x86: Fix code patching for paravirt-alternatives on 486
PATCH v2 1/1 x86: Fix code patching for paravirt-alternatives on 486
As reported in <http://bugs.debian.org/511703> and
<http://bugs.debian.org/515982>, kernels with paravirt-alternatives enabled crash in text_poke_early() on at least some 486-class processors.
The problem is that text_poke_early() itself uses inline functions affected by paravirt-alternatives and so will modify instructions that have already been prefetched. Pentium and later processors will invalidate the prefetched instructions in this case, but 486-class processors do not.
Change sync_core() to limit prefetching on 486-class (and 386-class) processors, and move the call to sync_core() above the call to the modifiable local_irq_restore().
---
Second try, incorporating the jmp into sync_core().
Also not signed as I know git has trouble with MIME.
Ben.
arch/x86/include/asm/processor.h | 16 +++++++++++++--- arch/x86/kernel/alternative.c | 2 +- 2 files changed, 14 insertions(+), 4 deletions(-)
arch/x86/include/asm/processor.h +13 -3
@@ -703,13 +703,23 @@ static inline void cpu_relax(void)rep_nop (); } -/* Stop speculative execution: */ +/* Stop speculative execution and prefetching of modified code. */ static inline void sync_core (void ) {int tmp; - asmvolatile ("cpuid" :"=a" (tmp) :"0" (1 ) - :"ebx" ,"ecx" ,"edx" ,"memory" ); +#if defined(CONFIG_M386 ) || defined(CONFIG_M486 ) +if (boot_cpu_data.x86 <5 ) +/* There is no speculative execution. +* jmp is a barrier to prefetching. */ + asmvolatile ("jmp 1f\n1:\n" :::"memory" ); + else +#endif +/* cpuid is a barrier to speculative execution. +* Prefetched instructions are automatically +* invalidated when modified. */ + asmvolatile ("cpuid" :"=a" (tmp) :"0" (1 ) + :"ebx" ,"ecx" ,"edx" ,"memory" ); } static inline void __monitor(const void *eax, unsigned long ecx,
arch/x86/kernel/alternative.c +1 -1
@@ -498,8 +498,8 @@ static void *__init_or_module text_poke_early(void *addr, const void *opcode,unsigned long flags;local_irq_save (flags);memcpy (addr, opcode, len); -local_irq_restore (flags);sync_core (); +local_irq_restore (flags);/* Could also do a CLFLUSH here to speed up CPU recovery; but that causes hangs on some VIA CPUs. */ return addr;
Signature · 1 line
1.6.3.3
Commit-ID: 5367b6887e7d8c870a5da7d9b8c6e9c207684e43 Gitweb: http://git.kernel.org/tip/5367b6887e7d8c870a5da7d9b8c6e9c207684e43 Author: Ben Hutchings <[email protected]> AuthorDate: Thu, 10 Sep 2009 02:53:50 +0100 Committer: H. Peter Anvin <[email protected]> CommitDate: Thu, 10 Sep 2009 16:50:19 -0700
x86: Fix code patching for paravirt-alternatives on 486
As reported in <http://bugs.debian.org/511703> and
<http://bugs.debian.org/515982>, kernels with paravirt-alternatives enabled crash in text_poke_early() on at least some 486-class processors.
The problem is that text_poke_early() itself uses inline functions affected by paravirt-alternatives and so will modify instructions that have already been prefetched. Pentium and later processors will invalidate the prefetched instructions in this case, but 486-class processors do not.
Change sync_core() to limit prefetching on 486-class (and 386-class) processors, and move the call to sync_core() above the call to the modifiable local_irq_restore().
LKML-Reference: <1252547631.3423.134.camel@localhost>
--- arch/x86/include/asm/processor.h | 16 +++++++++++++--- arch/x86/kernel/alternative.c | 2 +- 2 files changed, 14 insertions(+), 4 deletions(-)
arch/x86/include/asm/processor.h +13 -3
@@ -703,13 +703,23 @@ static inline void cpu_relax(void)rep_nop (); } -/* Stop speculative execution: */ +/* Stop speculative execution and prefetching of modified code. */ static inline void sync_core (void ) {int tmp; - asmvolatile ("cpuid" :"=a" (tmp) :"0" (1 ) - :"ebx" ,"ecx" ,"edx" ,"memory" ); +#if defined(CONFIG_M386 ) || defined(CONFIG_M486 ) +if (boot_cpu_data.x86 <5 ) +/* There is no speculative execution. +* jmp is a barrier to prefetching. */ + asmvolatile ("jmp 1f\n1:\n" :::"memory" ); + else +#endif +/* cpuid is a barrier to speculative execution. +* Prefetched instructions are automatically +* invalidated when modified. */ + asmvolatile ("cpuid" :"=a" (tmp) :"0" (1 ) + :"ebx" ,"ecx" ,"edx" ,"memory" ); } static inline void __monitor(const void *eax, unsigned long ecx,
arch/x86/kernel/alternative.c +1 -1
@@ -490,8 +490,8 @@ void *text_poke_early(void *addr, const void *opcode, size_t len)unsigned long flags;local_irq_save (flags);memcpy (addr, opcode, len); -local_irq_restore (flags);sync_core (); +local_irq_restore (flags);/* Could also do a CLFLUSH here to speed up CPU recovery; but that causes hangs on some VIA CPUs. */ return addr;