arch/powerpc: advertise ISA2.07, HTM, DSCR, EBB and ISEL bits in HWCAP2
PATCH 1/1 arch/powerpc: advertise ISA2.07, HTM, DSCR, EBB and ISEL bits in HWCAP2
Now that we have AT_HWCAP2 support, start exposing some of the new POWER8 features via it.
---
Note: there are, I think, some Freescale processors that also should be updated to indicate they support ISEL, but I don't know which ones. Since this is a new feature bit (and vector), it seems like we can fix that up in a follow-on patch. Also, this is my first patch trying to manipulate these bits, so please let me know if I'm doing something wrong (for instance, I don't see any particular order to the bits in PPC_FEATURE_*)
arch/powerpc/include/uapi/asm/cputable.h +8 -0
@@ -1,6 +1,7 @@#ifndef _UAPI__ASM_POWERPC_CPUTABLE_H#define _UAPI__ASM_POWERPC_CPUTABLE_H +/* in AT_HWCAP */ #define PPC_FEATURE_32 0x80000000#define PPC_FEATURE_64 0x40000000#define PPC_FEATURE_601_INSTR 0x20000000
@@ -33,4 +34,11 @@#define PPC_FEATURE_TRUE_LE 0x00000002#define PPC_FEATURE_PPC_LE 0x00000001 +/* in AT_HWCAP2 */ +#define PPC_FEATURE2_ARCH_2_07 0x80000000 +#define PPC_FEATURE2_HTM 0x40000000 +#define PPC_FEATURE2_DSCR 0x20000000 +#define PPC_FEATURE2_EBB 0x10000000 +#define PPC_FEATURE2_ISEL 0x08000000 +#endif /* _UAPI__ASM_POWERPC_CPUTABLE_H */
arch/powerpc/kernel/cputable.c +5 -0
@@ -102,6 +102,9 @@ extern void __restore_cpu_e6500(void); PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \ PPC_FEATURE_TRUE_LE | \ PPC_FEATURE_PSERIES_PERFMON_COMPAT) +#define COMMON_USER2_POWER8 (PPC_FEATURE2_ARCH_2_07 | PPC_FEATURE2_HTM | \ + PPC_FEATURE2_DSCR | PPC_FEATURE2_EBB | \ + PPC_FEATURE2_ISEL)#define COMMON_USER_PA6T (COMMON_USER_PPC64 | PPC_FEATURE_PA6T |\ PPC_FEATURE_TRUE_LE | \ PPC_FEATURE_HAS_ALTIVEC_COMP)
@@ -443,6 +446,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER8 (architected)", .cpu_features = CPU_FTRS_POWER8, .cpu_user_features = COMMON_USER_POWER8, + .cpu_user_features2 = COMMON_USER2_POWER8, .mmu_features = MMU_FTRS_POWER8, .icache_bsize = 128, .dcache_bsize = 128,
@@ -492,6 +496,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER8 (raw)", .cpu_features = CPU_FTRS_POWER8, .cpu_user_features = COMMON_USER_POWER8, + .cpu_user_features2 = COMMON_USER2_POWER8, .mmu_features = MMU_FTRS_POWER8, .icache_bsize = 128, .dcache_bsize = 128,
arch/powerpc/include/uapi/asm/cputable.h line 42
Quoted 6 lines · Nishanth Aravamudan
arch/powerpc/include/uapi/asm/cputable.h +6
-0+/* in AT_HWCAP2 */ +#define PPC_FEATURE2_ARCH_2_07 0x80000000 +#define PPC_FEATURE2_HTM 0x40000000 +#define PPC_FEATURE2_DSCR 0x20000000 +#define PPC_FEATURE2_EBB 0x10000000 +#define PPC_FEATURE2_ISEL 0x08000000
Should we "adjust" (ie filter out) some of these based on CONFIG_ options (such as transactional memory enabled, EBB supported by the hypervisor, etc...) ?
Cheers,
Ben.
Quoted 11 lines · Benjamin Herrenschmidt
arch/powerpc/include/uapi/asm/cputable.h line 42
Quoted 6 lines · Nishanth Aravamudan
arch/powerpc/include/uapi/asm/cputable.h +6
-0+/* in AT_HWCAP2 */ +#define PPC_FEATURE2_ARCH_2_07 0x80000000 +#define PPC_FEATURE2_HTM 0x40000000 +#define PPC_FEATURE2_DSCR 0x20000000 +#define PPC_FEATURE2_EBB 0x10000000 +#define PPC_FEATURE2_ISEL 0x08000000Should we "adjust" (ie filter out) some of these based on
CONFIG_options (such as transactional memory enabled, EBB supported by the hypervisor, etc...) ?
