* [PATCH 1/6] x86/fpu/regset: Use boot_cpu_has()
2016-04-05 6:29 [PATCH 0/6] x86: Misc cleanups Borislav Petkov
@ 2016-04-05 6:29 ` Borislav Petkov
2016-04-05 13:51 ` Brian Gerst
2016-04-13 11:41 ` [tip:x86/asm] x86/fpu/regset: Replace static_cpu_has() usage with boot_cpu_has() tip-bot for Borislav Petkov
2016-04-05 6:29 ` [PATCH 2/6] x86/cpu: Simplify extended APIC ID detection on AMD Borislav Petkov
` (4 subsequent siblings)
5 siblings, 2 replies; 17+ messages in thread
From: Borislav Petkov @ 2016-04-05 6:29 UTC (permalink / raw)
To: X86 ML; +Cc: LKML
From: Borislav Petkov <bp@suse.de>
fpregs_{g,s}et() are not sizzling-hot paths to justify the need for
static_cpu_has(). Use the normal boot_cpu_has() helper.
Signed-off-by: Borislav Petkov <bp@suse.de>
---
arch/x86/kernel/fpu/regset.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/fpu/regset.c b/arch/x86/kernel/fpu/regset.c
index bc5e76c1d7c5..81422dfb152b 100644
--- a/arch/x86/kernel/fpu/regset.c
+++ b/arch/x86/kernel/fpu/regset.c
@@ -278,7 +278,7 @@ int fpregs_get(struct task_struct *target, const struct user_regset *regset,
fpu__activate_fpstate_read(fpu);
- if (!static_cpu_has(X86_FEATURE_FPU))
+ if (!boot_cpu_has(X86_FEATURE_FPU))
return fpregs_soft_get(target, regset, pos, count, kbuf, ubuf);
if (!boot_cpu_has(X86_FEATURE_FXSR))
@@ -309,7 +309,7 @@ int fpregs_set(struct task_struct *target, const struct user_regset *regset,
fpu__activate_fpstate_write(fpu);
fpstate_sanitize_xstate(fpu);
- if (!static_cpu_has(X86_FEATURE_FPU))
+ if (!boot_cpu_has(X86_FEATURE_FPU))
return fpregs_soft_set(target, regset, pos, count, kbuf, ubuf);
if (!boot_cpu_has(X86_FEATURE_FXSR))
--
2.7.3
^ permalink raw reply related [flat|nested] 17+ messages in thread* Re: [PATCH 1/6] x86/fpu/regset: Use boot_cpu_has()
2016-04-05 6:29 ` [PATCH 1/6] x86/fpu/regset: Use boot_cpu_has() Borislav Petkov
@ 2016-04-05 13:51 ` Brian Gerst
2016-04-05 15:11 ` Borislav Petkov
2016-04-13 11:41 ` [tip:x86/asm] x86/fpu/regset: Replace static_cpu_has() usage with boot_cpu_has() tip-bot for Borislav Petkov
1 sibling, 1 reply; 17+ messages in thread
From: Brian Gerst @ 2016-04-05 13:51 UTC (permalink / raw)
To: Borislav Petkov; +Cc: X86 ML, LKML
On Tue, Apr 5, 2016 at 2:29 AM, Borislav Petkov <bp@alien8.de> wrote:
> From: Borislav Petkov <bp@suse.de>
>
> fpregs_{g,s}et() are not sizzling-hot paths to justify the need for
> static_cpu_has(). Use the normal boot_cpu_has() helper.
>
> Signed-off-by: Borislav Petkov <bp@suse.de>
> ---
> arch/x86/kernel/fpu/regset.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/kernel/fpu/regset.c b/arch/x86/kernel/fpu/regset.c
> index bc5e76c1d7c5..81422dfb152b 100644
> --- a/arch/x86/kernel/fpu/regset.c
> +++ b/arch/x86/kernel/fpu/regset.c
> @@ -278,7 +278,7 @@ int fpregs_get(struct task_struct *target, const struct user_regset *regset,
>
> fpu__activate_fpstate_read(fpu);
>
> - if (!static_cpu_has(X86_FEATURE_FPU))
> + if (!boot_cpu_has(X86_FEATURE_FPU))
> return fpregs_soft_get(target, regset, pos, count, kbuf, ubuf);
>
> if (!boot_cpu_has(X86_FEATURE_FXSR))
> @@ -309,7 +309,7 @@ int fpregs_set(struct task_struct *target, const struct user_regset *regset,
> fpu__activate_fpstate_write(fpu);
> fpstate_sanitize_xstate(fpu);
>
> - if (!static_cpu_has(X86_FEATURE_FPU))
> + if (!boot_cpu_has(X86_FEATURE_FPU))
> return fpregs_soft_set(target, regset, pos, count, kbuf, ubuf);
>
> if (!boot_cpu_has(X86_FEATURE_FXSR))
I think the general rule should be always use static_cpu_has() for
runtime checks, since it reduces down to a single jmp/nop instruction
after alternatives run. Even if it's not a hot path, it saves a bit
of runtime memory. boot_cpu_has() is fine for run-once init code.
