Remove 32-bit Xen PV guest support
Subsystems (7)
Juergen Gross
maintainerv2 cover Remove 32-bit Xen PV guest support
The long term plan has been to replace Xen PV guests by PVH. The first victim of that plan are now 32-bit PV guests, as those are used only rather seldom these days. Xen on x86 requires 64-bit support and with Grub2 now supporting PVH officially since version 2.04 there is no need to keep 32-bit PV guest support alive in the Linux kernel. Additionally Meltdown mitigation is not available in the kernel running as 32-bit PV guest, so dropping this mode makes sense from security point of view, too.
Changes in V2:
- rebase to 5.8 kernel
- addressed comments to V1
- new patches 3 and 4
Juergen Gross (4):
x86/xen: remove 32-bit Xen PV guest support x86/paravirt: remove 32-bit support from PARAVIRT_XXL x86/paravirt: cleanup paravirt macros
x86/paravirt: use CONFIG_PARAVIRT_XXL instead of CONFIG_PARAVIRT
arch/x86/entry/entry_32.S | 109 +------ arch/x86/entry/entry_64.S | 4 +- arch/x86/entry/vdso/vdso32/vclock_gettime.c | 1 + arch/x86/include/asm/fixmap.h | 2 +- arch/x86/include/asm/paravirt.h | 107 +----- arch/x86/include/asm/paravirt_types.h | 21 -- arch/x86/include/asm/pgtable-3level_types.h | 5 - arch/x86/include/asm/proto.h | 2 +- arch/x86/include/asm/required-features.h | 2 +- arch/x86/include/asm/segment.h | 6 +- arch/x86/kernel/cpu/common.c | 8 - arch/x86/kernel/head_32.S | 31 -- arch/x86/kernel/kprobes/core.c | 1 - arch/x86/kernel/kprobes/opt.c | 1 - arch/x86/kernel/paravirt.c | 18 -- arch/x86/kernel/paravirt_patch.c | 17 - arch/x86/xen/Kconfig | 3 +- arch/x86/xen/Makefile | 3 +- arch/x86/xen/apic.c | 17 - arch/x86/xen/enlighten_pv.c | 52 +-- arch/x86/xen/mmu_pv.c | 340 +++----------------- arch/x86/xen/p2m.c | 6 +- arch/x86/xen/setup.c | 35 +- arch/x86/xen/smp_pv.c | 18 -- arch/x86/xen/xen-asm.S | 182 ++++++++++- arch/x86/xen/xen-asm_32.S | 185 ----------- arch/x86/xen/xen-asm_64.S | 181 ----------- arch/x86/xen/xen-head.S | 6 - drivers/xen/Kconfig | 4 +- 29 files changed, 232 insertions(+), 1135 deletions(-) delete mode 100644 arch/x86/xen/xen-asm_32.S delete mode 100644 arch/x86/xen/xen-asm_64.S
Signature · 1 line
2.26.2
Juergen Gross
maintainerPATCH v2 1/4 x86/xen: remove 32-bit Xen PV guest support
Xen is requiring 64-bit machines today and since Xen 4.14 it can be built without 32-bit PV guest support. There is no need to carry the burden of 32-bit PV guest support in the kernel any longer, as new guests can be either HVM or PVH, or they can use a 64 bit kernel.
Remove the 32-bit Xen PV support from the kernel.
--- arch/x86/entry/entry_32.S | 109 +---------- arch/x86/include/asm/proto.h | 2 +- arch/x86/include/asm/segment.h | 2 +- arch/x86/kernel/head_32.S | 31 --- arch/x86/xen/Kconfig | 3 +- arch/x86/xen/Makefile | 3 +- arch/x86/xen/apic.c | 17 -- arch/x86/xen/enlighten_pv.c | 48 +---- arch/x86/xen/mmu_pv.c | 340 ++++----------------------------- arch/x86/xen/p2m.c | 6 +- arch/x86/xen/setup.c | 35 +--- arch/x86/xen/smp_pv.c | 18 -- arch/x86/xen/xen-asm.S | 182 ++++++++++++++++-- arch/x86/xen/xen-asm_32.S | 185 ------------------ arch/x86/xen/xen-asm_64.S | 181 ------------------ arch/x86/xen/xen-head.S | 6 - drivers/xen/Kconfig | 4 +- 17 files changed, 216 insertions(+), 956 deletions(-) delete mode 100644 arch/x86/xen/xen-asm_32.S delete mode 100644 arch/x86/xen/xen-asm_64.S
arch/x86/entry/entry_32.S +2 -107
@@ -449,8 +449,6 @@ .macro SWITCH_TO_KERNEL_STACK - ALTERNATIVE "", "jmp .Lend_\@", X86_FEATURE_XENPV - BUG_IF_WRONG_CR3 SWITCH_TO_KERNEL_CR3 scratch_reg=%eax
@@ -599,8 +597,6 @@ */ .macro SWITCH_TO_ENTRY_STACK - ALTERNATIVE "", "jmp .Lend_\@", X86_FEATURE_XENPV - /* Bytes to copy */ movl $PTREGS_SIZE, %ecx
@@ -872,17 +868,6 @@ SYM_ENTRY(__begin_SYSENTER_singlestep_region, SYM_L_GLOBAL, SYM_A_NONE) * will ignore all of the single-step traps generated in this range. */ -#ifdef CONFIG_XEN_PV -/* - * Xen doesn't set %esp to be precisely what the normal SYSENTER - * entry point expects, so fix it up before using the normal path. - */ -SYM_CODE_START(xen_sysenter_target) - addl $5*4, %esp /* remove xen-provided frame */ - jmp .Lsysenter_past_esp -SYM_CODE_END(xen_sysenter_target) -#endif - /* * 32-bit SYSENTER entry. *
@@ -966,9 +951,8 @@ SYM_FUNC_START(entry_SYSENTER_32) movl %esp, %eax call do_fast_syscall_32 - /* XEN PV guests always use IRET path */ - ALTERNATIVE "testl %eax, %eax; jz .Lsyscall_32_done", \ - "jmp .Lsyscall_32_done", X86_FEATURE_XENPV + testl %eax, %eax + jz .Lsyscall_32_done STACKLEAK_ERASE
