGIN(hmi_exception, 0xe60, 0x20) - INT_HANDLER hmi_exception, 0xe60, ool=1, early=1, hsrr=EXC_HV, ri=0, kvm=1 + GEN_INT_ENTRY hmi_exception_early, virt=0, ool=1 EXC_REAL_END(hmi_exception, 0xe60, 0x20) EXC_VIRT_NONE(0x4e60, 0x20) -INT_KVM_HANDLER hmi_exception, 0xe60, EXC_HV, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(hmi_exception_early_kvm) + GEN_KVM hmi_exception_early +TRAMP_KVM_BEGIN(hmi_exception_kvm) + GEN_KVM hmi_exception + EXC_COMMON_BEGIN(hmi_exception_early_common) mtctr r10 /* Restore ctr */ mfspr r11,SPRN_HSRR0 /* Save HSRR0 */
@@ -1688,8 +1874,7 @@ EXC_COMMON_BEGIN(hmi_exception_early_common) ld r1,PACAEMERGSP(r13) /* Use emergency stack for realmode */ subi r1,r1,INT_FRAME_SIZE /* alloc stack frame */ - /* We don't touch AMR here, we never go to virtual mode */ - INT_COMMON 0xe60, PACA_EXGEN, 0, 0, 0, 0, 0 + GEN_COMMON hmi_exception_early addi r3,r1,STACK_FRAME_OVERHEAD bl hmi_exception_realmode
@@ -1705,10 +1890,10 @@ EXC_COMMON_BEGIN(hmi_exception_early_common) * firmware. */ EXCEPTION_RESTORE_REGS EXC_HV - INT_HANDLER hmi_exception, 0xe60, hsrr=EXC_HV, bitmask=IRQS_DISABLED, kvm=1 + GEN_INT_ENTRY hmi_exception, virt=0 EXC_COMMON_BEGIN(hmi_exception_common) - INT_COMMON 0xe60, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON hmi_exception FINISH_NAP RUNLATCH_ON bl save_nvgprs
@@ -1717,15 +1902,24 @@ EXC_COMMON_BEGIN(hmi_exception_common) b ret_from_except +INT_DEFINE_BEGIN(h_doorbell) + IVEC=0xe80 + IHSRR=EXC_HV + IMASK=IRQS_DISABLED + IKVM_REAL=1 + IKVM_VIRT=1 +INT_DEFINE_END(h_doorbell) + EXC_REAL_BEGIN(h_doorbell, 0xe80, 0x20) - INT_HANDLER h_doorbell, 0xe80, ool=1, hsrr=EXC_HV, bitmask=IRQS_DISABLED, kvm=1 + GEN_INT_ENTRY h_doorbell, virt=0, ool=1 EXC_REAL_END(h_doorbell, 0xe80, 0x20) EXC_VIRT_BEGIN(h_doorbell, 0x4e80, 0x20) - INT_HANDLER h_doorbell, 0xe80, ool=1, virt=1, hsrr=EXC_HV, bitmask=IRQS_DISABLED, kvm=1 + GEN_INT_ENTRY h_doorbell, virt=1, ool=1 EXC_VIRT_END(h_doorbell, 0x4e80, 0x20) -INT_KVM_HANDLER h_doorbell, 0xe80, EXC_HV, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(h_doorbell_kvm) + GEN_KVM h_doorbell EXC_COMMON_BEGIN(h_doorbell_common) - INT_COMMON 0xe80, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON h_doorbell FINISH_NAP RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD
@@ -1737,15 +1931,24 @@ EXC_COMMON_BEGIN(h_doorbell_common) b ret_from_except_lite +INT_DEFINE_BEGIN(h_virt_irq) + IVEC=0xea0 + IHSRR=EXC_HV + IMASK=IRQS_DISABLED + IKVM_REAL=1 + IKVM_VIRT=1 +INT_DEFINE_END(h_virt_irq) + EXC_REAL_BEGIN(h_virt_irq, 0xea0, 0x20) - INT_HANDLER h_virt_irq, 0xea0, ool=1, hsrr=EXC_HV, bitmask=IRQS_DISABLED, kvm=1 + GEN_INT_ENTRY h_virt_irq, virt=0, ool=1 EXC_REAL_END(h_virt_irq, 0xea0, 0x20) EXC_VIRT_BEGIN(h_virt_irq, 0x4ea0, 0x20) - INT_HANDLER h_virt_irq, 0xea0, ool=1, virt=1, hsrr=EXC_HV, bitmask=IRQS_DISABLED, kvm=1 + GEN_INT_ENTRY h_virt_irq, virt=1, ool=1 EXC_VIRT_END(h_virt_irq, 0x4ea0, 0x20) -INT_KVM_HANDLER h_virt_irq, 0xea0, EXC_HV, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(h_virt_irq_kvm) + GEN_KVM h_virt_irq EXC_COMMON_BEGIN(h_virt_irq_common) - INT_COMMON 0xea0, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON h_virt_irq FINISH_NAP RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD
@@ -1759,15 +1962,22 @@ EXC_REAL_NONE(0xee0, 0x20) EXC_VIRT_NONE(0x4ee0, 0x20) +INT_DEFINE_BEGIN(performance_monitor) + IVEC=0xf00 + IMASK=IRQS_PMI_DISABLED + IKVM_REAL=1 +INT_DEFINE_END(performance_monitor) + EXC_REAL_BEGIN(performance_monitor, 0xf00, 0x20) - INT_HANDLER performance_monitor, 0xf00, ool=1, bitmask=IRQS_PMI_DISABLED, kvm=1 + GEN_INT_ENTRY performance_monitor, virt=0, ool=1 EXC_REAL_END(performance_monitor, 0xf00, 0x20) EXC_VIRT_BEGIN(performance_monitor, 0x4f00, 0x20) - INT_HANDLER performance_monitor, 0xf00, ool=1, virt=1, bitmask=IRQS_PMI_DISABLED + GEN_INT_ENTRY performance_monitor, virt=1, ool=1 EXC_VIRT_END(performance_monitor, 0x4f00, 0x20) -INT_KVM_HANDLER performance_monitor, 0xf00, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(performance_monitor_kvm) + GEN_KVM performance_monitor EXC_COMMON_BEGIN(performance_monitor_common) - INT_COMMON 0xf00, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON performance_monitor FINISH_NAP RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD
@@ -1775,15 +1985,22 @@ EXC_COMMON_BEGIN(performance_monitor_common) b ret_from_except_lite +INT_DEFINE_BEGIN(altivec_unavailable) + IVEC=0xf20 + IRECONCILE=0 + IKVM_REAL=1 +INT_DEFINE_END(altivec_unavailable) + EXC_REAL_BEGIN(altivec_unavailable, 0xf20, 0x20) - INT_HANDLER altivec_unavailable, 0xf20, ool=1, kvm=1 + GEN_INT_ENTRY altivec_unavailable, virt=0, ool=1 EXC_REAL_END(altivec_unavailable, 0xf20, 0x20) EXC_VIRT_BEGIN(altivec_unavailable, 0x4f20, 0x20) - INT_HANDLER altivec_unavailable, 0xf20, ool=1, virt=1 + GEN_INT_ENTRY altivec_unavailable, virt=1, ool=1 EXC_VIRT_END(altivec_unavailable, 0x4f20, 0x20) -INT_KVM_HANDLER altivec_unavailable, 0xf20, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(altivec_unavailable_kvm) + GEN_KVM altivec_unavailable EXC_COMMON_BEGIN(altivec_unavailable_common) - INT_COMMON 0xf20, PACA_EXGEN, 1, 1, 0, 0, 0 + GEN_COMMON altivec_unavailable #ifdef CONFIG_ALTIVEC BEGIN_FTR_SECTION beq 1f
@@ -1816,15 +2033,22 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) b ret_from_except +INT_DEFINE_BEGIN(vsx_unavailable) + IVEC=0xf40 + IRECONCILE=0 + IKVM_REAL=1 +INT_DEFINE_END(vsx_unavailable) + EXC_REAL_BEGIN(vsx_unavailable, 0xf40, 0x20) - INT_HANDLER vsx_unavailable, 0xf40, ool=1, kvm=1 + GEN_INT_ENTRY vsx_unavailable, virt=0, ool=1 EXC_REAL_END(vsx_unavailable, 0xf40, 0x20) EXC_VIRT_BEGIN(vsx_unavailable, 0x4f40, 0x20) - INT_HANDLER vsx_unavailable, 0xf40, ool=1, virt=1 + GEN_INT_ENTRY vsx_unavailable, virt=1, ool=1 EXC_VIRT_END(vsx_unavailable, 0x4f40, 0x20) -INT_KVM_HANDLER vsx_unavailable, 0xf40, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(vsx_unavailable_kvm) + GEN_KVM vsx_unavailable EXC_COMMON_BEGIN(vsx_unavailable_common) - INT_COMMON 0xf40, PACA_EXGEN, 1, 1, 0, 0, 0 + GEN_COMMON vsx_unavailable #ifdef CONFIG_VSX BEGIN_FTR_SECTION beq 1f
@@ -1856,30 +2080,44 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX) b ret_from_except +INT_DEFINE_BEGIN(facility_unavailable) + IVEC=0xf60 + IKVM_REAL=1 +INT_DEFINE_END(facility_unavailable) + EXC_REAL_BEGIN(facility_unavailable, 0xf60, 0x20) - INT_HANDLER facility_unavailable, 0xf60, ool=1, kvm=1 + GEN_INT_ENTRY facility_unavailable, virt=0, ool=1 EXC_REAL_END(facility_unavailable, 0xf60, 0x20) EXC_VIRT_BEGIN(facility_unavailable, 0x4f60, 0x20) - INT_HANDLER facility_unavailable, 0xf60, ool=1, virt=1 + GEN_INT_ENTRY facility_unavailable, virt=1, ool=1 EXC_VIRT_END(facility_unavailable, 0x4f60, 0x20) -INT_KVM_HANDLER facility_unavailable, 0xf60, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(facility_unavailable_kvm) + GEN_KVM facility_unavailable EXC_COMMON_BEGIN(facility_unavailable_common) - INT_COMMON 0xf60, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON facility_unavailable bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl facility_unavailable_exception b ret_from_except +INT_DEFINE_BEGIN(h_facility_unavailable) + IVEC=0xf80 + IHSRR=EXC_HV + IKVM_REAL=1 + IKVM_VIRT=1 +INT_DEFINE_END(h_facility_unavailable) + EXC_REAL_BEGIN(h_facility_unavailable, 0xf80, 0x20) - INT_HANDLER h_facility_unavailable, 0xf80, ool=1, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY h_facility_unavailable, virt=0, ool=1 EXC_REAL_END(h_facility_unavailable, 0xf80, 0x20) EXC_VIRT_BEGIN(h_facility_unavailable, 0x4f80, 0x20) - INT_HANDLER h_facility_unavailable, 0xf80, ool=1, virt=1, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY h_facility_unavailable, virt=1, ool=1 EXC_VIRT_END(h_facility_unavailable, 0x4f80, 0x20) -INT_KVM_HANDLER h_facility_unavailable, 0xf80, EXC_HV, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(h_facility_unavailable_kvm) + GEN_KVM h_facility_unavailable EXC_COMMON_BEGIN(h_facility_unavailable_common) - INT_COMMON 0xf80, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON h_facility_unavailable bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl facility_unavailable_exception
@@ -1899,13 +2137,21 @@ EXC_REAL_NONE(0x1100, 0x100) EXC_VIRT_NONE(0x5100, 0x100) #ifdef CONFIG_CBE_RAS +INT_DEFINE_BEGIN(cbe_system_error) + IVEC=0x1200 + IHSRR=EXC_HV + IKVM_SKIP=1 + IKVM_REAL=1 +INT_DEFINE_END(cbe_system_error) + EXC_REAL_BEGIN(cbe_system_error, 0x1200, 0x100) - INT_HANDLER cbe_system_error, 0x1200, ool=1, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY cbe_system_error, virt=0 EXC_REAL_END(cbe_system_error, 0x1200, 0x100) EXC_VIRT_NONE(0x5200, 0x100) -INT_KVM_HANDLER cbe_system_error, 0x1200, EXC_HV, PACA_EXGEN, 1 +TRAMP_KVM_BEGIN(cbe_system_error_kvm) + GEN_KVM cbe_system_error EXC_COMMON_BEGIN(cbe_system_error_common) - INT_COMMON 0x1200, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON cbe_system_error bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl cbe_system_error_exception
@@ -1916,15 +2162,22 @@ EXC_VIRT_NONE(0x5200, 0x100) #endif +INT_DEFINE_BEGIN(instruction_breakpoint) + IVEC=0x1300 + IKVM_SKIP=1 + IKVM_REAL=1 +INT_DEFINE_END(instruction_breakpoint) + EXC_REAL_BEGIN(instruction_breakpoint, 0x1300, 0x100) - INT_HANDLER instruction_breakpoint, 0x1300, kvm=1 + GEN_INT_ENTRY instruction_breakpoint, virt=0 EXC_REAL_END(instruction_breakpoint, 0x1300, 0x100) EXC_VIRT_BEGIN(instruction_breakpoint, 0x5300, 0x100) - INT_HANDLER instruction_breakpoint, 0x1300, virt=1 + GEN_INT_ENTRY instruction_breakpoint, virt=1 EXC_VIRT_END(instruction_breakpoint, 0x5300, 0x100) -INT_KVM_HANDLER instruction_breakpoint, 0x1300, EXC_STD, PACA_EXGEN, 1 +TRAMP_KVM_BEGIN(instruction_breakpoint_kvm) + GEN_KVM instruction_breakpoint EXC_COMMON_BEGIN(instruction_breakpoint_common) - INT_COMMON 0x1300, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON instruction_breakpoint bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl instruction_breakpoint_exception
@@ -1934,30 +2187,35 @@ EXC_COMMON_BEGIN(instruction_breakpoint_common) EXC_REAL_NONE(0x1400, 0x100) EXC_VIRT_NONE(0x5400, 0x100) -EXC_REAL_BEGIN(denorm_exception_hv, 0x1500, 0x100) - INT_HANDLER denorm_exception_hv, 0x1500, early=2, hsrr=EXC_HV +INT_DEFINE_BEGIN(denorm_exception) + IVEC=0x1500 + IHSRR=EXC_HV + IEARLY=2 +INT_DEFINE_END(denorm_exception) + +EXC_REAL_BEGIN(denorm_exception, 0x1500, 0x100) + GEN_INT_ENTRY denorm_exception, virt=0 #ifdef CONFIG_PPC_DENORMALISATION mfspr r10,SPRN_HSRR1 andis. r10,r10,(HSRR1_DENORM)@h /* denorm? */ bne+ denorm_assist #endif - KVMTEST denorm_exception_hv, EXC_HV 0x1500 - INT_SAVE_SRR_AND_JUMP denorm_common, EXC_HV, 1 -EXC_REAL_END(denorm_exception_hv, 0x1500, 0x100) - + KVMTEST denorm_exception, EXC_HV, 0x1500 + INT_SAVE_SRR_AND_JUMP denorm_exception_common, EXC_HV, 1 +EXC_REAL_END(denorm_exception, 0x1500, 0x100) #ifdef CONFIG_PPC_DENORMALISATION EXC_VIRT_BEGIN(denorm_exception, 0x5500, 0x100) - INT_HANDLER denorm_exception, 0x1500, 0, 2, 1, EXC_HV, PACA_EXGEN, 1, 0, 0, 0, 0 + GEN_INT_ENTRY denorm_exception, virt=1 mfspr r10,SPRN_HSRR1 andis. r10,r10,(HSRR1_DENORM)@h /* denorm? */ bne+ denorm_assist - INT_VIRT_SAVE_SRR_AND_JUMP denorm_common, EXC_HV + INT_VIRT_SAVE_SRR_AND_JUMP denorm_exception_common, EXC_HV EXC_VIRT_END(denorm_exception, 0x5500, 0x100) #else EXC_VIRT_NONE(0x5500, 0x100) #endif - -INT_KVM_HANDLER denorm_exception_hv, 0x1500, EXC_HV, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(denorm_exception_kvm) + GEN_KVM denorm_exception #ifdef CONFIG_PPC_DENORMALISATION TRAMP_REAL_BEGIN(denorm_assist)
@@ -2028,8 +2286,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR) b . #endif -EXC_COMMON_BEGIN(denorm_common) - INT_COMMON 0x1500, PACA_EXGEN, 1, 1, 1, 0, 0 +EXC_COMMON_BEGIN(denorm_exception_common) + GEN_COMMON denorm_exception bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl unknown_exception
@@ -2037,13 +2295,21 @@ EXC_COMMON_BEGIN(denorm_common) #ifdef CONFIG_CBE_RAS +INT_DEFINE_BEGIN(cbe_maintenance) + IVEC=0x1600 + IHSRR=EXC_HV + IKVM_SKIP=1 + IKVM_REAL=1 +INT_DEFINE_END(cbe_maintenance) + EXC_REAL_BEGIN(cbe_maintenance, 0x1600, 0x100) - INT_HANDLER cbe_maintenance, 0x1600, ool=1, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY cbe_maintenance, virt=0 EXC_REAL_END(cbe_maintenance, 0x1600, 0x100) EXC_VIRT_NONE(0x5600, 0x100) -INT_KVM_HANDLER cbe_maintenance, 0x1600, EXC_HV, PACA_EXGEN, 1 +TRAMP_KVM_BEGIN(cbe_maintenance_kvm) + GEN_KVM cbe_maintenance EXC_COMMON_BEGIN(cbe_maintenance_common) - INT_COMMON 0x1600, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON cbe_maintenance bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl cbe_maintenance_exception
@@ -2054,15 +2320,21 @@ EXC_VIRT_NONE(0x5600, 0x100) #endif +INT_DEFINE_BEGIN(altivec_assist) + IVEC=0x1700 + IKVM_REAL=1 +INT_DEFINE_END(altivec_assist) + EXC_REAL_BEGIN(altivec_assist, 0x1700, 0x100) - INT_HANDLER altivec_assist, 0x1700, kvm=1 + GEN_INT_ENTRY altivec_assist, virt=0 EXC_REAL_END(altivec_assist, 0x1700, 0x100) EXC_VIRT_BEGIN(altivec_assist, 0x5700, 0x100) - INT_HANDLER altivec_assist, 0x1700, virt=1 + GEN_INT_ENTRY altivec_assist, virt=1 EXC_VIRT_END(altivec_assist, 0x5700, 0x100) -INT_KVM_HANDLER altivec_assist, 0x1700, EXC_STD, PACA_EXGEN, 0 +TRAMP_KVM_BEGIN(altivec_assist_kvm) + GEN_KVM altivec_assist EXC_COMMON_BEGIN(altivec_assist_common) - INT_COMMON 0x1700, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON altivec_assist bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD #ifdef CONFIG_ALTIVEC
@@ -2074,13 +2346,21 @@ EXC_COMMON_BEGIN(altivec_assist_common) #ifdef CONFIG_CBE_RAS +INT_DEFINE_BEGIN(cbe_thermal) + IVEC=0x1800 + IHSRR=EXC_HV + IKVM_SKIP=1 + IKVM_REAL=1 +INT_DEFINE_END(cbe_thermal) + EXC_REAL_BEGIN(cbe_thermal, 0x1800, 0x100) - INT_HANDLER cbe_thermal, 0x1800, ool=1, hsrr=EXC_HV, kvm=1 + GEN_INT_ENTRY cbe_thermal, virt=0 EXC_REAL_END(cbe_thermal, 0x1800, 0x100) EXC_VIRT_NONE(0x5800, 0x100) -INT_KVM_HANDLER cbe_thermal, 0x1800, EXC_HV, PACA_EXGEN, 1 +TRAMP_KVM_BEGIN(cbe_thermal_kvm) + GEN_KVM cbe_thermal EXC_COMMON_BEGIN(cbe_thermal_common) - INT_COMMON 0x1800, PACA_EXGEN, 1, 1, 1, 0, 0 + GEN_COMMON cbe_thermal bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl cbe_thermal_exception
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 6/32 powerpc/64s/exception: Remove old INT_ENTRY macro
In reply to Nicholas Piggin--- arch/powerpc/kernel/exceptions-64s.S | 68 ++++++++++++---------------- 1 file changed, 30 insertions(+), 38 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +30 -38
@@ -482,13 +482,13 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) * - Fall through and continue executing in real, unrelocated mode. * This is done if early=2. */ -.macro INT_HANDLER name, vec, ool=0, early=0, virt=0, hsrr=0, area=PACA_EXGEN, ri=1, dar=0, dsisr=0, bitmask=0, kvm=0 +.macro GEN_INT_ENTRY name, virt, ool=0 SET_SCRATCH0(r13) /* save r13 */ GET_PACA(r13) - std r9,\area\()+EX_R9(r13) /* save r9 */ + std r9,IAREA+EX_R9(r13) /* save r9 */ OPT_GET_SPR(r9, SPRN_PPR, CPU_FTR_HAS_PPR) HMT_MEDIUM - std r10,\area\()+EX_R10(r13) /* save r10 - r12 */ + std r10,IAREA+EX_R10(r13) /* save r10 - r12 */ OPT_GET_SPR(r10, SPRN_CFAR, CPU_FTR_CFAR) .if \ool .if !\virt
@@ -502,47 +502,47 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) .endif .endif - OPT_SAVE_REG_TO_PACA(\area\()+EX_PPR, r9, CPU_FTR_HAS_PPR) - OPT_SAVE_REG_TO_PACA(\area\()+EX_CFAR, r10, CPU_FTR_CFAR) + OPT_SAVE_REG_TO_PACA(IAREA+EX_PPR, r9, CPU_FTR_HAS_PPR) + OPT_SAVE_REG_TO_PACA(IAREA+EX_CFAR, r10, CPU_FTR_CFAR) INTERRUPT_TO_KERNEL - SAVE_CTR(r10, \area\()) + SAVE_CTR(r10, IAREA) mfcr r9 - .if \kvm - KVMTEST \name \hsrr \vec + .if (!\virt && IKVM_REAL) || (\virt && IKVM_VIRT) + KVMTEST \name IHSRR IVEC .endif - .if \bitmask + .if IMASK lbz r10,PACAIRQSOFTMASK(r13) - andi. r10,r10,\bitmask + andi. r10,r10,IMASK /* Associate vector numbers with bits in paca->irq_happened */ - .if \vec == 0x500 || \vec == 0xea0 + .if IVEC == 0x500 || IVEC == 0xea0 li r10,PACA_IRQ_EE - .elseif \vec == 0x900 + .elseif IVEC == 0x900 li r10,PACA_IRQ_DEC - .elseif \vec == 0xa00 || \vec == 0xe80 + .elseif IVEC == 0xa00 || IVEC == 0xe80 li r10,PACA_IRQ_DBELL - .elseif \vec == 0xe60 + .elseif IVEC == 0xe60 li r10,PACA_IRQ_HMI - .elseif \vec == 0xf00 + .elseif IVEC == 0xf00 li r10,PACA_IRQ_PMI .else .abort "Bad maskable vector" .endif - .if \hsrr == EXC_HV_OR_STD + .if IHSRR == EXC_HV_OR_STD BEGIN_FTR_SECTION bne masked_Hinterrupt FTR_SECTION_ELSE bne masked_interrupt ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) - .elseif \hsrr + .elseif IHSRR bne masked_Hinterrupt .else bne masked_interrupt .endif .endif - std r11,\area\()+EX_R11(r13) - std r12,\area\()+EX_R12(r13) + std r11,IAREA+EX_R11(r13) + std r12,IAREA+EX_R12(r13) /* * DAR/DSISR, SCRATCH0 must be read before setting MSR[RI],
@@ -550,47 +550,39 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) * not recoverable if they are live. */ GET_SCRATCH0(r10) - std r10,\area\()+EX_R13(r13) - .if \dar == 1 - .if \hsrr + std r10,IAREA+EX_R13(r13) + .if IDAR == 1 + .if IHSRR mfspr r10,SPRN_HDAR .else mfspr r10,SPRN_DAR .endif - std r10,\area\()+EX_DAR(r13) + std r10,IAREA+EX_DAR(r13) .endif - .if \dsisr == 1 - .if \hsrr + .if IDSISR == 1 + .if IHSRR mfspr r10,SPRN_HDSISR .else mfspr r10,SPRN_DSISR .endif - stw r10,\area\()+EX_DSISR(r13) + stw r10,IAREA+EX_DSISR(r13) .endif - .if \early == 2 + .if IEARLY == 2 /* nothing more */ - .elseif \early + .elseif IEARLY mfctr r10 /* save ctr, even for !RELOCATABLE */ BRANCH_TO_C000(r11, \name\()_common) .elseif !\virt - INT_SAVE_SRR_AND_JUMP \name\()_common, \hsrr, \ri + INT_SAVE_SRR_AND_JUMP \name\()_common, IHSRR, ISET_RI .else - INT_VIRT_SAVE_SRR_AND_JUMP \name\()_common, \hsrr + INT_VIRT_SAVE_SRR_AND_JUMP \name\()_common, IHSRR .endif .if \ool .popsection .endif .endm -.macro GEN_INT_ENTRY name, virt, ool=0 - .if ! \virt - INT_HANDLER \name, IVEC, \ool, IEARLY, \virt, IHSRR, IAREA, ISET_RI, IDAR, IDSISR, IMASK, IKVM_REAL - .else - INT_HANDLER \name, IVEC, \ool, IEARLY, \virt, IHSRR, IAREA, ISET_RI, IDAR, IDSISR, IMASK, IKVM_VIRT - .endif -.endm - /* * On entry r13 points to the paca, r9-r13 are saved in the paca, * r9 contains the saved CR, r11 and r12 contain the saved SRR0 and
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 7/32 powerpc/64s/exception: Remove old INT_COMMON macro
In reply to Nicholas Piggin--- arch/powerpc/kernel/exceptions-64s.S | 51 +++++++++++++--------------- 1 file changed, 24 insertions(+), 27 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +24 -27
@@ -591,8 +591,8 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) * If stack=0, then the stack is already set in r1, and r1 is saved in r10. * PPR save and CPU accounting is not done for the !stack case (XXX why not?) */ -.macro INT_COMMON vec, area, stack, kuap, reconcile, dar, dsisr - .if \stack +.macro GEN_COMMON name + .if ISTACK andi. r10,r12,MSR_PR /* See if coming from user */ mr r10,r1 /* Save r1 */ subi r1,r1,INT_FRAME_SIZE /* alloc frame on kernel stack */
@@ -609,54 +609,54 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) std r0,GPR0(r1) /* save r0 in stackframe */ std r10,GPR1(r1) /* save r1 in stackframe */ - .if \stack - .if \kuap + .if ISTACK + .if IKUAP kuap_save_amr_and_lock r9, r10, cr1, cr0 .endif beq 101f /* if from kernel mode */ ACCOUNT_CPU_USER_ENTRY(r13, r9, r10) - SAVE_PPR(\area, r9) + SAVE_PPR(IAREA, r9) 101: .else - .if \kuap + .if IKUAP kuap_save_amr_and_lock r9, r10, cr1 .endif .endif /* Save original regs values from save area to stack frame. */ - ld r9,\area+EX_R9(r13) /* move r9, r10 to stackframe */ - ld r10,\area+EX_R10(r13) + ld r9,IAREA+EX_R9(r13) /* move r9, r10 to stackframe */ + ld r10,IAREA+EX_R10(r13) std r9,GPR9(r1) std r10,GPR10(r1) - ld r9,\area+EX_R11(r13) /* move r11 - r13 to stackframe */ - ld r10,\area+EX_R12(r13) - ld r11,\area+EX_R13(r13) + ld r9,IAREA+EX_R11(r13) /* move r11 - r13 to stackframe */ + ld r10,IAREA+EX_R12(r13) + ld r11,IAREA+EX_R13(r13) std r9,GPR11(r1) std r10,GPR12(r1) std r11,GPR13(r1) - .if \dar - .if \dar == 2 + .if IDAR + .if IDAR == 2 ld r10,_NIP(r1) .else - ld r10,\area+EX_DAR(r13) + ld r10,IAREA+EX_DAR(r13) .endif std r10,_DAR(r1) .endif - .if \dsisr - .if \dsisr == 2 + .if IDSISR + .if IDSISR == 2 ld r10,_MSR(r1) lis r11,DSISR_SRR1_MATCH_64S@h and r10,r10,r11 .else - lwz r10,\area+EX_DSISR(r13) + lwz r10,IAREA+EX_DSISR(r13) .endif std r10,_DSISR(r1) .endif BEGIN_FTR_SECTION_NESTED(66) - ld r10,\area+EX_CFAR(r13) + ld r10,IAREA+EX_CFAR(r13) std r10,ORIG_GPR3(r1) END_FTR_SECTION_NESTED(CPU_FTR_CFAR, CPU_FTR_CFAR, 66) - GET_CTR(r10, \area) + GET_CTR(r10, IAREA) std r10,_CTR(r1) std r2,GPR2(r1) /* save r2 in stackframe */ SAVE_4GPRS(3, r1) /* save r3 - r6 in stackframe */
@@ -668,26 +668,22 @@ END_FTR_SECTION_NESTED(CPU_FTR_CFAR, CPU_FTR_CFAR, 66) mfspr r11,SPRN_XER /* save XER in stackframe */ std r10,SOFTE(r1) std r11,_XER(r1) - li r9,(\vec)+1 + li r9,(IVEC)+1 std r9,_TRAP(r1) /* set trap number */ li r10,0 ld r11,exception_marker@toc(r2) std r10,RESULT(r1) /* clear regs->result */ std r11,STACK_FRAME_OVERHEAD-16(r1) /* mark the frame */ - .if \stack + .if ISTACK ACCOUNT_STOLEN_TIME .endif - .if \reconcile + .if IRECONCILE RECONCILE_IRQ_STATE(r10, r11) .endif .endm -.macro GEN_COMMON name - INT_COMMON IVEC, IAREA, ISTACK, IKUAP, IRECONCILE, IDAR, IDSISR -.endm - /* * Restore all registers including H/SRR0/1 saved in a stack frame of a * standard exception.
@@ -2387,7 +2383,8 @@ EXC_COMMON_BEGIN(soft_nmi_common) mr r10,r1 ld r1,PACAEMERGSP(r13) subi r1,r1,INT_FRAME_SIZE - INT_COMMON 0x900, PACA_EXGEN, 0, 1, 1, 0, 0 + __ISTACK(decrementer)=0 + GEN_COMMON decrementer bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl soft_nmi_interrupt
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 8/32 powerpc/64s/exception: Remove old INT_KVM_HANDLER
In reply to Nicholas Piggin--- arch/powerpc/kernel/exceptions-64s.S | 55 +++++++++++++--------------- 1 file changed, 26 insertions(+), 29 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +26 -29
@@ -266,15 +266,6 @@ do_define_int n .endif .endm -.macro INT_KVM_HANDLER name, vec, hsrr, area, skip - TRAMP_KVM_BEGIN(\name\()_kvm) - KVM_HANDLER \vec, \hsrr, \area, \skip -.endm - -.macro GEN_KVM name - KVM_HANDLER IVEC, IHSRR, IAREA, IKVM_SKIP -.endm - #ifdef CONFIG_KVM_BOOK3S_64_HANDLER #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE /*
@@ -293,35 +284,35 @@ do_define_int n bne \name\()_kvm .endm -.macro KVM_HANDLER vec, hsrr, area, skip - .if \skip +.macro GEN_KVM name + .if IKVM_SKIP cmpwi r10,KVM_GUEST_MODE_SKIP beq 89f .else BEGIN_FTR_SECTION_NESTED(947) - ld r10,\area+EX_CFAR(r13) + ld r10,IAREA+EX_CFAR(r13) std r10,HSTATE_CFAR(r13) END_FTR_SECTION_NESTED(CPU_FTR_CFAR,CPU_FTR_CFAR,947) .endif BEGIN_FTR_SECTION_NESTED(948) - ld r10,\area+EX_PPR(r13) + ld r10,IAREA+EX_PPR(r13) std r10,HSTATE_PPR(r13) END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) - ld r10,\area+EX_R10(r13) + ld r10,IAREA+EX_R10(r13) std r12,HSTATE_SCRATCH0(r13) sldi r12,r9,32 /* HSRR variants have the 0x2 bit added to their trap number */ - .if \hsrr == EXC_HV_OR_STD + .if IHSRR == EXC_HV_OR_STD BEGIN_FTR_SECTION - ori r12,r12,(\vec + 0x2) + ori r12,r12,(IVEC + 0x2) FTR_SECTION_ELSE - ori r12,r12,(\vec) + ori r12,r12,(IVEC) ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) - .elseif \hsrr - ori r12,r12,(\vec + 0x2) + .elseif IHSRR + ori r12,r12,(IVEC+ 0x2) .else - ori r12,r12,(\vec) + ori r12,r12,(IVEC) .endif #ifdef CONFIG_RELOCATABLE
@@ -334,25 +325,25 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) std r9,HSTATE_SCRATCH1(r13) __LOAD_FAR_HANDLER(r9, kvmppc_interrupt) mtctr r9 - ld r9,\area+EX_R9(r13) + ld r9,IAREA+EX_R9(r13) bctr #else - ld r9,\area+EX_R9(r13) + ld r9,IAREA+EX_R9(r13) b kvmppc_interrupt #endif - .if \skip + .if IKVM_SKIP 89: mtocrf 0x80,r9 - ld r9,\area+EX_R9(r13) - ld r10,\area+EX_R10(r13) - .if \hsrr == EXC_HV_OR_STD + ld r9,IAREA+EX_R9(r13) + ld r10,IAREA+EX_R10(r13) + .if IHSRR == EXC_HV_OR_STD BEGIN_FTR_SECTION b kvmppc_skip_Hinterrupt FTR_SECTION_ELSE b kvmppc_skip_interrupt ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) - .elseif \hsrr + .elseif IHSRR b kvmppc_skip_Hinterrupt .else b kvmppc_skip_interrupt
@@ -363,7 +354,7 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) #else .macro KVMTEST name, hsrr, n .endm -.macro KVM_HANDLER name, vec, hsrr, area, skip +.macro GEN_KVM name .endm #endif
@@ -1627,6 +1618,12 @@ EXC_VIRT_NONE(0x4b00, 0x100) * without saving, though xer is not a good idea to use, as hardware may * interpret some bits so it may be costly to change them. */ +INT_DEFINE_BEGIN(system_call) + IVEC=0xc00 + IKVM_REAL=1 + IKVM_VIRT=1 +INT_DEFINE_END(system_call) + .macro SYSTEM_CALL virt #ifdef CONFIG_KVM_BOOK3S_64_HANDLER /*
@@ -1720,7 +1717,7 @@ TRAMP_KVM_BEGIN(system_call_kvm) SET_SCRATCH0(r10) std r9,PACA_EXGEN+EX_R9(r13) mfcr r9 - KVM_HANDLER 0xc00, EXC_STD, PACA_EXGEN, 0 + GEN_KVM system_call #endif
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 9/32 powerpc/64s/exception: Add ISIDE option
In reply to Nicholas PigginRather than using DAR=2 to select the i-side registers, add an explicit option.
--- arch/powerpc/kernel/exceptions-64s.S | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +16 -7
@@ -199,6 +199,7 @@ END_FTR_SECTION_NESTED(ftr,ftr,943) #define IVEC .L_IVEC_\name\() #define IHSRR .L_IHSRR_\name\() #define IAREA .L_IAREA_\name\() +#define IISIDE .L_IISIDE_\name\() #define IDAR .L_IDAR_\name\() #define IDSISR .L_IDSISR_\name\() #define ISET_RI .L_ISET_RI_\name\()
@@ -231,6 +232,9 @@ do_define_int n .ifndef IAREA IAREA=PACA_EXGEN .endif + .ifndef IISIDE + IISIDE=0 + .endif .ifndef IDAR IDAR=0 .endif
@@ -542,7 +546,7 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) */ GET_SCRATCH0(r10) std r10,IAREA+EX_R13(r13) - .if IDAR == 1 + .if IDAR && !IISIDE .if IHSRR mfspr r10,SPRN_HDAR .else
@@ -550,7 +554,7 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) .endif std r10,IAREA+EX_DAR(r13) .endif - .if IDSISR == 1 + .if IDSISR && !IISIDE .if IHSRR mfspr r10,SPRN_HDSISR .else
@@ -625,16 +629,18 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) std r9,GPR11(r1) std r10,GPR12(r1) std r11,GPR13(r1) + .if IDAR - .if IDAR == 2 + .if IISIDE ld r10,_NIP(r1) .else ld r10,IAREA+EX_DAR(r13) .endif std r10,_DAR(r1) .endif + .if IDSISR - .if IDSISR == 2 + .if IISIDE ld r10,_MSR(r1) lis r11,DSISR_SRR1_MATCH_64S@h and r10,r10,r11
@@ -643,6 +649,7 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) .endif std r10,_DSISR(r1) .endif + BEGIN_FTR_SECTION_NESTED(66) ld r10,IAREA+EX_CFAR(r13) std r10,ORIG_GPR3(r1)
@@ -1311,8 +1318,9 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) INT_DEFINE_BEGIN(instruction_access) IVEC=0x400 - IDAR=2 - IDSISR=2 + IISIDE=1 + IDAR=1 + IDSISR=1 IKVM_REAL=1 INT_DEFINE_END(instruction_access)
@@ -1341,7 +1349,8 @@ INT_DEFINE_BEGIN(instruction_access_slb) IVEC=0x480 IAREA=PACA_EXSLB IRECONCILE=0 - IDAR=2 + IISIDE=1 + IDAR=1 IKVM_REAL=1 INT_DEFINE_END(instruction_access_slb)
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 10/32 powerpc/64s/exception: move real->virt switch into the common handler
In reply to Nicholas PigginThe real mode interrupt entry points currently use rfid to branch to the common handler in virtual mode. This is a significant amount of code, and forces other code (notably the KVM test) to live in the real mode handler.
In the interest of minimising the amount of code that runs unrelocated move the switch to virt mode into the common code, and do it with mtmsrd, which avoids clobbering SRRs (although the post-KVMTEST performance of real-mode interrupt handlers is not a big concern these days).
This requires CTR to always be saved (real-mode needs to reach 0xc...) but that's not a huge impact these days. It could be optimized away in future.
--- arch/powerpc/include/asm/exception-64s.h | 4 - arch/powerpc/kernel/exceptions-64s.S | 251 ++++++++++------------- 2 files changed, 109 insertions(+), 146 deletions(-)
arch/powerpc/include/asm/exception-64s.h +0 -4
@@ -33,11 +33,7 @@#include <asm/feature-fixups.h> /* PACA save area size in u64 units (exgen, exmc, etc) */ -#if defined(CONFIG_RELOCATABLE )#define EX_SIZE 10 -#else -#define EX_SIZE 9 -#endif /* * maximum recursive depth of MCE exceptions
arch/powerpc/kernel/exceptions-64s.S +109 -142
431 lines not on this page · raw message
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 11/32 powerpc/64s/exception: move soft-mask test to common code
In reply to Nicholas PigginAs well as moving code out of the unrelocated vectors, this allows the masked handlers to be moved to common code, and allows the soft_nmi handler to be generated more like a regular handler.
--- arch/powerpc/kernel/exceptions-64s.S | 106 +++++++++++++-------------- 1 file changed, 49 insertions(+), 57 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +49 -57
@@ -411,36 +411,6 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) .if (!\virt && IKVM_REAL) || (\virt && IKVM_VIRT) KVMTEST \name IHSRR IVEC .endif - .if IMASK - lbz r10,PACAIRQSOFTMASK(r13) - andi. r10,r10,IMASK - /* Associate vector numbers with bits in paca->irq_happened */ - .if IVEC == 0x500 || IVEC == 0xea0 - li r10,PACA_IRQ_EE - .elseif IVEC == 0x900 - li r10,PACA_IRQ_DEC - .elseif IVEC == 0xa00 || IVEC == 0xe80 - li r10,PACA_IRQ_DBELL - .elseif IVEC == 0xe60 - li r10,PACA_IRQ_HMI - .elseif IVEC == 0xf00 - li r10,PACA_IRQ_PMI - .else - .abort "Bad maskable vector" - .endif - - .if IHSRR == EXC_HV_OR_STD - BEGIN_FTR_SECTION - bne masked_Hinterrupt - FTR_SECTION_ELSE - bne masked_interrupt - ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) - .elseif IHSRR - bne masked_Hinterrupt - .else - bne masked_interrupt - .endif - .endif std r11,IAREA+EX_R11(r13) std r12,IAREA+EX_R12(r13)
@@ -524,6 +494,37 @@ DEFINE_FIXED_SYMBOL(\name\()_common_virt) .endm .macro __GEN_COMMON_BODY name + .if IMASK + lbz r10,PACAIRQSOFTMASK(r13) + andi. r10,r10,IMASK + /* Associate vector numbers with bits in paca->irq_happened */ + .if IVEC == 0x500 || IVEC == 0xea0 + li r10,PACA_IRQ_EE + .elseif IVEC == 0x900 + li r10,PACA_IRQ_DEC + .elseif IVEC == 0xa00 || IVEC == 0xe80 + li r10,PACA_IRQ_DBELL + .elseif IVEC == 0xe60 + li r10,PACA_IRQ_HMI + .elseif IVEC == 0xf00 + li r10,PACA_IRQ_PMI + .else + .abort "Bad maskable vector" + .endif + + .if IHSRR == EXC_HV_OR_STD + BEGIN_FTR_SECTION + bne masked_Hinterrupt + FTR_SECTION_ELSE + bne masked_interrupt + ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) + .elseif IHSRR + bne masked_Hinterrupt + .else + bne masked_interrupt + .endif + .endif + .if ISTACK andi. r10,r12,MSR_PR /* See if coming from user */ mr r10,r1 /* Save r1 */
@@ -2330,18 +2331,10 @@ EXC_VIRT_NONE(0x5800, 0x100) #ifdef CONFIG_PPC_WATCHDOG -#define MASKED_DEC_HANDLER_LABEL 3f - -#define MASKED_DEC_HANDLER(_H) \ -3: /* soft-nmi */ \ - std r12,PACA_EXGEN+EX_R12(r13); \ - GET_SCRATCH0(r10); \ - std r10,PACA_EXGEN+EX_R13(r13); \ - mfspr r11,SPRN_SRR0; /* save SRR0 */ \ - mfspr r12,SPRN_SRR1; /* and SRR1 */ \ - LOAD_HANDLER(r10, soft_nmi_common); \ - mtctr r10; \ - bctr +INT_DEFINE_BEGIN(soft_nmi) + IVEC=0x900 + ISTACK=0 +INT_DEFINE_END(soft_nmi) /* * Branch to soft_nmi_interrupt using the emergency stack. The emergency
@@ -2353,19 +2346,16 @@ EXC_VIRT_NONE(0x5800, 0x100) * and run it entirely with interrupts hard disabled. */ EXC_COMMON_BEGIN(soft_nmi_common) + mfspr r11,SPRN_SRR0 mr r10,r1 ld r1,PACAEMERGSP(r13) subi r1,r1,INT_FRAME_SIZE - __ISTACK(decrementer)=0 - __GEN_COMMON_BODY decrementer + __GEN_COMMON_BODY soft_nmi bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl soft_nmi_interrupt b ret_from_except -#else /* CONFIG_PPC_WATCHDOG */ -#define MASKED_DEC_HANDLER_LABEL 2f /* normal return */ -#define MASKED_DEC_HANDLER(_H) #endif /* CONFIG_PPC_WATCHDOG */ /*
@@ -2384,7 +2374,6 @@ masked_Hinterrupt: .else masked_interrupt: .endif - std r11,PACA_EXGEN+EX_R11(r13) lbz r11,PACAIRQHAPPENED(r13) or r11,r11,r10 stb r11,PACAIRQHAPPENED(r13)
@@ -2393,26 +2382,30 @@ masked_interrupt: lis r10,0x7fff ori r10,r10,0xffff mtspr SPRN_DEC,r10 - b MASKED_DEC_HANDLER_LABEL +#ifdef CONFIG_PPC_WATCHDOG + b soft_nmi_common +#else + b 2f +#endif 1: andi. r10,r10,PACA_IRQ_MUST_HARD_MASK beq 2f + xori r12,r12,MSR_EE /* clear MSR_EE */ .if \hsrr - mfspr r10,SPRN_HSRR1 - xori r10,r10,MSR_EE /* clear MSR_EE */ - mtspr SPRN_HSRR1,r10 + mtspr SPRN_HSRR1,r12 .else - mfspr r10,SPRN_SRR1 - xori r10,r10,MSR_EE /* clear MSR_EE */ - mtspr SPRN_SRR1,r10 + mtspr SPRN_SRR1,r12 .endif ori r11,r11,PACA_IRQ_HARD_DIS stb r11,PACAIRQHAPPENED(r13) 2: /* done */ + ld r10,PACA_EXGEN+EX_CTR(r13) + mtctr r10 mtcrf 0x80,r9 std r1,PACAR1(r13) ld r9,PACA_EXGEN+EX_R9(r13) ld r10,PACA_EXGEN+EX_R10(r13) ld r11,PACA_EXGEN+EX_R11(r13) + ld r12,PACA_EXGEN+EX_R12(r13) /* returns to kernel where r13 must be set up, so don't restore it */ .if \hsrr HRFI_TO_KERNEL
@@ -2420,7 +2413,6 @@ masked_interrupt: RFI_TO_KERNEL .endif b . - MASKED_DEC_HANDLER(\hsrr\()) .endm TRAMP_REAL_BEGIN(stf_barrier_fallback)
@@ -2527,7 +2519,7 @@ TRAMP_REAL_BEGIN(hrfi_flush_fallback) * instruction code patches (which end up in the common .text area) * cannot reach these if they are put there. */ -USE_FIXED_SECTION(virt_trampolines) +USE_TEXT_SECTION() MASKED_INTERRUPT EXC_STD MASKED_INTERRUPT EXC_HV
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 12/32 powerpc/64s/exception: move KVM test to common code
In reply to Nicholas PigginThis allows more code to be moved out of unrelocated regions. The system call KVMTEST is changed to be open-coded and remain in the tramp area to avoid having to move it to entry_64.S. The custom nature of the system call entry code means the hcall case can be made more streamlined than regular interrupt handlers.