--
Brian Gerst
^ permalink raw reply [flat|nested] 17+ messages in thread* Re: [PATCH 1/6] x86/fpu/regset: Use boot_cpu_has()
2016-04-05 13:51 ` Brian Gerst
@ 2016-04-05 15:11 ` Borislav Petkov
0 siblings, 0 replies; 17+ messages in thread
From: Borislav Petkov @ 2016-04-05 15:11 UTC (permalink / raw)
To: Brian Gerst; +Cc: X86 ML, LKML
On Tue, Apr 05, 2016 at 09:51:55AM -0400, Brian Gerst wrote:
> I think the general rule should be always use static_cpu_has() for
> runtime checks, since it reduces down to a single jmp/nop instruction
> after alternatives run. Even if it's not a hot path, it saves a bit
> of runtime memory.
So gcc generates a MOV and a TEST for boot_cpu_has(). Not exactly the
bloat or slowdown.
So I look at those differently: static_cpu_has()
a) should be used on hotpaths
b) is used as such to annotate those - i.e., "ah, there's a
static_cpu_has(), this is a hot path, I better be careful what I'm doing
here."
boot_cpu_has() in all the rest.
And saving a couple of bytes per call site but increasing alternatives
application time by milli- or microseconds... Meh. I don't see it being
really worth the trouble. But this is just me.
--
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
^ permalink raw reply [flat|nested] 17+ messages in thread
* [tip:x86/asm] x86/fpu/regset: Replace static_cpu_has() usage with boot_cpu_has()
2016-04-05 6:29 ` [PATCH 1/6] x86/fpu/regset: Use boot_cpu_has() Borislav Petkov
2016-04-05 13:51 ` Brian Gerst
@ 2016-04-13 11:41 ` tip-bot for Borislav Petkov
1 sibling, 0 replies; 17+ messages in thread
From: tip-bot for Borislav Petkov @ 2016-04-13 11:41 UTC (permalink / raw)
To: linux-tip-commits
Cc: hpa, mingo, torvalds, bp, luto, brgerst, bp, linux-kernel,
dvlasenk, peterz, tglx
Commit-ID: 78df526c74a4db696e1e058b9869471937d0773b
Gitweb: http://git.kernel.org/tip/78df526c74a4db696e1e058b9869471937d0773b
Author: Borislav Petkov <bp@suse.de>
AuthorDate: Tue, 5 Apr 2016 08:29:50 +0200
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 13 Apr 2016 11:37:42 +0200
x86/fpu/regset: Replace static_cpu_has() usage with boot_cpu_has()
fpregs_{g,s}et() are not sizzling-hot paths to justify the need for
static_cpu_has(). Use the normal boot_cpu_has() helper.
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Andy Lutomirski <luto@amacapital.net>
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: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1459837795-2588-2-git-send-email-bp@alien8.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
arch/x86/kernel/fpu/regset.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/fpu/regset.c b/arch/x86/kernel/fpu/regset.c
index bc5e76c..81422df 100644
--- a/arch/x86/kernel/fpu/regset.c
+++ b/arch/x86/kernel/fpu/regset.c
@@ -278,7 +278,7 @@ int fpregs_get(struct task_struct *target, const struct user_regset *regset,
fpu__activate_fpstate_read(fpu);
- if (!static_cpu_has(X86_FEATURE_FPU))
+ if (!boot_cpu_has(X86_FEATURE_FPU))
return fpregs_soft_get(target, regset, pos, count, kbuf, ubuf);
if (!boot_cpu_has(X86_FEATURE_FXSR))
@@ -309,7 +309,7 @@ int fpregs_set(struct task_struct *target, const struct user_regset *regset,
fpu__activate_fpstate_write(fpu);
fpstate_sanitize_xstate(fpu);
- if (!static_cpu_has(X86_FEATURE_FPU))
+ if (!boot_cpu_has(X86_FEATURE_FPU))
return fpregs_soft_set(target, regset, pos, count, kbuf, ubuf);
if (!boot_cpu_has(X86_FEATURE_FXSR))
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 2/6] x86/cpu: Simplify extended APIC ID detection on AMD
2016-04-05 6:29 [PATCH 0/6] x86: Misc cleanups Borislav Petkov
2016-04-05 6:29 ` [PATCH 1/6] x86/fpu/regset: Use boot_cpu_has() Borislav Petkov
@ 2016-04-05 6:29 ` Borislav Petkov
2016-04-13 11:41 ` [tip:x86/asm] " tip-bot for Borislav Petkov
2016-04-05 6:29 ` [PATCH 3/6] x86/tsc: Do not check X86_FEATURE_CONSTANT_TSC in notifier call Borislav Petkov
` (3 subsequent siblings)
5 siblings, 1 reply; 17+ messages in thread
From: Borislav Petkov @ 2016-04-05 6:29 UTC (permalink / raw)
To: X86 ML; +Cc: LKML
From: Borislav Petkov <bp@suse.de>
Both if-branches are under if (boot_cpu_has(X86_FEATURE_APIC)), unify
them.
Also, simplify the test for bits:
- 17 ("ApicExtBrdCst: APIC extended broadcast enable") and
- 18 ("ApicExtId: APIC extended ID enable.")
in "D18F0x68 Link Transaction Control."
No functionality change.
Signed-off-by: Borislav Petkov <bp@suse.de>
---
arch/x86/kernel/cpu/amd.c | 17 ++++++++++-------
1 file changed, 10 insertions(+), 7 deletions(-)
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 140eb4351a7b..2a380d7606d7 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -565,14 +565,17 @@ static void early_init_amd(struct cpuinfo_x86 *c)
* can safely set X86_FEATURE_EXTD_APICID unconditionally for families
* after 16h.
*/
- if (boot_cpu_has(X86_FEATURE_APIC) && c->x86 > 0x16) {
- set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
- } else if (boot_cpu_has(X86_FEATURE_APIC) && c->x86 >= 0xf) {
- /* check CPU config space for extended APIC ID */
- unsigned int val;
- val = read_pci_config(0, 24, 0, 0x68);
- if ((val & ((1 << 17) | (1 << 18))) == ((1 << 17) | (1 << 18)))
+ if (boot_cpu_has(X86_FEATURE_APIC)) {
+ if (c->x86 > 0x16)
set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
+ else if (c->x86 >= 0xf) {
+ /* check CPU config space for extended APIC ID */
+ unsigned int val;
+
+ val = read_pci_config(0, 24, 0, 0x68);
+ if ((val >> 17 & 0x3) == 0x3)
+ set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
+ }
}
#endif
--
2.7.3
^ permalink raw reply related [flat|nested] 17+ messages in thread* [tip:x86/asm] x86/cpu: Simplify extended APIC ID detection on AMD
2016-04-05 6:29 ` [PATCH 2/6] x86/cpu: Simplify extended APIC ID detection on AMD Borislav Petkov
@ 2016-04-13 11:41 ` tip-bot for Borislav Petkov
0 siblings, 0 replies; 17+ messages in thread
From: tip-bot for Borislav Petkov @ 2016-04-13 11:41 UTC (permalink / raw)
To: linux-tip-commits
Cc: mingo, tglx, brgerst, bp, linux-kernel, torvalds, dvlasenk, hpa,
bp, peterz, luto
Commit-ID: 425d8c2fc5e6dddbad083502bb77c7beae545620
Gitweb: http://git.kernel.org/tip/425d8c2fc5e6dddbad083502bb77c7beae545620
Author: Borislav Petkov <bp@suse.de>
AuthorDate: Tue, 5 Apr 2016 08:29:51 +0200
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 13 Apr 2016 11:37:42 +0200
x86/cpu: Simplify extended APIC ID detection on AMD
Both if-branches are under if (boot_cpu_has(X86_FEATURE_APIC)), unify
them.
Also, simplify the test for bits:
- 17 ("ApicExtBrdCst: APIC extended broadcast enable") and
- 18 ("ApicExtId: APIC extended ID enable.")
in "D18F0x68 Link Transaction Control."
No functionality change.
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Andy Lutomirski <luto@amacapital.net>
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: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1459837795-2588-3-git-send-email-bp@alien8.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
arch/x86/kernel/cpu/amd.c | 17 ++++++++++-------
1 file changed, 10 insertions(+), 7 deletions(-)
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 54f7b44..c343a54 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -565,14 +565,17 @@ static void early_init_amd(struct cpuinfo_x86 *c)
* can safely set X86_FEATURE_EXTD_APICID unconditionally for families
* after 16h.
*/
- if (boot_cpu_has(X86_FEATURE_APIC) && c->x86 > 0x16) {
- set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
- } else if (boot_cpu_has(X86_FEATURE_APIC) && c->x86 >= 0xf) {
- /* check CPU config space for extended APIC ID */
- unsigned int val;
- val = read_pci_config(0, 24, 0, 0x68);
- if ((val & ((1 << 17) | (1 << 18))) == ((1 << 17) | (1 << 18)))
+ if (boot_cpu_has(X86_FEATURE_APIC)) {
+ if (c->x86 > 0x16)
set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
+ else if (c->x86 >= 0xf) {
+ /* check CPU config space for extended APIC ID */
+ unsigned int val;
+
+ val = read_pci_config(0, 24, 0, 0x68);
+ if ((val >> 17 & 0x3) == 0x3)
+ set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
+ }
}
#endif
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 3/6] x86/tsc: Do not check X86_FEATURE_CONSTANT_TSC in notifier call
2016-04-05 6:29 [PATCH 0/6] x86: Misc cleanups Borislav Petkov
2016-04-05 6:29 ` [PATCH 1/6] x86/fpu/regset: Use boot_cpu_has() Borislav Petkov
2016-04-05 6:29 ` [PATCH 2/6] x86/cpu: Simplify extended APIC ID detection on AMD Borislav Petkov
@ 2016-04-05 6:29 ` Borislav Petkov
2016-04-05 8:00 ` Peter Zijlstra
2016-04-13 11:41 ` [tip:x86/asm] " tip-bot for Borislav Petkov
2016-04-05 6:29 ` [PATCH 4/6] x86/tsc: Save an indentation level in recalibrate_cpu_khz() Borislav Petkov
` (2 subsequent siblings)
5 siblings, 2 replies; 17+ messages in thread
From: Borislav Petkov @ 2016-04-05 6:29 UTC (permalink / raw)
To: X86 ML; +Cc: LKML, Thomas Gleixner, Peter Zijlstra
From: Borislav Petkov <bp@suse.de>
... because the notifier-registering routine already does that. Also,
rename cpufreq_tsc() init call to something more telling.
