Skip to contentSkip to conversation

powerpc/64: Add INTERRUPT_SANITIZE_REGISTERS Kconfig

PATCH Superseded Rohan McLure 2 participants 12 messages Nov 29, 2022 PowerPC

v4 of 5 Superseded Latest is v5 →

lore
Share & tools
lore
#0 lore raw

PATCH v4 1/7 powerpc/64: Add INTERRUPT_SANITIZE_REGISTERS Kconfig

Add Kconfig option for enabling clearing of registers on arrival in an interrupt handler. This reduces the speculation influence of registers on kernel internals. The option will be consumed by 64-bit systems that feature speculation and wish to implement this mitigation.

This patch only introduces the Kconfig option, no actual mitigations.

The primary overhead of this mitigation lies in an increased number of registers that must be saved and restored by interrupt handlers on Book3S systems. Enable by default on Book3E systems, which prior to this patch eagerly save and restore register state, meaning that the mitigation when implemented will have minimal overhead.

---
Resubmitting patches as their own series after v6 partially merged:

---
 arch/powerpc/Kconfig | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/powerpc/Kconfig +9 -0
index 4fd4924f6d50..280c797e0f30 100644
@@ -531,6 +531,15 @@ config HOTPLUG_CPU
 
 	  Say N if you are unsure.
 
+config INTERRUPT_SANITIZE_REGISTERS
+	bool "Clear gprs on interrupt arrival"
+	depends on PPC64 && ARCH_HAS_SYSCALL_WRAPPER
+	default PPC_BOOK3E_64
+	help
+	  Reduce the influence of user register state on interrupt handlers and
+	  syscalls through clearing user state from registers before handling
+	  the exception.
+
 config PPC_QUEUED_SPINLOCKS
 	bool "Queued spinlocks" if EXPERT
 	depends on SMP
Signature · 1 line
2.37.2
#1 lore raw

PATCH v4 2/7 powerpc/64: Add interrupt register sanitisation macros

Include in asm/ppc_asm.h macros to be used in multiple successive patches to implement zeroising architected registers in interrupt handlers. Registers will be sanitised in this fashion in future patches to reduce the speculation influence of user-controlled register values. These mitigations will be configurable through the CONFIG_INTERRUPT_SANITIZE_REGISTERS Kconfig option.

Included are macros for conditionally zeroising registers and restoring as required with the mitigation enabled. With the mitigation disabled, non-volatiles must be restored on demand at separate locations to those required by the mitigation.

---
v4: New patch

---
 arch/powerpc/include/asm/ppc_asm.h | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)
arch/powerpc/include/asm/ppc_asm.h +17 -0
index 753a2757bcd4..272b2795c36a 100644
@@ -74,6 +74,23 @@
 #define SAVE_GPR(n, base)		SAVE_GPRS(n, n, base)
 #define REST_GPR(n, base)		REST_GPRS(n, n, base)
 
+/* macros for handling user register sanitisation */
+#ifdef CONFIG_INTERRUPT_SANITIZE_REGISTERS
+#define SANITIZE_ZEROIZE_SYSCALL_GPRS()		ZEROIZE_GPR(0);		\
+						ZEROIZE_GPRS(5, 12);	\
+						ZEROIZE_NVGPRS()
+#define SANITIZE_ZEROIZE_INTERRUPT_NVGPRS()	ZEROIZE_NVGPRS()
+#define SANITIZE_ZEROIZE_NVGPRS()		ZEROIZE_NVGPRS()
+#define SANITIZE_RESTORE_NVGPRS()		REST_NVGPRS(r1)
+#define HANDLER_RESTORE_NVGPRS()
+#else
+#define SANITIZE_ZEROIZE_INTERRUPT_NVGPRS()
+#define SANITIZE_ZEROIZE_SYSCALL_GPRS()
+#define SANITIZE_ZEROIZE_NVGPRS()
+#define SANITIZE_RESTORE_NVGPRS()
+#define HANDLER_RESTORE_NVGPRS()		REST_NVGPRS(r1)
+#endif /* CONFIG_INTERRUPT_SANITIZE_REGISTERS */
+
 #define SAVE_FPR(n, base)	stfd	n,8*TS_FPRWIDTH*(n)(base)
 #define SAVE_2FPRS(n, base)	SAVE_FPR(n, base); SAVE_FPR(n+1, base)
 #define SAVE_4FPRS(n, base)	SAVE_2FPRS(n, base); SAVE_2FPRS(n+2, base)
Signature · 1 line
2.37.2
reviewer Reviewed-by on v4 2/7
#2 lore raw

Re: [PATCH v4 2/7] powerpc/64: Add interrupt register sanitisation macros

Quoted 7 lines · Rohan McLure

Include in asm/ppc_asm.h macros to be used in multiple successive patches to implement zeroising architected registers in interrupt handlers. Registers will be sanitised in this fashion in future patches to reduce the speculation influence of user-controlled register values. These mitigations will be configurable through the CONFIG_INTERRUPT_SANITIZE_REGISTERS Kconfig option.