Err, yeah, that seems reasonable :) However, it seems like glibc uses these values rather directly so it knows what bits to check for each feature. Therefore, it seems like it would be better to do the ifdeffery/checking in the user in cputable.c, but that seems like it could get quite complicated.
Would it be ok (I guess I'm asking Ryan & co. here) to have an #ifdef in the definition that may or may not mean the bit is set in the aux vector, but the bit, if set, would always be the same bit?
-Nish
Nishanth Aravamudan <[email protected]> wrote on 05/03/2013 06:40:19 PM:
Quoted 38 lines
Nishanth Aravamudan <[email protected]>
05/03/2013 06:40 PMTo
Benjamin Herrenschmidt <[email protected]>
cc
Michael Neuling <[email protected]>, Michael R Meissner/ Cambridge/IBM@IBMUS, Steve Munroe/Rochester/IBM@IBMUS, Peter Bergner/Rochester/IBM@IBMUS, Ryan Arnold/Rochester/IBM@IBMUS, [email protected]
Subject
Re: [PATCH]
arch/powerpc: advertise ISA2.07, HTM, DSCR, EBB and ISEL bits in HWCAP2Quoted 11 lines · Benjamin Herrenschmidt
arch/powerpc/include/uapi/asm/cputable.h line 42
Quoted 6 lines · Nishanth Aravamudan
arch/powerpc/include/uapi/asm/cputable.h +6
-0+/* in AT_HWCAP2 */ +#define PPC_FEATURE2_ARCH_2_07 0x80000000 +#define PPC_FEATURE2_HTM 0x40000000 +#define PPC_FEATURE2_DSCR 0x20000000 +#define PPC_FEATURE2_EBB 0x10000000 +#define PPC_FEATURE2_ISEL 0x08000000Should we "adjust" (ie filter out) some of these based on
CONFIG_options (such as transactional memory enabled, EBB supported by the hypervisor, etc...) ?Err, yeah, that seems reasonable :) However, it seems like glibc uses these values rather directly so it knows what bits to check for each feature. Therefore, it seems like it would be better to do the ifdeffery/checking in the user in
cputable.c, but that seems like it could get quite complicated.
Quoted 3 lines
Would it be ok (I guess I'm asking Ryan & co. here) to have an #ifdef in the definition that may or may not mean the bit is set in the aux vector, but the bit, if set, would always be the same bit?
My understanding was that these bits being 'on' is an indication of what features the hardware supports (or what the kernel emulates) and a not an indication of whether that facility is currently enabled or not. If the hardware supports a particular feature but it is not enabled I'd expect that user-space usage of that feature would cause the kernel to trap on a facility availability exception (which is how Altivec/VMX is implemented, being defaulted to turned off).
Otherwise there's no way I could know whether an ISA [optional] feature is actually available on a particular machine.
And yes, the bits can't change. My usage of hwcap.h has to coincide with the kernel's asm/cputable.h
Ryan
Quoted 7 lines · Ryan Arnold
My understanding was that these bits being 'on' is an indication of what features the hardware supports (or what the kernel emulates) and a not an indication of whether that facility is currently enabled or not. If the hardware supports a particular feature but it is not enabled I'd expect that user-space usage of that feature would cause the kernel to trap on a facility availability exception (which is how Altivec/VMX is implemented, being defaulted to turned off).
Right but the discussion is about whether we should expose the bits when the kernel doesn't have the ability to handle the feature :-)
IE. We need to remove the HTM feature if the kernel is compiled without transactional memory support.
Similarily, Nish, you may need to check that we remove those bits if pHyp has the partition in a mode that doesn't support them (P7 compatibility for example) for migration purposes.
Cheers,
Ben.
Quoted 5 lines
Otherwise there's no way I could know whether an ISA [optional] feature is actually available on a particular machine.
And yes, the bits can't change. My usage of
hwcap.hhas to coincide with the kernel'sasm/cputable.h
Benjamin Herrenschmidt <[email protected]> wrote on 05/06/2013 04:37:08 PM:
Quoted 28 lines
Benjamin Herrenschmidt <[email protected]>
05/06/2013 04:37 PMTo
Ryan Arnold/Rochester/IBM@IBMUS
cc
Nishanth Aravamudan <[email protected]>, linuxppc- [email protected], Michael R Meissner/Cambridge/IBM@IBMUS, Michael Neuling <[email protected]>, Peter Bergner/ Rochester/IBM@IBMUS, Steve Munroe/Rochester/IBM@IBMUS
Subject
Re: [PATCH]
arch/powerpc: advertise ISA2.07, HTM, DSCR, EBB and ISEL bits in HWCAP2Quoted 7 lines · Ryan Arnold
My understanding was that these bits being 'on' is an indication of what features the hardware supports (or what the kernel emulates) and a not an indication of whether that facility is currently enabled or not. If the hardware supports a particular feature but it is not enabled I'd expect that user-space usage of that feature would cause the kernel to trap on a facility availability exception (which is how Altivec/VMX is implemented, being defaulted to turned off).