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
---
arch/x86/kernel/tsc.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index a0346bc51833..5bb702c77e8f 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -922,9 +922,6 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
struct cpufreq_freqs *freq = data;
unsigned long *lpj;
- if (cpu_has(&cpu_data(freq->cpu), X86_FEATURE_CONSTANT_TSC))
- return 0;
-
lpj = &boot_cpu_data.loops_per_jiffy;
#ifdef CONFIG_SMP
if (!(freq->flags & CPUFREQ_CONST_LOOPS))
@@ -954,7 +951,7 @@ static struct notifier_block time_cpufreq_notifier_block = {
.notifier_call = time_cpufreq_notifier
};
-static int __init cpufreq_tsc(void)
+static int __init cpufreq_register_tsc_scaling(void)
{
if (!boot_cpu_has(X86_FEATURE_TSC))
return 0;
@@ -965,7 +962,7 @@ static int __init cpufreq_tsc(void)
return 0;
}
-core_initcall(cpufreq_tsc);
+core_initcall(cpufreq_register_tsc_scaling);
#endif /* CONFIG_CPU_FREQ */
--
2.7.3
^ permalink raw reply related [flat|nested] 17+ messages in thread* Re: [PATCH 3/6] x86/tsc: Do not check X86_FEATURE_CONSTANT_TSC in notifier call
2016-04-05 6:29 ` [PATCH 3/6] x86/tsc: Do not check X86_FEATURE_CONSTANT_TSC in notifier call Borislav Petkov
@ 2016-04-05 8:00 ` Peter Zijlstra
2016-04-13 11:41 ` [tip:x86/asm] " tip-bot for Borislav Petkov
1 sibling, 0 replies; 17+ messages in thread
From: Peter Zijlstra @ 2016-04-05 8:00 UTC (permalink / raw)
To: Borislav Petkov; +Cc: X86 ML, LKML, Thomas Gleixner
On Tue, Apr 05, 2016 at 08:29:52AM +0200, Borislav Petkov wrote:
> From: Borislav Petkov <bp@suse.de>
>
> ... because the notifier-registering routine already does that. Also,
> rename cpufreq_tsc() init call to something more telling.
>
> Signed-off-by: Borislav Petkov <bp@suse.de>
> Cc: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
^ permalink raw reply [flat|nested] 17+ messages in thread
* [tip:x86/asm] x86/tsc: Do not check X86_FEATURE_CONSTANT_TSC in notifier call
2016-04-05 6:29 ` [PATCH 3/6] x86/tsc: Do not check X86_FEATURE_CONSTANT_TSC in notifier call Borislav Petkov
2016-04-05 8:00 ` Peter Zijlstra
@ 2016-04-13 11:41 ` tip-bot for Borislav Petkov
1 sibling, 0 replies; 17+ messages in thread
From: tip-bot for Borislav Petkov @ 2016-04-13 11:41 UTC (permalink / raw)
To: linux-tip-commits
Cc: bp, torvalds, mingo, bp, peterz, brgerst, dvlasenk, hpa, tglx,
linux-kernel, luto
Commit-ID: a841cca74ea7612508aee161c89987b2646ed769
Gitweb: http://git.kernel.org/tip/a841cca74ea7612508aee161c89987b2646ed769
Author: Borislav Petkov <bp@suse.de>
AuthorDate: Tue, 5 Apr 2016 08:29:52 +0200
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 13 Apr 2016 11:37:43 +0200
x86/tsc: Do not check X86_FEATURE_CONSTANT_TSC in notifier call
... because the notifier-registering routine already does that. Also,
rename cpufreq_tsc() init call to something more telling.