@@ -1166,95 +1150,6 @@ SYM_FUNC_END(entry_INT80_32) #endif .endm -#ifdef CONFIG_PARAVIRT -SYM_CODE_START(native_iret) - iret - _ASM_EXTABLE(native_iret, asm_iret_error) -SYM_CODE_END(native_iret) -#endif - -#ifdef CONFIG_XEN_PV -/* - * See comment in entry_64.S for further explanation - * - * Note: This is not an actual IDT entry point. It's a XEN specific entry - * point and therefore named to match the 64-bit trampoline counterpart. - */ -SYM_FUNC_START(xen_asm_exc_xen_hypervisor_callback) - /* - * Check to see if we got the event in the critical - * region in xen_iret_direct, after we've reenabled - * events and checked for pending events. This simulates - * iret instruction's behaviour where it delivers a - * pending interrupt when enabling interrupts: - */ - cmpl $xen_iret_start_crit, (%esp) - jb 1f - cmpl $xen_iret_end_crit, (%esp) - jae 1f - call xen_iret_crit_fixup -1: - pushl $-1 /* orig_ax = -1 => not a system call */ - SAVE_ALL - ENCODE_FRAME_POINTER - - mov %esp, %eax - call xen_pv_evtchn_do_upcall - jmp handle_exception_return -SYM_FUNC_END(xen_asm_exc_xen_hypervisor_callback) - -/* - * Hypervisor uses this for application faults while it executes. - * We get here for two reasons: - * 1. Fault while reloading DS, ES, FS or GS - * 2. Fault while executing IRET - * Category 1 we fix up by reattempting the load, and zeroing the segment - * register if the load fails. - * Category 2 we fix up by jumping to do_iret_error. We cannot use the - * normal Linux return path in this case because if we use the IRET hypercall - * to pop the stack frame we end up in an infinite loop of failsafe callbacks. - * We distinguish between categories by maintaining a status value in EAX. - */ -SYM_FUNC_START(xen_failsafe_callback) - pushl %eax - movl $1, %eax -1: mov 4(%esp), %ds -2: mov 8(%esp), %es -3: mov 12(%esp), %fs -4: mov 16(%esp), %gs - /* EAX == 0 => Category 1 (Bad segment) - EAX != 0 => Category 2 (Bad IRET) */ - testl %eax, %eax - popl %eax - lea 16(%esp), %esp - jz 5f - jmp asm_iret_error -5: pushl $-1 /* orig_ax = -1 => not a system call */ - SAVE_ALL - ENCODE_FRAME_POINTER - jmp handle_exception_return - -.section .fixup, "ax" -6: xorl %eax, %eax - movl %eax, 4(%esp) - jmp 1b -7: xorl %eax, %eax - movl %eax, 8(%esp) - jmp 2b -8: xorl %eax, %eax - movl %eax, 12(%esp) - jmp 3b -9: xorl %eax, %eax - movl %eax, 16(%esp) - jmp 4b -.previous - _ASM_EXTABLE(1b, 6b) - _ASM_EXTABLE(2b, 7b) - _ASM_EXTABLE(3b, 8b) - _ASM_EXTABLE(4b, 9b) -SYM_FUNC_END(xen_failsafe_callback) -#endif /* CONFIG_XEN_PV */ - SYM_CODE_START_LOCAL_NOALIGN(handle_exception) /* the function address is in %gs's slot on the stack */ SAVE_ALL switch_stacks=1 skip_gs=1 unwind_espfix=1
arch/x86/include/asm/proto.h +1 -1
@@ -25,7 +25,7 @@ void entry_SYSENTER_compat(void);void __end_entry_SYSENTER_compat (void );void entry_SYSCALL_compat (void );void entry_INT80_compat (void ); -#if defined(CONFIG_X86_64 ) && defined(CONFIG_XEN_PV ) +#ifdef CONFIG_XEN_PV void xen_entry_INT80_compat (void );#endif #endif
arch/x86/include/asm/segment.h +1 -1
@@ -301,7 +301,7 @@ static inline void vdso_read_cpunode(unsigned *cpu, unsigned *node)extern const char early_idt_handler_array[NUM_EXCEPTION_VECTORS ][EARLY_IDT_HANDLER_SIZE ];extern void early_ignore_irq (void ); -#if defined(CONFIG_X86_64 ) && defined(CONFIG_XEN_PV ) +#ifdef CONFIG_XEN_PV extern const char xen_early_idt_handler_array[NUM_EXCEPTION_VECTORS ][XEN_EARLY_IDT_HANDLER_SIZE ];#endif
arch/x86/kernel/head_32.S +0 -31
@@ -134,38 +134,7 @@ SYM_CODE_START(startup_32) movl %eax,pa(initial_page_table+0xffc) #endif -#ifdef CONFIG_PARAVIRT - /* This is can only trip for a broken bootloader... */ - cmpw $0x207, pa(boot_params + BP_version) - jb .Ldefault_entry - - /* Paravirt-compatible boot parameters. Look to see what architecture - we're booting under. */ - movl pa(boot_params + BP_hardware_subarch), %eax - cmpl $num_subarch_entries, %eax - jae .Lbad_subarch - - movl pa(subarch_entries)(,%eax,4), %eax - subl $__PAGE_OFFSET, %eax - jmp *%eax - -.Lbad_subarch: -SYM_INNER_LABEL_ALIGN(xen_entry, SYM_L_WEAK) - /* Unknown implementation; there's really - nothing we can do at this point. */ - ud2a - - __INITDATA - -subarch_entries: - .long .Ldefault_entry /* normal x86/PC */ - .long xen_entry /* Xen hypervisor */ - .long .Ldefault_entry /* Moorestown MID */ -num_subarch_entries = (. - subarch_entries) / 4 -.previous -#else jmp .Ldefault_entry -#endif /* CONFIG_PARAVIRT */ SYM_CODE_END(startup_32) #ifdef CONFIG_HOTPLUG_CPU
arch/x86/xen/Kconfig +2 -1
@@ -19,6 +19,7 @@ config XEN_PV bool "Xen PV guest support" default y depends on XEN + depends on X86_64 select PARAVIRT_XXL select XEN_HAVE_PVMMU select XEN_HAVE_VPMU
@@ -50,7 +51,7 @@ config XEN_PVHVM_SMP config XEN_512GB bool "Limit Xen pv-domain memory to 512GB" - depends on XEN_PV && X86_64 + depends on XEN_PV default y help Limit paravirtualized user domains to 512GB of RAM.