Quoted 6 lines

Included are macros for conditionally zeroising registers and restoring as required with the mitigation enabled. With the mitigation disabled, non-volatiles must be restored on demand at separate locations to those required by the mitigation.

Thanks. You might just call them SANITIZE_NVGPRS() etc if it's not functionally important that they're zero. But I don't mind long names too much.

Quoted 36 lines

---
v4: New patch

---
 arch/powerpc/include/asm/ppc_asm.h | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)
arch/powerpc/include/asm/ppc_asm.h +17 -0
index 753a2757bcd4..272b2795c36a 100644
@@ -74,6 +74,23 @@
 #define SAVE_GPR(n, base)		SAVE_GPRS(n, n, base)
 #define REST_GPR(n, base)		REST_GPRS(n, n, base)
 
+/* macros for handling user register sanitisation */
+#ifdef CONFIG_INTERRUPT_SANITIZE_REGISTERS
+#define SANITIZE_ZEROIZE_SYSCALL_GPRS()		ZEROIZE_GPR(0);		\
+						ZEROIZE_GPRS(5, 12);	\
+						ZEROIZE_NVGPRS()
+#define SANITIZE_ZEROIZE_INTERRUPT_NVGPRS()	ZEROIZE_NVGPRS()
+#define SANITIZE_ZEROIZE_NVGPRS()		ZEROIZE_NVGPRS()
+#define SANITIZE_RESTORE_NVGPRS()		REST_NVGPRS(r1)
+#define HANDLER_RESTORE_NVGPRS()
+#else
+#define SANITIZE_ZEROIZE_INTERRUPT_NVGPRS()
+#define SANITIZE_ZEROIZE_SYSCALL_GPRS()
+#define SANITIZE_ZEROIZE_NVGPRS()
+#define SANITIZE_RESTORE_NVGPRS()
+#define HANDLER_RESTORE_NVGPRS()		REST_NVGPRS(r1)
+#endif /* CONFIG_INTERRUPT_SANITIZE_REGISTERS */
+
 #define SAVE_FPR(n, base)	stfd	n,8*TS_FPRWIDTH*(n)(base)
 #define SAVE_2FPRS(n, base)	SAVE_FPR(n, base); SAVE_FPR(n+1, base)
 #define SAVE_4FPRS(n, base)	SAVE_2FPRS(n, base); SAVE_2FPRS(n+2, base)
Signature · 1 line
2.37.2
#3 lore raw

PATCH v4 3/7 powerpc/64: Sanitise common exit code for interrupts

In reply to Rohan McLure

Interrupt code is shared between Book3E/S 64-bit systems for interrupt handlers. Ensure that exit code correctly restores non-volatile gprs on each system when CONFIG_INTERRUPT_SANITIZE_REGISTERS is enabled.

Also introduce macros for clearing/restoring registers on interrupt entry for when this configuration option is either disabled or enabled.

---
v4: New patch

---
 arch/powerpc/kernel/interrupt_64.S | 6 ++++++
 1 file changed, 6 insertions(+)
arch/powerpc/kernel/interrupt_64.S +6 -0
index 978a173eb339..1ef4fdef74fb 100644
@@ -408,9 +408,11 @@ interrupt_return_\srr\()_user: /* make backtraces match the _kernel variant */
 _ASM_NOKPROBE_SYMBOL(interrupt_return_\srr\()_user)
 	addi	r3,r1,STACK_FRAME_OVERHEAD
 	bl	interrupt_exit_user_prepare
+#ifndef CONFIG_INTERRUPT_SANITIZE_REGISTERS
 	cmpdi	r3,0
 	bne-	.Lrestore_nvgprs_\srr
 .Lrestore_nvgprs_\srr\()_cont:
+#endif
 	std	r1,PACA_EXIT_SAVE_R1(r13) /* save r1 for restart */
 #ifdef CONFIG_PPC_BOOK3S
 .Linterrupt_return_\srr\()_user_rst_start:
@@ -424,6 +426,7 @@ _ASM_NOKPROBE_SYMBOL(interrupt_return_\srr\()_user)
 	stb	r11,PACAIRQHAPPENED(r13) # clear out possible HARD_DIS
 
