[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <48225DEC.2030502@keyaccess.nl>
Date: Thu, 08 May 2008 03:57:00 +0200
From: Rene Herman <rene.herman@...access.nl>
To: Linus Torvalds <torvalds@...ux-foundation.org>
CC: "H. Peter Anvin" <hpa@...or.com>,
Thomas Gleixner <tglx@...utronix.de>,
Adrian Bunk <bunk@...nel.org>,
Yinghai Lu <yhlu.kernel@...il.com>,
Ingo Molnar <mingo@...e.hu>,
Linux Kernel <linux-kernel@...r.kernel.org>,
akpm@...ux-foundation.org, Pavel Machek <pavel@...e.cz>
Subject: [PATCH] x86: introduce a new Linux defined feature flag for PAT support
On 08-05-08 02:28, Rene Herman wrote:
> I completely and fully agree with your new flag suggestion. Was just
> in a git bisect and am actually on the lowly thing in question so I'm
> not fast but was looking already. X86_FEATURE_PAT_GOOD. Is it safe to
> reuse bit 14 now?
Okay, so how's this? Seem to work well for me and makes me happy. Only
tested on UP.
(hope the inline thing works)
arch/x86/kernel/cpu/common.c | 12 ++++--------
arch/x86/kernel/cpu/feature_names.c | 2 +-
arch/x86/kernel/setup_64.c | 7 ++-----
arch/x86/mm/pat.c | 2 +-
include/asm-x86/cpufeature.h | 3 ++-
5 files changed, 10 insertions(+), 16 deletions(-)
x86: introduce a new Linux defined feature flag for PAT support
Use a new Linux defined X86_FEATURE_PAT_GOOD feature flag to
indicate we're on a CPU we want to support PAT on, so as to
keep the CPU defined X86_FEATURE_PAT readily available for
indicating if the CPU even has the PAT feature.
Signed-off-by: Rene Herman <rene.herman@...il.com>
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 35b4f6a..7bec5e0 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -308,16 +308,14 @@ static void __cpuinit early_get_cap(struct cpuinfo_x86 *c)
}
- clear_cpu_cap(c, X86_FEATURE_PAT);
-
switch (c->x86_vendor) {
case X86_VENDOR_AMD:
if (c->x86 >= 0xf && c->x86 <= 0x11)
- set_cpu_cap(c, X86_FEATURE_PAT);
+ set_cpu_cap(c, X86_FEATURE_PAT_GOOD);
break;
case X86_VENDOR_INTEL:
if (c->x86 == 0xF || (c->x86 == 6 && c->x86_model >= 15))
- set_cpu_cap(c, X86_FEATURE_PAT);
+ set_cpu_cap(c, X86_FEATURE_PAT_GOOD);
break;
}
@@ -409,16 +407,14 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
init_scattered_cpuid_features(c);
}
- clear_cpu_cap(c, X86_FEATURE_PAT);
-
switch (c->x86_vendor) {
case X86_VENDOR_AMD:
if (c->x86 >= 0xf && c->x86 <= 0x11)
- set_cpu_cap(c, X86_FEATURE_PAT);
+ set_cpu_cap(c, X86_FEATURE_PAT_GOOD);
break;
case X86_VENDOR_INTEL:
if (c->x86 == 0xF || (c->x86 == 6 && c->x86_model >= 15))
- set_cpu_cap(c, X86_FEATURE_PAT);
+ set_cpu_cap(c, X86_FEATURE_PAT_GOOD);
break;
}
}
diff --git a/arch/x86/kernel/cpu/feature_names.c b/arch/x86/kernel/cpu/feature_names.c
index e43ad4a..04f996d 100644
--- a/arch/x86/kernel/cpu/feature_names.c
+++ b/arch/x86/kernel/cpu/feature_names.c
@@ -38,7 +38,7 @@ const char * const x86_cap_flags[NCAPINTS*32] = {
"cxmmx", "k6_mtrr", "cyrix_arr", "centaur_mcr",
NULL, NULL, NULL, NULL,
"constant_tsc", "up", NULL, "arch_perfmon",
- "pebs", "bts", NULL, NULL,
+ "pebs", "bts", "pat_good", NULL,
"rep_good", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 22c14e2..460134e 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -1063,19 +1063,16 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
if (c->extended_cpuid_level >= 0x80000007)
c->x86_power = cpuid_edx(0x80000007);
-
- clear_cpu_cap(c, X86_FEATURE_PAT);
-
switch (c->x86_vendor) {
case X86_VENDOR_AMD:
early_init_amd(c);
if (c->x86 >= 0xf && c->x86 <= 0x11)
- set_cpu_cap(c, X86_FEATURE_PAT);
+ set_cpu_cap(c, X86_FEATURE_PAT_GOOD);
break;
case X86_VENDOR_INTEL:
early_init_intel(c);
if (c->x86 == 0xF || (c->x86 == 6 && c->x86_model >= 15))
- set_cpu_cap(c, X86_FEATURE_PAT);
+ set_cpu_cap(c, X86_FEATURE_PAT_GOOD);
break;
case X86_VENDOR_CENTAUR:
early_init_centaur(c);
diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c
index 277446c..6ee3efb 100644
--- a/arch/x86/mm/pat.c
+++ b/arch/x86/mm/pat.c
@@ -43,7 +43,7 @@ static int pat_known_cpu(void)
if (!pat_wc_enabled)
return 0;
- if (cpu_has_pat)
+ if (cpu_has_pat && cpu_has_pat_good)
return 1;
pat_wc_enabled = 0;
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 0d609c8..63c14e4 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -74,7 +74,7 @@
#define X86_FEATURE_ARCH_PERFMON (3*32+11) /* Intel Architectural PerfMon */
#define X86_FEATURE_PEBS (3*32+12) /* Precise-Event Based Sampling */
#define X86_FEATURE_BTS (3*32+13) /* Branch Trace Store */
-/* 14 free */
+#define X86_FEATURE_PAT_GOOD (3*32+14) /* PAT works correctly */
/* 15 free */
#define X86_FEATURE_REP_GOOD (3*32+16) /* rep microcode works well on this CPU */
#define X86_FEATURE_MFENCE_RDTSC (3*32+17) /* Mfence synchronizes RDTSC */
@@ -187,6 +187,7 @@ extern const char * const x86_power_flags[32];
#define cpu_has_gbpages boot_cpu_has(X86_FEATURE_GBPAGES)
#define cpu_has_arch_perfmon boot_cpu_has(X86_FEATURE_ARCH_PERFMON)
#define cpu_has_pat boot_cpu_has(X86_FEATURE_PAT)
+#define cpu_has_pat_good boot_cpu_has(X86_FEATURE_PAT_GOOD)
#if defined(CONFIG_X86_INVLPG) || defined(CONFIG_X86_64)
# define cpu_has_invlpg 1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists