Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1684989 > unrolled thread
| Started by | Mikulas Patocka <mpatocka@redhat.com> |
|---|---|
| First post | 2017-07-11 13:50 +0200 |
| Last post | 2017-07-16 11:30 +0200 |
| Articles | 5 — 3 participants |
Back to article view | Back to linux.kernel
[PATCH] x86/kernel/cpu/amd.c: use indirect call Mikulas Patocka <mpatocka@redhat.com> - 2017-07-11 13:50 +0200
[tip:x86/urgent] x86/cpu: Use indirect call to measure performance in init_amd_k6() tip-bot for Mikulas Patocka <tipbot@zytor.com> - 2017-07-12 15:20 +0200
Re: [tip:x86/urgent] x86/cpu: Use indirect call to measure performance in init_amd_k6() Borislav Petkov <bp@alien8.de> - 2017-07-12 15:50 +0200
Re: [tip:x86/urgent] x86/cpu: Use indirect call to measure performance in init_amd_k6() Mikulas Patocka <mpatocka@redhat.com> - 2017-07-13 04:30 +0200
[tip:x86/urgent] x86/cpu: Use indirect call to measure performance in init_amd_k6() tip-bot for Mikulas Patocka <tipbot@zytor.com> - 2017-07-16 11:30 +0200
| From | Mikulas Patocka <mpatocka@redhat.com> |
|---|---|
| Date | 2017-07-11 13:50 +0200 |
| Subject | [PATCH] x86/kernel/cpu/amd.c: use indirect call |
| Message-ID | <u21PX-1KO-11@gated-at.bofh.it> |
This old piece of code is supposed to measure the performance of indirect calls to determine if the processor is buggy or not, however the compiler optimizer turns it into a direct call. Use the macro OPTIMIZER_HIDE_VAR to thwart the optimization, so that real indirect call is generated. Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> --- arch/x86/kernel/cpu/amd.c | 1 + 1 file changed, 1 insertion(+) Index: linux-2.6/arch/x86/kernel/cpu/amd.c =================================================================== --- linux-2.6.orig/arch/x86/kernel/cpu/amd.c +++ linux-2.6/arch/x86/kernel/cpu/amd.c @@ -134,6 +134,7 @@ static void init_amd_k6(struct cpuinfo_x n = K6_BUG_LOOP; f_vide = vide; + OPTIMIZER_HIDE_VAR(f_vide); d = rdtsc(); while (n--) f_vide();
[toc] | [next] | [standalone]
| From | tip-bot for Mikulas Patocka <tipbot@zytor.com> |
|---|---|
| Date | 2017-07-12 15:20 +0200 |
| Subject | [tip:x86/urgent] x86/cpu: Use indirect call to measure performance in init_amd_k6() |
| Message-ID | <u2pID-g1-23@gated-at.bofh.it> |
| In reply to | #1684989 |
Commit-ID: 429ed86b5c99b388543f6f7f6983b1230326f2bb Gitweb: http://git.kernel.org/tip/429ed86b5c99b388543f6f7f6983b1230326f2bb Author: Mikulas Patocka <mpatocka@redhat.com> AuthorDate: Tue, 11 Jul 2017 07:44:05 -0400 Committer: Ingo Molnar <mingo@kernel.org> CommitDate: Wed, 12 Jul 2017 10:42:32 +0200 x86/cpu: Use indirect call to measure performance in init_amd_k6() This old piece of code is supposed to measure the performance of indirect calls to determine if the processor is buggy or not, however the compiler optimizer turns it into a direct call. Use the OPTIMIZER_HIDE_VAR() macro to thwart the optimization, so that a real indirect call is generated. Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/alpine.LRH.2.02.1707110737530.8746@file01.intranet.prod.int.rdu2.redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org> --- arch/x86/kernel/cpu/amd.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index bb5abe8..3b9e220 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -134,6 +134,7 @@ static void init_amd_k6(struct cpuinfo_x86 *c) n = K6_BUG_LOOP; f_vide = vide; + OPTIMIZER_HIDE_VAR(f_vide); d = rdtsc(); while (n--) f_vide();
[toc] | [prev] | [next] | [standalone]
| From | Borislav Petkov <bp@alien8.de> |
|---|---|
| Date | 2017-07-12 15:50 +0200 |
| Subject | Re: [tip:x86/urgent] x86/cpu: Use indirect call to measure performance in init_amd_k6() |
| Message-ID | <u2qbE-pT-13@gated-at.bofh.it> |
| In reply to | #1685772 |
On Wed, Jul 12, 2017 at 06:09:57AM -0700, tip-bot for Mikulas Patocka wrote:
> Commit-ID: 429ed86b5c99b388543f6f7f6983b1230326f2bb
> Gitweb: http://git.kernel.org/tip/429ed86b5c99b388543f6f7f6983b1230326f2bb
> Author: Mikulas Patocka <mpatocka@redhat.com>
> AuthorDate: Tue, 11 Jul 2017 07:44:05 -0400
> Committer: Ingo Molnar <mingo@kernel.org>
> CommitDate: Wed, 12 Jul 2017 10:42:32 +0200
>
> x86/cpu: Use indirect call to measure performance in init_amd_k6()
>
> This old piece of code is supposed to measure the performance of indirect
> calls to determine if the processor is buggy or not, however the compiler
> optimizer turns it into a direct call.