Quoted 5 lines
Right but the discussion is about whether we should expose the bits when the kernel doesn't have the ability to handle the feature :-)
IE. We need to remove the HTM feature if the kernel is compiled without transactional memory support.
Thanks for explaining. This is exactly how it should work.
Ryan
Quoted 12 lines · Benjamin Herrenschmidt
Quoted 7 lines · Ryan Arnold
My understanding was that these bits being 'on' is an indication of what features the hardware supports (or what the kernel emulates) and a not an indication of whether that facility is currently enabled or not. If the hardware supports a particular feature but it is not enabled I'd expect that user-space usage of that feature would cause the kernel to trap on a facility availability exception (which is how Altivec/VMX is implemented, being defaulted to turned off).
Right but the discussion is about whether we should expose the bits when the kernel doesn't have the ability to handle the feature :-)
Quoted 6 lines
IE. We need to remove the HTM feature if the kernel is compiled without transactional memory support.
Similarily, Nish, you may need to check that we remove those bits if pHyp has the partition in a mode that doesn't support them (P7 compatibility for example) for migration purposes.
Yep, I'll need to talk with Mikey about this part. Will be a follow-on patch if needed. Minimally, the bit defines will stay the same, which is the important part to get going right now.
Thanks,
Nish
Quoted 7 lines · Nishanth Aravamudan
Quoted 3 lines
Similarily, Nish, you may need to check that we remove those bits if pHyp has the partition in a mode that doesn't support them (P7 compatibility for example) for migration purposes.
Yep, I'll need to talk with Mikey about this part. Will be a follow-on patch if needed. Minimally, the bit defines will stay the same, which is the important part to get going right now.
Actually in such a mode we'd get a back-version architected PVR so we should be fine now that I think twice, but of course that need to be tested.
Cheers,
Ben.
Quoted 5 lines · Benjamin Herrenschmidt
Quoted 4 lines · Nishanth Aravamudan
Quoted 3 lines
Similarily, Nish, you may need to check that we remove those bits if pHyp has the partition in a mode that doesn't support them (P7 compatibility for example) for migration purposes.
Quoted 7 lines
Quoted 3 lines
Yep, I'll need to talk with Mikey about this part. Will be a follow-on patch if needed. Minimally, the bit defines will stay the same, which is the important part to get going right now.
Actually in such a mode we'd get a back-version architected PVR so we should be fine now that I think twice, but of course that need to be tested.
True, I'll make sure it does get tested.
-Nish
According to the GCC sources, ISEL is enabled by default for the 8540, 8548, e500mc, e500mc64, e6500 processors.
From:
Nishanth Aravamudan <[email protected]>
To:
[email protected],
Cc:
Steve Munroe/Rochester/IBM@IBMUS, Peter Bergner/Rochester/IBM@IBMUS, Michael R Meissner/Cambridge/IBM@IBMUS, Michael Neuling <[email protected]>, [email protected], Ryan Arnold/Rochester/IBM@IBMUS
Date:
05/03/2013 07:19 PM
Subject:
[PATCH] arch/powerpc: advertise ISA2.07, HTM, DSCR, EBB and ISEL bits in HWCAP2
Now that we have AT_HWCAP2 support, start exposing some of the new POWER8 features via it.