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Andy Lutomirski <luto@amacapital.net>
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: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1459837795-2588-4-git-send-email-bp@alien8.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
arch/x86/kernel/tsc.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index a0346bc..5bb702c 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -922,9 +922,6 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
struct cpufreq_freqs *freq = data;
unsigned long *lpj;
- if (cpu_has(&cpu_data(freq->cpu), X86_FEATURE_CONSTANT_TSC))
- return 0;
-
lpj = &boot_cpu_data.loops_per_jiffy;
#ifdef CONFIG_SMP
if (!(freq->flags & CPUFREQ_CONST_LOOPS))
@@ -954,7 +951,7 @@ static struct notifier_block time_cpufreq_notifier_block = {
.notifier_call = time_cpufreq_notifier
};
-static int __init cpufreq_tsc(void)
+static int __init cpufreq_register_tsc_scaling(void)
{
if (!boot_cpu_has(X86_FEATURE_TSC))
return 0;
@@ -965,7 +962,7 @@ static int __init cpufreq_tsc(void)
return 0;
}
-core_initcall(cpufreq_tsc);
+core_initcall(cpufreq_register_tsc_scaling);
#endif /* CONFIG_CPU_FREQ */
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 4/6] x86/tsc: Save an indentation level in recalibrate_cpu_khz()
2016-04-05 6:29 [PATCH 0/6] x86: Misc cleanups Borislav Petkov
` (2 preceding siblings ...)
2016-04-05 6:29 ` [PATCH 3/6] x86/tsc: Do not check X86_FEATURE_CONSTANT_TSC in notifier call Borislav Petkov
@ 2016-04-05 6:29 ` Borislav Petkov
2016-04-05 8:05 ` Peter Zijlstra
2016-04-13 11:42 ` [tip:x86/asm] " tip-bot for Borislav Petkov
2016-04-05 6:29 ` [PATCH 5/6] x86/fpu: Remove check_fpu() indirection Borislav Petkov
2016-04-05 6:29 ` [PATCH 6/6] x86/fpu: Get rid of x87 math exception helpers Borislav Petkov
5 siblings, 2 replies; 17+ messages in thread
From: Borislav Petkov @ 2016-04-05 6:29 UTC (permalink / raw)
To: X86 ML; +Cc: LKML, Thomas Gleixner, Peter Zijlstra
From: Borislav Petkov <bp@suse.de>
... by flipping the check.
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
---
arch/x86/kernel/tsc.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 5bb702c77e8f..38ba6de56ede 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -834,15 +834,15 @@ int recalibrate_cpu_khz(void)
#ifndef CONFIG_SMP
unsigned long cpu_khz_old = cpu_khz;
- if (boot_cpu_has(X86_FEATURE_TSC)) {
- tsc_khz = x86_platform.calibrate_tsc();
- cpu_khz = tsc_khz;
- cpu_data(0).loops_per_jiffy =
- cpufreq_scale(cpu_data(0).loops_per_jiffy,
- cpu_khz_old, cpu_khz);
- return 0;
- } else
+ if (!boot_cpu_has(X86_FEATURE_TSC))
return -ENODEV;
+
+ tsc_khz = x86_platform.calibrate_tsc();
+ cpu_khz = tsc_khz;
+ cpu_data(0).loops_per_jiffy = cpufreq_scale(cpu_data(0).loops_per_jiffy,
+ cpu_khz_old, cpu_khz);
+
+ return 0;
#else
return -ENODEV;
#endif
--
2.7.3
^ permalink raw reply related [flat|nested] 17+ messages in thread* Re: [PATCH 4/6] x86/tsc: Save an indentation level in recalibrate_cpu_khz()
2016-04-05 6:29 ` [PATCH 4/6] x86/tsc: Save an indentation level in recalibrate_cpu_khz() Borislav Petkov
@ 2016-04-05 8:05 ` Peter Zijlstra
2016-04-13 11:42 ` [tip:x86/asm] " tip-bot for Borislav Petkov
1 sibling, 0 replies; 17+ messages in thread
From: Peter Zijlstra @ 2016-04-05 8:05 UTC (permalink / raw)
To: Borislav Petkov; +Cc: X86 ML, LKML, Thomas Gleixner
On Tue, Apr 05, 2016 at 08:29:53AM +0200, Borislav Petkov wrote:
> From: Borislav Petkov <bp@suse.de>
>
> ... by flipping the check.
>
> Signed-off-by: Borislav Petkov <bp@suse.de>
> Cc: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
^ permalink raw reply [flat|nested] 17+ messages in thread
* [tip:x86/asm] x86/tsc: Save an indentation level in recalibrate_cpu_khz()
2016-04-05 6:29 ` [PATCH 4/6] x86/tsc: Save an indentation level in recalibrate_cpu_khz() Borislav Petkov
2016-04-05 8:05 ` Peter Zijlstra
@ 2016-04-13 11:42 ` tip-bot for Borislav Petkov
1 sibling, 0 replies; 17+ messages in thread
From: tip-bot for Borislav Petkov @ 2016-04-13 11:42 UTC (permalink / raw)
To: linux-tip-commits
Cc: luto, brgerst, bp, dvlasenk, hpa, torvalds, mingo, tglx, bp,
peterz, linux-kernel
Commit-ID: eff4677e9fb9b680d1d5f6ba079116548d072b7e
Gitweb: http://git.kernel.org/tip/eff4677e9fb9b680d1d5f6ba079116548d072b7e
Author: Borislav Petkov <bp@suse.de>
AuthorDate: Tue, 5 Apr 2016 08:29:53 +0200
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 13 Apr 2016 11:37:43 +0200
x86/tsc: Save an indentation level in recalibrate_cpu_khz()
... by flipping the check.