arch/x86/xen/Makefile +1 -2
@@ -1,5 +1,5 @@ # SPDX-License-Identifier: GPL-2.0 -OBJECT_FILES_NON_STANDARD_xen-asm_$(BITS).o := y +OBJECT_FILES_NON_STANDARD_xen-asm.o := y ifdef CONFIG_FUNCTION_TRACER # Do not profile debug and lowlevel utilities
@@ -34,7 +34,6 @@ obj-$(CONFIG_XEN_PV) += mmu_pv.o obj-$(CONFIG_XEN_PV) += irq.o obj-$(CONFIG_XEN_PV) += multicalls.o obj-$(CONFIG_XEN_PV) += xen-asm.o -obj-$(CONFIG_XEN_PV) += xen-asm_$(BITS).o obj-$(CONFIG_XEN_PVH) += enlighten_pvh.o
arch/x86/xen/apic.c +0 -17
@@ -58,10 +58,6 @@ static u32 xen_apic_read(u32 reg)if (reg ==APIC_LVR )return 0x14 ; -#ifdef CONFIG_X86_32 -if (reg ==APIC_LDR ) -return SET_APIC_LOGICAL_ID (1UL <<smp_processor_id ()); -#endif if (reg !=APIC_ID )return 0 ;
@@ -127,14 +123,6 @@ static int xen_phys_pkg_id(int initial_apic_id, int index_msb)return initial_apic_id >> index_msb; } -#ifdef CONFIG_X86_32 -static int xen_x86_32_early_logical_apicid (int cpu) -{ -/* Match with APIC_LDR read. Otherwise setup_local_APIC complains. */ -return 1 << cpu; -} -#endif -static void xen_noop (void ) { }
@@ -197,11 +185,6 @@ static struct apic xen_pv_apic = { .icr_write = xen_apic_icr_write, .wait_icr_idle = xen_noop, .safe_wait_icr_idle = xen_safe_apic_wait_icr_idle, - -#ifdef CONFIG_X86_32 -/* generic_processor_info and setup_local_APIC. */ - .x86_32_early_logical_apicid = xen_x86_32_early_logical_apicid, -#endif };static void __init xen_apic_check (void )
arch/x86/xen/enlighten_pv.c +2 -46
152 lines not on this page · raw message
arch/x86/xen/mmu_pv.c +36 -304
576 lines not on this page · raw message
arch/x86/xen/p2m.c +1 -5
@@ -379,12 +379,8 @@ static void __init xen_rebuild_p2m_list(unsigned long *p2m)if (type ==P2M_TYPE_PFN || i < chunk) {/* Use initial p2m page contents. */ -#ifdef CONFIG_X86_64 mfns =alloc_p2m_page ();copy_page (mfns, xen_p2m_addr + pfn); -#else - mfns = xen_p2m_addr + pfn; -#endif ptep =populate_extra_pte ((unsigned long )(p2m + pfn));set_pte (ptep,pfn_pte (PFN_DOWN (__pa (mfns)),PAGE_KERNEL ));
@@ -467,7 +463,7 @@ EXPORT_SYMBOL_GPL(get_phys_to_machine);* Allocate new pmd(s). It is checked whether the old pmd is still in place. * If not, nothing is changed. This is okay as the only reason for allocating * a new pmd is to replace p2m_missing_pte or p2m_identity_pte by a individual -* pmd. In case of PAE/x86-32 there are multiple pmds to allocate! +* pmd. */ static pte_t *alloc_p2m_pmd (unsigned long addr,pte_t *pte_pg) {
arch/x86/xen/setup.c +1 -34
@@ -545,13 +545,10 @@ static unsigned long __init xen_get_pages_limit(void) {unsigned long limit; -#ifdef CONFIG_X86_32 - limit =GB (64 ) /PAGE_SIZE ; -#else limit =MAXMEM /PAGE_SIZE ;if (!xen_initial_domain () && xen_512gb_limit) limit =GB (512 ) /PAGE_SIZE ; -#endif +return limit; }
@@ -722,17 +719,8 @@ static void __init xen_reserve_xen_mfnlist(void)if (!xen_is_e820_reserved (start, size))return ; -#ifdef CONFIG_X86_32 -/* -* Relocating the p2m on 32 bit system to an arbitrary virtual address -* is not supported, so just give up. -*/ -xen_raw_console_write ("Xen hypervisor allocated p2m list conflicts with E820 map\n" ); -BUG (); -#else xen_relocate_p2m ();memblock_free (start, size); -#endif }/**
@@ -921,20 +909,6 @@ char * __init xen_memory_setup(void)return "Xen" ; } -/* -* Set the bit indicating "nosegneg" library variants should be used. -* We only need to bother in pure 32-bit mode; compat 32-bit processes -* can have un-truncated segments, so wrapping around is allowed. -*/ -static void __init fiddle_vdso (void ) -{ -#ifdef CONFIG_X86_32 -u32 *mask = vdso_image_32.data + - vdso_image_32.sym_VDSO32_NOTE_MASK; - *mask |=1 <<VDSO_NOTE_NONEGSEG_BIT ; -#endif -} -static int register_callback (unsigned type,const void *func) {struct callback_register callback = {
@@ -951,11 +925,7 @@ void xen_enable_sysenter(void)int ret;unsigned sysenter_feature; -#ifdef CONFIG_X86_32 - sysenter_feature =X86_FEATURE_SEP ; -#else sysenter_feature =X86_FEATURE_SYSENTER32 ; -#endif if (!boot_cpu_has (sysenter_feature))return ;
@@ -967,7 +937,6 @@ void xen_enable_sysenter(void)void xen_enable_syscall (void ) { -#ifdef CONFIG_X86_64 int ret; ret =register_callback (CALLBACKTYPE_syscall, xen_syscall_target);
@@ -983,7 +952,6 @@ void xen_enable_syscall(void)if (ret !=0 )setup_clear_cpu_cap (X86_FEATURE_SYSCALL32 ); } -#endif /* CONFIG_X86_64 */ }static void __init xen_pvmmu_arch_setup (void )
@@ -1024,7 +992,6 @@ void __init xen_arch_setup(void)disable_cpuidle ();disable_cpufreq ();WARN_ON (xen_set_default_idle ()); -fiddle_vdso ();#ifdef CONFIG_NUMA numa_off =1 ;#endif
arch/x86/xen/smp_pv.c +0 -18
@@ -212,15 +212,6 @@ static void __init xen_pv_smp_prepare_boot_cpu(void)* sure the old memory can be recycled. */ make_lowmem_page_readwrite (xen_initial_gdt); -#ifdef CONFIG_X86_32 -/* -* Xen starts us with XEN_FLAT_RING1_DS, but linux code -* expects __USER_DS -*/ -loadsegment (ds, __USER_DS); -loadsegment (es, __USER_DS); -#endif -xen_filter_cpu_maps ();xen_setup_vcpu_info_placement ();
@@ -301,10 +292,6 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle) gdt =get_cpu_gdt_rw (cpu); -#ifdef CONFIG_X86_32 - ctxt->user_regs.fs = __KERNEL_PERCPU; - ctxt->user_regs.gs = __KERNEL_STACK_CANARY; -#endif memset (&ctxt->fpu_ctxt,0 ,sizeof (ctxt->fpu_ctxt));/*
@@ -342,12 +329,7 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle) ctxt->kernel_ss = __KERNEL_DS; ctxt->kernel_sp =task_top_of_stack (idle); -#ifdef CONFIG_X86_32 - ctxt->event_callback_cs = __KERNEL_CS; - ctxt->failsafe_callback_cs = __KERNEL_CS; -#else ctxt->gs_base_kernel =per_cpu_offset (cpu); -#endif ctxt->event_callback_eip = (unsigned long )xen_asm_exc_xen_hypervisor_callback; ctxt->failsafe_callback_eip =
arch/x86/xen/xen-asm.S +168 -14
216 lines not on this page · raw message
arch/x86/xen/xen-asm_32.S deleted +0 -185
185 lines not on this page · raw message
arch/x86/xen/xen-asm_64.S deleted +0 -181
181 lines not on this page · raw message
arch/x86/xen/xen-head.S +0 -6
@@ -35,13 +35,8 @@ SYM_CODE_START(startup_xen) rep __ASM_SIZE(stos) mov %_ASM_SI, xen_start_info -#ifdef CONFIG_X86_64 mov initial_stack(%rip), %rsp -#else - mov initial_stack, %esp -#endif -#ifdef CONFIG_X86_64 /* Set up %gs. * * The base of %gs always points to fixed_percpu_data. If the
@@ -53,7 +48,6 @@ SYM_CODE_START(startup_xen) movq $INIT_PER_CPU_VAR(fixed_percpu_data),%rax cdq wrmsr -#endif call xen_start_kernel SYM_CODE_END(startup_xen)
drivers/xen/Kconfig +1 -3
@@ -52,9 +52,7 @@ config XEN_BALLOON_MEMORY_HOTPLUG config XEN_BALLOON_MEMORY_HOTPLUG_LIMIT int "Hotplugged memory limit (in GiB) for a PV guest" - default 512 if X86_64 - default 4 if X86_32 - range 0 64 if X86_32 + default 512 depends on XEN_HAVE_PVMMU depends on XEN_BALLOON_MEMORY_HOTPLUG help
Signature · 1 line
2.26.2
Re: [PATCH v2 1/4] x86/xen: remove 32-bit Xen PV guest support
Quoted 7 lines · Juergen Gross
Xen is requiring 64-bit machines today and since Xen 4.14 it can be built without 32-bit PV guest support. There is no need to carry the burden of 32-bit PV guest support in the kernel any longer, as new guests can be either HVM or PVH, or they can use a 64 bit kernel.
Remove the 32-bit Xen PV support from the kernel.
If you send a v3, it would be better to split the move of the 64-bit code into xen-asm.S into a separate patch.
--
Brian Gerst
Boris Ostrovsky
reviewerRe: [PATCH v2 1/4] x86/xen: remove 32-bit Xen PV guest support
In reply to Juergen GrossQuoted 8 lines · Juergen Gross
Xen is requiring 64-bit machines today and since Xen 4.14 it can be built without 32-bit PV guest support. There is no need to carry the burden of 32-bit PV guest support in the kernel any longer, as new guests can be either HVM or PVH, or they can use a 64 bit kernel.
Remove the 32-bit Xen PV support from the kernel.
Signed-off-by Juergen Gross <[email protected]>
Quoted 9 lines
--- arch/x86/entry/entry_32.S | 109 +---------- arch/x86/include/asm/proto.h | 2 +- arch/x86/include/asm/segment.h | 2 +- arch/x86/kernel/head_32.S | 31 --- arch/x86/xen/Kconfig | 3 +- arch/x86/xen/Makefile | 3 +- arch/x86/xen/apic.c | 17 -- arch/x86/xen/enlighten_pv.c | 48 +----
Should we drop PageHighMem() test in set_aliased_prot()?