---
Note: there are, I think, some Freescale processors that also should be updated to indicate they support ISEL, but I don't know which ones. Since this is a new feature bit (and vector), it seems like we can fix that up in a follow-on patch. Also, this is my first patch trying to manipulate these bits, so please let me know if I'm doing something wrong (for instance, I don't see any particular order to the bits in PPC_FEATURE_*)
(quoted diff) +0 -0
b/arch/powerpc/include/uapi/asm/cputable.h
index ed9dd81..78db4e2 100644
arch/powerpc/include/uapi/asm/cputable.h +8 -0
@@ -1,6 +1,7 @@#ifndef _UAPI__ASM_POWERPC_CPUTABLE_H#define _UAPI__ASM_POWERPC_CPUTABLE_H +/* in AT_HWCAP */ #define PPC_FEATURE_32 0x80000000#define PPC_FEATURE_64 0x40000000#define PPC_FEATURE_601_INSTR 0x20000000
@@ -33,4 +34,11 @@#define PPC_FEATURE_TRUE_LE 0x00000002#define PPC_FEATURE_PPC_LE 0x00000001 +/* in AT_HWCAP2 */ +#define PPC_FEATURE2_ARCH_2_07 0x80000000 +#define PPC_FEATURE2_HTM 0x40000000 +#define PPC_FEATURE2_DSCR 0x20000000 +#define PPC_FEATURE2_EBB 0x10000000 +#define PPC_FEATURE2_ISEL 0x08000000 +#endif /* _UAPI__ASM_POWERPC_CPUTABLE_H */
(quoted diff) +0 -0
b/arch/powerpc/kernel/cputable.c
index ae9f433..871c741 100644
arch/powerpc/kernel/cputable.c +1 -0
@@ -102,6 +102,9 @@ extern void __restore_cpu_e6500(void); PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \ PPC_FEATURE_TRUE_LE | \ PPC_FEATURE_PSERIES_PERFMON_COMPAT) +#define COMMON_USER2_POWER8 (PPC_FEATURE2_ARCH_2_07 |
PPC_FEATURE2_HTM | \
+ PPC_FEATURE2_DSCR | PPC_FEATURE2_EBB | \
+ PPC_FEATURE2_ISEL)
#define COMMON_USER_PA6T (COMMON_USER_PPC64 |PPC_FEATURE_PA6T |\
PPC_FEATURE_TRUE_LE | \
PPC_FEATURE_HAS_ALTIVEC_COMP)
@@ -443,6 +446,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
.cpu_name =
"POWER8 (architected)",
.cpu_features =CPU_FTRS_POWER8,
.cpu_user_features =COMMON_USER_POWER8,
+ .cpu_user_features2 =COMMON_USER2_POWER8,
.mmu_features =MMU_FTRS_POWER8,
.icache_bsize =
128,
.dcache_bsize =
128,
@@ -492,6 +496,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
.cpu_name =
"POWER8 (raw)",
.cpu_features =CPU_FTRS_POWER8,
.cpu_user_features =COMMON_USER_POWER8,
+ .cpu_user_features2 =COMMON_USER2_POWER8,
.mmu_features =MMU_FTRS_POWER8,
.icache_bsize =
128,
.dcache_bsize =
128,
Quoted 1 line
According to the GCC sources, ISEL is enabled by default for the 8540, 8548, e500mc, e500mc64, e6500 processors.
And e5500. POWER7 is said to support it as well.
Segher
Quoted 4 lines · Nishanth Aravamudan
Now that we have
AT_HWCAP2support, start exposing some of the new POWER8 features via it.Signed-off-by Nishanth Aravamudan <[email protected]>
A few points:
We need a TAR bit as well, although this could be covered in 207?
For TM we need to turn it off if CONFIG_PPC_TRANSACTIONAL_MEM is not set. Look at PPC_FEATURE_HAS_ALTIVEC_COMP for how we do that and follow the bouncing ball.
Please add ISEL on other processors.
Please add DSCR on other processors.
EBB can be reserved, but there is no enablement at this point so don't turn it on yet. We'll turn it on when we send the perf API.