--- arch/powerpc/kernel/exceptions-64s.S | 239 ++++++++++++------------ arch/powerpc/kvm/book3s_hv_rmhandlers.S | 11 -- arch/powerpc/kvm/book3s_segment.S | 7 - 3 files changed, 119 insertions(+), 138 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +119 -120
@@ -44,7 +44,6 @@ * EXC_VIRT_BEGIN/END - virt (AIL), unrelocated exception vectors * TRAMP_REAL_BEGIN - real, unrelocated helpers (virt may call these) * TRAMP_VIRT_BEGIN - virt, unreloc helpers (in practice, real can use) - * TRAMP_KVM_BEGIN - KVM handlers, these are put into real, unrelocated * EXC_COMMON - After switching to virtual, relocated mode. */
@@ -74,13 +73,6 @@ name: #define TRAMP_VIRT_BEGIN(name) \ FIXED_SECTION_ENTRY_BEGIN(virt_trampolines, name) -#ifdef CONFIG_KVM_BOOK3S_64_HANDLER -#define TRAMP_KVM_BEGIN(name) \ - TRAMP_VIRT_BEGIN(name) -#else -#define TRAMP_KVM_BEGIN(name) -#endif - #define EXC_REAL_NONE(start, size) \ FIXED_SECTION_ENTRY_BEGIN_LOCATION(real_vectors, exc_real_##start##_##unused, start, size); \ FIXED_SECTION_ENTRY_END_LOCATION(real_vectors, exc_real_##start##_##unused, start, size)
@@ -271,6 +263,9 @@ do_define_int n .endm .macro GEN_KVM name + .balign IFETCH_ALIGN_BYTES +\name\()_kvm: + .if IKVM_SKIP cmpwi r10,KVM_GUEST_MODE_SKIP beq 89f
@@ -281,13 +276,18 @@ BEGIN_FTR_SECTION_NESTED(947) END_FTR_SECTION_NESTED(CPU_FTR_CFAR,CPU_FTR_CFAR,947) .endif + ld r10,PACA_EXGEN+EX_CTR(r13) + mtctr r10 BEGIN_FTR_SECTION_NESTED(948) ld r10,IAREA+EX_PPR(r13) std r10,HSTATE_PPR(r13) END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) - ld r10,IAREA+EX_R10(r13) + ld r11,IAREA+EX_R11(r13) + ld r12,IAREA+EX_R12(r13) std r12,HSTATE_SCRATCH0(r13) sldi r12,r9,32 + ld r9,IAREA+EX_R9(r13) + ld r10,IAREA+EX_R10(r13) /* HSRR variants have the 0x2 bit added to their trap number */ .if IHSRR == EXC_HV_OR_STD BEGIN_FTR_SECTION
@@ -300,29 +300,16 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) .else ori r12,r12,(IVEC) .endif - -#ifdef CONFIG_RELOCATABLE - /* - * KVM requires __LOAD_FAR_HANDLER beause kvmppc_interrupt lives - * outside the head section. CONFIG_RELOCATABLE KVM expects CTR - * to be saved in HSTATE_SCRATCH1. - */ - ld r9,IAREA+EX_CTR(r13) - std r9,HSTATE_SCRATCH1(r13) - __LOAD_FAR_HANDLER(r9, kvmppc_interrupt) - mtctr r9 - ld r9,IAREA+EX_R9(r13) - bctr -#else - ld r9,IAREA+EX_R9(r13) b kvmppc_interrupt -#endif - .if IKVM_SKIP 89: mtocrf 0x80,r9 + ld r10,PACA_EXGEN+EX_CTR(r13) + mtctr r10 ld r9,IAREA+EX_R9(r13) ld r10,IAREA+EX_R10(r13) + ld r11,IAREA+EX_R11(r13) + ld r12,IAREA+EX_R12(r13) .if IHSRR == EXC_HV_OR_STD BEGIN_FTR_SECTION b kvmppc_skip_Hinterrupt
@@ -407,11 +394,6 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) mfctr r10 std r10,IAREA+EX_CTR(r13) mfcr r9 - - .if (!\virt && IKVM_REAL) || (\virt && IKVM_VIRT) - KVMTEST \name IHSRR IVEC - .endif - std r11,IAREA+EX_R11(r13) std r12,IAREA+EX_R12(r13)
@@ -475,6 +457,10 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) .macro __GEN_COMMON_ENTRY name DEFINE_FIXED_SYMBOL(\name\()_common_real) \name\()_common_real: + .if IKVM_REAL + KVMTEST \name IHSRR IVEC + .endif + ld r10,PACAKMSR(r13) /* get MSR value for kernel */ /* MSR[RI] is clear iff using SRR regs */ .if IHSRR == EXC_HV_OR_STD
@@ -487,9 +473,17 @@ DEFINE_FIXED_SYMBOL(\name\()_common_real) mtmsrd r10 .if IVIRT + .if IKVM_VIRT + b 1f /* skip the virt test coming from real */ + .endif + .balign IFETCH_ALIGN_BYTES DEFINE_FIXED_SYMBOL(\name\()_common_virt) \name\()_common_virt: + .if IKVM_VIRT + KVMTEST \name IHSRR IVEC +1: + .endif .endif /* IVIRT */ .endm
@@ -848,8 +842,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) */ EXC_REAL_END(system_reset, 0x100, 0x100) EXC_VIRT_NONE(0x4100, 0x100) -TRAMP_KVM_BEGIN(system_reset_kvm) - GEN_KVM system_reset #ifdef CONFIG_PPC_P7_NAP TRAMP_REAL_BEGIN(system_reset_idle_wake)
@@ -927,6 +919,8 @@ EXC_COMMON_BEGIN(system_reset_common) EXCEPTION_RESTORE_REGS EXC_STD RFI_TO_USER_OR_KERNEL + GEN_KVM system_reset + INT_DEFINE_BEGIN(machine_check_early) IVEC=0x200
@@ -968,9 +962,6 @@ TRAMP_REAL_BEGIN(machine_check_fwnmi) GEN_INT_ENTRY machine_check_early, virt=0 #endif -TRAMP_KVM_BEGIN(machine_check_kvm) - GEN_KVM machine_check - #define MACHINE_CHECK_HANDLER_WINDUP \ /* Clear MSR_RI before setting SRR0 and SRR1. */\ li r9,0; \
@@ -1126,6 +1117,9 @@ EXC_COMMON_BEGIN(machine_check_common) bl machine_check_exception b ret_from_except + GEN_KVM machine_check + + #ifdef CONFIG_PPC_P7_NAP /* * This is an idle wakeup. Low level machine check has already been
@@ -1218,8 +1212,6 @@ EXC_REAL_END(data_access, 0x300, 0x80) EXC_VIRT_BEGIN(data_access, 0x4300, 0x80) GEN_INT_ENTRY data_access, virt=1 EXC_VIRT_END(data_access, 0x4300, 0x80) -TRAMP_KVM_BEGIN(data_access_kvm) - GEN_KVM data_access EXC_COMMON_BEGIN(data_access_common) GEN_COMMON data_access ld r4,_DAR(r1)
@@ -1232,6 +1224,8 @@ MMU_FTR_SECTION_ELSE b handle_page_fault ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) + GEN_KVM data_access + INT_DEFINE_BEGIN(data_access_slb) IVEC=0x380
@@ -1248,8 +1242,6 @@ EXC_REAL_END(data_access_slb, 0x380, 0x80) EXC_VIRT_BEGIN(data_access_slb, 0x4380, 0x80) GEN_INT_ENTRY data_access_slb, virt=1 EXC_VIRT_END(data_access_slb, 0x4380, 0x80) -TRAMP_KVM_BEGIN(data_access_slb_kvm) - GEN_KVM data_access_slb EXC_COMMON_BEGIN(data_access_slb_common) GEN_COMMON data_access_slb ld r4,_DAR(r1)
@@ -1274,6 +1266,8 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) bl do_bad_slb_fault b ret_from_except + GEN_KVM data_access_slb + INT_DEFINE_BEGIN(instruction_access) IVEC=0x400
@@ -1289,8 +1283,6 @@ EXC_REAL_END(instruction_access, 0x400, 0x80) EXC_VIRT_BEGIN(instruction_access, 0x4400, 0x80) GEN_INT_ENTRY instruction_access, virt=1 EXC_VIRT_END(instruction_access, 0x4400, 0x80) -TRAMP_KVM_BEGIN(instruction_access_kvm) - GEN_KVM instruction_access EXC_COMMON_BEGIN(instruction_access_common) GEN_COMMON instruction_access ld r4,_DAR(r1)
@@ -1303,6 +1295,8 @@ MMU_FTR_SECTION_ELSE b handle_page_fault ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) + GEN_KVM instruction_access + INT_DEFINE_BEGIN(instruction_access_slb) IVEC=0x480
@@ -1319,8 +1313,6 @@ EXC_REAL_END(instruction_access_slb, 0x480, 0x80) EXC_VIRT_BEGIN(instruction_access_slb, 0x4480, 0x80) GEN_INT_ENTRY instruction_access_slb, virt=1 EXC_VIRT_END(instruction_access_slb, 0x4480, 0x80) -TRAMP_KVM_BEGIN(instruction_access_slb_kvm) - GEN_KVM instruction_access_slb EXC_COMMON_BEGIN(instruction_access_slb_common) GEN_COMMON instruction_access_slb ld r4,_DAR(r1)
@@ -1345,6 +1337,9 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) bl do_bad_slb_fault b ret_from_except + GEN_KVM instruction_access_slb + + INT_DEFINE_BEGIN(hardware_interrupt) IVEC=0x500 IHSRR=EXC_HV_OR_STD
@@ -1359,8 +1354,6 @@ EXC_REAL_END(hardware_interrupt, 0x500, 0x100) EXC_VIRT_BEGIN(hardware_interrupt, 0x4500, 0x100) GEN_INT_ENTRY hardware_interrupt, virt=1 EXC_VIRT_END(hardware_interrupt, 0x4500, 0x100) -TRAMP_KVM_BEGIN(hardware_interrupt_kvm) - GEN_KVM hardware_interrupt EXC_COMMON_BEGIN(hardware_interrupt_common) GEN_COMMON hardware_interrupt FINISH_NAP
@@ -1369,6 +1362,8 @@ EXC_COMMON_BEGIN(hardware_interrupt_common) bl do_IRQ b ret_from_except_lite + GEN_KVM hardware_interrupt + INT_DEFINE_BEGIN(alignment) IVEC=0x600
@@ -1383,8 +1378,6 @@ EXC_REAL_END(alignment, 0x600, 0x100) EXC_VIRT_BEGIN(alignment, 0x4600, 0x100) GEN_INT_ENTRY alignment, virt=1 EXC_VIRT_END(alignment, 0x4600, 0x100) -TRAMP_KVM_BEGIN(alignment_kvm) - GEN_KVM alignment EXC_COMMON_BEGIN(alignment_common) GEN_COMMON alignment bl save_nvgprs
@@ -1392,6 +1385,8 @@ EXC_COMMON_BEGIN(alignment_common) bl alignment_exception b ret_from_except + GEN_KVM alignment + INT_DEFINE_BEGIN(program_check) IVEC=0x700
@@ -1404,8 +1399,6 @@ EXC_REAL_END(program_check, 0x700, 0x100) EXC_VIRT_BEGIN(program_check, 0x4700, 0x100) GEN_INT_ENTRY program_check, virt=1 EXC_VIRT_END(program_check, 0x4700, 0x100) -TRAMP_KVM_BEGIN(program_check_kvm) - GEN_KVM program_check EXC_COMMON_BEGIN(program_check_common) __GEN_COMMON_ENTRY program_check
@@ -1445,6 +1438,8 @@ EXC_COMMON_BEGIN(program_check_common) bl program_check_exception b ret_from_except + GEN_KVM program_check + INT_DEFINE_BEGIN(fp_unavailable) IVEC=0x800
@@ -1458,8 +1453,6 @@ EXC_REAL_END(fp_unavailable, 0x800, 0x100) EXC_VIRT_BEGIN(fp_unavailable, 0x4800, 0x100) GEN_INT_ENTRY fp_unavailable, virt=1 EXC_VIRT_END(fp_unavailable, 0x4800, 0x100) -TRAMP_KVM_BEGIN(fp_unavailable_kvm) - GEN_KVM fp_unavailable EXC_COMMON_BEGIN(fp_unavailable_common) GEN_COMMON fp_unavailable bne 1f /* if from user, just load it up */
@@ -1490,6 +1483,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_TM) b ret_from_except #endif + GEN_KVM fp_unavailable + INT_DEFINE_BEGIN(decrementer) IVEC=0x900
@@ -1503,8 +1498,6 @@ EXC_REAL_END(decrementer, 0x900, 0x80) EXC_VIRT_BEGIN(decrementer, 0x4900, 0x80) GEN_INT_ENTRY decrementer, virt=1 EXC_VIRT_END(decrementer, 0x4900, 0x80) -TRAMP_KVM_BEGIN(decrementer_kvm) - GEN_KVM decrementer EXC_COMMON_BEGIN(decrementer_common) GEN_COMMON decrementer FINISH_NAP
@@ -1513,6 +1506,8 @@ EXC_COMMON_BEGIN(decrementer_common) bl timer_interrupt b ret_from_except_lite + GEN_KVM decrementer + INT_DEFINE_BEGIN(hdecrementer) IVEC=0x980
@@ -1527,8 +1522,6 @@ EXC_REAL_END(hdecrementer, 0x980, 0x80) EXC_VIRT_BEGIN(hdecrementer, 0x4980, 0x80) GEN_INT_ENTRY hdecrementer, virt=1 EXC_VIRT_END(hdecrementer, 0x4980, 0x80) -TRAMP_KVM_BEGIN(hdecrementer_kvm) - GEN_KVM hdecrementer EXC_COMMON_BEGIN(hdecrementer_common) GEN_COMMON hdecrementer bl save_nvgprs
@@ -1536,6 +1529,8 @@ EXC_COMMON_BEGIN(hdecrementer_common) bl hdec_interrupt b ret_from_except + GEN_KVM hdecrementer + INT_DEFINE_BEGIN(doorbell_super) IVEC=0xa00
@@ -1549,8 +1544,6 @@ EXC_REAL_END(doorbell_super, 0xa00, 0x100) EXC_VIRT_BEGIN(doorbell_super, 0x4a00, 0x100) GEN_INT_ENTRY doorbell_super, virt=1 EXC_VIRT_END(doorbell_super, 0x4a00, 0x100) -TRAMP_KVM_BEGIN(doorbell_super_kvm) - GEN_KVM doorbell_super EXC_COMMON_BEGIN(doorbell_super_common) GEN_COMMON doorbell_super FINISH_NAP
@@ -1563,6 +1556,8 @@ EXC_COMMON_BEGIN(doorbell_super_common) #endif b ret_from_except_lite + GEN_KVM doorbell_super + EXC_REAL_NONE(0xb00, 0x100) EXC_VIRT_NONE(0x4b00, 0x100)
@@ -1667,6 +1662,7 @@ EXC_VIRT_BEGIN(system_call, 0x4c00, 0x100) EXC_VIRT_END(system_call, 0x4c00, 0x100) #ifdef CONFIG_KVM_BOOK3S_64_HANDLER +TRAMP_REAL_BEGIN(system_call_kvm) /* * This is a hcall, so register convention is as above, with these * differences:
@@ -1674,20 +1670,35 @@ EXC_VIRT_END(system_call, 0x4c00, 0x100) * ctr = orig r13 * orig r10 saved in PACA */ -TRAMP_KVM_BEGIN(system_call_kvm) /* * Save the PPR (on systems that support it) before changing to * HMT_MEDIUM. That allows the KVM code to save that value into the * guest state (it is the guest's PPR value). */ - OPT_GET_SPR(r10, SPRN_PPR, CPU_FTR_HAS_PPR) +BEGIN_FTR_SECTION_NESTED(948) + mfspr r10,SPRN_PPR + std r10,HSTATE_PPR(r13) +END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) HMT_MEDIUM - OPT_SAVE_REG_TO_PACA(PACA_EXGEN+EX_PPR, r10, CPU_FTR_HAS_PPR) mfctr r10 SET_SCRATCH0(r10) - std r9,PACA_EXGEN+EX_R9(r13) - mfcr r9 - GEN_KVM system_call + mfcr r10 + std r12,HSTATE_SCRATCH0(r13) + sldi r12,r10,32 + ori r12,r12,0xc00 +#ifdef CONFIG_RELOCATABLE + /* + * Requires __LOAD_FAR_HANDLER beause kvmppc_interrupt lives + * outside the head section. + */ + __LOAD_FAR_HANDLER(r10, kvmppc_interrupt) + mtctr r10 + ld r10,PACA_EXGEN+EX_R10(r13) + bctr +#else + ld r10,PACA_EXGEN+EX_R10(r13) + b kvmppc_interrupt +#endif #endif
@@ -1702,8 +1713,6 @@ EXC_REAL_END(single_step, 0xd00, 0x100) EXC_VIRT_BEGIN(single_step, 0x4d00, 0x100) GEN_INT_ENTRY single_step, virt=1 EXC_VIRT_END(single_step, 0x4d00, 0x100) -TRAMP_KVM_BEGIN(single_step_kvm) - GEN_KVM single_step EXC_COMMON_BEGIN(single_step_common) GEN_COMMON single_step bl save_nvgprs
@@ -1711,6 +1720,8 @@ EXC_COMMON_BEGIN(single_step_common) bl single_step_exception b ret_from_except + GEN_KVM single_step + INT_DEFINE_BEGIN(h_data_storage) IVEC=0xe00
@@ -1728,8 +1739,6 @@ EXC_REAL_END(h_data_storage, 0xe00, 0x20) EXC_VIRT_BEGIN(h_data_storage, 0x4e00, 0x20) GEN_INT_ENTRY h_data_storage, virt=1, ool=1 EXC_VIRT_END(h_data_storage, 0x4e00, 0x20) -TRAMP_KVM_BEGIN(h_data_storage_kvm) - GEN_KVM h_data_storage EXC_COMMON_BEGIN(h_data_storage_common) GEN_COMMON h_data_storage bl save_nvgprs
@@ -1743,6 +1752,8 @@ MMU_FTR_SECTION_ELSE ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_TYPE_RADIX) b ret_from_except + GEN_KVM h_data_storage + INT_DEFINE_BEGIN(h_instr_storage) IVEC=0xe20
@@ -1757,8 +1768,6 @@ EXC_REAL_END(h_instr_storage, 0xe20, 0x20) EXC_VIRT_BEGIN(h_instr_storage, 0x4e20, 0x20) GEN_INT_ENTRY h_instr_storage, virt=1, ool=1 EXC_VIRT_END(h_instr_storage, 0x4e20, 0x20) -TRAMP_KVM_BEGIN(h_instr_storage_kvm) - GEN_KVM h_instr_storage EXC_COMMON_BEGIN(h_instr_storage_common) GEN_COMMON h_instr_storage bl save_nvgprs
@@ -1766,6 +1775,8 @@ EXC_COMMON_BEGIN(h_instr_storage_common) bl unknown_exception b ret_from_except + GEN_KVM h_instr_storage + INT_DEFINE_BEGIN(emulation_assist) IVEC=0xe40
@@ -1780,8 +1791,6 @@ EXC_REAL_END(emulation_assist, 0xe40, 0x20) EXC_VIRT_BEGIN(emulation_assist, 0x4e40, 0x20) GEN_INT_ENTRY emulation_assist, virt=1, ool=1 EXC_VIRT_END(emulation_assist, 0x4e40, 0x20) -TRAMP_KVM_BEGIN(emulation_assist_kvm) - GEN_KVM emulation_assist EXC_COMMON_BEGIN(emulation_assist_common) GEN_COMMON emulation_assist bl save_nvgprs
@@ -1789,6 +1798,8 @@ EXC_COMMON_BEGIN(emulation_assist_common) bl emulation_assist_interrupt b ret_from_except + GEN_KVM emulation_assist + /* * hmi_exception trampoline is a special case. It jumps to hmi_exception_early
@@ -1816,10 +1827,6 @@ EXC_REAL_BEGIN(hmi_exception, 0xe60, 0x20) GEN_INT_ENTRY hmi_exception_early, virt=0, ool=1 EXC_REAL_END(hmi_exception, 0xe60, 0x20) EXC_VIRT_NONE(0x4e60, 0x20) -TRAMP_KVM_BEGIN(hmi_exception_early_kvm) - GEN_KVM hmi_exception_early -TRAMP_KVM_BEGIN(hmi_exception_kvm) - GEN_KVM hmi_exception EXC_COMMON_BEGIN(hmi_exception_early_common) mfspr r11,SPRN_HSRR0 /* Save HSRR0 */
@@ -1846,6 +1853,8 @@ EXC_COMMON_BEGIN(hmi_exception_early_common) EXCEPTION_RESTORE_REGS EXC_HV GEN_INT_ENTRY hmi_exception, virt=0 + GEN_KVM hmi_exception_early + EXC_COMMON_BEGIN(hmi_exception_common) GEN_COMMON hmi_exception FINISH_NAP
@@ -1855,6 +1864,8 @@ EXC_COMMON_BEGIN(hmi_exception_common) bl handle_hmi_exception b ret_from_except + GEN_KVM hmi_exception + INT_DEFINE_BEGIN(h_doorbell) IVEC=0xe80
@@ -1870,8 +1881,6 @@ EXC_REAL_END(h_doorbell, 0xe80, 0x20) EXC_VIRT_BEGIN(h_doorbell, 0x4e80, 0x20) GEN_INT_ENTRY h_doorbell, virt=1, ool=1 EXC_VIRT_END(h_doorbell, 0x4e80, 0x20) -TRAMP_KVM_BEGIN(h_doorbell_kvm) - GEN_KVM h_doorbell EXC_COMMON_BEGIN(h_doorbell_common) GEN_COMMON h_doorbell FINISH_NAP
@@ -1884,6 +1893,8 @@ EXC_COMMON_BEGIN(h_doorbell_common) #endif b ret_from_except_lite + GEN_KVM h_doorbell + INT_DEFINE_BEGIN(h_virt_irq) IVEC=0xea0
@@ -1899,8 +1910,6 @@ EXC_REAL_END(h_virt_irq, 0xea0, 0x20) EXC_VIRT_BEGIN(h_virt_irq, 0x4ea0, 0x20) GEN_INT_ENTRY h_virt_irq, virt=1, ool=1 EXC_VIRT_END(h_virt_irq, 0x4ea0, 0x20) -TRAMP_KVM_BEGIN(h_virt_irq_kvm) - GEN_KVM h_virt_irq EXC_COMMON_BEGIN(h_virt_irq_common) GEN_COMMON h_virt_irq FINISH_NAP
@@ -1909,6 +1918,8 @@ EXC_COMMON_BEGIN(h_virt_irq_common) bl do_IRQ b ret_from_except_lite + GEN_KVM h_virt_irq + EXC_REAL_NONE(0xec0, 0x20) EXC_VIRT_NONE(0x4ec0, 0x20)
@@ -1928,8 +1939,6 @@ EXC_REAL_END(performance_monitor, 0xf00, 0x20) EXC_VIRT_BEGIN(performance_monitor, 0x4f00, 0x20) GEN_INT_ENTRY performance_monitor, virt=1, ool=1 EXC_VIRT_END(performance_monitor, 0x4f00, 0x20) -TRAMP_KVM_BEGIN(performance_monitor_kvm) - GEN_KVM performance_monitor EXC_COMMON_BEGIN(performance_monitor_common) GEN_COMMON performance_monitor FINISH_NAP
@@ -1938,6 +1947,8 @@ EXC_COMMON_BEGIN(performance_monitor_common) bl performance_monitor_exception b ret_from_except_lite + GEN_KVM performance_monitor + INT_DEFINE_BEGIN(altivec_unavailable) IVEC=0xf20
@@ -1951,8 +1962,6 @@ EXC_REAL_END(altivec_unavailable, 0xf20, 0x20) EXC_VIRT_BEGIN(altivec_unavailable, 0x4f20, 0x20) GEN_INT_ENTRY altivec_unavailable, virt=1, ool=1 EXC_VIRT_END(altivec_unavailable, 0x4f20, 0x20) -TRAMP_KVM_BEGIN(altivec_unavailable_kvm) - GEN_KVM altivec_unavailable EXC_COMMON_BEGIN(altivec_unavailable_common) GEN_COMMON altivec_unavailable #ifdef CONFIG_ALTIVEC
@@ -1986,6 +1995,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) bl altivec_unavailable_exception b ret_from_except + GEN_KVM altivec_unavailable + INT_DEFINE_BEGIN(vsx_unavailable) IVEC=0xf40
@@ -1999,8 +2010,6 @@ EXC_REAL_END(vsx_unavailable, 0xf40, 0x20) EXC_VIRT_BEGIN(vsx_unavailable, 0x4f40, 0x20) GEN_INT_ENTRY vsx_unavailable, virt=1, ool=1 EXC_VIRT_END(vsx_unavailable, 0x4f40, 0x20) -TRAMP_KVM_BEGIN(vsx_unavailable_kvm) - GEN_KVM vsx_unavailable EXC_COMMON_BEGIN(vsx_unavailable_common) GEN_COMMON vsx_unavailable #ifdef CONFIG_VSX
@@ -2033,6 +2042,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX) bl vsx_unavailable_exception b ret_from_except + GEN_KVM vsx_unavailable + INT_DEFINE_BEGIN(facility_unavailable) IVEC=0xf60
@@ -2045,8 +2056,6 @@ EXC_REAL_END(facility_unavailable, 0xf60, 0x20) EXC_VIRT_BEGIN(facility_unavailable, 0x4f60, 0x20) GEN_INT_ENTRY facility_unavailable, virt=1, ool=1 EXC_VIRT_END(facility_unavailable, 0x4f60, 0x20) -TRAMP_KVM_BEGIN(facility_unavailable_kvm) - GEN_KVM facility_unavailable EXC_COMMON_BEGIN(facility_unavailable_common) GEN_COMMON facility_unavailable bl save_nvgprs
@@ -2054,6 +2063,8 @@ EXC_COMMON_BEGIN(facility_unavailable_common) bl facility_unavailable_exception b ret_from_except + GEN_KVM facility_unavailable + INT_DEFINE_BEGIN(h_facility_unavailable) IVEC=0xf80
@@ -2068,8 +2079,6 @@ EXC_REAL_END(h_facility_unavailable, 0xf80, 0x20) EXC_VIRT_BEGIN(h_facility_unavailable, 0x4f80, 0x20) GEN_INT_ENTRY h_facility_unavailable, virt=1, ool=1 EXC_VIRT_END(h_facility_unavailable, 0x4f80, 0x20) -TRAMP_KVM_BEGIN(h_facility_unavailable_kvm) - GEN_KVM h_facility_unavailable EXC_COMMON_BEGIN(h_facility_unavailable_common) GEN_COMMON h_facility_unavailable bl save_nvgprs
@@ -2077,6 +2086,8 @@ EXC_COMMON_BEGIN(h_facility_unavailable_common) bl facility_unavailable_exception b ret_from_except + GEN_KVM h_facility_unavailable + EXC_REAL_NONE(0xfa0, 0x20) EXC_VIRT_NONE(0x4fa0, 0x20)
@@ -2102,14 +2113,15 @@ EXC_REAL_BEGIN(cbe_system_error, 0x1200, 0x100) GEN_INT_ENTRY cbe_system_error, virt=0 EXC_REAL_END(cbe_system_error, 0x1200, 0x100) EXC_VIRT_NONE(0x5200, 0x100) -TRAMP_KVM_BEGIN(cbe_system_error_kvm) - GEN_KVM cbe_system_error EXC_COMMON_BEGIN(cbe_system_error_common) GEN_COMMON cbe_system_error bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl cbe_system_error_exception b ret_from_except + + GEN_KVM cbe_system_error + #else /* CONFIG_CBE_RAS */ EXC_REAL_NONE(0x1200, 0x100) EXC_VIRT_NONE(0x5200, 0x100)
@@ -2128,8 +2140,6 @@ EXC_REAL_END(instruction_breakpoint, 0x1300, 0x100) EXC_VIRT_BEGIN(instruction_breakpoint, 0x5300, 0x100) GEN_INT_ENTRY instruction_breakpoint, virt=1 EXC_VIRT_END(instruction_breakpoint, 0x5300, 0x100) -TRAMP_KVM_BEGIN(instruction_breakpoint_kvm) - GEN_KVM instruction_breakpoint EXC_COMMON_BEGIN(instruction_breakpoint_common) GEN_COMMON instruction_breakpoint bl save_nvgprs
@@ -2137,6 +2147,8 @@ EXC_COMMON_BEGIN(instruction_breakpoint_common) bl instruction_breakpoint_exception b ret_from_except + GEN_KVM instruction_breakpoint + EXC_REAL_NONE(0x1400, 0x100) EXC_VIRT_NONE(0x5400, 0x100)
@@ -2145,6 +2157,7 @@ INT_DEFINE_BEGIN(denorm_exception) IVEC=0x1500 IHSRR=EXC_HV IEARLY=2 + IKVM_REAL=1 INT_DEFINE_END(denorm_exception) EXC_REAL_BEGIN(denorm_exception, 0x1500, 0x100)
@@ -2154,7 +2167,6 @@ EXC_REAL_BEGIN(denorm_exception, 0x1500, 0x100) andis. r10,r10,(HSRR1_DENORM)@h /* denorm? */ bne+ denorm_assist #endif - KVMTEST denorm_exception, EXC_HV, 0x1500 mfspr r11,SPRN_HSRR0 mfspr r12,SPRN_HSRR1 GEN_BRANCH_TO_COMMON denorm_exception, virt=0
@@ -2172,8 +2184,6 @@ EXC_VIRT_END(denorm_exception, 0x5500, 0x100) #else EXC_VIRT_NONE(0x5500, 0x100) #endif -TRAMP_KVM_BEGIN(denorm_exception_kvm) - GEN_KVM denorm_exception #ifdef CONFIG_PPC_DENORMALISATION TRAMP_REAL_BEGIN(denorm_assist)
@@ -2251,6 +2261,8 @@ EXC_COMMON_BEGIN(denorm_exception_common) bl unknown_exception b ret_from_except + GEN_KVM denorm_exception + #ifdef CONFIG_CBE_RAS INT_DEFINE_BEGIN(cbe_maintenance)
@@ -2264,14 +2276,15 @@ EXC_REAL_BEGIN(cbe_maintenance, 0x1600, 0x100) GEN_INT_ENTRY cbe_maintenance, virt=0 EXC_REAL_END(cbe_maintenance, 0x1600, 0x100) EXC_VIRT_NONE(0x5600, 0x100) -TRAMP_KVM_BEGIN(cbe_maintenance_kvm) - GEN_KVM cbe_maintenance EXC_COMMON_BEGIN(cbe_maintenance_common) GEN_COMMON cbe_maintenance bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl cbe_maintenance_exception b ret_from_except + + GEN_KVM cbe_maintenance + #else /* CONFIG_CBE_RAS */ EXC_REAL_NONE(0x1600, 0x100) EXC_VIRT_NONE(0x5600, 0x100)
@@ -2289,8 +2302,6 @@ EXC_REAL_END(altivec_assist, 0x1700, 0x100) EXC_VIRT_BEGIN(altivec_assist, 0x5700, 0x100) GEN_INT_ENTRY altivec_assist, virt=1 EXC_VIRT_END(altivec_assist, 0x5700, 0x100) -TRAMP_KVM_BEGIN(altivec_assist_kvm) - GEN_KVM altivec_assist EXC_COMMON_BEGIN(altivec_assist_common) GEN_COMMON altivec_assist bl save_nvgprs
@@ -2302,6 +2313,8 @@ EXC_COMMON_BEGIN(altivec_assist_common) #endif b ret_from_except + GEN_KVM altivec_assist + #ifdef CONFIG_CBE_RAS INT_DEFINE_BEGIN(cbe_thermal)
@@ -2315,14 +2328,15 @@ EXC_REAL_BEGIN(cbe_thermal, 0x1800, 0x100) GEN_INT_ENTRY cbe_thermal, virt=0 EXC_REAL_END(cbe_thermal, 0x1800, 0x100) EXC_VIRT_NONE(0x5800, 0x100) -TRAMP_KVM_BEGIN(cbe_thermal_kvm) - GEN_KVM cbe_thermal EXC_COMMON_BEGIN(cbe_thermal_common) GEN_COMMON cbe_thermal bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl cbe_thermal_exception b ret_from_except + + GEN_KVM cbe_thermal + #else /* CONFIG_CBE_RAS */ EXC_REAL_NONE(0x1800, 0x100) EXC_VIRT_NONE(0x5800, 0x100)
@@ -2514,17 +2528,12 @@ TRAMP_REAL_BEGIN(hrfi_flush_fallback) GET_SCRATCH0(r13); hrfid -/* - * Real mode exceptions actually use this too, but alternate - * instruction code patches (which end up in the common .text area) - * cannot reach these if they are put there. - */ USE_TEXT_SECTION() MASKED_INTERRUPT EXC_STD MASKED_INTERRUPT EXC_HV #ifdef CONFIG_KVM_BOOK3S_64_HANDLER -TRAMP_REAL_BEGIN(kvmppc_skip_interrupt) +kvmppc_skip_interrupt: /* * Here all GPRs are unchanged from when the interrupt happened * except for r13, which is saved in SPRG_SCRATCH0.
@@ -2536,7 +2545,7 @@ TRAMP_REAL_BEGIN(kvmppc_skip_interrupt) RFI_TO_KERNEL b . -TRAMP_REAL_BEGIN(kvmppc_skip_Hinterrupt) +kvmppc_skip_Hinterrupt: /* * Here all GPRs are unchanged from when the interrupt happened * except for r13, which is saved in SPRG_SCRATCH0.
@@ -2549,16 +2558,6 @@ TRAMP_REAL_BEGIN(kvmppc_skip_Hinterrupt) b . #endif -/* - * Ensure that any handlers that get invoked from the exception prologs - * above are below the first 64KB (0x10000) of the kernel image because - * the prologs assemble the addresses of these handlers using the - * LOAD_HANDLER macro, which uses an ori instruction. - */ - -/*** Common interrupt handlers ***/ - - /* * Relocation-on interrupts: A subset of the interrupts can be delivered * with IR=1/DR=1, if AIL==2 and MSR.HV won't be changed by delivering
arch/powerpc/kvm/book3s_hv_rmhandlers.S +0 -11
@@ -1266,7 +1266,6 @@ kvmppc_interrupt_hv: * R12 = (guest CR << 32) | interrupt vector * R13 = PACA * guest R12 saved in shadow VCPU SCRATCH0 - * guest CTR saved in shadow VCPU SCRATCH1 if RELOCATABLE * guest R13 saved in SPRN_SCRATCH0 */ std r9, HSTATE_SCRATCH2(r13)
@@ -1367,12 +1366,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) 11: stw r3,VCPU_HEIR(r9) /* these are volatile across C function calls */ -#ifdef CONFIG_RELOCATABLE - ld r3, HSTATE_SCRATCH1(r13) - mtctr r3 -#else mfctr r3 -#endif mfxer r4 std r3, VCPU_CTR(r9) std r4, VCPU_XER(r9)
@@ -3258,7 +3252,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_P9_TM_HV_ASSIST) * r12 is (CR << 32) | vector * r13 points to our PACA * r12 is saved in HSTATE_SCRATCH0(r13) - * ctr is saved in HSTATE_SCRATCH1(r13) if RELOCATABLE * r9 is saved in HSTATE_SCRATCH2(r13) * r13 is saved in HSPRG1 * cfar is saved in HSTATE_CFAR(r13)
@@ -3307,11 +3300,7 @@ kvmppc_bad_host_intr: ld r5, HSTATE_CFAR(r13) std r5, ORIG_GPR3(r1) mflr r3 -#ifdef CONFIG_RELOCATABLE - ld r4, HSTATE_SCRATCH1(r13) -#else mfctr r4 -#endif mfxer r5 lbz r6, PACAIRQSOFTMASK(r13) std r3, _LINK(r1)
arch/powerpc/kvm/book3s_segment.S +0 -7
@@ -167,16 +167,9 @@ kvmppc_interrupt_pr: * R12 = (guest CR << 32) | exit handler id * R13 = PACA * HSTATE.SCRATCH0 = guest R12 - * HSTATE.SCRATCH1 = guest CTR if RELOCATABLE */ #ifdef CONFIG_PPC64 /* Match 32-bit entry */ -#ifdef CONFIG_RELOCATABLE - std r9, HSTATE_SCRATCH2(r13) - ld r9, HSTATE_SCRATCH1(r13) - mtctr r9 - ld r9, HSTATE_SCRATCH2(r13) -#endif rotldi r12, r12, 32 /* Flip R12 halves for stw */ stw r12, HSTATE_SCRATCH1(r13) /* CR is now in the low half */ srdi r12, r12, 32 /* shift trap into low half */
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 13/32 powerpc/64s/exception: remove confusing IEARLY option
In reply to Nicholas PigginReplace IEARLY=1 and IEARLY=2 with IBRANCH_COMMON, which controls if the entry code branches to a common handler; and IREALMODE_COMMON, which controls whether the common handler should remain in real mode.
These special cases no longer avoid loading the SRR registers, there is no point as most of them load the registers immediately anyway.
--- arch/powerpc/kernel/exceptions-64s.S | 48 ++++++++++++++-------------- 1 file changed, 24 insertions(+), 24 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +24 -24
@@ -174,7 +174,8 @@ END_FTR_SECTION_NESTED(ftr,ftr,943) #define IDAR .L_IDAR_\name\() #define IDSISR .L_IDSISR_\name\() #define ISET_RI .L_ISET_RI_\name\() -#define IEARLY .L_IEARLY_\name\() +#define IBRANCH_TO_COMMON .L_IBRANCH_TO_COMMON_\name\() +#define IREALMODE_COMMON .L_IREALMODE_COMMON_\name\() #define IMASK .L_IMASK_\name\() #define IKVM_SKIP .L_IKVM_SKIP_\name\() #define IKVM_REAL .L_IKVM_REAL_\name\()
@@ -218,8 +219,15 @@ do_define_int n .ifndef ISET_RI ISET_RI=1 .endif - .ifndef IEARLY - IEARLY=0 + .ifndef IBRANCH_TO_COMMON + IBRANCH_TO_COMMON=1 + .endif + .ifndef IREALMODE_COMMON + IREALMODE_COMMON=0 + .else + .if ! IBRANCH_TO_COMMON + .error "IREALMODE_COMMON=1 but IBRANCH_TO_COMMON=0" + .endif .endif .ifndef IMASK IMASK=0
@@ -353,6 +361,11 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) */ .macro GEN_BRANCH_TO_COMMON name, virt + .if IREALMODE_COMMON + LOAD_HANDLER(r10, \name\()_common) + mtctr r10 + bctr + .else .if \virt #ifndef CONFIG_RELOCATABLE b \name\()_common_virt
@@ -366,6 +379,7 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) mtctr r10 bctr .endif + .endif .endm .macro GEN_INT_ENTRY name, virt, ool=0
@@ -421,11 +435,6 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) stw r10,IAREA+EX_DSISR(r13) .endif - .if IEARLY == 2 - /* nothing more */ - .elseif IEARLY - BRANCH_TO_C000(r11, \name\()_common) - .else .if IHSRR == EXC_HV_OR_STD BEGIN_FTR_SECTION mfspr r11,SPRN_HSRR0 /* save HSRR0 */
@@ -441,6 +450,8 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) mfspr r11,SPRN_SRR0 /* save SRR0 */ mfspr r12,SPRN_SRR1 /* and SRR1 */ .endif + + .if IBRANCH_TO_COMMON GEN_BRANCH_TO_COMMON \name \virt .endif
@@ -926,6 +937,7 @@ INT_DEFINE_BEGIN(machine_check_early) IVEC=0x200 IAREA=PACA_EXMC IVIRT=0 /* no virt entry point */ + IREALMODE_COMMON=1 /* * MSR_RI is not enabled, because PACA_EXMC is being used, so a * nested machine check corrupts it. machine_check_common enables
@@ -933,7 +945,6 @@ INT_DEFINE_BEGIN(machine_check_early) */ ISET_RI=0 ISTACK=0 - IEARLY=1 IDAR=1 IDSISR=1 IRECONCILE=0
@@ -973,9 +984,6 @@ TRAMP_REAL_BEGIN(machine_check_fwnmi) EXCEPTION_RESTORE_REGS EXC_STD EXC_COMMON_BEGIN(machine_check_early_common) - mfspr r11,SPRN_SRR0 - mfspr r12,SPRN_SRR1 - /* * Switch to mc_emergency stack and handle re-entrancy (we limit * the nested MCE upto level 4 to avoid stack overflow).
@@ -1809,7 +1817,7 @@ EXC_COMMON_BEGIN(emulation_assist_common) INT_DEFINE_BEGIN(hmi_exception_early) IVEC=0xe60 IHSRR=EXC_HV - IEARLY=1 + IREALMODE_COMMON=1 ISTACK=0 IRECONCILE=0 IKUAP=0 /* We don't touch AMR here, we never go to virtual mode */
@@ -1829,8 +1837,6 @@ EXC_REAL_END(hmi_exception, 0xe60, 0x20) EXC_VIRT_NONE(0x4e60, 0x20) EXC_COMMON_BEGIN(hmi_exception_early_common) - mfspr r11,SPRN_HSRR0 /* Save HSRR0 */ - mfspr r12,SPRN_HSRR1 /* Save HSRR1 */ mr r10,r1 /* Save r1 */ ld r1,PACAEMERGSP(r13) /* Use emergency stack for realmode */ subi r1,r1,INT_FRAME_SIZE /* alloc stack frame */
@@ -2156,29 +2162,23 @@ EXC_VIRT_NONE(0x5400, 0x100) INT_DEFINE_BEGIN(denorm_exception) IVEC=0x1500 IHSRR=EXC_HV - IEARLY=2 + IBRANCH_TO_COMMON=0 IKVM_REAL=1 INT_DEFINE_END(denorm_exception) EXC_REAL_BEGIN(denorm_exception, 0x1500, 0x100) GEN_INT_ENTRY denorm_exception, virt=0 #ifdef CONFIG_PPC_DENORMALISATION - mfspr r10,SPRN_HSRR1 - andis. r10,r10,(HSRR1_DENORM)@h /* denorm? */ + andis. r10,r12,(HSRR1_DENORM)@h /* denorm? */ bne+ denorm_assist #endif - mfspr r11,SPRN_HSRR0 - mfspr r12,SPRN_HSRR1 GEN_BRANCH_TO_COMMON denorm_exception, virt=0 EXC_REAL_END(denorm_exception, 0x1500, 0x100) #ifdef CONFIG_PPC_DENORMALISATION EXC_VIRT_BEGIN(denorm_exception, 0x5500, 0x100) GEN_INT_ENTRY denorm_exception, virt=1 - mfspr r10,SPRN_HSRR1 - andis. r10,r10,(HSRR1_DENORM)@h /* denorm? */ + andis. r10,r12,(HSRR1_DENORM)@h /* denorm? */ bne+ denorm_assist - mfspr r11,SPRN_HSRR0 - mfspr r12,SPRN_HSRR1 GEN_BRANCH_TO_COMMON denorm_exception, virt=1 EXC_VIRT_END(denorm_exception, 0x5500, 0x100) #else
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 14/32 powerpc/64s/exception: remove the SPR saving patch code macros
In reply to Nicholas PigginThese are used infrequently enough they don't provide much help, so inline them.
--- arch/powerpc/kernel/exceptions-64s.S | 82 ++++++++++------------------ 1 file changed, 28 insertions(+), 54 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +28 -54
@@ -110,46 +110,6 @@ name: #define EXC_HV 1 #define EXC_STD 0 -/* - * PPR save/restore macros used in exceptions-64s.S - * Used for P7 or later processors - */ -#define SAVE_PPR(area, ra) \ -BEGIN_FTR_SECTION_NESTED(940) \ - ld ra,area+EX_PPR(r13); /* Read PPR from paca */ \ - std ra,_PPR(r1); \ -END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,940) - -#define RESTORE_PPR_PACA(area, ra) \ -BEGIN_FTR_SECTION_NESTED(941) \ - ld ra,area+EX_PPR(r13); \ - mtspr SPRN_PPR,ra; \ -END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,941) - -/* - * Get an SPR into a register if the CPU has the given feature - */ -#define OPT_GET_SPR(ra, spr, ftr) \ -BEGIN_FTR_SECTION_NESTED(943) \ - mfspr ra,spr; \ -END_FTR_SECTION_NESTED(ftr,ftr,943) - -/* - * Set an SPR from a register if the CPU has the given feature - */ -#define OPT_SET_SPR(ra, spr, ftr) \ -BEGIN_FTR_SECTION_NESTED(943) \ - mtspr spr,ra; \ -END_FTR_SECTION_NESTED(ftr,ftr,943) - -/* - * Save a register to the PACA if the CPU has the given feature - */ -#define OPT_SAVE_REG_TO_PACA(offset, ra, ftr) \ -BEGIN_FTR_SECTION_NESTED(943) \ - std ra,offset(r13); \ -END_FTR_SECTION_NESTED(ftr,ftr,943) - /* * Branch to label using its 0xC000 address. This results in instruction * address suitable for MSR[IR]=0 or 1, which allows relocation to be turned
@@ -278,18 +238,18 @@ do_define_int n cmpwi r10,KVM_GUEST_MODE_SKIP beq 89f .else -BEGIN_FTR_SECTION_NESTED(947) +BEGIN_FTR_SECTION ld r10,IAREA+EX_CFAR(r13) std r10,HSTATE_CFAR(r13) -END_FTR_SECTION_NESTED(CPU_FTR_CFAR,CPU_FTR_CFAR,947) +END_FTR_SECTION_IFSET(CPU_FTR_CFAR) .endif ld r10,PACA_EXGEN+EX_CTR(r13) mtctr r10 -BEGIN_FTR_SECTION_NESTED(948) +BEGIN_FTR_SECTION ld r10,IAREA+EX_PPR(r13) std r10,HSTATE_PPR(r13) -END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) +END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) ld r11,IAREA+EX_R11(r13) ld r12,IAREA+EX_R12(r13) std r12,HSTATE_SCRATCH0(r13)
@@ -386,10 +346,14 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) SET_SCRATCH0(r13) /* save r13 */ GET_PACA(r13) std r9,IAREA+EX_R9(r13) /* save r9 */ - OPT_GET_SPR(r9, SPRN_PPR, CPU_FTR_HAS_PPR) +BEGIN_FTR_SECTION + mfspr r9,SPRN_PPR +END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) HMT_MEDIUM std r10,IAREA+EX_R10(r13) /* save r10 - r12 */ - OPT_GET_SPR(r10, SPRN_CFAR, CPU_FTR_CFAR) +BEGIN_FTR_SECTION + mfspr r10,SPRN_CFAR +END_FTR_SECTION_IFSET(CPU_FTR_CFAR) .if \ool .if !\virt b tramp_real_\name
@@ -402,8 +366,12 @@ END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) .endif .endif - OPT_SAVE_REG_TO_PACA(IAREA+EX_PPR, r9, CPU_FTR_HAS_PPR) - OPT_SAVE_REG_TO_PACA(IAREA+EX_CFAR, r10, CPU_FTR_CFAR) +BEGIN_FTR_SECTION + std r9,IAREA+EX_PPR(r13) +END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) +BEGIN_FTR_SECTION + std r10,IAREA+EX_CFAR(r13) +END_FTR_SECTION_IFSET(CPU_FTR_CFAR) INTERRUPT_TO_KERNEL mfctr r10 std r10,IAREA+EX_CTR(r13)
@@ -558,7 +526,10 @@ DEFINE_FIXED_SYMBOL(\name\()_common_virt) .endif beq 101f /* if from kernel mode */ ACCOUNT_CPU_USER_ENTRY(r13, r9, r10) - SAVE_PPR(IAREA, r9) +BEGIN_FTR_SECTION + ld r9,IAREA+EX_PPR(r13) /* Read PPR from paca */ + std r9,_PPR(r1) +END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) 101: .else .if IKUAP
@@ -598,10 +569,10 @@ DEFINE_FIXED_SYMBOL(\name\()_common_virt) std r10,_DSISR(r1) .endif -BEGIN_FTR_SECTION_NESTED(66) +BEGIN_FTR_SECTION ld r10,IAREA+EX_CFAR(r13) std r10,ORIG_GPR3(r1) -END_FTR_SECTION_NESTED(CPU_FTR_CFAR, CPU_FTR_CFAR, 66) +END_FTR_SECTION_IFSET(CPU_FTR_CFAR) ld r10,IAREA+EX_CTR(r13) std r10,_CTR(r1) std r2,GPR2(r1) /* save r2 in stackframe */
@@ -1683,10 +1654,10 @@ TRAMP_REAL_BEGIN(system_call_kvm) * HMT_MEDIUM. That allows the KVM code to save that value into the * guest state (it is the guest's PPR value). */ -BEGIN_FTR_SECTION_NESTED(948) +BEGIN_FTR_SECTION mfspr r10,SPRN_PPR std r10,HSTATE_PPR(r13) -END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948) +END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) HMT_MEDIUM mfctr r10 SET_SCRATCH0(r10)
@@ -2241,7 +2212,10 @@ denorm_done: mtspr SPRN_HSRR0,r11 mtcrf 0x80,r9 ld r9,PACA_EXGEN+EX_R9(r13) - RESTORE_PPR_PACA(PACA_EXGEN, r10) +BEGIN_FTR_SECTION + ld r10,PACA_EXGEN+EX_PPR(r13) + mtspr SPRN_PPR,r10 +END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) BEGIN_FTR_SECTION ld r10,PACA_EXGEN+EX_CFAR(r13) mtspr SPRN_CFAR,r10
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 15/32 powerpc/64s/exception: trim unused arguments from KVMTEST macro
In reply to Nicholas Piggin--- arch/powerpc/kernel/exceptions-64s.S | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +5 -5
@@ -224,7 +224,7 @@ do_define_int n #define kvmppc_interrupt kvmppc_interrupt_pr #endif -.macro KVMTEST name, hsrr, n +.macro KVMTEST name lbz r10,HSTATE_IN_GUEST(r13) cmpwi r10,0 bne \name\()_kvm
@@ -293,7 +293,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) .endm #else -.macro KVMTEST name, hsrr, n +.macro KVMTEST name .endm .macro GEN_KVM name .endm
@@ -437,7 +437,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR) DEFINE_FIXED_SYMBOL(\name\()_common_real) \name\()_common_real: .if IKVM_REAL - KVMTEST \name IHSRR IVEC + KVMTEST \name .endif ld r10,PACAKMSR(r13) /* get MSR value for kernel */
@@ -460,7 +460,7 @@ DEFINE_FIXED_SYMBOL(\name\()_common_real) DEFINE_FIXED_SYMBOL(\name\()_common_virt) \name\()_common_virt: .if IKVM_VIRT - KVMTEST \name IHSRR IVEC + KVMTEST \name 1: .endif .endif /* IVIRT */
@@ -1582,7 +1582,7 @@ INT_DEFINE_END(system_call) GET_PACA(r13) std r10,PACA_EXGEN+EX_R10(r13) INTERRUPT_TO_KERNEL - KVMTEST system_call EXC_STD 0xc00 /* uses r10, branch to system_call_kvm */ + KVMTEST system_call /* uses r10, branch to system_call_kvm */ mfctr r9 #else mr r9,r13
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 16/32 powerpc/64s/exception: hdecrementer avoid touching the stack
In reply to Nicholas PigginThe hdec interrupt handler is reported to sometimes fire in Linux if KVM leaves it pending after a guest exists. This is harmless, so there is a no-op handler for it.
The interrupt handler currently uses the regular kernel stack. Change this to avoid touching the stack entirely.
This should be the last place where the regular Linux stack can be accessed with asynchronous interrupts (including PMI) soft-masked. It might be possible to take advantage of this invariant, e.g., to context switch the kernel stack SLB entry without clearing MSR[EE].
--- arch/powerpc/include/asm/time.h | 1 - arch/powerpc/kernel/exceptions-64s.S | 25 ++++++++++++++++++++----- arch/powerpc/kernel/time.c | 9 --------- 3 files changed, 20 insertions(+), 15 deletions(-)
arch/powerpc/include/asm/time.h +0 -1
@@ -24,7 +24,6 @@ extern struct clock_event_device decrementer_clockevent;extern void generic_calibrate_decr (void ); -extern void hdec_interrupt (struct pt_regs *regs);/* Some sane defaults: 125 MHz timebase, 1GHz processor */ extern unsigned long ppc_proc_freq;
arch/powerpc/kernel/exceptions-64s.S +20 -5
@@ -1491,6 +1491,8 @@ EXC_COMMON_BEGIN(decrementer_common) INT_DEFINE_BEGIN(hdecrementer) IVEC=0x980 IHSRR=EXC_HV + ISTACK=0 + IRECONCILE=0 IKVM_REAL=1 IKVM_VIRT=1 INT_DEFINE_END(hdecrementer)
@@ -1502,11 +1504,24 @@ EXC_VIRT_BEGIN(hdecrementer, 0x4980, 0x80) GEN_INT_ENTRY hdecrementer, virt=1 EXC_VIRT_END(hdecrementer, 0x4980, 0x80) EXC_COMMON_BEGIN(hdecrementer_common) - GEN_COMMON hdecrementer - bl save_nvgprs - addi r3,r1,STACK_FRAME_OVERHEAD - bl hdec_interrupt - b ret_from_except + __GEN_COMMON_ENTRY hdecrementer + /* + * Hypervisor decrementer interrupts not caught by the KVM test + * shouldn't occur but are sometimes left pending on exit from a KVM + * guest. We don't need to do anything to clear them, as they are + * edge-triggered. + * + * Be careful to avoid touching the kernel stack. + */ + ld r10,PACA_EXGEN+EX_CTR(r13) + mtctr r10 + mtcrf 0x80,r9 + ld r9,PACA_EXGEN+EX_R9(r13) + ld r10,PACA_EXGEN+EX_R10(r13) + ld r11,PACA_EXGEN+EX_R11(r13) + ld r12,PACA_EXGEN+EX_R12(r13) + ld r13,PACA_EXGEN+EX_R13(r13) + HRFI_TO_KERNEL GEN_KVM hdecrementer
arch/powerpc/kernel/time.c +0 -9
@@ -663,15 +663,6 @@ void timer_broadcast_interrupt(void) }#endif -/* -* Hypervisor decrementer interrupts shouldn't occur but are sometimes -* left pending on exit from a KVM guest. We don't need to do anything -* to clear them, as they are edge-triggered. -*/ -void hdec_interrupt (struct pt_regs *regs) -{ -} -#ifdef CONFIG_SUSPEND static void generic_suspend_disable_irqs (void ) {
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 17/32 powerpc/64s/exception: re-inline some handlers
In reply to Nicholas PigginThe reduction in interrupt entry size allows some handlers to be re-inlined.
--- arch/powerpc/kernel/exceptions-64s.S | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +3 -3
@@ -1186,7 +1186,7 @@ INT_DEFINE_BEGIN(data_access) INT_DEFINE_END(data_access) EXC_REAL_BEGIN(data_access, 0x300, 0x80) - GEN_INT_ENTRY data_access, virt=0, ool=1 + GEN_INT_ENTRY data_access, virt=0 EXC_REAL_END(data_access, 0x300, 0x80) EXC_VIRT_BEGIN(data_access, 0x4300, 0x80) GEN_INT_ENTRY data_access, virt=1
@@ -1216,7 +1216,7 @@ INT_DEFINE_BEGIN(data_access_slb) INT_DEFINE_END(data_access_slb) EXC_REAL_BEGIN(data_access_slb, 0x380, 0x80) - GEN_INT_ENTRY data_access_slb, virt=0, ool=1 + GEN_INT_ENTRY data_access_slb, virt=0 EXC_REAL_END(data_access_slb, 0x380, 0x80) EXC_VIRT_BEGIN(data_access_slb, 0x4380, 0x80) GEN_INT_ENTRY data_access_slb, virt=1
@@ -1472,7 +1472,7 @@ INT_DEFINE_BEGIN(decrementer) INT_DEFINE_END(decrementer) EXC_REAL_BEGIN(decrementer, 0x900, 0x80) - GEN_INT_ENTRY decrementer, virt=0, ool=1 + GEN_INT_ENTRY decrementer, virt=0 EXC_REAL_END(decrementer, 0x900, 0x80) EXC_VIRT_BEGIN(decrementer, 0x4900, 0x80) GEN_INT_ENTRY decrementer, virt=1
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 18/32 powerpc/64s/exception: Clean up SRR specifiers
In reply to Nicholas PigginRemove more magic numbers and replace with nicely named bools.
--- arch/powerpc/kernel/exceptions-64s.S | 68 +++++++++++++--------------- 1 file changed, 32 insertions(+), 36 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +32 -36
@@ -105,11 +105,6 @@ name: ori reg,reg,(ABS_ADDR(label))@l; \ addis reg,reg,(ABS_ADDR(label))@h -/* Exception register prefixes */ -#define EXC_HV_OR_STD 2 /* depends on HVMODE */ -#define EXC_HV 1 -#define EXC_STD 0 - /* * Branch to label using its 0xC000 address. This results in instruction * address suitable for MSR[IR]=0 or 1, which allows relocation to be turned
@@ -128,6 +123,7 @@ name: */ #define IVEC .L_IVEC_\name\() #define IHSRR .L_IHSRR_\name\() +#define IHSRR_IF_HVMODE .L_IHSRR_IF_HVMODE_\name\() #define IAREA .L_IAREA_\name\() #define IVIRT .L_IVIRT_\name\() #define IISIDE .L_IISIDE_\name\()
@@ -159,7 +155,10 @@ do_define_int n .error "IVEC not defined" .endif .ifndef IHSRR - IHSRR=EXC_STD + IHSRR=0 + .endif + .ifndef IHSRR_IF_HVMODE + IHSRR_IF_HVMODE=0 .endif .ifndef IAREA IAREA=PACA_EXGEN
@@ -257,7 +256,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) ld r9,IAREA+EX_R9(r13) ld r10,IAREA+EX_R10(r13) /* HSRR variants have the 0x2 bit added to their trap number */ - .if IHSRR == EXC_HV_OR_STD + .if IHSRR_IF_HVMODE BEGIN_FTR_SECTION ori r12,r12,(IVEC + 0x2) FTR_SECTION_ELSE
@@ -278,7 +277,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) ld r10,IAREA+EX_R10(r13) ld r11,IAREA+EX_R11(r13) ld r12,IAREA+EX_R12(r13) - .if IHSRR == EXC_HV_OR_STD + .if IHSRR_IF_HVMODE BEGIN_FTR_SECTION b kvmppc_skip_Hinterrupt FTR_SECTION_ELSE
@@ -403,7 +402,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR) stw r10,IAREA+EX_DSISR(r13) .endif - .if IHSRR == EXC_HV_OR_STD + .if IHSRR_IF_HVMODE BEGIN_FTR_SECTION mfspr r11,SPRN_HSRR0 /* save HSRR0 */ mfspr r12,SPRN_HSRR1 /* and HSRR1 */
@@ -485,7 +484,7 @@ DEFINE_FIXED_SYMBOL(\name\()_common_virt) .abort "Bad maskable vector" .endif - .if IHSRR == EXC_HV_OR_STD + .if IHSRR_IF_HVMODE BEGIN_FTR_SECTION bne masked_Hinterrupt FTR_SECTION_ELSE
@@ -618,12 +617,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR) * Restore all registers including H/SRR0/1 saved in a stack frame of a * standard exception. */ -.macro EXCEPTION_RESTORE_REGS hsrr +.macro EXCEPTION_RESTORE_REGS hsrr=0 /* Move original SRR0 and SRR1 into the respective regs */ ld r9,_MSR(r1) - .if \hsrr == EXC_HV_OR_STD - .error "EXC_HV_OR_STD Not implemented for EXCEPTION_RESTORE_REGS" - .endif .if \hsrr mtspr SPRN_HSRR1,r9 .else
@@ -898,7 +894,7 @@ EXC_COMMON_BEGIN(system_reset_common) ld r10,SOFTE(r1) stb r10,PACAIRQSOFTMASK(r13) - EXCEPTION_RESTORE_REGS EXC_STD + EXCEPTION_RESTORE_REGS RFI_TO_USER_OR_KERNEL GEN_KVM system_reset
@@ -952,7 +948,7 @@ TRAMP_REAL_BEGIN(machine_check_fwnmi) lhz r12,PACA_IN_MCE(r13); \ subi r12,r12,1; \ sth r12,PACA_IN_MCE(r13); \ - EXCEPTION_RESTORE_REGS EXC_STD + EXCEPTION_RESTORE_REGS EXC_COMMON_BEGIN(machine_check_early_common) /*
@@ -1321,7 +1317,7 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) INT_DEFINE_BEGIN(hardware_interrupt) IVEC=0x500 - IHSRR=EXC_HV_OR_STD + IHSRR_IF_HVMODE=1 IMASK=IRQS_DISABLED IKVM_REAL=1 IKVM_VIRT=1
@@ -1490,7 +1486,7 @@ EXC_COMMON_BEGIN(decrementer_common) INT_DEFINE_BEGIN(hdecrementer) IVEC=0x980 - IHSRR=EXC_HV + IHSRR=1 ISTACK=0 IRECONCILE=0 IKVM_REAL=1
@@ -1719,7 +1715,7 @@ EXC_COMMON_BEGIN(single_step_common) INT_DEFINE_BEGIN(h_data_storage) IVEC=0xe00 - IHSRR=EXC_HV + IHSRR=1 IDAR=1 IDSISR=1 IKVM_SKIP=1
@@ -1751,7 +1747,7 @@ ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_TYPE_RADIX) INT_DEFINE_BEGIN(h_instr_storage) IVEC=0xe20 - IHSRR=EXC_HV + IHSRR=1 IKVM_REAL=1 IKVM_VIRT=1 INT_DEFINE_END(h_instr_storage)
@@ -1774,7 +1770,7 @@ EXC_COMMON_BEGIN(h_instr_storage_common) INT_DEFINE_BEGIN(emulation_assist) IVEC=0xe40 - IHSRR=EXC_HV + IHSRR=1 IKVM_REAL=1 IKVM_VIRT=1 INT_DEFINE_END(emulation_assist)
@@ -1802,7 +1798,7 @@ EXC_COMMON_BEGIN(emulation_assist_common) */ INT_DEFINE_BEGIN(hmi_exception_early) IVEC=0xe60 - IHSRR=EXC_HV + IHSRR=1 IREALMODE_COMMON=1 ISTACK=0 IRECONCILE=0
@@ -1812,7 +1808,7 @@ INT_DEFINE_END(hmi_exception_early) INT_DEFINE_BEGIN(hmi_exception) IVEC=0xe60 - IHSRR=EXC_HV + IHSRR=1 IMASK=IRQS_DISABLED IKVM_REAL=1 INT_DEFINE_END(hmi_exception)
@@ -1834,7 +1830,7 @@ EXC_COMMON_BEGIN(hmi_exception_early_common) cmpdi cr0,r3,0 bne 1f - EXCEPTION_RESTORE_REGS EXC_HV + EXCEPTION_RESTORE_REGS hsrr=1 HRFI_TO_USER_OR_KERNEL 1:
@@ -1842,7 +1838,7 @@ EXC_COMMON_BEGIN(hmi_exception_early_common) * Go to virtual mode and pull the HMI event information from * firmware. */ - EXCEPTION_RESTORE_REGS EXC_HV + EXCEPTION_RESTORE_REGS hsrr=1 GEN_INT_ENTRY hmi_exception, virt=0 GEN_KVM hmi_exception_early
@@ -1861,7 +1857,7 @@ EXC_COMMON_BEGIN(hmi_exception_common) INT_DEFINE_BEGIN(h_doorbell) IVEC=0xe80 - IHSRR=EXC_HV + IHSRR=1 IMASK=IRQS_DISABLED IKVM_REAL=1 IKVM_VIRT=1
@@ -1890,7 +1886,7 @@ EXC_COMMON_BEGIN(h_doorbell_common) INT_DEFINE_BEGIN(h_virt_irq) IVEC=0xea0 - IHSRR=EXC_HV + IHSRR=1 IMASK=IRQS_DISABLED IKVM_REAL=1 IKVM_VIRT=1
@@ -2060,7 +2056,7 @@ EXC_COMMON_BEGIN(facility_unavailable_common) INT_DEFINE_BEGIN(h_facility_unavailable) IVEC=0xf80 - IHSRR=EXC_HV + IHSRR=1 IKVM_REAL=1 IKVM_VIRT=1 INT_DEFINE_END(h_facility_unavailable)
@@ -2096,7 +2092,7 @@ EXC_VIRT_NONE(0x5100, 0x100) #ifdef CONFIG_CBE_RAS INT_DEFINE_BEGIN(cbe_system_error) IVEC=0x1200 - IHSRR=EXC_HV + IHSRR=1 IKVM_SKIP=1 IKVM_REAL=1 INT_DEFINE_END(cbe_system_error)
@@ -2147,8 +2143,8 @@ EXC_VIRT_NONE(0x5400, 0x100) INT_DEFINE_BEGIN(denorm_exception) IVEC=0x1500 - IHSRR=EXC_HV - IBRANCH_TO_COMMON=0 + IHSRR=1 + IBRANCH_COMMON=0 IKVM_REAL=1 INT_DEFINE_END(denorm_exception)
@@ -2256,7 +2252,7 @@ EXC_COMMON_BEGIN(denorm_exception_common) #ifdef CONFIG_CBE_RAS INT_DEFINE_BEGIN(cbe_maintenance) IVEC=0x1600 - IHSRR=EXC_HV + IHSRR=1 IKVM_SKIP=1 IKVM_REAL=1 INT_DEFINE_END(cbe_maintenance)
@@ -2308,7 +2304,7 @@ EXC_COMMON_BEGIN(altivec_assist_common) #ifdef CONFIG_CBE_RAS INT_DEFINE_BEGIN(cbe_thermal) IVEC=0x1800 - IHSRR=EXC_HV + IHSRR=1 IKVM_SKIP=1 IKVM_REAL=1 INT_DEFINE_END(cbe_thermal)
@@ -2371,7 +2367,7 @@ EXC_COMMON_BEGIN(soft_nmi_common) * - Else it is one of PACA_IRQ_MUST_HARD_MASK, so hard disable and return. * This is called with r10 containing the value to OR to the paca field. */ -.macro MASKED_INTERRUPT hsrr +.macro MASKED_INTERRUPT hsrr=0 .if \hsrr masked_Hinterrupt: .else
@@ -2518,8 +2514,8 @@ TRAMP_REAL_BEGIN(hrfi_flush_fallback) hrfid USE_TEXT_SECTION() - MASKED_INTERRUPT EXC_STD - MASKED_INTERRUPT EXC_HV + MASKED_INTERRUPT + MASKED_INTERRUPT hsrr=1 #ifdef CONFIG_KVM_BOOK3S_64_HANDLER kvmppc_skip_interrupt:
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 19/32 powerpc/64s/exception: add more comments for interrupt handlers
In reply to Nicholas PigginA few of the non-standard handlers are left uncommented. Some more description could be added to some.