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Andy Lutomirski <luto@amacapital.net>
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: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1459837795-2588-5-git-send-email-bp@alien8.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
arch/x86/kernel/tsc.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 5bb702c..38ba6de 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -834,15 +834,15 @@ int recalibrate_cpu_khz(void)
#ifndef CONFIG_SMP
unsigned long cpu_khz_old = cpu_khz;
- if (boot_cpu_has(X86_FEATURE_TSC)) {
- tsc_khz = x86_platform.calibrate_tsc();
- cpu_khz = tsc_khz;
- cpu_data(0).loops_per_jiffy =
- cpufreq_scale(cpu_data(0).loops_per_jiffy,
- cpu_khz_old, cpu_khz);
- return 0;
- } else
+ if (!boot_cpu_has(X86_FEATURE_TSC))
return -ENODEV;
+
+ tsc_khz = x86_platform.calibrate_tsc();
+ cpu_khz = tsc_khz;
+ cpu_data(0).loops_per_jiffy = cpufreq_scale(cpu_data(0).loops_per_jiffy,
+ cpu_khz_old, cpu_khz);
+
+ return 0;
#else
return -ENODEV;
#endif
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 5/6] x86/fpu: Remove check_fpu() indirection
2016-04-05 6:29 [PATCH 0/6] x86: Misc cleanups Borislav Petkov
` (3 preceding siblings ...)
2016-04-05 6:29 ` [PATCH 4/6] x86/tsc: Save an indentation level in recalibrate_cpu_khz() Borislav Petkov
@ 2016-04-05 6:29 ` Borislav Petkov
2016-04-13 11:42 ` [tip:x86/asm] " tip-bot for Borislav Petkov
2016-04-05 6:29 ` [PATCH 6/6] x86/fpu: Get rid of x87 math exception helpers Borislav Petkov
5 siblings, 1 reply; 17+ messages in thread
From: Borislav Petkov @ 2016-04-05 6:29 UTC (permalink / raw)
To: X86 ML; +Cc: LKML
From: Borislav Petkov <bp@suse.de>
Rename it to fpu__init_check_bugs() and do the CPU feature check at
entry, thus getting rid of the old fpu__init_check_bugs() wrapper.
Signed-off-by: Borislav Petkov <bp@suse.de>
---
arch/x86/kernel/fpu/bugs.c | 16 +++++-----------
1 file changed, 5 insertions(+), 11 deletions(-)
diff --git a/arch/x86/kernel/fpu/bugs.c b/arch/x86/kernel/fpu/bugs.c
index 224b5ec52195..aad34aafc0e0 100644
--- a/arch/x86/kernel/fpu/bugs.c
+++ b/arch/x86/kernel/fpu/bugs.c
@@ -21,11 +21,15 @@ static double __initdata y = 3145727.0;
* We should really only care about bugs here
* anyway. Not features.
*/
-static void __init check_fpu(void)
+void __init fpu__init_check_bugs(void)
{
u32 cr0_saved;
s32 fdiv_bug;
+ /* kernel_fpu_begin/end() relies on patched alternative instructions. */
+ if (!boot_cpu_has(X86_FEATURE_FPU))
+ return;
+
/* We might have CR0::TS set already, clear it: */
cr0_saved = read_cr0();
write_cr0(cr0_saved & ~X86_CR0_TS);
@@ -59,13 +63,3 @@ static void __init check_fpu(void)
pr_warn("Hmm, FPU with FDIV bug\n");
}
}
-
-void __init fpu__init_check_bugs(void)
-{
- /*
- * kernel_fpu_begin/end() in check_fpu() relies on the patched
- * alternative instructions.
- */
- if (boot_cpu_has(X86_FEATURE_FPU))
- check_fpu();
-}
--
2.7.3
^ permalink raw reply related [flat|nested] 17+ messages in thread* [tip:x86/asm] x86/fpu: Remove check_fpu() indirection
2016-04-05 6:29 ` [PATCH 5/6] x86/fpu: Remove check_fpu() indirection Borislav Petkov
@ 2016-04-13 11:42 ` tip-bot for Borislav Petkov
0 siblings, 0 replies; 17+ messages in thread
From: tip-bot for Borislav Petkov @ 2016-04-13 11:42 UTC (permalink / raw)
To: linux-tip-commits
Cc: bp, brgerst, dvlasenk, linux-kernel, tglx, mingo, hpa, bp, luto,
peterz, torvalds
Commit-ID: de82fbc3823b7b15ee03466ebfb1c5ec7cc1a941
Gitweb: http://git.kernel.org/tip/de82fbc3823b7b15ee03466ebfb1c5ec7cc1a941
Author: Borislav Petkov <bp@suse.de>
AuthorDate: Tue, 5 Apr 2016 08:29:54 +0200
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 13 Apr 2016 11:37:43 +0200
x86/fpu: Remove check_fpu() indirection
Rename it to fpu__init_check_bugs() and do the CPU feature check at
entry, thus getting rid of the old fpu__init_check_bugs() wrapper.