Mikey
Quoted 64 lines
---
Note: there are, I think, some Freescale processors that also should be updated to indicate they support ISEL, but I don't know which ones. Since this is a new feature bit (and vector), it seems like we can fix that up in a follow-on patch. Also, this is my first patch trying to manipulate these bits, so please let me know if I'm doing something wrong (for instance, I don't see any particular order to the bits in PPC_FEATURE_*)arch/powerpc/include/uapi/asm/cputable.h +8
-0@@ -1,6 +1,7 @@#ifndef _UAPI__ASM_POWERPC_CPUTABLE_H#define _UAPI__ASM_POWERPC_CPUTABLE_H +/* in AT_HWCAP */ #define PPC_FEATURE_32 0x80000000#define PPC_FEATURE_64 0x40000000#define PPC_FEATURE_601_INSTR 0x20000000@@ -33,4 +34,11 @@#define PPC_FEATURE_TRUE_LE 0x00000002#define PPC_FEATURE_PPC_LE 0x00000001 +/* in AT_HWCAP2 */ +#define PPC_FEATURE2_ARCH_2_07 0x80000000 +#define PPC_FEATURE2_HTM 0x40000000 +#define PPC_FEATURE2_DSCR 0x20000000 +#define PPC_FEATURE2_EBB 0x10000000 +#define PPC_FEATURE2_ISEL 0x08000000 +#endif /* _UAPI__ASM_POWERPC_CPUTABLE_H */ arch/powerpc/kernel/cputable.c +5
-0@@ -102,6 +102,9 @@ extern void __restore_cpu_e6500(void); PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \ PPC_FEATURE_TRUE_LE | \ PPC_FEATURE_PSERIES_PERFMON_COMPAT) +#define COMMON_USER2_POWER8 (PPC_FEATURE2_ARCH_2_07 | PPC_FEATURE2_HTM | \ + PPC_FEATURE2_DSCR | PPC_FEATURE2_EBB | \ + PPC_FEATURE2_ISEL)#define COMMON_USER_PA6T (COMMON_USER_PPC64 | PPC_FEATURE_PA6T |\ PPC_FEATURE_TRUE_LE | \ PPC_FEATURE_HAS_ALTIVEC_COMP)@@ -443,6 +446,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER8 (architected)", .cpu_features = CPU_FTRS_POWER8, .cpu_user_features = COMMON_USER_POWER8, + .cpu_user_features2 = COMMON_USER2_POWER8, .mmu_features = MMU_FTRS_POWER8, .icache_bsize = 128, .dcache_bsize = 128,@@ -492,6 +496,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER8 (raw)", .cpu_features = CPU_FTRS_POWER8, .cpu_user_features = COMMON_USER_POWER8, + .cpu_user_features2 = COMMON_USER2_POWER8, .mmu_features = MMU_FTRS_POWER8, .icache_bsize = 128, .dcache_bsize = 128,
PATCH v2 1/4 powerpc/cputable: reserve bits in HWCAP2 for new features
Also, make HTM's presence dependent on the .config option.
---
Changes since v1:
- Add TAR.
- Make HTM config dependent.
arch/powerpc/include/asm/cputable.h +2 -0
@@ -224,8 +224,10 @@ extern const char *powerpc_base_platform;/* We only set the TM feature if the kernel was compiled with TM supprt */ #ifdef CONFIG_PPC_TRANSACTIONAL_MEM #define CPU_FTR_TM_COMP CPU_FTR_TM +#define PPC_FEATURE2_HTM_COMP PPC_FEATURE2_HTM#else #define CPU_FTR_TM_COMP 0 +#define PPC_FEATURE2_HTM_COMP 0#endif /* We need to mark all pages as being coherent if we're SMP or we have a
arch/powerpc/include/uapi/asm/cputable.h +9 -0
@@ -1,6 +1,7 @@#ifndef _UAPI__ASM_POWERPC_CPUTABLE_H#define _UAPI__ASM_POWERPC_CPUTABLE_H +/* in AT_HWCAP */ #define PPC_FEATURE_32 0x80000000#define PPC_FEATURE_64 0x40000000#define PPC_FEATURE_601_INSTR 0x20000000
@@ -33,4 +34,12 @@#define PPC_FEATURE_TRUE_LE 0x00000002#define PPC_FEATURE_PPC_LE 0x00000001 +/* in AT_HWCAP2 */ +#define PPC_FEATURE2_ARCH_2_07 0x80000000 +#define PPC_FEATURE2_HTM 0x40000000 +#define PPC_FEATURE2_DSCR 0x20000000 +#define PPC_FEATURE2_EBB 0x10000000 +#define PPC_FEATURE2_ISEL 0x08000000 +#define PPC_FEATURE2_TAR 0x04000000 +#endif /* _UAPI__ASM_POWERPC_CPUTABLE_H */
PATCH v2 2/4 powerpc/cputable: advertise DSCR support on P7/P7+
arch/powerpc/kernel/cputable.c +4 -0
@@ -98,6 +98,7 @@ extern void __restore_cpu_e6500(void); PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \ PPC_FEATURE_TRUE_LE | \ PPC_FEATURE_PSERIES_PERFMON_COMPAT) +#define COMMON_USER2_POWER7 (PPC_FEATURE2_DSCR)#define COMMON_USER_POWER8 (COMMON_USER_PPC64 | PPC_FEATURE_ARCH_2_06 |\ PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \ PPC_FEATURE_TRUE_LE | \