(And there are few other places where is is used, in mmu_pv.c)
arch/x86/xen/enlighten_pv.c static void xen_load_tls(struct thread_struct *t, unsigned int cpu) · line 561
Quoted 8 lines
arch/x86/xen/enlighten_pv.c +0
-4@@ -555,13 +547,8 @@ static void xen_load_tls(struct thread_struct *t, unsigned int cpu)* exception between the new %fs descriptor being loaded and * %fs being effectively cleared at __switch_to(). */ -if (paravirt_get_lazy_mode () ==PARAVIRT_LAZY_CPU ) { -#ifdef CONFIG_X86_32 -lazy_load_gs (0 ); -#else
I think this also needs an adjustment to the preceding comment.
arch/x86/xen/mmu_pv.c static void xen_set_pud(pud_t *ptr, pud_t val) · line 446
Quoted 6 lines
arch/x86/xen/mmu_pv.c +0
-5-#ifdef CONFIG_X86_PAE -static void xen_set_pte_atomic (pte_t *ptep,pte_t pte) -{ -trace_xen_mmu_set_pte_atomic (ptep, pte); -__xen_set_pte (ptep, pte);
Probably not for this series but I wonder whether __xen_set_pte() should continue to use hypercall now that we are 64-bit only.
Review static int __xen_pgd_walk(struct mm_struct *mm, pgd_t *pgd,
Quoted 1 line
(quoted diff) +0
-0@@ -654,14 +621,12 @@ static int __xen_pgd_walk(struct mm_struct *mm, pgd_t *pgd,
Comment above should be updated.
-boris
Re: [PATCH v2 1/4] x86/xen: remove 32-bit Xen PV guest support
Quoted 10 lines · Boris Ostrovsky
arch/x86/xen/mmu_pv.c static void xen_set_pud(pud_t *ptr, pud_t val) · line 446
Quoted 6 lines · Juergen Gross
arch/x86/xen/mmu_pv.c +0
-5-#ifdef CONFIG_X86_PAE -static void xen_set_pte_atomic (pte_t *ptep,pte_t pte) -{ -trace_xen_mmu_set_pte_atomic (ptep, pte); -__xen_set_pte (ptep, pte);Probably not for this series but I wonder whether
__xen_set_pte()should continue to use hypercall now that we are 64-bit only.
The hypercall path is a SYSCALL (and SYSRET out).
The "writeable" PTE path is a #PF, followed by an x86 instruction emulation, which then reaches the same logic as the hypercall path (and an IRET out).
~Andrew
Boris Ostrovsky
reviewerRe: [PATCH v2 1/4] x86/xen: remove 32-bit Xen PV guest support
Quoted 10 lines · Andrew Cooper
Quoted 9 lines · Boris Ostrovsky
arch/x86/xen/mmu_pv.c static void xen_set_pud(pud_t *ptr, pud_t val) · line 446
Quoted 6 lines · Juergen Gross
arch/x86/xen/mmu_pv.c +0
-5-#ifdef CONFIG_X86_PAE -static void xen_set_pte_atomic (pte_t *ptep,pte_t pte) -{ -trace_xen_mmu_set_pte_atomic (ptep, pte); -__xen_set_pte (ptep, pte);Probably not for this series but I wonder whether
__xen_set_pte()should continue to use hypercall now that we are 64-bit only.
Quoted 5 lines
The hypercall path is a SYSCALL (and SYSRET out).
The "writeable" PTE path is a #PF, followed by an x86 instruction emulation, which then reaches the same logic as the hypercall path (and an IRET out).
Then we should at least update the comment.
-boris
Jürgen Groß
maintainerRe: [PATCH v2 1/4] x86/xen: remove 32-bit Xen PV guest support
In reply to Boris OstrovskyQuoted 20 lines · Boris Ostrovsky
Quoted 8 lines · Juergen Gross
Xen is requiring 64-bit machines today and since Xen 4.14 it can be built without 32-bit PV guest support. There is no need to carry the burden of 32-bit PV guest support in the kernel any longer, as new guests can be either HVM or PVH, or they can use a 64 bit kernel.
Remove the 32-bit Xen PV support from the kernel.
Signed-off-by Juergen Gross <[email protected]>Quoted 9 lines
--- arch/x86/entry/entry_32.S | 109 +---------- arch/x86/include/asm/proto.h | 2 +- arch/x86/include/asm/segment.h | 2 +- arch/x86/kernel/head_32.S | 31 --- arch/x86/xen/Kconfig | 3 +- arch/x86/xen/Makefile | 3 +- arch/x86/xen/apic.c | 17 -- arch/x86/xen/enlighten_pv.c | 48 +----
Quoted 4 lines
Should we drop
PageHighMem()test inset_aliased_prot()?(And there are few other places where is is used, in
mmu_pv.c)
Yes, will drop those.
Quoted 14 lines
arch/x86/xen/enlighten_pv.c static void xen_load_tls(struct thread_struct *t, unsigned int cpu) · line 561
Quoted 11 lines
arch/x86/xen/enlighten_pv.c +0
-4@@ -555,13 +547,8 @@ static void xen_load_tls(struct thread_struct *t, unsigned int cpu)* exception between the new %fs descriptor being loaded and * %fs being effectively cleared at __switch_to(). */ -if (paravirt_get_lazy_mode () ==PARAVIRT_LAZY_CPU ) { -#ifdef CONFIG_X86_32 -lazy_load_gs (0 ); -#else I think this also needs an adjustment to the preceding comment.
Yes.
Quoted 12 lines
arch/x86/xen/mmu_pv.c static void xen_set_pud(pud_t *ptr, pud_t val) · line 446
Quoted 8 lines
arch/x86/xen/mmu_pv.c +0
-5-#ifdef CONFIG_X86_PAE -static void xen_set_pte_atomic (pte_t *ptep,pte_t pte) -{ -trace_xen_mmu_set_pte_atomic (ptep, pte); -__xen_set_pte (ptep, pte);Probably not for this series but I wonder whether
__xen_set_pte()should continue to use hypercall now that we are 64-bit only.
As Andrew wrote already the hypercall will be cheaper.
I'll adjust the comment, though.
Quoted 6 lines
Review static int __xen_pgd_walk(struct mm_struct *mm, pgd_t *pgd,
Quoted 3 lines
(quoted diff) +0
-0@@ -654,14 +621,12 @@ static int __xen_pgd_walk(struct mm_struct *mm, pgd_t *pgd,Comment above should be updated.
Yes.
Juergen
Juergen Gross
maintainerPATCH v2 2/4 x86/paravirt: remove 32-bit support from PARAVIRT_XXL
In reply to Juergen GrossThe last 32-bit user of stuff under CONFIG_PARAVIRT_XXL is gone.