--- arch/powerpc/kernel/exceptions-64s.S | 391 ++++++++++++++++++++++++--- 1 file changed, 353 insertions(+), 38 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +353 -38
@@ -121,26 +121,26 @@ name: /* * Interrupt code generation macros */ -#define IVEC .L_IVEC_\name\() -#define IHSRR .L_IHSRR_\name\() -#define IHSRR_IF_HVMODE .L_IHSRR_IF_HVMODE_\name\() -#define IAREA .L_IAREA_\name\() -#define IVIRT .L_IVIRT_\name\() -#define IISIDE .L_IISIDE_\name\() -#define IDAR .L_IDAR_\name\() -#define IDSISR .L_IDSISR_\name\() -#define ISET_RI .L_ISET_RI_\name\() -#define IBRANCH_TO_COMMON .L_IBRANCH_TO_COMMON_\name\() -#define IREALMODE_COMMON .L_IREALMODE_COMMON_\name\() -#define IMASK .L_IMASK_\name\() -#define IKVM_SKIP .L_IKVM_SKIP_\name\() -#define IKVM_REAL .L_IKVM_REAL_\name\() +#define IVEC .L_IVEC_\name\() /* Interrupt vector address */ +#define IHSRR .L_IHSRR_\name\() /* Sets SRR or HSRR registers */ +#define IHSRR_IF_HVMODE .L_IHSRR_IF_HVMODE_\name\() /* HSRR if HV else SRR */ +#define IAREA .L_IAREA_\name\() /* PACA save area */ +#define IVIRT .L_IVIRT_\name\() /* Has virt mode entry point */ +#define IISIDE .L_IISIDE_\name\() /* Uses SRR0/1 not DAR/DSISR */ +#define IDAR .L_IDAR_\name\() /* Uses DAR (or SRR0) */ +#define IDSISR .L_IDSISR_\name\() /* Uses DSISR (or SRR1) */ +#define ISET_RI .L_ISET_RI_\name\() /* Run common code w/ MSR[RI]=1 */ +#define IBRANCH_TO_COMMON .L_IBRANCH_TO_COMMON_\name\() /* ENTRY branch to common */ +#define IREALMODE_COMMON .L_IREALMODE_COMMON_\name\() /* Common runs in realmode */ +#define IMASK .L_IMASK_\name\() /* IRQ soft-mask bit */ +#define IKVM_SKIP .L_IKVM_SKIP_\name\() /* Generate KVM skip handler */ +#define IKVM_REAL .L_IKVM_REAL_\name\() /* Real entry tests KVM */ #define __IKVM_REAL(name) .L_IKVM_REAL_ ## name -#define IKVM_VIRT .L_IKVM_VIRT_\name\() -#define ISTACK .L_ISTACK_\name\() +#define IKVM_VIRT .L_IKVM_VIRT_\name\() /* Virt entry tests KVM */ +#define ISTACK .L_ISTACK_\name\() /* Set regular kernel stack */ #define __ISTACK(name) .L_ISTACK_ ## name -#define IRECONCILE .L_IRECONCILE_\name\() -#define IKUAP .L_IKUAP_\name\() +#define IRECONCILE .L_IRECONCILE_\name\() /* Do RECONCILE_IRQ_STATE */ +#define IKUAP .L_IKUAP_\name\() /* Do KUAP lock */ #define INT_DEFINE_BEGIN(n) \ .macro int_define_ ## n name
@@ -759,6 +759,39 @@ __start_interrupts: EXC_VIRT_NONE(0x4000, 0x100) +/** + * Interrupt 0x100 - System Reset Interrupt (SRESET aka NMI). + * This is a non-maskable, asynchronous interrupt always taken in real-mode. + * It is caused by: + * - Wake from power-saving state, on powernv. + * - An NMI from another CPU, triggered by firmware or hypercall. + * - As crash/debug signal injected from BMC, firmware or hypervisor. + * + * Handling: + * Power-save wakeup is the only performance critical path, so this is + * determined quickly as possible first. In this case volatile registers + * can be discarded and SPRs like CFAR don't need to be read. + * + * If not a powersave wakeup, then it's run as a regular interrupt, however + * it uses its own stack and PACA save area to preserve the regular kernel + * environment for debugging. + * + * This interrupt is not maskable, so triggering it when MSR[RI] is clear, + * or SCRATCH0 is in use, etc. may cause a crash. It's also not entirely + * correct to switch to virtual mode to run the regular interrupt handler + * because it might be interrupted when the MMU is in a bad state (e.g., SLB + * is clear). + * + * FWNMI: + * PAPR specifies a "fwnmi" facility which sends the sreset to a different + * entry point with a different register set up. Some hypervisors will + * send the sreset to 0x100 in the guest if it is not fwnmi capable. + * + * KVM: + * Unlike most SRR interrupts, this may be taken by the host while executing + * in a guest, so a KVM test is required. KVM will pull the CPU out of guest + * mode and then raise the sreset. + */ INT_DEFINE_BEGIN(system_reset) IVEC=0x100 IAREA=PACA_EXNMI
@@ -834,6 +867,7 @@ TRAMP_REAL_BEGIN(system_reset_idle_wake) * Vectors for the FWNMI option. Share common code. */ TRAMP_REAL_BEGIN(system_reset_fwnmi) + /* XXX: fwnmi guest could run a nested/PR guest, so why no test? */ __IKVM_REAL(system_reset)=0 GEN_INT_ENTRY system_reset, virt=0
@@ -900,6 +934,44 @@ EXC_COMMON_BEGIN(system_reset_common) GEN_KVM system_reset +/** + * Interrupt 0x200 - Machine Check Interrupt (MCE). + * This is a non-maskable interrupt always taken in real-mode. It can be + * synchronous or asynchronous, caused by hardware or software, and it may be + * taken in a power-saving state. + * + * Handling: + * Similarly to system reset, this uses its own stack and PACA save area, + * the difference is re-entrancy is allowed on the machine check stack. + * + * machine_check_early is run in real mode, and carefully decodes the + * machine check and tries to handle it (e.g., flush the SLB if there was an + * error detected there), determines if it was recoverable and logs the + * event. + * + * Then, depending on the execution context when the interrupt is taken, there + * are 3 main actions: + * - Executing in kernel mode. The event is queued with irq_work, which means + * it is handled when it is next safe to do so (i.e., the kernel has enabled + * interrupts), which could be immediately when the interrupt returns. This + * avoids nasty issues like switching to virtual mode when the MMU is in a + * bad state, or when executing OPAL code. (SRESET is exposed to such issues, + * but it has different priorities). Check to see if the CPU was in power + * save, and return via the wake up code if it was. + * + * - Executing in user mode. machine_check_exception is run like a normal + * interrupt handler, which processes the data generated by the early handler. + * + * - Executing in guest mode. The interrupt is run with its KVM test, and + * branches to KVM to deal with. KVM may queue the event for the host + * to report later. + * + * This interrupt is not maskable, so if it triggers when MSR[RI] is clear, + * or SCRATCH0 is in use, it may cause a crash. + * + * KVM: + * See SRESET. + */ INT_DEFINE_BEGIN(machine_check_early) IVEC=0x200 IAREA=PACA_EXMC
@@ -1159,19 +1231,28 @@ END_FTR_SECTION_IFSET(CPU_FTR_HVMODE) /** - * 0x300 - Data Storage Interrupt (DSI) - * This interrupt is generated due to a data access which does not have a valid - * page table entry with permissions to allow the data access to be performed. - * DAWR matches also fault here, as do RC updates, and minor misc errors e.g., - * copy/paste, AMO, certain invalid CI accesses, etc. + * Interrupt 0x300 - Data Storage Interrupt (DSI). + * This is a synchronous interrupt generated due to a data access exception, + * e.g., a load orstore which does not have a valid page table entry with + * permissions. DAWR matches also fault here, as do RC updates, and minor misc + * errors e.g., copy/paste, AMO, certain invalid CI accesses, etc. + * + * Handling: + * - Hash MMU + * Go to do_hash_page first to see if the HPT can be filled from an entry in + * the Linux page table. Hash faults can hit in kernel mode in a fairly + * arbitrary state (e.g., interrupts disabled, locks held) when accessing + * "non-bolted" regions, e.g., vmalloc space. However these should always be + * backed by Linux page tables. * - * This interrupt is delivered to the guest (HV bit unchanged). + * If none is found, do a Linux page fault. Linux page faults can happen in + * kernel mode due to user copy operations of course. * - * Linux HPT responds by first attempting to refill the hash table from the - * Linux page table, then going to a full page fault if the Linux page table - * entry was insufficient. RPT goes straight to full page fault. + * - Radix MMU + * The hardware loads from the Linux page table directly, so a fault goes + * immediately to Linux page fault. * - * PR KVM ...? + * Conditions like DAWR match are handled on the way in to Linux page fault. */ INT_DEFINE_BEGIN(data_access) IVEC=0x300
@@ -1202,6 +1283,24 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) GEN_KVM data_access +/** + * Interrupt 0x380 - Data Segment Interrupt (DSLB). + * This is a synchronous interrupt in response to an MMU fault missing SLB + * entry for HPT, or an address outside RPT translation range. + * + * Handling: + * - HPT: + * This refills the SLB, or reports an access fault similarly to a bad page + * fault. When coming from user-mode, the SLB handler may access any kernel + * data, though it may itself take a DSLB. When coming from kernel mode, + * recursive faults must be avoided so access is restricted to the kernel + * image text/data, kernel stack, and any data allocated below + * ppc64_bolted_size (first segment). The kernel handler must avoid stomping + * on user-handler data structures. + * + * A dedicated save area EXSLB is used (XXX: but it actually need not be + * these days, we could use EXGEN). + */ INT_DEFINE_BEGIN(data_access_slb) IVEC=0x380 IAREA=PACA_EXSLB
@@ -1244,6 +1343,15 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) GEN_KVM data_access_slb +/** + * Interrupt 0x400 - Instruction Storage Interrupt (ISI). + * This is a synchronous interrupt in response to an MMU fault due to an + * instruction fetch. + * + * Handling: + * Similar to DSI, though in response to fetch. The faulting address is found + * in SRR0 (rather than DAR), and status in SRR1 (rather than DSISR). + */ INT_DEFINE_BEGIN(instruction_access) IVEC=0x400 IISIDE=1
@@ -1273,6 +1381,15 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) GEN_KVM instruction_access +/** + * Interrupt 0x480 - Instruction Segment Interrupt (ISLB). + * This is a synchronous interrupt in response to an MMU fault due to an + * instruction fetch. + * + * Handling: + * Similar to DSLB, though in response to fetch. The faulting address is found + * in SRR0 (rather than DAR). + */ INT_DEFINE_BEGIN(instruction_access_slb) IVEC=0x480 IAREA=PACA_EXSLB
@@ -1315,6 +1432,29 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) GEN_KVM instruction_access_slb +/** + * Interrupt 0x500 - External Interrupt. + * This is an asynchronous maskable interrupt in response to an "external + * exception" from the interrupt controller or hypervisor (e.g., device + * interrupt). It is maskable in hardware by clearing MSR[EE], and + * soft-maskable with IRQS_DISABLED mask (i.e., local_irq_disable()). + * + * When running in HV mode, Linux sets up the LPCR[LPES] bit such that + * interrupts are delivered with HSRR registers, guests use SRRs, which + * reqiures IHSRR_IF_HVMODE. + * + * On bare metal POWER9 and later, Linux sets the LPCR[HVICE] bit such that + * external interrupts are delivered as Hypervisor Virtualization Interrupts + * rather than External Interrupts. + * + * Handling: + * This calls into Linux IRQ handler. NVGPRs are not saved to reduce overhead, + * because registers at the time of the interrupt are not so important as it is + * asynchronous. + * + * If soft masked, the masked handler will note the pending interrupt for + * replay, and clear MSR[EE] in the interrupted context. + */ INT_DEFINE_BEGIN(hardware_interrupt) IVEC=0x500 IHSRR_IF_HVMODE=1
@@ -1340,6 +1480,10 @@ EXC_COMMON_BEGIN(hardware_interrupt_common) GEN_KVM hardware_interrupt +/** + * Interrupt 0x600 - Alignment Interrupt + * This is a synchronous interrupt in response to data alignment fault. + */ INT_DEFINE_BEGIN(alignment) IVEC=0x600 IDAR=1
@@ -1363,6 +1507,15 @@ EXC_COMMON_BEGIN(alignment_common) GEN_KVM alignment +/** + * Interrupt 0x700 - Program Interrupt (program check). + * This is a synchronous interrupt in response to various instruction faults: + * traps, privilege errors, TM errors, floating point exceptions. + * + * Handling: + * This interrupt may use the "emergency stack" in some cases when being taken + * from kernel context, which complicates handling. + */ INT_DEFINE_BEGIN(program_check) IVEC=0x700 IKVM_REAL=1
@@ -1416,6 +1569,15 @@ EXC_COMMON_BEGIN(program_check_common) GEN_KVM program_check +/* + * Interrupt 0x800 - Floating-Point Unavailable Interrupt. + * This is a synchronous interrupt in response to executing an fp instruction + * with MSR[FP]=0. + * + * Handling: + * This will load FP registers and enable the FP bit if coming from userspace, + * otherwise report a bad kernel use of FP. + */ INT_DEFINE_BEGIN(fp_unavailable) IVEC=0x800 IRECONCILE=0
@@ -1461,6 +1623,23 @@ END_FTR_SECTION_IFSET(CPU_FTR_TM) GEN_KVM fp_unavailable +/** + * Interrupt 0x900 - Decrementer Interrupt. + * This is an asynchronous interrupt in response to a decrementer exception + * (e.g., DEC has wrapped below zero). It is maskable in hardware by clearing + * MSR[EE], and soft-maskable with IRQS_DISABLED mask (i.e., + * local_irq_disable()). + * + * Handling: + * This calls into Linux timer handler. NVGPRs are not saved (see 0x500). + * + * If soft masked, the masked handler will note the pending interrupt for + * replay, and bump the decrementer to a high value, leaving MSR[EE] enabled + * in the interrupted context. + * If PPC_WATCHDOG is configured, the soft masked handler will actually set + * things back up to run soft_nmi_interrupt as a regular interrupt handler + * on the emergency stack. + */ INT_DEFINE_BEGIN(decrementer) IVEC=0x900 IMASK=IRQS_DISABLED
@@ -1484,6 +1663,16 @@ EXC_COMMON_BEGIN(decrementer_common) GEN_KVM decrementer +/** + * Interrupt 0x980 - Hypervisor Decrementer Interrupt. + * This is an asynchronous interrupt, similar to 0x900 but for the HDEC + * register. + * + * Handling: + * Linux does not use this outside KVM where it's used to keep a host timer + * while the guest is given control of DEC. It should normally be caught by + * the KVM test and routed there. + */ INT_DEFINE_BEGIN(hdecrementer) IVEC=0x980 IHSRR=1
@@ -1522,6 +1711,20 @@ EXC_COMMON_BEGIN(hdecrementer_common) GEN_KVM hdecrementer +/** + * Interrupt 0xa00 - Directed Privileged Doorbell Interrupt. + * This is an asynchronous interrupt in response to a msgsndp doorbell. + * It is maskable in hardware by clearing MSR[EE], and soft-maskable with + * IRQS_DISABLED mask (i.e., local_irq_disable()). + * + * Handling: + * Guests may use this for IPIs between threads in a core if the + * hypervisor supports it. NVGPRS are not saved (see 0x500). + * + * If soft masked, the masked handler will note the pending interrupt for + * replay, leaving MSR[EE] enabled in the interrupted context because the + * doorbells are edge triggered. + */ INT_DEFINE_BEGIN(doorbell_super) IVEC=0xa00 IMASK=IRQS_DISABLED
@@ -1552,16 +1755,20 @@ EXC_COMMON_BEGIN(doorbell_super_common) EXC_REAL_NONE(0xb00, 0x100) EXC_VIRT_NONE(0x4b00, 0x100) -/* - * system call / hypercall (0xc00, 0x4c00) - * - * The system call exception is invoked with "sc 0" and does not alter HV bit. - * - * The hypercall is invoked with "sc 1" and sets HV=1. +/** + * Interrupt 0xc00 - System Call Interrupt (syscall, hcall). + * This is a synchronous interrupt invoked with the "sc" instruction. The + * system call is invoked with "sc 0" and does not alter the HV bit, so it + * is directed to the currently running OS. The hypercall is invoked with + * "sc 1" and it sets HV=1, so it elevates to hypervisor. * * In HPT, sc 1 always goes to 0xc00 real mode. In RADIX, sc 1 can go to * 0x4c00 virtual mode. * + * Handling: + * If the KVM test fires then it was due to a hypercall and is accordingly + * routed to KVM. Otherwise this executes a normal Linux system call. + * * Call convention: * * syscall and hypercalls register conventions are documented in
@@ -1692,6 +1899,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) #endif +/** + * Interrupt 0xd00 - Trace Interrupt. + * This is a synchronous interrupt in response to instruction step or + * breakpoint faults. + */ INT_DEFINE_BEGIN(single_step) IVEC=0xd00 IKVM_REAL=1
@@ -1713,6 +1925,18 @@ EXC_COMMON_BEGIN(single_step_common) GEN_KVM single_step +/** + * Interrupt 0xe00 - Hypervisor Data Storage Interrupt (HDSI). + * This is a synchronous interrupt in response to an MMU fault caused by a + * guest data access. + * + * Handling: + * This should always get routed to KVM. In radix MMU mode, this is caused + * by a guest nested radix access that can't be performed due to the + * partition scope page table. In hash mode, this can be caused by guests + * running with translation disabled (virtual real mode) or with VPM enabled. + * KVM will update the page table structures or disallow the access. + */ INT_DEFINE_BEGIN(h_data_storage) IVEC=0xe00 IHSRR=1
@@ -1745,6 +1969,11 @@ ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_TYPE_RADIX) GEN_KVM h_data_storage +/** + * Interrupt 0xe20 - Hypervisor Instruction Storage Interrupt (HISI). + * This is a synchronous interrupt in response to an MMU fault caused by a + * guest instruction fetch, similar to HDSI. + */ INT_DEFINE_BEGIN(h_instr_storage) IVEC=0xe20 IHSRR=1
@@ -1768,6 +1997,9 @@ EXC_COMMON_BEGIN(h_instr_storage_common) GEN_KVM h_instr_storage +/** + * Interrupt 0xe40 - Hypervisor Emulation Assistance Interrupt. + */ INT_DEFINE_BEGIN(emulation_assist) IVEC=0xe40 IHSRR=1
@@ -1791,10 +2023,29 @@ EXC_COMMON_BEGIN(emulation_assist_common) GEN_KVM emulation_assist -/* - * hmi_exception trampoline is a special case. It jumps to hmi_exception_early - * first, and then eventaully from there to the trampoline to get into virtual - * mode. +/** + * Interrupt 0xe60 - Hypervisor Maintenance Interrupt (HMI). + * This is an asynchronous interrupt caused by a Hypervisor Maintenance + * Exception. It is always taken in real mode but uses HSRR registers + * unlike SRESET and MCE. + * + * It is maskable in hardware by clearing MSR[EE], and partially soft-maskable + * with IRQS_DISABLED mask (i.e., local_irq_disable()). + * + * Handling: + * This is a special case, this is handled similarly to machine checks, with an + * initial real mode handler that is not soft-masked, which attempts to fix the + * problem. Then a regular handler which is soft-maskable and reports the + * problem. + * + * The emergency stack is used for the early real mode handler. + * + * XXX: unclear why MCE and HMI schemes could not be made common, e.g., + * either use soft-masking for the MCE, or use irq_work for the HMI. + * + * KVM: + * Unlike MCE, this calls into KVM without calling the real mode handler + * first. */ INT_DEFINE_BEGIN(hmi_exception_early) IVEC=0xe60
@@ -1855,6 +2106,11 @@ EXC_COMMON_BEGIN(hmi_exception_common) GEN_KVM hmi_exception +/** + * Interrupt 0xe80 - Directed Hypervisor Doorbell Interrupt. + * This is an asynchronous interrupt in response to a msgsnd doorbell. + * Similar to the 0xa00 doorbell but for host rather than guest. + */ INT_DEFINE_BEGIN(h_doorbell) IVEC=0xe80 IHSRR=1
@@ -1884,6 +2140,11 @@ EXC_COMMON_BEGIN(h_doorbell_common) GEN_KVM h_doorbell +/** + * Interrupt 0xea0 - Hypervisor Virtualization Interrupt. + * This is an asynchronous interrupt in response to an "external exception". + * Similar to 0x500 but for host only. + */ INT_DEFINE_BEGIN(h_virt_irq) IVEC=0xea0 IHSRR=1
@@ -1915,6 +2176,22 @@ EXC_REAL_NONE(0xee0, 0x20) EXC_VIRT_NONE(0x4ee0, 0x20) +/* + * Interrupt 0xf00 - Performance Monitor Interrupt (PMI, PMU). + * This is an asynchronous interrupt in response to a PMU exception. + * It is maskable in hardware by clearing MSR[EE], and soft-maskable with + * IRQS_PMI_DISABLED mask (NOTE: NOT local_irq_disable()). + * + * Handling: + * This calls into the perf subsystem. + * + * Like the watchdog soft-nmi, it appears an NMI interrupt to Linux, in that it + * runs under local_irq_disable. However it may be soft-masked in + * powerpc-specific code. + * + * If soft masked, the masked handler will note the pending interrupt for + * replay, and clear MSR[EE] in the interrupted context. + */ INT_DEFINE_BEGIN(performance_monitor) IVEC=0xf00 IMASK=IRQS_PMI_DISABLED
@@ -1938,6 +2215,12 @@ EXC_COMMON_BEGIN(performance_monitor_common) GEN_KVM performance_monitor +/** + * Interrupt 0xf20 - Vector Unavailable Interrupt. + * This is a synchronous interrupt in response to + * executing a vector (or altivec) instruction with MSR[VEC]=0. + * Similar to FP unavailable. + */ INT_DEFINE_BEGIN(altivec_unavailable) IVEC=0xf20 IRECONCILE=0
@@ -1986,6 +2269,12 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) GEN_KVM altivec_unavailable +/** + * Interrupt 0xf40 - VSX Unavailable Interrupt. + * This is a synchronous interrupt in response to + * executing a VSX instruction with MSR[VSX]=0. + * Similar to FP unavailable. + */ INT_DEFINE_BEGIN(vsx_unavailable) IVEC=0xf40 IRECONCILE=0
@@ -2033,6 +2322,13 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX) GEN_KVM vsx_unavailable +/** + * Interrupt 0xf60 - Facility Unavailable Interrupt. + * This is a synchronous interrupt in response to + * executing an instruction without access to the facility that can be + * resolved by the OS (e.g., FSCR, MSR). + * Similar to FP unavailable. + */ INT_DEFINE_BEGIN(facility_unavailable) IVEC=0xf60 IKVM_REAL=1
@@ -2054,6 +2350,13 @@ EXC_COMMON_BEGIN(facility_unavailable_common) GEN_KVM facility_unavailable +/** + * Interrupt 0xf60 - Hypervisor Facility Unavailable Interrupt. + * This is a synchronous interrupt in response to + * executing an instruction without access to the facility that can only + * be resolved in HV mode (e.g., HFSCR). + * Similar to FP unavailable. + */ INT_DEFINE_BEGIN(h_facility_unavailable) IVEC=0xf80 IHSRR=1
@@ -2141,6 +2444,18 @@ EXC_COMMON_BEGIN(instruction_breakpoint_common) EXC_REAL_NONE(0x1400, 0x100) EXC_VIRT_NONE(0x5400, 0x100) +/** + * Interrupt 0x1500 - Soft Patch Interrupt + * + * Handling: + * This is an implementation specific interrupt which can be used for a + * range of exceptions. + * + * This interrupt handler is unique in that it runs the denormal assist + * code even for guests (and even in guest context) without going to KVM, + * for speed. POWER9 does not raise denorm exceptions, so this special case + * could be phased out in future to reduce special cases. + */ INT_DEFINE_BEGIN(denorm_exception) IVEC=0x1500 IHSRR=1
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 20/32 powerpc/64s/exception: only test KVM in SRR interrupts when PR KVM is supported
In reply to Nicholas PigginApart from SRESET, MCE, and syscall (hcall variant), the SRR type interrupts are not escalated to hypervisor mode, so delivered to the OS.
When running PR KVM, the OS is the hypervisor, and the guest runs with MSR[PR]=1, so these interrupts must test if a guest was running when interrupted. These tests are required at the real-mode entry points because the PR KVM host runs with LPCR[AIL]=0.
In HV KVM and nested HV KVM, the guest always receives these interrupts, so there is no need for the host to make this test. So remove the tests if PR KVM is not configured.
--- arch/powerpc/kernel/exceptions-64s.S | 65 ++++++++++++++++++++++++++-- 1 file changed, 62 insertions(+), 3 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +62 -3
@@ -214,9 +214,36 @@ do_define_int n #ifdef CONFIG_KVM_BOOK3S_64_HANDLER #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE /* - * If hv is possible, interrupts come into to the hv version - * of the kvmppc_interrupt code, which then jumps to the PR handler, - * kvmppc_interrupt_pr, if the guest is a PR guest. + * All interrupts which set HSRR registers, as well as SRESET and MCE and + * syscall when invoked with "sc 1" switch to MSR[HV]=1 (HVMODE) to be taken, + * so they all generally need to test whether they were taken in guest context. + * + * Note: SRESET and MCE may also be sent to the guest by the hypervisor, and be + * taken with MSR[HV]=0. + * + * Interrupts which set SRR registers (with the above exceptions) do not + * elevate to MSR[HV]=1 mode, though most can be taken when running with + * MSR[HV]=1 (e.g., bare metal kernel and userspace). So these interrupts do + * not need to test whether a guest is running because they get delivered to + * the guest directly, including nested HV KVM guests. + * + * The exception is PR KVM, where the guest runs with MSR[PR]=1 and the host + * runs with MSR[HV]=0, so the host takes all interrupts on behalf of the + * guest. PR KVM runs with LPCR[AIL]=0 which causes interrupts to always be + * delivered to the real-mode entry point, therefore such interrupts only test + * KVM in their real mode handlers, and only when PR KVM is possible. + * + * Interrupts that are taken in MSR[HV]=0 and escalate to MSR[HV]=1 are always + * delivered in real-mode when the MMU is in hash mode because the MMU + * registers are not set appropriately to translate host addresses. In nested + * radix mode these can be delivered in virt-mode as the host translations are + * used implicitly (see: effective LPID, effective PID). + */ + +/* + * If an interrupt is taken while a guest is running, it is immediately routed + * to KVM to handle. If both HV and PR KVM arepossible, KVM interrupts go first + * to kvmppc_interrupt_hv, which handles the PR guest case. */ #define kvmppc_interrupt kvmppc_interrupt_hv #else
@@ -1258,8 +1285,10 @@ INT_DEFINE_BEGIN(data_access) IVEC=0x300 IDAR=1 IDSISR=1 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_SKIP=1 IKVM_REAL=1 +#endif INT_DEFINE_END(data_access) EXC_REAL_BEGIN(data_access, 0x300, 0x80)
@@ -1306,8 +1335,10 @@ INT_DEFINE_BEGIN(data_access_slb) IAREA=PACA_EXSLB IRECONCILE=0 IDAR=1 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_SKIP=1 IKVM_REAL=1 +#endif INT_DEFINE_END(data_access_slb) EXC_REAL_BEGIN(data_access_slb, 0x380, 0x80)
@@ -1357,7 +1388,9 @@ INT_DEFINE_BEGIN(instruction_access) IISIDE=1 IDAR=1 IDSISR=1 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(instruction_access) EXC_REAL_BEGIN(instruction_access, 0x400, 0x80)
@@ -1396,7 +1429,9 @@ INT_DEFINE_BEGIN(instruction_access_slb) IRECONCILE=0 IISIDE=1 IDAR=1 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(instruction_access_slb) EXC_REAL_BEGIN(instruction_access_slb, 0x480, 0x80)
@@ -1488,7 +1523,9 @@ INT_DEFINE_BEGIN(alignment) IVEC=0x600 IDAR=1 IDSISR=1 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(alignment) EXC_REAL_BEGIN(alignment, 0x600, 0x100)
@@ -1518,7 +1555,9 @@ EXC_COMMON_BEGIN(alignment_common) */ INT_DEFINE_BEGIN(program_check) IVEC=0x700 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(program_check) EXC_REAL_BEGIN(program_check, 0x700, 0x100)
@@ -1581,7 +1620,9 @@ EXC_COMMON_BEGIN(program_check_common) INT_DEFINE_BEGIN(fp_unavailable) IVEC=0x800 IRECONCILE=0 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(fp_unavailable) EXC_REAL_BEGIN(fp_unavailable, 0x800, 0x100)
@@ -1643,7 +1684,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_TM) INT_DEFINE_BEGIN(decrementer) IVEC=0x900 IMASK=IRQS_DISABLED +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(decrementer) EXC_REAL_BEGIN(decrementer, 0x900, 0x80)
@@ -1728,7 +1771,9 @@ EXC_COMMON_BEGIN(hdecrementer_common) INT_DEFINE_BEGIN(doorbell_super) IVEC=0xa00 IMASK=IRQS_DISABLED +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(doorbell_super) EXC_REAL_BEGIN(doorbell_super, 0xa00, 0x100)
@@ -1906,7 +1951,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) */ INT_DEFINE_BEGIN(single_step) IVEC=0xd00 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(single_step) EXC_REAL_BEGIN(single_step, 0xd00, 0x100)
@@ -2195,7 +2242,9 @@ EXC_VIRT_NONE(0x4ee0, 0x20) INT_DEFINE_BEGIN(performance_monitor) IVEC=0xf00 IMASK=IRQS_PMI_DISABLED +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(performance_monitor) EXC_REAL_BEGIN(performance_monitor, 0xf00, 0x20)
@@ -2224,7 +2273,9 @@ EXC_COMMON_BEGIN(performance_monitor_common) INT_DEFINE_BEGIN(altivec_unavailable) IVEC=0xf20 IRECONCILE=0 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(altivec_unavailable) EXC_REAL_BEGIN(altivec_unavailable, 0xf20, 0x20)
@@ -2278,7 +2329,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) INT_DEFINE_BEGIN(vsx_unavailable) IVEC=0xf40 IRECONCILE=0 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(vsx_unavailable) EXC_REAL_BEGIN(vsx_unavailable, 0xf40, 0x20)
@@ -2331,7 +2384,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX) */ INT_DEFINE_BEGIN(facility_unavailable) IVEC=0xf60 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(facility_unavailable) EXC_REAL_BEGIN(facility_unavailable, 0xf60, 0x20)
@@ -2421,8 +2476,10 @@ EXC_VIRT_NONE(0x5200, 0x100) INT_DEFINE_BEGIN(instruction_breakpoint) IVEC=0x1300 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_SKIP=1 IKVM_REAL=1 +#endif INT_DEFINE_END(instruction_breakpoint) EXC_REAL_BEGIN(instruction_breakpoint, 0x1300, 0x100)
@@ -2593,7 +2650,9 @@ EXC_VIRT_NONE(0x5600, 0x100) INT_DEFINE_BEGIN(altivec_assist) IVEC=0x1700 +#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE IKVM_REAL=1 +#endif INT_DEFINE_END(altivec_assist) EXC_REAL_BEGIN(altivec_assist, 0x1700, 0x100)
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 21/32 powerpc/64s/exception: sreset interrupts reconcile fix
In reply to Nicholas PigginThis adds IRQ_HARD_DIS to irq_happened. Although it doesn't seem to matter much because we're not allowed to enable irqs in an NMI handler, the soft-irq debugging code is becoming more strict about ensuring IRQ_HARD_DIS is in sync with MSR[EE], this may help avoid asserts or other issues.
Add a comments explaining why MCE does not have this. Early machine check is generally much smaller and more contained code which will explode if you look at it wrong anyway as it runs in real mode, though there's an argument that we should do similar reconciling for the MCE as well.
--- arch/powerpc/kernel/exceptions-64s.S | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +10 -4
@@ -920,18 +920,19 @@ EXC_COMMON_BEGIN(system_reset_common) __GEN_COMMON_BODY system_reset bl save_nvgprs /* - * Set IRQS_ALL_DISABLED unconditionally so arch_irqs_disabled does + * Set IRQS_ALL_DISABLED unconditionally so irqs_disabled() does * the right thing. We do not want to reconcile because that goes * through irq tracing which we don't want in NMI. * - * Save PACAIRQHAPPENED because some code will do a hard disable - * (e.g., xmon). So we want to restore this back to where it was - * when we return. DAR is unused in the stack, so save it there. + * Save PACAIRQHAPPENED to _DAR (otherwise unused), and set HARD_DIS + * as we are running with MSR[EE]=0. */ li r10,IRQS_ALL_DISABLED stb r10,PACAIRQSOFTMASK(r13) lbz r10,PACAIRQHAPPENED(r13) std r10,_DAR(r1) + ori r10,r10,PACA_IRQ_HARD_DIS + stb r10,PACAIRQHAPPENED(r13) addi r3,r1,STACK_FRAME_OVERHEAD bl system_reset_exception
@@ -976,6 +977,11 @@ EXC_COMMON_BEGIN(system_reset_common) * error detected there), determines if it was recoverable and logs the * event. * + * This early code does not "reconcile" irq soft-mask state like SRESET or + * regular interrupts do, so irqs_disabled() among other things may not work + * properly (irq disable/enable already doesn't work because irq tracing can + * not work in real mode). + * * Then, depending on the execution context when the interrupt is taken, there * are 3 main actions: * - Executing in kernel mode. The event is queued with irq_work, which means
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 22/32 powerpc/64s/exception: soft nmi interrupt should not use ret_from_except
In reply to Nicholas PigginThe soft nmi handler does not reconcile interrupt state, so it should not return via the normal ret_from_except path. Return like other NMIs, using the EXCEPTION_RESTORE_REGS macro.
This becomes important when the scv interrupt is implemented, which must handle soft-masked interrupts that have r13 set to something other than the PACA -- returning to kernel in this case must restore r13.
---
v3:
- save/restore irq soft mask state like other NMIs rather than a normal reconcile, to avoid soft mask warnings or possibly worse.
arch/powerpc/kernel/exceptions-64s.S | 29 +++++++++++++++++++++++++++- 1 file changed, 28 insertions(+), 1 deletion(-)
arch/powerpc/kernel/exceptions-64s.S +28 -1
@@ -2713,6 +2713,7 @@ EXC_VIRT_NONE(0x5800, 0x100) INT_DEFINE_BEGIN(soft_nmi) IVEC=0x900 ISTACK=0 + IRECONCILE=0 /* Soft-NMI may fire under local_irq_disable */ INT_DEFINE_END(soft_nmi) /*
@@ -2731,9 +2732,35 @@ EXC_COMMON_BEGIN(soft_nmi_common) subi r1,r1,INT_FRAME_SIZE __GEN_COMMON_BODY soft_nmi bl save_nvgprs + + /* + * Set IRQS_ALL_DISABLED and save PACAIRQHAPPENED (see + * system_reset_common) + */ + li r10,IRQS_ALL_DISABLED + stb r10,PACAIRQSOFTMASK(r13) + lbz r10,PACAIRQHAPPENED(r13) + std r10,_DAR(r1) + ori r10,r10,PACA_IRQ_HARD_DIS + stb r10,PACAIRQHAPPENED(r13) + addi r3,r1,STACK_FRAME_OVERHEAD bl soft_nmi_interrupt - b ret_from_except + + /* Clear MSR_RI before setting SRR0 and SRR1. */ + li r9,0 + mtmsrd r9,1 + + /* + * Restore soft mask settings. + */ + ld r10,_DAR(r1) + stb r10,PACAIRQHAPPENED(r13) + ld r10,SOFTE(r1) + stb r10,PACAIRQSOFTMASK(r13) + + EXCEPTION_RESTORE_REGS hsrr=0 + RFI_TO_KERNEL #endif /* CONFIG_PPC_WATCHDOG */
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 23/32 powerpc/64: system call remove non-volatile GPR save optimisation
In reply to Nicholas Pigginpowerpc has an optimisation where interrupts avoid saving the non-volatile (or callee saved) registers to the interrupt stack frame if they are not required.
Two problems with this are that an interrupt does not always know whether it will need non-volatiles; and if it does need them, they can only be saved from the entry-scoped asm code (because we don't control what the C compiler does with these registers).
system calls are the most difficult: some system calls always require all registers (e.g., fork, to copy regs into the child). Sometimes registers are only required under certain conditions (e.g., tracing, signal delivery). These cases require ugly logic in the call chains (e.g., ppc_fork), and require a lot of logic to be implemented in asm.
So remove the optimisation for system calls, and always save NVGPRs on entry. Modern high performance CPUs are not so sensitive, because the stores are dense in cache and can be hidden by other expensive work in the syscall path -- the null syscall selftests benchmark on POWER9 is not slowed (124.40ns before and 123.64ns after, i.e., within the noise).
Other interrupts retain the NVGPR optimisation for now.
--- arch/powerpc/kernel/entry_64.S | 72 +++++------------------- arch/powerpc/kernel/syscalls/syscall.tbl | 22 +++++--- 2 files changed, 28 insertions(+), 66 deletions(-)
arch/powerpc/kernel/entry_64.S +13 -59
@@ -98,13 +98,14 @@ END_BTB_FLUSH_SECTION std r11,_XER(r1) std r11,_CTR(r1) std r9,GPR13(r1) + SAVE_NVGPRS(r1) mflr r10 /* * This clears CR0.SO (bit 28), which is the error indication on * return from this system call. */ rldimi r2,r11,28,(63-28) - li r11,0xc01 + li r11,0xc00 std r10,_LINK(r1) std r11,_TRAP(r1) std r3,ORIG_GPR3(r1)
@@ -323,7 +324,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) /* Traced system call support */ .Lsyscall_dotrace: - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl do_syscall_trace_enter
@@ -408,7 +408,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) mtmsrd r10,1 #endif /* CONFIG_PPC_BOOK3E */ - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl do_syscall_trace_leave b ret_from_except
@@ -442,62 +441,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) _ASM_NOKPROBE_SYMBOL(system_call_common); _ASM_NOKPROBE_SYMBOL(system_call_exit); -/* Save non-volatile GPRs, if not already saved. */ -_GLOBAL(save_nvgprs) - ld r11,_TRAP(r1) - andi. r0,r11,1 - beqlr- - SAVE_NVGPRS(r1) - clrrdi r0,r11,1 - std r0,_TRAP(r1) - blr -_ASM_NOKPROBE_SYMBOL(save_nvgprs); - - -/* - * The sigsuspend and rt_sigsuspend system calls can call do_signal - * and thus put the process into the stopped state where we might - * want to examine its user state with ptrace. Therefore we need - * to save all the nonvolatile registers (r14 - r31) before calling - * the C code. Similarly, fork, vfork and clone need the full - * register state on the stack so that it can be copied to the child. - */ - -_GLOBAL(ppc_fork) - bl save_nvgprs - bl sys_fork - b .Lsyscall_exit - -_GLOBAL(ppc_vfork) - bl save_nvgprs - bl sys_vfork - b .Lsyscall_exit - -_GLOBAL(ppc_clone) - bl save_nvgprs - bl sys_clone - b .Lsyscall_exit - -_GLOBAL(ppc_clone3) - bl save_nvgprs - bl sys_clone3 - b .Lsyscall_exit - -_GLOBAL(ppc32_swapcontext) - bl save_nvgprs - bl compat_sys_swapcontext - b .Lsyscall_exit - -_GLOBAL(ppc64_swapcontext) - bl save_nvgprs - bl sys_swapcontext - b .Lsyscall_exit - -_GLOBAL(ppc_switch_endian) - bl save_nvgprs - bl sys_switch_endian - b .Lsyscall_exit - _GLOBAL(ret_from_fork) bl schedule_tail REST_NVGPRS(r1)
@@ -516,6 +459,17 @@ _GLOBAL(ret_from_kernel_thread) li r3,0 b .Lsyscall_exit +/* Save non-volatile GPRs, if not already saved. */ +_GLOBAL(save_nvgprs) + ld r11,_TRAP(r1) + andi. r0,r11,1 + beqlr- + SAVE_NVGPRS(r1) + clrrdi r0,r11,1 + std r0,_TRAP(r1) + blr +_ASM_NOKPROBE_SYMBOL(save_nvgprs); + #ifdef CONFIG_PPC_BOOK3S_64 #define FLUSH_COUNT_CACHE \
arch/powerpc/kernel/syscalls/syscall.tbl +15 -7
@@ -9,7 +9,9 @@ # 0 nospu restart_syscall sys_restart_syscall 1 nospu exit sys_exit -2 nospu fork ppc_fork +2 32 fork ppc_fork sys_fork +2 64 fork sys_fork +2 spu fork sys_ni_syscall 3 common read sys_read 4 common write sys_write 5 common open sys_open compat_sys_open
@@ -158,7 +160,9 @@ 119 32 sigreturn sys_sigreturn compat_sys_sigreturn 119 64 sigreturn sys_ni_syscall 119 spu sigreturn sys_ni_syscall -120 nospu clone ppc_clone +120 32 clone ppc_clone sys_clone +120 64 clone sys_clone +120 spu clone sys_ni_syscall 121 common setdomainname sys_setdomainname 122 common uname sys_newuname 123 common modify_ldt sys_ni_syscall
@@ -240,7 +244,9 @@ 186 spu sendfile sys_sendfile64 187 common getpmsg sys_ni_syscall 188 common putpmsg sys_ni_syscall -189 nospu vfork ppc_vfork +189 32 vfork ppc_vfork sys_vfork +189 64 vfork sys_vfork +189 spu vfork sys_ni_syscall 190 common ugetrlimit sys_getrlimit compat_sys_getrlimit 191 common readahead sys_readahead compat_sys_readahead 192 32 mmap2 sys_mmap2 compat_sys_mmap2
@@ -316,8 +322,8 @@ 248 32 clock_nanosleep sys_clock_nanosleep_time32 248 64 clock_nanosleep sys_clock_nanosleep 248 spu clock_nanosleep sys_clock_nanosleep -249 32 swapcontext ppc_swapcontext ppc32_swapcontext -249 64 swapcontext ppc64_swapcontext +249 32 swapcontext ppc_swapcontext compat_sys_swapcontext +249 64 swapcontext sys_swapcontext 249 spu swapcontext sys_ni_syscall 250 common tgkill sys_tgkill 251 32 utimes sys_utimes_time32
@@ -456,7 +462,7 @@ 361 common bpf sys_bpf 362 nospu execveat sys_execveat compat_sys_execveat 363 32 switch_endian sys_ni_syscall -363 64 switch_endian ppc_switch_endian +363 64 switch_endian sys_switch_endian 363 spu switch_endian sys_ni_syscall 364 common userfaultfd sys_userfaultfd 365 common membarrier sys_membarrier
@@ -516,6 +522,8 @@ 432 common fsmount sys_fsmount 433 common fspick sys_fspick 434 common pidfd_open sys_pidfd_open -435 nospu clone3 ppc_clone3 +435 32 clone3 ppc_clone3 sys_clone3 +435 64 clone3 sys_clone3 +435 spu clone3 sys_ni_syscall 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 24/32 powerpc/64: sstep ifdef the deprecated fast endian switch syscall
In reply to Nicholas Piggin--- arch/powerpc/lib/sstep.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
arch/powerpc/lib/sstep.c +3 -2
@@ -3179,8 +3179,9 @@ int emulate_step(struct pt_regs *regs, unsigned int instr)* entry code works. If that is changed, this will * need to be changed also. */ -if (regs->gpr[0 ] ==0x1ebe && -cpu_has_feature (CPU_FTR_REAL_LE )) { +if (IS_ENABLED (CONFIG_PPC_FAST_ENDIAN_SWITCH ) && +cpu_has_feature (CPU_FTR_REAL_LE ) && + regs->gpr[0 ] ==0x1ebe ) { regs->msr ^=MSR_LE ; goto instr_done; }
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 25/32 powerpc/64: system call implement entry/exit logic in C
In reply to Nicholas PigginSystem call entry and particularly exit code is beyond the limit of what is reasonable to implement in asm.
This conversion moves all conditional branches out of the asm code, except for the case that all GPRs should be restored at exit.
Null syscall test is about 5% faster after this patch, because the exit work is handled under local_irq_disable, and the hard mask and pending interrupt replay is handled after that, which avoids games with MSR.
---
v2,rebase (from Michal):
- Add endian conversion for dtl_idx (ms)
- Fix sparse warning about missing declaration (ms)
- Add unistd.h to fix some defconfigs, add SPDX, minor formatting (mpe)
v3: Fixes thanks to reports from mpe and selftests errors:
- Several soft-mask debug and unsafe smp_processor_id() warnings due to tracing and other false positives due to checks in "unreconciled" code.
- Fix a bug with syscall tracing functions that set registers (e.g., PTRACE_SETREG) not setting GPRs properly.
- Fix silly tabort_syscall bug that causes kernel crashes when making system calls in transactional state.
arch/powerpc/include/asm/asm-prototypes.h | 17 +- .../powerpc/include/asm/book3s/64/kup-radix.h | 14 +- arch/powerpc/include/asm/cputime.h | 29 ++ arch/powerpc/include/asm/hw_irq.h | 4 + arch/powerpc/include/asm/ptrace.h | 3 + arch/powerpc/include/asm/signal.h | 3 + arch/powerpc/include/asm/switch_to.h | 5 + arch/powerpc/include/asm/time.h | 3 + arch/powerpc/kernel/Makefile | 3 +- arch/powerpc/kernel/entry_64.S | 338 +++--------------- arch/powerpc/kernel/signal.h | 2 - arch/powerpc/kernel/syscall_64.c | 213 +++++++++++ arch/powerpc/kernel/systbl.S | 9 +- 13 files changed, 328 insertions(+), 315 deletions(-) create mode 100644 arch/powerpc/kernel/syscall_64.c
arch/powerpc/include/asm/asm-prototypes.h +6 -11
@@ -97,6 +97,12 @@ ppc_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp,unsigned long __init early_init (unsigned long dt_ptr);void __init machine_init (u64 dt_ptr);#endif +#ifdef CONFIG_PPC64 +long system_call_exception (long r3,long r4,long r5,long r6,long r7,long r8,unsigned long r0,struct pt_regs *regs); +notrace unsigned long syscall_exit_prepare (unsigned long r3,struct pt_regs *regs); +notrace unsigned long interrupt_exit_user_prepare (struct pt_regs *regs,unsigned long msr); +notrace unsigned long interrupt_exit_kernel_prepare (struct pt_regs *regs,unsigned long msr); +#endif long ppc_fadvise64_64 (int fd,int advice,u32 offset_high,u32 offset_low,u32 len_high,u32 len_low);
@@ -104,14 +110,6 @@ long sys_switch_endian(void); notrace unsigned int __check_irq_replay(void); void notrace restore_interrupts(void); -/* ptrace */ -long do_syscall_trace_enter (struct pt_regs *regs); -void do_syscall_trace_leave (struct pt_regs *regs); - -/* process */ -void restore_math (struct pt_regs *regs); -void restore_tm_state (struct pt_regs *regs); -/* prom_init (OpenFirmware) */ unsigned long __init prom_init(unsigned long r3, unsigned long r4, unsigned long pp,
@@ -122,9 +120,6 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,void __init early_setup (unsigned long dt_ptr);void early_setup_secondary (void ); -/* time */ -void accumulate_stolen_time (void ); -/* misc runtime */ extern u64 __bswapdi2 (u64 );extern s64 __lshrdi3 (s64 ,int );
arch/powerpc/include/asm/book3s/64/kup-radix.h +13 -1
@@ -3,6 +3,7 @@#define _ASM_POWERPC_BOOK3S_64_KUP_RADIX_H#include <linux/const.h> +#include <asm/reg.h> #define AMR_KUAP_BLOCK_READ UL(0x4000000000000000)#define AMR_KUAP_BLOCK_WRITE UL(0x8000000000000000)
@@ -56,7 +57,14 @@#ifdef CONFIG_PPC_KUAP -#include <asm/reg.h> +#include <asm/mmu.h> +#include <asm/ptrace.h> + +static inline void kuap_check_amr (void ) +{ +if (IS_ENABLED (CONFIG_PPC_KUAP_DEBUG ) &&mmu_has_feature (MMU_FTR_RADIX_KUAP )) +WARN_ON_ONCE (mfspr (SPRN_AMR ) !=AMR_KUAP_BLOCKED ); +}/* * We support individually allowing read or write, but we don't support nesting
@@ -127,6 +135,10 @@ bad_kuap_fault(struct pt_regs *regs, unsigned long address, bool is_write) (regs->kuap & (is_write ?AMR_KUAP_BLOCK_WRITE :AMR_KUAP_BLOCK_READ )), "Bug: %s fault blocked by AMR!", is_write ? "Write" : "Read"); } +#else /* CONFIG_PPC_KUAP */ +static inline void kuap_check_amr (void ) +{ +}#endif /* CONFIG_PPC_KUAP */ #endif /* __ASSEMBLY__ */
arch/powerpc/include/asm/cputime.h +29 -0
@@ -44,6 +44,28 @@ static inline unsigned long cputime_to_usecs(const cputime_t ct)#ifdef CONFIG_PPC64 #define get_accounting (tsk) (&get_paca()->accounting)static inline void arch_vtime_task_switch (struct task_struct *tsk) { } + +/* +* account_cpu_user_entry/exit runs "unreconciled", so can't trace, +* can't use use get_paca() +*/ +static notrace inline void account_cpu_user_entry(void) +{ +unsigned long tb =mftb (); +struct cpu_accounting_data *acct = &local_paca->accounting; + + acct->utime += (tb - acct->starttime_user); + acct->starttime = tb; +} +static notrace inline void account_cpu_user_exit(void) +{ +unsigned long tb =mftb (); +struct cpu_accounting_data *acct = &local_paca->accounting; + + acct->stime += (tb - acct->starttime); + acct->starttime_user = tb; +} +#else #define get_accounting (tsk) (&task_thread_info(tsk)->accounting)/*
@@ -61,5 +83,12 @@ static inline void arch_vtime_task_switch(struct task_struct *prev)#endif #endif /* __KERNEL__ */ +#else /* CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */ +static inline void account_cpu_user_entry (void ) +{ +} +static inline void account_cpu_user_exit (void ) +{ +}#endif /* CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */ #endif /* __POWERPC_CPUTIME_H */
arch/powerpc/include/asm/hw_irq.h +4 -0
@@ -228,9 +228,13 @@ static inline bool arch_irqs_disabled(void)#ifdef CONFIG_PPC_BOOK3E #define __hard_irq_enable () wrtee(MSR_EE)#define __hard_irq_disable () wrtee(0) +#define __hard_EE_RI_disable () wrtee(0) +#define __hard_RI_enable () do { } while (0)#else #define __hard_irq_enable () __mtmsrd(MSR_EE|MSR_RI, 1)#define __hard_irq_disable () __mtmsrd(MSR_RI, 1) +#define __hard_EE_RI_disable () __mtmsrd(0, 1) +#define __hard_RI_enable () __mtmsrd(MSR_RI, 1)#endif #define hard_irq_disable () do { \
arch/powerpc/include/asm/ptrace.h +3 -0
@@ -138,6 +138,9 @@ extern unsigned long profile_pc(struct pt_regs *regs);#define profile_pc (regs) instruction_pointer(regs)#endif +long do_syscall_trace_enter (struct pt_regs *regs); +void do_syscall_trace_leave (struct pt_regs *regs); +#define kernel_stack_pointer (regs) ((regs)->gpr[1])static inline int is_syscall_success (struct pt_regs *regs) {
arch/powerpc/include/asm/signal.h +3 -0
@@ -6,4 +6,7 @@#include <uapi/asm/signal.h> #include <uapi/asm/ptrace.h> +struct pt_regs ; +void do_notify_resume (struct pt_regs *regs,unsigned long thread_info_flags); +#endif /* _ASM_POWERPC_SIGNAL_H */
arch/powerpc/include/asm/switch_to.h +5 -0
@@ -5,6 +5,7 @@#ifndef _ASM_POWERPC_SWITCH_TO_H#define _ASM_POWERPC_SWITCH_TO_H +#include <linux/sched.h> #include <asm/reg.h> struct thread_struct ;
@@ -22,6 +23,10 @@ extern void switch_booke_debug_regs(struct debug_reg *new_debug);extern int emulate_altivec (struct pt_regs *); +void restore_math (struct pt_regs *regs); + +void restore_tm_state (struct pt_regs *regs); +extern void flush_all_to_thread (struct task_struct *);extern void giveup_all (struct task_struct *);
arch/powerpc/include/asm/time.h +3 -0
@@ -194,5 +194,8 @@ DECLARE_PER_CPU(u64, decrementers_next_tb);/* Convert timebase ticks to nanoseconds */ unsigned long long tb_to_ns (unsigned long long tb_ticks); +/* SPLPAR */ +void accumulate_stolen_time (void ); +#endif /* __KERNEL__ */ #endif /* __POWERPC_TIME_H */
arch/powerpc/kernel/Makefile +2 -1
@@ -50,7 +50,8 @@ obj-y := cputable.o ptrace.o syscalls.o \ of_platform.o prom_parse.o obj-$(CONFIG_PPC64) += setup_64.o sys_ppc32.o \ signal_64.o ptrace32.o \ - paca.o nvram_64.o firmware.o note.o + paca.o nvram_64.o firmware.o note.o \ + syscall_64.o obj-$(CONFIG_VDSO32) += vdso32/ obj-$(CONFIG_PPC_WATCHDOG) += watchdog.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o
arch/powerpc/kernel/entry_64.S +44 -294
413 lines not on this page · raw message
arch/powerpc/kernel/signal.h +0 -2
@@ -10,8 +10,6 @@#ifndef _POWERPC_ARCH_SIGNAL_H#define _POWERPC_ARCH_SIGNAL_H -extern void do_notify_resume (struct pt_regs *regs,unsigned long thread_info_flags); -extern void __user *get_sigframe (struct ksignal *ksig,unsigned long sp,size_t frame_size,int is_32);
arch/powerpc/kernel/syscall_64.c new file +213 -0
213 lines not on this page · raw message
arch/powerpc/kernel/systbl.S +3 -6
@@ -16,25 +16,22 @@ #ifdef CONFIG_PPC64 .p2align 3 +#define __SYSCALL(nr, entry) .8byte entry +#else +#define __SYSCALL(nr, entry) .long entry #endif .globl sys_call_table sys_call_table: #ifdef CONFIG_PPC64 -#define __SYSCALL(nr, entry) .8byte DOTSYM(entry) #include <asm/syscall_table_64.h> -#undef __SYSCALL #else -#define __SYSCALL(nr, entry) .long entry #include <asm/syscall_table_32.h> -#undef __SYSCALL #endif #ifdef CONFIG_COMPAT .globl compat_sys_call_table compat_sys_call_table: #define compat_sys_sigsuspend sys_sigsuspend -#define __SYSCALL(nr, entry) .8byte DOTSYM(entry) #include <asm/syscall_table_c32.h> -#undef __SYSCALL #endif
Signature · 1 line
2.23.0
Re: [PATCH v3 25/32] powerpc/64: system call implement entry/exit logic in C
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 48 lines
System call entry and particularly exit code is beyond the limit of what is reasonable to implement in asm.
This conversion moves all conditional branches out of the asm code, except for the case that all GPRs should be restored at exit.
Null syscall test is about 5% faster after this patch, because the exit work is handled under
local_irq_disable, and the hard mask and pending interrupt replay is handled after that, which avoids games with MSR.Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michal Suchanek <[email protected]>---
v2,rebase (from Michal):
- Add endian conversion fordtl_idx(ms)
- Fix sparse warning about missing declaration (ms)
- Addunistd.hto fix some defconfigs, add SPDX, minor formatting (mpe)v3: Fixes thanks to reports from mpe and selftests errors:
- Several soft-mask debug and unsafesmp_processor_id()warnings due to tracing and other false positives due to checks in "unreconciled" code.
- Fix a bug with syscall tracing functions that set registers (e.g.,PTRACE_SETREG) not setting GPRs properly.
- Fix sillytabort_syscallbug that causes kernel crashes when making system calls in transactional state.arch/powerpc/include/asm/asm-prototypes.h | 17 +- .../powerpc/include/asm/book3s/64/kup-radix.h | 14 +- arch/powerpc/include/asm/cputime.h | 29 ++ arch/powerpc/include/asm/hw_irq.h | 4 + arch/powerpc/include/asm/ptrace.h | 3 + arch/powerpc/include/asm/signal.h | 3 + arch/powerpc/include/asm/switch_to.h | 5 + arch/powerpc/include/asm/time.h | 3 + arch/powerpc/kernel/Makefile | 3 +- arch/powerpc/kernel/entry_64.S | 338 +++--------------- arch/powerpc/kernel/signal.h | 2 - arch/powerpc/kernel/syscall_64.c | 213 +++++++++++ arch/powerpc/kernel/systbl.S | 9 +- 13 files changed, 328 insertions(+), 315 deletions(-)create mode 100644
arch/powerpc/kernel/syscall_64.carch/powerpc/include/asm/asm-prototypes.h +0
-0
arch/powerpc/include/asm/asm-prototypes.h ppc_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, · line 100
Quoted 4 lines
arch/powerpc/include/asm/asm-prototypes.h +1
-0@@ -97,6 +97,12 @@ ppc_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp,unsigned long __init early_init (unsigned long dt_ptr);void __init machine_init (u64 dt_ptr);#endif +#ifdef CONFIG_PPC64
This ifdef is not necessary as it has no pending #else. Having function declaration without definition is not an issue. Keeping in mind that we are aiming at generalising this to PPC32.
Quoted 77 lines
arch/powerpc/include/asm/asm-prototypes.h +5
-11+long system_call_exception (long r3,long r4,long r5,long r6,long r7,long r8,unsigned long r0,struct pt_regs *regs); +notrace unsigned long syscall_exit_prepare (unsigned long r3,struct pt_regs *regs); +notrace unsigned long interrupt_exit_user_prepare (struct pt_regs *regs,unsigned long msr); +notrace unsigned long interrupt_exit_kernel_prepare (struct pt_regs *regs,unsigned long msr); +#endif long ppc_fadvise64_64 (int fd,int advice,u32 offset_high,u32 offset_low,u32 len_high,u32 len_low);@@ -104,14 +110,6 @@ long sys_switch_endian(void); notrace unsigned int __check_irq_replay(void); void notrace restore_interrupts(void); -/* ptrace */ -long do_syscall_trace_enter (struct pt_regs *regs); -void do_syscall_trace_leave (struct pt_regs *regs); - -/* process */ -void restore_math (struct pt_regs *regs); -void restore_tm_state (struct pt_regs *regs); -/* prom_init (OpenFirmware) */ unsigned long __init prom_init(unsigned long r3, unsigned long r4, unsigned long pp,@@ -122,9 +120,6 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,void __init early_setup (unsigned long dt_ptr);void early_setup_secondary (void ); -/* time */ -void accumulate_stolen_time (void ); -/* misc runtime */ extern u64 __bswapdi2 (u64 );extern s64 __lshrdi3 (s64 ,int );arch/powerpc/include/asm/book3s/64/kup-radix.h +13
-1@@ -3,6 +3,7 @@#define _ASM_POWERPC_BOOK3S_64_KUP_RADIX_H#include <linux/const.h> +#include <asm/reg.h> #define AMR_KUAP_BLOCK_READ UL(0x4000000000000000)#define AMR_KUAP_BLOCK_WRITE UL(0x8000000000000000)@@ -56,7 +57,14 @@#ifdef CONFIG_PPC_KUAP -#include <asm/reg.h> +#include <asm/mmu.h> +#include <asm/ptrace.h> + +static inline void kuap_check_amr (void ) +{ +if (IS_ENABLED (CONFIG_PPC_KUAP_DEBUG ) &&mmu_has_feature (MMU_FTR_RADIX_KUAP )) +WARN_ON_ONCE (mfspr (SPRN_AMR ) !=AMR_KUAP_BLOCKED ); +}/* * We support individually allowing read or write, but we don't support nesting @@ -127,6 +135,10 @@ bad_kuap_fault(struct pt_regs *regs, unsigned long address, bool is_write) (regs->kuap & (is_write ?AMR_KUAP_BLOCK_WRITE :AMR_KUAP_BLOCK_READ )), "Bug: %s fault blocked by AMR!", is_write ? "Write" : "Read"); } +#else /* CONFIG_PPC_KUAP */ +static inline void kuap_check_amr (void ) +{ +}#endif /* CONFIG_PPC_KUAP */ #endif /* __ASSEMBLY__ */ arch/powerpc/include/asm/cputime.h +0
-0
arch/powerpc/include/asm/cputime.h static inline unsigned long cputime_to_usecs(const cputime_t ct) · line 46
Quoted 3 lines
arch/powerpc/include/asm/cputime.h +0
-0@@ -44,6 +44,28 @@ static inline unsigned long cputime_to_usecs(const cputime_t ct)#ifdef CONFIG_PPC64 #define get_accounting (tsk) (&get_paca()->accounting)static inline void arch_vtime_task_switch (struct task_struct *tsk) { }
Could we have the below additions sit outside of this PPC64 ifdef, to be reused on PPC32 ?
arch/powerpc/include/asm/cputime.h static inline unsigned long cputime_to_usecs(const cputime_t ct) · line 55
Quoted 9 lines
arch/powerpc/include/asm/cputime.h +9
-0+ +/* +* account_cpu_user_entry/exit runs "unreconciled", so can't trace, +* can't use use get_paca() +*/ +static notrace inline void account_cpu_user_entry(void) +{ +unsigned long tb =mftb (); +struct cpu_accounting_data *acct = &local_paca->accounting;
In the spirit of reusing that code on PPC32, can we use get_accounting() ? Or an alternate version of get_accounting(), eg get_accounting_notrace() to be defined ?