@@ -428,6 +429,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7 (architected)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, + .cpu_user_features2 = COMMON_USER2_POWER7, .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,
@@ -458,6 +460,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7 (raw)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, + .cpu_user_features2 = COMMON_USER2_POWER7, .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,
@@ -475,6 +478,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7+ (raw)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, + .cpu_user_features = COMMON_USER2_POWER7, .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,
PATCH v2 3/4 powerpc/cputable: advertise ISEL support on appropriate embedded processors
arch/powerpc/kernel/cputable.c +4 -0
@@ -1999,6 +1999,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_user_features = COMMON_USER_BOOKE |PPC_FEATURE_HAS_SPE_COMP | PPC_FEATURE_HAS_EFP_SINGLE_COMP, + .cpu_user_features2 = PPC_FEATURE2_ISEL, .mmu_features = MMU_FTR_TYPE_FSL_E, .icache_bsize = 32, .dcache_bsize = 32,
@@ -2018,6 +2019,7 @@ static struct cpu_spec __initdata cpu_specs[] = {PPC_FEATURE_HAS_SPE_COMP |PPC_FEATURE_HAS_EFP_SINGLE_COMP | PPC_FEATURE_HAS_EFP_DOUBLE_COMP, + .cpu_user_features2 = PPC_FEATURE2_ISEL, .mmu_features = MMU_FTR_TYPE_FSL_E | MMU_FTR_BIG_PHYS, .icache_bsize = 32, .dcache_bsize = 32,
@@ -2034,6 +2036,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "e500mc", .cpu_features = CPU_FTRS_E500MC, .cpu_user_features = COMMON_USER_BOOKE | PPC_FEATURE_HAS_FPU, + .cpu_user_features2 = PPC_FEATURE2_ISEL, .mmu_features =MMU_FTR_TYPE_FSL_E |MMU_FTR_BIG_PHYS | MMU_FTR_USE_TLBILX, .icache_bsize = 64,
@@ -2073,6 +2076,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_features = CPU_FTRS_E6500, .cpu_user_features =COMMON_USER_BOOKE |PPC_FEATURE_HAS_FPU | PPC_FEATURE_HAS_ALTIVEC_COMP, + .cpu_user_features2 = PPC_FEATURE2_ISEL, .mmu_features =MMU_FTR_TYPE_FSL_E |MMU_FTR_BIG_PHYS | MMU_FTR_USE_TLBILX, .icache_bsize = 64,
PATCH v2 4/4 powerpc/cputable: advertise support for ISEL/HTM/DSCR/TAR on POWER8
---
Changes since v1:
- Add TAR.
- Use config dependent symbol for HTM.
arch/powerpc/kernel/cputable.c +5 -0
@@ -103,6 +103,9 @@ extern void __restore_cpu_e6500(void); PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \ PPC_FEATURE_TRUE_LE | \ PPC_FEATURE_PSERIES_PERFMON_COMPAT) +#define COMMON_USER2_POWER8 (PPC_FEATURE2_ARCH_2_07 | \ + PPC_FEATURE2_HTM_COMP | PPC_FEATURE2_DSCR | \ + PPC_FEATURE2_ISEL | PPC_FEATURE2_TAR)#define COMMON_USER_PA6T (COMMON_USER_PPC64 | PPC_FEATURE_PA6T |\ PPC_FEATURE_TRUE_LE | \ PPC_FEATURE_HAS_ALTIVEC_COMP)
@@ -445,6 +448,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER8 (architected)", .cpu_features = CPU_FTRS_POWER8, .cpu_user_features = COMMON_USER_POWER8, + .cpu_user_features2 = COMMON_USER2_POWER8, .mmu_features = MMU_FTRS_POWER8, .icache_bsize = 128, .dcache_bsize = 128,
@@ -496,6 +500,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER8 (raw)", .cpu_features = CPU_FTRS_POWER8, .cpu_user_features = COMMON_USER_POWER8, + .cpu_user_features2 = COMMON_USER2_POWER8, .mmu_features = MMU_FTRS_POWER8, .icache_bsize = 128, .dcache_bsize = 128,
PATCH v3 3/4 powerpc/cputable: advertise ISEL support on appropriate embedded processors
In reply to Nishanth Aravamudan---
Changes since v2:
- Added e5500 per Segher.
I've not added POWER7 yet, as I'm waiting to confirm with others on that. I will send it as a follow-up patch if needed.