Remove 32-bit specific parts.
--- arch/x86/entry/vdso/vdso32/vclock_gettime.c | 1 + arch/x86/include/asm/paravirt.h | 92 +++------------------ arch/x86/include/asm/paravirt_types.h | 21 ----- arch/x86/include/asm/pgtable-3level_types.h | 5 -- arch/x86/include/asm/segment.h | 4 - arch/x86/kernel/cpu/common.c | 8 -- arch/x86/kernel/kprobes/core.c | 1 - arch/x86/kernel/kprobes/opt.c | 1 - arch/x86/kernel/paravirt.c | 18 ---- arch/x86/kernel/paravirt_patch.c | 17 ---- arch/x86/xen/enlighten_pv.c | 6 -- 11 files changed, 13 insertions(+), 161 deletions(-)
arch/x86/entry/vdso/vdso32/vclock_gettime.c +1 -0
@@ -14,6 +14,7 @@#undef CONFIG_ILLEGAL_POINTER_VALUE#undef CONFIG_SPARSEMEM_VMEMMAP#undef CONFIG_NR_CPUS +#undef CONFIG_PARAVIRT_XXL#define CONFIG_X86_32 1#define CONFIG_PGTABLE_LEVELS 2
arch/x86/include/asm/paravirt.h +12 -80
184 lines not on this page · raw message
arch/x86/include/asm/paravirt_types.h +0 -21
@@ -68,12 +68,7 @@ struct paravirt_callee_save {/* general info */ struct pv_info {#ifdef CONFIG_PARAVIRT_XXL -unsigned int kernel_rpl; -int shared_kernel_pmd; - -#ifdef CONFIG_X86_64 u16 extra_user_64bit_cs;/* __USER_CS if none */ -#endif #endif const char *name;
@@ -126,9 +121,7 @@ struct pv_cpu_ops {void (*set_ldt)(const void *desc,unsigned entries);unsigned long (*store_tr)(void );void (*load_tls)(struct thread_struct *t,unsigned int cpu); -#ifdef CONFIG_X86_64 void (*load_gs_index)(unsigned int idx); -#endif void (*write_ldt_entry)(struct desc_struct *ldt,int entrynum,const void *desc); void (*write_gdt_entry)(struct desc_struct *,
@@ -263,21 +256,11 @@ struct pv_mmu_ops {struct paravirt_callee_save pgd_val;struct paravirt_callee_save make_pgd; -#if CONFIG_PGTABLE_LEVELS >=3 -#ifdef CONFIG_X86_PAE -void (*set_pte_atomic)(pte_t *ptep,pte_t pteval); -void (*pte_clear)(struct mm_struct *mm,unsigned long addr, -pte_t *ptep); -void (*pmd_clear)(pmd_t *pmdp); - -#endif /* CONFIG_X86_PAE */ -void (*set_pud)(pud_t *pudp,pud_t pudval);struct paravirt_callee_save pmd_val;struct paravirt_callee_save make_pmd; -#if CONFIG_PGTABLE_LEVELS >=4 struct paravirt_callee_save pud_val;struct paravirt_callee_save make_pud;
@@ -290,10 +273,6 @@ struct pv_mmu_ops {void (*set_pgd)(pgd_t *pgdp,pgd_t pgdval);#endif /* CONFIG_PGTABLE_LEVELS >= 5 */ -#endif /* CONFIG_PGTABLE_LEVELS >= 4 */ - -#endif /* CONFIG_PGTABLE_LEVELS >= 3 */ -struct pv_lazy_ops lazy_mode;/* dom0 ops */
arch/x86/include/asm/pgtable-3level_types.h +0 -5
@@ -20,12 +20,7 @@ typedef union { } pte_t;#endif /* !__ASSEMBLY__ */ -#ifdef CONFIG_PARAVIRT_XXL -#define SHARED_KERNEL_PMD ((!static_cpu_has(X86_FEATURE_PTI) && \ - (pv_info.shared_kernel_pmd))) -#else #define SHARED_KERNEL_PMD (!static_cpu_has(X86_FEATURE_PTI)) -#endif #define ARCH_PAGE_TABLE_SYNC_MASK (SHARED_KERNEL_PMD ? 0 : PGTBL_PMD_MODIFIED)
arch/x86/include/asm/segment.h +0 -4
@@ -222,10 +222,6 @@#endif -#ifndef CONFIG_PARAVIRT_XXL -# define get_kernel_rpl () 0 -#endif -#define IDT_ENTRIES 256#define NUM_EXCEPTION_VECTORS 32
arch/x86/kernel/cpu/common.c +0 -8
@@ -1396,15 +1396,7 @@ static void generic_identify(struct cpuinfo_x86 *c)* ESPFIX issue, we can change this. */ #ifdef CONFIG_X86_32 -# ifdef CONFIG_PARAVIRT_XXL -do { -extern void native_iret (void ); -if (pv_ops.cpu.iret == native_iret) -set_cpu_bug (c,X86_BUG_ESPFIX ); - }while (0 ); -# else set_cpu_bug (c,X86_BUG_ESPFIX ); -# endif #endif }
arch/x86/kernel/kprobes/core.c +0 -1
@@ -780,7 +780,6 @@ __used __visible void *trampoline_handler(struct pt_regs *regs)/* fixup registers */ regs->cs = __KERNEL_CS;#ifdef CONFIG_X86_32 - regs->cs |=get_kernel_rpl (); regs->gs =0 ;#endif /* We use pt_regs->sp for return address holder. */
arch/x86/kernel/kprobes/opt.c +0 -1
@@ -180,7 +180,6 @@ optimized_callback(struct optimized_kprobe *op, struct pt_regs *regs)/* Save skipped registers */ regs->cs = __KERNEL_CS;#ifdef CONFIG_X86_32 - regs->cs |=get_kernel_rpl (); regs->gs =0 ;#endif regs->ip = (unsigned long )op->kp.addr +INT3_INSN_SIZE ;
arch/x86/kernel/paravirt.c +0 -18
@@ -263,13 +263,8 @@ enum paravirt_lazy_mode paravirt_get_lazy_mode(void)struct pv_info pv_info = { .name ="bare hardware" , #ifdef CONFIG_PARAVIRT_XXL - .kernel_rpl =0 , - .shared_kernel_pmd =1 ,/* Only used when CONFIG_X86_PAE is set */ - -#ifdef CONFIG_X86_64 .extra_user_64bit_cs = __USER_CS, #endif -#endif };/* 64-bit pagetable entries */