Quoted 123 lines
arch/powerpc/include/asm/cputime.h +20
-0+ + acct->utime += (tb - acct->starttime_user); + acct->starttime = tb; +} +static notrace inline void account_cpu_user_exit(void) +{ +unsigned long tb =mftb (); +struct cpu_accounting_data *acct = &local_paca->accounting; + + acct->stime += (tb - acct->starttime); + acct->starttime_user = tb; +} +#else #define get_accounting (tsk) (&task_thread_info(tsk)->accounting)/* @@ -61,5 +83,12 @@ static inline void arch_vtime_task_switch(struct task_struct *prev)#endif #endif /* __KERNEL__ */ +#else /* CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */ +static inline void account_cpu_user_entry (void ) +{ +} +static inline void account_cpu_user_exit (void ) +{ +}#endif /* CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */ #endif /* __POWERPC_CPUTIME_H */ arch/powerpc/include/asm/hw_irq.h +4
-0@@ -228,9 +228,13 @@ static inline bool arch_irqs_disabled(void)#ifdef CONFIG_PPC_BOOK3E #define __hard_irq_enable () wrtee(MSR_EE)#define __hard_irq_disable () wrtee(0) +#define __hard_EE_RI_disable () wrtee(0) +#define __hard_RI_enable () do { } while (0)#else #define __hard_irq_enable () __mtmsrd(MSR_EE|MSR_RI, 1)#define __hard_irq_disable () __mtmsrd(MSR_RI, 1) +#define __hard_EE_RI_disable () __mtmsrd(0, 1) +#define __hard_RI_enable () __mtmsrd(MSR_RI, 1)#endif #define hard_irq_disable () do { \arch/powerpc/include/asm/ptrace.h +3
-0@@ -138,6 +138,9 @@ extern unsigned long profile_pc(struct pt_regs *regs);#define profile_pc (regs) instruction_pointer(regs)#endif +long do_syscall_trace_enter (struct pt_regs *regs); +void do_syscall_trace_leave (struct pt_regs *regs); +#define kernel_stack_pointer (regs) ((regs)->gpr[1])static inline int is_syscall_success (struct pt_regs *regs) {arch/powerpc/include/asm/signal.h +3
-0@@ -6,4 +6,7 @@#include <uapi/asm/signal.h> #include <uapi/asm/ptrace.h> +struct pt_regs ; +void do_notify_resume (struct pt_regs *regs,unsigned long thread_info_flags); +#endif /* _ASM_POWERPC_SIGNAL_H */ arch/powerpc/include/asm/switch_to.h +5
-0@@ -5,6 +5,7 @@#ifndef _ASM_POWERPC_SWITCH_TO_H#define _ASM_POWERPC_SWITCH_TO_H +#include <linux/sched.h> #include <asm/reg.h> struct thread_struct ;@@ -22,6 +23,10 @@ extern void switch_booke_debug_regs(struct debug_reg *new_debug);extern int emulate_altivec (struct pt_regs *); +void restore_math (struct pt_regs *regs); + +void restore_tm_state (struct pt_regs *regs); +extern void flush_all_to_thread (struct task_struct *);extern void giveup_all (struct task_struct *);arch/powerpc/include/asm/time.h +3
-0@@ -194,5 +194,8 @@ DECLARE_PER_CPU(u64, decrementers_next_tb);/* Convert timebase ticks to nanoseconds */ unsigned long long tb_to_ns (unsigned long long tb_ticks); +/* SPLPAR */ +void accumulate_stolen_time (void ); +#endif /* __KERNEL__ */ #endif /* __POWERPC_TIME_H */ arch/powerpc/kernel/Makefile +2
-1@@ -50,7 +50,8 @@ obj-y := cputable.o ptrace.o syscalls.o \ of_platform.o prom_parse.o obj-$(CONFIG_PPC64) += setup_64.o sys_ppc32.o \ signal_64.o ptrace32.o \ - paca.o nvram_64.o firmware.o note.o + paca.o nvram_64.o firmware.o note.o \ + syscall_64.o obj-$(CONFIG_VDSO32) += vdso32/ obj-$(CONFIG_PPC_WATCHDOG) += watchdog.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o
[snip]
arch/powerpc/kernel/syscall_64.c
Quoted 14 lines
arch/powerpc/kernel/signal.h +0
-2@@ -10,8 +10,6 @@#ifndef _POWERPC_ARCH_SIGNAL_H#define _POWERPC_ARCH_SIGNAL_H -extern void do_notify_resume (struct pt_regs *regs,unsigned long thread_info_flags); -extern void __user *get_sigframe (struct ksignal *ksig,unsigned long sp,size_t frame_size,int is_32);arch/powerpc/kernel/syscall_64.c +0
-0
Could some part of it go in a syscall.c to be reused on PPC32 ?
Quoted 60 lines
new file mode 100644
index 000000000000..20f77cc19df8arch/powerpc/kernel/syscall_64.c new file +55
-0@@ -0,0 +1,213 @@ +// SPDX-License-Identifier: GPL-2.0-or-later + +#include <linux/err.h> +#include <asm/asm-prototypes.h> +#include <asm/book3s/64/kup-radix.h> +#include <asm/cputime.h> +#include <asm/hw_irq.h> +#include <asm/kprobes.h> +#include <asm/paca.h> +#include <asm/ptrace.h> +#include <asm/reg.h> +#include <asm/signal.h> +#include <asm/switch_to.h> +#include <asm/syscall.h> +#include <asm/time.h> +#include <asm/unistd.h> + +extern void __noreturn tabort_syscall (unsigned long nip,unsigned long msr); + +typedef long (*syscall_fn )(long ,long ,long ,long ,long ,long ); + +/* Has to run notrace because it is entered "unreconciled" */ +notrace long system_call_exception (long r3,long r4,long r5,long r6,long r7,long r8, +unsigned long r0,struct pt_regs *regs) +{ +unsigned long ti_flags; +syscall_fn f; + +BUG_ON (!(regs->msr &MSR_PR )); + +if (IS_ENABLED (CONFIG_PPC_TRANSACTIONAL_MEM ) && +unlikely (regs->msr &MSR_TS_T )) +tabort_syscall (regs->nip, regs->msr); + +account_cpu_user_entry (); + +#ifdef CONFIG_PPC_SPLPAR +if (IS_ENABLED (CONFIG_VIRT_CPU_ACCOUNTING_NATIVE ) && +firmware_has_feature (FW_FEATURE_SPLPAR )) { +struct lppaca *lp = local_paca->lppaca_ptr; + +if (unlikely (local_paca->dtl_ridx !=be64_to_cpu (lp->dtl_idx))) +accumulate_stolen_time (); + } +#endif + +kuap_check_amr (); + +/* +* A syscall should always be called with interrupts enabled +* so we just unconditionally hard-enable here. When some kind +* of irq tracing is used, we additionally check that condition +* is correct +*/ +if (IS_ENABLED (CONFIG_PPC_IRQ_SOFT_MASK_DEBUG )) {
arch/powerpc/kernel/syscall_64.c line 65
Quoted 10 lines
arch/powerpc/kernel/syscall_64.c new file +10
-0+WARN_ON (irq_soft_mask_return () !=IRQS_ENABLED ); +WARN_ON (local_paca->irq_happened); + } +/* +* This is not required for the syscall exit path, but makes the +* stack frame look nicer. If this was initialised in the first stack +* frame, or if the unwinder was taught the first stack frame always +* returns to user with IRQS_ENABLED, this store could be avoided! +*/ + regs->softe =IRQS_ENABLED ;
softe doesn't exist on PPC32. Can we do that through a helper ?
arch/powerpc/kernel/syscall_64.c line 67
Quoted 2 lines
arch/powerpc/kernel/syscall_64.c new file +2
-0+ +__hard_irq_enable ();
This doesn't exist on PPC32. Should we define __hard_irq_enable() as arch_local_irq_enable() on PPC32 ?
Quoted 18 lines
arch/powerpc/kernel/syscall_64.c new file +18
-0+ + ti_flags =current_thread_info ()->flags; +if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) { +/* +* We use the return value of do_syscall_trace_enter() as the +* syscall number. If the syscall was rejected for any reason +* do_syscall_trace_enter() returns an invalid syscall number +* and the test against NR_syscalls will fail and the return +* value to be used is in regs->gpr[3]. +*/ + r0 =do_syscall_trace_enter (regs); +if (unlikely (r0 >= NR_syscalls)) +return regs->gpr[3 ]; + r3 = regs->gpr[3 ]; + r4 = regs->gpr[4 ]; + r5 = regs->gpr[5 ]; + r6 = regs->gpr[6 ]; + r7 = regs->gpr[7 ];
arch/powerpc/kernel/syscall_64.c line 95
Quoted 10 lines
arch/powerpc/kernel/syscall_64.c new file +10
-0+ r8 = regs->gpr[8 ]; + + }else if (unlikely (r0 >= NR_syscalls)) { +return -ENOSYS ; + } + +/* May be faster to do array_index_nospec? */ +barrier_nospec (); + +if (unlikely (ti_flags & _TIF_32BIT)) {
Use is_compat_task() instead ?
Quoted 142 lines
arch/powerpc/kernel/syscall_64.c new file +118
-0+ f = (void *)compat_sys_call_table[r0]; + + r3 &=0x00000000ffffffffULL ; + r4 &=0x00000000ffffffffULL ; + r5 &=0x00000000ffffffffULL ; + r6 &=0x00000000ffffffffULL ; + r7 &=0x00000000ffffffffULL ; + r8 &=0x00000000ffffffffULL ; + + }else { + f = (void *)sys_call_table[r0]; + } + +return f (r3, r4, r5, r6, r7, r8); +} + +/* +* This should be called after a syscall returns, with r3 the return value +* from the syscall. If this function returns non-zero, the system call +* exit assembly should additionally load all GPR registers and CTR and XER +* from the interrupt frame. +* +* The function graph tracer can not trace the return side of this function, +* because RI=0 and soft mask state is "unreconciled", so it is marked notrace. +*/ +notrace unsigned long syscall_exit_prepare (unsigned long r3, +struct pt_regs *regs) +{ +unsigned long *ti_flagsp = ¤t_thread_info ()->flags; +unsigned long ti_flags; +unsigned long ret =0 ; + + regs->result = r3; + +/* Check whether the syscall is issued inside a restartable sequence */ +rseq_syscall (regs); + + ti_flags = *ti_flagsp; + +if (unlikely (r3 >= (unsigned long )-MAX_ERRNO )) { +if (likely (!(ti_flags & (_TIF_NOERROR | _TIF_RESTOREALL)))) { + r3 = -r3; + regs->ccr |=0x10000000 ;/* Set SO bit in CR */ + } + } + +if (unlikely (ti_flags & _TIF_PERSYSCALL_MASK)) { +if (ti_flags & _TIF_RESTOREALL) + ret = _TIF_RESTOREALL; +else + regs->gpr[3 ] = r3; +clear_bits (_TIF_PERSYSCALL_MASK, ti_flagsp); + }else { + regs->gpr[3 ] = r3; + } + +if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) { +do_syscall_trace_leave (regs); + ret |= _TIF_RESTOREALL; + } + +again: +local_irq_disable (); + ti_flags =READ_ONCE (*ti_flagsp); +while (unlikely (ti_flags & _TIF_USER_WORK_MASK)) { +local_irq_enable (); +if (ti_flags & _TIF_NEED_RESCHED) { +schedule (); + }else { +/* +* SIGPENDING must restore signal handler function +* argument GPRs, and some non-volatiles (e.g., r1). +* Restore all for now. This could be made lighter. +*/ +if (ti_flags & _TIF_SIGPENDING) + ret |= _TIF_RESTOREALL; +do_notify_resume (regs, ti_flags); + } +local_irq_disable (); + ti_flags =READ_ONCE (*ti_flagsp); + } + +if (IS_ENABLED (CONFIG_PPC_BOOK3S ) &&IS_ENABLED (CONFIG_PPC_FPU )) { +unsigned long mathflags =MSR_FP ; + +if (IS_ENABLED (CONFIG_ALTIVEC )) + mathflags |=MSR_VEC ; + +if ((regs->msr & mathflags) != mathflags) +restore_math (regs); + } + +/* This must be done with RI=1 because tracing may touch vmaps */ +trace_hardirqs_on (); + +/* This pattern matches prep_irq_for_idle */ +__hard_EE_RI_disable (); +if (unlikely (lazy_irq_pending ())) { +__hard_RI_enable (); +trace_hardirqs_off (); + local_paca->irq_happened |=PACA_IRQ_HARD_DIS ; +local_irq_enable (); +/* Took an interrupt which may have more exit work to do. */ + goto again; + } + local_paca->irq_happened =0 ; +irq_soft_mask_set (IRQS_ENABLED ); + +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif + +kuap_check_amr (); + +account_cpu_user_exit (); + +return ret; +}arch/powerpc/kernel/systbl.S +3
-4@@ -16,25 +16,22 @@ #ifdef CONFIG_PPC64 .p2align 3 +#define __SYSCALL(nr, entry) .8byte entry +#else +#define __SYSCALL(nr, entry) .long entry #endif .globl sys_call_table sys_call_table: #ifdef CONFIG_PPC64 -#define __SYSCALL(nr, entry) .8byte DOTSYM(entry) #include <asm/syscall_table_64.h> -#undef __SYSCALL #else -#define __SYSCALL(nr, entry) .long entry #include <asm/syscall_table_32.h> -#undef __SYSCALL
arch/powerpc/kernel/systbl.S line 37
Quoted 10 lines
arch/powerpc/kernel/systbl.S +0
-2#endif #ifdef CONFIG_COMPAT .globl compat_sys_call_table compat_sys_call_table: #define compat_sys_sigsuspend sys_sigsuspend -#define __SYSCALL(nr, entry) .8byte DOTSYM(entry) #include <asm/syscall_table_c32.h> -#undef __SYSCALL #endif
Christophe
Nicholas Piggin
reviewerRe: [PATCH v3 25/32] powerpc/64: system call implement entry/exit logic in C
Christophe Leroy's on March 19, 2020 7:18 pm:
Quoted 46 lines
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 43 lines
System call entry and particularly exit code is beyond the limit of what is reasonable to implement in asm.
This conversion moves all conditional branches out of the asm code, except for the case that all GPRs should be restored at exit.
Null syscall test is about 5% faster after this patch, because the exit work is handled under
local_irq_disable, and the hard mask and pending interrupt replay is handled after that, which avoids games with MSR.Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michal Suchanek <[email protected]>---
v2,rebase (from Michal):
- Add endian conversion fordtl_idx(ms)
- Fix sparse warning about missing declaration (ms)
- Addunistd.hto fix some defconfigs, add SPDX, minor formatting (mpe)v3: Fixes thanks to reports from mpe and selftests errors:
- Several soft-mask debug and unsafesmp_processor_id()warnings due to tracing and other false positives due to checks in "unreconciled" code.
- Fix a bug with syscall tracing functions that set registers (e.g.,PTRACE_SETREG) not setting GPRs properly.
- Fix sillytabort_syscallbug that causes kernel crashes when making system calls in transactional state.arch/powerpc/include/asm/asm-prototypes.h | 17 +- .../powerpc/include/asm/book3s/64/kup-radix.h | 14 +- arch/powerpc/include/asm/cputime.h | 29 ++ arch/powerpc/include/asm/hw_irq.h | 4 + arch/powerpc/include/asm/ptrace.h | 3 + arch/powerpc/include/asm/signal.h | 3 + arch/powerpc/include/asm/switch_to.h | 5 + arch/powerpc/include/asm/time.h | 3 + arch/powerpc/kernel/Makefile | 3 +- arch/powerpc/kernel/entry_64.S | 338 +++--------------- arch/powerpc/kernel/signal.h | 2 - arch/powerpc/kernel/syscall_64.c | 213 +++++++++++ arch/powerpc/kernel/systbl.S | 9 +- 13 files changed, 328 insertions(+), 315 deletions(-)create mode 100644
arch/powerpc/kernel/syscall_64.c
Quoted 13 lines
arch/powerpc/include/asm/asm-prototypes.h ppc_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, · line 100
Quoted 9 lines
arch/powerpc/include/asm/asm-prototypes.h +1
-0@@ -97,6 +97,12 @@ ppc_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp,unsigned long __init early_init (unsigned long dt_ptr);void __init machine_init (u64 dt_ptr);#endif +#ifdef CONFIG_PPC64 This ifdef is not necessary as it has no pending #else. Having function declaration without definition is not an issue. Keeping in mind that we are aiming at generalising this to PPC32.
Well there's other unnecessary ifdefs in there too I think. But sure. This patch also got the interrupt_exit_ prototypes leaked in from the later patch so I could fix those.
Quoted 11 lines
arch/powerpc/include/asm/cputime.h static inline unsigned long cputime_to_usecs(const cputime_t ct) · line 46
Quoted 8 lines
arch/powerpc/include/asm/cputime.h +0
-0@@ -44,6 +44,28 @@ static inline unsigned long cputime_to_usecs(const cputime_t ct)#ifdef CONFIG_PPC64 #define get_accounting (tsk) (&get_paca()->accounting)static inline void arch_vtime_task_switch (struct task_struct *tsk) { }Could we have the below additions sit outside of this PPC64 ifdef, to be reused on PPC32 ?
Okay.
Quoted 5 lines
Quoted 5 lines
arch/powerpc/include/asm/cputime.h +5
-0+ +/* +* account_cpu_user_entry/exit runs "unreconciled", so can't trace, +* can't use use get_paca() +*/
Quoted 8 lines
arch/powerpc/include/asm/cputime.h static inline unsigned long cputime_to_usecs(const cputime_t ct) · line 55
Quoted 4 lines
arch/powerpc/include/asm/cputime.h +4
-0+static notrace inline void account_cpu_user_entry(void) +{ +unsigned long tb =mftb (); +struct cpu_accounting_data *acct = &local_paca->accounting;In the spirit of reusing that code on PPC32, can we use
get_accounting()? Or an alternate version ofget_accounting(), egget_accounting_notrace()to be defined ?
Okay.
Quoted 3 lines
arch/powerpc/kernel/syscall_64.c
Quoted 1 line
arch/powerpc/kernel/syscall_64.c +0
-0Could some part of it go in a
syscall.cto be reused on PPC32 ?
I could put it all in syscall.c and then we can adjust with some ifdefs or helpers. I don't think there is enough to be worth syscall.c, syscall_32.c, and syscall_64.c.
I wonder about the interrupt returns as well, that doesn't really make sense in a file called syscall.c, but the code is very similar to system call exit. Should we just call it interrupts.c?
Quoted 9 lines
arch/powerpc/kernel/syscall_64.c
Quoted 7 lines
arch/powerpc/kernel/syscall_64.c +7
-0+/* +* This is not required for the syscall exit path, but makes the +* stack frame look nicer. If this was initialised in the first stack +* frame, or if the unwinder was taught the first stack frame always +* returns to user with IRQS_ENABLED, this store could be avoided! +*/ + regs->softe =IRQS_ENABLED ;softe doesn't exist on PPC32. Can we do that through a helper ?
I guess, we can have regs_set_irq_state(regs, IRQS_ENABLED); or something like that.
We make that helper and a _get_ counterpart in a later patch which covers other cases in the tree as well.
Quoted 5 lines
arch/powerpc/kernel/syscall_64.c
Quoted 2 lines
arch/powerpc/kernel/syscall_64.c +2
-0+ +__hard_irq_enable ();This doesn't exist on PPC32. Should we define
__hard_irq_enable()asarch_local_irq_enable()on PPC32 ?
This goes away with patch 29. Better not to have this ugly thing spill into ppc32 code at all if we can avoid it :)
Quoted 23 lines
Quoted 23 lines
arch/powerpc/kernel/syscall_64.c +22
-0+ + ti_flags =current_thread_info ()->flags; +if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) { +/* +* We use the return value of do_syscall_trace_enter() as the +* syscall number. If the syscall was rejected for any reason +* do_syscall_trace_enter() returns an invalid syscall number +* and the test against NR_syscalls will fail and the return +* value to be used is in regs->gpr[3]. +*/ + r0 =do_syscall_trace_enter (regs); +if (unlikely (r0 >= NR_syscalls)) +return regs->gpr[3 ]; + r3 = regs->gpr[3 ]; + r4 = regs->gpr[4 ]; + r5 = regs->gpr[5 ]; + r6 = regs->gpr[6 ]; + r7 = regs->gpr[7 ]; + r8 = regs->gpr[8 ]; + + }else if (unlikely (r0 >= NR_syscalls)) { +return -ENOSYS ;
Quoted 8 lines
arch/powerpc/kernel/syscall_64.c
Quoted 6 lines
arch/powerpc/kernel/syscall_64.c +6
-0+ } + +/* May be faster to do array_index_nospec? */ +barrier_nospec (); + +if (unlikely (ti_flags & _TIF_32BIT)) {Use
is_compat_task()instead ?
Michal pointed this out, he's got patches that do this on top of this series.
Incremental diff for your suggestions below. Now there is likely we're going to have a few ifdefs, particularly in the exit paths where we have complexity handling irq soft masked state where helpers dont make much sense. I don't think that will be such a bad thing, but we can come to it as we go.
Thanks,
Nick
--- arch/powerpc/include/asm/asm-prototypes.h | 4 --- arch/powerpc/include/asm/cputime.h | 38 +++++++++++++---------- 2 files changed, 21 insertions(+), 21 deletions(-)
arch/powerpc/include/asm/asm-prototypes.h +0 -4
@@ -97,12 +97,8 @@ ppc_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp,unsigned long __init early_init (unsigned long dt_ptr);void __init machine_init (u64 dt_ptr);#endif -#ifdef CONFIG_PPC64 long system_call_exception (long r3,long r4,long r5,long r6,long r7,long r8,unsigned long r0,struct pt_regs *regs);notrace unsigned long syscall_exit_prepare (unsigned long r3,struct pt_regs *regs); -notrace unsigned long interrupt_exit_user_prepare (struct pt_regs *regs,unsigned long msr); -notrace unsigned long interrupt_exit_kernel_prepare (struct pt_regs *regs,unsigned long msr); -#endif long ppc_fadvise64_64 (int fd,int advice,u32 offset_high,u32 offset_low,u32 len_high,u32 len_low);
arch/powerpc/include/asm/cputime.h +21 -17
@@ -43,8 +43,26 @@ static inline unsigned long cputime_to_usecs(const cputime_t ct)*/ #ifdef CONFIG_PPC64 #define get_accounting (tsk) (&get_paca()->accounting) +#define raw_get_accounting (tsk) (&local_paca->accounting)static inline void arch_vtime_task_switch (struct task_struct *tsk) { } +#else +#define get_accounting (tsk) (&task_thread_info(tsk)->accounting) +#define raw_get_accounting (tsk) get_accounting(tsk) +/* +* Called from the context switch with interrupts disabled, to charge all +* accumulated times to the current process, and to prepare accounting on +* the next process. +*/ +static inline void arch_vtime_task_switch (struct task_struct *prev) +{ +struct cpu_accounting_data *acct =get_accounting (current); +struct cpu_accounting_data *acct0 =get_accounting (prev); + + acct->starttime = acct0->starttime; +} +#endif +/* * account_cpu_user_entry/exit runs "unreconciled", so can't trace, * can't use use get_paca()
@@ -52,35 +70,21 @@ static inline void arch_vtime_task_switch(struct task_struct *tsk) { } static notrace inline void account_cpu_user_entry(void) {unsigned long tb =mftb (); -struct cpu_accounting_data *acct = &local_paca->accounting; +struct cpu_accounting_data *acct =raw_get_accounting (current); acct->utime += (tb - acct->starttime_user); acct->starttime = tb; } + static notrace inline void account_cpu_user_exit(void) {unsigned long tb =mftb (); -struct cpu_accounting_data *acct = &local_paca->accounting; +struct cpu_accounting_data *acct =raw_get_accounting (current); acct->stime += (tb - acct->starttime); acct->starttime_user = tb; } -#else -#define get_accounting (tsk) (&task_thread_info(tsk)->accounting) -/* -* Called from the context switch with interrupts disabled, to charge all -* accumulated times to the current process, and to prepare accounting on -* the next process. -*/ -static inline void arch_vtime_task_switch (struct task_struct *prev) -{ -struct cpu_accounting_data *acct =get_accounting (current); -struct cpu_accounting_data *acct0 =get_accounting (prev); - - acct->starttime = acct0->starttime; -} -#endif #endif /* __KERNEL__ */ #else /* CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 26/32 powerpc/64: system call zero volatile registers when returning
In reply to Nicholas PigginKernel addresses and potentially other sensitive data could be leaked in volatile registers after a syscall.
--- arch/powerpc/kernel/entry_64.S | 12 ++++++++++++ 1 file changed, 12 insertions(+)
arch/powerpc/kernel/entry_64.S +12 -0
@@ -135,6 +135,18 @@ END_FTR_SECTION_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS) cmpdi r3,0 bne .Lsyscall_restore_regs + li r0,0 + li r4,0 + li r5,0 + li r6,0 + li r7,0 + li r8,0 + li r9,0 + li r10,0 + li r11,0 + li r12,0 + mtctr r0 + mtspr SPRN_XER,r0 .Lsyscall_restore_regs_cont: BEGIN_FTR_SECTION
Signature · 1 line
2.23.0
Re: [PATCH v3 26/32] powerpc/64: system call zero volatile registers when returning
Hi!
Quoted 8 lines · Nicholas Piggin
Kernel addresses and potentially other sensitive data could be leaked in volatile registers after a syscall.
arch/powerpc/kernel/entry_64.S +3
-0cmpdi r3,0 bne .Lsyscall_restore_regs + li r0,0 + li r4,0 + li r5,0
arch/powerpc/kernel/entry_64.S END_FTR_SECTION_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS) · line 150
Quoted 10 lines
arch/powerpc/kernel/entry_64.S +9
-0+ li r6,0 + li r7,0 + li r8,0 + li r9,0 + li r10,0 + li r11,0 + li r12,0 + mtctr r0 + mtspr SPRN_XER,r0 .Lsyscall_restore_regs_cont:
What about LR? Is that taken care of later?
This also deserves a big fat comment imo, it is very important after all, and not so obvious.
Segher
Nicholas Piggin
reviewerRe: [PATCH v3 26/32] powerpc/64: system call zero volatile registers when returning
Segher Boessenkool's on February 26, 2020 7:20 am:
Quoted 15 lines
Hi!
Quoted 12 lines · Nicholas Piggin
Kernel addresses and potentially other sensitive data could be leaked in volatile registers after a syscall.
arch/powerpc/kernel/entry_64.S +7
-0cmpdi r3,0 bne .Lsyscall_restore_regs + li r0,0 + li r4,0 + li r5,0 + li r6,0 + li r7,0 + li r8,0 + li r9,0
Quoted 8 lines
arch/powerpc/kernel/entry_64.S END_FTR_SECTION_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS) · line 150
Quoted 6 lines
arch/powerpc/kernel/entry_64.S +5
-0+ li r10,0 + li r11,0 + li r12,0 + mtctr r0 + mtspr SPRN_XER,r0 .Lsyscall_restore_regs_cont:What about LR? Is that taken care of later?
LR is preserved by sc as per ABI.
Quoted 2 lines
This also deserves a big fat comment imo, it is very important after all, and not so obvious.
Sure I can add something.
Thanks,
Nick
Nicholas Piggin
reviewerPATCH 1/1 Fix powerpc/64: system call zero volatile registers when returning
In reply to Segher BoessenkoolHere's an incremental fix that can be folded into the patch.
Segher Boessenkool's on February 26, 2020 7:20 am:
Quoted 24 lines
Hi!
Quoted 8 lines · Nicholas Piggin
Kernel addresses and potentially other sensitive data could be leaked in volatile registers after a syscall.
(quoted diff) +3
-0cmpdi r3,0 bne .Lsyscall_restore_regs + li r0,0 + li r4,0 + li r5,0Review
Quoted 10 lines
(quoted diff) +9
-0+ li r6,0 + li r7,0 + li r8,0 + li r9,0 + li r10,0 + li r11,0 + li r12,0 + mtctr r0 + mtspr SPRN_XER,r0 .Lsyscall_restore_regs_cont:What about LR? Is that taken care of later?
Quoted 6 lines
This also deserves a big fat comment imo, it is very important after all, and not so obvious.
Segher
--- arch/powerpc/kernel/entry_64.S | 1 + 1 file changed, 1 insertion(+)
arch/powerpc/kernel/entry_64.S +1 -0
@@ -135,6 +135,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS) cmpdi r3,0 bne .Lsyscall_restore_regs + /* Zero volatile regs that may contain sensitive kernel data */ li r0,0 li r4,0 li r5,0
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 27/32 powerpc/64: implement soft interrupt replay in C
In reply to Nicholas PigginWhen local_irq_enable() finds a pending soft-masked interrupt, it "replays" it by setting up registers like the initial interrupt entry, then calls into the low level handler to set up an interrupt stack frame and process the interrupt.
This is not necessary, and uses more stack than needed. The high level interrupt handler can be called directly from C, with just pt_regs set up on stack. This should be faster and use less stack.
--- arch/powerpc/include/asm/hw_irq.h | 1 - arch/powerpc/kernel/exceptions-64e.S | 32 ------ arch/powerpc/kernel/exceptions-64s.S | 47 -------- arch/powerpc/kernel/irq.c | 165 +++++++++++++++++++++------ 4 files changed, 130 insertions(+), 115 deletions(-)
arch/powerpc/include/asm/hw_irq.h +0 -1
@@ -52,7 +52,6 @@#ifndef __ASSEMBLY__extern void replay_system_reset (void ); -extern void __replay_interrupt (unsigned int vector);extern void timer_interrupt (struct pt_regs *);extern void timer_broadcast_interrupt (void );
arch/powerpc/kernel/exceptions-64e.S +0 -32
@@ -1002,38 +1002,6 @@ masked_interrupt_book3e_0x280: masked_interrupt_book3e_0x2c0: masked_interrupt_book3e PACA_IRQ_DBELL 0 -/* - * Called from arch_local_irq_enable when an interrupt needs - * to be resent. r3 contains either 0x500,0x900,0x260 or 0x280 - * to indicate the kind of interrupt. MSR:EE is already off. - * We generate a stackframe like if a real interrupt had happened. - * - * Note: While MSR:EE is off, we need to make sure that _MSR - * in the generated frame has EE set to 1 or the exception - * handler will not properly re-enable them. - */ -_GLOBAL(__replay_interrupt) - /* We are going to jump to the exception common code which - * will retrieve various register values from the PACA which - * we don't give a damn about. - */ - mflr r10 - mfmsr r11 - mfcr r4 - mtspr SPRN_SPRG_GEN_SCRATCH,r13; - std r1,PACA_EXGEN+EX_R1(r13); - stw r4,PACA_EXGEN+EX_CR(r13); - ori r11,r11,MSR_EE - subi r1,r1,INT_FRAME_SIZE; - cmpwi cr0,r3,0x500 - beq exc_0x500_common - cmpwi cr0,r3,0x900 - beq exc_0x900_common - cmpwi cr0,r3,0x280 - beq exc_0x280_common - blr - - /* * This is called from 0x300 and 0x400 handlers after the prologs with * r14 and r15 containing the fault address and error code, with the
arch/powerpc/kernel/exceptions-64s.S +0 -47
50 lines not on this page · raw message
arch/powerpc/kernel/irq.c +130 -35
205 lines not on this page · raw message
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 28/32 powerpc/64s: interrupt implement exit logic in C
In reply to Nicholas PigginImplement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
The stack store emulation is significantly simplfied, rather than creating a new return frame and switching to that before performing the store, it uses the PACA to keep a scratch register around to perform thestore.
The asm return code is moved into 64e for now. The new logic has made allowance for 64e, but I don't have a full environment that works well to test it, and even booting in emulated qemu is not great for stress testing. 64e shouldn't be too far off working with this, given a bit more testing and auditing of the logic.
This is slightly faster on a POWER9 (page fault speed increases about 1.1%), probably due to reduced mtmsrd.
---
v2,rebase (from Michal):
- Move the FP restore functions to restore_math. They are not used anywhere else and when restore_math is not built gcc warns about them being unused (ms)
- Add asm/context_tracking.h include to exceptions-64e.S for SCHEDULE_USER definition
v3:
- Fix return from interrupt replay problem by replaying interrupts rather than enabling irqs. This ends up being cleaner and __check_irq_replay goes away completely for 64s. Should bring 64e up to speed and kill a lot of cruft after it's proven on 64s.
- Don't use _GLOBAL if it's not called from C
- Simplify stack store emulation code further, add a bit more commenting.
- Some missing no probe annotations
.../powerpc/include/asm/book3s/64/kup-radix.h | 10 + arch/powerpc/include/asm/hw_irq.h | 1 + arch/powerpc/include/asm/switch_to.h | 6 + arch/powerpc/kernel/entry_64.S | 486 +++++------------- arch/powerpc/kernel/exceptions-64e.S | 255 ++++++++- arch/powerpc/kernel/exceptions-64s.S | 119 ++--- arch/powerpc/kernel/irq.c | 36 +- arch/powerpc/kernel/process.c | 89 ++-- arch/powerpc/kernel/syscall_64.c | 164 +++++- arch/powerpc/kernel/vector.S | 2 +- 10 files changed, 642 insertions(+), 526 deletions(-)
arch/powerpc/include/asm/book3s/64/kup-radix.h +10 -0
@@ -60,6 +60,12 @@#include <asm/mmu.h> #include <asm/ptrace.h> +static inline void kuap_restore_amr (struct pt_regs *regs) +{ +if (mmu_has_feature (MMU_FTR_RADIX_KUAP )) +mtspr (SPRN_AMR , regs->kuap); +} +static inline void kuap_check_amr (void ) {if (IS_ENABLED (CONFIG_PPC_KUAP_DEBUG ) &&mmu_has_feature (MMU_FTR_RADIX_KUAP ))
@@ -136,6 +142,10 @@ bad_kuap_fault(struct pt_regs *regs, unsigned long address, bool is_write) "Bug: %s fault blocked by AMR!", is_write ? "Write" : "Read"); }#else /* CONFIG_PPC_KUAP */ +static inline void kuap_restore_amr (struct pt_regs *regs) +{ +} +static inline void kuap_check_amr (void ) { }
arch/powerpc/include/asm/hw_irq.h +1 -0
@@ -52,6 +52,7 @@#ifndef __ASSEMBLY__extern void replay_system_reset (void ); +extern void replay_soft_interrupts (void );extern void timer_interrupt (struct pt_regs *);extern void timer_broadcast_interrupt (void );
arch/powerpc/include/asm/switch_to.h +6 -0
@@ -23,7 +23,13 @@ extern void switch_booke_debug_regs(struct debug_reg *new_debug);extern int emulate_altivec (struct pt_regs *); +#ifdef CONFIG_PPC_BOOK3S_64 void restore_math (struct pt_regs *regs); +#else +static inline void restore_math (struct pt_regs *regs) +{ +} +#endif void restore_tm_state (struct pt_regs *regs);
arch/powerpc/kernel/entry_64.S +116 -370
549 lines not on this page · raw message
arch/powerpc/kernel/exceptions-64e.S +248 -7
277 lines not on this page · raw message
arch/powerpc/kernel/exceptions-64s.S +45 -74
449 lines not on this page · raw message
arch/powerpc/kernel/irq.c +4 -32
@@ -110,6 +110,8 @@ static inline notrace int decrementer_check_overflow(void)return now >= *next_tb; } +#ifdef CONFIG_PPC_BOOK3E +/* This is called whenever we are re-enabling interrupts * and returns either 0 (nothing to do) or 500/900/280/a00/e80 if * there's an EE, DEC or DBELL to generate.
@@ -169,41 +171,16 @@ notrace unsigned int __check_irq_replay(void) } } -/* -* Force the delivery of pending soft-disabled interrupts on PS3. -* Any HV call will have this side effect. -*/ -if (firmware_has_feature (FW_FEATURE_PS3_LV1 )) { -u64 tmp, tmp2; -lv1_get_version_info (&tmp, &tmp2); - } - -/* -* Check if an hypervisor Maintenance interrupt happened. -* This is a higher priority interrupt than the others, so -* replay it first. -*/ -if (happened &PACA_IRQ_HMI ) { - local_paca->irq_happened &= ~PACA_IRQ_HMI ; -return 0xe60 ; - } -if (happened &PACA_IRQ_DEC ) { local_paca->irq_happened &= ~PACA_IRQ_DEC ;return 0x900 ; } -if (happened &PACA_IRQ_PMI ) { - local_paca->irq_happened &= ~PACA_IRQ_PMI ; -return 0xf00 ; - } -if (happened &PACA_IRQ_EE ) { local_paca->irq_happened &= ~PACA_IRQ_EE ;return 0x500 ; } -#ifdef CONFIG_PPC_BOOK3E /* * Check if an EPR external interrupt happened this bit is typically * set if we need to handle another "edge" interrupt from within the
@@ -218,20 +195,15 @@ notrace unsigned int __check_irq_replay(void) local_paca->irq_happened &= ~PACA_IRQ_DBELL ; return 0x280; } -#else -if (happened &PACA_IRQ_DBELL ) { - local_paca->irq_happened &= ~PACA_IRQ_DBELL ; -return 0xa00 ; - } -#endif /* CONFIG_PPC_BOOK3E */ /* There should be nothing left ! */ BUG_ON (local_paca->irq_happened !=0 );return 0 ; } +#endif /* CONFIG_PPC_BOOK3E */ -static void replay_soft_interrupts (void ) +void replay_soft_interrupts (void ) {/* * We use local_paca rather than get_paca() to avoid all
arch/powerpc/kernel/process.c +48 -41
122 lines not on this page · raw message
arch/powerpc/kernel/syscall_64.c +163 -1
180 lines not on this page · raw message
arch/powerpc/kernel/vector.S +1 -1
@@ -134,7 +134,7 @@ _GLOBAL(load_up_vsx) /* enable use of VSX after return */ oris r12,r12,MSR_VSX@h std r12,_MSR(r1) - b fast_exception_return + b fast_interrupt_return #endif /* CONFIG_VSX */
Signature · 1 line
2.23.0
Re: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 7 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
The stack store emulation is significantly simplfied, rather than creating a new return frame and switching to that before performing the store, it uses the PACA to keep a scratch register around to perform thestore.
Quoted 9 lines
The asm return code is moved into 64e for now. The new logic has made allowance for 64e, but I don't have a full environment that works well to test it, and even booting in emulated qemu is not great for stress testing. 64e shouldn't be too far off working with this, given a bit more testing and auditing of the logic.
This is slightly faster on a POWER9 (page fault speed increases about 1.1%), probably due to reduced mtmsrd.
How do you measure 'page fault' speed ?
Nicholas Piggin
reviewerRe: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Excerpts from Christophe Leroy's message of January 27, 2021 6:54 pm:
Quoted 16 lines
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 13 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
The stack store emulation is significantly simplfied, rather than creating a new return frame and switching to that before performing the store, it uses the PACA to keep a scratch register around to perform thestore.
The asm return code is moved into 64e for now. The new logic has made allowance for 64e, but I don't have a full environment that works well to test it, and even booting in emulated qemu is not great for stress testing. 64e shouldn't be too far off working with this, given a bit more testing and auditing of the logic.
Quoted 5 lines
Quoted 3 lines
This is slightly faster on a POWER9 (page fault speed increases about 1.1%), probably due to reduced mtmsrd.
How do you measure 'page fault' speed ?
mmap 1000 pages, store to each one, mprotect(PROT_READ) then mprotect(PROT_READ|PROT_WRITE), then store a byte to each page and measure the cost. Something like that IIRC.
Thanks,
Nick
Re: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
In reply to Nicholas PigginLe 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 55 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
arch/powerpc/kernel/syscall_64.c +50
-0+notrace unsigned long interrupt_exit_kernel_prepare (struct pt_regs *regs,unsigned long msr) +{ +unsigned long *ti_flagsp = ¤t_thread_info ()->flags; +unsigned long flags; + +if (IS_ENABLED (CONFIG_PPC_BOOK3S ) &&unlikely (!(regs->msr &MSR_RI ))) +unrecoverable_exception (regs); +BUG_ON (regs->msr &MSR_PR ); +BUG_ON (!FULL_REGS (regs)); + +local_irq_save (flags); + +if (regs->softe ==IRQS_ENABLED ) { +/* Returning to a kernel context with local irqs enabled. */ +WARN_ON_ONCE (!(regs->msr &MSR_EE )); +again: +if (IS_ENABLED (CONFIG_PREEMPT )) { +/* Return to preemptible kernel context */ +if (unlikely (*ti_flagsp & _TIF_NEED_RESCHED)) { +if (preempt_count () ==0 ) +preempt_schedule_irq (); + } + } + +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (unlikely (lazy_irq_pending ())) { +__hard_RI_enable (); +irq_soft_mask_set (IRQS_ALL_DISABLED ); +trace_hardirqs_off (); + local_paca->irq_happened |=PACA_IRQ_HARD_DIS ; +/* +* Can't local_irq_enable in case we are in interrupt +* context. Must replay directly. +*/ +replay_soft_interrupts (); +irq_soft_mask_set (flags); +/* Took an interrupt, may have more exit work to do. */ + goto again; + } + local_paca->irq_happened =0 ; +irq_soft_mask_set (IRQS_ENABLED ); + }else { +/* Returning to a kernel context with local irqs disabled. */ +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (regs->msr &MSR_EE ) + local_paca->irq_happened &= ~PACA_IRQ_HARD_DIS ; + } +
arch/powerpc/kernel/syscall_64.c notrace unsigned long syscall_exit_prepare(unsigned long r3, · line 367
Quoted 10 lines
arch/powerpc/kernel/syscall_64.c +10
-0+ +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif + +/* +* We don't need to restore AMR on the way back to userspace for KUAP. +* The value of AMR only matters while we're in the kernel. +*/ +kuap_restore_amr (regs);
Is that correct to restore KUAP state here ? Shouldn't we have it at lower level in assembly ?
Isn't there a risk that someone manages to call interrupt_exit_kernel_prepare() or the end of it in a way or another, and get the previous KUAP state restored by this way ?
Also, it looks a bit strange to have kuap_save_amr_and_lock() done at lowest level in assembly, and kuap_restore_amr() done in upper level. That looks unbalanced.
Christophe
Quoted 22 lines
arch/powerpc/kernel/syscall_64.c +8
-0+ +if (unlikely (*ti_flagsp & _TIF_EMULATE_STACK_STORE)) { +clear_bits (_TIF_EMULATE_STACK_STORE, ti_flagsp); +return 1 ; + } + return 0; +} +#endif arch/powerpc/kernel/vector.S +1
-1@@ -134,7 +134,7 @@ _GLOBAL(load_up_vsx) /* enable use of VSX after return */ oris r12,r12,MSR_VSX@h std r12,_MSR(r1) - b fast_exception_return + b fast_interrupt_return #endif /* CONFIG_VSX */
Nicholas Piggin
reviewerRe: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Excerpts from Christophe Leroy's message of February 4, 2021 2:25 am:
Quoted 71 lines
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 55 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
arch/powerpc/kernel/syscall_64.c +50
-0+notrace unsigned long interrupt_exit_kernel_prepare (struct pt_regs *regs,unsigned long msr) +{ +unsigned long *ti_flagsp = ¤t_thread_info ()->flags; +unsigned long flags; + +if (IS_ENABLED (CONFIG_PPC_BOOK3S ) &&unlikely (!(regs->msr &MSR_RI ))) +unrecoverable_exception (regs); +BUG_ON (regs->msr &MSR_PR ); +BUG_ON (!FULL_REGS (regs)); + +local_irq_save (flags); + +if (regs->softe ==IRQS_ENABLED ) { +/* Returning to a kernel context with local irqs enabled. */ +WARN_ON_ONCE (!(regs->msr &MSR_EE )); +again: +if (IS_ENABLED (CONFIG_PREEMPT )) { +/* Return to preemptible kernel context */ +if (unlikely (*ti_flagsp & _TIF_NEED_RESCHED)) { +if (preempt_count () ==0 ) +preempt_schedule_irq (); + } + } + +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (unlikely (lazy_irq_pending ())) { +__hard_RI_enable (); +irq_soft_mask_set (IRQS_ALL_DISABLED ); +trace_hardirqs_off (); + local_paca->irq_happened |=PACA_IRQ_HARD_DIS ; +/* +* Can't local_irq_enable in case we are in interrupt +* context. Must replay directly. +*/ +replay_soft_interrupts (); +irq_soft_mask_set (flags); +/* Took an interrupt, may have more exit work to do. */ + goto again; + } + local_paca->irq_happened =0 ; +irq_soft_mask_set (IRQS_ENABLED ); + }else { +/* Returning to a kernel context with local irqs disabled. */ +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (regs->msr &MSR_EE ) + local_paca->irq_happened &= ~PACA_IRQ_HARD_DIS ; + } +arch/powerpc/kernel/syscall_64.c notrace unsigned long syscall_exit_prepare(unsigned long r3, · line 367
Quoted 10 lines
arch/powerpc/kernel/syscall_64.c +10
-0+ +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif + +/* +* We don't need to restore AMR on the way back to userspace for KUAP. +* The value of AMR only matters while we're in the kernel. +*/ +kuap_restore_amr (regs);Is that correct to restore KUAP state here ? Shouldn't we have it at lower level in assembly ?
Quoted 2 lines
Isn't there a risk that someone manages to call
interrupt_exit_kernel_prepare()or the end of it in a way or another, and get the previous KUAP state restored by this way ?
I'm not sure if there much more risk if it's here rather than the instruction being in another place in the code.
There's a lot of user access around the kernel too if you want to find a gadget to unlock KUAP then I suppose there is a pretty large attack surface.
Quoted 2 lines
Also, it looks a bit strange to have
kuap_save_amr_and_lock()done at lowest level in assembly, andkuap_restore_amr()done in upper level. That looks unbalanced.
I'd like to bring the entry assembly into C.
Thanks,
Nick
Re: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Le 04/02/2021 à 04:27, Nicholas Piggin a écrit :
Quoted 75 lines
Excerpts from Christophe Leroy's message of February 4, 2021 2:25 am:
Quoted 71 lines
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 55 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
arch/powerpc/kernel/syscall_64.c +50
-0+notrace unsigned long interrupt_exit_kernel_prepare (struct pt_regs *regs,unsigned long msr) +{ +unsigned long *ti_flagsp = ¤t_thread_info ()->flags; +unsigned long flags; + +if (IS_ENABLED (CONFIG_PPC_BOOK3S ) &&unlikely (!(regs->msr &MSR_RI ))) +unrecoverable_exception (regs); +BUG_ON (regs->msr &MSR_PR ); +BUG_ON (!FULL_REGS (regs)); + +local_irq_save (flags); + +if (regs->softe ==IRQS_ENABLED ) { +/* Returning to a kernel context with local irqs enabled. */ +WARN_ON_ONCE (!(regs->msr &MSR_EE )); +again: +if (IS_ENABLED (CONFIG_PREEMPT )) { +/* Return to preemptible kernel context */ +if (unlikely (*ti_flagsp & _TIF_NEED_RESCHED)) { +if (preempt_count () ==0 ) +preempt_schedule_irq (); + } + } + +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (unlikely (lazy_irq_pending ())) { +__hard_RI_enable (); +irq_soft_mask_set (IRQS_ALL_DISABLED ); +trace_hardirqs_off (); + local_paca->irq_happened |=PACA_IRQ_HARD_DIS ; +/* +* Can't local_irq_enable in case we are in interrupt +* context. Must replay directly. +*/ +replay_soft_interrupts (); +irq_soft_mask_set (flags); +/* Took an interrupt, may have more exit work to do. */ + goto again; + } + local_paca->irq_happened =0 ; +irq_soft_mask_set (IRQS_ENABLED ); + }else { +/* Returning to a kernel context with local irqs disabled. */ +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (regs->msr &MSR_EE ) + local_paca->irq_happened &= ~PACA_IRQ_HARD_DIS ; + } +arch/powerpc/kernel/syscall_64.c notrace unsigned long syscall_exit_prepare(unsigned long r3, · line 367
Quoted 10 lines
arch/powerpc/kernel/syscall_64.c +10
-0+ +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif + +/* +* We don't need to restore AMR on the way back to userspace for KUAP. +* The value of AMR only matters while we're in the kernel. +*/ +kuap_restore_amr (regs);Is that correct to restore KUAP state here ? Shouldn't we have it at lower level in assembly ?
Quoted 2 lines
Isn't there a risk that someone manages to call
interrupt_exit_kernel_prepare()or the end of it in a way or another, and get the previous KUAP state restored by this way ?
Quoted 6 lines
I'm not sure if there much more risk if it's here rather than the instruction being in another place in the code.
There's a lot of user access around the kernel too if you want to find a gadget to unlock KUAP then I suppose there is a pretty large attack surface.
My understanding is that user access scope is strictly limited, for instance we enforce the begin/end of user access to be in the same function, and we refrain from calling any other function inside the user access window. x86 even have 'objtool' to enforce it at build time. So in theory there is no way to get out of the function while user access is open.
Here with the interrupt exit function it is free beer. You have a place where you re-open user access and return with a simple blr. So that's open bar. If someone manages to just call the interrupt exit function, then user access remains open
Quoted 6 lines
Quoted 3 lines
Also, it looks a bit strange to have
kuap_save_amr_and_lock()done at lowest level in assembly, andkuap_restore_amr()done in upper level. That looks unbalanced.I'd like to bring the entry assembly into C.
I really think it's not a good idea. We'll get better control and readability by keeping it at the lowest possible level in assembly.
x86 only save and restore SMAC state in assembly.
Christophe
Nicholas Piggin
reviewerRe: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Excerpts from Christophe Leroy's message of February 4, 2021 6:03 pm:
Quoted 85 lines
Le 04/02/2021 à 04:27, Nicholas Piggin a écrit :
Quoted 75 lines
Excerpts from Christophe Leroy's message of February 4, 2021 2:25 am:
Quoted 71 lines
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 55 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
arch/powerpc/kernel/syscall_64.c +50
-0+notrace unsigned long interrupt_exit_kernel_prepare (struct pt_regs *regs,unsigned long msr) +{ +unsigned long *ti_flagsp = ¤t_thread_info ()->flags; +unsigned long flags; + +if (IS_ENABLED (CONFIG_PPC_BOOK3S ) &&unlikely (!(regs->msr &MSR_RI ))) +unrecoverable_exception (regs); +BUG_ON (regs->msr &MSR_PR ); +BUG_ON (!FULL_REGS (regs)); + +local_irq_save (flags); + +if (regs->softe ==IRQS_ENABLED ) { +/* Returning to a kernel context with local irqs enabled. */ +WARN_ON_ONCE (!(regs->msr &MSR_EE )); +again: +if (IS_ENABLED (CONFIG_PREEMPT )) { +/* Return to preemptible kernel context */ +if (unlikely (*ti_flagsp & _TIF_NEED_RESCHED)) { +if (preempt_count () ==0 ) +preempt_schedule_irq (); + } + } + +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (unlikely (lazy_irq_pending ())) { +__hard_RI_enable (); +irq_soft_mask_set (IRQS_ALL_DISABLED ); +trace_hardirqs_off (); + local_paca->irq_happened |=PACA_IRQ_HARD_DIS ; +/* +* Can't local_irq_enable in case we are in interrupt +* context. Must replay directly. +*/ +replay_soft_interrupts (); +irq_soft_mask_set (flags); +/* Took an interrupt, may have more exit work to do. */ + goto again; + } + local_paca->irq_happened =0 ; +irq_soft_mask_set (IRQS_ENABLED ); + }else { +/* Returning to a kernel context with local irqs disabled. */ +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (regs->msr &MSR_EE ) + local_paca->irq_happened &= ~PACA_IRQ_HARD_DIS ; + } +arch/powerpc/kernel/syscall_64.c notrace unsigned long syscall_exit_prepare(unsigned long r3, · line 367
Quoted 10 lines
arch/powerpc/kernel/syscall_64.c +10
-0+ +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif + +/* +* We don't need to restore AMR on the way back to userspace for KUAP. +* The value of AMR only matters while we're in the kernel. +*/ +kuap_restore_amr (regs);Is that correct to restore KUAP state here ? Shouldn't we have it at lower level in assembly ?
Quoted 2 lines
Isn't there a risk that someone manages to call
interrupt_exit_kernel_prepare()or the end of it in a way or another, and get the previous KUAP state restored by this way ?Quoted 6 lines
I'm not sure if there much more risk if it's here rather than the instruction being in another place in the code.
There's a lot of user access around the kernel too if you want to find a gadget to unlock KUAP then I suppose there is a pretty large attack surface.
Quoted 8 lines
My understanding is that user access scope is strictly limited, for instance we enforce the begin/end of user access to be in the same function, and we refrain from calling any other function inside the user access window. x86 even have 'objtool' to enforce it at build time. So in theory there is no way to get out of the function while user access is open.
Here with the interrupt exit function it is free beer. You have a place where you re-open user access and return with a simple blr. So that's open bar. If someone manages to just call the interrupt exit function, then user access remains open
Hmm okay maybe that's a good point.
Quoted 10 lines
Quoted 5 lines
Quoted 2 lines
Also, it looks a bit strange to have
kuap_save_amr_and_lock()done at lowest level in assembly, andkuap_restore_amr()done in upper level. That looks unbalanced.I'd like to bring the entry assembly into C.
I really think it's not a good idea. We'll get better control and readability by keeping it at the lowest possible level in assembly.
x86 only save and restore SMAC state in assembly.
Okay we could go the other way and move the unlock into asm then.
Thanks,
Nick
Michael Ellerman
reviewerRe: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Quoted 4 lines · Nicholas Piggin
Excerpts from Christophe Leroy's message of February 4, 2021 6:03 pm:
Quoted 3 lines
Le 04/02/2021 à 04:27, Nicholas Piggin a écrit :
Quoted 2 lines
Excerpts from Christophe Leroy's message of February 4, 2021 2:25 am:
Quoted 1 line
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
...
Quoted 24 lines
Quoted 24 lines
Quoted 12 lines
Quoted 11 lines
arch/powerpc/kernel/syscall_64.c notrace unsigned long syscall_exit_prepare(unsigned long r3, · line 367
Quoted 6 lines
arch/powerpc/kernel/syscall_64.c +6
-0+ +/* +* We don't need to restore AMR on the way back to userspace for KUAP. +* The value of AMR only matters while we're in the kernel. +*/ +kuap_restore_amr (regs);Is that correct to restore KUAP state here ? Shouldn't we have it at lower level in assembly ?
Isn't there a risk that someone manages to call
interrupt_exit_kernel_prepare()or the end of it in a way or another, and get the previous KUAP state restored by this way ?Quoted 6 lines
I'm not sure if there much more risk if it's here rather than the instruction being in another place in the code.
There's a lot of user access around the kernel too if you want to find a gadget to unlock KUAP then I suppose there is a pretty large attack surface.
My understanding is that user access scope is strictly limited, for instance we enforce the begin/end of user access to be in the same function, and we refrain from calling any other function inside the user access window. x86 even have 'objtool' to enforce it at build time. So in theory there is no way to get out of the function while user access is open.
Quoted 5 lines
Quoted 3 lines
Here with the interrupt exit function it is free beer. You have a place where you re-open user access and return with a simple blr. So that's open bar. If someone manages to just call the interrupt exit function, then user access remains open
Hmm okay maybe that's a good point.
I don't think it's a very attractive gadget, it's not just a plain blr, it does a full stack frame tear down before the return. And there's no LR reloads anywhere very close.
Obviously it depends on what the compiler decides to do, it's possible it could be a usable gadget. But there are other places that are more attractive I think, eg:
c00000000061d768: a6 03 3d 7d mtspr 29,r9c00000000061d76c: 2c 01 00 4c isyncc00000000061d770: 00 00 00 60 nopc00000000061d774: 30 00 21 38 addi r1,r1,48c00000000061d778: 20 00 80 4e blr
So I don't think we should redesign the code *purely* because we're worried about interrupt_exit_kernel_prepare() being a useful gadget. If we can come up with a way to restructure it that reads well and is maintainable, and also reduces the chance of it being a good gadget then sure.
cheers
Nicholas Piggin
reviewerRe: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Excerpts from Michael Ellerman's message of February 5, 2021 10:22 am:
Quoted 51 lines
Quoted 4 lines · Nicholas Piggin
Excerpts from Christophe Leroy's message of February 4, 2021 6:03 pm:
Quoted 3 lines
Le 04/02/2021 à 04:27, Nicholas Piggin a écrit :
Quoted 2 lines
Excerpts from Christophe Leroy's message of February 4, 2021 2:25 am:
Quoted 1 line
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
...