 .Lfast_user_interrupt_return_\srr\():
+	SANITIZE_RESTORE_NVGPRS()
 #ifdef CONFIG_PPC_BOOK3S
 	.ifc \srr,srr
 	lbz	r4,PACASRR_VALID(r13)
@@ -493,9 +496,11 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
 	b	.	/* prevent speculative execution */
 .Linterrupt_return_\srr\()_user_rst_end:
 
+#ifndef CONFIG_INTERRUPT_SANITIZE_REGISTERS
 .Lrestore_nvgprs_\srr\():
 	REST_NVGPRS(r1)
 	b	.Lrestore_nvgprs_\srr\()_cont
+#endif
 
 #ifdef CONFIG_PPC_BOOK3S
 interrupt_return_\srr\()_user_restart:
@@ -576,6 +581,7 @@ _ASM_NOKPROBE_SYMBOL(interrupt_return_\srr\()_kernel)
 	stb	r11,PACAIRQHAPPENED(r13) // clear the possible HARD_DIS
 
 .Lfast_kernel_interrupt_return_\srr\():
+	SANITIZE_RESTORE_NVGPRS()
 	cmpdi	cr1,r3,0
 #ifdef CONFIG_PPC_BOOK3S
 	.ifc \srr,srr
Signature · 1 line
2.37.2
reviewer Reviewed-by on v4 3/7
#4 lore raw

Re: [PATCH v4 3/7] powerpc/64: Sanitise common exit code for interrupts

Quoted 19 lines · Rohan McLure

Interrupt code is shared between Book3E/S 64-bit systems for interrupt handlers. Ensure that exit code correctly restores non-volatile gprs on each system when CONFIG_INTERRUPT_SANITIZE_REGISTERS is enabled.

Also introduce macros for clearing/restoring registers on interrupt entry for when this configuration option is either disabled or enabled.

---
v4: New patch

---
 arch/powerpc/kernel/interrupt_64.S | 6 ++++++
 1 file changed, 6 insertions(+)
arch/powerpc/kernel/interrupt_64.S +0 -0
index 978a173eb339..1ef4fdef74fb 100644

arch/powerpc/kernel/interrupt_64.S interrupt_return_\srr\()_user: /* make backtraces match the _kernel variant */ · line 415

Quoted 8 lines
arch/powerpc/kernel/interrupt_64.S +2 -0
@@ -408,9 +408,11 @@ interrupt_return_\srr\()_user: /* make backtraces match the _kernel variant */
 _ASM_NOKPROBE_SYMBOL(interrupt_return_\srr\()_user)
 	addi	r3,r1,STACK_FRAME_OVERHEAD
 	bl	interrupt_exit_user_prepare
+#ifndef CONFIG_INTERRUPT_SANITIZE_REGISTERS
 	cmpdi	r3,0
 	bne-	.Lrestore_nvgprs_\srr
 .Lrestore_nvgprs_\srr\()_cont:
+#endif

Looks pretty good. You might add a comment here to say nvgprs are always restored, in the sanitize case. Not that it's hard to grep for.

Thanks,
Nick

#5 lore raw

PATCH v4 4/7 powerpc/64s: IOption for MSR stored in r12

In reply to Rohan McLure

Interrupt handlers in asm/exceptions-64s.S contain a great deal of common code produced by the GEN_COMMON macros. Currently, at the exit point of the macro, r12 will contain the contents of the MSR. A future patch will cause these macros to zeroise architected registers to avoid potential speculation influence of user data.

Provide an IOption that signals that r12 must be retained, as the interrupt handler assumes it to hold the contents of the MSR.

