call stop_machine_text_poke() on all cpus
PATCH 1/1 call stop_machine_text_poke() on all cpus
Currently, text_poke_smp() passes a NULL as the third argument to __stop_machine(), which will only run stop_machine_text_poke() on 1 cpu. Change NULL -> cpu_online_mask, as stop_machine_text_poke() is intended to be run on all cpus.
I actually didn't notice any problems with stop_machine_text_poke() only being called on 1 cpu, but found this via code inspection.
--- arch/x86/kernel/alternative.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
arch/x86/kernel/alternative.c +1 -1
@@ -705,6 +705,6 @@ void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len)atomic_set (&stop_machine_first,1 ); wrote_text =0 ;/* Use __stop_machine() because the caller already got online_cpus. */ -__stop_machine (stop_machine_text_poke, (void *)&tpp,NULL ); +__stop_machine (stop_machine_text_poke, (void *)&tpp, cpu_online_mask);return addr; }
Signature · 1 line
1.7.1
Quoted 9 lines · * Jason Baron ([email protected]) wrote:
Currently,
text_poke_smp()passes a NULL as the third argument to__stop_machine(), which will only runstop_machine_text_poke()on 1 cpu. Change NULL ->cpu_online_mask, asstop_machine_text_poke()is intended to be run on all cpus.I actually didn't notice any problems with
stop_machine_text_poke()only being called on 1 cpu, but found this via code inspection.Signed-off-by Jason Baron <[email protected]>
Good catch!
The side-effect of this is that the other CPUs are not executing the memory barrier and not flushing the icache, as we would normally expect.
Thanks,
Quoted 19 lines
--- arch/x86/kernel/alternative.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)arch/x86/kernel/alternative.c +1
-1@@ -705,6 +705,6 @@ void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len)atomic_set (&stop_machine_first,1 ); wrote_text =0 ;/* Use __stop_machine() because the caller already got online_cpus. */ -__stop_machine (stop_machine_text_poke, (void *)&tpp,NULL ); +__stop_machine (stop_machine_text_poke, (void *)&tpp, cpu_online_mask);return addr; }Signature · 1 line
1.7.1
Signature · 4 lines
Mathieu Desnoyers Operating System Efficiency R&D Consultant EfficiOS Inc. http://www.efficios.com
Quoted 4 lines · Jason Baron
Currently,
text_poke_smp()passes a NULL as the third argument to__stop_machine(), which will only runstop_machine_text_poke()on 1 cpu. Change NULL ->cpu_online_mask, asstop_machine_text_poke()is intended to be run on all cpus.
That should be my misunderstand...
Quoted 3 lines
I actually didn't notice any problems with
stop_machine_text_poke()only being called on 1 cpu, but found this via code inspection.
Thank you very much!
Quoted 18 lines
Signed-off-by Jason Baron <[email protected]>--- arch/x86/kernel/alternative.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)arch/x86/kernel/alternative.c +1
-1@@ -705,6 +705,6 @@ void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len)atomic_set (&stop_machine_first,1 ); wrote_text =0 ;/* Use __stop_machine() because the caller already got online_cpus. */ -__stop_machine (stop_machine_text_poke, (void *)&tpp,NULL ); +__stop_machine (stop_machine_text_poke, (void *)&tpp, cpu_online_mask);return addr; }
Signature · 4 lines
Masami HIRAMATSU 2nd Dept. Linux Technology Center Hitachi, Ltd., Systems Development Laboratory E-mail: [email protected]
[tip:x86/urgent] x86, alternative: Call stop_machine_text_poke() on all cpus
In reply to Jason BaronCommit-ID: 404ba5d7bb958d3d788bdaa0debc0bdf60f13ffe Gitweb: http://git.kernel.org/tip/404ba5d7bb958d3d788bdaa0debc0bdf60f13ffe Author: Jason Baron <[email protected]> AuthorDate: Thu, 28 Oct 2010 11:20:27 -0400 Committer: H. Peter Anvin <[email protected]> CommitDate: Fri, 29 Oct 2010 16:42:58 -0700
x86, alternative: Call stop_machine_text_poke() on all cpus
Currently, text_poke_smp() passes a NULL as the third argument to __stop_machine(), which will only run stop_machine_text_poke() on 1 cpu. Change NULL -> cpu_online_mask, as stop_machine_text_poke() is intended to be run on all cpus.
I actually didn't notice any problems with stop_machine_text_poke() only being called on 1 cpu, but found this via code inspection.
LKML-Reference: <[email protected]>
--- arch/x86/kernel/alternative.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
arch/x86/kernel/alternative.c +1 -1
@@ -638,7 +638,7 @@ void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len)atomic_set (&stop_machine_first,1 ); wrote_text =0 ;/* Use __stop_machine() because the caller already got online_cpus. */ -__stop_machine (stop_machine_text_poke, (void *)&tpp,NULL ); +__stop_machine (stop_machine_text_poke, (void *)&tpp, cpu_online_mask);return addr; }