Skip to contentSkip to conversation

x86: update AMD CPUID bits

PATCH New Andre Przywara 1 participant 5 messages Sep 3, 2010

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

cover x86: update AMD CPUID bits

Recently the public AMD CPUID specification
http://support.amd.com/us/Processor_TechDocs/25481.pdf
has been updated and revealed new CPUID flag feature names. The following patches introduce them to the kernel to properly display them in /proc/cpuinfo and allows KVM guests to use them.
Note: One bit has been renamed, so I propose patch 1/4 for inclusion in the stable series.

Please apply!

Regards,
Andre.

--
Andre Przywara
AMD-Operating System Research Center (OSRC), Dresden, Germany

#1 lore raw

PATCH 1/4 x86: Fix misnamed AMD CPUID feature bit

The AMD SSE5 feature set as-it has been replaced by some extensions to the AVX instruction set. Thus the bit formerly advertised as SSE5 is re-used for one of these extensions (XOP).
Although this changes the /proc/cpuinfo output, it is not user visible, as there are no CPUs (yet) having this feature.
To avoid confusion this should be added to the stable series, too.

Cc [email protected] [.32.x .34.x, .35.x]
---
 arch/x86/include/asm/cpufeature.h |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
arch/x86/include/asm/cpufeature.h +1 -1
index 781a50b..c9c73d8 100644
@@ -152,7 +152,7 @@
 #define X86_FEATURE_3DNOWPREFETCH (6*32+ 8) /* 3DNow prefetch instructions */
 #define X86_FEATURE_OSVW	(6*32+ 9) /* OS Visible Workaround */
 #define X86_FEATURE_IBS		(6*32+10) /* Instruction Based Sampling */
-#define X86_FEATURE_SSE5	(6*32+11) /* SSE-5 */
+#define X86_FEATURE_XOP		(6*32+11) /* extended AVX instructions */
 #define X86_FEATURE_SKINIT	(6*32+12) /* SKINIT/STGI instructions */
 #define X86_FEATURE_WDT		(6*32+13) /* Watchdog timer */
 #define X86_FEATURE_NODEID_MSR	(6*32+19) /* NodeId MSR */
Signature · 1 line
1.6.4
#2 lore raw

PATCH 2/4 x86: Update AMD CPUID feature bits

In reply to Andre Przywara

AMD's public CPUID specification has been updated and some bits have got names. Add them to properly describe new CPU features.

---
 arch/x86/include/asm/cpufeature.h |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)
arch/x86/include/asm/cpufeature.h +4 -0
index c9c73d8..341835d 100644
@@ -155,7 +155,11 @@
 #define X86_FEATURE_XOP		(6*32+11) /* extended AVX instructions */
 #define X86_FEATURE_SKINIT	(6*32+12) /* SKINIT/STGI instructions */
 #define X86_FEATURE_WDT		(6*32+13) /* Watchdog timer */
+#define X86_FEATURE_LWP		(6*32+15) /* Light Weight Profiling */
+#define X86_FEATURE_FMA4	(6*32+16) /* 4 operands MAC instructions */
 #define X86_FEATURE_NODEID_MSR	(6*32+19) /* NodeId MSR */
+#define X86_FEATURE_TBM		(6*32+21) /* trailing bit manipulations */
+#define X86_FEATURE_TOPOEXT	(6*32+22) /* topology extensions CPUID leafs */
 
 /*
  * Auxiliary flags: Linux defined - For features scattered in various
Signature · 1 line
1.6.4
#3 lore raw

PATCH 3/4 x86: Fix allowed CPUID bits for KVM guests

In reply to Andre Przywara

The AMD extension to AVX (FMA4, XOP) work on the same YMM register set as AVX, so they are safe for guests to use, as long as AVX itself is allowed.