@@ -305,9 +300,7 @@ struct paravirt_patch_template pv_ops = { .cpu.load_idt = native_load_idt, .cpu.store_tr = native_store_tr, .cpu.load_tls = native_load_tls, -#ifdef CONFIG_X86_64 .cpu.load_gs_index = native_load_gs_index, -#endif .cpu.write_ldt_entry = native_write_ldt_entry, .cpu.write_gdt_entry = native_write_gdt_entry, .cpu.write_idt_entry = native_write_idt_entry,
@@ -317,9 +310,7 @@ struct paravirt_patch_template pv_ops = { .cpu.load_sp0 = native_load_sp0, -#ifdef CONFIG_X86_64 .cpu.usergs_sysret64 = native_usergs_sysret64, -#endif .cpu.iret = native_iret, .cpu.swapgs = native_swapgs,
@@ -374,18 +365,11 @@ struct paravirt_patch_template pv_ops = { .mmu.ptep_modify_prot_start = __ptep_modify_prot_start, .mmu.ptep_modify_prot_commit = __ptep_modify_prot_commit, -#if CONFIG_PGTABLE_LEVELS >=3 -#ifdef CONFIG_X86_PAE - .mmu.set_pte_atomic = native_set_pte_atomic, - .mmu.pte_clear = native_pte_clear, - .mmu.pmd_clear = native_pmd_clear, -#endif .mmu.set_pud = native_set_pud, .mmu.pmd_val = PTE_IDENT, .mmu.make_pmd = PTE_IDENT, -#if CONFIG_PGTABLE_LEVELS >= 4 .mmu.pud_val = PTE_IDENT, .mmu.make_pud = PTE_IDENT,
@@ -397,8 +381,6 @@ struct paravirt_patch_template pv_ops = { .mmu.set_pgd = native_set_pgd,#endif /* CONFIG_PGTABLE_LEVELS >= 5 */ -#endif /* CONFIG_PGTABLE_LEVELS >= 4 */ -#endif /* CONFIG_PGTABLE_LEVELS >= 3 */ .mmu.pte_val = PTE_IDENT, .mmu.pgd_val = PTE_IDENT,
arch/x86/kernel/paravirt_patch.c +0 -17
@@ -26,14 +26,10 @@ struct patch_xxl {const unsigned char mmu_read_cr3[3 ];const unsigned char mmu_write_cr3[3 ];const unsigned char irq_restore_fl[2 ]; -# ifdef CONFIG_X86_64 const unsigned char cpu_wbinvd[2 ];const unsigned char cpu_usergs_sysret64[6 ];const unsigned char cpu_swapgs[3 ];const unsigned char mov64[3 ]; -# else -const unsigned char cpu_iret[1 ]; -# endif };static const struct patch_xxl patch_data_xxl = {
@@ -42,7 +38,6 @@ static const struct patch_xxl patch_data_xxl = { .irq_save_fl = { 0x9c, 0x58 }, // pushf; pop %[re]ax .mmu_read_cr2 = { 0x0f, 0x20, 0xd0 }, // mov %cr2, %[re]ax .mmu_read_cr3 = { 0x0f, 0x20, 0xd8 }, // mov %cr3, %[re]ax -# ifdef CONFIG_X86_64 .mmu_write_cr3 = { 0x0f, 0x22, 0xdf }, // mov %rdi, %cr3 .irq_restore_fl = { 0x57, 0x9d }, // push %rdi; popfq .cpu_wbinvd = { 0x0f, 0x09 }, // wbinvd
@@ -50,19 +45,11 @@ static const struct patch_xxl patch_data_xxl = { 0x48, 0x0f, 0x07 }, // swapgs; sysretq .cpu_swapgs = { 0x0f, 0x01, 0xf8 }, // swapgs .mov64 = { 0x48, 0x89, 0xf8 }, // mov %rdi, %rax -# else - .mmu_write_cr3 = { 0x0f, 0x22, 0xd8 }, // mov %eax, %cr3 - .irq_restore_fl = { 0x50, 0x9d }, // push %eax; popf - .cpu_iret = { 0xcf }, // iret -# endif };unsigned int paravirt_patch_ident_64 (void *insn_buff,unsigned int len) { -#ifdef CONFIG_X86_64 return PATCH (xxl, mov64, insn_buff, len); -#endif -return 0 ; }# endif /* CONFIG_PARAVIRT_XXL */
@@ -98,13 +85,9 @@ unsigned int native_patch(u8 type, void *insn_buff, unsigned long addr,PATCH_CASE (mmu, read_cr3, xxl, insn_buff, len);PATCH_CASE (mmu, write_cr3, xxl, insn_buff, len); -# ifdef CONFIG_X86_64 PATCH_CASE (cpu, usergs_sysret64, xxl, insn_buff, len);PATCH_CASE (cpu, swapgs, xxl, insn_buff, len);PATCH_CASE (cpu, wbinvd, xxl, insn_buff, len); -# else -PATCH_CASE (cpu, iret, xxl, insn_buff, len); -# endif #endif #ifdef CONFIG_PARAVIRT_SPINLOCKS
arch/x86/xen/enlighten_pv.c +0 -6
@@ -1002,8 +1002,6 @@ void __init xen_setup_vcpu_info_placement(void) }static const struct pv_info xen_info__initconst = { - .shared_kernel_pmd =0 , - .extra_user_64bit_cs =FLAT_USER_CS64 , .name ="Xen" , };
@@ -1301,10 +1299,6 @@ asmlinkage __visible void __init xen_start_kernel(void) xen_start_info->nr_pages);xen_reserve_special_pages (); -/* keep using Xen gdt for now; no urgent need to change it */ - - pv_info.kernel_rpl =0 ; -/* set the limit of our address space */ xen_reserve_top ();
Signature · 1 line
2.26.2
Juergen Gross
maintainerPATCH v2 3/4 x86/paravirt: cleanup paravirt macros
In reply to Juergen GrossSome paravirt macros are no longer used, delete them.