---
v4: Split 64s register sanitisation commit to establish this IOption

---
 arch/powerpc/kernel/exceptions-64s.S | 7 +++++++
 1 file changed, 7 insertions(+)
arch/powerpc/kernel/exceptions-64s.S +7 -0
index 5381a43e50fe..58d72db1d484 100644
@@ -111,6 +111,7 @@ name:
 #define ISTACK		.L_ISTACK_\name\()	/* Set regular kernel stack */
 #define __ISTACK(name)	.L_ISTACK_ ## name
 #define IKUAP		.L_IKUAP_\name\()	/* Do KUAP lock */
+#define IMSR_R12	.L_IMSR_R12_\name\()	/* Assumes MSR saved to r12 */
 
 #define INT_DEFINE_BEGIN(n)						\
 .macro int_define_ ## n name
@@ -176,6 +177,9 @@ do_define_int n
 	.ifndef IKUAP
 		IKUAP=1
 	.endif
+	.ifndef IMSR_R12
+		IMSR_R12=0
+	.endif
 .endm
 
 /*
@@ -1751,6 +1755,7 @@ INT_DEFINE_BEGIN(fp_unavailable)
 #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 	IKVM_REAL=1
 #endif
+	IMSR_R12=1
 INT_DEFINE_END(fp_unavailable)
 
 EXC_REAL_BEGIN(fp_unavailable, 0x800, 0x100)
@@ -2372,6 +2377,7 @@ INT_DEFINE_BEGIN(altivec_unavailable)
 #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 	IKVM_REAL=1
 #endif
+	IMSR_R12=1
 INT_DEFINE_END(altivec_unavailable)
 
 EXC_REAL_BEGIN(altivec_unavailable, 0xf20, 0x20)
@@ -2421,6 +2427,7 @@ INT_DEFINE_BEGIN(vsx_unavailable)
 #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 	IKVM_REAL=1
 #endif
+	IMSR_R12=1
 INT_DEFINE_END(vsx_unavailable)
 
 EXC_REAL_BEGIN(vsx_unavailable, 0xf40, 0x20)
Signature · 1 line
2.37.2
reviewer Reviewed-by on v4 4/7
#6 lore raw

Re: [PATCH v4 4/7] powerpc/64s: IOption for MSR stored in r12

Quoted 8 lines · Rohan McLure

Interrupt handlers in asm/exceptions-64s.S contain a great deal of common code produced by the GEN_COMMON macros. Currently, at the exit point of the macro, r12 will contain the contents of the MSR. A future patch will cause these macros to zeroise architected registers to avoid potential speculation influence of user data.

Provide an IOption that signals that r12 must be retained, as the interrupt handler assumes it to hold the contents of the MSR.

Quoted 56 lines

---
v4: Split 64s register sanitisation commit to establish this IOption

---
 arch/powerpc/kernel/exceptions-64s.S | 7 +++++++
 1 file changed, 7 insertions(+)
arch/powerpc/kernel/exceptions-64s.S +7 -0
index 5381a43e50fe..58d72db1d484 100644
@@ -111,6 +111,7 @@ name:
 #define ISTACK		.L_ISTACK_\name\()	/* Set regular kernel stack */
 #define __ISTACK(name)	.L_ISTACK_ ## name
 #define IKUAP		.L_IKUAP_\name\()	/* Do KUAP lock */
+#define IMSR_R12	.L_IMSR_R12_\name\()	/* Assumes MSR saved to r12 */
 
 #define INT_DEFINE_BEGIN(n)						\
 .macro int_define_ ## n name
@@ -176,6 +177,9 @@ do_define_int n
 	.ifndef IKUAP
 		IKUAP=1
 	.endif
+	.ifndef IMSR_R12
+		IMSR_R12=0
+	.endif
 .endm
 
 /*
@@ -1751,6 +1755,7 @@ INT_DEFINE_BEGIN(fp_unavailable)
 #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 	IKVM_REAL=1
 #endif
+	IMSR_R12=1
 INT_DEFINE_END(fp_unavailable)
 
 EXC_REAL_BEGIN(fp_unavailable, 0x800, 0x100)
@@ -2372,6 +2377,7 @@ INT_DEFINE_BEGIN(altivec_unavailable)
 #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 	IKVM_REAL=1
 #endif
+	IMSR_R12=1
 INT_DEFINE_END(altivec_unavailable)
 
 EXC_REAL_BEGIN(altivec_unavailable, 0xf20, 0x20)
@@ -2421,6 +2427,7 @@ INT_DEFINE_BEGIN(vsx_unavailable)
 #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 	IKVM_REAL=1
 #endif
+	IMSR_R12=1
 INT_DEFINE_END(vsx_unavailable)
 
 EXC_REAL_BEGIN(vsx_unavailable, 0xf40, 0x20)
Signature · 1 line
2.37.2
#7 lore raw

PATCH v4 5/7 powerpc/64s: Zeroise gprs on interrupt routine entry on Book3S

In reply to Rohan McLure

Zeroise user state in gprs (assign to zero) to reduce the influence of user registers on speculation within kernel syscall handlers. Clears occur at the very beginning of the sc and scv 0 interrupt handlers, with restores occurring following the execution of the syscall handler.