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Andy Lutomirski <luto@amacapital.net>
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: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1459837795-2588-6-git-send-email-bp@alien8.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
arch/x86/kernel/fpu/bugs.c | 16 +++++-----------
1 file changed, 5 insertions(+), 11 deletions(-)
diff --git a/arch/x86/kernel/fpu/bugs.c b/arch/x86/kernel/fpu/bugs.c
index 224b5ec..aad34aa 100644
--- a/arch/x86/kernel/fpu/bugs.c
+++ b/arch/x86/kernel/fpu/bugs.c
@@ -21,11 +21,15 @@ static double __initdata y = 3145727.0;
* We should really only care about bugs here
* anyway. Not features.
*/
-static void __init check_fpu(void)
+void __init fpu__init_check_bugs(void)
{
u32 cr0_saved;
s32 fdiv_bug;
+ /* kernel_fpu_begin/end() relies on patched alternative instructions. */
+ if (!boot_cpu_has(X86_FEATURE_FPU))
+ return;
+
/* We might have CR0::TS set already, clear it: */
cr0_saved = read_cr0();
write_cr0(cr0_saved & ~X86_CR0_TS);
@@ -59,13 +63,3 @@ static void __init check_fpu(void)
pr_warn("Hmm, FPU with FDIV bug\n");
}
}
-
-void __init fpu__init_check_bugs(void)
-{
- /*
- * kernel_fpu_begin/end() in check_fpu() relies on the patched
- * alternative instructions.
- */
- if (boot_cpu_has(X86_FEATURE_FPU))
- check_fpu();
-}
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 6/6] x86/fpu: Get rid of x87 math exception helpers
2016-04-05 6:29 [PATCH 0/6] x86: Misc cleanups Borislav Petkov
` (4 preceding siblings ...)
2016-04-05 6:29 ` [PATCH 5/6] x86/fpu: Remove check_fpu() indirection Borislav Petkov
@ 2016-04-05 6:29 ` Borislav Petkov
2016-04-13 11:42 ` [tip:x86/asm] " tip-bot for Borislav Petkov
5 siblings, 1 reply; 17+ messages in thread
From: Borislav Petkov @ 2016-04-05 6:29 UTC (permalink / raw)
To: X86 ML; +Cc: LKML
From: Borislav Petkov <bp@suse.de>
... and integrate their functionality into their single user
fpu__exception_code().
No functionality change.
Signed-off-by: Borislav Petkov <bp@suse.de>
---
arch/x86/kernel/fpu/core.c | 44 +++++++++++++-------------------------------
1 file changed, 13 insertions(+), 31 deletions(-)
diff --git a/arch/x86/kernel/fpu/core.c b/arch/x86/kernel/fpu/core.c
index 1551b28398a4..97027545a72d 100644
--- a/arch/x86/kernel/fpu/core.c
+++ b/arch/x86/kernel/fpu/core.c
@@ -506,33 +506,6 @@ void fpu__clear(struct fpu *fpu)
* x87 math exception handling:
*/
-static inline unsigned short get_fpu_cwd(struct fpu *fpu)
-{
- if (boot_cpu_has(X86_FEATURE_FXSR)) {
- return fpu->state.fxsave.cwd;
- } else {
- return (unsigned short)fpu->state.fsave.cwd;
- }
-}
-
-static inline unsigned short get_fpu_swd(struct fpu *fpu)
-{
- if (boot_cpu_has(X86_FEATURE_FXSR)) {
- return fpu->state.fxsave.swd;
- } else {
- return (unsigned short)fpu->state.fsave.swd;
- }
-}
-
-static inline unsigned short get_fpu_mxcsr(struct fpu *fpu)
-{
- if (boot_cpu_has(X86_FEATURE_XMM)) {
- return fpu->state.fxsave.mxcsr;
- } else {
- return MXCSR_DEFAULT;
- }
-}
-
int fpu__exception_code(struct fpu *fpu, int trap_nr)
{
int err;
@@ -547,10 +520,15 @@ int fpu__exception_code(struct fpu *fpu, int trap_nr)
* so if this combination doesn't produce any single exception,
* then we have a bad program that isn't synchronizing its FPU usage
* and it will suffer the consequences since we won't be able to
- * fully reproduce the context of the exception
+ * fully reproduce the context of the exception.
*/
- cwd = get_fpu_cwd(fpu);
- swd = get_fpu_swd(fpu);
+ if (boot_cpu_has(X86_FEATURE_FXSR)) {
+ cwd = fpu->state.fxsave.cwd;
+ swd = fpu->state.fxsave.swd;
+ } else {
+ cwd = (unsigned short)fpu->state.fsave.cwd;
+ swd = (unsigned short)fpu->state.fsave.swd;
+ }
err = swd & ~cwd;
} else {
@@ -560,7 +538,11 @@ int fpu__exception_code(struct fpu *fpu, int trap_nr)
* unmasked exception was caught we must mask the exception mask bits
* at 0x1f80, and then use these to mask the exception bits at 0x3f.