Quoted 24 lines
Quoted 24 lines
Quoted 12 lines
Quoted 11 lines
arch/powerpc/kernel/syscall_64.c notrace unsigned long syscall_exit_prepare(unsigned long r3, · line 367
Quoted 6 lines
arch/powerpc/kernel/syscall_64.c +6
-0+ +/* +* We don't need to restore AMR on the way back to userspace for KUAP. +* The value of AMR only matters while we're in the kernel. +*/ +kuap_restore_amr (regs);Is that correct to restore KUAP state here ? Shouldn't we have it at lower level in assembly ?
Isn't there a risk that someone manages to call
interrupt_exit_kernel_prepare()or the end of it in a way or another, and get the previous KUAP state restored by this way ?Quoted 6 lines
I'm not sure if there much more risk if it's here rather than the instruction being in another place in the code.
There's a lot of user access around the kernel too if you want to find a gadget to unlock KUAP then I suppose there is a pretty large attack surface.
My understanding is that user access scope is strictly limited, for instance we enforce the begin/end of user access to be in the same function, and we refrain from calling any other function inside the user access window. x86 even have 'objtool' to enforce it at build time. So in theory there is no way to get out of the function while user access is open.
Quoted 5 lines
Quoted 3 lines
Here with the interrupt exit function it is free beer. You have a place where you re-open user access and return with a simple blr. So that's open bar. If someone manages to just call the interrupt exit function, then user access remains open
Hmm okay maybe that's a good point.
I don't think it's a very attractive gadget, it's not just a plain blr, it does a full stack frame tear down before the return. And there's no LR reloads anywhere very close.
Obviously it depends on what the compiler decides to do, it's possible it could be a usable gadget. But there are other places that are more attractive I think, eg:
c00000000061d768: a6 03 3d 7d mtspr 29,r9c00000000061d76c: 2c 01 00 4c isyncc00000000061d770: 00 00 00 60 nopc00000000061d774: 30 00 21 38 addi r1,r1,48c00000000061d778: 20 00 80 4e blr
Quoted 5 lines
So I don't think we should redesign the code *purely* because we're worried about
interrupt_exit_kernel_prepare()being a useful gadget. If we can come up with a way to restructure it that reads well and is maintainable, and also reduces the chance of it being a good gadget then sure.
Okay. That would be good if we can keep it in C, the pkeys + kuap combo is fairly complicated and we might want to something cleverer with it, so that would make it even more difficult in asm.
Thanks,
Nick
Re: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Le 05/02/2021 à 03:16, Nicholas Piggin a écrit :
Quoted 58 lines
Excerpts from Michael Ellerman's message of February 5, 2021 10:22 am:
Quoted 51 lines
Quoted 4 lines · Nicholas Piggin
Excerpts from Christophe Leroy's message of February 4, 2021 6:03 pm:
Quoted 3 lines
Le 04/02/2021 à 04:27, Nicholas Piggin a écrit :
Quoted 2 lines
Excerpts from Christophe Leroy's message of February 4, 2021 2:25 am:
Quoted 1 line
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
...
Quoted 24 lines
Quoted 24 lines
Quoted 12 lines
Quoted 11 lines
Review
Quoted 6 lines
(quoted diff) +6
-0+ + /* + * We don't need to restore AMR on the way back to userspace for KUAP. + * The value of AMR only matters while we're in the kernel. + */ + kuap_restore_amr(regs);Is that correct to restore KUAP state here ? Shouldn't we have it at lower level in assembly ?
Isn't there a risk that someone manages to call
interrupt_exit_kernel_prepare()or the end of it in a way or another, and get the previous KUAP state restored by this way ?Quoted 6 lines
I'm not sure if there much more risk if it's here rather than the instruction being in another place in the code.
There's a lot of user access around the kernel too if you want to find a gadget to unlock KUAP then I suppose there is a pretty large attack surface.
My understanding is that user access scope is strictly limited, for instance we enforce the begin/end of user access to be in the same function, and we refrain from calling any other function inside the user access window. x86 even have 'objtool' to enforce it at build time. So in theory there is no way to get out of the function while user access is open.
Quoted 5 lines
Quoted 3 lines
Here with the interrupt exit function it is free beer. You have a place where you re-open user access and return with a simple blr. So that's open bar. If someone manages to just call the interrupt exit function, then user access remains open
Hmm okay maybe that's a good point.
I don't think it's a very attractive gadget, it's not just a plain blr, it does a full stack frame tear down before the return. And there's no LR reloads anywhere very close.
Obviously it depends on what the compiler decides to do, it's possible it could be a usable gadget. But there are other places that are more attractive I think, eg:
c00000000061d768: a6 03 3d 7d mtspr 29,r9c00000000061d76c: 2c 01 00 4c isyncc00000000061d770: 00 00 00 60 nopc00000000061d774: 30 00 21 38 addi r1,r1,48c00000000061d778: 20 00 80 4e blrQuoted 5 lines
So I don't think we should redesign the code *purely* because we're worried about
interrupt_exit_kernel_prepare()being a useful gadget. If we can come up with a way to restructure it that reads well and is maintainable, and also reduces the chance of it being a good gadget then sure.
Quoted 4 lines
Okay. That would be good if we can keep it in C, the pkeys + kuap combo is fairly complicated and we might want to something cleverer with it, so that would make it even more difficult in asm.
Ok.
For ppc32, I prefer to keep it in assembly for the time being and move everything from ASM to C at once after porting syscall and interrupts to C and wrappers.
Hope this is OK for you.
Christophe
Nicholas Piggin
reviewerRe: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Excerpts from Christophe Leroy's message of February 5, 2021 4:04 pm:
Quoted 66 lines
Le 05/02/2021 à 03:16, Nicholas Piggin a écrit :
Quoted 58 lines
Excerpts from Michael Ellerman's message of February 5, 2021 10:22 am:
Quoted 51 lines
Quoted 4 lines · Nicholas Piggin
Excerpts from Christophe Leroy's message of February 4, 2021 6:03 pm:
Quoted 3 lines
Le 04/02/2021 à 04:27, Nicholas Piggin a écrit :
Quoted 2 lines
Excerpts from Christophe Leroy's message of February 4, 2021 2:25 am:
Quoted 1 line
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
...
Quoted 24 lines
Quoted 24 lines
Quoted 12 lines
Quoted 11 lines
Review
Quoted 6 lines
(quoted diff) +6
-0+ + /* + * We don't need to restore AMR on the way back to userspace for KUAP. + * The value of AMR only matters while we're in the kernel. + */ + kuap_restore_amr(regs);Is that correct to restore KUAP state here ? Shouldn't we have it at lower level in assembly ?
Isn't there a risk that someone manages to call
interrupt_exit_kernel_prepare()or the end of it in a way or another, and get the previous KUAP state restored by this way ?Quoted 6 lines
I'm not sure if there much more risk if it's here rather than the instruction being in another place in the code.
There's a lot of user access around the kernel too if you want to find a gadget to unlock KUAP then I suppose there is a pretty large attack surface.
My understanding is that user access scope is strictly limited, for instance we enforce the begin/end of user access to be in the same function, and we refrain from calling any other function inside the user access window. x86 even have 'objtool' to enforce it at build time. So in theory there is no way to get out of the function while user access is open.
Quoted 5 lines
Quoted 3 lines
Here with the interrupt exit function it is free beer. You have a place where you re-open user access and return with a simple blr. So that's open bar. If someone manages to just call the interrupt exit function, then user access remains open
Hmm okay maybe that's a good point.
I don't think it's a very attractive gadget, it's not just a plain blr, it does a full stack frame tear down before the return. And there's no LR reloads anywhere very close.
Obviously it depends on what the compiler decides to do, it's possible it could be a usable gadget. But there are other places that are more attractive I think, eg:
c00000000061d768: a6 03 3d 7d mtspr 29,r9c00000000061d76c: 2c 01 00 4c isyncc00000000061d770: 00 00 00 60 nopc00000000061d774: 30 00 21 38 addi r1,r1,48c00000000061d778: 20 00 80 4e blrQuoted 5 lines
So I don't think we should redesign the code *purely* because we're worried about
interrupt_exit_kernel_prepare()being a useful gadget. If we can come up with a way to restructure it that reads well and is maintainable, and also reduces the chance of it being a good gadget then sure.Quoted 4 lines
Okay. That would be good if we can keep it in C, the pkeys + kuap combo is fairly complicated and we might want to something cleverer with it, so that would make it even more difficult in asm.
Quoted 6 lines
Ok.
For ppc32, I prefer to keep it in assembly for the time being and move everything from ASM to C at once after porting syscall and interrupts to C and wrappers.
Hope this is OK for you.
I don't see a problem with that.
Thanks,
Nick
Re: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
In reply to Nicholas PigginLe 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 7 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
The stack store emulation is significantly simplfied, rather than creating a new return frame and switching to that before performing the store, it uses the PACA to keep a scratch register around to perform thestore.
Quoted 8 lines
The asm return code is moved into 64e for now. The new logic has made allowance for 64e, but I don't have a full environment that works well to test it, and even booting in emulated qemu is not great for stress testing. 64e shouldn't be too far off working with this, given a bit more testing and auditing of the logic.
This is slightly faster on a POWER9 (page fault speed increases about 1.1%), probably due to reduced mtmsrd.
This series, and especially this patch has added a awfull number of BUG_ON() traps.
We have an issue open at https://github.com/linuxppc/issues/issues/88 since 2017 for reducing the number of BUG_ON()s
And the kernel Documentation is explicit on the willingness to deprecate BUG_ON(), see https://www.kernel.org/doc/html/latest/process/deprecated.html?highlight=bug_on :
BUG() and BUG_ON()
Use WARN() and WARN_ON() instead, and handle the “impossible” error condition as gracefully as possible. While the BUG()-family of APIs were originally designed to act as an “impossible situation” assert and to kill a kernel thread “safely”, they turn out to just be too risky. (e.g. “In what order do locks need to be released? Have various states been restored?”) Very commonly, using BUG() will destabilize a system or entirely break it, which makes it impossible to debug or even get viable crash reports. Linus has very strong feelings about this.
So ... can we do something cleaner with all the BUG_ON()s recently added ?
Christophe
Quoted 1,833 lines
Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michal Suchanek <[email protected]>---
v2,rebase (from Michal):
- Move the FP restore functions torestore_math. They are not used anywhere else and whenrestore_mathis not built gcc warns about them being unused (ms)
- Addasm/context_tracking.hinclude toexceptions-64e.SforSCHEDULE_USERdefinitionv3:
- Fix return from interrupt replay problem by replaying interrupts rather than enabling irqs. This ends up being cleaner and__check_irq_replaygoes away completely for 64s. Should bring 64e up to speed and kill a lot of cruft after it's proven on 64s.
- Don't use_GLOBALif it's not called from C
- Simplify stack store emulation code further, add a bit more commenting.
- Some missing no probe annotations.../powerpc/include/asm/book3s/64/kup-radix.h | 10 + arch/powerpc/include/asm/hw_irq.h | 1 + arch/powerpc/include/asm/switch_to.h | 6 + arch/powerpc/kernel/entry_64.S | 486 +++++------------- arch/powerpc/kernel/exceptions-64e.S | 255 ++++++++- arch/powerpc/kernel/exceptions-64s.S | 119 ++--- arch/powerpc/kernel/irq.c | 36 +- arch/powerpc/kernel/process.c | 89 ++-- arch/powerpc/kernel/syscall_64.c | 164 +++++- arch/powerpc/kernel/vector.S | 2 +- 10 files changed, 642 insertions(+), 526 deletions(-)arch/powerpc/include/asm/book3s/64/kup-radix.h +10
-0@@ -60,6 +60,12 @@#include <asm/mmu.h> #include <asm/ptrace.h> +static inline void kuap_restore_amr (struct pt_regs *regs) +{ +if (mmu_has_feature (MMU_FTR_RADIX_KUAP )) +mtspr (SPRN_AMR , regs->kuap); +} +static inline void kuap_check_amr (void ) {if (IS_ENABLED (CONFIG_PPC_KUAP_DEBUG ) &&mmu_has_feature (MMU_FTR_RADIX_KUAP ))@@ -136,6 +142,10 @@ bad_kuap_fault(struct pt_regs *regs, unsigned long address, bool is_write) "Bug: %s fault blocked by AMR!", is_write ? "Write" : "Read"); }#else /* CONFIG_PPC_KUAP */ +static inline void kuap_restore_amr (struct pt_regs *regs) +{ +} +static inline void kuap_check_amr (void ) { }arch/powerpc/include/asm/hw_irq.h +1
-0@@ -52,6 +52,7 @@#ifndef __ASSEMBLY__extern void replay_system_reset (void ); +extern void replay_soft_interrupts (void );extern void timer_interrupt (struct pt_regs *);extern void timer_broadcast_interrupt (void );arch/powerpc/include/asm/switch_to.h +6
-0@@ -23,7 +23,13 @@ extern void switch_booke_debug_regs(struct debug_reg *new_debug);extern int emulate_altivec (struct pt_regs *); +#ifdef CONFIG_PPC_BOOK3S_64 void restore_math (struct pt_regs *regs); +#else +static inline void restore_math (struct pt_regs *regs) +{ +} +#endif void restore_tm_state (struct pt_regs *regs);arch/powerpc/kernel/entry_64.S +116
-370@@ -16,6 +16,7 @@ #include <linux/errno.h> #include <linux/err.h> +#include <asm/cache.h> #include <asm/unistd.h> #include <asm/processor.h> #include <asm/page.h>@@ -221,6 +222,7 @@ _GLOBAL(ret_from_kernel_thread) li r3,0 b .Lsyscall_exit +#ifdef CONFIG_PPC_BOOK3E /* Save non-volatile GPRs, if not already saved. */ _GLOBAL(save_nvgprs) ld r11,_TRAP(r1)@@ -231,6 +233,7 @@ _GLOBAL(save_nvgprs) std r0,_TRAP(r1) blr _ASM_NOKPROBE_SYMBOL(save_nvgprs); +#endif #ifdef CONFIG_PPC_BOOK3S_64@@ -294,7 +297,7 @@ flush_count_cache: * state of one is saved on its kernel stack. Then the state * of the other is restored from its kernel stack. The memory * management hardware is updated to the second process's state. - * Finally, we can return to the second process, via ret_from_except. + * Finally, we can return to the second process, via interrupt_return. * On entry, r3 points to the THREAD for the current task, r4 * points to the THREAD for the new task. *@@ -446,408 +449,151 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S) addi r1,r1,SWITCH_FRAME_SIZE blr - .align 7 -_GLOBAL(ret_from_except) - ld r11,_TRAP(r1) - andi. r0,r11,1 - bne ret_from_except_lite - REST_NVGPRS(r1) - -_GLOBAL(ret_from_except_lite) +#ifdef CONFIG_PPC_BOOK3S /* - * Disable interrupts so that current_thread_info()->flags - * can't change between when we test it and when we return - * from the interrupt. - */ -#ifdef CONFIG_PPC_BOOK3E - wrteei 0 -#else - li r10,MSR_RI - mtmsrd r10,1 /* Update machine state */ -#endif /* CONFIG_PPC_BOOK3E */ + * If MSR EE/RI was never enabled, IRQs not reconciled, NVGPRs not + * touched, AMR not set, no exit work created, then this can be used. + */ + .balign IFETCH_ALIGN_BYTES + .globl fast_interrupt_return +fast_interrupt_return: +_ASM_NOKPROBE_SYMBOL(fast_interrupt_return) + ld r4,_MSR(r1) + andi. r0,r4,MSR_PR + bne .Lfast_user_interrupt_return + andi. r0,r4,MSR_RI + bne+ .Lfast_kernel_interrupt_return + addi r3,r1,STACK_FRAME_OVERHEAD + bl unrecoverable_exception + b . /* should not get here */ - ld r9, PACA_THREAD_INFO(r13) - ld r3,_MSR(r1) -#ifdef CONFIG_PPC_BOOK3E - ld r10,PACACURRENT(r13) -#endif /* CONFIG_PPC_BOOK3E */ - ld r4,TI_FLAGS(r9) - andi. r3,r3,MSR_PR - beq resume_kernel -#ifdef CONFIG_PPC_BOOK3E - lwz r3,(THREAD+THREAD_DBCR0)(r10) -#endif /* CONFIG_PPC_BOOK3E */ + .balign IFETCH_ALIGN_BYTES + .globl interrupt_return +interrupt_return: +_ASM_NOKPROBE_SYMBOL(interrupt_return) + REST_NVGPRS(r1) - /* Check current_thread_info()->flags */ - andi. r0,r4,_TIF_USER_WORK_MASK - bne 1f -#ifdef CONFIG_PPC_BOOK3E - /* - * Check to see if the dbcr0 register is set up to debug. - * Use the internal debug mode bit to do this. - */ - andis. r0,r3,DBCR0_IDM@h - beq restore - mfmsr r0 - rlwinm r0,r0,0,~MSR_DE /* Clear MSR.DE */ - mtmsr r0 - mtspr SPRN_DBCR0,r3 - li r10, -1 - mtspr SPRN_DBSR,r10 - b restore -#else - addi r3,r1,STACK_FRAME_OVERHEAD - bl restore_math - b restore -#endif -1: andi. r0,r4,_TIF_NEED_RESCHED - beq 2f - bl restore_interrupts - SCHEDULE_USER - b ret_from_except_lite -2: -#ifdef CONFIG_PPC_TRANSACTIONAL_MEM - andi. r0,r4,_TIF_USER_WORK_MASK & ~_TIF_RESTORE_TM - bne 3f /* only restore TM if nothing else to do */ + .balign IFETCH_ALIGN_BYTES + .globl interrupt_return_lite +interrupt_return_lite: +_ASM_NOKPROBE_SYMBOL(interrupt_return_lite) + ld r4,_MSR(r1) + andi. r0,r4,MSR_PR + beq .Lkernel_interrupt_return addi r3,r1,STACK_FRAME_OVERHEAD - bl restore_tm_state - b restore -3: -#endif - bl save_nvgprs - /* - * Use a non volatile GPR to save and restore our thread_info flags - * across the call to restore_interrupts. - */ - mr r30,r4 - bl restore_interrupts - mr r4,r30 - addi r3,r1,STACK_FRAME_OVERHEAD - bl do_notify_resume - b ret_from_except - -resume_kernel: - /* check current_thread_info, _TIF_EMULATE_STACK_STORE */ - andis. r8,r4,_TIF_EMULATE_STACK_STORE@h - beq+ 1f + bl interrupt_exit_user_prepare + cmpdi r3,0 + bne- .Lrestore_nvgprs - addi r8,r1,INT_FRAME_SIZE /* Get the kprobed function entry */ +.Lfast_user_interrupt_return: + ld r11,_NIP(r1) + ld r12,_MSR(r1) +BEGIN_FTR_SECTION + ld r10,_PPR(r1) + mtspr SPRN_PPR,r10 +END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) + mtspr SPRN_SRR0,r11 + mtspr SPRN_SRR1,r12 - ld r3,GPR1(r1) - subi r3,r3,INT_FRAME_SIZE /* dst: Allocate a trampoline exception frame */ - mr r4,r1 /* src: current exception frame */ - mr r1,r3 /* Reroute the trampoline frame to r1 */ +BEGIN_FTR_SECTION + stdcx. r0,0,r1 /* to clear the reservation */ +FTR_SECTION_ELSE + ldarx r0,0,r1 +ALT_FTR_SECTION_END_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS) - /* Copy from the original to the trampoline. */ - li r5,INT_FRAME_SIZE/8 /* size: INT_FRAME_SIZE */ - li r6,0 /* start offset: 0 */ - mtctr r5 -2: ldx r0,r6,r4 - stdx r0,r6,r3 - addi r6,r6,8 - bdnz 2b - - /* Do real store operation to complete stdu */ - ld r5,GPR1(r1) - std r8,0(r5) - - /* Clear _TIF_EMULATE_STACK_STORE flag */ - lis r11,_TIF_EMULATE_STACK_STORE@h - addi r5,r9,TI_FLAGS -0: ldarx r4,0,r5 - andc r4,r4,r11 - stdcx. r4,0,r5 - bne- 0b -1: - -#ifdef CONFIG_PREEMPTION - /* Check if we need to preempt */ - andi. r0,r4,_TIF_NEED_RESCHED - beq+ restore - /* Check that preempt_count() == 0 and interrupts are enabled */ - lwz r8,TI_PREEMPT(r9) - cmpwi cr0,r8,0 - bne restore - ld r0,SOFTE(r1) - andi. r0,r0,IRQS_DISABLED - bne restore + ld r3,_CCR(r1) + ld r4,_LINK(r1) + ld r5,_CTR(r1) + ld r6,_XER(r1) + li r0,0 - /* - * Here we are preempting the current task. We want to make - * sure we are soft-disabled first and reconcile irq state. - */ - RECONCILE_IRQ_STATE(r3,r4) - bl preempt_schedule_irq + REST_4GPRS(7, r1) + REST_2GPRS(11, r1) + REST_GPR(13, r1) - /* - * arch_local_irq_restore() from preempt_schedule_irq above may - * enable hard interrupt but we really should disable interrupts - * when we return from the interrupt, and so that we don't get - * interrupted after loading SRR0/1. - */ -#ifdef CONFIG_PPC_BOOK3E - wrteei 0 -#else - li r10,MSR_RI - mtmsrd r10,1 /* Update machine state */ -#endif /* CONFIG_PPC_BOOK3E */ -#endif /* CONFIG_PREEMPTION */ + mtcr r3 + mtlr r4 + mtctr r5 + mtspr SPRN_XER,r6 - .globl fast_exc_return_irq -fast_exc_return_irq: -restore: - /* - * This is the main kernel exit path. First we check if we - * are about to re-enable interrupts - */ - ld r5,SOFTE(r1) - lbz r6,PACAIRQSOFTMASK(r13) - andi. r5,r5,IRQS_DISABLED - bne .Lrestore_irq_off + REST_4GPRS(2, r1) + REST_GPR(6, r1) + REST_GPR(0, r1) + REST_GPR(1, r1) + RFI_TO_USER + b . /* prevent speculative execution */ - /* We are enabling, were we already enabled ? Yes, just return */ - andi. r6,r6,IRQS_DISABLED - beq cr0,.Ldo_restore +.Lrestore_nvgprs: + REST_NVGPRS(r1) + b .Lfast_user_interrupt_return - /* - * We are about to soft-enable interrupts (we are hard disabled - * at this point). We check if there's anything that needs to - * be replayed first. - */ - lbz r0,PACAIRQHAPPENED(r13) - cmpwi cr0,r0,0 - bne- .Lrestore_check_irq_replay + .balign IFETCH_ALIGN_BYTES +.Lkernel_interrupt_return: + addi r3,r1,STACK_FRAME_OVERHEAD + bl interrupt_exit_kernel_prepare + cmpdi cr1,r3,0 - /* - * Get here when nothing happened while soft-disabled, just - * soft-enable and move-on. We will hard-enable as a side - * effect of rfi - */ -.Lrestore_no_replay: - TRACE_ENABLE_INTS - li r0,IRQS_ENABLED - stb r0,PACAIRQSOFTMASK(r13); +.Lfast_kernel_interrupt_return: + ld r11,_NIP(r1) + ld r12,_MSR(r1) + mtspr SPRN_SRR0,r11 + mtspr SPRN_SRR1,r12 - /* - * Final return path. BookE is handled in a different file - */ -.Ldo_restore: -#ifdef CONFIG_PPC_BOOK3E - b exception_return_book3e -#else - /* - * Clear the reservation. If we know the CPU tracks the address of - * the reservation then we can potentially save some cycles and use - * a larx. On POWER6 and POWER7 this is significantly faster. - */ BEGIN_FTR_SECTION stdcx. r0,0,r1 /* to clear the reservation */ FTR_SECTION_ELSE - ldarx r4,0,r1 + ldarx r0,0,r1 ALT_FTR_SECTION_END_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS) - /* - * Some code path such as load_up_fpu or altivec return directly - * here. They run entirely hard disabled and do not alter the - * interrupt state. They also don't use lwarx/stwcx. and thus - * are known not to leave dangling reservations. - */ - .globl fast_exception_return -fast_exception_return: - ld r3,_MSR(r1) + ld r3,_LINK(r1) ld r4,_CTR(r1) - ld r0,_LINK(r1) - mtctr r4 - mtlr r0 - ld r4,_XER(r1) - mtspr SPRN_XER,r4 - - kuap_check_amr r5, r6 - - REST_8GPRS(5, r1) - - andi. r0,r3,MSR_RI - beq- .Lunrecov_restore - - /* - * Clear RI before restoring r13. If we are returning to - * userspace and we take an exception after restoring r13, - * we end up corrupting the userspace r13 value. - */ - li r4,0 - mtmsrd r4,1 - -#ifdef CONFIG_PPC_TRANSACTIONAL_MEM - /* TM debug */ - std r3, PACATMSCRATCH(r13) /* Stash returned-to MSR */ -#endif - /* - * r13 is our per cpu area, only restore it if we are returning to - * userspace the value stored in the stack frame may belong to - * another CPU. - */ - andi. r0,r3,MSR_PR - beq 1f -BEGIN_FTR_SECTION - /* Restore PPR */ - ld r2,_PPR(r1) - mtspr SPRN_PPR,r2 -END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) - ACCOUNT_CPU_USER_EXIT(r13, r2, r4) - REST_GPR(13, r1) - - /* - * We don't need to restore AMR on the way back to userspace for KUAP. - * The value of AMR only matters while we're in the kernel. - */ - mtspr SPRN_SRR1,r3 - - ld r2,_CCR(r1) - mtcrf 0xFF,r2 - ld r2,_NIP(r1) - mtspr SPRN_SRR0,r2 - - ld r0,GPR0(r1) - ld r2,GPR2(r1) - ld r3,GPR3(r1) - ld r4,GPR4(r1) - ld r1,GPR1(r1) - RFI_TO_USER - b . /* prevent speculative execution */ + ld r5,_XER(r1) + ld r6,_CCR(r1) + li r0,0 -1: mtspr SPRN_SRR1,r3 + REST_4GPRS(7, r1) + REST_2GPRS(11, r1) - ld r2,_CCR(r1) - mtcrf 0xFF,r2 - ld r2,_NIP(r1) - mtspr SPRN_SRR0,r2 + mtlr r3 + mtctr r4 + mtspr SPRN_XER,r5 /* * Leaving a stale exception_marker on the stack can confuse * the reliable stack unwinder later on. Clear it. */ - li r2,0 - std r2,STACK_FRAME_OVERHEAD-16(r1) + std r0,STACK_FRAME_OVERHEAD-16(r1) - ld r0,GPR0(r1) - ld r2,GPR2(r1) - ld r3,GPR3(r1) + REST_4GPRS(2, r1) - kuap_restore_amr r4 - - ld r4,GPR4(r1) - ld r1,GPR1(r1) + bne- cr1,1f /* emulate stack store */ + mtcr r6 + REST_GPR(6, r1) + REST_GPR(0, r1) + REST_GPR(1, r1) RFI_TO_KERNEL b . /* prevent speculative execution */ -#endif /* CONFIG_PPC_BOOK3E */ - - /* - * We are returning to a context with interrupts soft disabled. - * - * However, we may also about to hard enable, so we need to - * make sure that in this case, we also clear PACA_IRQ_HARD_DIS - * or that bit can get out of sync and bad things will happen - */ -.Lrestore_irq_off: - ld r3,_MSR(r1) - lbz r7,PACAIRQHAPPENED(r13) - andi. r0,r3,MSR_EE - beq 1f - rlwinm r7,r7,0,~PACA_IRQ_HARD_DIS - stb r7,PACAIRQHAPPENED(r13) -1: -#if defined(CONFIG_PPC_IRQ_SOFT_MASK_DEBUG) && defined(CONFIG_BUG) - /* The interrupt should not have soft enabled. */ - lbz r7,PACAIRQSOFTMASK(r13) -1: tdeqi r7,IRQS_ENABLED - EMIT_BUG_ENTRY 1b,__FILE__,__LINE__,BUGFLAG_WARNING -#endif - b .Ldo_restore - - /* - * Something did happen, check if a re-emit is needed - * (this also clears paca->irq_happened) - */ -.Lrestore_check_irq_replay: - /* XXX: We could implement a fast path here where we check - * for irq_happened being just 0x01, in which case we can - * clear it and return. That means that we would potentially - * miss a decrementer having wrapped all the way around. - * - * Still, this might be useful for things like hash_page - */ - bl __check_irq_replay - cmpwi cr0,r3,0 - beq .Lrestore_no_replay - - /* - * We need to re-emit an interrupt. We do so by re-using our - * existing exception frame. We first change the trap value, - * but we need to ensure we preserve the low nibble of it - */ - ld r4,_TRAP(r1) - clrldi r4,r4,60 - or r4,r4,r3 - std r4,_TRAP(r1) - - /* - * PACA_IRQ_HARD_DIS won't always be set here, so set it now - * to reconcile the IRQ state. Tracing is already accounted for. - */ - lbz r4,PACAIRQHAPPENED(r13) - ori r4,r4,PACA_IRQ_HARD_DIS - stb r4,PACAIRQHAPPENED(r13) - - /* - * Then find the right handler and call it. Interrupts are - * still soft-disabled and we keep them that way. - */ - cmpwi cr0,r3,0x500 - bne 1f - addi r3,r1,STACK_FRAME_OVERHEAD; - bl do_IRQ - b ret_from_except -1: cmpwi cr0,r3,0xf00 - bne 1f - addi r3,r1,STACK_FRAME_OVERHEAD; - bl performance_monitor_exception - b ret_from_except -1: cmpwi cr0,r3,0xe60 - bne 1f - addi r3,r1,STACK_FRAME_OVERHEAD; - bl handle_hmi_exception - b ret_from_except -1: cmpwi cr0,r3,0x900 - bne 1f - addi r3,r1,STACK_FRAME_OVERHEAD; - bl timer_interrupt - b ret_from_except -#ifdef CONFIG_PPC_DOORBELL -1: -#ifdef CONFIG_PPC_BOOK3E - cmpwi cr0,r3,0x280 -#else - cmpwi cr0,r3,0xa00 -#endif /* CONFIG_PPC_BOOK3E */ - bne 1f - addi r3,r1,STACK_FRAME_OVERHEAD; - bl doorbell_exception -#endif /* CONFIG_PPC_DOORBELL */ -1: b ret_from_except /* What else to do here ? */ - -.Lunrecov_restore: - addi r3,r1,STACK_FRAME_OVERHEAD - bl unrecoverable_exception - b .Lunrecov_restore - -_ASM_NOKPROBE_SYMBOL(ret_from_except); -_ASM_NOKPROBE_SYMBOL(ret_from_except_lite); -_ASM_NOKPROBE_SYMBOL(resume_kernel); -_ASM_NOKPROBE_SYMBOL(fast_exc_return_irq); -_ASM_NOKPROBE_SYMBOL(restore); -_ASM_NOKPROBE_SYMBOL(fast_exception_return); +1: /* + * Emulate stack store with update. New r1 value was already calculated + * and updated in our interrupt regs by emulate_loadstore, but we can't + * store the previous value of r1 to the stack before re-loading our + * registers from it, otherwise they could be clobbered. Use + * PACA_EXGEN as temporary storage to hold the store data, as + * interrupts are disabled here so it won't be clobbered. + */ + mtcr r6 + std r9,PACA_EXGEN+0(r13) + addi r9,r1,INT_FRAME_SIZE /* get original r1 */ + REST_GPR(6, r1) + REST_GPR(0, r1) + REST_GPR(1, r1) + std r9,0(r1) /* perform store component of stdu */ + ld r9,PACA_EXGEN+0(r13) + RFI_TO_KERNEL + b . /* prevent speculative execution */ +#endif /* CONFIG_PPC_BOOK3S */ #ifdef CONFIG_PPC_RTAS /*arch/powerpc/kernel/exceptions-64e.S +248
-7@@ -24,6 +24,7 @@ #include <asm/kvm_asm.h> #include <asm/kvm_booke_hv_asm.h> #include <asm/feature-fixups.h> +#include <asm/context_tracking.h> /* XXX This will ultimately add space for a special exception save * structure used to save things like SRR0/SRR1, SPRGs, MAS, etc...@@ -1041,17 +1042,161 @@ alignment_more: bl alignment_exception b ret_from_except -/* - * We branch here from entry_64.S for the last stage of the exception - * return code path. MSR:EE is expected to be off at that point - */ -_GLOBAL(exception_return_book3e) - b 1f + .align 7 +_GLOBAL(ret_from_except) + ld r11,_TRAP(r1) + andi. r0,r11,1 + bne ret_from_except_lite + REST_NVGPRS(r1) + +_GLOBAL(ret_from_except_lite) + /* + * Disable interrupts so that current_thread_info()->flags + * can't change between when we test it and when we return + * from the interrupt. + */ + wrteei 0 + + ld r9, PACA_THREAD_INFO(r13) + ld r3,_MSR(r1) + ld r10,PACACURRENT(r13) + ld r4,TI_FLAGS(r9) + andi. r3,r3,MSR_PR + beq resume_kernel + lwz r3,(THREAD+THREAD_DBCR0)(r10) + + /* Check current_thread_info()->flags */ + andi. r0,r4,_TIF_USER_WORK_MASK + bne 1f + /* + * Check to see if the dbcr0 register is set up to debug. + * Use the internal debug mode bit to do this. + */ + andis. r0,r3,DBCR0_IDM@h + beq restore + mfmsr r0 + rlwinm r0,r0,0,~MSR_DE /* Clear MSR.DE */ + mtmsr r0 + mtspr SPRN_DBCR0,r3 + li r10, -1 + mtspr SPRN_DBSR,r10 + b restore +1: andi. r0,r4,_TIF_NEED_RESCHED + beq 2f + bl restore_interrupts + SCHEDULE_USER + b ret_from_except_lite +2: + bl save_nvgprs + /* + * Use a non volatile GPR to save and restore our thread_info flags + * across the call to restore_interrupts. + */ + mr r30,r4 + bl restore_interrupts + mr r4,r30 + addi r3,r1,STACK_FRAME_OVERHEAD + bl do_notify_resume + b ret_from_except + +resume_kernel: + /* check current_thread_info, _TIF_EMULATE_STACK_STORE */ + andis. r8,r4,_TIF_EMULATE_STACK_STORE@h + beq+ 1f + + addi r8,r1,INT_FRAME_SIZE /* Get the kprobed function entry */ + + ld r3,GPR1(r1) + subi r3,r3,INT_FRAME_SIZE /* dst: Allocate a trampoline exception frame */ + mr r4,r1 /* src: current exception frame */ + mr r1,r3 /* Reroute the trampoline frame to r1 */ + + /* Copy from the original to the trampoline. */ + li r5,INT_FRAME_SIZE/8 /* size: INT_FRAME_SIZE */ + li r6,0 /* start offset: 0 */ + mtctr r5 +2: ldx r0,r6,r4 + stdx r0,r6,r3 + addi r6,r6,8 + bdnz 2b + + /* Do real store operation to complete stdu */ + ld r5,GPR1(r1) + std r8,0(r5) + + /* Clear _TIF_EMULATE_STACK_STORE flag */ + lis r11,_TIF_EMULATE_STACK_STORE@h + addi r5,r9,TI_FLAGS +0: ldarx r4,0,r5 + andc r4,r4,r11 + stdcx. r4,0,r5 + bne- 0b +1: + +#ifdef CONFIG_PREEMPT + /* Check if we need to preempt */ + andi. r0,r4,_TIF_NEED_RESCHED + beq+ restore + /* Check that preempt_count() == 0 and interrupts are enabled */ + lwz r8,TI_PREEMPT(r9) + cmpwi cr0,r8,0 + bne restore + ld r0,SOFTE(r1) + andi. r0,r0,IRQS_DISABLED + bne restore + + /* + * Here we are preempting the current task. We want to make + * sure we are soft-disabled first and reconcile irq state. + */ + RECONCILE_IRQ_STATE(r3,r4) + bl preempt_schedule_irq + + /* + * arch_local_irq_restore() from preempt_schedule_irq above may + * enable hard interrupt but we really should disable interrupts + * when we return from the interrupt, and so that we don't get + * interrupted after loading SRR0/1. + */ + wrteei 0 +#endif /* CONFIG_PREEMPT */ + +restore: + /* + * This is the main kernel exit path. First we check if we + * are about to re-enable interrupts + */ + ld r5,SOFTE(r1) + lbz r6,PACAIRQSOFTMASK(r13) + andi. r5,r5,IRQS_DISABLED + bne .Lrestore_irq_off + + /* We are enabling, were we already enabled ? Yes, just return */ + andi. r6,r6,IRQS_DISABLED + beq cr0,fast_exception_return + + /* + * We are about to soft-enable interrupts (we are hard disabled + * at this point). We check if there's anything that needs to + * be replayed first. + */ + lbz r0,PACAIRQHAPPENED(r13) + cmpwi cr0,r0,0 + bne- .Lrestore_check_irq_replay + + /* + * Get here when nothing happened while soft-disabled, just + * soft-enable and move-on. We will hard-enable as a side + * effect of rfi + */ +.Lrestore_no_replay: + TRACE_ENABLE_INTS + li r0,IRQS_ENABLED + stb r0,PACAIRQSOFTMASK(r13); /* This is the return from load_up_fpu fast path which could do with * less GPR restores in fact, but for now we have a single return path */ - .globl fast_exception_return fast_exception_return: wrteei 0 1: mr r0,r13@@ -1092,6 +1237,102 @@ fast_exception_return: mfspr r13,SPRN_SPRG_GEN_SCRATCH rfi + /* + * We are returning to a context with interrupts soft disabled. + * + * However, we may also about to hard enable, so we need to + * make sure that in this case, we also clear PACA_IRQ_HARD_DIS + * or that bit can get out of sync and bad things will happen + */ +.Lrestore_irq_off: + ld r3,_MSR(r1) + lbz r7,PACAIRQHAPPENED(r13) + andi. r0,r3,MSR_EE + beq 1f + rlwinm r7,r7,0,~PACA_IRQ_HARD_DIS + stb r7,PACAIRQHAPPENED(r13) +1: +#if defined(CONFIG_PPC_IRQ_SOFT_MASK_DEBUG) && defined(CONFIG_BUG) + /* The interrupt should not have soft enabled. */ + lbz r7,PACAIRQSOFTMASK(r13) +1: tdeqi r7,IRQS_ENABLED + EMIT_BUG_ENTRY 1b,__FILE__,__LINE__,BUGFLAG_WARNING +#endif + b fast_exception_return + + /* + * Something did happen, check if a re-emit is needed + * (this also clears paca->irq_happened) + */ +.Lrestore_check_irq_replay: + /* XXX: We could implement a fast path here where we check + * for irq_happened being just 0x01, in which case we can + * clear it and return. That means that we would potentially + * miss a decrementer having wrapped all the way around. + * + * Still, this might be useful for things like hash_page + */ + bl __check_irq_replay + cmpwi cr0,r3,0 + beq .Lrestore_no_replay + + /* + * We need to re-emit an interrupt. We do so by re-using our + * existing exception frame. We first change the trap value, + * but we need to ensure we preserve the low nibble of it + */ + ld r4,_TRAP(r1) + clrldi r4,r4,60 + or r4,r4,r3 + std r4,_TRAP(r1) + + /* + * PACA_IRQ_HARD_DIS won't always be set here, so set it now + * to reconcile the IRQ state. Tracing is already accounted for. + */ + lbz r4,PACAIRQHAPPENED(r13) + ori r4,r4,PACA_IRQ_HARD_DIS + stb r4,PACAIRQHAPPENED(r13) + + /* + * Then find the right handler and call it. Interrupts are + * still soft-disabled and we keep them that way. + */ + cmpwi cr0,r3,0x500 + bne 1f + addi r3,r1,STACK_FRAME_OVERHEAD; + bl do_IRQ + b ret_from_except +1: cmpwi cr0,r3,0xf00 + bne 1f + addi r3,r1,STACK_FRAME_OVERHEAD; + bl performance_monitor_exception + b ret_from_except +1: cmpwi cr0,r3,0xe60 + bne 1f + addi r3,r1,STACK_FRAME_OVERHEAD; + bl handle_hmi_exception + b ret_from_except +1: cmpwi cr0,r3,0x900 + bne 1f + addi r3,r1,STACK_FRAME_OVERHEAD; + bl timer_interrupt + b ret_from_except +#ifdef CONFIG_PPC_DOORBELL +1: + cmpwi cr0,r3,0x280 + bne 1f + addi r3,r1,STACK_FRAME_OVERHEAD; + bl doorbell_exception +#endif /* CONFIG_PPC_DOORBELL */ +1: b ret_from_except /* What else to do here ? */ + +_ASM_NOKPROBE_SYMBOL(ret_from_except); +_ASM_NOKPROBE_SYMBOL(ret_from_except_lite); +_ASM_NOKPROBE_SYMBOL(resume_kernel); +_ASM_NOKPROBE_SYMBOL(restore); +_ASM_NOKPROBE_SYMBOL(fast_exception_return); + /* * Trampolines used when spotting a bad kernel stack pointer in * the exception entry code.arch/powerpc/kernel/exceptions-64s.S +45
-74@@ -575,6 +575,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) std r10,GPR12(r1) std r11,GPR13(r1) + SAVE_NVGPRS(r1) + .if IDAR .if IISIDE ld r10,_NIP(r1)@@ -611,7 +613,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR) mfspr r11,SPRN_XER /* save XER in stackframe */ std r10,SOFTE(r1) std r11,_XER(r1) - li r9,(IVEC)+1 + li r9,IVEC std r9,_TRAP(r1) /* set trap number */ li r10,0 ld r11,exception_marker@toc(r2)@@ -918,7 +920,6 @@ EXC_COMMON_BEGIN(system_reset_common) ld r1,PACA_NMI_EMERG_SP(r13) subi r1,r1,INT_FRAME_SIZE __GEN_COMMON_BODY system_reset - bl save_nvgprs /* * Set IRQS_ALL_DISABLED unconditionally so irqs_disabled() does * the right thing. We do not want to reconcile because that goes@@ -1099,7 +1100,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HVMODE) li r10,MSR_RI mtmsrd r10,1 - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl machine_check_early std r3,RESULT(r1) /* Save result */@@ -1192,10 +1192,9 @@ EXC_COMMON_BEGIN(machine_check_common) /* Enable MSR_RI when finished with PACA_EXMC */ li r10,MSR_RI mtmsrd r10,1 - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl machine_check_exception - b ret_from_except + b interrupt_return GEN_KVM machine_check@@ -1362,20 +1361,19 @@ BEGIN_MMU_FTR_SECTION bl do_slb_fault cmpdi r3,0 bne- 1f - b fast_exception_return + b fast_interrupt_return 1: /* Error case */ MMU_FTR_SECTION_ELSE /* Radix case, access is outside page table range */ li r3,-EFAULT ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) std r3,RESULT(r1) - bl save_nvgprs RECONCILE_IRQ_STATE(r10, r11) ld r4,_DAR(r1) ld r5,RESULT(r1) addi r3,r1,STACK_FRAME_OVERHEAD bl do_bad_slb_fault - b ret_from_except + b interrupt_return GEN_KVM data_access_slb@@ -1455,20 +1453,19 @@ BEGIN_MMU_FTR_SECTION bl do_slb_fault cmpdi r3,0 bne- 1f - b fast_exception_return + b fast_interrupt_return 1: /* Error case */ MMU_FTR_SECTION_ELSE /* Radix case, access is outside page table range */ li r3,-EFAULT ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX) std r3,RESULT(r1) - bl save_nvgprs RECONCILE_IRQ_STATE(r10, r11) ld r4,_DAR(r1) ld r5,RESULT(r1) addi r3,r1,STACK_FRAME_OVERHEAD bl do_bad_slb_fault - b ret_from_except + b interrupt_return GEN_KVM instruction_access_slb@@ -1516,7 +1513,7 @@ EXC_COMMON_BEGIN(hardware_interrupt_common) RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD bl do_IRQ - b ret_from_except_lite + b interrupt_return_lite GEN_KVM hardware_interrupt@@ -1542,10 +1539,9 @@ EXC_VIRT_BEGIN(alignment, 0x4600, 0x100) EXC_VIRT_END(alignment, 0x4600, 0x100) EXC_COMMON_BEGIN(alignment_common) GEN_COMMON alignment - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl alignment_exception - b ret_from_except + b interrupt_return GEN_KVM alignment@@ -1606,10 +1602,9 @@ EXC_COMMON_BEGIN(program_check_common) __ISTACK(program_check)=1 __GEN_COMMON_BODY program_check 3: - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl program_check_exception - b ret_from_except + b interrupt_return GEN_KVM program_check@@ -1640,7 +1635,6 @@ EXC_VIRT_END(fp_unavailable, 0x4800, 0x100) EXC_COMMON_BEGIN(fp_unavailable_common) GEN_COMMON fp_unavailable bne 1f /* if from user, just load it up */ - bl save_nvgprs RECONCILE_IRQ_STATE(r10, r11) addi r3,r1,STACK_FRAME_OVERHEAD bl kernel_fp_unavailable_exception@@ -1657,14 +1651,13 @@ BEGIN_FTR_SECTION END_FTR_SECTION_IFSET(CPU_FTR_TM) #endif bl load_up_fpu - b fast_exception_return + b fast_interrupt_return #ifdef CONFIG_PPC_TRANSACTIONAL_MEM 2: /* User process was in a transaction */ - bl save_nvgprs RECONCILE_IRQ_STATE(r10, r11) addi r3,r1,STACK_FRAME_OVERHEAD bl fp_unavailable_tm - b ret_from_except + b interrupt_return #endif GEN_KVM fp_unavailable@@ -1707,7 +1700,7 @@ EXC_COMMON_BEGIN(decrementer_common) RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD bl timer_interrupt - b ret_from_except_lite + b interrupt_return_lite GEN_KVM decrementer@@ -1798,7 +1791,7 @@ EXC_COMMON_BEGIN(doorbell_super_common) #else bl unknown_exception #endif - b ret_from_except_lite + b interrupt_return_lite GEN_KVM doorbell_super@@ -1970,10 +1963,9 @@ EXC_VIRT_BEGIN(single_step, 0x4d00, 0x100) EXC_VIRT_END(single_step, 0x4d00, 0x100) EXC_COMMON_BEGIN(single_step_common) GEN_COMMON single_step - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl single_step_exception - b ret_from_except + b interrupt_return GEN_KVM single_step@@ -2008,7 +2000,6 @@ EXC_VIRT_BEGIN(h_data_storage, 0x4e00, 0x20) EXC_VIRT_END(h_data_storage, 0x4e00, 0x20) EXC_COMMON_BEGIN(h_data_storage_common) GEN_COMMON h_data_storage - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD BEGIN_MMU_FTR_SECTION ld r4,_DAR(r1)@@ -2017,7 +2008,7 @@ BEGIN_MMU_FTR_SECTION MMU_FTR_SECTION_ELSE bl unknown_exception ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_TYPE_RADIX) - b ret_from_except + b interrupt_return GEN_KVM h_data_storage@@ -2042,10 +2033,9 @@ EXC_VIRT_BEGIN(h_instr_storage, 0x4e20, 0x20) EXC_VIRT_END(h_instr_storage, 0x4e20, 0x20) EXC_COMMON_BEGIN(h_instr_storage_common) GEN_COMMON h_instr_storage - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl unknown_exception - b ret_from_except + b interrupt_return GEN_KVM h_instr_storage@@ -2068,10 +2058,9 @@ EXC_VIRT_BEGIN(emulation_assist, 0x4e40, 0x20) EXC_VIRT_END(emulation_assist, 0x4e40, 0x20) EXC_COMMON_BEGIN(emulation_assist_common) GEN_COMMON emulation_assist - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl emulation_assist_interrupt - b ret_from_except + b interrupt_return GEN_KVM emulation_assist@@ -2151,10 +2140,9 @@ EXC_COMMON_BEGIN(hmi_exception_common) GEN_COMMON hmi_exception FINISH_NAP RUNLATCH_ON - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl handle_hmi_exception - b ret_from_except + b interrupt_return GEN_KVM hmi_exception@@ -2188,7 +2176,7 @@ EXC_COMMON_BEGIN(h_doorbell_common) #else bl unknown_exception #endif - b ret_from_except_lite + b interrupt_return_lite GEN_KVM h_doorbell@@ -2218,7 +2206,7 @@ EXC_COMMON_BEGIN(h_virt_irq_common) RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD bl do_IRQ - b ret_from_except_lite + b interrupt_return_lite GEN_KVM h_virt_irq@@ -2265,7 +2253,7 @@ EXC_COMMON_BEGIN(performance_monitor_common) RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD bl performance_monitor_exception - b ret_from_except_lite + b interrupt_return_lite GEN_KVM performance_monitor@@ -2305,23 +2293,21 @@ BEGIN_FTR_SECTION END_FTR_SECTION_NESTED(CPU_FTR_TM, CPU_FTR_TM, 69) #endif bl load_up_altivec - b fast_exception_return + b fast_interrupt_return #ifdef CONFIG_PPC_TRANSACTIONAL_MEM 2: /* User process was in a transaction */ - bl save_nvgprs RECONCILE_IRQ_STATE(r10, r11) addi r3,r1,STACK_FRAME_OVERHEAD bl altivec_unavailable_tm - b ret_from_except + b interrupt_return #endif 1: END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) #endif - bl save_nvgprs RECONCILE_IRQ_STATE(r10, r11) addi r3,r1,STACK_FRAME_OVERHEAD bl altivec_unavailable_exception - b ret_from_except + b interrupt_return GEN_KVM altivec_unavailable@@ -2363,20 +2349,18 @@ BEGIN_FTR_SECTION b load_up_vsx #ifdef CONFIG_PPC_TRANSACTIONAL_MEM 2: /* User process was in a transaction */ - bl save_nvgprs RECONCILE_IRQ_STATE(r10, r11) addi r3,r1,STACK_FRAME_OVERHEAD bl vsx_unavailable_tm - b ret_from_except + b interrupt_return #endif 1: END_FTR_SECTION_IFSET(CPU_FTR_VSX) #endif - bl save_nvgprs RECONCILE_IRQ_STATE(r10, r11) addi r3,r1,STACK_FRAME_OVERHEAD bl vsx_unavailable_exception - b ret_from_except + b interrupt_return GEN_KVM vsx_unavailable@@ -2403,10 +2387,9 @@ EXC_VIRT_BEGIN(facility_unavailable, 0x4f60, 0x20) EXC_VIRT_END(facility_unavailable, 0x4f60, 0x20) EXC_COMMON_BEGIN(facility_unavailable_common) GEN_COMMON facility_unavailable - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl facility_unavailable_exception - b ret_from_except + b interrupt_return GEN_KVM facility_unavailable@@ -2433,10 +2416,9 @@ EXC_VIRT_BEGIN(h_facility_unavailable, 0x4f80, 0x20) EXC_VIRT_END(h_facility_unavailable, 0x4f80, 0x20) EXC_COMMON_BEGIN(h_facility_unavailable_common) GEN_COMMON h_facility_unavailable - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl facility_unavailable_exception - b ret_from_except + b interrupt_return GEN_KVM h_facility_unavailable@@ -2467,10 +2449,9 @@ EXC_REAL_END(cbe_system_error, 0x1200, 0x100) EXC_VIRT_NONE(0x5200, 0x100) EXC_COMMON_BEGIN(cbe_system_error_common) GEN_COMMON cbe_system_error - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl cbe_system_error_exception - b ret_from_except + b interrupt_return GEN_KVM cbe_system_error@@ -2496,10 +2477,9 @@ EXC_VIRT_BEGIN(instruction_breakpoint, 0x5300, 0x100) EXC_VIRT_END(instruction_breakpoint, 0x5300, 0x100) EXC_COMMON_BEGIN(instruction_breakpoint_common) GEN_COMMON instruction_breakpoint - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl instruction_breakpoint_exception - b ret_from_except + b interrupt_return GEN_KVM instruction_breakpoint@@ -2619,10 +2599,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR) EXC_COMMON_BEGIN(denorm_exception_common) GEN_COMMON denorm_exception - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl unknown_exception - b ret_from_except + b interrupt_return GEN_KVM denorm_exception@@ -2641,10 +2620,9 @@ EXC_REAL_END(cbe_maintenance, 0x1600, 0x100) EXC_VIRT_NONE(0x5600, 0x100) EXC_COMMON_BEGIN(cbe_maintenance_common) GEN_COMMON cbe_maintenance - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl cbe_maintenance_exception - b ret_from_except + b interrupt_return GEN_KVM cbe_maintenance@@ -2669,14 +2647,13 @@ EXC_VIRT_BEGIN(altivec_assist, 0x5700, 0x100) EXC_VIRT_END(altivec_assist, 0x5700, 0x100) EXC_COMMON_BEGIN(altivec_assist_common) GEN_COMMON altivec_assist - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD #ifdef CONFIG_ALTIVEC bl altivec_assist_exception #else bl unknown_exception #endif - b ret_from_except + b interrupt_return GEN_KVM altivec_assist@@ -2695,10 +2672,9 @@ EXC_REAL_END(cbe_thermal, 0x1800, 0x100) EXC_VIRT_NONE(0x5800, 0x100) EXC_COMMON_BEGIN(cbe_thermal_common) GEN_COMMON cbe_thermal - bl save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD bl cbe_thermal_exception - b ret_from_except + b interrupt_return GEN_KVM cbe_thermal@@ -2731,7 +2707,6 @@ EXC_COMMON_BEGIN(soft_nmi_common) ld r1,PACAEMERGSP(r13) subi r1,r1,INT_FRAME_SIZE __GEN_COMMON_BODY soft_nmi - bl save_nvgprs /* * Set IRQS_ALL_DISABLED and save PACAIRQHAPPENED (see@@ -3063,7 +3038,7 @@ do_hash_page: cmpdi r3,0 /* see if __hash_page succeeded */ /* Success */ - beq fast_exc_return_irq /* Return from exception on success */ + beq interrupt_return_lite /* Return from exception on success */ /* Error */ blt- 13f@@ -3080,17 +3055,15 @@ handle_page_fault: addi r3,r1,STACK_FRAME_OVERHEAD bl do_page_fault cmpdi r3,0 - beq+ ret_from_except_lite - bl save_nvgprs + beq+ interrupt_return_lite mr r5,r3 addi r3,r1,STACK_FRAME_OVERHEAD ld r4,_DAR(r1) bl bad_page_fault - b ret_from_except + b interrupt_return /* We have a data breakpoint exception - handle it */ handle_dabr_fault: - bl save_nvgprs ld r4,_DAR(r1) ld r5,_DSISR(r1) addi r3,r1,STACK_FRAME_OVERHEAD@@ -3098,21 +3071,20 @@ handle_dabr_fault: /* * do_break() may have changed the NV GPRS while handling a breakpoint. * If so, we need to restore them with their updated values. Don't use - * ret_from_except_lite here. + * interrupt_return_lite here. */ - b ret_from_except + b interrupt_return #ifdef CONFIG_PPC_BOOK3S_64 /* We have a page fault that hash_page could handle but HV refused * the PTE insertion */ -13: bl save_nvgprs - mr r5,r3 +13: mr r5,r3 addi r3,r1,STACK_FRAME_OVERHEAD ld r4,_DAR(r1) bl low_hash_fault - b ret_from_except + b interrupt_return #endif /*@@ -3122,11 +3094,10 @@ handle_dabr_fault: * were soft-disabled. We want to invoke the exception handler for * the access, or panic if there isn't a handler. */ -77: bl save_nvgprs - addi r3,r1,STACK_FRAME_OVERHEAD +77: addi r3,r1,STACK_FRAME_OVERHEAD li r5,SIGSEGV bl bad_page_fault - b ret_from_except + b interrupt_return /* * When doorbell is triggered from system reset wakeup, the message isarch/powerpc/kernel/irq.c +4
-32@@ -110,6 +110,8 @@ static inline notrace int decrementer_check_overflow(void)return now >= *next_tb; } +#ifdef CONFIG_PPC_BOOK3E +/* This is called whenever we are re-enabling interrupts * and returns either 0 (nothing to do) or 500/900/280/a00/e80 if * there's an EE, DEC or DBELL to generate. @@ -169,41 +171,16 @@ notrace unsigned int __check_irq_replay(void) } } -/* -* Force the delivery of pending soft-disabled interrupts on PS3. -* Any HV call will have this side effect. -*/ -if (firmware_has_feature (FW_FEATURE_PS3_LV1 )) { -u64 tmp, tmp2; -lv1_get_version_info (&tmp, &tmp2); - } - -/* -* Check if an hypervisor Maintenance interrupt happened. -* This is a higher priority interrupt than the others, so -* replay it first. -*/ -if (happened &PACA_IRQ_HMI ) { - local_paca->irq_happened &= ~PACA_IRQ_HMI ; -return 0xe60 ; - } -if (happened &PACA_IRQ_DEC ) { local_paca->irq_happened &= ~PACA_IRQ_DEC ;return 0x900 ; } -if (happened &PACA_IRQ_PMI ) { - local_paca->irq_happened &= ~PACA_IRQ_PMI ; -return 0xf00 ; - } -if (happened &PACA_IRQ_EE ) { local_paca->irq_happened &= ~PACA_IRQ_EE ;return 0x500 ; } -#ifdef CONFIG_PPC_BOOK3E /* * Check if an EPR external interrupt happened this bit is typically * set if we need to handle another "edge" interrupt from within the @@ -218,20 +195,15 @@ notrace unsigned int __check_irq_replay(void) local_paca->irq_happened &= ~PACA_IRQ_DBELL ; return 0x280; } -#else -if (happened &PACA_IRQ_DBELL ) { - local_paca->irq_happened &= ~PACA_IRQ_DBELL ; -return 0xa00 ; - } -#endif /* CONFIG_PPC_BOOK3E */ /* There should be nothing left ! */ BUG_ON (local_paca->irq_happened !=0 );return 0 ; } +#endif /* CONFIG_PPC_BOOK3E */ -static void replay_soft_interrupts (void ) +void replay_soft_interrupts (void ) {/* * We use local_paca rather than get_paca() to avoid all arch/powerpc/kernel/process.c +48
-41@@ -236,23 +236,9 @@ void enable_kernel_fp(void) } }EXPORT_SYMBOL (enable_kernel_fp); - -static int restore_fp (struct task_struct *tsk) -{ -if (tsk->thread.load_fp) { -load_fp_state (¤t->thread.fp_state); - current->thread.load_fp++; -return 1 ; - } -return 0 ; -} -#else -static int restore_fp (struct task_struct *tsk) {return 0 ; }#endif /* CONFIG_PPC_FPU */ #ifdef CONFIG_ALTIVEC -#define loadvec (thr) ((thr).load_vec) -static void __giveup_altivec (struct task_struct *tsk) {unsigned long msr;@@ -318,21 +304,6 @@ void flush_altivec_to_thread(struct task_struct *tsk) } }EXPORT_SYMBOL_GPL (flush_altivec_to_thread); - -static int restore_altivec (struct task_struct *tsk) -{ -if (cpu_has_feature (CPU_FTR_ALTIVEC ) && (tsk->thread.load_vec)) { -load_vr_state (&tsk->thread.vr_state); - tsk->thread.used_vr =1 ; - tsk->thread.load_vec++; - -return 1 ; - } -return 0 ; -} -#else -#define loadvec (thr) 0 -static inline int restore_altivec (struct task_struct *tsk) {return 0 ; }#endif /* CONFIG_ALTIVEC */ #ifdef CONFIG_VSX @@ -400,18 +371,6 @@ void flush_vsx_to_thread(struct task_struct *tsk) } }EXPORT_SYMBOL_GPL (flush_vsx_to_thread); - -static int restore_vsx (struct task_struct *tsk) -{ -if (cpu_has_feature (CPU_FTR_VSX )) { - tsk->thread.used_vsr =1 ; -return 1 ; - } - -return 0 ; -} -#else -static inline int restore_vsx (struct task_struct *tsk) {return 0 ; }#endif /* CONFIG_VSX */ #ifdef CONFIG_SPE @@ -511,6 +470,53 @@ void giveup_all(struct task_struct *tsk) }EXPORT_SYMBOL (giveup_all); +#ifdef CONFIG_PPC_BOOK3S_64 +#ifdef CONFIG_PPC_FPU +static int restore_fp (struct task_struct *tsk) +{ +if (tsk->thread.load_fp) { +load_fp_state (¤t->thread.fp_state); + current->thread.load_fp++; +return 1 ; + } +return 0 ; +} +#else +static int restore_fp (struct task_struct *tsk) {return 0 ; } +#endif /* CONFIG_PPC_FPU */ + +#ifdef CONFIG_ALTIVEC +#define loadvec (thr) ((thr).load_vec) +static int restore_altivec (struct task_struct *tsk) +{ +if (cpu_has_feature (CPU_FTR_ALTIVEC ) && (tsk->thread.load_vec)) { +load_vr_state (&tsk->thread.vr_state); + tsk->thread.used_vr =1 ; + tsk->thread.load_vec++; + +return 1 ; + } +return 0 ; +} +#else +#define loadvec (thr) 0 +static inline int restore_altivec (struct task_struct *tsk) {return 0 ; } +#endif /* CONFIG_ALTIVEC */ + +#ifdef CONFIG_VSX +static int restore_vsx (struct task_struct *tsk) +{ +if (cpu_has_feature (CPU_FTR_VSX )) { + tsk->thread.used_vsr =1 ; +return 1 ; + } + +return 0 ; +} +#else +static inline int restore_vsx (struct task_struct *tsk) {return 0 ; } +#endif /* CONFIG_VSX */ +/* * The exception exit path calls restore_math() with interrupts hard disabled * but the soft irq state not "reconciled". ftrace code that calls @@ -551,6 +557,7 @@ void notrace restore_math(struct pt_regs *regs) regs->msr = msr; } +#endif static void save_all (struct task_struct *tsk) {arch/powerpc/kernel/syscall_64.c +163
-1@@ -26,7 +26,11 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7,unsigned long ti_flags;syscall_fn f; +if (IS_ENABLED (CONFIG_PPC_BOOK3S )) +BUG_ON (!(regs->msr &MSR_RI ));BUG_ON (!(regs->msr &MSR_PR )); +BUG_ON (!FULL_REGS (regs)); +BUG_ON (regs->softe !=IRQS_ENABLED );if (IS_ENABLED (CONFIG_PPC_TRANSACTIONAL_MEM ) &&unlikely (regs->msr &MSR_TS_T ))@@ -195,7 +199,7 @@ notrace unsigned long syscall_exit_prepare(unsigned long r3,trace_hardirqs_off (); local_paca->irq_happened |=PACA_IRQ_HARD_DIS ;local_irq_enable (); -/* Took an interrupt which may have more exit work to do. */ +/* Took an interrupt, may have more exit work to do. */ goto again; } local_paca->irq_happened =0 ;@@ -211,3 +215,161 @@ notrace unsigned long syscall_exit_prepare(unsigned long r3,return ret; } + +#ifdef CONFIG_PPC_BOOK3S /* BOOK3E not yet using this */ +notrace unsigned long interrupt_exit_user_prepare (struct pt_regs *regs,unsigned long msr) +{ +#ifdef CONFIG_PPC_BOOK3E +struct thread_struct *ts = ¤t->thread; +#endif +unsigned long *ti_flagsp = ¤t_thread_info ()->flags; +unsigned long ti_flags; +unsigned long flags; +unsigned long ret =0 ; + +if (IS_ENABLED (CONFIG_PPC_BOOK3S )) +BUG_ON (!(regs->msr &MSR_RI )); +BUG_ON (!(regs->msr &MSR_PR )); +BUG_ON (!FULL_REGS (regs)); +BUG_ON (regs->softe !=IRQS_ENABLED ); + +local_irq_save (flags); + +again: + ti_flags =READ_ONCE (*ti_flagsp); +while (unlikely (ti_flags & (_TIF_USER_WORK_MASK & ~_TIF_RESTORE_TM))) { +local_irq_enable ();/* returning to user: may enable */ +if (ti_flags & _TIF_NEED_RESCHED) { +schedule (); + }else { +if (ti_flags & _TIF_SIGPENDING) + ret |= _TIF_RESTOREALL; +do_notify_resume (regs, ti_flags); + } +local_irq_disable (); + ti_flags =READ_ONCE (*ti_flagsp); + } + +if (IS_ENABLED (CONFIG_PPC_BOOK3S )) { +unsigned long mathflags =0 ; + +if (IS_ENABLED (CONFIG_PPC_FPU )) + mathflags |=MSR_FP ; +if (IS_ENABLED (CONFIG_ALTIVEC )) + mathflags |=MSR_VEC ; + +if (IS_ENABLED (CONFIG_PPC_TRANSACTIONAL_MEM ) && + (ti_flags & _TIF_RESTORE_TM)) +restore_tm_state (regs); +else if ((regs->msr & mathflags) != mathflags) +restore_math (regs); + } + +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (unlikely (lazy_irq_pending ())) { +__hard_RI_enable (); +trace_hardirqs_off (); + local_paca->irq_happened |=PACA_IRQ_HARD_DIS ; +local_irq_enable (); +local_irq_disable (); +/* Took an interrupt, may have more exit work to do. */ + goto again; + } + local_paca->irq_happened =0 ; +irq_soft_mask_set (IRQS_ENABLED ); + +#ifdef CONFIG_PPC_BOOK3E +if (unlikely (ts->debug.dbcr0 &DBCR0_IDM )) { +/* +* Check to see if the dbcr0 register is set up to debug. +* Use the internal debug mode bit to do this. +*/ +mtmsr (mfmsr () & ~MSR_DE ); +mtspr (SPRN_DBCR0 , ts->debug.dbcr0); +mtspr (SPRN_DBSR ,-1 ); + } +#endif + +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif + +kuap_check_amr (); + +account_cpu_user_exit (); + +return ret; +} + +void unrecoverable_exception (struct pt_regs *regs); +void preempt_schedule_irq (void ); + +notrace unsigned long interrupt_exit_kernel_prepare (struct pt_regs *regs,unsigned long msr) +{ +unsigned long *ti_flagsp = ¤t_thread_info ()->flags; +unsigned long flags; + +if (IS_ENABLED (CONFIG_PPC_BOOK3S ) &&unlikely (!(regs->msr &MSR_RI ))) +unrecoverable_exception (regs); +BUG_ON (regs->msr &MSR_PR ); +BUG_ON (!FULL_REGS (regs)); + +local_irq_save (flags); + +if (regs->softe ==IRQS_ENABLED ) { +/* Returning to a kernel context with local irqs enabled. */ +WARN_ON_ONCE (!(regs->msr &MSR_EE )); +again: +if (IS_ENABLED (CONFIG_PREEMPT )) { +/* Return to preemptible kernel context */ +if (unlikely (*ti_flagsp & _TIF_NEED_RESCHED)) { +if (preempt_count () ==0 ) +preempt_schedule_irq (); + } + } + +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (unlikely (lazy_irq_pending ())) { +__hard_RI_enable (); +irq_soft_mask_set (IRQS_ALL_DISABLED ); +trace_hardirqs_off (); + local_paca->irq_happened |=PACA_IRQ_HARD_DIS ; +/* +* Can't local_irq_enable in case we are in interrupt +* context. Must replay directly. +*/ +replay_soft_interrupts (); +irq_soft_mask_set (flags); +/* Took an interrupt, may have more exit work to do. */ + goto again; + } + local_paca->irq_happened =0 ; +irq_soft_mask_set (IRQS_ENABLED ); + }else { +/* Returning to a kernel context with local irqs disabled. */ +trace_hardirqs_on (); +__hard_EE_RI_disable (); +if (regs->msr &MSR_EE ) + local_paca->irq_happened &= ~PACA_IRQ_HARD_DIS ; + } + + +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif + +/* +* We don't need to restore AMR on the way back to userspace for KUAP. +* The value of AMR only matters while we're in the kernel. +*/ +kuap_restore_amr (regs); + +if (unlikely (*ti_flagsp & _TIF_EMULATE_STACK_STORE)) { +clear_bits (_TIF_EMULATE_STACK_STORE, ti_flagsp); +return 1 ; + } +return 0 ; +} +#endif arch/powerpc/kernel/vector.S +1
-1@@ -134,7 +134,7 @@ _GLOBAL(load_up_vsx) /* enable use of VSX after return */ oris r12,r12,MSR_VSX@h std r12,_MSR(r1) - b fast_exception_return + b fast_interrupt_return #endif /* CONFIG_VSX */
Nicholas Piggin
reviewerRe: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Excerpts from Christophe Leroy's message of February 27, 2021 8:07 pm:
Quoted 36 lines
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 7 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
The stack store emulation is significantly simplfied, rather than creating a new return frame and switching to that before performing the store, it uses the PACA to keep a scratch register around to perform thestore.