Are you booting on a real K6 or qemu?
--
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
--
[toc] | [prev] | [next] | [standalone]
| From | Mikulas Patocka <mpatocka@redhat.com> |
|---|---|
| Date | 2017-07-13 04:30 +0200 |
| Subject | Re: [tip:x86/urgent] x86/cpu: Use indirect call to measure performance in init_amd_k6() |
| Message-ID | <u2C38-84q-33@gated-at.bofh.it> |
| In reply to | #1685785 |
On Wed, 12 Jul 2017, Borislav Petkov wrote: > On Wed, Jul 12, 2017 at 06:09:57AM -0700, tip-bot for Mikulas Patocka wrote: > > Commit-ID: 429ed86b5c99b388543f6f7f6983b1230326f2bb > > Gitweb: http://git.kernel.org/tip/429ed86b5c99b388543f6f7f6983b1230326f2bb > > Author: Mikulas Patocka <mpatocka@redhat.com> > > AuthorDate: Tue, 11 Jul 2017 07:44:05 -0400 > > Committer: Ingo Molnar <mingo@kernel.org> > > CommitDate: Wed, 12 Jul 2017 10:42:32 +0200 > > > > x86/cpu: Use indirect call to measure performance in init_amd_k6() > > > > This old piece of code is supposed to measure the performance of indirect > > calls to determine if the processor is buggy or not, however the compiler > > optimizer turns it into a direct call. > > Are you booting on a real K6 or qemu? I tested it on K6-3 and it doesn't trigger the warning, the loop with the indirect call takes 9 million cycles (it took 4 million cycles with the direct call). I don't have the old K6 revision with the bug. Mikulas > -- > Regards/Gruss, > Boris. > > ECO tip #101: Trim your mails when you reply. > --
[toc] | [prev] | [next] | [standalone]
| From | tip-bot for Mikulas Patocka <tipbot@zytor.com> |
|---|---|
| Date | 2017-07-16 11:30 +0200 |
| Subject | [tip:x86/urgent] x86/cpu: Use indirect call to measure performance in init_amd_k6() |
| Message-ID | <u3O2e-5hz-7@gated-at.bofh.it> |
| In reply to | #1684989 |
Commit-ID: 5f8a16156aa1b2d0223eaee9dacdfb9bc096f610 Gitweb: http://git.kernel.org/tip/5f8a16156aa1b2d0223eaee9dacdfb9bc096f610 Author: Mikulas Patocka <mpatocka@redhat.com> AuthorDate: Tue, 11 Jul 2017 07:44:05 -0400 Committer: Ingo Molnar <mingo@kernel.org> CommitDate: Sun, 16 Jul 2017 11:05:04 +0200 x86/cpu: Use indirect call to measure performance in init_amd_k6() This old piece of code is supposed to measure the performance of indirect calls to determine if the processor is buggy or not, however the compiler optimizer turns it into a direct call. Use the OPTIMIZER_HIDE_VAR() macro to thwart the optimization, so that a real indirect call is generated. Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/alpine.LRH.2.02.1707110737530.8746@file01.intranet.prod.int.rdu2.redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org> --- arch/x86/kernel/cpu/amd.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index bb5abe8..3b9e220 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -134,6 +134,7 @@ static void init_amd_k6(struct cpuinfo_x86 *c) n = K6_BUG_LOOP; f_vide = vide; + OPTIMIZER_HIDE_VAR(f_vide); d = rdtsc(); while (n--) f_vide();
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web