arch/powerpc/kernel/cputable.c +5 -0
@@ -1999,6 +1999,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_user_features = COMMON_USER_BOOKE |PPC_FEATURE_HAS_SPE_COMP | PPC_FEATURE_HAS_EFP_SINGLE_COMP, + .cpu_user_features2 = PPC_FEATURE2_ISEL, .mmu_features = MMU_FTR_TYPE_FSL_E, .icache_bsize = 32, .dcache_bsize = 32,
@@ -2018,6 +2019,7 @@ static struct cpu_spec __initdata cpu_specs[] = {PPC_FEATURE_HAS_SPE_COMP |PPC_FEATURE_HAS_EFP_SINGLE_COMP | PPC_FEATURE_HAS_EFP_DOUBLE_COMP, + .cpu_user_features2 = PPC_FEATURE2_ISEL, .mmu_features = MMU_FTR_TYPE_FSL_E | MMU_FTR_BIG_PHYS, .icache_bsize = 32, .dcache_bsize = 32,
@@ -2034,6 +2036,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "e500mc", .cpu_features = CPU_FTRS_E500MC, .cpu_user_features = COMMON_USER_BOOKE | PPC_FEATURE_HAS_FPU, + .cpu_user_features2 = PPC_FEATURE2_ISEL, .mmu_features =MMU_FTR_TYPE_FSL_E |MMU_FTR_BIG_PHYS | MMU_FTR_USE_TLBILX, .icache_bsize = 64,
@@ -2052,6 +2055,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "e5500", .cpu_features = CPU_FTRS_E5500, .cpu_user_features = COMMON_USER_BOOKE | PPC_FEATURE_HAS_FPU, + .cpu_user_features2 = PPC_FEATURE2_ISEL, .mmu_features =MMU_FTR_TYPE_FSL_E |MMU_FTR_BIG_PHYS | MMU_FTR_USE_TLBILX, .icache_bsize = 64,
@@ -2073,6 +2077,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_features = CPU_FTRS_E6500, .cpu_user_features =COMMON_USER_BOOKE |PPC_FEATURE_HAS_FPU | PPC_FEATURE_HAS_ALTIVEC_COMP, + .cpu_user_features2 = PPC_FEATURE2_ISEL, .mmu_features =MMU_FTR_TYPE_FSL_E |MMU_FTR_BIG_PHYS | MMU_FTR_USE_TLBILX, .icache_bsize = 64,
Re: [PATCH v2 2/4] powerpc/cputable: advertise DSCR support on P7/P7+
In reply to Nishanth AravamudanQuoted 31 lines · Nishanth Aravamudan
Signed-off-by Nishanth Aravamudan <[email protected]>arch/powerpc/kernel/cputable.c +3
-0@@ -98,6 +98,7 @@ extern void __restore_cpu_e6500(void); PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \ PPC_FEATURE_TRUE_LE | \ PPC_FEATURE_PSERIES_PERFMON_COMPAT) +#define COMMON_USER2_POWER7 (PPC_FEATURE2_DSCR)#define COMMON_USER_POWER8 (COMMON_USER_PPC64 | PPC_FEATURE_ARCH_2_06 |\ PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \ PPC_FEATURE_TRUE_LE | \@@ -428,6 +429,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7 (architected)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, + .cpu_user_features2 = COMMON_USER2_POWER7, .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,@@ -458,6 +460,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7 (raw)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, + .cpu_user_features2 = COMMON_USER2_POWER7, .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,
arch/powerpc/kernel/cputable.c static struct cpu_spec __initdata cpu_specs[] = { · line 481
Quoted 4 lines
arch/powerpc/kernel/cputable.c +1
-0@@ -475,6 +478,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7+ (raw)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, + .cpu_user_features = COMMON_USER2_POWER7,
^ Oops here, I think. Please consider applying this on top.
Untested, but seems obvious.
Fix a typo in setting COMMON_USER2_POWER7 bits to .cpu_user_features2 cpu specs table.
(quoted diff) +0 -0
b/arch/powerpc/kernel/cputable.c
index c60bbec..51eecb5 100644
arch/powerpc/kernel/cputable.c +1 -1
@@ -482,7 +482,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7+ (raw)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, - .cpu_user_features = COMMON_USER2_POWER7, + .cpu_user_features2 = COMMON_USER2_POWER7, .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,
Quoted 9 lines
arch/powerpc/kernel/cputable.c +0
-0.mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,_______________________________________________
Linuxppc-dev mailing list
[email protected]
https://lists.ozlabs.org/listinfo/linuxppc-dev
Michael Neuling
Acked-byRe: [PATCH v2 2/4] powerpc/cputable: advertise DSCR support on P7/P7+
Quoted 40 lines · Will Schmidt
Quoted 31 lines · Nishanth Aravamudan
Signed-off-by Nishanth Aravamudan <[email protected]>arch/powerpc/kernel/cputable.c +3
-0@@ -98,6 +98,7 @@ extern void __restore_cpu_e6500(void); PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \ PPC_FEATURE_TRUE_LE | \ PPC_FEATURE_PSERIES_PERFMON_COMPAT) +#define COMMON_USER2_POWER7 (PPC_FEATURE2_DSCR)#define COMMON_USER_POWER8 (COMMON_USER_PPC64 | PPC_FEATURE_ARCH_2_06 |\ PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \ PPC_FEATURE_TRUE_LE | \@@ -428,6 +429,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7 (architected)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, + .cpu_user_features2 = COMMON_USER2_POWER7, .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,@@ -458,6 +460,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7 (raw)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, + .cpu_user_features2 = COMMON_USER2_POWER7, .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,arch/powerpc/kernel/cputable.c static struct cpu_spec __initdata cpu_specs[] = { · line 481
Quoted 4 lines
arch/powerpc/kernel/cputable.c +1
-0@@ -475,6 +478,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7+ (raw)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, + .cpu_user_features = COMMON_USER2_POWER7,^ Oops here, I think. Please consider applying this on top.