Quoted 8 lines
The asm return code is moved into 64e for now. The new logic has made allowance for 64e, but I don't have a full environment that works well to test it, and even booting in emulated qemu is not great for stress testing. 64e shouldn't be too far off working with this, given a bit more testing and auditing of the logic.
This is slightly faster on a POWER9 (page fault speed increases about 1.1%), probably due to reduced mtmsrd.
This series, and especially this patch has added a awfull number of
BUG_ON()traps.We have an issue open at https://github.com/linuxppc/issues/issues/88 since 2017 for reducing the number of BUG_ON()s
And the kernel Documentation is explicit on the willingness to deprecate
BUG_ON(), see
https://www.kernel.org/doc/html/latest/process/deprecated.html?highlight=bug_on :
BUG()andBUG_ON()
UseWARN()andWARN_ON()instead, and handle the “impossible” error condition as gracefully as possible. While the BUG()-family of APIs were originally designed to act as an “impossible situation” assert and to kill a kernel thread “safely”, they turn out to just be too risky. (e.g. “In what order do locks need to be released? Have various states been restored?”) Very commonly, usingBUG()will destabilize a system or entirely break it, which makes it impossible to debug or even get viable crash reports. Linus has very strong feelings about this.
Quoted 1 line
So ... can we do something cleaner with all the BUG_ON()s recently added ?
Yeah you're right. Some of it is probably overkill due to paranoia when developing the series.
Now we have a bit more confidence we could probably look at cutting down on these.
I do get a bit concerned about detecting a problem in some code like this and attempting to just continue, it usually means the system is going to crash pretty badly anyway (and the WARN_ON trap interrupt is probably going to finish you off anyway). So I think removing the more obvious checks entirely (maybe with a PPC DEBUG config option) is the right way to go.
Thanks,
Nick
Re: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
In reply to Nicholas PigginLe 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 19 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
The stack store emulation is significantly simplfied, rather than creating a new return frame and switching to that before performing the store, it uses the PACA to keep a scratch register around to perform thestore.
The asm return code is moved into 64e for now. The new logic has made allowance for 64e, but I don't have a full environment that works well to test it, and even booting in emulated qemu is not great for stress testing. 64e shouldn't be too far off working with this, given a bit more testing and auditing of the logic.
This is slightly faster on a POWER9 (page fault speed increases about 1.1%), probably due to reduced mtmsrd.
Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michal Suchanek <[email protected]>---
...
Quoted 2 lines
arch/powerpc/kernel/syscall_64.c +2
-0+notrace unsigned long interrupt_exit_user_prepare (struct pt_regs *regs,unsigned long msr) +{
...
arch/powerpc/kernel/syscall_64.c notrace unsigned long syscall_exit_prepare(unsigned long r3, · line 296
Quoted 4 lines
arch/powerpc/kernel/syscall_64.c +4
-0+ +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif
Could we define a helper for that in asm/tm.h, that voids when CONFIG_PPC_TRANSACTIONAL_MEM is not selected ?
Quoted 14 lines
arch/powerpc/kernel/syscall_64.c +14
-0+ +kuap_check_amr (); + +account_cpu_user_exit (); + +return ret; +} + +void unrecoverable_exception (struct pt_regs *regs); +void preempt_schedule_irq (void ); + +notrace unsigned long interrupt_exit_kernel_prepare (struct pt_regs *regs,unsigned long msr) +{ +
...
Quoted 7 lines
arch/powerpc/kernel/syscall_64.c +7
-0+#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif + +/* +* We don't need to restore AMR on the way back to userspace for KUAP. +* The value of AMR only matters while we're in the kernel.
arch/powerpc/kernel/syscall_64.c notrace unsigned long syscall_exit_prepare(unsigned long r3, · line 375
Quoted 10 lines
arch/powerpc/kernel/syscall_64.c +10
-0+*/ +kuap_restore_amr (regs); + +if (unlikely (*ti_flagsp & _TIF_EMULATE_STACK_STORE)) { +clear_bits (_TIF_EMULATE_STACK_STORE, ti_flagsp); +return 1 ; + } + return 0; +} +#endif
Christophe
Nicholas Piggin
reviewerRe: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Excerpts from Christophe Leroy's message of March 15, 2021 11:41 pm:
Quoted 30 lines
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 19 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
The stack store emulation is significantly simplfied, rather than creating a new return frame and switching to that before performing the store, it uses the PACA to keep a scratch register around to perform thestore.
The asm return code is moved into 64e for now. The new logic has made allowance for 64e, but I don't have a full environment that works well to test it, and even booting in emulated qemu is not great for stress testing. 64e shouldn't be too far off working with this, given a bit more testing and auditing of the logic.
This is slightly faster on a POWER9 (page fault speed increases about 1.1%), probably due to reduced mtmsrd.
Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michal Suchanek <[email protected]>---
...
Quoted 2 lines
arch/powerpc/kernel/syscall_64.c +2
-0+notrace unsigned long interrupt_exit_user_prepare (struct pt_regs *regs,unsigned long msr) +{...
Quoted 7 lines
arch/powerpc/kernel/syscall_64.c notrace unsigned long syscall_exit_prepare(unsigned long r3, · line 296
Quoted 4 lines
arch/powerpc/kernel/syscall_64.c +4
-0+ +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif Could we define a helper for that in
asm/tm.h, that voids whenCONFIG_PPC_TRANSACTIONAL_MEMis not selected ?
Yeah I wanted to do something about that. I don't know what it's used for here. I guess it saves the return MSR so if that causes a crash then the next oops would see it, but I wonder if we can just get that from SRR1 + program check error codes, or if there is something we can't reconstruct from there. Have to check with someone who knows TM better.
Thanks,
Nick
Michael Ellerman
reviewerRe: [PATCH v3 28/32] powerpc/64s: interrupt implement exit logic in C
Quoted 38 lines · Nicholas Piggin
Excerpts from Christophe Leroy's message of March 15, 2021 11:41 pm:
Quoted 29 lines
Le 25/02/2020 à 18:35, Nicholas Piggin a écrit :
Quoted 19 lines
Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store.
The stack store emulation is significantly simplfied, rather than creating a new return frame and switching to that before performing the store, it uses the PACA to keep a scratch register around to perform thestore.
The asm return code is moved into 64e for now. The new logic has made allowance for 64e, but I don't have a full environment that works well to test it, and even booting in emulated qemu is not great for stress testing. 64e shouldn't be too far off working with this, given a bit more testing and auditing of the logic.
This is slightly faster on a POWER9 (page fault speed increases about 1.1%), probably due to reduced mtmsrd.
Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michal Suchanek <[email protected]>---
...
Quoted 2 lines
arch/powerpc/kernel/syscall_64.c +2
-0+notrace unsigned long interrupt_exit_user_prepare (struct pt_regs *regs,unsigned long msr) +{...
Quoted 7 lines
arch/powerpc/kernel/syscall_64.c notrace unsigned long syscall_exit_prepare(unsigned long r3, · line 296
Quoted 4 lines
arch/powerpc/kernel/syscall_64.c +4
-0+ +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + local_paca->tm_scratch = regs->msr; +#endif Could we define a helper for that in
asm/tm.h, that voids whenCONFIG_PPC_TRANSACTIONAL_MEMis not selected ?
Quoted 5 lines
Yeah I wanted to do something about that. I don't know what it's used for here. I guess it saves the return MSR so if that causes a crash then the next oops would see it, but I wonder if we can just get that from SRR1 + program check error codes, or if there is something we can't reconstruct from there.
In the cases when you need it, you can't reconstruct it :)
But given the TM code is on life support we could probably drop tm_scratch.
I don't think we've used it in anger for several years. Probably since 265e60a170d0 ("powerpc/64s: Use emergency stack for kernel TM Bad Thing program checks") (Oct 2017).
If one of us has to debug some hairy TM issue we can always add it back temporarily in a dev kernel.
cheers
Nicholas Piggin
reviewerPATCH v3 29/32 powerpc/64s/exception: remove lite interrupt return
In reply to Nicholas PigginThe difference between lite and regular returns is that the lite case restores all NVGPRs, whereas lite skips that. This is quite clumsy though, most interrupts want the NVGPRs saved for debugging, not to modify in the caller, so the NVGPRs restore is not necessary most of the time. Restore NVGPRs explicitly for one case that requires it, and move everything else over to avoiding the restore unless the interrupt return demands it (e.g., handling a signal).
---
v3:
- Add a copule of missing restore cases for instruction emulation
arch/powerpc/kernel/entry_64.S | 6 ------ arch/powerpc/kernel/exceptions-64s.S | 24 ++++++++++++++---------- 2 files changed, 14 insertions(+), 16 deletions(-)
arch/powerpc/kernel/entry_64.S +0 -6
@@ -471,12 +471,6 @@ _ASM_NOKPROBE_SYMBOL(fast_interrupt_return) .globl interrupt_return interrupt_return: _ASM_NOKPROBE_SYMBOL(interrupt_return) - REST_NVGPRS(r1) - - .balign IFETCH_ALIGN_BYTES - .globl interrupt_return_lite -interrupt_return_lite: -_ASM_NOKPROBE_SYMBOL(interrupt_return_lite) ld r4,_MSR(r1) andi. r0,r4,MSR_PR beq .Lkernel_interrupt_return
arch/powerpc/kernel/exceptions-64s.S +14 -10
@@ -1513,7 +1513,7 @@ EXC_COMMON_BEGIN(hardware_interrupt_common) RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD bl do_IRQ - b interrupt_return_lite + b interrupt_return GEN_KVM hardware_interrupt
@@ -1541,6 +1541,7 @@ EXC_COMMON_BEGIN(alignment_common) GEN_COMMON alignment addi r3,r1,STACK_FRAME_OVERHEAD bl alignment_exception + REST_NVGPRS(r1) /* instruction emulation may change GPRs */ b interrupt_return GEN_KVM alignment
@@ -1604,6 +1605,7 @@ EXC_COMMON_BEGIN(program_check_common) 3: addi r3,r1,STACK_FRAME_OVERHEAD bl program_check_exception + REST_NVGPRS(r1) /* instruction emulation may change GPRs */ b interrupt_return GEN_KVM program_check
@@ -1700,7 +1702,7 @@ EXC_COMMON_BEGIN(decrementer_common) RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD bl timer_interrupt - b interrupt_return_lite + b interrupt_return GEN_KVM decrementer
@@ -1791,7 +1793,7 @@ EXC_COMMON_BEGIN(doorbell_super_common) #else bl unknown_exception #endif - b interrupt_return_lite + b interrupt_return GEN_KVM doorbell_super
@@ -2060,6 +2062,7 @@ EXC_COMMON_BEGIN(emulation_assist_common) GEN_COMMON emulation_assist addi r3,r1,STACK_FRAME_OVERHEAD bl emulation_assist_interrupt + REST_NVGPRS(r1) /* instruction emulation may change GPRs */ b interrupt_return GEN_KVM emulation_assist
@@ -2176,7 +2179,7 @@ EXC_COMMON_BEGIN(h_doorbell_common) #else bl unknown_exception #endif - b interrupt_return_lite + b interrupt_return GEN_KVM h_doorbell
@@ -2206,7 +2209,7 @@ EXC_COMMON_BEGIN(h_virt_irq_common) RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD bl do_IRQ - b interrupt_return_lite + b interrupt_return GEN_KVM h_virt_irq
@@ -2253,7 +2256,7 @@ EXC_COMMON_BEGIN(performance_monitor_common) RUNLATCH_ON addi r3,r1,STACK_FRAME_OVERHEAD bl performance_monitor_exception - b interrupt_return_lite + b interrupt_return GEN_KVM performance_monitor
@@ -2650,6 +2653,7 @@ EXC_COMMON_BEGIN(altivec_assist_common) addi r3,r1,STACK_FRAME_OVERHEAD #ifdef CONFIG_ALTIVEC bl altivec_assist_exception + REST_NVGPRS(r1) /* instruction emulation may change GPRs */ #else bl unknown_exception #endif
@@ -3038,7 +3042,7 @@ do_hash_page: cmpdi r3,0 /* see if __hash_page succeeded */ /* Success */ - beq interrupt_return_lite /* Return from exception on success */ + beq interrupt_return /* Return from exception on success */ /* Error */ blt- 13f
@@ -3055,7 +3059,7 @@ handle_page_fault: addi r3,r1,STACK_FRAME_OVERHEAD bl do_page_fault cmpdi r3,0 - beq+ interrupt_return_lite + beq+ interrupt_return mr r5,r3 addi r3,r1,STACK_FRAME_OVERHEAD ld r4,_DAR(r1)
@@ -3070,9 +3074,9 @@ handle_dabr_fault: bl do_break /* * do_break() may have changed the NV GPRS while handling a breakpoint. - * If so, we need to restore them with their updated values. Don't use - * interrupt_return_lite here. + * If so, we need to restore them with their updated values. */ + REST_NVGPRS(r1) b interrupt_return
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 30/32 powerpc/64: system call reconcile interrupts
In reply to Nicholas PigginThis reconciles interrupts in the system call case like all other interrupts. This allows system_call_common to be shared with the scv system call implementation in a subsequent patch.
--- arch/powerpc/kernel/entry_64.S | 11 +++++++++++ arch/powerpc/kernel/syscall_64.c | 28 +++++++++++++--------------- 2 files changed, 24 insertions(+), 15 deletions(-)
arch/powerpc/kernel/entry_64.S +11 -0
@@ -113,6 +113,17 @@ END_BTB_FLUSH_SECTION ld r11,exception_marker@toc(r2) std r11,-16(r10) /* "regshere" marker */ + /* + * RECONCILE_IRQ_STATE without calling trace_hardirqs_off(), which + * would clobber syscall parameters. Also we always enter with IRQs + * enabled and nothing pending. system_call_exception() will call + * trace_hardirqs_off(). + */ + li r11,IRQS_ALL_DISABLED + li r12,PACA_IRQ_HARD_DIS + stb r11,PACAIRQSOFTMASK(r13) + stb r12,PACAIRQHAPPENED(r13) + /* Calling convention has r9 = orig r0, r10 = regs */ mr r9,r0 bl system_call_exception
arch/powerpc/kernel/syscall_64.c +13 -15
@@ -19,13 +19,19 @@ extern void __noreturn tabort_syscall(unsigned long nip, unsigned long msr);typedef long (*syscall_fn )(long ,long ,long ,long ,long ,long ); -/* Has to run notrace because it is entered "unreconciled" */ -notrace long system_call_exception (long r3,long r4,long r5,long r6,long r7,long r8, -unsigned long r0,struct pt_regs *regs) +/* Has to run notrace because it is entered not completely "reconciled" */ +notrace long system_call_exception (long r3,long r4,long r5, +long r6,long r7,long r8, +unsigned long r0,struct pt_regs *regs) {unsigned long ti_flags;syscall_fn f; +if (IS_ENABLED (CONFIG_PPC_IRQ_SOFT_MASK_DEBUG )) +BUG_ON (irq_soft_mask_return () !=IRQS_ALL_DISABLED ); + +trace_hardirqs_off ();/* finish reconciling */ +if (IS_ENABLED (CONFIG_PPC_BOOK3S ))BUG_ON (!(regs->msr &MSR_RI ));BUG_ON (!(regs->msr &MSR_PR ));
@@ -33,8 +39,10 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7,BUG_ON (regs->softe !=IRQS_ENABLED );if (IS_ENABLED (CONFIG_PPC_TRANSACTIONAL_MEM ) && -unlikely (regs->msr &MSR_TS_T )) +unlikely (regs->msr &MSR_TS_T )) { +local_irq_enable ();tabort_syscall (regs->nip, regs->msr); + }account_cpu_user_entry ();
@@ -50,16 +58,6 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7,kuap_check_amr (); -/* -* A syscall should always be called with interrupts enabled -* so we just unconditionally hard-enable here. When some kind -* of irq tracing is used, we additionally check that condition -* is correct -*/ -if (IS_ENABLED (CONFIG_PPC_IRQ_SOFT_MASK_DEBUG )) { -WARN_ON (irq_soft_mask_return () !=IRQS_ENABLED ); -WARN_ON (local_paca->irq_happened); - }/* * This is not required for the syscall exit path, but makes the * stack frame look nicer. If this was initialised in the first stack
@@ -68,7 +66,7 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7,*/ regs->softe =IRQS_ENABLED ; -__hard_irq_enable (); +local_irq_enable (); ti_flags =current_thread_info ()->flags;if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) {
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 31/32 powerpc/64s/exception: treat NIA below __end_interrupts as soft-masked
In reply to Nicholas PigginThe scv instruction causes an interrupt which can enter the kernel with MSR[EE]=1, thus allowing interrupts to hit at any time. These must not be taken as normal interrupts, because they come from MSR[PR]=0 context, and yet the kernel stack is not yet set up and r13 is not set to the PACA).
Treat this as a soft-masked interrupt regardless of the soft masked state. This does not affect behaviour yet, because currently all interrupts are taken with MSR[EE]=0.
--- arch/powerpc/kernel/exceptions-64s.S | 27 ++++++++++++++++++++++++--- 1 file changed, 24 insertions(+), 3 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +24 -3
@@ -494,8 +494,24 @@ DEFINE_FIXED_SYMBOL(\name\()_common_virt) .macro __GEN_COMMON_BODY name .if IMASK + .if ! ISTACK + .error "No support for masked interrupt to use custom stack" + .endif + + /* If coming from user, skip soft-mask tests. */ + andi. r10,r12,MSR_PR + bne 2f + + /* Kernel code running below __end_interrupts is implicitly + * soft-masked */ + LOAD_HANDLER(r10, __end_interrupts) + cmpd r11,r10 + li r10,IMASK + blt- 1f + + /* Test the soft mask state against our interrupt's bit */ lbz r10,PACAIRQSOFTMASK(r13) - andi. r10,r10,IMASK +1: andi. r10,r10,IMASK /* Associate vector numbers with bits in paca->irq_happened */ .if IVEC == 0x500 || IVEC == 0xea0 li r10,PACA_IRQ_EE
@@ -526,7 +542,7 @@ DEFINE_FIXED_SYMBOL(\name\()_common_virt) .if ISTACK andi. r10,r12,MSR_PR /* See if coming from user */ - mr r10,r1 /* Save r1 */ +2: mr r10,r1 /* Save r1 */ subi r1,r1,INT_FRAME_SIZE /* alloc frame on kernel stack */ beq- 100f ld r1,PACAKSAVE(r13) /* kernel stack to use */
@@ -2791,7 +2807,8 @@ masked_interrupt: ld r10,PACA_EXGEN+EX_R10(r13) ld r11,PACA_EXGEN+EX_R11(r13) ld r12,PACA_EXGEN+EX_R12(r13) - /* returns to kernel where r13 must be set up, so don't restore it */ + ld r13,PACA_EXGEN+EX_R13(r13) + /* May return to masked low address where r13 is not set up */ .if \hsrr HRFI_TO_KERNEL .else
@@ -2950,6 +2967,10 @@ EXC_COMMON_BEGIN(ppc64_runlatch_on_trampoline) USE_FIXED_SECTION(virt_trampolines) /* + * All code below __end_interrupts is treated as soft-masked. If + * any code runs here with MSR[EE]=1, it must then cope with pending + * soft interrupt being raised (i.e., by ensuring it is replayed). + * * The __end_interrupts marker must be past the out-of-line (OOL) * handlers, so that they are copied to real address 0x100 when running * a relocatable kernel. This ensures they can be reached from the short
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerPATCH v3 32/32 powerpc/64s: system call support for scv/rfscv instructions
In reply to Nicholas PigginAdd support for the scv instruction on POWER9 and later CPUs.
For now this implements the zeroth scv vector 'scv 0', as identical to 'sc' system calls, with the exception that lr is not preserved, and it is 64-bit only. There may yet be changes made to this ABI, so it's for testing only.
rfscv is implemented to return from scv type system calls. It can not be used to return from sc system calls because those are defined to preserve lr.
In a comparison of getpid syscall, the test program had scv taking about 3 more cycles in user mode (92 vs 89 for sc), due to lr handling. getpid syscall throughput on POWER9 is improved by 33%, mostly due to reducing mtmsr and mtspr.
--- Documentation/powerpc/syscall64-abi.rst | 42 +++++--- arch/powerpc/include/asm/asm-prototypes.h | 2 +- arch/powerpc/include/asm/exception-64s.h | 6 ++ arch/powerpc/include/asm/head-64.h | 2 +- arch/powerpc/include/asm/ppc_asm.h | 2 + arch/powerpc/include/asm/processor.h | 2 +- arch/powerpc/include/asm/setup.h | 4 +- arch/powerpc/kernel/cpu_setup_power.S | 2 +- arch/powerpc/kernel/cputable.c | 3 +- arch/powerpc/kernel/dt_cpu_ftrs.c | 1 + arch/powerpc/kernel/entry_64.S | 114 +++++++++++++++++++++ arch/powerpc/kernel/exceptions-64s.S | 119 +++++++++++++++++++++- arch/powerpc/kernel/setup_64.c | 5 +- arch/powerpc/kernel/syscall_64.c | 14 ++- arch/powerpc/platforms/pseries/setup.c | 8 +- 15 files changed, 295 insertions(+), 31 deletions(-)
Documentation/powerpc/syscall64-abi.rst +30 -12
@@ -5,6 +5,15 @@ Power Architecture 64-bit Linux system call ABI syscall ======= +Invocation +---------- +The syscall is made with the sc instruction, and returns with execution +continuing at the instruction following the sc instruction. + +If PPC_FEATURE2_SCV appears in the AT_HWCAP2 ELF auxiliary vector, the +scv 0 instruction is an alternative that may be used, with some differences +to calling sequence. + syscall calling sequence\ [1]_ matches the Power Architecture 64-bit ELF ABI specification C function calling sequence, including register preservation rules, with the following differences.
@@ -12,16 +21,23 @@ rules, with the following differences. .. [1] Some syscalls (typically low-level management functions) may have different calling sequences (e.g., rt_sigreturn). -Parameters and return value ---------------------------- +Parameters +---------- The system call number is specified in r0. There is a maximum of 6 integer parameters to a syscall, passed in r3-r8. -Both a return value and a return error code are returned. cr0.SO is the return -error code, and r3 is the return value or error code. When cr0.SO is clear, -the syscall succeeded and r3 is the return value. When cr0.SO is set, the -syscall failed and r3 is the error code that generally corresponds to errno. +Return value +------------ +- For the sc instruction, both a return value and a return error code are + returned. cr0.SO is the return error code, and r3 is the return value or + error code. When cr0.SO is clear, the syscall succeeded and r3 is the return + value. When cr0.SO is set, the syscall failed and r3 is the error code that + generally corresponds to errno. + +- For the scv 0 instruction, there is a return value indicates failure if it + is >= -MAX_ERRNO (-4095) as an unsigned comparison, in which case it is the + negated return error code. Otherwise it is the successful return value. Stack -----
@@ -34,22 +50,23 @@ Register preservation rules match the ELF ABI calling sequence with the following differences: =========== ============= ======================================== +--- For the sc instruction --- r0 Volatile (System call number.) r3 Volatile (Parameter 1, and return value.) r4-r8 Volatile (Parameters 2-6.) -cr0 Volatile (cr0.SO is the return error condition) +cr0 Volatile (cr0.SO is the return error condition.) cr1, cr5-7 Nonvolatile lr Nonvolatile + +--- For the scv 0 instruction --- +r0 Volatile (System call number.) +r3 Volatile (Parameter 1, and return value.) +r4-r8 Volatile (Parameters 2-6.) =========== ============= ======================================== All floating point and vector data registers as well as control and status registers are nonvolatile. -Invocation ----------- -The syscall is performed with the sc instruction, and returns with execution -continuing at the instruction following the sc instruction. - Transactional Memory -------------------- Syscall behavior can change if the processor is in transactional or suspended
@@ -75,6 +92,7 @@ auxiliary vector. returning to the caller. This case is not well defined or supported, so this behavior should not be relied upon. +scv 0 syscalls will always behave as PPC_FEATURE2_HTM_NOSC. vsyscall ========
arch/powerpc/include/asm/asm-prototypes.h +1 -1
@@ -99,7 +99,7 @@ void __init machine_init(u64 dt_ptr);#endif #ifdef CONFIG_PPC64 long system_call_exception (long r3,long r4,long r5,long r6,long r7,long r8,unsigned long r0,struct pt_regs *regs); -notrace unsigned long syscall_exit_prepare (unsigned long r3,struct pt_regs *regs); +notrace unsigned long syscall_exit_prepare (unsigned long r3,struct pt_regs *regs,long scv);notrace unsigned long interrupt_exit_user_prepare (struct pt_regs *regs,unsigned long msr);notrace unsigned long interrupt_exit_kernel_prepare (struct pt_regs *regs,unsigned long msr);#endif
arch/powerpc/include/asm/exception-64s.h +6 -0
@@ -123,6 +123,12 @@hrfid ; \ b hrfi_flush_fallback +#define RFSCV_TO_USER \ + STF_EXIT_BARRIER_SLOT; \ + RFI_FLUSH_SLOT; \ + RFSCV; \ + b rfscv_flush_fallback +#endif /* __ASSEMBLY__ */ #endif /* _ASM_POWERPC_EXCEPTION_H */
arch/powerpc/include/asm/head-64.h +1 -1
@@ -128,7 +128,7 @@ end_##sname: .if ((start) % (size) != 0); \ .error "Fixed section exception vector misalignment"; \ .endif; \ - .if ((size) != 0x20) && ((size) != 0x80) && ((size) != 0x100); \ + .if ((size) != 0x20) && ((size) != 0x80) && ((size) != 0x100) && ((size) != 0x1000); \ .error "Fixed section exception vector bad size"; \ .endif; \ .if (start) < sname##_start; \
arch/powerpc/include/asm/ppc_asm.h +2 -0
@@ -755,6 +755,8 @@ END_FTR_SECTION_NESTED(CPU_FTR_CELL_TB_BUG, CPU_FTR_CELL_TB_BUG, 96)#define N_SLINE 68#define N_SO 100 +#define RFSCV .long 0x4c0000a4 +/* * Create an endian fixup trampoline *
arch/powerpc/include/asm/processor.h +1 -1
@@ -304,7 +304,7 @@ struct thread_struct { .regs = (struct pt_regs *)INIT_SP - 1, /* XXX bogus, I think */ \ .addr_limit = KERNEL_DS, \ .fpexc_mode = 0, \ - .fscr = FSCR_TAR | FSCR_EBB \ + .fscr = FSCR_TAR | FSCR_EBB | FSCR_SCV \ }#endif
arch/powerpc/include/asm/setup.h +2 -2
@@ -30,12 +30,12 @@ void setup_panic(void);#define ARCH_PANIC_TIMEOUT 180#ifdef CONFIG_PPC_PSERIES -extern void pseries_enable_reloc_on_exc (void ); +extern bool pseries_enable_reloc_on_exc (void );extern void pseries_disable_reloc_on_exc (void );extern void pseries_big_endian_exceptions (void );extern void pseries_little_endian_exceptions (void );#else -static inline void pseries_enable_reloc_on_exc (void ) {} +static inline bool pseries_enable_reloc_on_exc (void ) {return false; }static inline void pseries_disable_reloc_on_exc (void ) {}static inline void pseries_big_endian_exceptions (void ) {}static inline void pseries_little_endian_exceptions (void ) {}
arch/powerpc/kernel/cpu_setup_power.S +1 -1
@@ -184,7 +184,7 @@ __init_LPCR_ISA300: __init_FSCR: mfspr r3,SPRN_FSCR - ori r3,r3,FSCR_TAR|FSCR_DSCR|FSCR_EBB + ori r3,r3,FSCR_SCV|FSCR_TAR|FSCR_DSCR|FSCR_EBB mtspr SPRN_FSCR,r3 blr
arch/powerpc/kernel/cputable.c +2 -1
@@ -118,7 +118,8 @@ extern void __restore_cpu_e6500(void);#define COMMON_USER2_POWER9 (COMMON_USER2_POWER8 | \ PPC_FEATURE2_ARCH_3_00 | \ PPC_FEATURE2_HAS_IEEE128 | \ - PPC_FEATURE2_DARN ) + PPC_FEATURE2_DARN | \ + PPC_FEATURE2_SCV)#ifdef CONFIG_PPC_BOOK3E_64 #define COMMON_USER_BOOKE (COMMON_USER_PPC64 | PPC_FEATURE_BOOKE)
arch/powerpc/kernel/dt_cpu_ftrs.c +1 -0
@@ -566,6 +566,7 @@ static struct dt_cpu_feature_match __initdata {"little-endian", feat_enable_le, CPU_FTR_REAL_LE}, {"smt", feat_enable_smt, 0}, {"interrupt-facilities", feat_enable, 0}, + {"system-call-vectored", feat_enable, 0}, {"timer-facilities", feat_enable, 0}, {"timer-facilities-v3", feat_enable, 0}, {"debug-facilities", feat_enable, 0},
arch/powerpc/kernel/entry_64.S +114 -0
126 lines not on this page · raw message
arch/powerpc/kernel/exceptions-64s.S +115 -4
149 lines not on this page · raw message
arch/powerpc/kernel/setup_64.c +4 -1
@@ -196,7 +196,10 @@ static void __init configure_exceptions(void)/* Under a PAPR hypervisor, we need hypercalls */ if (firmware_has_feature (FW_FEATURE_SET_MODE )) {/* Enable AIL if possible */ -pseries_enable_reloc_on_exc (); +if (!pseries_enable_reloc_on_exc ()) { + init_task.thread.fscr &= ~FSCR_SCV ; + cur_cpu_spec->cpu_user_features2 &= ~PPC_FEATURE2_SCV ; + }/* * Tell the hypervisor that we want our exceptions to
arch/powerpc/kernel/syscall_64.c +10 -4
@@ -121,7 +121,8 @@ notrace long system_call_exception(long r3, long r4, long r5,* because RI=0 and soft mask state is "unreconciled", so it is marked notrace. */ notrace unsigned long syscall_exit_prepare (unsigned long r3, -struct pt_regs *regs) +struct pt_regs *regs, +long scv) {unsigned long *ti_flagsp = ¤t_thread_info ()->flags;unsigned long ti_flags;
@@ -134,7 +135,7 @@ notrace unsigned long syscall_exit_prepare(unsigned long r3, ti_flags = *ti_flagsp; -if (unlikely (r3 >= (unsigned long )-MAX_ERRNO )) { +if (unlikely (r3 >= (unsigned long )-MAX_ERRNO ) && !scv) {if (likely (!(ti_flags & (_TIF_NOERROR | _TIF_RESTOREALL)))) { r3 = -r3; regs->ccr |=0x10000000 ;/* Set SO bit in CR */
@@ -191,9 +192,14 @@ notrace unsigned long syscall_exit_prepare(unsigned long r3,trace_hardirqs_on ();/* This pattern matches prep_irq_for_idle */ -__hard_EE_RI_disable (); +/* scv need not set RI=0 because SRRs are not used */ +if (scv) +__hard_irq_disable (); +else +__hard_EE_RI_disable ();if (unlikely (lazy_irq_pending ())) { -__hard_RI_enable (); +if (!scv) +__hard_RI_enable ();trace_hardirqs_off (); local_paca->irq_happened |=PACA_IRQ_HARD_DIS ;local_irq_enable ();
arch/powerpc/platforms/pseries/setup.c +5 -3
@@ -349,7 +349,7 @@ static void pseries_lpar_idle(void)* to ever be a problem in practice we can move this into a kernel thread to * finish off the process later in boot. */ -void pseries_enable_reloc_on_exc (void ) +bool pseries_enable_reloc_on_exc (void ) {long rc;unsigned int delay, total_delay =0 ;
@@ -360,11 +360,13 @@ void pseries_enable_reloc_on_exc(void)if (rc ==H_P2 ) {pr_info ("Relocation on exceptions not" " supported\n" ); +return false; }else if (rc !=H_SUCCESS ) {pr_warn ("Unable to enable relocation" " on exceptions: %ld\n" , rc); +return false; } -break ; +return true; } delay =get_longbusy_msecs (rc);
@@ -373,7 +375,7 @@ void pseries_enable_reloc_on_exc(void)pr_warn ("Warning: Giving up waiting to enable " "relocation on exceptions (%u msec)!\n" , total_delay); -return ; +return false; }mdelay (delay);
Signature · 1 line
2.23.0
kbuild test robot
build botRe: [PATCH v3 32/32] powerpc/64s: system call support for scv/rfscv instructions
I love your patch! Yet something to improve:
Hi Nicholas,
I love your patch! Yet something to improve:
[auto build test ERROR on powerpc/next]
[also build test ERROR on v5.6-rc3 next-20200228]
[cannot apply to kvm-ppc/kvm-ppc-next scottwood/next]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system. BTW, we also suggest to use '--base' option to specify the base tree in git format-patch, please see https://stackoverflow.com/a/37406982]
url: https://github.com/0day-ci/linux/commits/Nicholas-Piggin/powerpc-64-interrupts-and-syscalls-series/20200226-043224 base: https://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux.git next config: powerpc-ppc64e_defconfig (attached as .config) compiler: powerpc64-linux-gcc (GCC) 7.5.0
reproduce:
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
GCC_VERSION=7.5.0 make.cross ARCH=powerpc
If you fix the issue, kindly add following tag
All errors (new ones prefixed by >>):
arch/powerpc/kernel/entry_64.S: Assembler messages:
>> arch/powerpc/kernel/entry_64.S:67: Error: unrecognized opcode: `interrupt_to_kernel' >> arch/powerpc/kernel/entry_64.S:164: Error: unrecognized opcode: `rfscv_to_user'
vim +67 arch/powerpc/kernel/entry_64.S
47
48 /*
49 * System calls.
50 */
51 .section ".toc","aw"
52 SYS_CALL_TABLE:
53 .tc sys_call_table[TC],sys_call_table
54
55 COMPAT_SYS_CALL_TABLE:
56 .tc compat_sys_call_table[TC],compat_sys_call_table
57
58 /* This value is used to mark exception frames on the stack. */
59 exception_marker:
60 .tc ID_EXC_MARKER[TC],STACK_FRAME_REGS_MARKER
61
62 .section ".text"
63 .align 7
64
65 .globl system_call_vectored_common
66 system_call_vectored_common:
> 67 INTERRUPT_TO_KERNEL
68 mr r10,r1
69 ld r1,PACAKSAVE(r13)
70 std r10,0(r1)
71 std r11,_NIP(r1)
72 std r12,_MSR(r1)
73 std r0,GPR0(r1)
74 std r10,GPR1(r1)
75 std r2,GPR2(r1)
76 ld r2,PACATOC(r13)
77 mfcr r12
78 li r11,0
79 /* Can we avoid saving r3-r8 in common case? */
80 std r3,GPR3(r1)
81 std r4,GPR4(r1)
82 std r5,GPR5(r1)
83 std r6,GPR6(r1)
84 std r7,GPR7(r1)
85 std r8,GPR8(r1)
86 /* Zero r9-r12, this should only be required when restoring all GPRs */
87 std r11,GPR9(r1)
88 std r11,GPR10(r1)
89 std r11,GPR11(r1)
90 std r11,GPR12(r1)
91 std r9,GPR13(r1)
92 SAVE_NVGPRS(r1)
93 std r11,_XER(r1)
94 std r11,_LINK(r1)
95 std r11,_CTR(r1)
96
97 li r11,0xc00
98 std r11,_TRAP(r1)
99 std r12,_CCR(r1)
100 std r3,ORIG_GPR3(r1)
101 addi r10,r1,STACK_FRAME_OVERHEAD
102 ld r11,exception_marker(a)toc(r2)
103 std r11,-16(r10) /* "regshere" marker */
104
105 /*
106 * RECONCILE_IRQ_STATE without calling trace_hardirqs_off(), which
107 * would clobber syscall parameters. Also we always enter with IRQs
108 * enabled and nothing pending. system_call_exception() will call
109 * trace_hardirqs_off().
110 *
111 * scv enters with MSR[EE]=1, so don't set PACA_IRQ_HARD_DIS.
112 */
113 li r9,IRQS_ALL_DISABLED
114 stb r9,PACAIRQSOFTMASK(r13)
115
116 /* Calling convention has r9 = orig r0, r10 = regs */
117 mr r9,r0
118 bl system_call_exception
119
120 .Lsyscall_vectored_exit:
121 addi r4,r1,STACK_FRAME_OVERHEAD
122 li r5,1 /* scv */
123 bl syscall_exit_prepare
124
125 ld r2,_CCR(r1)
126 ld r4,_NIP(r1)
127 ld r5,_MSR(r1)
128
129 BEGIN_FTR_SECTION
130 stdcx. r0,0,r1 /* to clear the reservation */
131 END_FTR_SECTION_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
132
133 mtlr r4
134 mtctr r5
135
136 cmpdi r3,0
137 bne syscall_vectored_restore_regs
138 li r0,0
139 li r4,0
140 li r5,0
141 li r6,0
142 li r7,0
143 li r8,0
144 li r9,0
145 li r10,0
146 li r11,0
147 li r12,0
148 mtspr SPRN_XER,r0
149 .Lsyscall_vectored_restore_regs_cont:
150
151 BEGIN_FTR_SECTION
152 HMT_MEDIUM_LOW
153 END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
154
155 /*
156 * We don't need to restore AMR on the way back to userspace for KUAP.
157 * The value of AMR only matters while we're in the kernel.
158 */
159 mtcr r2
160 ld r2,GPR2(r1)
161 ld r3,GPR3(r1)
162 ld r13,GPR13(r1)
163 ld r1,GPR1(r1)
> 164 RFSCV_TO_USER 165 b . /* prevent speculative execution */
166 _ASM_NOKPROBE_SYMBOL(system_call_vectored_common);
167
168 syscall_vectored_restore_regs:
169 ld r4,_XER(r1)
170 REST_NVGPRS(r1)
171 mtspr SPRN_XER,r4
172 ld r0,GPR0(r1)
173 REST_8GPRS(4, r1)
174 ld r12,GPR12(r1)
175 b .Lsyscall_vectored_restore_regs_cont
176
177 .balign IFETCH_ALIGN_BYTES
178 .globl system_call_common
179 system_call_common:
180 _ASM_NOKPROBE_SYMBOL(system_call_common)
181 mr r10,r1
182 ld r1,PACAKSAVE(r13)
183 std r10,0(r1)
184 std r11,_NIP(r1)
185 std r12,_MSR(r1)
186 std r0,GPR0(r1)
187 std r10,GPR1(r1)
188 std r2,GPR2(r1)
189 #ifdef CONFIG_PPC_FSL_BOOK3E
190 START_BTB_FLUSH_SECTION
191 BTB_FLUSH(r10)
192 END_BTB_FLUSH_SECTION
193 #endif
194 ld r2,PACATOC(r13)
195 mfcr r12
196 li r11,0
197 /* Can we avoid saving r3-r8 in common case? */
198 std r3,GPR3(r1)
199 std r4,GPR4(r1)
200 std r5,GPR5(r1)
201 std r6,GPR6(r1)
202 std r7,GPR7(r1)
203 std r8,GPR8(r1)
204 /* Zero r9-r12, this should only be required when restoring all GPRs */
205 std r11,GPR9(r1)
206 std r11,GPR10(r1)
207 std r11,GPR11(r1)
208 std r11,GPR12(r1)
209 std r9,GPR13(r1)
210 SAVE_NVGPRS(r1)
211 std r11,_XER(r1)
212 std r11,_CTR(r1)
213 mflr r10
214
215 /*
216 * This clears CR0.SO (bit 28), which is the error indication on
217 * return from this system call.
218 */
219 rldimi r12,r11,28,(63-28)
220 li r11,0xc00
221 std r10,_LINK(r1)
222 std r11,_TRAP(r1)
223 std r12,_CCR(r1)
224 std r3,ORIG_GPR3(r1)
225 addi r10,r1,STACK_FRAME_OVERHEAD
226 ld r11,exception_marker(a)toc(r2)
227 std r11,-16(r10) /* "regshere" marker */
228
229 /*
230 * RECONCILE_IRQ_STATE without calling trace_hardirqs_off(), which
231 * would clobber syscall parameters. Also we always enter with IRQs
232 * enabled and nothing pending. system_call_exception() will call
233 * trace_hardirqs_off().
234 */
235 li r11,IRQS_ALL_DISABLED
236 li r12,PACA_IRQ_HARD_DIS
237 stb r11,PACAIRQSOFTMASK(r13)
238 stb r12,PACAIRQHAPPENED(r13)
239
240 /* Calling convention has r9 = orig r0, r10 = regs */
241 mr r9,r0
242 bl system_call_exception
243
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
v11 cover Disable compat cruft on ppc64le v11
In reply to Nicholas PigginLess code means less bugs so add a knob to skip the compat stuff.
Changes in v2: saner CONFIG_COMPAT ifdefs
Changes in v3:
- change llseek to 32bit instead of builing it unconditionally in fs
- clanup the makefile conditionals
- remove some ifdefs or convert to IS_DEFINED where possible
Changes in v4:
- cleanup is_32bit_task and current_is_64bit
- more makefile cleanup
Changes in v5:
- more current_is_64bit cleanup
- split off callchain.c 32bit and 64bit parts
Changes in v6:
- cleanup makefile after split
- consolidate read_user_stack_32
- fix some checkpatch warnings
Changes in v7:
- add back __ARCH_WANT_SYS_LLSEEK to fix build with llseek
- remove leftover hunk
- add review tags
Changes in v8:
- consolidate valid_user_sp to fix it in the split callchain.c
- fix build errors/warnings with PPC64 !COMPAT and PPC32
Changes in v9:
- remove current_is_64bit()
Chanegs in v10:
- rebase, sent together with the syscall cleanup
Changes in v11:
- rebase
- add MAINTAINERS pattern for ppc perf
Michal Suchanek (8):
powerpc: Add back __ARCH_WANT_SYS_LLSEEK macro
powerpc: move common register copy functions from signal_32.c to
signal.c
powerpc/perf: consolidate read_user_stack_32 powerpc/perf: consolidate valid_user_sp
powerpc/64: make buildable without CONFIG_COMPAT powerpc/64: Make COMPAT user-selectable disabled on littleendian by
default.
powerpc/perf: split callchain.c by bitness
MAINTAINERS: perf: Add pattern that matches ppc perf to the perf
entry.
MAINTAINERS | 2 + arch/powerpc/Kconfig | 5 +- arch/powerpc/include/asm/thread_info.h | 4 +- arch/powerpc/include/asm/unistd.h | 1 + arch/powerpc/kernel/Makefile | 6 +- arch/powerpc/kernel/entry_64.S | 2 + arch/powerpc/kernel/signal.c | 144 +++++++++- arch/powerpc/kernel/signal_32.c | 140 ---------- arch/powerpc/kernel/syscall_64.c | 6 +- arch/powerpc/kernel/vdso.c | 3 +- arch/powerpc/perf/Makefile | 5 +- arch/powerpc/perf/callchain.c | 356 +------------------------ arch/powerpc/perf/callchain.h | 20 ++ arch/powerpc/perf/callchain_32.c | 196 ++++++++++++++ arch/powerpc/perf/callchain_64.c | 174 ++++++++++++ fs/read_write.c | 3 +- 16 files changed, 556 insertions(+), 511 deletions(-) create mode 100644 arch/powerpc/perf/callchain.h create mode 100644 arch/powerpc/perf/callchain_32.c create mode 100644 arch/powerpc/perf/callchain_64.c
Signature · 1 line
2.23.0
PATCH v11 1/8 powerpc: Add back __ARCH_WANT_SYS_LLSEEK macro
This partially reverts commit caf6f9c8a326 ("asm-generic: Remove unneeded __ARCH_WANT_SYS_LLSEEK macro")
When CONFIG_COMPAT is disabled on ppc64 the kernel does not build.
There is resistance to both removing the llseek syscall from the 64bit syscall tables and building the llseek interface unconditionally.
---
v7: new patch
--- arch/powerpc/include/asm/unistd.h | 1 + fs/read_write.c | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-)
arch/powerpc/include/asm/unistd.h +1 -0
@@ -31,6 +31,7 @@#define __ARCH_WANT_SYS_SOCKETCALL#define __ARCH_WANT_SYS_FADVISE64#define __ARCH_WANT_SYS_GETPGRP +#define __ARCH_WANT_SYS_LLSEEK#define __ARCH_WANT_SYS_NICE#define __ARCH_WANT_SYS_OLD_GETRLIMIT#define __ARCH_WANT_SYS_OLD_UNAME
fs/read_write.c +2 -1
@@ -331,7 +331,8 @@ COMPAT_SYSCALL_DEFINE3(lseek, unsigned int, fd, compat_off_t, offset, unsigned i }#endif -#if !defined(CONFIG_64BIT ) || defined(CONFIG_COMPAT ) +#if !defined(CONFIG_64BIT ) || defined(CONFIG_COMPAT ) || \ + defined(__ARCH_WANT_SYS_LLSEEK) SYSCALL_DEFINE5(llseek, unsigned int, fd, unsigned long, offset_high, unsigned long, offset_low, loff_t __user *, result, unsigned int, whence)
Signature · 1 line
2.23.0
PATCH v11 2/8 powerpc: move common register copy functions from signal_32.c to signal.c
In reply to Michal SuchanekThese functions are required for 64bit as well.