Zeroise GPRS r0, r2-r11, r14-r31, on entry into the kernel for all other interrupt sources. The remaining gprs are overwritten by entry macros to interrupt handlers, irrespective of whether or not a given handler consumes these register values. If an interrupt does not select the IMSR_R12 IOption, zeroise r12.

Prior to this commit, r14-r31 are restored on a per-interrupt basis at exit, but now they are always restored on 64bit Book3S. Remove explicit REST_NVGPRS invocations on 64-bit Book3S. 32-bit systems do not clear user registers on interrupt, and continue to depend on the return value of interrupt_exit_user_prepare to determine whether or not to restore non-volatiles.

The mmap_bench benchmark in selftests should rapidly invoke pagefaults. See ~0.8% performance regression with this mitigation, but this indicates the worst-case performance due to heavier-weight interrupt handlers. This mitigation is able to be enabled/disabled through CONFIG_INTERRUPT_SANITIZE_REGISTERS.

---
v2: REST_NVGPRS should be conditional on mitigation in scv handler. Fix improper multi-line preprocessor macro in interrupt_64.S
v4: Split off IMSR_R12 definition into its own patch. Move macro definitions for register sanitisation into asm/ppc_asm.h

---
 arch/powerpc/kernel/exceptions-64s.S | 27 ++++++++++++++++++---------
 arch/powerpc/kernel/interrupt_64.S   | 16 ++++++++++++++--
 2 files changed, 32 insertions(+), 11 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +18 -9
index 58d72db1d484..8ee3db3b6595 100644
@@ -506,6 +506,7 @@ DEFINE_FIXED_SYMBOL(\name\()_common_real, text)
 	std	r10,0(r1)		/* make stack chain pointer	*/
 	std	r0,GPR0(r1)		/* save r0 in stackframe	*/
 	std	r10,GPR1(r1)		/* save r1 in stackframe	*/
+	ZEROIZE_GPR(0)
 
 	/* Mark our [H]SRRs valid for return */
 	li	r10,1
@@ -548,8 +549,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
 	std	r9,GPR11(r1)
 	std	r10,GPR12(r1)
 	std	r11,GPR13(r1)
+	.if !IMSR_R12
+	ZEROIZE_GPRS(9, 12)
+	.else
+	ZEROIZE_GPRS(9, 11)
+	.endif
 
 	SAVE_NVGPRS(r1)
+	SANITIZE_ZEROIZE_NVGPRS()
 
 	.if IDAR
 	.if IISIDE
@@ -581,8 +588,8 @@ BEGIN_FTR_SECTION
 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	*/
-	SAVE_GPRS(3, 8, r1)		/* save r3 - r8 in stackframe   */
+	SAVE_GPRS(2, 8, r1)		/* save r2 - r8 in stackframe   */
+	ZEROIZE_GPRS(2, 8)
 	mflr	r9			/* Get LR, later save to stack	*/
 	LOAD_PACA_TOC()			/* get kernel TOC into r2	*/
 	std	r9,_LINK(r1)
@@ -700,6 +707,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
 	mtlr	r9
 	ld	r9,_CCR(r1)
 	mtcr	r9
+	SANITIZE_RESTORE_NVGPRS()
 	REST_GPRS(2, 13, r1)
 	REST_GPR(0, r1)
 	/* restore original r1. */
@@ -1445,7 +1453,7 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
 	 * do_break() may have changed the NV GPRS while handling a breakpoint.
 	 * If so, we need to restore them with their updated values.
 	 */
-	REST_NVGPRS(r1)
+	HANDLER_RESTORE_NVGPRS()
 	b	interrupt_return_srr
 
 
@@ -1671,7 +1679,7 @@ 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 */
+	HANDLER_RESTORE_NVGPRS() /* instruction emulation may change GPRs */
 	b	interrupt_return_srr
 
 
@@ -1737,7 +1745,7 @@ EXC_COMMON_BEGIN(program_check_common)
 .Ldo_program_check:
 	addi	r3,r1,STACK_FRAME_OVERHEAD
 	bl	program_check_exception
-	REST_NVGPRS(r1) /* instruction emulation may change GPRs */
+	HANDLER_RESTORE_NVGPRS() /* instruction emulation may change GPRs */
 	b	interrupt_return_srr
 