*/
- unsigned short mxcsr = get_fpu_mxcsr(fpu);
+ unsigned short mxcsr = MXCSR_DEFAULT;
+
+ if (boot_cpu_has(X86_FEATURE_XMM))
+ mxcsr = fpu->state.fxsave.mxcsr;
+
err = ~(mxcsr >> 7) & mxcsr;
}
--
2.7.3
^ permalink raw reply related [flat|nested] 17+ messages in thread* [tip:x86/asm] x86/fpu: Get rid of x87 math exception helpers
2016-04-05 6:29 ` [PATCH 6/6] x86/fpu: Get rid of x87 math exception helpers Borislav Petkov
@ 2016-04-13 11:42 ` tip-bot for Borislav Petkov
0 siblings, 0 replies; 17+ messages in thread
From: tip-bot for Borislav Petkov @ 2016-04-13 11:42 UTC (permalink / raw)
To: linux-tip-commits
Cc: bp, luto, torvalds, bp, peterz, dvlasenk, linux-kernel, tglx,
mingo, hpa, brgerst
Commit-ID: 6aa6dbfced51dec6cde159c6167ad3dad6add823
Gitweb: http://git.kernel.org/tip/6aa6dbfced51dec6cde159c6167ad3dad6add823
Author: Borislav Petkov <bp@suse.de>
AuthorDate: Tue, 5 Apr 2016 08:29:55 +0200
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 13 Apr 2016 11:37:44 +0200
x86/fpu: Get rid of x87 math exception helpers
... and integrate their functionality into their single user
fpu__exception_code().
No functionality change.
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Andy Lutomirski <luto@amacapital.net>
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: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1459837795-2588-7-git-send-email-bp@alien8.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
arch/x86/kernel/fpu/core.c | 44 +++++++++++++-------------------------------
1 file changed, 13 insertions(+), 31 deletions(-)
diff --git a/arch/x86/kernel/fpu/core.c b/arch/x86/kernel/fpu/core.c
index 1551b28..9702754 100644
--- a/arch/x86/kernel/fpu/core.c
+++ b/arch/x86/kernel/fpu/core.c
@@ -506,33 +506,6 @@ void fpu__clear(struct fpu *fpu)
* x87 math exception handling:
*/
-static inline unsigned short get_fpu_cwd(struct fpu *fpu)
-{
- if (boot_cpu_has(X86_FEATURE_FXSR)) {
- return fpu->state.fxsave.cwd;
- } else {
- return (unsigned short)fpu->state.fsave.cwd;
- }
-}
-
-static inline unsigned short get_fpu_swd(struct fpu *fpu)
-{
- if (boot_cpu_has(X86_FEATURE_FXSR)) {
- return fpu->state.fxsave.swd;
- } else {
- return (unsigned short)fpu->state.fsave.swd;
- }
-}
-
-static inline unsigned short get_fpu_mxcsr(struct fpu *fpu)
-{
- if (boot_cpu_has(X86_FEATURE_XMM)) {
- return fpu->state.fxsave.mxcsr;
- } else {
- return MXCSR_DEFAULT;
- }
-}
-
int fpu__exception_code(struct fpu *fpu, int trap_nr)
{
int err;
@@ -547,10 +520,15 @@ int fpu__exception_code(struct fpu *fpu, int trap_nr)
* so if this combination doesn't produce any single exception,
* then we have a bad program that isn't synchronizing its FPU usage
* and it will suffer the consequences since we won't be able to
- * fully reproduce the context of the exception
+ * fully reproduce the context of the exception.
*/
- cwd = get_fpu_cwd(fpu);
- swd = get_fpu_swd(fpu);
+ if (boot_cpu_has(X86_FEATURE_FXSR)) {
+ cwd = fpu->state.fxsave.cwd;
+ swd = fpu->state.fxsave.swd;
+ } else {
+ cwd = (unsigned short)fpu->state.fsave.cwd;
+ swd = (unsigned short)fpu->state.fsave.swd;
+ }
err = swd & ~cwd;
} else {
@@ -560,7 +538,11 @@ int fpu__exception_code(struct fpu *fpu, int trap_nr)
* unmasked exception was caught we must mask the exception mask bits
* at 0x1f80, and then use these to mask the exception bits at 0x3f.
*/
- unsigned short mxcsr = get_fpu_mxcsr(fpu);
+ unsigned short mxcsr = MXCSR_DEFAULT;
+
+ if (boot_cpu_has(X86_FEATURE_XMM))
+ mxcsr = fpu->state.fxsave.mxcsr;
+
err = ~(mxcsr >> 7) & mxcsr;
}
^ permalink raw reply related [flat|nested] 17+ messages in thread