--- arch/powerpc/kernel/signal.c | 141 ++++++++++++++++++++++++++++++++ arch/powerpc/kernel/signal_32.c | 140 ------------------------------- 2 files changed, 141 insertions(+), 140 deletions(-)
arch/powerpc/kernel/signal.c +141 -0
@@ -18,12 +18,153 @@#include <linux/syscalls.h> #include <asm/hw_breakpoint.h> #include <linux/uaccess.h> +#include <asm/switch_to.h> #include <asm/unistd.h> #include <asm/debug.h> #include <asm/tm.h> #include "signal.h" +#ifdef CONFIG_VSX +unsigned long copy_fpr_to_user (void __user *to, +struct task_struct *task) +{ +u64 buf[ELF_NFPREG ]; +int i; + +/* save FPR copy to local buffer then write to the thread_struct */ +for (i =0 ; i < (ELF_NFPREG -1 ) ; i++) + buf[i] = task->thread.TS_FPR (i); + buf[i] = task->thread.fp_state.fpscr; +return __copy_to_user (to, buf,ELF_NFPREG *sizeof (double )); +} + +unsigned long copy_fpr_from_user (struct task_struct *task, +void __user *from) +{ +u64 buf[ELF_NFPREG ]; +int i; + +if (__copy_from_user (buf, from,ELF_NFPREG *sizeof (double ))) +return 1 ; +for (i =0 ; i < (ELF_NFPREG -1 ) ; i++) + task->thread.TS_FPR (i) = buf[i]; + task->thread.fp_state.fpscr = buf[i]; + +return 0 ; +} + +unsigned long copy_vsx_to_user (void __user *to, +struct task_struct *task) +{ +u64 buf[ELF_NVSRHALFREG ]; +int i; + +/* save FPR copy to local buffer then write to the thread_struct */ +for (i =0 ; i <ELF_NVSRHALFREG ; i++) + buf[i] = task->thread.fp_state.fpr[i][TS_VSRLOWOFFSET ]; +return __copy_to_user (to, buf,ELF_NVSRHALFREG *sizeof (double )); +} + +unsigned long copy_vsx_from_user (struct task_struct *task, +void __user *from) +{ +u64 buf[ELF_NVSRHALFREG ]; +int i; + +if (__copy_from_user (buf, from,ELF_NVSRHALFREG *sizeof (double ))) +return 1 ; +for (i =0 ; i <ELF_NVSRHALFREG ; i++) + task->thread.fp_state.fpr[i][TS_VSRLOWOFFSET ] = buf[i]; +return 0 ; +} + +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM +unsigned long copy_ckfpr_to_user (void __user *to, +struct task_struct *task) +{ +u64 buf[ELF_NFPREG ]; +int i; + +/* save FPR copy to local buffer then write to the thread_struct */ +for (i =0 ; i < (ELF_NFPREG -1 ) ; i++) + buf[i] = task->thread.TS_CKFPR (i); + buf[i] = task->thread.ckfp_state.fpscr; +return __copy_to_user (to, buf,ELF_NFPREG *sizeof (double )); +} + +unsigned long copy_ckfpr_from_user (struct task_struct *task, +void __user *from) +{ +u64 buf[ELF_NFPREG ]; +int i; + +if (__copy_from_user (buf, from,ELF_NFPREG *sizeof (double ))) +return 1 ; +for (i =0 ; i < (ELF_NFPREG -1 ) ; i++) + task->thread.TS_CKFPR (i) = buf[i]; + task->thread.ckfp_state.fpscr = buf[i]; + +return 0 ; +} + +unsigned long copy_ckvsx_to_user (void __user *to, +struct task_struct *task) +{ +u64 buf[ELF_NVSRHALFREG ]; +int i; + +/* save FPR copy to local buffer then write to the thread_struct */ +for (i =0 ; i <ELF_NVSRHALFREG ; i++) + buf[i] = task->thread.ckfp_state.fpr[i][TS_VSRLOWOFFSET ]; +return __copy_to_user (to, buf,ELF_NVSRHALFREG *sizeof (double )); +} + +unsigned long copy_ckvsx_from_user (struct task_struct *task, +void __user *from) +{ +u64 buf[ELF_NVSRHALFREG ]; +int i; + +if (__copy_from_user (buf, from,ELF_NVSRHALFREG *sizeof (double ))) +return 1 ; +for (i =0 ; i <ELF_NVSRHALFREG ; i++) + task->thread.ckfp_state.fpr[i][TS_VSRLOWOFFSET ] = buf[i]; +return 0 ; +} +#endif /* CONFIG_PPC_TRANSACTIONAL_MEM */ +#else +inline unsigned long copy_fpr_to_user (void __user *to, +struct task_struct *task) +{ +return __copy_to_user (to, task->thread.fp_state.fpr, +ELF_NFPREG *sizeof (double )); +} + +inline unsigned long copy_fpr_from_user (struct task_struct *task, +void __user *from) +{ +return __copy_from_user (task->thread.fp_state.fpr, from, +ELF_NFPREG *sizeof (double )); +} + +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM +inline unsigned long copy_ckfpr_to_user (void __user *to, +struct task_struct *task) +{ +return __copy_to_user (to, task->thread.ckfp_state.fpr, +ELF_NFPREG *sizeof (double )); +} + +inline unsigned long copy_ckfpr_from_user (struct task_struct *task, +void __user *from) +{ +return __copy_from_user (task->thread.ckfp_state.fpr, from, +ELF_NFPREG *sizeof (double )); +} +#endif /* CONFIG_PPC_TRANSACTIONAL_MEM */ +#endif +/* Log an error when sending an unhandled signal to a process. Controlled * through debug.exception-trace sysctl. */
arch/powerpc/kernel/signal_32.c +0 -140
146 lines not on this page · raw message
Signature · 1 line
2.23.0
PATCH v11 3/8 powerpc/perf: consolidate read_user_stack_32
In reply to Michal SuchanekThere are two almost identical copies for 32bit and 64bit.
The function is used only in 32bit code which will be split out in next patch so consolidate to one function.
---
v6: new patch v8: move the consolidated function out of the ifdef block. v11: rebase on top of def0bfdbd603 --- arch/powerpc/perf/callchain.c | 48 +++++++++++++++++------------------ 1 file changed, 24 insertions(+), 24 deletions(-)
arch/powerpc/perf/callchain.c +24 -24
@@ -161,18 +161,6 @@ static int read_user_stack_64(unsigned long __user *ptr, unsigned long *ret)return read_user_stack_slow (ptr, ret,8 ); } -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) -{ -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,4 ); -} -static inline int valid_user_sp (unsigned long sp,int is_64) {if (!sp || (sp &7 ) || sp > (is_64 ?TASK_SIZE :0x100000000UL ) -32 )
@@ -277,19 +265,9 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry, }#else /* CONFIG_PPC64 */ -/* -* On 32-bit we just access the address and let hash_page create a -* HPTE if necessary, so there is no need to fall back to reading -* the page tables. Since this is called at interrupt level, -* do_page_fault() won't treat a DSI as a page fault. -*/ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_slow (void __user *ptr,void *buf,int nb) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -return probe_user_read (ret, ptr,sizeof (*ret)); +return 0 ; } static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,
@@ -312,6 +290,28 @@ static inline int valid_user_sp(unsigned long sp, int is_64)#endif /* CONFIG_PPC64 */ +/* +* On 32-bit we just access the address and let hash_page create a +* HPTE if necessary, so there is no need to fall back to reading +* the page tables. Since this is called at interrupt level, +* do_page_fault() won't treat a DSI as a page fault. +*/ +static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +{ +int rc; + +if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || + ((unsigned long )ptr &3 )) +return -EFAULT ; + + rc =probe_user_read (ret, ptr,sizeof (*ret)); + +if (IS_ENABLED (CONFIG_PPC64 ) && rc) +return read_user_stack_slow (ptr, ret,4 ); + +return rc; +} +/* * Layout for non-RT signal frames */
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerRe: [PATCH v11 3/8] powerpc/perf: consolidate read_user_stack_32
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 61 lines
There are two almost identical copies for 32bit and 64bit.
The function is used only in 32bit code which will be split out in next patch so consolidate to one function.
Signed-off-by Michal Suchanek <[email protected]>Reviewed-by Christophe Leroy <[email protected]>---
v6: new patch v8: move the consolidated function out of the ifdef block. v11: rebase on top of def0bfdbd603 --- arch/powerpc/perf/callchain.c | 48 +++++++++++++++++------------------ 1 file changed, 24 insertions(+), 24 deletions(-)arch/powerpc/perf/callchain.c +2
-24@@ -161,18 +161,6 @@ static int read_user_stack_64(unsigned long __user *ptr, unsigned long *ret)return read_user_stack_slow (ptr, ret,8 ); } -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) -{ -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,4 ); -} -static inline int valid_user_sp (unsigned long sp,int is_64) {if (!sp || (sp &7 ) || sp > (is_64 ?TASK_SIZE :0x100000000UL ) -32 )@@ -277,19 +265,9 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry, }#else /* CONFIG_PPC64 */ -/* -* On 32-bit we just access the address and let hash_page create a -* HPTE if necessary, so there is no need to fall back to reading -* the page tables. Since this is called at interrupt level, -* do_page_fault() won't treat a DSI as a page fault. -*/ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_slow (void __user *ptr,void *buf,int nb) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -return probe_user_read (ret, ptr,sizeof (*ret)); +return 0 ; } static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,
arch/powerpc/perf/callchain.c static inline int valid_user_sp(unsigned long sp, int is_64) · line 298
Quoted 9 lines
arch/powerpc/perf/callchain.c +6
-0@@ -312,6 +290,28 @@ static inline int valid_user_sp(unsigned long sp, int is_64)#endif /* CONFIG_PPC64 */ +/* +* On 32-bit we just access the address and let hash_page create a +* HPTE if necessary, so there is no need to fall back to reading +* the page tables. Since this is called at interrupt level, +* do_page_fault() won't treat a DSI as a page fault. +*/
The comment is actually probably better to stay in the 32-bit read_user_stack_slow implementation. Is that function defined on 32-bit purely so that you can use IS_ENABLED()? In that case I would prefer to put a BUG() there which makes it self documenting.
Thanks,
Nick
Quoted 23 lines
arch/powerpc/perf/callchain.c +16
-0+static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +{ +int rc; + +if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || + ((unsigned long )ptr &3 )) +return -EFAULT ; + + rc =probe_user_read (ret, ptr,sizeof (*ret)); + +if (IS_ENABLED (CONFIG_PPC64 ) && rc) +return read_user_stack_slow (ptr, ret,4 ); + +return rc; +} +/* * Layout for non-RT signal frames */ Signature · 1 line
2.23.0
Re: [PATCH v11 3/8] powerpc/perf: consolidate read_user_stack_32
Quoted 65 lines · Nicholas Piggin
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 64 lines
There are two almost identical copies for 32bit and 64bit.
The function is used only in 32bit code which will be split out in next patch so consolidate to one function.
Signed-off-by Michal Suchanek <[email protected]>Reviewed-by Christophe Leroy <[email protected]>---
v6: new patch v8: move the consolidated function out of the ifdef block. v11: rebase on top of def0bfdbd603 --- arch/powerpc/perf/callchain.c | 48 +++++++++++++++++------------------ 1 file changed, 24 insertions(+), 24 deletions(-)arch/powerpc/perf/callchain.c +2
-24@@ -161,18 +161,6 @@ static int read_user_stack_64(unsigned long __user *ptr, unsigned long *ret)return read_user_stack_slow (ptr, ret,8 ); } -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) -{ -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,4 ); -} -static inline int valid_user_sp (unsigned long sp,int is_64) {if (!sp || (sp &7 ) || sp > (is_64 ?TASK_SIZE :0x100000000UL ) -32 )@@ -277,19 +265,9 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry, }#else /* CONFIG_PPC64 */ -/* -* On 32-bit we just access the address and let hash_page create a -* HPTE if necessary, so there is no need to fall back to reading -* the page tables. Since this is called at interrupt level, -* do_page_fault() won't treat a DSI as a page fault. -*/ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_slow (void __user *ptr,void *buf,int nb) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -return probe_user_read (ret, ptr,sizeof (*ret)); +return 0 ; } static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,@@ -312,6 +290,28 @@ static inline int valid_user_sp(unsigned long sp, int is_64)#endif /* CONFIG_PPC64 */
Quoted 10 lines
arch/powerpc/perf/callchain.c static inline int valid_user_sp(unsigned long sp, int is_64) · line 298
Quoted 6 lines
arch/powerpc/perf/callchain.c +6
-0+/* +* On 32-bit we just access the address and let hash_page create a +* HPTE if necessary, so there is no need to fall back to reading +* the page tables. Since this is called at interrupt level, +* do_page_fault() won't treat a DSI as a page fault. +*/ The comment is actually probably better to stay in the 32-bit
read_user_stack_slowimplementation. Is that function defined on 32-bit purely so that you can useIS_ENABLED()? In that case
It documents the IS_ENABLED() and that's where it is. The 32bit definition is only a technical detail.
Quoted 1 line
I would prefer to put a
BUG()there which makes it self documenting.
Which will cause checkpatch complaints about introducing new BUG() which is frowned on.
Thanks
Michal
Nicholas Piggin
reviewerRe: [PATCH v11 3/8] powerpc/perf: consolidate read_user_stack_32
Michal Suchánek's on March 25, 2020 5:38 am:
Quoted 76 lines · Nicholas Piggin
Quoted 65 lines
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 64 lines
There are two almost identical copies for 32bit and 64bit.
The function is used only in 32bit code which will be split out in next patch so consolidate to one function.
Signed-off-by Michal Suchanek <[email protected]>Reviewed-by Christophe Leroy <[email protected]>---
v6: new patch v8: move the consolidated function out of the ifdef block. v11: rebase on top of def0bfdbd603 --- arch/powerpc/perf/callchain.c | 48 +++++++++++++++++------------------ 1 file changed, 24 insertions(+), 24 deletions(-)arch/powerpc/perf/callchain.c +2
-24@@ -161,18 +161,6 @@ static int read_user_stack_64(unsigned long __user *ptr, unsigned long *ret)return read_user_stack_slow (ptr, ret,8 ); } -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) -{ -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,4 ); -} -static inline int valid_user_sp (unsigned long sp,int is_64) {if (!sp || (sp &7 ) || sp > (is_64 ?TASK_SIZE :0x100000000UL ) -32 )@@ -277,19 +265,9 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry, }#else /* CONFIG_PPC64 */ -/* -* On 32-bit we just access the address and let hash_page create a -* HPTE if necessary, so there is no need to fall back to reading -* the page tables. Since this is called at interrupt level, -* do_page_fault() won't treat a DSI as a page fault. -*/ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_slow (void __user *ptr,void *buf,int nb) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -return probe_user_read (ret, ptr,sizeof (*ret)); +return 0 ; } static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,@@ -312,6 +290,28 @@ static inline int valid_user_sp(unsigned long sp, int is_64)#endif /* CONFIG_PPC64 */ Quoted 10 lines
arch/powerpc/perf/callchain.c static inline int valid_user_sp(unsigned long sp, int is_64) · line 298
Quoted 6 lines
arch/powerpc/perf/callchain.c +6
-0+/* +* On 32-bit we just access the address and let hash_page create a +* HPTE if necessary, so there is no need to fall back to reading +* the page tables. Since this is called at interrupt level, +* do_page_fault() won't treat a DSI as a page fault. +*/ The comment is actually probably better to stay in the 32-bit
read_user_stack_slowimplementation. Is that function defined on 32-bit purely so that you can useIS_ENABLED()? In that case
Quoted 2 lines
It documents the
IS_ENABLED()and that's where it is. The 32bit definition is only a technical detail.
Sorry for the late reply, busy trying to fix bugs in the C rewrite series. I don't think it is the right place, it should be in the ppc32 implementation detail. ppc64 has an equivalent comment at the top of its read_user_stack functions.
Quoted 3 lines
Quoted 1 line
I would prefer to put a
BUG()there which makes it self documenting.Which will cause checkpatch complaints about introducing new
BUG()which is frowned on.
It's fine in this case, that warning is about not introducing runtime bugs, but this wouldn't be.
But... I actually don't like adding read_user_stack_slow on 32-bit and especially not just to make IS_ENABLED work.
IMO this would be better if you really want to consolidate it
---
arch/powerpc/perf/callchain.c +17 -19
@@ -108,7 +108,7 @@ perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *re* interrupt context, so if the access faults, we read the page tables * to find which page (if any) is mapped and access it directly. */ -static int read_user_stack_slow (void __user *ptr,void *buf,int nb) +static int read_user_stack_slow (const void __user *ptr,void *buf,int nb) {int ret = -EFAULT ;pgd_t *pgdir;
@@ -149,28 +149,21 @@ static int read_user_stack_slow(void __user *ptr, void *buf, int nb)return ret; } -static int read_user_stack_64 (unsigned long __user *ptr,unsigned long *ret) +static int __read_user_stack (const void __user *ptr,void *ret,size_t size) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned long ) || - ((unsigned long )ptr &7 )) +if ((unsigned long )ptr >TASK_SIZE - size || + ((unsigned long )ptr & (size -1 )))return -EFAULT ; -if (!probe_user_read (ret, ptr,sizeof (*ret))) +if (!probe_user_read (ret, ptr, size))return 0 ; -return read_user_stack_slow (ptr, ret,8 ); +return read_user_stack_slow (ptr, ret, size); } -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_64 (unsigned long __user *ptr,unsigned long *ret) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,4 ); +return __read_user_stack (ptr, ret,sizeof (*ret)); }static inline int valid_user_sp (unsigned long sp,int is_64)
@@ -283,13 +276,13 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,* the page tables. Since this is called at interrupt level, * do_page_fault() won't treat a DSI as a page fault. */ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int __read_user_stack (const void __user *ptr,void *ret,size_t size) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) +if ((unsigned long )ptr >TASK_SIZE - size || + ((unsigned long )ptr & (size -1 )))return -EFAULT ; -return probe_user_read (ret, ptr,sizeof (*ret)); +return probe_user_read (ret, ptr, size); } static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,
@@ -312,6 +305,11 @@ static inline int valid_user_sp(unsigned long sp, int is_64)#endif /* CONFIG_PPC64 */ +static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +{ +return __read_user_stack (ptr, ret,sizeof (*ret)); +} +/* * Layout for non-RT signal frames */
Re: [PATCH v11 3/8] powerpc/perf: consolidate read_user_stack_32
Hello,
there are 3 variants of the function
read_user_stack_64
32bit read_user_stack_32
64bit read_user_Stack_32
Quoted 167 lines · Nicholas Piggin
Michal Suchánek's on March 25, 2020 5:38 am:
Quoted 76 lines · Nicholas Piggin
Quoted 65 lines
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 64 lines
There are two almost identical copies for 32bit and 64bit.
The function is used only in 32bit code which will be split out in next patch so consolidate to one function.
Signed-off-by Michal Suchanek <[email protected]>Reviewed-by Christophe Leroy <[email protected]>---
v6: new patch v8: move the consolidated function out of the ifdef block. v11: rebase on top of def0bfdbd603 --- arch/powerpc/perf/callchain.c | 48 +++++++++++++++++------------------ 1 file changed, 24 insertions(+), 24 deletions(-)arch/powerpc/perf/callchain.c +2
-24@@ -161,18 +161,6 @@ static int read_user_stack_64(unsigned long __user *ptr, unsigned long *ret)return read_user_stack_slow (ptr, ret,8 ); } -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) -{ -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,4 ); -} -static inline int valid_user_sp (unsigned long sp,int is_64) {if (!sp || (sp &7 ) || sp > (is_64 ?TASK_SIZE :0x100000000UL ) -32 )@@ -277,19 +265,9 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry, }#else /* CONFIG_PPC64 */ -/* -* On 32-bit we just access the address and let hash_page create a -* HPTE if necessary, so there is no need to fall back to reading -* the page tables. Since this is called at interrupt level, -* do_page_fault() won't treat a DSI as a page fault. -*/ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_slow (void __user *ptr,void *buf,int nb) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -return probe_user_read (ret, ptr,sizeof (*ret)); +return 0 ; } static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,@@ -312,6 +290,28 @@ static inline int valid_user_sp(unsigned long sp, int is_64)#endif /* CONFIG_PPC64 */ Quoted 10 lines
arch/powerpc/perf/callchain.c static inline int valid_user_sp(unsigned long sp, int is_64) · line 298
Quoted 6 lines
arch/powerpc/perf/callchain.c +6
-0+/* +* On 32-bit we just access the address and let hash_page create a +* HPTE if necessary, so there is no need to fall back to reading +* the page tables. Since this is called at interrupt level, +* do_page_fault() won't treat a DSI as a page fault. +*/ The comment is actually probably better to stay in the 32-bit
read_user_stack_slowimplementation. Is that function defined on 32-bit purely so that you can useIS_ENABLED()? In that caseQuoted 2 lines
It documents the
IS_ENABLED()and that's where it is. The 32bit definition is only a technical detail.Sorry for the late reply, busy trying to fix bugs in the C rewrite series. I don't think it is the right place, it should be in the ppc32 implementation detail. ppc64 has an equivalent comment at the top of its
read_user_stackfunctions.Quoted 3 lines
Quoted 1 line
I would prefer to put a
BUG()there which makes it self documenting.Which will cause checkpatch complaints about introducing new
BUG()which is frowned on.It's fine in this case, that warning is about not introducing runtime bugs, but this wouldn't be.
But... I actually don't like adding
read_user_stack_slowon 32-bit and especially not just to makeIS_ENABLEDwork.IMO this would be better if you really want to consolidate it
---
arch/powerpc/perf/callchain.c +12
-19@@ -108,7 +108,7 @@ perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *re* interrupt context, so if the access faults, we read the page tables * to find which page (if any) is mapped and access it directly. */ -static int read_user_stack_slow (void __user *ptr,void *buf,int nb) +static int read_user_stack_slow (const void __user *ptr,void *buf,int nb) {int ret = -EFAULT ;pgd_t *pgdir;@@ -149,28 +149,21 @@ static int read_user_stack_slow(void __user *ptr, void *buf, int nb)return ret; } -static int read_user_stack_64 (unsigned long __user *ptr,unsigned long *ret) +static int __read_user_stack (const void __user *ptr,void *ret,size_t size) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned long ) || - ((unsigned long )ptr &7 )) +if ((unsigned long )ptr >TASK_SIZE - size || + ((unsigned long )ptr & (size -1 )))return -EFAULT ; -if (!probe_user_read (ret, ptr,sizeof (*ret))) +if (!probe_user_read (ret, ptr, size))return 0 ; -return read_user_stack_slow (ptr, ret,8 ); +return read_user_stack_slow (ptr, ret, size); } -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_64 (unsigned long __user *ptr,unsigned long *ret) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,4 ); +return __read_user_stack (ptr, ret,sizeof (*ret)); }static inline int valid_user_sp (unsigned long sp,int is_64)@@ -283,13 +276,13 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,* the page tables. Since this is called at interrupt level, * do_page_fault() won't treat a DSI as a page fault. */ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int __read_user_stack (const void __user *ptr,void *ret,size_t size) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) +if ((unsigned long )ptr >TASK_SIZE - size || + ((unsigned long )ptr & (size -1 )))return -EFAULT ; -return probe_user_read (ret, ptr,sizeof (*ret)); +return probe_user_read (ret, ptr, size); } static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,
arch/powerpc/perf/callchain.c static inline int valid_user_sp(unsigned long sp, int is_64) · line 310
Quoted 6 lines
arch/powerpc/perf/callchain.c +3
-0@@ -312,6 +305,11 @@ static inline int valid_user_sp(unsigned long sp, int is_64)#endif /* CONFIG_PPC64 */ +static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +{ +return __read_user_stack (ptr, ret,sizeof (*ret));
Does not work for 64bit read_user_stack_32 ^ this should be 4.
Other than that it should preserve the existing logic just fine.
Thanks
Michal
Nicholas Piggin
reviewerRe: [PATCH v11 3/8] powerpc/perf: consolidate read_user_stack_32
Michal Suchánek's on April 3, 2020 8:52 pm:
Quoted 8 lines
Hello,
there are 3 variants of the function
read_user_stack_6432bit
read_user_stack_32
64bitread_user_Stack_32
Right.
Quoted 1 line · Nicholas Piggin
[...]
Quoted 8 lines
arch/powerpc/perf/callchain.c static inline int valid_user_sp(unsigned long sp, int is_64) · line 310
Quoted 5 lines
#endif /*
CONFIG_PPC64*/arch/powerpc/perf/callchain.c +3
-0+static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +{ +return __read_user_stack (ptr, ret,sizeof (*ret));Does not work for 64bit
read_user_stack_32^ this should be 4.Other than that it should preserve the existing logic just fine.
sizeof(int) == 4 on 64bit so it should work.
Thanks,
Nick
Re: [PATCH v11 3/8] powerpc/perf: consolidate read_user_stack_32
Quoted 16 lines · Nicholas Piggin
Michal Suchánek's on April 3, 2020 8:52 pm:
Quoted 8 lines
Hello,
there are 3 variants of the function
read_user_stack_6432bit
read_user_stack_32
64bitread_user_Stack_32Right.
Quoted 1 line · Nicholas Piggin
[...]
Quoted 2 lines
Quoted 2 lines
#endif /*
CONFIG_PPC64*/
Quoted 9 lines
Quoted 6 lines
arch/powerpc/perf/callchain.c static inline int valid_user_sp(unsigned long sp, int is_64) · line 310
Quoted 3 lines
arch/powerpc/perf/callchain.c +3
-0+static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +{ +return __read_user_stack (ptr, ret,sizeof (*ret));Does not work for 64bit
read_user_stack_32^ this should be 4.Other than that it should preserve the existing logic just fine.
sizeof(int) == 4 on 64bit so it should work.
Right, the type is different for the 32bit and 64bit version.
Thanks
Michal
Re: [PATCH v11 3/8] powerpc/perf: consolidate read_user_stack_32
In reply to Nicholas PigginQuoted 77 lines · Nicholas Piggin
Michal Suchánek's on March 25, 2020 5:38 am:
Quoted 76 lines · Nicholas Piggin
Quoted 75 lines
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 61 lines
There are two almost identical copies for 32bit and 64bit.
The function is used only in 32bit code which will be split out in next patch so consolidate to one function.
Signed-off-by Michal Suchanek <[email protected]>Reviewed-by Christophe Leroy <[email protected]>---
v6: new patch v8: move the consolidated function out of the ifdef block. v11: rebase on top of def0bfdbd603 --- arch/powerpc/perf/callchain.c | 48 +++++++++++++++++------------------ 1 file changed, 24 insertions(+), 24 deletions(-)arch/powerpc/perf/callchain.c +2
-24@@ -161,18 +161,6 @@ static int read_user_stack_64(unsigned long __user *ptr, unsigned long *ret)return read_user_stack_slow (ptr, ret,8 ); } -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) -{ -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,4 ); -} -static inline int valid_user_sp (unsigned long sp,int is_64) {if (!sp || (sp &7 ) || sp > (is_64 ?TASK_SIZE :0x100000000UL ) -32 )@@ -277,19 +265,9 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry, }#else /* CONFIG_PPC64 */ -/* -* On 32-bit we just access the address and let hash_page create a -* HPTE if necessary, so there is no need to fall back to reading -* the page tables. Since this is called at interrupt level, -* do_page_fault() won't treat a DSI as a page fault. -*/ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_slow (void __user *ptr,void *buf,int nb) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -return probe_user_read (ret, ptr,sizeof (*ret)); +return 0 ; } static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,arch/powerpc/perf/callchain.c static inline int valid_user_sp(unsigned long sp, int is_64) · line 298
Quoted 9 lines
arch/powerpc/perf/callchain.c +6
-0@@ -312,6 +290,28 @@ static inline int valid_user_sp(unsigned long sp, int is_64)#endif /* CONFIG_PPC64 */ +/* +* On 32-bit we just access the address and let hash_page create a +* HPTE if necessary, so there is no need to fall back to reading +* the page tables. Since this is called at interrupt level, +* do_page_fault() won't treat a DSI as a page fault. +*/ The comment is actually probably better to stay in the 32-bit
read_user_stack_slowimplementation. Is that function defined on 32-bit purely so that you can useIS_ENABLED()? In that case
Quoted 6 lines
Quoted 2 lines
It documents the
IS_ENABLED()and that's where it is. The 32bit definition is only a technical detail.Sorry for the late reply, busy trying to fix bugs in the C rewrite series. I don't think it is the right place, it should be in the ppc32 implementation detail.
Which does not exist anymore after the 32bit and 64bit part is split.
Quoted 11 lines
ppc64 has an equivalent comment at the top of its
read_user_stackfunctions.Quoted 3 lines
Quoted 1 line
I would prefer to put a
BUG()there which makes it self documenting.Which will cause checkpatch complaints about introducing new
BUG()which is frowned on.It's fine in this case, that warning is about not introducing runtime bugs, but this wouldn't be.
But... I actually don't like adding
read_user_stack_slowon 32-bit and especially not just to makeIS_ENABLEDwork.
That's to not break build at this point. Later the function is removed.
Quoted 84 lines
IMO this would be better if you really want to consolidate it
---
arch/powerpc/perf/callchain.c +17
-19@@ -108,7 +108,7 @@ perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *re* interrupt context, so if the access faults, we read the page tables * to find which page (if any) is mapped and access it directly. */ -static int read_user_stack_slow (void __user *ptr,void *buf,int nb) +static int read_user_stack_slow (const void __user *ptr,void *buf,int nb) {int ret = -EFAULT ;pgd_t *pgdir;@@ -149,28 +149,21 @@ static int read_user_stack_slow(void __user *ptr, void *buf, int nb)return ret; } -static int read_user_stack_64 (unsigned long __user *ptr,unsigned long *ret) +static int __read_user_stack (const void __user *ptr,void *ret,size_t size) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned long ) || - ((unsigned long )ptr &7 )) +if ((unsigned long )ptr >TASK_SIZE - size || + ((unsigned long )ptr & (size -1 )))return -EFAULT ; -if (!probe_user_read (ret, ptr,sizeof (*ret))) +if (!probe_user_read (ret, ptr, size))return 0 ; -return read_user_stack_slow (ptr, ret,8 ); +return read_user_stack_slow (ptr, ret, size); } -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_64 (unsigned long __user *ptr,unsigned long *ret) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,4 ); +return __read_user_stack (ptr, ret,sizeof (*ret)); }static inline int valid_user_sp (unsigned long sp,int is_64)@@ -283,13 +276,13 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,* the page tables. Since this is called at interrupt level, * do_page_fault() won't treat a DSI as a page fault. */ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int __read_user_stack (const void __user *ptr,void *ret,size_t size) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) +if ((unsigned long )ptr >TASK_SIZE - size || + ((unsigned long )ptr & (size -1 )))return -EFAULT ; -return probe_user_read (ret, ptr,sizeof (*ret)); +return probe_user_read (ret, ptr, size); } static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,@@ -312,6 +305,11 @@ static inline int valid_user_sp(unsigned long sp, int is_64)#endif /* CONFIG_PPC64 */ +static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +{ +return __read_user_stack (ptr, ret,sizeof (*ret)); +} +/* * Layout for non-RT signal frames */
PATCH 1/1 powerpcs: perf: consolidate perf_callchain_user_64 and perf_callchain_user_32
In reply to Nicholas Pigginperf_callchain_user_64 and perf_callchain_user_32 are nearly identical. Consolidate into one function with thin wrappers.
--- arch/powerpc/perf/callchain.h | 24 +++++++++++++++++++++++- arch/powerpc/perf/callchain_32.c | 21 ++------------------- arch/powerpc/perf/callchain_64.c | 14 ++++---------- 3 files changed, 29 insertions(+), 30 deletions(-)
arch/powerpc/perf/callchain.h +23 -1
@@ -2,7 +2,7 @@#ifndef _POWERPC_PERF_CALLCHAIN_H#define _POWERPC_PERF_CALLCHAIN_H -int read_user_stack_slow (void __user *ptr,void *buf,int nb); +int read_user_stack_slow (const void __user *ptr,void *buf,int nb);void perf_callchain_user_64 (struct perf_callchain_entry_ctx *entry,struct pt_regs *regs); void perf_callchain_user_32(struct perf_callchain_entry_ctx *entry,
@@ -16,4 +16,26 @@ static inline bool invalid_user_sp(unsigned long sp) return (!sp || (sp & mask) || (sp > top)); } +/* +* On 32-bit we just access the address and let hash_page create a +* HPTE if necessary, so there is no need to fall back to reading +* the page tables. Since this is called at interrupt level, +* do_page_fault() won't treat a DSI as a page fault. +*/ +static inline int __read_user_stack (const void __user *ptr,void *ret, +size_t size) +{ +int rc; + +if ((unsigned long )ptr >TASK_SIZE - size || + ((unsigned long )ptr & (size -1 ))) +return -EFAULT ; + rc =probe_user_read (ret, ptr, size); + +if (rc &&IS_ENABLED (CONFIG_PPC64 )) +return read_user_stack_slow (ptr, ret, size); + +return rc; +} +#endif /* _POWERPC_PERF_CALLCHAIN_H */
arch/powerpc/perf/callchain_32.c +2 -19
@@ -31,26 +31,9 @@#endif /* CONFIG_PPC64 */ -/* -* On 32-bit we just access the address and let hash_page create a -* HPTE if necessary, so there is no need to fall back to reading -* the page tables. Since this is called at interrupt level, -* do_page_fault() won't treat a DSI as a page fault. -*/ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_32 (const unsigned int __user *ptr,unsigned int *ret) { -int rc; - -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - - rc =probe_user_read (ret, ptr,sizeof (*ret)); - -if (IS_ENABLED (CONFIG_PPC64 ) && rc) -return read_user_stack_slow (ptr, ret,4 ); - -return rc; + return __read_user_stack(ptr, ret, sizeof(*ret); }/*
arch/powerpc/perf/callchain_64.c +4 -10
@@ -24,7 +24,7 @@* interrupt context, so if the access faults, we read the page tables * to find which page (if any) is mapped and access it directly. */ -int read_user_stack_slow (void __user *ptr,void *buf,int nb) +int read_user_stack_slow (const void __user *ptr,void *buf,int nb) {int ret = -EFAULT ;pgd_t *pgdir;
@@ -65,16 +65,10 @@ int read_user_stack_slow(void __user *ptr, void *buf, int nb)return ret; } -static int read_user_stack_64 (unsigned long __user *ptr,unsigned long *ret) +static int read_user_stack_64 (const unsigned long __user *ptr, +unsigned long *ret) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned long ) || - ((unsigned long )ptr &7 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,8 ); +return __read_user_stack (ptr, ret,sizeof (*ret)); }/*
Signature · 1 line
2.23.0
Re: [PATCH] powerpcs: perf: consolidate perf_callchain_user_64 and perf_callchain_user_32
Le 06/04/2020 à 23:00, Michal Suchanek a écrit :
Quoted 16 lines
perf_callchain_user_64andperf_callchain_user_32are nearly identical. Consolidate into one function with thin wrappers.Suggested-by Nicholas Piggin <[email protected]>Signed-off-by Michal Suchanek <[email protected]>--- arch/powerpc/perf/callchain.h | 24 +++++++++++++++++++++++- arch/powerpc/perf/callchain_32.c | 21 ++------------------- arch/powerpc/perf/callchain_64.c | 14 ++++---------- 3 files changed, 29 insertions(+), 30 deletions(-)arch/powerpc/perf/callchain.h +0
-0
arch/powerpc/perf/callchain.h line 5
Quoted 5 lines
arch/powerpc/perf/callchain.h +1
-1@@ -2,7 +2,7 @@#ifndef _POWERPC_PERF_CALLCHAIN_H#define _POWERPC_PERF_CALLCHAIN_H -int read_user_stack_slow (void __user *ptr,void *buf,int nb); +int read_user_stack_slow (const void __user *ptr,void *buf,int nb);
Does the constification of ptr has to be in this patch ? Wouldn't it be better to have it as a separate patch ?
Quoted 14 lines
arch/powerpc/perf/callchain.h +7
-0void perf_callchain_user_64 (struct perf_callchain_entry_ctx *entry,struct pt_regs *regs); void perf_callchain_user_32(struct perf_callchain_entry_ctx *entry,@@ -16,4 +16,26 @@ static inline bool invalid_user_sp(unsigned long sp)return (!sp || (sp & mask) || (sp > top)); } +/* +* On 32-bit we just access the address and let hash_page create a +* HPTE if necessary, so there is no need to fall back to reading +* the page tables. Since this is called at interrupt level, +* do_page_fault() won't treat a DSI as a page fault. +*/ +static inline int __read_user_stack(const void __user *ptr, void *ret,
arch/powerpc/perf/callchain.h static inline bool invalid_user_sp(unsigned long sp) · line 35
Quoted 10 lines
arch/powerpc/perf/callchain.h +10
-0+ size_t size) +{ +int rc; + +if ((unsigned long )ptr >TASK_SIZE - size || + ((unsigned long )ptr & (size -1 ))) +return -EFAULT ; + rc =probe_user_read (ret, ptr, size); + +if (rc &&IS_ENABLED (CONFIG_PPC64 ))
gcc is probably smart enough to deal with it efficiently, but it would be more correct to test rc after checking CONFIG_PPC64.
Quoted 63 lines
arch/powerpc/perf/callchain.h +5
-0+return read_user_stack_slow (ptr, ret, size); + + return rc; +} +#endif /* _POWERPC_PERF_CALLCHAIN_H */ arch/powerpc/perf/callchain_32.c +2
-19@@ -31,26 +31,9 @@#endif /* CONFIG_PPC64 */ -/* -* On 32-bit we just access the address and let hash_page create a -* HPTE if necessary, so there is no need to fall back to reading -* the page tables. Since this is called at interrupt level, -* do_page_fault() won't treat a DSI as a page fault. -*/ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_32 (const unsigned int __user *ptr,unsigned int *ret) { -int rc; - -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - - rc =probe_user_read (ret, ptr,sizeof (*ret)); - -if (IS_ENABLED (CONFIG_PPC64 ) && rc) -return read_user_stack_slow (ptr, ret,4 ); - -return rc; + return __read_user_stack(ptr, ret, sizeof(*ret); }/* arch/powerpc/perf/callchain_64.c +3
-4@@ -24,7 +24,7 @@* interrupt context, so if the access faults, we read the page tables * to find which page (if any) is mapped and access it directly. */ -int read_user_stack_slow (void __user *ptr,void *buf,int nb) +int read_user_stack_slow (const void __user *ptr,void *buf,int nb) {int ret = -EFAULT ;pgd_t *pgdir;@@ -65,16 +65,10 @@ int read_user_stack_slow(void __user *ptr, void *buf, int nb)return ret; } -static int read_user_stack_64 (unsigned long __user *ptr,unsigned long *ret) +static int read_user_stack_64 (const unsigned long __user *ptr, +unsigned long *ret) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned long ) || - ((unsigned long )ptr &7 ))
arch/powerpc/perf/callchain_64.c int read_user_stack_slow(void __user *ptr, void *buf, int nb) · line 74
Quoted 10 lines
arch/powerpc/perf/callchain_64.c +1
-6-return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,8 ); +return __read_user_stack (ptr, ret,sizeof (*ret)); }/*
Christophe
Re: [PATCH] powerpcs: perf: consolidate perf_callchain_user_64 and perf_callchain_user_32
Quoted 14 lines · Christophe Leroy
Le 06/04/2020 à 23:00, Michal Suchanek a écrit :
Quoted 11 lines
perf_callchain_user_64andperf_callchain_user_32are nearly identical. Consolidate into one function with thin wrappers.Suggested-by Nicholas Piggin <[email protected]>Signed-off-by Michal Suchanek <[email protected]>--- arch/powerpc/perf/callchain.h | 24 +++++++++++++++++++++++- arch/powerpc/perf/callchain_32.c | 21 ++------------------- arch/powerpc/perf/callchain_64.c | 14 ++++---------- 3 files changed, 29 insertions(+), 30 deletions(-)
Quoted 11 lines
arch/powerpc/perf/callchain.h line 4
Quoted 9 lines
arch/powerpc/perf/callchain.h +1
-1@@ -2,7 +2,7 @@#ifndef _POWERPC_PERF_CALLCHAIN_H#define _POWERPC_PERF_CALLCHAIN_H -int read_user_stack_slow (void __user *ptr,void *buf,int nb); +int read_user_stack_slow (const void __user *ptr,void *buf,int nb);Does the constification of ptr has to be in this patch ?
It was in the original patch. The code is touched anyway.
Quoted 1 line
Wouldn't it be better to have it as a separate patch ?
Don't care much either way. Can resend it as separate patches.
Quoted 19 lines
Quoted 19 lines
arch/powerpc/perf/callchain.h +12
-0void perf_callchain_user_64 (struct perf_callchain_entry_ctx *entry,struct pt_regs *regs); void perf_callchain_user_32(struct perf_callchain_entry_ctx *entry,@@ -16,4 +16,26 @@ static inline bool invalid_user_sp(unsigned long sp)return (!sp || (sp & mask) || (sp > top)); } +/* +* On 32-bit we just access the address and let hash_page create a +* HPTE if necessary, so there is no need to fall back to reading +* the page tables. Since this is called at interrupt level, +* do_page_fault() won't treat a DSI as a page fault. +*/ +static inline int __read_user_stack (const void __user *ptr,void *ret, +size_t size) +{ +int rc; + + if ((unsigned long)ptr > TASK_SIZE - size ||
Quoted 8 lines
arch/powerpc/perf/callchain.h static inline bool invalid_user_sp(unsigned long sp) · line 35
Quoted 5 lines
arch/powerpc/perf/callchain.h +5
-0+ ((unsigned long)ptr & (size - 1))) +return -EFAULT ; + rc =probe_user_read (ret, ptr, size); + +if (rc &&IS_ENABLED (CONFIG_PPC64 ))gcc is probably smart enough to deal with it efficiently, but it would be more correct to test rc after checking
CONFIG_PPC64.
IS_ENABLED(CONFIG_PPC64) is constant so that part of the check should be compiled out in any case.
Thanks
Michal
PATCH v11 4/8 powerpc/perf: consolidate valid_user_sp
In reply to Michal SuchanekMerge the 32bit and 64bit version.
Halve the check constants on 32bit.
Use STACK_TOP since it is defined.
Passing is_64 is now redundant since is_32bit_task() is used to determine which callchain variant should be used. Use STACK_TOP and is_32bit_task() directly.
This removes a page from the valid 32bit area on 64bit:
#define TASK_SIZE_USER32 (0x0000000100000000UL - (1 * PAGE_SIZE)) #define STACK_TOP_USER32 TASK_SIZE_USER32
---
v8: new patch
v11: simplify by using is_32bit_task()
--- arch/powerpc/perf/callchain.c | 27 +++++++++++---------------- 1 file changed, 11 insertions(+), 16 deletions(-)
arch/powerpc/perf/callchain.c +11 -16
@@ -102,6 +102,15 @@ perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *re } } +static inline int valid_user_sp (unsigned long sp) +{ +bool is_64 = !is_32bit_task (); + +if (!sp || (sp & (is_64 ?7 :3 )) || sp >STACK_TOP - (is_64 ?32 :16 )) +return 0 ; +return 1 ; +} +#ifdef CONFIG_PPC64 /* * On 64-bit we don't want to invoke hash_page on user addresses from
@@ -161,13 +170,6 @@ static int read_user_stack_64(unsigned long __user *ptr, unsigned long *ret)return read_user_stack_slow (ptr, ret,8 ); } -static inline int valid_user_sp (unsigned long sp,int is_64) -{ -if (!sp || (sp &7 ) || sp > (is_64 ?TASK_SIZE :0x100000000UL ) -32 ) -return 0 ; -return 1 ; -} -/* * 64-bit user processes use the same stack frame for RT and non-RT signals. */
@@ -226,7 +228,7 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,while (entry->nr < entry->max_stack) { fp = (unsigned long __user *) sp; -if (!valid_user_sp (sp,1 ) ||read_user_stack_64 (fp, &next_sp)) +if (!valid_user_sp (sp) ||read_user_stack_64 (fp, &next_sp))return ;if (level >0 &&read_user_stack_64 (&fp[2 ], &next_ip))return ;
@@ -275,13 +277,6 @@ static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry { } -static inline int valid_user_sp (unsigned long sp,int is_64) -{ -if (!sp || (sp &7 ) || sp >TASK_SIZE -32 ) -return 0 ; -return 1 ; -} -#define __SIGNAL_FRAMESIZE32 __SIGNAL_FRAMESIZE#define sigcontext32 sigcontext#define mcontext32 mcontext
@@ -423,7 +418,7 @@ static void perf_callchain_user_32(struct perf_callchain_entry_ctx *entry,while (entry->nr < entry->max_stack) { fp = (unsigned int __user *) (unsigned long ) sp; -if (!valid_user_sp (sp,0 ) ||read_user_stack_32 (fp, &next_sp)) +if (!valid_user_sp (sp) ||read_user_stack_32 (fp, &next_sp))return ;if (level >0 &&read_user_stack_32 (&fp[1 ], &next_ip))return ;
Signature · 1 line
2.23.0
PATCH v11 5/8 powerpc/64: make buildable without CONFIG_COMPAT
In reply to Michal SuchanekThere are numerous references to 32bit functions in generic and 64bit code so ifdef them out.
---
v2:
- fix 32bit ifdef condition in signal.c
- simplify the compat ifdef condition in vdso.c - 64bit is redundant
- simplify the compat ifdef condition in callchain.c - 64bit is redundant v3:
- use IS_ENABLED and maybe_unused where possible
- do not ifdef declarations
- clean up Makefile
v4:
- further makefile cleanup
- simplify is_32bit_task conditions
- avoid ifdef in condition by using return
v5:
- avoid unreachable code on 32bit
- make is_current_64bit constant on !COMPAT
- add stub perf_callchain_user_32 to avoid some ifdefs
v6:
- consolidate current_is_64bit
v7:
- remove leftover perf_callchain_user_32 stub from previous series version v8:
- fix build again - too trigger-happy with stub removal
- remove a vdso.c hunk that causes warning according to kbuild test robot v9:
- removed current_is_64bit in previous patch
v10:
- rebase on top of 70ed86f4de5bd
--- arch/powerpc/include/asm/thread_info.h | 4 ++-- arch/powerpc/kernel/Makefile | 6 +++--- arch/powerpc/kernel/entry_64.S | 2 ++ arch/powerpc/kernel/signal.c | 3 +-- arch/powerpc/kernel/syscall_64.c | 6 ++---- arch/powerpc/kernel/vdso.c | 3 ++- arch/powerpc/perf/callchain.c | 8 +++++++- 7 files changed, 19 insertions(+), 13 deletions(-)
arch/powerpc/include/asm/thread_info.h +2 -2
@@ -162,10 +162,10 @@ static inline bool test_thread_local_flags(unsigned int flags)return (ti->local_flags & flags) !=0 ; } -#ifdef CONFIG_PPC64 +#ifdef CONFIG_COMPAT #define is_32bit_task () (test_thread_flag(TIF_32BIT))#else -#define is_32bit_task () (1) +#define is_32bit_task () (IS_ENABLED(CONFIG_PPC32))#endif #if defined(CONFIG_PPC64 )
arch/powerpc/kernel/Makefile +3 -3
@@ -42,16 +42,16 @@ CFLAGS_btext.o += -DDISABLE_BRANCH_PROFILING endif obj-y := cputable.o ptrace.o syscalls.o \ - irq.o align.o signal_32.o pmc.o vdso.o \ + irq.o align.o signal_$(BITS).o pmc.o vdso.o \ process.o systbl.o idle.o \ signal.o sysfs.o cacheinfo.o time.o \ prom.o traps.o setup-common.o \ udbg.o misc.o io.o misc_$(BITS).o \ of_platform.o prom_parse.o -obj-$(CONFIG_PPC64) += setup_64.o sys_ppc32.o \ - signal_64.o ptrace32.o \ +obj-$(CONFIG_PPC64) += setup_64.o \ paca.o nvram_64.o firmware.o note.o \ syscall_64.o +obj-$(CONFIG_COMPAT) += sys_ppc32.o ptrace32.o signal_32.o obj-$(CONFIG_VDSO32) += vdso32/ obj-$(CONFIG_PPC_WATCHDOG) += watchdog.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o
arch/powerpc/kernel/entry_64.S +2 -0
@@ -52,8 +52,10 @@ SYS_CALL_TABLE: .tc sys_call_table[TC],sys_call_table +#ifdef CONFIG_COMPAT COMPAT_SYS_CALL_TABLE: .tc compat_sys_call_table[TC],compat_sys_call_table +#endif /* This value is used to mark exception frames on the stack. */ exception_marker:
arch/powerpc/kernel/signal.c +1 -2
@@ -247,7 +247,6 @@ static void do_signal(struct task_struct *tsk)sigset_t *oldset =sigmask_to_save ();struct ksignal ksig = { .sig =0 };int ret; -int is32 =is_32bit_task ();BUG_ON (tsk != current);
@@ -277,7 +276,7 @@ static void do_signal(struct task_struct *tsk)rseq_signal_deliver (&ksig, tsk->thread.regs); -if (is32) { +if (is_32bit_task ()) {if (ksig.ka.sa.sa_flags &SA_SIGINFO ) ret =handle_rt_signal32 (&ksig, oldset, tsk);else
arch/powerpc/kernel/syscall_64.c +2 -4
@@ -24,7 +24,6 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7, long r8, unsigned long r0, struct pt_regs *regs) { - unsigned long ti_flags; syscall_fn f;if (IS_ENABLED (CONFIG_PPC_IRQ_SOFT_MASK_DEBUG ))
@@ -68,8 +67,7 @@ notrace long system_call_exception(long r3, long r4, long r5,local_irq_enable (); - ti_flags =current_thread_info ()->flags; -if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) { +if (unlikely (current_thread_info ()->flags & _TIF_SYSCALL_DOTRACE)) {/* * We use the return value of do_syscall_trace_enter() as the * syscall number. If the syscall was rejected for any reason
@@ -94,7 +92,7 @@ notrace long system_call_exception(long r3, long r4, long r5,/* May be faster to do array_index_nospec? */ barrier_nospec (); -if (unlikely (ti_flags & _TIF_32BIT)) { +if (unlikely (is_32bit_task ())) { f = (void *)compat_sys_call_table[r0]; r3 &=0x00000000ffffffffULL ;
arch/powerpc/kernel/vdso.c +2 -1
@@ -656,7 +656,8 @@ static void __init vdso_setup_syscall_map(void)if (sys_call_table[i] != sys_ni_syscall) vdso_data->syscall_map_64[i >>5 ] |=0x80000000UL >> (i &0x1f ); -if (compat_sys_call_table[i] != sys_ni_syscall) +if (IS_ENABLED (CONFIG_COMPAT ) && + compat_sys_call_table[i] != sys_ni_syscall) vdso_data->syscall_map_32[i >>5 ] |=0x80000000UL >> (i &0x1f );#else /* CONFIG_PPC64 */
arch/powerpc/perf/callchain.c +7 -1
@@ -15,7 +15,7 @@#include <asm/sigcontext.h> #include <asm/ucontext.h> #include <asm/vdso.h> -#ifdef CONFIG_PPC64 +#ifdef CONFIG_COMPAT #include "../kernel/ppc32.h" #endif #include <asm/pte-walk.h>
@@ -285,6 +285,7 @@ static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry#endif /* CONFIG_PPC64 */ +#if defined(CONFIG_PPC32 ) || defined(CONFIG_COMPAT )/* * On 32-bit we just access the address and let hash_page create a * HPTE if necessary, so there is no need to fall back to reading
@@ -448,6 +449,11 @@ static void perf_callchain_user_32(struct perf_callchain_entry_ctx *entry, sp = next_sp; } } +#else /* 32bit */ +static void perf_callchain_user_32 (struct perf_callchain_entry_ctx *entry, +struct pt_regs *regs) +{} +#endif /* 32bit */ void perf_callchain_user (struct perf_callchain_entry_ctx *entry,struct pt_regs *regs)
Signature · 1 line
2.23.0
Nicholas Piggin
reviewerRe: [PATCH v11 5/8] powerpc/64: make buildable without CONFIG_COMPAT
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 13 lines
arch/powerpc/kernel/signal.c +0
-1@@ -247,7 +247,6 @@ static void do_signal(struct task_struct *tsk)sigset_t *oldset =sigmask_to_save ();struct ksignal ksig = { .sig =0 };int ret; -int is32 =is_32bit_task ();BUG_ON (tsk != current);
arch/powerpc/kernel/signal.c static void do_signal(struct task_struct *tsk) · line 282
Quoted 8 lines
arch/powerpc/kernel/signal.c +1
-1@@ -277,7 +276,7 @@ static void do_signal(struct task_struct *tsk)rseq_signal_deliver (&ksig, tsk->thread.regs); -if (is32) { +if (is_32bit_task ()) {if (ksig.ka.sa.sa_flags &SA_SIGINFO ) ret =handle_rt_signal32 (&ksig, oldset, tsk);else
Unnecessary?
Quoted 23 lines
arch/powerpc/kernel/syscall_64.c +1
-3@@ -24,7 +24,6 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7, long r8, unsigned long r0, struct pt_regs *regs) { - unsigned long ti_flags; syscall_fn f;if (IS_ENABLED (CONFIG_PPC_IRQ_SOFT_MASK_DEBUG ))@@ -68,8 +67,7 @@ notrace long system_call_exception(long r3, long r4, long r5,local_irq_enable (); - ti_flags =current_thread_info ()->flags; -if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) { +if (unlikely (current_thread_info ()->flags & _TIF_SYSCALL_DOTRACE)) {/* * We use the return value of do_syscall_trace_enter() as the * syscall number. If the syscall was rejected for any reason
arch/powerpc/kernel/syscall_64.c notrace long system_call_exception(long r3, long r4, long r5, · line 95
Quoted 5 lines
arch/powerpc/kernel/syscall_64.c +1
-1@@ -94,7 +92,7 @@ notrace long system_call_exception(long r3, long r4, long r5,/* May be faster to do array_index_nospec? */ barrier_nospec (); -if (unlikely (ti_flags & _TIF_32BIT)) { +if (unlikely (is_32bit_task ())) {
Problem is, does this allow the load of ti_flags to be used for both tests, or does test_bit make it re-load?
This could maybe be fixed by testing if(IS_ENABLED(CONFIG_COMPAT) &&
Other than these, the patches all look pretty good to me.
Thanks,
Nick
Re: [PATCH v11 5/8] powerpc/64: make buildable without CONFIG_COMPAT
Quoted 51 lines · Nicholas Piggin
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 13 lines
arch/powerpc/kernel/signal.c +0
-1@@ -247,7 +247,6 @@ static void do_signal(struct task_struct *tsk)sigset_t *oldset =sigmask_to_save ();struct ksignal ksig = { .sig =0 };int ret; -int is32 =is_32bit_task ();BUG_ON (tsk != current);arch/powerpc/kernel/signal.c static void do_signal(struct task_struct *tsk) · line 282
Quoted 8 lines
arch/powerpc/kernel/signal.c +1
-1@@ -277,7 +276,7 @@ static void do_signal(struct task_struct *tsk)rseq_signal_deliver (&ksig, tsk->thread.regs); -if (is32) { +if (is_32bit_task ()) {if (ksig.ka.sa.sa_flags &SA_SIGINFO ) ret =handle_rt_signal32 (&ksig, oldset, tsk);else Unnecessary?
Quoted 26 lines
arch/powerpc/kernel/syscall_64.c +1
-3@@ -24,7 +24,6 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7, long r8, unsigned long r0, struct pt_regs *regs) { - unsigned long ti_flags; syscall_fn f;if (IS_ENABLED (CONFIG_PPC_IRQ_SOFT_MASK_DEBUG ))@@ -68,8 +67,7 @@ notrace long system_call_exception(long r3, long r4, long r5,local_irq_enable (); - ti_flags =current_thread_info ()->flags; -if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) { +if (unlikely (current_thread_info ()->flags & _TIF_SYSCALL_DOTRACE)) {/* * We use the return value of do_syscall_trace_enter() as the * syscall number. If the syscall was rejected for any reason @@ -94,7 +92,7 @@ notrace long system_call_exception(long r3, long r4, long r5,/* May be faster to do array_index_nospec? */ barrier_nospec ();
Quoted 7 lines
arch/powerpc/kernel/syscall_64.c notrace long system_call_exception(long r3, long r4, long r5, · line 95
Quoted 2 lines
arch/powerpc/kernel/syscall_64.c +1
-1-if (unlikely (ti_flags & _TIF_32BIT)) { +if (unlikely (is_32bit_task ())) {Problem is, does this allow the load of
ti_flagsto be used for both tests, or doestest_bitmake it re-load?This could maybe be fixed by testing if(IS_ENABLED(CONFIG_COMPAT) &&
Both points already discussed here:
Thanks
Michal
Nicholas Piggin
reviewerRe: [PATCH v11 5/8] powerpc/64: make buildable without CONFIG_COMPAT
Michal Suchánek's on March 25, 2020 5:30 am:
Quoted 52 lines · Nicholas Piggin
Quoted 51 lines
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 13 lines
arch/powerpc/kernel/signal.c +0
-1@@ -247,7 +247,6 @@ static void do_signal(struct task_struct *tsk)sigset_t *oldset =sigmask_to_save ();struct ksignal ksig = { .sig =0 };int ret; -int is32 =is_32bit_task ();BUG_ON (tsk != current);arch/powerpc/kernel/signal.c static void do_signal(struct task_struct *tsk) · line 282
Quoted 8 lines
arch/powerpc/kernel/signal.c +1
-1@@ -277,7 +276,7 @@ static void do_signal(struct task_struct *tsk)rseq_signal_deliver (&ksig, tsk->thread.regs); -if (is32) { +if (is_32bit_task ()) {if (ksig.ka.sa.sa_flags &SA_SIGINFO ) ret =handle_rt_signal32 (&ksig, oldset, tsk);else Unnecessary?
Quoted 26 lines
arch/powerpc/kernel/syscall_64.c +1
-3@@ -24,7 +24,6 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7, long r8, unsigned long r0, struct pt_regs *regs) { - unsigned long ti_flags; syscall_fn f;if (IS_ENABLED (CONFIG_PPC_IRQ_SOFT_MASK_DEBUG ))@@ -68,8 +67,7 @@ notrace long system_call_exception(long r3, long r4, long r5,local_irq_enable (); - ti_flags =current_thread_info ()->flags; -if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) { +if (unlikely (current_thread_info ()->flags & _TIF_SYSCALL_DOTRACE)) {/* * We use the return value of do_syscall_trace_enter() as the * syscall number. If the syscall was rejected for any reason @@ -94,7 +92,7 @@ notrace long system_call_exception(long r3, long r4, long r5,/* May be faster to do array_index_nospec? */ barrier_nospec ();
Quoted 8 lines
Quoted 7 lines
arch/powerpc/kernel/syscall_64.c notrace long system_call_exception(long r3, long r4, long r5, · line 95
Quoted 2 lines
arch/powerpc/kernel/syscall_64.c +1
-1-if (unlikely (ti_flags & _TIF_32BIT)) { +if (unlikely (is_32bit_task ())) {Problem is, does this allow the load of
ti_flagsto be used for both tests, or doestest_bitmake it re-load?This could maybe be fixed by testing if(IS_ENABLED(CONFIG_COMPAT) &&
Both points already discussed here:
Agh, I'm hopeless.