 
@@ -2169,7 +2177,7 @@ 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 */
+	HANDLER_RESTORE_NVGPRS() /* instruction emulation may change GPRs */
 	b	interrupt_return_hsrr
 
 
@@ -2489,7 +2497,7 @@ EXC_COMMON_BEGIN(facility_unavailable_common)
 	GEN_COMMON facility_unavailable
 	addi	r3,r1,STACK_FRAME_OVERHEAD
 	bl	facility_unavailable_exception
-	REST_NVGPRS(r1) /* instruction emulation may change GPRs */
+	HANDLER_RESTORE_NVGPRS() /* instruction emulation may change GPRs */
 	b	interrupt_return_srr
 
 
@@ -2517,7 +2525,8 @@ EXC_COMMON_BEGIN(h_facility_unavailable_common)
 	GEN_COMMON h_facility_unavailable
 	addi	r3,r1,STACK_FRAME_OVERHEAD
 	bl	facility_unavailable_exception
-	REST_NVGPRS(r1) /* XXX Shouldn't be necessary in practice */
+	/* XXX Shouldn't be necessary in practice */
+	HANDLER_RESTORE_NVGPRS()
 	b	interrupt_return_hsrr
 
 
@@ -2743,7 +2752,7 @@ 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 */
+	HANDLER_RESTORE_NVGPRS() /* instruction emulation may change GPRs */
 #else
 	bl	unknown_exception
 #endif
arch/powerpc/kernel/interrupt_64.S +14 -2
index 1ef4fdef74fb..66a56659e8c8 100644
@@ -96,6 +96,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
 	 * but this is the best we can do.
 	 */
 
+	/*
+	 * Zero user registers to prevent influencing speculative execution
+	 * state of kernel code.
+	 */
+	SANITIZE_ZEROIZE_SYSCALL_GPRS()
 	bl	system_call_exception
 
 .Lsyscall_vectored_\name\()_exit:
@@ -124,6 +129,7 @@ BEGIN_FTR_SECTION
 	HMT_MEDIUM_LOW
 END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
 
+	SANITIZE_RESTORE_NVGPRS()
 	cmpdi	r3,0
 	bne	.Lsyscall_vectored_\name\()_restore_regs
 
@@ -159,7 +165,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
 	ld	r4,_LINK(r1)
 	ld	r5,_XER(r1)
 
-	REST_NVGPRS(r1)
+	HANDLER_RESTORE_NVGPRS()
 	REST_GPR(0, r1)
 	mtcr	r2
 	mtctr	r3
@@ -275,6 +281,11 @@ END_BTB_FLUSH_SECTION
 	wrteei	1
 #endif
 
+	/*
+	 * Zero user registers to prevent influencing speculative execution
+	 * state of kernel code.
+	 */
+	SANITIZE_ZEROIZE_SYSCALL_GPRS()
 	bl	system_call_exception
 
 .Lsyscall_exit:
@@ -315,6 +326,7 @@ BEGIN_FTR_SECTION
 	stdcx.	r0,0,r1			/* to clear the reservation */
 END_FTR_SECTION_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
 
+	SANITIZE_RESTORE_NVGPRS()
 	cmpdi	r3,0
 	bne	.Lsyscall_restore_regs
 	/* Zero volatile regs that may contain sensitive kernel data */
@@ -342,7 +354,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
 .Lsyscall_restore_regs:
 	ld	r3,_CTR(r1)
 	ld	r4,_XER(r1)
-	REST_NVGPRS(r1)
+	HANDLER_RESTORE_NVGPRS()
 	mtctr	r3
 	mtspr	SPRN_XER,r4
 	REST_GPR(0, r1)
Signature · 1 line
2.37.2
#8 lore raw

Re: [PATCH v4 5/7] powerpc/64s: Zeroise gprs on interrupt routine entry on Book3S

Quoted 49 lines · Rohan McLure

Zeroise user state in gprs (assign to zero) to reduce the influence of user registers on speculation within kernel syscall handlers. Clears occur at the very beginning of the sc and scv 0 interrupt handlers, with restores occurring following the execution of the syscall handler.