--- arch/x86/include/asm/paravirt.h | 15 --------------- 1 file changed, 15 deletions(-)
arch/x86/include/asm/paravirt.h +0 -15
@@ -609,16 +609,9 @@ bool __raw_callee_save___native_vcpu_is_preempted(long cpu);#endif /* SMP && PARAVIRT_SPINLOCKS */ #ifdef CONFIG_X86_32 -#define PV_SAVE_REGS "pushl %ecx; pushl %edx;" -#define PV_RESTORE_REGS "popl %edx; popl %ecx;" -/* save and restore all caller-save registers, except return value */ #define PV_SAVE_ALL_CALLER_REGS "pushl %ecx;"#define PV_RESTORE_ALL_CALLER_REGS "popl %ecx;" - -#define PV_FLAGS_ARG "0" -#define PV_EXTRA_CLOBBERS -#define PV_VEXTRA_CLOBBERS #else /* save and restore all caller-save registers, except return value */ #define PV_SAVE_ALL_CALLER_REGS \
@@ -639,14 +632,6 @@ bool __raw_callee_save___native_vcpu_is_preempted(long cpu);"pop %rsi;" \"pop %rdx;" \ "pop %rcx;" - -/* We save some registers, but all of them, that's too much. We clobber all - * caller saved registers but the argument parameter */ -#define PV_SAVE_REGS "pushq %%rdi;" -#define PV_RESTORE_REGS "popq %%rdi;" -#define PV_EXTRA_CLOBBERS EXTRA_CLOBBERS, "rcx" , "rdx", "rsi" -#define PV_VEXTRA_CLOBBERS EXTRA_CLOBBERS, "rdi", "rcx" , "rdx", "rsi" -#define PV_FLAGS_ARG "D"#endif /*
Signature · 1 line
2.26.2
Juergen Gross
maintainerPATCH v2 4/4 x86/paravirt: use CONFIG_PARAVIRT_XXL instead of CONFIG_PARAVIRT
In reply to Juergen GrossThere are some code parts using CONFIG_PARAVIRT for Xen pvops related issues instead of the more stringent CONFIG_PARAVIRT_XXL.
--- arch/x86/entry/entry_64.S | 4 ++-- arch/x86/include/asm/fixmap.h | 2 +- arch/x86/include/asm/required-features.h | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-)
arch/x86/entry/entry_64.S +2 -2
@@ -45,13 +45,13 @@ .code64 .section .entry.text, "ax" -#ifdef CONFIG_PARAVIRT +#ifdef CONFIG_PARAVIRT_XXL SYM_CODE_START(native_usergs_sysret64) UNWIND_HINT_EMPTY swapgs sysretq SYM_CODE_END(native_usergs_sysret64) -#endif /* CONFIG_PARAVIRT */ +#endif /* CONFIG_PARAVIRT_XXL */ /* * 64-bit SYSCALL instruction entry. Up to 6 arguments in registers.
arch/x86/include/asm/fixmap.h +1 -1
@@ -99,7 +99,7 @@ enum fixed_addresses { FIX_PCIE_MCFG,#endif #endif -#ifdef CONFIG_PARAVIRT +#ifdef CONFIG_PARAVIRT_XXL FIX_PARAVIRT_BOOTMAP,#endif #ifdef CONFIG_X86_INTEL_MID
arch/x86/include/asm/required-features.h +1 -1
@@ -54,7 +54,7 @@#endif #ifdef CONFIG_X86_64 -#ifdef CONFIG_PARAVIRT +#ifdef CONFIG_PARAVIRT_XXL /* Paravirtualized systems may not have PSE or PGE available */ #define NEED_PSE 0#define NEED_PGE 0
Signature · 1 line
2.26.2
Quoted 8 lines · Juergen Gross
The long term plan has been to replace Xen PV guests by PVH. The first victim of that plan are now 32-bit PV guests, as those are used only rather seldom these days. Xen on x86 requires 64-bit support and with Grub2 now supporting PVH officially since version 2.04 there is no need to keep 32-bit PV guest support alive in the Linux kernel. Additionally Meltdown mitigation is not available in the kernel running as 32-bit PV guest, so dropping this mode makes sense from security point of view, too.
Hooray!!! Much thanks!
Quoted 9 lines · Juergen Gross
The long term plan has been to replace Xen PV guests by PVH. The first victim of that plan are now 32-bit PV guests, as those are used only rather seldom these days. Xen on x86 requires 64-bit support and with Grub2 now supporting PVH officially since version 2.04 there is no need to keep 32-bit PV guest support alive in the Linux kernel. Additionally Meltdown mitigation is not available in the kernel running as 32-bit PV guest, so dropping this mode makes sense from security point of view, too.
One thing that you missed is removing VDSO_NOTE_NONEGSEG_BIT from vdso32/note.S. With that removed there is no difference from the 64-bit version.
Otherwise this series looks good to me.
--
Brian Gerst
Jürgen Groß
maintainerQuoted 11 lines · Brian Gerst
Quoted 9 lines · Juergen Gross
The long term plan has been to replace Xen PV guests by PVH. The first victim of that plan are now 32-bit PV guests, as those are used only rather seldom these days. Xen on x86 requires 64-bit support and with Grub2 now supporting PVH officially since version 2.04 there is no need to keep 32-bit PV guest support alive in the Linux kernel. Additionally Meltdown mitigation is not available in the kernel running as 32-bit PV guest, so dropping this mode makes sense from security point of view, too.
Quoted 3 lines
One thing that you missed is removing
VDSO_NOTE_NONEGSEG_BITfromvdso32/note.S. With that removed there is no difference from the 64-bit version.
Oh, this means we can probably remove arch/x86/xen/vdso.h completely.
Quoted 2 lines
Otherwise this series looks good to me.
Thanks,
Juergen