I don't think it really resolves this issue. But probably don't have time to look at generated asm, and might never because it won't really hit LE unless we add a 32-bit ABI. It's pretty minor though either way.
Sorry for being difficult, I really do like your patches :)
Thanks,
Nick
PATCH v11 6/8 powerpc/64: Make COMPAT user-selectable disabled on littleendian by default.
In reply to Michal SuchanekOn bigendian ppc64 it is common to have 32bit legacy binaries but much less so on littleendian.
---
v3: make configurable
--- arch/powerpc/Kconfig | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
arch/powerpc/Kconfig +3 -2
@@ -264,8 +264,9 @@ config PANIC_TIMEOUT default 180 config COMPAT - bool - default y if PPC64 + bool "Enable support for 32bit binaries" + depends on PPC64 + default y if !CPU_LITTLE_ENDIAN select COMPAT_BINFMT_ELF select ARCH_WANT_OLD_COMPAT_IPC select COMPAT_OLD_SIGACTION
Signature · 1 line
2.23.0
PATCH v11 7/8 powerpc/perf: split callchain.c by bitness
In reply to Michal SuchanekBuilding callchain.c with !COMPAT proved quite ugly with all the defines. Splitting out the 32bit and 64bit parts looks better.
No code change intended.
---
v6:
- move current_is_64bit consolidetaion to earlier patch
- move defines to the top of callchain_32.c
- Makefile cleanup
v8:
- fix valid_user_sp
v11:
- rebase on top of def0bfdbd603
--- arch/powerpc/perf/Makefile | 5 +- arch/powerpc/perf/callchain.c | 357 +------------------------------ arch/powerpc/perf/callchain.h | 20 ++ arch/powerpc/perf/callchain_32.c | 196 +++++++++++++++++ arch/powerpc/perf/callchain_64.c | 174 +++++++++++++++ 5 files changed, 395 insertions(+), 357 deletions(-) create mode 100644 arch/powerpc/perf/callchain.h create mode 100644 arch/powerpc/perf/callchain_32.c create mode 100644 arch/powerpc/perf/callchain_64.c
arch/powerpc/perf/Makefile +4 -1
@@ -1,6 +1,9 @@ # SPDX-License-Identifier: GPL-2.0 -obj-$(CONFIG_PERF_EVENTS) += callchain.o perf_regs.o +obj-$(CONFIG_PERF_EVENTS) += callchain.o callchain_$(BITS).o perf_regs.o +ifdef CONFIG_COMPAT +obj-$(CONFIG_PERF_EVENTS) += callchain_32.o +endif obj-$(CONFIG_PPC_PERF_CTRS) += core-book3s.o bhrb.o obj64-$(CONFIG_PPC_PERF_CTRS) += ppc970-pmu.o power5-pmu.o \
arch/powerpc/perf/callchain.c +1 -356
371 lines not on this page · raw message
arch/powerpc/perf/callchain.h new file +20 -0
@@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +#ifndef _POWERPC_PERF_CALLCHAIN_H +#define _POWERPC_PERF_CALLCHAIN_H + +int read_user_stack_slow (void __user *ptr,void *buf,int nb); +void perf_callchain_user_64 (struct perf_callchain_entry_ctx *entry, +struct pt_regs *regs); +void perf_callchain_user_32 (struct perf_callchain_entry_ctx *entry, +struct pt_regs *regs); + +static inline int valid_user_sp (unsigned long sp) +{ +bool is_64 = !is_32bit_task (); + +if (!sp || (sp & (is_64 ?7 :3 )) || sp >STACK_TOP - (is_64 ?32 :16 )) +return 0 ; +return 1 ; +} + +#endif /* _POWERPC_PERF_CALLCHAIN_H */
arch/powerpc/perf/callchain_32.c new file +196 -0
196 lines not on this page · raw message
arch/powerpc/perf/callchain_64.c new file +174 -0
174 lines not on this page · raw message
Signature · 1 line
2.23.0
PATCH v11 8/8 MAINTAINERS: perf: Add pattern that matches ppc perf to the perf entry.
In reply to Michal Suchanek---
v10: new patch
--- MAINTAINERS | 2 ++ 1 file changed, 2 insertions(+)
MAINTAINERS +2 -0
@@ -13088,6 +13088,8 @@ F: arch/*/kernel/*/perf_event*.c F: arch/*/kernel/*/*/perf_event*.c F: arch/*/include/asm/perf_event.h F: arch/*/kernel/perf_callchain.c +F: arch/*/perf/* +F: arch/*/perf/*/* F: arch/*/events/* F: arch/*/events/*/* F: tools/perf/
Signature · 1 line
2.23.0
Re: [PATCH v11 8/8] MAINTAINERS: perf: Add pattern that matches ppc perf to the perf entry.
Quoted 13 lines · Michal Suchanek
Signed-off-by Michal Suchanek <[email protected]>---
v10: new patch--- MAINTAINERS | 2 ++ 1 file changed, 2 insertions(+)MAINTAINERS +0
-0
MAINTAINERS F: arch/*/kernel/*/perf_event*.c · line 13095
Quoted 8 lines
MAINTAINERS +2
-0@@ -13088,6 +13088,8 @@ F: arch/*/kernel/*/perf_event*.c F: arch/*/kernel/*/*/perf_event*.c F: arch/*/include/asm/perf_event.h F: arch/*/kernel/perf_callchain.c +F: arch/*/perf/* +F: arch/*/perf/*/* F: arch/*/events/* F: arch/*/events/*/* F: tools/perf/
Had you run parse-maintainers.pl?
Signature · 2 lines
With Best Regards, Andy Shevchenko
Re: [PATCH v11 8/8] MAINTAINERS: perf: Add pattern that matches ppc perf to the perf entry.
Quoted 9 lines · Andy Shevchenko
Quoted 8 lines · Michal Suchanek
Signed-off-by Michal Suchanek <[email protected]>---
v10: new patch--- MAINTAINERS | 2 ++ 1 file changed, 2 insertions(+)
Quoted 15 lines
MAINTAINERS F: arch/*/kernel/*/perf_event*.c · line 13095
Quoted 13 lines
MAINTAINERS +2
-0@@ -13088,6 +13088,8 @@ F: arch/*/kernel/*/perf_event*.c F: arch/*/kernel/*/*/perf_event*.c F: arch/*/include/asm/perf_event.h F: arch/*/kernel/perf_callchain.c +F: arch/*/perf/* +F: arch/*/perf/*/* F: arch/*/events/* F: arch/*/events/*/* F: tools/perf/Had you run parse-maintainers.pl?
Did not know it exists. The output is:
scripts/parse-maintainers.pl
Odd non-pattern line ' Documentation/devicetree/bindings/media/ti,cal.yaml
' for 'TI VPE/CAL DRIVERS' at scripts/parse-maintainers.pl line 147, <$file> line 16756.
Thanks
Michal
Re: [PATCH v11 8/8] MAINTAINERS: perf: Add pattern that matches ppc perf to the perf entry.
Quoted 25 lines · Michal Suchánek
Quoted 9 lines · Andy Shevchenko
Quoted 8 lines · Michal Suchanek
Signed-off-by Michal Suchanek <[email protected]>---
v10: new patch--- MAINTAINERS | 2 ++ 1 file changed, 2 insertions(+)Quoted 15 lines
MAINTAINERS F: arch/*/kernel/*/perf_event*.c · line 13095
Quoted 13 lines
MAINTAINERS +2
-0@@ -13088,6 +13088,8 @@ F: arch/*/kernel/*/perf_event*.c F: arch/*/kernel/*/*/perf_event*.c F: arch/*/include/asm/perf_event.h F: arch/*/kernel/perf_callchain.c +F: arch/*/perf/* +F: arch/*/perf/*/* F: arch/*/events/* F: arch/*/events/*/* F: tools/perf/Had you run parse-maintainers.pl?
Quoted 7 lines
Did not know it exists. The output is:
scripts/parse-maintainers.pl
Odd non-pattern line 'Documentation/devicetree/bindings/media/ti,cal.yaml
' for 'TI VPE/CAL DRIVERS' atscripts/parse-maintainers.plline 147, <$file> line 16756.
It is fixed in media tree and available in linux next as
d44535cb14c9 ("media: MAINTAINERS: Sort entries in database for TI VPE/CAL")
Signature · 2 lines
With Best Regards, Andy Shevchenko
Re: [PATCH v11 8/8] MAINTAINERS: perf: Add pattern that matches ppc perf to the perf entry.
In reply to Michal SuchanekQuoted 12 lines · Michal Suchanek
Signed-off-by Michal Suchanek <[email protected]>---
v10: new patch--- MAINTAINERS | 2 ++ 1 file changed, 2 insertions(+)MAINTAINERS +0
-0
MAINTAINERS F: arch/*/kernel/*/perf_event*.c · line 13092
Quoted 5 lines
MAINTAINERS +2
-0@@ -13088,6 +13088,8 @@ F: arch/*/kernel/*/perf_event*.c F: arch/*/kernel/*/*/perf_event*.c F: arch/*/include/asm/perf_event.h F: arch/*/kernel/perf_callchain.c +F: arch/*/perf/* +F: arch/*/perf/*/*
While I understand the desire, I believe that repetitive listings like this don't really help much.
Having a single entry of:
F: arch/*/perf/
would serve the same purpose.
Nominally, the difference between the 2 entries vs the 1 entry is this:
F: arch/*/perf/*
Only the specific files in any directory that matches this pattern but not any files in their subdirectories are maintained.
F: arch/*/perf/*/*
Only the files in any top level subdirectory of any directory that matches this pattern are maintained but not files in any directory of those subdirectories.
F: arch/*/perf/
Any file or any file in any subdirectory of any directory that matches this pattern is maintained.
Re: [PATCH v11 0/8] Disable compat cruft on ppc64le v11
In reply to Michal SuchanekYou sent it twice ? Any difference between the two dispatch ?
Christophe
Le 19/03/2020 à 13:19, Michal Suchanek a écrit :
Quoted 66 lines
Less code means less bugs so add a knob to skip the compat stuff.
Changes in v2: saner
CONFIG_COMPATifdefs
Changes in v3:
- change llseek to 32bit instead of builing it unconditionally in fs
- clanup the makefile conditionals
- remove some ifdefs or convert toIS_DEFINEDwhere possible
Changes in v4:
- cleanupis_32bit_taskandcurrent_is_64bit
- more makefile cleanup
Changes in v5:
- morecurrent_is_64bitcleanup
- split offcallchain.c32bit and 64bit parts
Changes in v6:
- cleanup makefile after split
- consolidateread_user_stack_32
- fix some checkpatch warnings
Changes in v7:
- add back__ARCH_WANT_SYS_LLSEEKto fix build with llseek
- remove leftover hunk
- add review tags
Changes in v8:
- consolidatevalid_user_spto fix it in the splitcallchain.c
- fix build errors/warnings with PPC64 !COMPAT and PPC32
Changes in v9:
- removecurrent_is_64bit()
Chanegs in v10:
- rebase, sent together with the syscall cleanup
Changes in v11:
- rebase
- addMAINTAINERSpattern for ppc perfMichal Suchanek (8):
powerpc: Add back__ARCH_WANT_SYS_LLSEEKmacro
powerpc: move common register copy functions fromsignal_32.cto
signal.c
powerpc/perf: consolidateread_user_stack_32powerpc/perf: consolidatevalid_user_sp
powerpc/64: make buildable withoutCONFIG_COMPATpowerpc/64: Make COMPAT user-selectable disabled on littleendian by
default.
powerpc/perf: splitcallchain.cby bitness
MAINTAINERS: perf: Add pattern that matches ppc perf to the perf
entry.MAINTAINERS | 2 + arch/powerpc/Kconfig | 5 +- arch/powerpc/include/asm/thread_info.h | 4 +- arch/powerpc/include/asm/unistd.h | 1 + arch/powerpc/kernel/Makefile | 6 +- arch/powerpc/kernel/entry_64.S | 2 + arch/powerpc/kernel/signal.c | 144 +++++++++- arch/powerpc/kernel/signal_32.c | 140 ---------- arch/powerpc/kernel/syscall_64.c | 6 +- arch/powerpc/kernel/vdso.c | 3 +- arch/powerpc/perf/Makefile | 5 +- arch/powerpc/perf/callchain.c | 356 +------------------------ arch/powerpc/perf/callchain.h | 20 ++ arch/powerpc/perf/callchain_32.c | 196 ++++++++++++++ arch/powerpc/perf/callchain_64.c | 174 ++++++++++++ fs/read_write.c | 3 +- 16 files changed, 556 insertions(+), 511 deletions(-)create mode 100644
arch/powerpc/perf/callchain.hcreate mode 100644arch/powerpc/perf/callchain_32.ccreate mode 100644arch/powerpc/perf/callchain_64.c
Re: [PATCH v11 0/8] Disable compat cruft on ppc64le v11
Quoted 1 line · Christophe Leroy
You sent it twice ? Any difference between the two dispatch ?
Some headers were broken the first time around.
Thanks
Michal
Quoted 70 lines
Christophe
Le 19/03/2020 à 13:19, Michal Suchanek a écrit :
Quoted 66 lines
Less code means less bugs so add a knob to skip the compat stuff.
Changes in v2: saner
CONFIG_COMPATifdefs
Changes in v3:
- change llseek to 32bit instead of builing it unconditionally in fs
- clanup the makefile conditionals
- remove some ifdefs or convert toIS_DEFINEDwhere possible
Changes in v4:
- cleanupis_32bit_taskandcurrent_is_64bit
- more makefile cleanup
Changes in v5:
- morecurrent_is_64bitcleanup
- split offcallchain.c32bit and 64bit parts
Changes in v6:
- cleanup makefile after split
- consolidateread_user_stack_32
- fix some checkpatch warnings
Changes in v7:
- add back__ARCH_WANT_SYS_LLSEEKto fix build with llseek
- remove leftover hunk
- add review tags
Changes in v8:
- consolidatevalid_user_spto fix it in the splitcallchain.c
- fix build errors/warnings with PPC64 !COMPAT and PPC32
Changes in v9:
- removecurrent_is_64bit()
Chanegs in v10:
- rebase, sent together with the syscall cleanup
Changes in v11:
- rebase
- addMAINTAINERSpattern for ppc perfMichal Suchanek (8):
powerpc: Add back__ARCH_WANT_SYS_LLSEEKmacro
powerpc: move common register copy functions fromsignal_32.cto
signal.c
powerpc/perf: consolidateread_user_stack_32powerpc/perf: consolidatevalid_user_sp
powerpc/64: make buildable withoutCONFIG_COMPATpowerpc/64: Make COMPAT user-selectable disabled on littleendian by
default.
powerpc/perf: splitcallchain.cby bitness
MAINTAINERS: perf: Add pattern that matches ppc perf to the perf
entry.MAINTAINERS | 2 + arch/powerpc/Kconfig | 5 +- arch/powerpc/include/asm/thread_info.h | 4 +- arch/powerpc/include/asm/unistd.h | 1 + arch/powerpc/kernel/Makefile | 6 +- arch/powerpc/kernel/entry_64.S | 2 + arch/powerpc/kernel/signal.c | 144 +++++++++- arch/powerpc/kernel/signal_32.c | 140 ---------- arch/powerpc/kernel/syscall_64.c | 6 +- arch/powerpc/kernel/vdso.c | 3 +- arch/powerpc/perf/Makefile | 5 +- arch/powerpc/perf/callchain.c | 356 +------------------------ arch/powerpc/perf/callchain.h | 20 ++ arch/powerpc/perf/callchain_32.c | 196 ++++++++++++++ arch/powerpc/perf/callchain_64.c | 174 ++++++++++++ fs/read_write.c | 3 +- 16 files changed, 556 insertions(+), 511 deletions(-)create mode 100644
arch/powerpc/perf/callchain.hcreate mode 100644arch/powerpc/perf/callchain_32.ccreate mode 100644arch/powerpc/perf/callchain_64.c
Nicholas Piggin
reviewerRe: [PATCH v11 0/8] Disable compat cruft on ppc64le v11
In reply to Michal SuchanekMichal Suchanek's on March 19, 2020 10:19 pm:
Quoted 23 lines
Less code means less bugs so add a knob to skip the compat stuff.
Changes in v2: saner
CONFIG_COMPATifdefs
Changes in v3:
- change llseek to 32bit instead of builing it unconditionally in fs
- clanup the makefile conditionals
- remove some ifdefs or convert toIS_DEFINEDwhere possible
Changes in v4:
- cleanupis_32bit_taskandcurrent_is_64bit
- more makefile cleanup
Changes in v5:
- morecurrent_is_64bitcleanup
- split offcallchain.c32bit and 64bit parts
Changes in v6:
- cleanup makefile after split
- consolidateread_user_stack_32
- fix some checkpatch warnings
Changes in v7:
- add back__ARCH_WANT_SYS_LLSEEKto fix build with llseek
- remove leftover hunk
- add review tags
Changes in v8:
- consolidatevalid_user_spto fix it in the splitcallchain.c
Quoted 8 lines
- fix build errors/warnings with PPC64 !COMPAT and PPC32
Changes in v9:
- removecurrent_is_64bit()
Chanegs in v10:
- rebase, sent together with the syscall cleanup
Changes in v11:
- rebase
- addMAINTAINERSpattern for ppc perf
These all look good to me. I had some minor comment about one patch but not really a big deal and there were more cleanups on top of it, so I don't mind if it's merged as is.
Actually I think we have a bit of stack reading fixes for 64s radix now (not a bug fix as such, but we don't need the hash fault logic in radix), so if I get around to that I can propose the changes in that series.
Thanks,
Nick
Re: [PATCH v11 0/8] Disable compat cruft on ppc64le v11
Le 03/04/2020 à 09:25, Nicholas Piggin a écrit :
Quoted 33 lines
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 23 lines
Less code means less bugs so add a knob to skip the compat stuff.
Changes in v2: saner
CONFIG_COMPATifdefs
Changes in v3:
- change llseek to 32bit instead of builing it unconditionally in fs
- clanup the makefile conditionals
- remove some ifdefs or convert toIS_DEFINEDwhere possible
Changes in v4:
- cleanupis_32bit_taskandcurrent_is_64bit
- more makefile cleanup
Changes in v5:
- morecurrent_is_64bitcleanup
- split offcallchain.c32bit and 64bit parts
Changes in v6:
- cleanup makefile after split
- consolidateread_user_stack_32
- fix some checkpatch warnings
Changes in v7:
- add back__ARCH_WANT_SYS_LLSEEKto fix build with llseek
- remove leftover hunk
- add review tags
Changes in v8:
- consolidatevalid_user_spto fix it in the splitcallchain.cQuoted 8 lines
- fix build errors/warnings with PPC64 !COMPAT and PPC32
Changes in v9:
- removecurrent_is_64bit()
Chanegs in v10:
- rebase, sent together with the syscall cleanup
Changes in v11:
- rebase
- addMAINTAINERSpattern for ppc perf
Quoted 8 lines
These all look good to me. I had some minor comment about one patch but not really a big deal and there were more cleanups on top of it, so I don't mind if it's merged as is.
Actually I think we have a bit of stack reading fixes for 64s radix now (not a bug fix as such, but we don't need the hash fault logic in radix), so if I get around to that I can propose the changes in that series.
As far as I can see, there is a v12
Christophe
Nicholas Piggin
reviewerRe: [PATCH v11 0/8] Disable compat cruft on ppc64le v11
Christophe Leroy's on April 3, 2020 5:26 pm:
Quoted 40 lines
Le 03/04/2020 à 09:25, Nicholas Piggin a écrit :
Quoted 37 lines
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 23 lines
Less code means less bugs so add a knob to skip the compat stuff.
Changes in v2: saner
CONFIG_COMPATifdefs
Changes in v3:
- change llseek to 32bit instead of builing it unconditionally in fs
- clanup the makefile conditionals
- remove some ifdefs or convert toIS_DEFINEDwhere possible
Changes in v4:
- cleanupis_32bit_taskandcurrent_is_64bit
- more makefile cleanup
Changes in v5:
- morecurrent_is_64bitcleanup
- split offcallchain.c32bit and 64bit parts
Changes in v6:
- cleanup makefile after split
- consolidateread_user_stack_32
- fix some checkpatch warnings
Changes in v7:
- add back__ARCH_WANT_SYS_LLSEEKto fix build with llseek
- remove leftover hunk
- add review tags
Changes in v8:
- consolidatevalid_user_spto fix it in the splitcallchain.cQuoted 8 lines
- fix build errors/warnings with PPC64 !COMPAT and PPC32
Changes in v9:
- removecurrent_is_64bit()
Chanegs in v10:
- rebase, sent together with the syscall cleanup
Changes in v11:
- rebase
- addMAINTAINERSpattern for ppc perfThese all look good to me. I had some minor comment about one patch but not really a big deal and there were more cleanups on top of it, so I don't mind if it's merged as is.
Quoted 6 lines
Quoted 4 lines
Actually I think we have a bit of stack reading fixes for 64s radix now (not a bug fix as such, but we don't need the hash fault logic in radix), so if I get around to that I can propose the changes in that series.
As far as I can see, there is a v12
For the most part I was looking at the patches in mpe's next-test tree on github, if that's the v12 series, same comment applies but it's a pretty small nitpick.
Thanks,
Nick
Michael Ellerman
reviewerRe: [PATCH v11 0/8] Disable compat cruft on ppc64le v11
Quoted 44 lines · Nicholas Piggin
Christophe Leroy's on April 3, 2020 5:26 pm:
Quoted 43 lines
Le 03/04/2020 à 09:25, Nicholas Piggin a écrit :
Quoted 41 lines
Michal Suchanek's on March 19, 2020 10:19 pm:
Quoted 23 lines
Less code means less bugs so add a knob to skip the compat stuff.
Changes in v2: saner
CONFIG_COMPATifdefs
Changes in v3:
- change llseek to 32bit instead of builing it unconditionally in fs
- clanup the makefile conditionals
- remove some ifdefs or convert toIS_DEFINEDwhere possible
Changes in v4:
- cleanupis_32bit_taskandcurrent_is_64bit
- more makefile cleanup
Changes in v5:
- morecurrent_is_64bitcleanup
- split offcallchain.c32bit and 64bit parts
Changes in v6:
- cleanup makefile after split
- consolidateread_user_stack_32
- fix some checkpatch warnings
Changes in v7:
- add back__ARCH_WANT_SYS_LLSEEKto fix build with llseek
- remove leftover hunk
- add review tags
Changes in v8:
- consolidatevalid_user_spto fix it in the splitcallchain.cQuoted 8 lines
- fix build errors/warnings with PPC64 !COMPAT and PPC32
Changes in v9:
- removecurrent_is_64bit()
Chanegs in v10:
- rebase, sent together with the syscall cleanup
Changes in v11:
- rebase
- addMAINTAINERSpattern for ppc perfThese all look good to me. I had some minor comment about one patch but not really a big deal and there were more cleanups on top of it, so I don't mind if it's merged as is.
Actually I think we have a bit of stack reading fixes for 64s radix now (not a bug fix as such, but we don't need the hash fault logic in radix), so if I get around to that I can propose the changes in that series.
Quoted 5 lines
Quoted 1 line
As far as I can see, there is a v12
For the most part I was looking at the patches in mpe's next-test tree on github, if that's the v12 series, same comment applies but it's a pretty small nitpick.
Yeah I have v12 in my tree.
This has floated around long enough (our fault), so I'm going to take it and we can fix anything up later.
cheers
v12 cover Disable compat cruft on ppc64le v12
In reply to Nicholas PigginLess code means less bugs so add a knob to skip the compat stuff.
Changes in v2: saner CONFIG_COMPAT ifdefs
Changes in v3:
- change llseek to 32bit instead of builing it unconditionally in fs
- clanup the makefile conditionals
- remove some ifdefs or convert to IS_DEFINED where possible
Changes in v4:
- cleanup is_32bit_task and current_is_64bit
- more makefile cleanup
Changes in v5:
- more current_is_64bit cleanup
- split off callchain.c 32bit and 64bit parts
Changes in v6:
- cleanup makefile after split
- consolidate read_user_stack_32
- fix some checkpatch warnings
Changes in v7:
- add back __ARCH_WANT_SYS_LLSEEK to fix build with llseek
- remove leftover hunk
- add review tags
Changes in v8:
- consolidate valid_user_sp to fix it in the split callchain.c
- fix build errors/warnings with PPC64 !COMPAT and PPC32
Changes in v9:
- remove current_is_64bit()
Chanegs in v10:
- rebase, sent together with the syscall cleanup
Changes in v11:
- rebase
- add MAINTAINERS pattern for ppc perf
Changes in v12:
- simplify valid_user_sp and change to invalid_user_sp
- remove superfluous perf patterns in MAINTAINERS
Michal Suchanek (8):
powerpc: Add back __ARCH_WANT_SYS_LLSEEK macro
powerpc: move common register copy functions from signal_32.c to
signal.c
powerpc/perf: consolidate read_user_stack_32 powerpc/perf: consolidate valid_user_sp -> invalid_user_sp powerpc/64: make buildable without CONFIG_COMPAT powerpc/64: Make COMPAT user-selectable disabled on littleendian by
default.
powerpc/perf: split callchain.c by bitness
MAINTAINERS: perf: Add pattern that matches ppc perf to the perf
entry.
MAINTAINERS | 6 +- arch/powerpc/Kconfig | 5 +- arch/powerpc/include/asm/thread_info.h | 4 +- arch/powerpc/include/asm/unistd.h | 1 + arch/powerpc/kernel/Makefile | 6 +- arch/powerpc/kernel/entry_64.S | 2 + arch/powerpc/kernel/signal.c | 144 +++++++++- arch/powerpc/kernel/signal_32.c | 140 ---------- arch/powerpc/kernel/syscall_64.c | 6 +- arch/powerpc/kernel/vdso.c | 3 +- arch/powerpc/perf/Makefile | 5 +- arch/powerpc/perf/callchain.c | 356 +------------------------ arch/powerpc/perf/callchain.h | 19 ++ arch/powerpc/perf/callchain_32.c | 196 ++++++++++++++ arch/powerpc/perf/callchain_64.c | 174 ++++++++++++ fs/read_write.c | 3 +- 16 files changed, 556 insertions(+), 514 deletions(-) create mode 100644 arch/powerpc/perf/callchain.h create mode 100644 arch/powerpc/perf/callchain_32.c create mode 100644 arch/powerpc/perf/callchain_64.c
Signature · 1 line
2.23.0
PATCH v12 1/8 powerpc: Add back __ARCH_WANT_SYS_LLSEEK macro
This partially reverts commit caf6f9c8a326 ("asm-generic: Remove unneeded __ARCH_WANT_SYS_LLSEEK macro")
When CONFIG_COMPAT is disabled on ppc64 the kernel does not build.
There is resistance to both removing the llseek syscall from the 64bit syscall tables and building the llseek interface unconditionally.
---
v7: new patch
--- arch/powerpc/include/asm/unistd.h | 1 + fs/read_write.c | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-)
arch/powerpc/include/asm/unistd.h +1 -0
@@ -31,6 +31,7 @@#define __ARCH_WANT_SYS_SOCKETCALL#define __ARCH_WANT_SYS_FADVISE64#define __ARCH_WANT_SYS_GETPGRP +#define __ARCH_WANT_SYS_LLSEEK#define __ARCH_WANT_SYS_NICE#define __ARCH_WANT_SYS_OLD_GETRLIMIT#define __ARCH_WANT_SYS_OLD_UNAME
fs/read_write.c +2 -1
@@ -331,7 +331,8 @@ COMPAT_SYSCALL_DEFINE3(lseek, unsigned int, fd, compat_off_t, offset, unsigned i }#endif -#if !defined(CONFIG_64BIT ) || defined(CONFIG_COMPAT ) +#if !defined(CONFIG_64BIT ) || defined(CONFIG_COMPAT ) || \ + defined(__ARCH_WANT_SYS_LLSEEK) SYSCALL_DEFINE5(llseek, unsigned int, fd, unsigned long, offset_high, unsigned long, offset_low, loff_t __user *, result, unsigned int, whence)
Signature · 1 line
2.23.0
Re: [PATCH v12 1/8] powerpc: Add back __ARCH_WANT_SYS_LLSEEK macro
Quoted 5 lines · Michal Suchanek
This partially reverts commit
caf6f9c8a326("asm-generic: Remove unneeded__ARCH_WANT_SYS_LLSEEKmacro")When
CONFIG_COMPATis disabled on ppc64 the kernel does not build.
Quoted 8 lines
There is resistance to both removing the llseek syscall from the 64bit syscall tables and building the llseek interface unconditionally.
Signed-off-by Michal Suchanek <[email protected]>Reviewed-by Arnd Bergmann <[email protected]>
Patches 1-7 applied to powerpc next, thanks.
https://git.kernel.org/powerpc/c/9e62ccec3ba0a17c8050ea78500dfdd0e4c5c0cc
cheers
PATCH v12 2/8 powerpc: move common register copy functions from signal_32.c to signal.c
In reply to Michal SuchanekThese functions are required for 64bit as well.
--- arch/powerpc/kernel/signal.c | 141 ++++++++++++++++++++++++++++++++ arch/powerpc/kernel/signal_32.c | 140 ------------------------------- 2 files changed, 141 insertions(+), 140 deletions(-)
arch/powerpc/kernel/signal.c +141 -0
@@ -18,12 +18,153 @@#include <linux/syscalls.h> #include <asm/hw_breakpoint.h> #include <linux/uaccess.h> +#include <asm/switch_to.h> #include <asm/unistd.h> #include <asm/debug.h> #include <asm/tm.h> #include "signal.h" +#ifdef CONFIG_VSX +unsigned long copy_fpr_to_user (void __user *to, +struct task_struct *task) +{ +u64 buf[ELF_NFPREG ]; +int i; + +/* save FPR copy to local buffer then write to the thread_struct */ +for (i =0 ; i < (ELF_NFPREG -1 ) ; i++) + buf[i] = task->thread.TS_FPR (i); + buf[i] = task->thread.fp_state.fpscr; +return __copy_to_user (to, buf,ELF_NFPREG *sizeof (double )); +} + +unsigned long copy_fpr_from_user (struct task_struct *task, +void __user *from) +{ +u64 buf[ELF_NFPREG ]; +int i; + +if (__copy_from_user (buf, from,ELF_NFPREG *sizeof (double ))) +return 1 ; +for (i =0 ; i < (ELF_NFPREG -1 ) ; i++) + task->thread.TS_FPR (i) = buf[i]; + task->thread.fp_state.fpscr = buf[i]; + +return 0 ; +} + +unsigned long copy_vsx_to_user (void __user *to, +struct task_struct *task) +{ +u64 buf[ELF_NVSRHALFREG ]; +int i; + +/* save FPR copy to local buffer then write to the thread_struct */ +for (i =0 ; i <ELF_NVSRHALFREG ; i++) + buf[i] = task->thread.fp_state.fpr[i][TS_VSRLOWOFFSET ]; +return __copy_to_user (to, buf,ELF_NVSRHALFREG *sizeof (double )); +} + +unsigned long copy_vsx_from_user (struct task_struct *task, +void __user *from) +{ +u64 buf[ELF_NVSRHALFREG ]; +int i; + +if (__copy_from_user (buf, from,ELF_NVSRHALFREG *sizeof (double ))) +return 1 ; +for (i =0 ; i <ELF_NVSRHALFREG ; i++) + task->thread.fp_state.fpr[i][TS_VSRLOWOFFSET ] = buf[i]; +return 0 ; +} + +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM +unsigned long copy_ckfpr_to_user (void __user *to, +struct task_struct *task) +{ +u64 buf[ELF_NFPREG ]; +int i; + +/* save FPR copy to local buffer then write to the thread_struct */ +for (i =0 ; i < (ELF_NFPREG -1 ) ; i++) + buf[i] = task->thread.TS_CKFPR (i); + buf[i] = task->thread.ckfp_state.fpscr; +return __copy_to_user (to, buf,ELF_NFPREG *sizeof (double )); +} + +unsigned long copy_ckfpr_from_user (struct task_struct *task, +void __user *from) +{ +u64 buf[ELF_NFPREG ]; +int i; + +if (__copy_from_user (buf, from,ELF_NFPREG *sizeof (double ))) +return 1 ; +for (i =0 ; i < (ELF_NFPREG -1 ) ; i++) + task->thread.TS_CKFPR (i) = buf[i]; + task->thread.ckfp_state.fpscr = buf[i]; + +return 0 ; +} + +unsigned long copy_ckvsx_to_user (void __user *to, +struct task_struct *task) +{ +u64 buf[ELF_NVSRHALFREG ]; +int i; + +/* save FPR copy to local buffer then write to the thread_struct */ +for (i =0 ; i <ELF_NVSRHALFREG ; i++) + buf[i] = task->thread.ckfp_state.fpr[i][TS_VSRLOWOFFSET ]; +return __copy_to_user (to, buf,ELF_NVSRHALFREG *sizeof (double )); +} + +unsigned long copy_ckvsx_from_user (struct task_struct *task, +void __user *from) +{ +u64 buf[ELF_NVSRHALFREG ]; +int i; + +if (__copy_from_user (buf, from,ELF_NVSRHALFREG *sizeof (double ))) +return 1 ; +for (i =0 ; i <ELF_NVSRHALFREG ; i++) + task->thread.ckfp_state.fpr[i][TS_VSRLOWOFFSET ] = buf[i]; +return 0 ; +} +#endif /* CONFIG_PPC_TRANSACTIONAL_MEM */ +#else +inline unsigned long copy_fpr_to_user (void __user *to, +struct task_struct *task) +{ +return __copy_to_user (to, task->thread.fp_state.fpr, +ELF_NFPREG *sizeof (double )); +} + +inline unsigned long copy_fpr_from_user (struct task_struct *task, +void __user *from) +{ +return __copy_from_user (task->thread.fp_state.fpr, from, +ELF_NFPREG *sizeof (double )); +} + +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM +inline unsigned long copy_ckfpr_to_user (void __user *to, +struct task_struct *task) +{ +return __copy_to_user (to, task->thread.ckfp_state.fpr, +ELF_NFPREG *sizeof (double )); +} + +inline unsigned long copy_ckfpr_from_user (struct task_struct *task, +void __user *from) +{ +return __copy_from_user (task->thread.ckfp_state.fpr, from, +ELF_NFPREG *sizeof (double )); +} +#endif /* CONFIG_PPC_TRANSACTIONAL_MEM */ +#endif +/* Log an error when sending an unhandled signal to a process. Controlled * through debug.exception-trace sysctl. */
arch/powerpc/kernel/signal_32.c +0 -140
146 lines not on this page · raw message
Signature · 1 line
2.23.0
PATCH v12 3/8 powerpc/perf: consolidate read_user_stack_32
In reply to Michal SuchanekThere are two almost identical copies for 32bit and 64bit.
The function is used only in 32bit code which will be split out in next patch so consolidate to one function.
---
v6: new patch v8: move the consolidated function out of the ifdef block. v11: rebase on top of def0bfdbd603 --- arch/powerpc/perf/callchain.c | 48 +++++++++++++++++------------------ 1 file changed, 24 insertions(+), 24 deletions(-)
arch/powerpc/perf/callchain.c +24 -24
@@ -161,18 +161,6 @@ static int read_user_stack_64(unsigned long __user *ptr, unsigned long *ret)return read_user_stack_slow (ptr, ret,8 ); } -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) -{ -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -if (!probe_user_read (ret, ptr,sizeof (*ret))) -return 0 ; - -return read_user_stack_slow (ptr, ret,4 ); -} -static inline int valid_user_sp (unsigned long sp,int is_64) {if (!sp || (sp &7 ) || sp > (is_64 ?TASK_SIZE :0x100000000UL ) -32 )
@@ -277,19 +265,9 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry, }#else /* CONFIG_PPC64 */ -/* -* On 32-bit we just access the address and let hash_page create a -* HPTE if necessary, so there is no need to fall back to reading -* the page tables. Since this is called at interrupt level, -* do_page_fault() won't treat a DSI as a page fault. -*/ -static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +static int read_user_stack_slow (void __user *ptr,void *buf,int nb) { -if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || - ((unsigned long )ptr &3 )) -return -EFAULT ; - -return probe_user_read (ret, ptr,sizeof (*ret)); +return 0 ; } static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,
@@ -312,6 +290,28 @@ static inline int valid_user_sp(unsigned long sp, int is_64)#endif /* CONFIG_PPC64 */ +/* +* On 32-bit we just access the address and let hash_page create a +* HPTE if necessary, so there is no need to fall back to reading +* the page tables. Since this is called at interrupt level, +* do_page_fault() won't treat a DSI as a page fault. +*/ +static int read_user_stack_32 (unsigned int __user *ptr,unsigned int *ret) +{ +int rc; + +if ((unsigned long )ptr >TASK_SIZE -sizeof (unsigned int ) || + ((unsigned long )ptr &3 )) +return -EFAULT ; + + rc =probe_user_read (ret, ptr,sizeof (*ret)); + +if (IS_ENABLED (CONFIG_PPC64 ) && rc) +return read_user_stack_slow (ptr, ret,4 ); + +return rc; +} +/* * Layout for non-RT signal frames */
Signature · 1 line
2.23.0
PATCH v12 4/8 powerpc/perf: consolidate valid_user_sp -> invalid_user_sp
In reply to Michal SuchanekMerge the 32bit and 64bit version.
Halve the check constants on 32bit.
Use STACK_TOP since it is defined.
Passing is_64 is now redundant since is_32bit_task() is used to determine which callchain variant should be used. Use STACK_TOP and is_32bit_task() directly.
This removes a page from the valid 32bit area on 64bit:
#define TASK_SIZE_USER32 (0x0000000100000000UL - (1 * PAGE_SIZE)) #define STACK_TOP_USER32 TASK_SIZE_USER32
Change return value to bool. It is inverted by users anyway.
Change to invalid_user_sp to avoid inverting the return value twice.
---
v8: new patch
v11: simplify by using is_32bit_task()
v12:
- simplify by precalculating subexpresions
- change return value to bool
- remove double inversion
--- arch/powerpc/perf/callchain.c | 26 ++++++++++---------------- 1 file changed, 10 insertions(+), 16 deletions(-)
arch/powerpc/perf/callchain.c +10 -16
@@ -102,6 +102,14 @@ perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *re } } +static inline bool invalid_user_sp (unsigned long sp) +{ +unsigned long mask =is_32bit_task () ?3 :7 ; +unsigned long top =STACK_TOP - (is_32bit_task () ?16 :32 ); + +return (!sp || (sp & mask) || (sp > top)); +} +#ifdef CONFIG_PPC64 /* * On 64-bit we don't want to invoke hash_page on user addresses from
@@ -161,13 +169,6 @@ static int read_user_stack_64(unsigned long __user *ptr, unsigned long *ret)return read_user_stack_slow (ptr, ret,8 ); } -static inline int valid_user_sp (unsigned long sp,int is_64) -{ -if (!sp || (sp &7 ) || sp > (is_64 ?TASK_SIZE :0x100000000UL ) -32 ) -return 0 ; -return 1 ; -} -/* * 64-bit user processes use the same stack frame for RT and non-RT signals. */
@@ -226,7 +227,7 @@ static void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry,while (entry->nr < entry->max_stack) { fp = (unsigned long __user *) sp; -if (!valid_user_sp (sp,1 ) ||read_user_stack_64 (fp, &next_sp)) +if (invalid_user_sp (sp) ||read_user_stack_64 (fp, &next_sp))return ;if (level >0 &&read_user_stack_64 (&fp[2 ], &next_ip))return ;
@@ -275,13 +276,6 @@ static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry { } -static inline int valid_user_sp (unsigned long sp,int is_64) -{ -if (!sp || (sp &7 ) || sp >TASK_SIZE -32 ) -return 0 ; -return 1 ; -} -#define __SIGNAL_FRAMESIZE32 __SIGNAL_FRAMESIZE#define sigcontext32 sigcontext#define mcontext32 mcontext
@@ -423,7 +417,7 @@ static void perf_callchain_user_32(struct perf_callchain_entry_ctx *entry,while (entry->nr < entry->max_stack) { fp = (unsigned int __user *) (unsigned long ) sp; -if (!valid_user_sp (sp,0 ) ||read_user_stack_32 (fp, &next_sp)) +if (invalid_user_sp (sp) ||read_user_stack_32 (fp, &next_sp))return ;if (level >0 &&read_user_stack_32 (&fp[1 ], &next_ip))return ;
Signature · 1 line
2.23.0
PATCH v12 5/8 powerpc/64: make buildable without CONFIG_COMPAT
In reply to Michal SuchanekThere are numerous references to 32bit functions in generic and 64bit code so ifdef them out.
---
v2:
- fix 32bit ifdef condition in signal.c
- simplify the compat ifdef condition in vdso.c - 64bit is redundant
- simplify the compat ifdef condition in callchain.c - 64bit is redundant v3:
- use IS_ENABLED and maybe_unused where possible
- do not ifdef declarations
- clean up Makefile
v4:
- further makefile cleanup
- simplify is_32bit_task conditions
- avoid ifdef in condition by using return
v5:
- avoid unreachable code on 32bit
- make is_current_64bit constant on !COMPAT
- add stub perf_callchain_user_32 to avoid some ifdefs
v6:
- consolidate current_is_64bit
v7:
- remove leftover perf_callchain_user_32 stub from previous series version v8:
- fix build again - too trigger-happy with stub removal
- remove a vdso.c hunk that causes warning according to kbuild test robot v9:
- removed current_is_64bit in previous patch
v10:
- rebase on top of 70ed86f4de5bd
--- arch/powerpc/include/asm/thread_info.h | 4 ++-- arch/powerpc/kernel/Makefile | 6 +++--- arch/powerpc/kernel/entry_64.S | 2 ++ arch/powerpc/kernel/signal.c | 3 +-- arch/powerpc/kernel/syscall_64.c | 6 ++---- arch/powerpc/kernel/vdso.c | 3 ++- arch/powerpc/perf/callchain.c | 8 +++++++- 7 files changed, 19 insertions(+), 13 deletions(-)
arch/powerpc/include/asm/thread_info.h +2 -2
@@ -162,10 +162,10 @@ static inline bool test_thread_local_flags(unsigned int flags)return (ti->local_flags & flags) !=0 ; } -#ifdef CONFIG_PPC64 +#ifdef CONFIG_COMPAT #define is_32bit_task () (test_thread_flag(TIF_32BIT))#else -#define is_32bit_task () (1) +#define is_32bit_task () (IS_ENABLED(CONFIG_PPC32))#endif #if defined(CONFIG_PPC64 )
arch/powerpc/kernel/Makefile +3 -3
@@ -42,16 +42,16 @@ CFLAGS_btext.o += -DDISABLE_BRANCH_PROFILING endif obj-y := cputable.o ptrace.o syscalls.o \ - irq.o align.o signal_32.o pmc.o vdso.o \ + irq.o align.o signal_$(BITS).o pmc.o vdso.o \ process.o systbl.o idle.o \ signal.o sysfs.o cacheinfo.o time.o \ prom.o traps.o setup-common.o \ udbg.o misc.o io.o misc_$(BITS).o \ of_platform.o prom_parse.o -obj-$(CONFIG_PPC64) += setup_64.o sys_ppc32.o \ - signal_64.o ptrace32.o \ +obj-$(CONFIG_PPC64) += setup_64.o \ paca.o nvram_64.o firmware.o note.o \ syscall_64.o +obj-$(CONFIG_COMPAT) += sys_ppc32.o ptrace32.o signal_32.o obj-$(CONFIG_VDSO32) += vdso32/ obj-$(CONFIG_PPC_WATCHDOG) += watchdog.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o
arch/powerpc/kernel/entry_64.S +2 -0
@@ -52,8 +52,10 @@ SYS_CALL_TABLE: .tc sys_call_table[TC],sys_call_table +#ifdef CONFIG_COMPAT COMPAT_SYS_CALL_TABLE: .tc compat_sys_call_table[TC],compat_sys_call_table +#endif /* This value is used to mark exception frames on the stack. */ exception_marker:
arch/powerpc/kernel/signal.c +1 -2
@@ -247,7 +247,6 @@ static void do_signal(struct task_struct *tsk)sigset_t *oldset =sigmask_to_save ();struct ksignal ksig = { .sig =0 };int ret; -int is32 =is_32bit_task ();BUG_ON (tsk != current);
@@ -277,7 +276,7 @@ static void do_signal(struct task_struct *tsk)rseq_signal_deliver (&ksig, tsk->thread.regs); -if (is32) { +if (is_32bit_task ()) {if (ksig.ka.sa.sa_flags &SA_SIGINFO ) ret =handle_rt_signal32 (&ksig, oldset, tsk);else
arch/powerpc/kernel/syscall_64.c +2 -4
@@ -24,7 +24,6 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7, long r8, unsigned long r0, struct pt_regs *regs) { - unsigned long ti_flags; syscall_fn f;if (IS_ENABLED (CONFIG_PPC_IRQ_SOFT_MASK_DEBUG ))
@@ -68,8 +67,7 @@ notrace long system_call_exception(long r3, long r4, long r5,local_irq_enable (); - ti_flags =current_thread_info ()->flags; -if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) { +if (unlikely (current_thread_info ()->flags & _TIF_SYSCALL_DOTRACE)) {/* * We use the return value of do_syscall_trace_enter() as the * syscall number. If the syscall was rejected for any reason
@@ -94,7 +92,7 @@ notrace long system_call_exception(long r3, long r4, long r5,/* May be faster to do array_index_nospec? */ barrier_nospec (); -if (unlikely (ti_flags & _TIF_32BIT)) { +if (unlikely (is_32bit_task ())) { f = (void *)compat_sys_call_table[r0]; r3 &=0x00000000ffffffffULL ;
arch/powerpc/kernel/vdso.c +2 -1
@@ -656,7 +656,8 @@ static void __init vdso_setup_syscall_map(void)if (sys_call_table[i] != sys_ni_syscall) vdso_data->syscall_map_64[i >>5 ] |=0x80000000UL >> (i &0x1f ); -if (compat_sys_call_table[i] != sys_ni_syscall) +if (IS_ENABLED (CONFIG_COMPAT ) && + compat_sys_call_table[i] != sys_ni_syscall) vdso_data->syscall_map_32[i >>5 ] |=0x80000000UL >> (i &0x1f );#else /* CONFIG_PPC64 */
arch/powerpc/perf/callchain.c +7 -1
@@ -15,7 +15,7 @@#include <asm/sigcontext.h> #include <asm/ucontext.h> #include <asm/vdso.h> -#ifdef CONFIG_PPC64 +#ifdef CONFIG_COMPAT #include "../kernel/ppc32.h" #endif #include <asm/pte-walk.h>
@@ -284,6 +284,7 @@ static inline void perf_callchain_user_64(struct perf_callchain_entry_ctx *entry#endif /* CONFIG_PPC64 */ +#if defined(CONFIG_PPC32 ) || defined(CONFIG_COMPAT )/* * On 32-bit we just access the address and let hash_page create a * HPTE if necessary, so there is no need to fall back to reading
@@ -447,6 +448,11 @@ static void perf_callchain_user_32(struct perf_callchain_entry_ctx *entry, sp = next_sp; } } +#else /* 32bit */ +static void perf_callchain_user_32 (struct perf_callchain_entry_ctx *entry, +struct pt_regs *regs) +{} +#endif /* 32bit */ void perf_callchain_user (struct perf_callchain_entry_ctx *entry,struct pt_regs *regs)
Signature · 1 line
2.23.0
Re: [PATCH v12 5/8] powerpc/64: make buildable without CONFIG_COMPAT
Le 20/03/2020 à 11:20, Michal Suchanek a écrit :
Quoted 42 lines
There are numerous references to 32bit functions in generic and 64bit code so ifdef them out.
Signed-off-by Michal Suchanek <[email protected]>---
v2:
- fix 32bit ifdef condition insignal.c
- simplify the compat ifdef condition invdso.c- 64bit is redundant
- simplify the compat ifdef condition incallchain.c- 64bit is redundant v3:
- useIS_ENABLEDandmaybe_unusedwhere possible
- do not ifdef declarations
- clean upMakefile
v4:
- further makefile cleanup
- simplifyis_32bit_taskconditions
- avoid ifdef in condition by using return
v5:
- avoid unreachable code on 32bit
- makeis_current_64bitconstant on !COMPAT
- add stubperf_callchain_user_32to avoid some ifdefs
v6:
- consolidatecurrent_is_64bit
v7:
- remove leftoverperf_callchain_user_32stub from previous series version v8:
- fix build again - too trigger-happy with stub removal
- remove avdso.chunk that causes warning according to kbuild test robot v9:
- removedcurrent_is_64bitin previous patch
v10:
- rebase on top of70ed86f4de5bd--- arch/powerpc/include/asm/thread_info.h | 4 ++-- arch/powerpc/kernel/Makefile | 6 +++--- arch/powerpc/kernel/entry_64.S | 2 ++ arch/powerpc/kernel/signal.c | 3 +-- arch/powerpc/kernel/syscall_64.c | 6 ++---- arch/powerpc/kernel/vdso.c | 3 ++- arch/powerpc/perf/callchain.c | 8 +++++++- 7 files changed, 19 insertions(+), 13 deletions(-)
[...]
Quoted 23 lines
arch/powerpc/kernel/syscall_64.c +1
-3@@ -24,7 +24,6 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7, long r8, unsigned long r0, struct pt_regs *regs) { - unsigned long ti_flags; syscall_fn f;if (IS_ENABLED (CONFIG_PPC_IRQ_SOFT_MASK_DEBUG ))@@ -68,8 +67,7 @@ notrace long system_call_exception(long r3, long r4, long r5,local_irq_enable (); - ti_flags =current_thread_info ()->flags; -if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) { +if (unlikely (current_thread_info ()->flags & _TIF_SYSCALL_DOTRACE)) {/* * We use the return value of do_syscall_trace_enter() as the * syscall number. If the syscall was rejected for any reason
arch/powerpc/kernel/syscall_64.c notrace long system_call_exception(long r3, long r4, long r5, · line 95
Quoted 5 lines
arch/powerpc/kernel/syscall_64.c +1
-1@@ -94,7 +92,7 @@ notrace long system_call_exception(long r3, long r4, long r5,/* May be faster to do array_index_nospec? */ barrier_nospec (); -if (unlikely (ti_flags & _TIF_32BIT)) { +if (unlikely (is_32bit_task ())) {
is_compat() should be used here instead, because we dont want to use compat_sys_call_table() on PPC32.
arch/powerpc/kernel/syscall_64.c notrace long system_call_exception(long r3, long r4, long r5, · line 98
Quoted 3 lines
arch/powerpc/kernel/syscall_64.c +0
-0f = (void *)compat_sys_call_table[r0]; r3 &=0x00000000ffffffffULL ;
Christophe
Re: [PATCH v12 5/8] powerpc/64: make buildable without CONFIG_COMPAT
Quoted 66 lines · Christophe Leroy
Le 20/03/2020 à 11:20, Michal Suchanek a écrit :
Quoted 42 lines
There are numerous references to 32bit functions in generic and 64bit code so ifdef them out.
Signed-off-by Michal Suchanek <[email protected]>---
v2:
- fix 32bit ifdef condition insignal.c
- simplify the compat ifdef condition invdso.c- 64bit is redundant
- simplify the compat ifdef condition incallchain.c- 64bit is redundant v3:
- useIS_ENABLEDandmaybe_unusedwhere possible
- do not ifdef declarations
- clean upMakefile
v4:
- further makefile cleanup
- simplifyis_32bit_taskconditions
- avoid ifdef in condition by using return
v5:
- avoid unreachable code on 32bit
- makeis_current_64bitconstant on !COMPAT
- add stubperf_callchain_user_32to avoid some ifdefs
v6:
- consolidatecurrent_is_64bit
v7:
- remove leftoverperf_callchain_user_32stub from previous series version v8:
- fix build again - too trigger-happy with stub removal
- remove avdso.chunk that causes warning according to kbuild test robot v9:
- removedcurrent_is_64bitin previous patch
v10:
- rebase on top of70ed86f4de5bd--- arch/powerpc/include/asm/thread_info.h | 4 ++-- arch/powerpc/kernel/Makefile | 6 +++--- arch/powerpc/kernel/entry_64.S | 2 ++ arch/powerpc/kernel/signal.c | 3 +-- arch/powerpc/kernel/syscall_64.c | 6 ++---- arch/powerpc/kernel/vdso.c | 3 ++- arch/powerpc/perf/callchain.c | 8 +++++++- 7 files changed, 19 insertions(+), 13 deletions(-)[...]
Quoted 18 lines
arch/powerpc/kernel/syscall_64.c +1
-3@@ -24,7 +24,6 @@ notrace long system_call_exception(long r3, long r4, long r5, long r6, long r7, long r8, unsigned long r0, struct pt_regs *regs) { - unsigned long ti_flags; syscall_fn f;if (IS_ENABLED (CONFIG_PPC_IRQ_SOFT_MASK_DEBUG ))@@ -68,8 +67,7 @@ notrace long system_call_exception(long r3, long r4, long r5,local_irq_enable (); - ti_flags =current_thread_info ()->flags; -if (unlikely (ti_flags & _TIF_SYSCALL_DOTRACE)) { +if (unlikely (current_thread_info ()->flags & _TIF_SYSCALL_DOTRACE)) {/* * We use the return value of do_syscall_trace_enter() as the
Quoted 8 lines
arch/powerpc/kernel/syscall_64.c notrace long system_call_exception(long r3, long r4, long r5, · line 94
Quoted 6 lines
arch/powerpc/kernel/syscall_64.c +1
-1* syscall number. If the syscall was rejected for any reason @@ -94,7 +92,7 @@ notrace long system_call_exception(long r3, long r4, long r5,/* May be faster to do array_index_nospec? */ barrier_nospec (); -if (unlikely (ti_flags & _TIF_32BIT)) { +if (unlikely (is_32bit_task ())) {
is_compat()should be used here instead, because we dont want to use
is_compat_task()
Quoted 5 lines
compat_sys_call_table()on PPC32.Quoted 2 lines
arch/powerpc/kernel/syscall_64.c +0
-0f = (void *)compat_sys_call_table[r0]; r3 &=0x00000000ffffffffULL ;
That only applies once you use this for 32bit as well. Right now it's 64bit only so the two are the same.
Thanks
Michal
PATCH v12 6/8 powerpc/64: Make COMPAT user-selectable disabled on littleendian by default.
In reply to Michal SuchanekOn bigendian ppc64 it is common to have 32bit legacy binaries but much less so on littleendian.
---
v3: make configurable
--- arch/powerpc/Kconfig | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
arch/powerpc/Kconfig +3 -2
@@ -264,8 +264,9 @@ config PANIC_TIMEOUT default 180 config COMPAT - bool - default y if PPC64 + bool "Enable support for 32bit binaries" + depends on PPC64 + default y if !CPU_LITTLE_ENDIAN select COMPAT_BINFMT_ELF select ARCH_WANT_OLD_COMPAT_IPC select COMPAT_OLD_SIGACTION
Signature · 1 line
2.23.0
PATCH v12 7/8 powerpc/perf: split callchain.c by bitness
In reply to Michal SuchanekBuilding callchain.c with !COMPAT proved quite ugly with all the defines. Splitting out the 32bit and 64bit parts looks better.
No code change intended.
---
v6:
- move current_is_64bit consolidetaion to earlier patch
- move defines to the top of callchain_32.c
- Makefile cleanup
v8:
- fix valid_user_sp
v11:
- rebase on top of def0bfdbd603
--- arch/powerpc/perf/Makefile | 5 +- arch/powerpc/perf/callchain.c | 356 +------------------------------ arch/powerpc/perf/callchain.h | 19 ++ arch/powerpc/perf/callchain_32.c | 196 +++++++++++++++++ arch/powerpc/perf/callchain_64.c | 174 +++++++++++++++ 5 files changed, 394 insertions(+), 356 deletions(-) create mode 100644 arch/powerpc/perf/callchain.h create mode 100644 arch/powerpc/perf/callchain_32.c create mode 100644 arch/powerpc/perf/callchain_64.c
arch/powerpc/perf/Makefile +4 -1
@@ -1,6 +1,9 @@ # SPDX-License-Identifier: GPL-2.0 -obj-$(CONFIG_PERF_EVENTS) += callchain.o perf_regs.o +obj-$(CONFIG_PERF_EVENTS) += callchain.o callchain_$(BITS).o perf_regs.o +ifdef CONFIG_COMPAT +obj-$(CONFIG_PERF_EVENTS) += callchain_32.o +endif obj-$(CONFIG_PPC_PERF_CTRS) += core-book3s.o bhrb.o obj64-$(CONFIG_PPC_PERF_CTRS) += ppc970-pmu.o power5-pmu.o \
arch/powerpc/perf/callchain.c +1 -355
370 lines not on this page · raw message
arch/powerpc/perf/callchain.h new file +19 -0
@@ -0,0 +1,19 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +#ifndef _POWERPC_PERF_CALLCHAIN_H +#define _POWERPC_PERF_CALLCHAIN_H + +int read_user_stack_slow (void __user *ptr,void *buf,int nb); +void perf_callchain_user_64 (struct perf_callchain_entry_ctx *entry, +struct pt_regs *regs); +void perf_callchain_user_32 (struct perf_callchain_entry_ctx *entry, +struct pt_regs *regs); + +static inline bool invalid_user_sp (unsigned long sp) +{ +unsigned long mask =is_32bit_task () ?3 :7 ; +unsigned long top =STACK_TOP - (is_32bit_task () ?16 :32 ); + +return (!sp || (sp & mask) || (sp > top)); +} + +#endif /* _POWERPC_PERF_CALLCHAIN_H */
arch/powerpc/perf/callchain_32.c new file +196 -0
196 lines not on this page · raw message
arch/powerpc/perf/callchain_64.c new file +174 -0
174 lines not on this page · raw message
Signature · 1 line
2.23.0