Untested, but seems obvious.
Quoted 2 lines
Fix a typo in setting
COMMON_USER2_POWER7bits to .cpu_user_features2 cpu specs table.
Nice catch. Thanks
Quoted 46 lines
Signed-off-by Will Schmidt <[email protected]>arch/powerpc/kernel/cputable.c +1
-1@@ -482,7 +482,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7+ (raw)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, - .cpu_user_features = COMMON_USER2_POWER7, + .cpu_user_features2 = COMMON_USER2_POWER7, .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,Quoted 26 lines
arch/powerpc/kernel/cputable.c +0
-0.mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128,_______________________________________________
Linuxppc-dev mailing list
[email protected]
https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH v2 1/4] powerpc/cputable: reserve bits in HWCAP2 for new features
In reply to Nishanth AravamudanHi Nish,
Nishanth Aravamudan <[email protected]> wrote on 05/03/2013 07:47:56 PM:
arch/powerpc/include/uapi/asm/cputable.h line 45
Quoted 9 lines
arch/powerpc/include/uapi/asm/cputable.h +8
-0+/* in AT_HWCAP2 */ +#define PPC_FEATURE2_ARCH_2_07 0x80000000 +#define PPC_FEATURE2_HTM 0x40000000 +#define PPC_FEATURE2_DSCR 0x20000000 +#define PPC_FEATURE2_EBB 0x10000000 +#define PPC_FEATURE2_ISEL 0x08000000 +#define PPC_FEATURE2_TAR 0x04000000 +#endif /* _UAPI__ASM_POWERPC_CPUTABLE_H */
Following the existing naming convention in cputable.h, the features should probably be amended to the following:
#define PPC_FEATURE2_ARCH_2_07 0x80000000 #define PPC_FEATURE2_HAS_HTM 0x40000000 #define PPC_FEATURE2_HAS_DSC 0x20000000 #define PPC_FEATURE2_HAS_EBB 0x10000000 #define PPC_FEATURE2_HAS_ISEL 0x08000000 #define PPC_FEATURE2_HAS_TAR 0x04000000
Notice that I changed DSCR to DSC. The 'R' wasn't descriptive.
Ryan S. Arnold
IBM Linux Technology Center
Re: [PATCH v2 1/4] powerpc/cputable: reserve bits in HWCAP2 for new features
Quoted 1 line · Ryan Arnold
Notice that I changed DSCR to DSC. The 'R' wasn't descriptive.
The "R" is the name of the register for which we are exposing the availability to userspace... it's also the name of the sysfs entry so I'd rather keep it for consistency.
Cheers,
Ben.
Re: [PATCH v2 1/4] powerpc/cputable: reserve bits in HWCAP2 for new features
Benjamin Herrenschmidt <[email protected]> wrote on 05/06/2013 04:41:01 PM:
Quoted 9 lines
Benjamin Herrenschmidt <[email protected]>
05/06/2013 04:41 PMTo
Ryan Arnold/Rochester/IBM@IBMUS
cc
Quoted 8 lines
Nishanth Aravamudan <[email protected]>, linuxppc- [email protected], [email protected], Michael R Meissner/ Cambridge/IBM@IBMUS, Michael Neuling <[email protected]>, Peter Bergner/Rochester/IBM@IBMUS, Steve Munroe/Rochester/IBM@IBMUS
Subject
Re: [PATCH v2 1/4] powerpc/cputable: reserve bits in HWCAP2 for new
features
Quoted 7 lines · Ryan Arnold
Quoted 1 line
Notice that I changed DSCR to DSC. The 'R' wasn't descriptive.
The "R" is the name of the register for which we are exposing the availability to userspace... it's also the name of the sysfs entry so I'd rather keep it for consistency.
I'm fine with keeping the 'R' in the name. Thanks for the input.
Ryan