---
 arch/x86/kvm/x86.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)
arch/x86/kvm/x86.c +2 -2
index 3a09c62..eb89e7b 100644
@@ -1996,8 +1996,8 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
 	const u32 kvm_supported_word6_x86_features =
 		F(LAHF_LM) | F(CMP_LEGACY) | F(SVM) | 0 /* ExtApicSpace */ |
 		F(CR8_LEGACY) | F(ABM) | F(SSE4A) | F(MISALIGNSSE) |
-		F(3DNOWPREFETCH) | 0 /* OSVW */ | 0 /* IBS */ | F(SSE5) |
-		0 /* SKINIT */ | 0 /* WDT */;
+		F(3DNOWPREFETCH) | 0 /* OSVW */ | 0 /* IBS */ | F(XOP) |
+		0 /* SKINIT, WDT, LWP */ | F(FMA4) | F(TBM);
 
 	/* all calls to cpuid_count() should be made on the same cpu */
 	get_cpu();
Signature · 1 line
1.6.4
#4 lore raw

PATCH 4/4 x86, kvm: add new AMD SVM feature bits

In reply to Andre Przywara

The recently updated CPUID specification names new SVM feature bits. Add them to the list of reported features.

---
 arch/x86/include/asm/cpufeature.h |    7 +++++++
 arch/x86/kernel/cpu/scattered.c   |    6 ++++++
 2 files changed, 13 insertions(+), 0 deletions(-)
arch/x86/include/asm/cpufeature.h +7 -0
index 341835d..bffeab7 100644
@@ -183,6 +183,13 @@
 #define X86_FEATURE_LBRV	(8*32+ 6) /* AMD LBR Virtualization support */
 #define X86_FEATURE_SVML	(8*32+ 7) /* "svm_lock" AMD SVM locking MSR */
 #define X86_FEATURE_NRIPS	(8*32+ 8) /* "nrip_save" AMD SVM next_rip save */
+#define X86_FEATURE_TSCRATEMSR  (8*32+ 9) /* "tsc_scale" AMD TSC scaling support */
+#define X86_FEATURE_VMCBCLEAN   (8*32+10) /* "vmcb_clean" AMD VMCB clean bits support */
+#define X86_FEATURE_FLUSHBYASID (8*32+11) /* AMD flush-by-ASID support */
+#define X86_FEATURE_DECODEASSISTS (8*32+12) /* AMD Decode Assists support */
+#define X86_FEATURE_PAUSEFILTER (8*32+13) /* AMD filtered pause intercept */
+#define X86_FEATURE_PFTHRESHOLD (8*32+14) /* AMD pause filter threshold */
+
 
 /* Intel-defined CPU features, CPUID level 0x00000007:0 (ebx), word 9 */
 #define X86_FEATURE_FSGSBASE	(9*32+ 0) /* {RD/WR}{FS/GS}BASE instructions*/
arch/x86/kernel/cpu/scattered.c +6 -0
index 34b4dad..2c77931 100644
@@ -43,6 +43,12 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c)
 		{ X86_FEATURE_LBRV,		CR_EDX, 1, 0x8000000a, 0 },
 		{ X86_FEATURE_SVML,		CR_EDX, 2, 0x8000000a, 0 },
 		{ X86_FEATURE_NRIPS,		CR_EDX, 3, 0x8000000a, 0 },
+		{ X86_FEATURE_TSCRATEMSR,	CR_EDX, 4, 0x8000000a, 0 },
+		{ X86_FEATURE_VMCBCLEAN,	CR_EDX, 5, 0x8000000a, 0 },
+		{ X86_FEATURE_FLUSHBYASID,	CR_EDX, 6, 0x8000000a, 0 },
+		{ X86_FEATURE_DECODEASSISTS,	CR_EDX, 7, 0x8000000a, 0 },
+		{ X86_FEATURE_PAUSEFILTER,	CR_EDX,10, 0x8000000a, 0 },
+		{ X86_FEATURE_PFTHRESHOLD,	CR_EDX,12, 0x8000000a, 0 },
 		{ 0, 0, 0, 0, 0 }
 	};
 
Signature · 1 line
1.6.4