Zeroise GPRS r0, r2-r11, r14-r31, on entry into the kernel for all other interrupt sources. The remaining gprs are overwritten by entry macros to interrupt handlers, irrespective of whether or not a given handler consumes these register values. If an interrupt does not select the IMSR_R12 IOption, zeroise r12.

Prior to this commit, r14-r31 are restored on a per-interrupt basis at exit, but now they are always restored on 64bit Book3S. Remove explicit REST_NVGPRS invocations on 64-bit Book3S. 32-bit systems do not clear user registers on interrupt, and continue to depend on the return value of interrupt_exit_user_prepare to determine whether or not to restore non-volatiles.

The mmap_bench benchmark in selftests should rapidly invoke pagefaults. See ~0.8% performance regression with this mitigation, but this indicates the worst-case performance due to heavier-weight interrupt handlers. This mitigation is able to be enabled/disabled through CONFIG_INTERRUPT_SANITIZE_REGISTERS.

---
v2: REST_NVGPRS should be conditional on mitigation in scv handler. Fix improper multi-line preprocessor macro in interrupt_64.S
v4: Split off IMSR_R12 definition into its own patch. Move macro definitions for register sanitisation into asm/ppc_asm.h

---
 arch/powerpc/kernel/exceptions-64s.S | 27 ++++++++++++++++++---------
 arch/powerpc/kernel/interrupt_64.S   | 16 ++++++++++++++--
 2 files changed, 32 insertions(+), 11 deletions(-)
arch/powerpc/kernel/exceptions-64s.S +1 -0
index 58d72db1d484..8ee3db3b6595 100644
@@ -506,6 +506,7 @@ DEFINE_FIXED_SYMBOL(\name\()_common_real, text)
 	std	r10,0(r1)		/* make stack chain pointer	*/
 	std	r0,GPR0(r1)		/* save r0 in stackframe	*/
 	std	r10,GPR1(r1)		/* save r1 in stackframe	*/
+	ZEROIZE_GPR(0)
 
 	/* Mark our [H]SRRs valid for return */
 	li	r10,1

arch/powerpc/kernel/exceptions-64s.S END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) · line 556

Quoted 8 lines
arch/powerpc/kernel/exceptions-64s.S +5 -0
@@ -548,8 +549,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
 	std	r9,GPR11(r1)
 	std	r10,GPR12(r1)
 	std	r11,GPR13(r1)
+	.if !IMSR_R12
+	ZEROIZE_GPRS(9, 12)
+	.else
+	ZEROIZE_GPRS(9, 11)
+	.endif

These unconditionally zero. Should be SANITIZE_ZEROIZE_GPRS()? Same for a few more cases.

Hmm. r12 actually contains come-from-MSR here, which isn't really user controlled. r9-r11 just got loaded with some user GPRs, but they're the usual scratch registers and get used for other things later.

The whole interrupt entry file could probably use a spring clean, some re-scheduling, data layout improvements, and more consistent use of scratch registers... so I'm overly concerned about removing every possible redundant instruction here if it makes things a bit harder to follow. But we might be able to get rid of the above zeroize, with a comment.

Quoted 7 lines
arch/powerpc/kernel/exceptions-64s.S +1 -0
 	SAVE_NVGPRS(r1)
+	SANITIZE_ZEROIZE_NVGPRS()
 
 	.if IDAR
 	.if IISIDE

arch/powerpc/kernel/exceptions-64s.S BEGIN_FTR_SECTION · line 595

Quoted 10 lines
arch/powerpc/kernel/exceptions-64s.S +2 -2
@@ -581,8 +588,8 @@ BEGIN_FTR_SECTION
 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	*/
-	SAVE_GPRS(3, 8, r1)		/* save r3 - r8 in stackframe   */
+	SAVE_GPRS(2, 8, r1)		/* save r2 - r8 in stackframe   */
+	ZEROIZE_GPRS(2, 8)
 	mflr	r9			/* Get LR, later save to stack	*/
 	LOAD_PACA_TOC()			/* get kernel TOC into r2	*/
 	std	r9,_LINK(r1)

r2 gets zeroed then used again in LOAD_PACA_TOC too.

Otherwise looks pretty good... Although CTR doesn't get zeroed and I suppose it could be speculatively used as a source (e.g., bctr). CRs other than 0 and sometimes 1 too, they're probably a bit less important than CTR though. We don't use TAR in the kernel so that one should be okay to leave (maybe with a comment).

That can be done another time, I'd like to see the GPR sanitising patches go in... It *might* just be a matter of one mtctr in the case of !RELOCATABLE kernel though to get the ctr...

Thanks,
Nick

#9 lore raw

PATCH v4 6/7 powerpc/64e: Clear gprs on interrupt routine entry on Book3E

In reply to Rohan McLure

Zero GPRS r14-r31 on entry into the kernel for interrupt sources to limit influence of user-space values in potential speculation gadgets. Prior to this commit, all other GPRS are reassigned during the common prologue to interrupt handlers and so need not be zeroised explicitly.

This may be done safely, without loss of register state prior to the interrupt, as the common prologue saves the initial values of non-volatiles, which are unconditionally restored in interrupt_64.S. Mitigation defaults to enabled by INTERRUPT_SANITIZE_REGISTERS.

---
Resubmitting patches as their own series after v6 partially merged:

---
 arch/powerpc/kernel/exceptions-64e.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/kernel/exceptions-64e.S +2 -2
index 930e36099015..6b842d5ea4e1 100644
@@ -358,7 +358,6 @@ ret_from_mc_except:
 	std	r14,PACA_EXMC+EX_R14(r13);				    \
 	std	r15,PACA_EXMC+EX_R15(r13)
 
-
 /* Core exception code for all exceptions except TLB misses. */
 #define EXCEPTION_COMMON_LVL(n, scratch, excf)				    \
 exc_##n##_common:							    \
@@ -394,7 +393,8 @@ exc_##n##_common:							    \
 	std	r12,STACK_FRAME_OVERHEAD-16(r1); /* mark the frame */	    \
 	std	r3,_TRAP(r1);		/* set trap number		*/  \
 	std	r0,RESULT(r1);		/* clear regs->result */	    \
-	SAVE_NVGPRS(r1);
+	SAVE_NVGPRS(r1);						    \
+	SANITIZE_ZEROIZE_NVGPRS();	/* minimise speculation influence */
 
 #define EXCEPTION_COMMON(n) \
 	EXCEPTION_COMMON_LVL(n, SPRN_SPRG_GEN_SCRATCH, PACA_EXGEN)
Signature · 1 line
2.37.2
#10 lore raw

PATCH v4 7/7 powerpc/64: Sanitise user registers on interrupt in pseries, POWERNV

In reply to Rohan McLure

Cause pseries and POWERNV platforms to default to zeroising all potentially user-defined registers when entering the kernel by means of any interrupt source, reducing user-influence of the kernel and the likelihood or producing speculation gadgets.

---
Resubmitting patches as their own series after v6 partially merged:

v4: Default on POWERNV as well.

---
 arch/powerpc/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/Kconfig +1 -1
index 280c797e0f30..2ab114a02f62 100644
@@ -534,7 +534,7 @@ config HOTPLUG_CPU
 config INTERRUPT_SANITIZE_REGISTERS
 	bool "Clear gprs on interrupt arrival"
 	depends on PPC64 && ARCH_HAS_SYSCALL_WRAPPER
-	default PPC_BOOK3E_64
+	default PPC_BOOK3E_64 || PPC_PSERIES || PPC_POWERNV
 	help
 	  Reduce the influence of user register state on interrupt handlers and
 	  syscalls through clearing user state from registers before handling
Signature · 1 line
2.37.2
reviewer Acked-by on v4 7/7
#11 lore raw

Re: [PATCH v4 7/7] powerpc/64: Sanitise user registers on interrupt in pseries, POWERNV

Quoted 6 lines · Rohan McLure

Cause pseries and POWERNV platforms to default to zeroising all potentially user-defined registers when entering the kernel by means of any interrupt source, reducing user-influence of the kernel and the likelihood or producing speculation gadgets.

Quoted 23 lines

---
Resubmitting patches as their own series after v6 partially merged:

v4: Default on POWERNV as well.

---
 arch/powerpc/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/Kconfig +1 -1
index 280c797e0f30..2ab114a02f62 100644
@@ -534,7 +534,7 @@ config HOTPLUG_CPU
 config INTERRUPT_SANITIZE_REGISTERS
 	bool "Clear gprs on interrupt arrival"
 	depends on PPC64 && ARCH_HAS_SYSCALL_WRAPPER
-	default PPC_BOOK3E_64
+	default PPC_BOOK3E_64 || PPC_PSERIES || PPC_POWERNV
 	help
 	  Reduce the influence of user register state on interrupt handlers and
 	  syscalls through clearing user state from registers before handling
Signature · 1 line
2.37.2