diff options
author | Nicholas Piggin <npiggin@gmail.com> | 2018-02-20 22:08:29 +0300 |
---|---|---|
committer | Michael Ellerman <mpe@ellerman.id.au> | 2018-03-31 16:47:50 +0300 |
commit | 471d7ff8b51b63521c8ea35c51966ab4caa434ee (patch) | |
tree | 697babe0784be6808eeb49f12cba44dc937af05e | |
parent | 3735eb850e6c3e7472329fea368e7d030a4d89dd (diff) | |
download | linux-471d7ff8b51b63521c8ea35c51966ab4caa434ee.tar.xz |
powerpc/64s: Remove POWER4 support
POWER4 has been broken since at least the change 49d09bf2a6
("powerpc/64s: Optimise MSR handling in exception handling"), which
requires mtmsrd L=1 support. This was introduced in ISA v2.01, and
POWER4 supports ISA v2.00.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
-rw-r--r-- | arch/powerpc/Makefile | 5 | ||||
-rw-r--r-- | arch/powerpc/include/asm/cputable.h | 8 | ||||
-rw-r--r-- | arch/powerpc/include/asm/mmu.h | 6 | ||||
-rw-r--r-- | arch/powerpc/kernel/cputable.c | 36 | ||||
-rw-r--r-- | arch/powerpc/kernel/prom_init.c | 10 | ||||
-rw-r--r-- | arch/powerpc/kvm/emulate.c | 6 | ||||
-rw-r--r-- | arch/powerpc/mm/hash_utils_64.c | 9 | ||||
-rw-r--r-- | arch/powerpc/perf/Makefile | 2 | ||||
-rw-r--r-- | arch/powerpc/perf/power4-pmu.c | 622 | ||||
-rw-r--r-- | arch/powerpc/platforms/Kconfig.cputype | 6 |
10 files changed, 18 insertions, 692 deletions
diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile index af46d22fa660..373d7f5b7cde 100644 --- a/arch/powerpc/Makefile +++ b/arch/powerpc/Makefile @@ -146,8 +146,8 @@ CFLAGS-$(CONFIG_PPC32) := -ffixed-r2 $(MULTIPLEWORD) CFLAGS-$(CONFIG_PPC32) += $(call cc-option,-mno-readonly-in-sdata) ifeq ($(CONFIG_PPC_BOOK3S_64),y) -CFLAGS-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mtune=power7,-mtune=power4) -CFLAGS-$(CONFIG_GENERIC_CPU) += -mcpu=power4 +CFLAGS-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mtune=power7,$(call cc-option,-mtune=power5)) +CFLAGS-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mcpu=power5,-mcpu=power4) else CFLAGS-$(CONFIG_GENERIC_CPU) += -mcpu=powerpc64 endif @@ -168,7 +168,6 @@ ifdef CONFIG_MPROFILE_KERNEL endif CFLAGS-$(CONFIG_CELL_CPU) += $(call cc-option,-mcpu=cell) -CFLAGS-$(CONFIG_POWER4_CPU) += $(call cc-option,-mcpu=power4) CFLAGS-$(CONFIG_POWER5_CPU) += $(call cc-option,-mcpu=power5) CFLAGS-$(CONFIG_POWER6_CPU) += $(call cc-option,-mcpu=power6) CFLAGS-$(CONFIG_POWER7_CPU) += $(call cc-option,-mcpu=power7) diff --git a/arch/powerpc/include/asm/cputable.h b/arch/powerpc/include/asm/cputable.h index 17c791277d02..7e22607e07b4 100644 --- a/arch/powerpc/include/asm/cputable.h +++ b/arch/powerpc/include/asm/cputable.h @@ -413,10 +413,6 @@ static inline void cpu_feature_keys_init(void) { } #define CPU_FTRS_GENERIC_32 (CPU_FTR_COMMON | CPU_FTR_NODSISRALIGN) /* 64-bit CPUs */ -#define CPU_FTRS_POWER4 (CPU_FTR_LWSYNC | \ - CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_CTRL | \ - CPU_FTR_MMCRA | CPU_FTR_CP_USE_DCBTZ | \ - CPU_FTR_STCX_CHECKS_ADDRESS) #define CPU_FTRS_PPC970 (CPU_FTR_LWSYNC | \ CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_CTRL | \ CPU_FTR_ALTIVEC_COMP | CPU_FTR_CAN_NAP | CPU_FTR_MMCRA | \ @@ -488,7 +484,7 @@ static inline void cpu_feature_keys_init(void) { } #define CPU_FTRS_POSSIBLE (CPU_FTRS_E6500 | CPU_FTRS_E5500) #else #define CPU_FTRS_POSSIBLE \ - (CPU_FTRS_POWER4 | CPU_FTRS_PPC970 | CPU_FTRS_POWER5 | \ + (CPU_FTRS_PPC970 | CPU_FTRS_POWER5 | \ CPU_FTRS_POWER6 | CPU_FTRS_POWER7 | CPU_FTRS_POWER8E | \ CPU_FTRS_POWER8 | CPU_FTRS_POWER8_DD1 | CPU_FTRS_CELL | \ CPU_FTRS_PA6T | CPU_FTR_VSX_COMP | CPU_FTR_ALTIVEC_COMP | \ @@ -541,7 +537,7 @@ enum { #define CPU_FTRS_ALWAYS (CPU_FTRS_E6500 & CPU_FTRS_E5500) #else #define CPU_FTRS_ALWAYS \ - (CPU_FTRS_POWER4 & CPU_FTRS_PPC970 & CPU_FTRS_POWER5 & \ + (CPU_FTRS_PPC970 & CPU_FTRS_POWER5 & \ CPU_FTRS_POWER6 & CPU_FTRS_POWER7 & CPU_FTRS_CELL & \ CPU_FTRS_PA6T & CPU_FTRS_POWER8 & CPU_FTRS_POWER8E & \ CPU_FTRS_POWER8_DD1 & ~CPU_FTR_HVMODE & CPU_FTRS_POSSIBLE & \ diff --git a/arch/powerpc/include/asm/mmu.h b/arch/powerpc/include/asm/mmu.h index bb38312cff28..61d15ce92278 100644 --- a/arch/powerpc/include/asm/mmu.h +++ b/arch/powerpc/include/asm/mmu.h @@ -111,9 +111,9 @@ /* MMU feature bit sets for various CPUs */ #define MMU_FTRS_DEFAULT_HPTE_ARCH_V2 \ MMU_FTR_HPTE_TABLE | MMU_FTR_PPCAS_ARCH_V2 -#define MMU_FTRS_POWER4 MMU_FTRS_DEFAULT_HPTE_ARCH_V2 -#define MMU_FTRS_PPC970 MMU_FTRS_POWER4 | MMU_FTR_TLBIE_CROP_VA -#define MMU_FTRS_POWER5 MMU_FTRS_POWER4 | MMU_FTR_LOCKLESS_TLBIE +#define MMU_FTRS_POWER MMU_FTRS_DEFAULT_HPTE_ARCH_V2 +#define MMU_FTRS_PPC970 MMU_FTRS_POWER | MMU_FTR_TLBIE_CROP_VA +#define MMU_FTRS_POWER5 MMU_FTRS_POWER | MMU_FTR_LOCKLESS_TLBIE #define MMU_FTRS_POWER6 MMU_FTRS_POWER5 | MMU_FTR_KERNEL_RO | MMU_FTR_68_BIT_VA #define MMU_FTRS_POWER7 MMU_FTRS_POWER6 #define MMU_FTRS_POWER8 MMU_FTRS_POWER6 diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index b3de017bcd71..c8fc9691f8c7 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -133,36 +133,6 @@ extern void __restore_cpu_e6500(void); static struct cpu_spec __initdata cpu_specs[] = { #ifdef CONFIG_PPC_BOOK3S_64 - { /* Power4 */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00350000, - .cpu_name = "POWER4 (gp)", - .cpu_features = CPU_FTRS_POWER4, - .cpu_user_features = COMMON_USER_POWER4, - .mmu_features = MMU_FTRS_POWER4 | MMU_FTR_TLBIE_CROP_VA, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .pmc_type = PPC_PMC_IBM, - .oprofile_cpu_type = "ppc64/power4", - .oprofile_type = PPC_OPROFILE_POWER4, - .platform = "power4", - }, - { /* Power4+ */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00380000, - .cpu_name = "POWER4+ (gq)", - .cpu_features = CPU_FTRS_POWER4, - .cpu_user_features = COMMON_USER_POWER4, - .mmu_features = MMU_FTRS_POWER4 | MMU_FTR_TLBIE_CROP_VA, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .pmc_type = PPC_PMC_IBM, - .oprofile_cpu_type = "ppc64/power4", - .oprofile_type = PPC_OPROFILE_POWER4, - .platform = "power4", - }, { /* PPC970 */ .pvr_mask = 0xffff0000, .pvr_value = 0x00390000, @@ -628,15 +598,15 @@ static struct cpu_spec __initdata cpu_specs[] = { { /* default match */ .pvr_mask = 0x00000000, .pvr_value = 0x00000000, - .cpu_name = "POWER4 (compatible)", + .cpu_name = "POWER5 (compatible)", .cpu_features = CPU_FTRS_COMPATIBLE, .cpu_user_features = COMMON_USER_PPC64, - .mmu_features = MMU_FTRS_DEFAULT_HPTE_ARCH_V2, + .mmu_features = MMU_FTRS_POWER, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 6, .pmc_type = PPC_PMC_IBM, - .platform = "power4", + .platform = "power5", } #endif /* CONFIG_PPC_BOOK3S_64 */ diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index 0323e073341d..e181fdea3da9 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -1810,16 +1810,8 @@ static void __init prom_initialize_tce_table(void) * size to 4 MB. This is enough to map 2GB of PCI DMA space. * By doing this, we avoid the pitfalls of trying to DMA to * MMIO space and the DMA alias hole. - * - * On POWER4, firmware sets the TCE region by assuming - * each TCE table is 8MB. Using this memory for anything - * else will impact performance, so we always allocate 8MB. - * Anton */ - if (pvr_version_is(PVR_POWER4) || pvr_version_is(PVR_POWER4p)) - minsize = 8UL << 20; - else - minsize = 4UL << 20; + minsize = 4UL << 20; /* Align to the greater of the align or size */ align = max(minalign, minsize); diff --git a/arch/powerpc/kvm/emulate.c b/arch/powerpc/kvm/emulate.c index 4d8b4d6cebff..fa888bfc347e 100644 --- a/arch/powerpc/kvm/emulate.c +++ b/arch/powerpc/kvm/emulate.c @@ -45,12 +45,6 @@ void kvmppc_emulate_dec(struct kvm_vcpu *vcpu) #ifdef CONFIG_PPC_BOOK3S /* mtdec lowers the interrupt line when positive. */ kvmppc_core_dequeue_dec(vcpu); - - /* POWER4+ triggers a dec interrupt if the value is < 0 */ - if (vcpu->arch.dec & 0x80000000) { - kvmppc_core_queue_dec(vcpu); - return; - } #endif #ifdef CONFIG_BOOKE diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c index 7587a2ec8874..0bd3790d35df 100644 --- a/arch/powerpc/mm/hash_utils_64.c +++ b/arch/powerpc/mm/hash_utils_64.c @@ -132,9 +132,10 @@ EXPORT_SYMBOL(mmu_hash_ops); * is provided by the firmware. */ -/* Pre-POWER4 CPUs (4k pages only) +/* + * Fallback (4k pages only) */ -static struct mmu_psize_def mmu_psize_defaults_old[] = { +static struct mmu_psize_def mmu_psize_defaults[] = { [MMU_PAGE_4K] = { .shift = 12, .sllp = 0, @@ -554,8 +555,8 @@ static void __init htab_scan_page_sizes(void) mmu_psize_set_default_penc(); /* Default to 4K pages only */ - memcpy(mmu_psize_defs, mmu_psize_defaults_old, - sizeof(mmu_psize_defaults_old)); + memcpy(mmu_psize_defs, mmu_psize_defaults, + sizeof(mmu_psize_defaults)); /* * Try to find the available page sizes in the device-tree diff --git a/arch/powerpc/perf/Makefile b/arch/powerpc/perf/Makefile index 57ebc655d2ac..82986d2acd9b 100644 --- a/arch/powerpc/perf/Makefile +++ b/arch/powerpc/perf/Makefile @@ -4,7 +4,7 @@ subdir-ccflags-$(CONFIG_PPC_WERROR) := -Werror obj-$(CONFIG_PERF_EVENTS) += callchain.o perf_regs.o obj-$(CONFIG_PPC_PERF_CTRS) += core-book3s.o bhrb.o -obj64-$(CONFIG_PPC_PERF_CTRS) += power4-pmu.o ppc970-pmu.o power5-pmu.o \ +obj64-$(CONFIG_PPC_PERF_CTRS) += ppc970-pmu.o power5-pmu.o \ power5+-pmu.o power6-pmu.o power7-pmu.o \ isa207-common.o power8-pmu.o power9-pmu.o obj32-$(CONFIG_PPC_PERF_CTRS) += mpc7450-pmu.o diff --git a/arch/powerpc/perf/power4-pmu.c b/arch/powerpc/perf/power4-pmu.c deleted file mode 100644 index ce6072fa481b..000000000000 --- a/arch/powerpc/perf/power4-pmu.c +++ /dev/null @@ -1,622 +0,0 @@ -/* - * Performance counter support for POWER4 (GP) and POWER4+ (GQ) processors. - * - * Copyright 2009 Paul Mackerras, IBM Corporation. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <linux/kernel.h> -#include <linux/perf_event.h> -#include <linux/string.h> -#include <asm/reg.h> -#include <asm/cputable.h> - -/* - * Bits in event code for POWER4 - */ -#define PM_PMC_SH 12 /* PMC number (1-based) for direct events */ -#define PM_PMC_MSK 0xf -#define PM_UNIT_SH 8 /* TTMMUX number and setting - unit select */ -#define PM_UNIT_MSK 0xf -#define PM_LOWER_SH 6 -#define PM_LOWER_MSK 1 -#define PM_LOWER_MSKS 0x40 -#define PM_BYTE_SH 4 /* Byte number of event bus to use */ -#define PM_BYTE_MSK 3 -#define PM_PMCSEL_MSK 7 - -/* - * Unit code values - */ -#define PM_FPU 1 -#define PM_ISU1 2 -#define PM_IFU 3 -#define PM_IDU0 4 -#define PM_ISU1_ALT 6 -#define PM_ISU2 7 -#define PM_IFU_ALT 8 -#define PM_LSU0 9 -#define PM_LSU1 0xc -#define PM_GPS 0xf - -/* - * Bits in MMCR0 for POWER4 - */ -#define MMCR0_PMC1SEL_SH 8 -#define MMCR0_PMC2SEL_SH 1 -#define MMCR_PMCSEL_MSK 0x1f - -/* - * Bits in MMCR1 for POWER4 - */ -#define MMCR1_TTM0SEL_SH 62 -#define MMCR1_TTC0SEL_SH 61 -#define MMCR1_TTM1SEL_SH 59 -#define MMCR1_TTC1SEL_SH 58 -#define MMCR1_TTM2SEL_SH 56 -#define MMCR1_TTC2SEL_SH 55 -#define MMCR1_TTM3SEL_SH 53 -#define MMCR1_TTC3SEL_SH 52 -#define MMCR1_TTMSEL_MSK 3 -#define MMCR1_TD_CP_DBG0SEL_SH 50 -#define MMCR1_TD_CP_DBG1SEL_SH 48 -#define MMCR1_TD_CP_DBG2SEL_SH 46 -#define MMCR1_TD_CP_DBG3SEL_SH 44 -#define MMCR1_DEBUG0SEL_SH 43 -#define MMCR1_DEBUG1SEL_SH 42 -#define MMCR1_DEBUG2SEL_SH 41 -#define MMCR1_DEBUG3SEL_SH 40 -#define MMCR1_PMC1_ADDER_SEL_SH 39 -#define MMCR1_PMC2_ADDER_SEL_SH 38 -#define MMCR1_PMC6_ADDER_SEL_SH 37 -#define MMCR1_PMC5_ADDER_SEL_SH 36 -#define MMCR1_PMC8_ADDER_SEL_SH 35 -#define MMCR1_PMC7_ADDER_SEL_SH 34 -#define MMCR1_PMC3_ADDER_SEL_SH 33 -#define MMCR1_PMC4_ADDER_SEL_SH 32 -#define MMCR1_PMC3SEL_SH 27 -#define MMCR1_PMC4SEL_SH 22 -#define MMCR1_PMC5SEL_SH 17 -#define MMCR1_PMC6SEL_SH 12 -#define MMCR1_PMC7SEL_SH 7 -#define MMCR1_PMC8SEL_SH 2 /* note bit 0 is in MMCRA for GP */ - -static short mmcr1_adder_bits[8] = { - MMCR1_PMC1_ADDER_SEL_SH, - MMCR1_PMC2_ADDER_SEL_SH, - MMCR1_PMC3_ADDER_SEL_SH, - MMCR1_PMC4_ADDER_SEL_SH, - MMCR1_PMC5_ADDER_SEL_SH, - MMCR1_PMC6_ADDER_SEL_SH, - MMCR1_PMC7_ADDER_SEL_SH, - MMCR1_PMC8_ADDER_SEL_SH -}; - -/* - * Bits in MMCRA - */ -#define MMCRA_PMC8SEL0_SH 17 /* PMC8SEL bit 0 for GP */ - -/* - * Layout of constraint bits: - * 6666555555555544444444443333333333222222222211111111110000000000 - * 3210987654321098765432109876543210987654321098765432109876543210 - * |[ >[ >[ >|||[ >[ >< >< >< >< ><><><><><><><><> - * | UC1 UC2 UC3 ||| PS1 PS2 B0 B1 B2 B3 P1P2P3P4P5P6P7P8 - * \SMPL ||\TTC3SEL - * |\TTC_IFU_SEL - * \TTM2SEL0 - * - * SMPL - SAMPLE_ENABLE constraint - * 56: SAMPLE_ENABLE value 0x0100_0000_0000_0000 - * - * UC1 - unit constraint 1: can't have all three of FPU/ISU1/IDU0|ISU2 - * 55: UC1 error 0x0080_0000_0000_0000 - * 54: FPU events needed 0x0040_0000_0000_0000 - * 53: ISU1 events needed 0x0020_0000_0000_0000 - * 52: IDU0|ISU2 events needed 0x0010_0000_0000_0000 - * - * UC2 - unit constraint 2: can't have all three of FPU/IFU/LSU0 - * 51: UC2 error 0x0008_0000_0000_0000 - * 50: FPU events needed 0x0004_0000_0000_0000 - * 49: IFU events needed 0x0002_0000_0000_0000 - * 48: LSU0 events needed 0x0001_0000_0000_0000 - * - * UC3 - unit constraint 3: can't have all four of LSU0/IFU/IDU0|ISU2/ISU1 - * 47: UC3 error 0x8000_0000_0000 - * 46: LSU0 events needed 0x4000_0000_0000 - * 45: IFU events needed 0x2000_0000_0000 - * 44: IDU0|ISU2 events needed 0x1000_0000_0000 - * 43: ISU1 events needed 0x0800_0000_0000 - * - * TTM2SEL0 - * 42: 0 = IDU0 events needed - * 1 = ISU2 events needed 0x0400_0000_0000 - * - * TTC_IFU_SEL - * 41: 0 = IFU.U events needed - * 1 = IFU.L events needed 0x0200_0000_0000 - * - * TTC3SEL - * 40: 0 = LSU1.U events needed - * 1 = LSU1.L events needed 0x0100_0000_0000 - * - * PS1 - * 39: PS1 error 0x0080_0000_0000 - * 36-38: count of events needing PMC1/2/5/6 0x0070_0000_0000 - * - * PS2 - * 35: PS2 error 0x0008_0000_0000 - * 32-34: count of events needing PMC3/4/7/8 0x0007_0000_0000 - * - * B0 - * 28-31: Byte 0 event source 0xf000_0000 - * 1 = FPU - * 2 = ISU1 - * 3 = IFU - * 4 = IDU0 - * 7 = ISU2 - * 9 = LSU0 - * c = LSU1 - * f = GPS - * - * B1, B2, B3 - * 24-27, 20-23, 16-19: Byte 1, 2, 3 event sources - * - * P8 - * 15: P8 error 0x8000 - * 14-15: Count of events needing PMC8 - * - * P1..P7 - * 0-13: Count of events needing PMC1..PMC7 - * - * Note: this doesn't allow events using IFU.U to be combined with events - * using IFU.L, though that is feasible (using TTM0 and TTM2). However - * there are no listed events for IFU.L (they are debug events not - * verified for performance monitoring) so this shouldn't cause a - * problem. - */ - -static struct unitinfo { - unsigned long value, mask; - int unit; - int lowerbit; -} p4_unitinfo[16] = { - [PM_FPU] = { 0x44000000000000ul, 0x88000000000000ul, PM_FPU, 0 }, - [PM_ISU1] = { 0x20080000000000ul, 0x88000000000000ul, PM_ISU1, 0 }, - [PM_ISU1_ALT] = - { 0x20080000000000ul, 0x88000000000000ul, PM_ISU1, 0 }, - [PM_IFU] = { 0x02200000000000ul, 0x08820000000000ul, PM_IFU, 41 }, - [PM_IFU_ALT] = - { 0x02200000000000ul, 0x08820000000000ul, PM_IFU, 41 }, - [PM_IDU0] = { 0x10100000000000ul, 0x80840000000000ul, PM_IDU0, 1 }, - [PM_ISU2] = { 0x10140000000000ul, 0x80840000000000ul, PM_ISU2, 0 }, - [PM_LSU0] = { 0x01400000000000ul, 0x08800000000000ul, PM_LSU0, 0 }, - [PM_LSU1] = { 0x00000000000000ul, 0x00010000000000ul, PM_LSU1, 40 }, - [PM_GPS] = { 0x00000000000000ul, 0x00000000000000ul, PM_GPS, 0 } -}; - -static unsigned char direct_marked_event[8] = { - (1<<2) | (1<<3), /* PMC1: PM_MRK_GRP_DISP, PM_MRK_ST_CMPL */ - (1<<3) | (1<<5), /* PMC2: PM_THRESH_TIMEO, PM_MRK_BRU_FIN */ - (1<<3), /* PMC3: PM_MRK_ST_CMPL_INT */ - (1<<4) | (1<<5), /* PMC4: PM_MRK_GRP_CMPL, PM_MRK_CRU_FIN */ - (1<<4) | (1<<5), /* PMC5: PM_MRK_GRP_TIMEO */ - (1<<3) | (1<<4) | (1<<5), - /* PMC6: PM_MRK_ST_GPS, PM_MRK_FXU_FIN, PM_MRK_GRP_ISSUED */ - (1<<4) | (1<<5), /* PMC7: PM_MRK_FPU_FIN, PM_MRK_INST_FIN */ - (1<<4), /* PMC8: PM_MRK_LSU_FIN */ -}; - -/* - * Returns 1 if event counts things relating to marked instructions - * and thus needs the MMCRA_SAMPLE_ENABLE bit set, or 0 if not. - */ -static int p4_marked_instr_event(u64 event) -{ - int pmc, psel, unit, byte, bit; - unsigned int mask; - - pmc = (event >> PM_PMC_SH) & PM_PMC_MSK; - psel = event & PM_PMCSEL_MSK; - if (pmc) { - if (direct_marked_event[pmc - 1] & (1 << psel)) - return 1; - if (psel == 0) /* add events */ - bit = (pmc <= 4)? pmc - 1: 8 - pmc; - else if (psel == 6) /* decode events */ - bit = 4; - else - return 0; - } else - bit = psel; - - byte = (event >> PM_BYTE_SH) & PM_BYTE_MSK; - unit = (event >> PM_UNIT_SH) & PM_UNIT_MSK; - mask = 0; - switch (unit) { - case PM_LSU1: - if (event & PM_LOWER_MSKS) - mask = 1 << 28; /* byte 7 bit 4 */ - else - mask = 6 << 24; /* byte 3 bits 1 and 2 */ - break; - case PM_LSU0: - /* byte 3, bit 3; byte 2 bits 0,2,3,4,5; byte 1 */ - mask = 0x083dff00; - } - return (mask >> (byte * 8 + bit)) & 1; -} - -static int p4_get_constraint(u64 event, unsigned long *maskp, - unsigned long *valp) -{ - int pmc, byte, unit, lower, sh; - unsigned long mask = 0, value = 0; - int grp = -1; - - pmc = (event >> PM_PMC_SH) & PM_PMC_MSK; - if (pmc) { - if (pmc > 8) - return -1; - sh = (pmc - 1) * 2; - mask |= 2 << sh; - value |= 1 << sh; - grp = ((pmc - 1) >> 1) & 1; - } - unit = (event >> PM_UNIT_SH) & PM_UNIT_MSK; - byte = (event >> PM_BYTE_SH) & PM_BYTE_MSK; - if (unit) { - lower = (event >> PM_LOWER_SH) & PM_LOWER_MSK; - - /* - * Bus events on bytes 0 and 2 can be counted - * on PMC1/2/5/6; bytes 1 and 3 on PMC3/4/7/8. - */ - if (!pmc) - grp = byte & 1; - - if (!p4_unitinfo[unit].unit) - return -1; - mask |= p4_unitinfo[unit].mask; - value |= p4_unitinfo[unit].value; - sh = p4_unitinfo[unit].lowerbit; - if (sh > 1) - value |= (unsigned long)lower << sh; - else if (lower != sh) - return -1; - unit = p4_unitinfo[unit].unit; - - /* Set byte lane select field */ - mask |= 0xfULL << (28 - 4 * byte); - value |= (unsigned long)unit << (28 - 4 * byte); - } - if (grp == 0) { - /* increment PMC1/2/5/6 field */ - mask |= 0x8000000000ull; - value |= 0x1000000000ull; - } else { - /* increment PMC3/4/7/8 field */ - mask |= 0x800000000ull; - value |= 0x100000000ull; - } - - /* Marked instruction events need sample_enable set */ - if (p4_marked_instr_event(event)) { - mask |= 1ull << 56; - value |= 1ull << 56; - } - - /* PMCSEL=6 decode events on byte 2 need sample_enable clear */ - if (pmc && (event & PM_PMCSEL_MSK) == 6 && byte == 2) - mask |= 1ull << 56; - - *maskp = mask; - *valp = value; - return 0; -} - -static unsigned int ppc_inst_cmpl[] = { - 0x1001, 0x4001, 0x6001, 0x7001, 0x8001 -}; - -static int p4_get_alternatives(u64 event, unsigned int flags, u64 alt[]) -{ - int i, j, na; - - alt[0] = event; - na = 1; - - /* 2 possibilities for PM_GRP_DISP_REJECT */ - if (event == 0x8003 || event == 0x0224) { - alt[1] = event ^ (0x8003 ^ 0x0224); - return 2; - } - - /* 2 possibilities for PM_ST_MISS_L1 */ - if (event == 0x0c13 || event == 0x0c23) { - alt[1] = event ^ (0x0c13 ^ 0x0c23); - return 2; - } - - /* several possibilities for PM_INST_CMPL */ - for (i = 0; i < ARRAY_SIZE(ppc_inst_cmpl); ++i) { - if (event == ppc_inst_cmpl[i]) { - for (j = 0; j < ARRAY_SIZE(ppc_inst_cmpl); ++j) - if (j != i) - alt[na++] = ppc_inst_cmpl[j]; - break; - } - } - - return na; -} - -static int p4_compute_mmcr(u64 event[], int n_ev, - unsigned int hwc[], unsigned long mmcr[], struct perf_event *pevents[]) -{ - unsigned long mmcr0 = 0, mmcr1 = 0, mmcra = 0; - unsigned int pmc, unit, byte, psel, lower; - unsigned int ttm, grp; - unsigned int pmc_inuse = 0; - unsigned int pmc_grp_use[2]; - unsigned char busbyte[4]; - unsigned char unituse[16]; - unsigned int unitlower = 0; - int i; - - if (n_ev > 8) - return -1; - - /* First pass to count resource use */ - pmc_grp_use[0] = pmc_grp_use[1] = 0; - memset(busbyte, 0, sizeof(busbyte)); - memset(unituse, 0, sizeof(unituse)); - for (i = 0; i < n_ev; ++i) { - pmc = (event[i] >> PM_PMC_SH) & PM_PMC_MSK; - if (pmc) { - if (pmc_inuse & (1 << (pmc - 1))) - return -1; - pmc_inuse |= 1 << (pmc - 1); - /* count 1/2/5/6 vs 3/4/7/8 use */ - ++pmc_grp_use[((pmc - 1) >> 1) & 1]; - } - unit = (event[i] >> PM_UNIT_SH) & PM_UNIT_MSK; - byte = (event[i] >> PM_BYTE_SH) & PM_BYTE_MSK; - lower = (event[i] >> PM_LOWER_SH) & PM_LOWER_MSK; - if (unit) { - if (!pmc) - ++pmc_grp_use[byte & 1]; - if (unit == 6 || unit == 8) - /* map alt ISU1/IFU codes: 6->2, 8->3 */ - unit = (unit >> 1) - 1; - if (busbyte[byte] && busbyte[byte] != unit) - return -1; - busbyte[byte] = unit; - lower <<= unit; - if (unituse[unit] && lower != (unitlower & lower)) - return -1; - unituse[unit] = 1; - unitlower |= lower; - } - } - if (pmc_grp_use[0] > 4 || pmc_grp_use[1] > 4) - return -1; - - /* - * Assign resources and set multiplexer selects. - * - * Units 1,2,3 are on TTM0, 4,6,7 on TTM1, 8,10 on TTM2. - * Each TTMx can only select one unit, but since - * units 2 and 6 are both ISU1, and 3 and 8 are both IFU, - * we have some choices. - */ - if (unituse[2] & (unituse[1] | (unituse[3] & unituse[9]))) { - unituse[6] = 1; /* Move 2 to 6 */ - unituse[2] = 0; - } - if (unituse[3] & (unituse[1] | unituse[2])) { - unituse[8] = 1; /* Move 3 to 8 */ - unituse[3] = 0; - unitlower = (unitlower & ~8) | ((unitlower & 8) << 5); - } - /* Check only one unit per TTMx */ - if (unituse[1] + unituse[2] + unituse[3] > 1 || - unituse[4] + unituse[6] + unituse[7] > 1 || - unituse[8] + unituse[9] > 1 || - (unituse[5] | unituse[10] | unituse[11] | - unituse[13] | unituse[14])) - return -1; - - /* Set TTMxSEL fields. Note, units 1-3 => TTM0SEL codes 0-2 */ - mmcr1 |= (unsigned long)(unituse[3] * 2 + unituse[2]) - << MMCR1_TTM0SEL_SH; - mmcr1 |= (unsigned long)(unituse[7] * 3 + unituse[6] * 2) - << MMCR1_TTM1SEL_SH; - mmcr1 |= (unsigned long)unituse[9] << MMCR1_TTM2SEL_SH; - - /* Set TTCxSEL fields. */ - if (unitlower & 0xe) - mmcr1 |= 1ull << MMCR1_TTC0SEL_SH; - if (unitlower & 0xf0) - mmcr1 |= 1ull << MMCR1_TTC1SEL_SH; - if (unitlower & 0xf00) - mmcr1 |= 1ull << MMCR1_TTC2SEL_SH; - if (unitlower & 0x7000) - mmcr1 |= 1ull << MMCR1_TTC3SEL_SH; - - /* Set byte lane select fields. */ - for (byte = 0; byte < 4; ++byte) { - unit = busbyte[byte]; - if (!unit) - continue; - if (unit == 0xf) { - /* special case for GPS */ - mmcr1 |= 1ull << (MMCR1_DEBUG0SEL_SH - byte); - } else { - if (!unituse[unit]) - ttm = unit - 1; /* 2->1, 3->2 */ - else - ttm = unit >> 2; - mmcr1 |= (unsigned long)ttm - << (MMCR1_TD_CP_DBG0SEL_SH - 2 * byte); - } - } - - /* Second pass: assign PMCs, set PMCxSEL and PMCx_ADDER_SEL fields */ - for (i = 0; i < n_ev; ++i) { - pmc = (event[i] >> PM_PMC_SH) & PM_PMC_MSK; - unit = (event[i] >> PM_UNIT_SH) & PM_UNIT_MSK; - byte = (event[i] >> PM_BYTE_SH) & PM_BYTE_MSK; - psel = event[i] & PM_PMCSEL_MSK; - if (!pmc) { - /* Bus event or 00xxx direct event (off or cycles) */ - if (unit) - psel |= 0x10 | ((byte & 2) << 2); - for (pmc = 0; pmc < 8; ++pmc) { - if (pmc_inuse & (1 << pmc)) - continue; - grp = (pmc >> 1) & 1; - if (unit) { - if (grp == (byte & 1)) - break; - } else if (pmc_grp_use[grp] < 4) { - ++pmc_grp_use[grp]; - break; - } - } - pmc_inuse |= 1 << pmc; - } else { - /* Direct event */ - --pmc; - if (psel == 0 && (byte & 2)) - /* add events on higher-numbered bus */ - mmcr1 |= 1ull << mmcr1_adder_bits[pmc]; - else if (psel == 6 && byte == 3) - /* seem to need to set sample_enable here */ - mmcra |= MMCRA_SAMPLE_ENABLE; - psel |= 8; - } - if (pmc <= 1) - mmcr0 |= psel << (MMCR0_PMC1SEL_SH - 7 * pmc); - else - mmcr1 |= psel << (MMCR1_PMC3SEL_SH - 5 * (pmc - 2)); - if (pmc == 7) /* PMC8 */ - mmcra |= (psel & 1) << MMCRA_PMC8SEL0_SH; - hwc[i] = pmc; - if (p4_marked_instr_event(event[i])) - mmcra |= MMCRA_SAMPLE_ENABLE; - } - - if (pmc_inuse & 1) - mmcr0 |= MMCR0_PMC1CE; - if (pmc_inuse & 0xfe) - mmcr0 |= MMCR0_PMCjCE; - - mmcra |= 0x2000; /* mark only one IOP per PPC instruction */ - - /* Return MMCRx values */ - mmcr[0] = mmcr0; - mmcr[1] = mmcr1; - mmcr[2] = mmcra; - return 0; -} - -static void p4_disable_pmc(unsigned int pmc, unsigned long mmcr[]) -{ - /* - * Setting the PMCxSEL field to 0 disables PMC x. - * (Note that pmc is 0-based here, not 1-based.) - */ - if (pmc <= 1) { - mmcr[0] &= ~(0x1fUL << (MMCR0_PMC1SEL_SH - 7 * pmc)); - } else { - mmcr[1] &= ~(0x1fUL << (MMCR1_PMC3SEL_SH - 5 * (pmc - 2))); - if (pmc == 7) - mmcr[2] &= ~(1UL << MMCRA_PMC8SEL0_SH); - } -} - -static int p4_generic_events[] = { - [PERF_COUNT_HW_CPU_CYCLES] = 7, - [PERF_COUNT_HW_INSTRUCTIONS] = 0x1001, - [PERF_COUNT_HW_CACHE_REFERENCES] = 0x8c10, /* PM_LD_REF_L1 */ - [PERF_COUNT_HW_CACHE_MISSES] = 0x3c10, /* PM_LD_MISS_L1 */ - [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x330, /* PM_BR_ISSUED */ - [PERF_COUNT_HW_BRANCH_MISSES] = 0x331, /* PM_BR_MPRED_CR */ -}; - -#define C(x) PERF_COUNT_HW_CACHE_##x - -/* - * Table of generalized cache-related events. - * 0 means not supported, -1 means nonsensical, other values - * are event codes. - */ -static int power4_cache_events[C(MAX)][C(OP_MAX)][C(RESULT_MAX)] = { - [C(L1D)] = { /* RESULT_ACCESS RESULT_MISS */ - [C(OP_READ)] = { 0x8c10, 0x3c10 }, - [C(OP_WRITE)] = { 0x7c10, 0xc13 }, - [C(OP_PREFETCH)] = { 0xc35, 0 }, - }, - [C(L1I)] = { /* RESULT_ACCESS RESULT_MISS */ - [C(OP_READ)] = { 0, 0 }, - [C(OP_WRITE)] = { -1, -1 }, - [C(OP_PREFETCH)] = { 0, 0 }, - }, - [C(LL)] = { /* RESULT_ACCESS RESULT_MISS */ - [C(OP_READ)] = { 0, 0 }, - [C(OP_WRITE)] = { 0, 0 }, - [C(OP_PREFETCH)] = { 0xc34, 0 }, - }, - [C(DTLB)] = { /* RESULT_ACCESS RESULT_MISS */ - [C(OP_READ)] = { 0, 0x904 }, - [C(OP_WRITE)] = { -1, -1 }, - [C(OP_PREFETCH)] = { -1, -1 }, - }, - [C(ITLB)] = { /* RESULT_ACCESS RESULT_MISS */ - [C(OP_READ)] = { 0, 0x900 }, - [C(OP_WRITE)] = { -1, -1 }, - [C(OP_PREFETCH)] = { -1, -1 }, - }, - [C(BPU)] = { /* RESULT_ACCESS RESULT_MISS */ - [C(OP_READ)] = { 0x330, 0x331 }, - [C(OP_WRITE)] = { -1, -1 }, - [C(OP_PREFETCH)] = { -1, -1 }, - }, - [C(NODE)] = { /* RESULT_ACCESS RESULT_MISS */ - [C(OP_READ)] = { -1, -1 }, - [C(OP_WRITE)] = { -1, -1 }, - [C(OP_PREFETCH)] = { -1, -1 }, - }, -}; - -static struct power_pmu power4_pmu = { - .name = "POWER4/4+", - .n_counter = 8, - .max_alternatives = 5, - .add_fields = 0x0000001100005555ul, - .test_adder = 0x0011083300000000ul, - .compute_mmcr = p4_compute_mmcr, - .get_constraint = p4_get_constraint, - .get_alternatives = p4_get_alternatives, - .disable_pmc = p4_disable_pmc, - .n_generic = ARRAY_SIZE(p4_generic_events), - .generic_events = p4_generic_events, - .cache_events = &power4_cache_events, - .flags = PPMU_NO_SIPR | PPMU_NO_CONT_SAMPLING, -}; - -static int __init init_power4_pmu(void) -{ - if (!cur_cpu_spec->oprofile_cpu_type || - strcmp(cur_cpu_spec->oprofile_cpu_type, "ppc64/power4")) - return -ENODEV; - - return register_power_pmu(&power4_pmu); -} - -early_initcall(init_power4_pmu); diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype index 5a8b1bf1e819..fdae4584a016 100644 --- a/arch/powerpc/platforms/Kconfig.cputype +++ b/arch/powerpc/platforms/Kconfig.cputype @@ -61,7 +61,7 @@ choice help There are two families of 64 bit PowerPC chips supported. The most common ones are the desktop and server CPUs - (POWER4, POWER5, 970, POWER5+, POWER6, POWER7, POWER8 ...) + (POWER5, 970, POWER5+, POWER6, POWER7, POWER8, POWER9 ...) The other are the "embedded" processors compliant with the "Book 3E" variant of the architecture @@ -103,10 +103,6 @@ config CELL_CPU bool "Cell Broadband Engine" depends on PPC_BOOK3S_64 && !CPU_LITTLE_ENDIAN -config POWER4_CPU - bool "POWER4" - depends on PPC_BOOK3S_64 && !CPU_LITTLE_ENDIAN - config POWER5_CPU bool "POWER5" depends on PPC_BOOK3S_64 && !CPU_LITTLE_ENDIAN |