* [Qemu-devel] [RFC v2 01/13] target-ppc: Introduce Power9 family
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-25 1:26 ` David Gibson
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 02/13] target-ppc: Introduce POWER ISA 3.0 flag Nikunj A Dadhania
` (11 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
[ rebased and added POWER9 alias and POWER9 SPAPR core ]
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
hw/ppc/spapr_cpu_core.c | 5 +++
target-ppc/cpu-models.c | 5 +++
target-ppc/cpu-models.h | 1 +
target-ppc/cpu-qom.h | 7 ++++
target-ppc/mmu_helper.c | 3 +-
target-ppc/translate_init.c | 85 ++++++++++++++++++++++++++++++++++++++++++++-
6 files changed, 104 insertions(+), 2 deletions(-)
diff --git a/hw/ppc/spapr_cpu_core.c b/hw/ppc/spapr_cpu_core.c
index 9347f07..766c515 100644
--- a/hw/ppc/spapr_cpu_core.c
+++ b/hw/ppc/spapr_cpu_core.c
@@ -351,6 +351,7 @@ SPAPR_CPU_CORE_INITFN(POWER7+_v2.1, POWER7plus);
SPAPR_CPU_CORE_INITFN(POWER8_v2.0, POWER8);
SPAPR_CPU_CORE_INITFN(POWER8E_v2.1, POWER8E);
SPAPR_CPU_CORE_INITFN(POWER8NVL_v1.0, POWER8NVL);
+SPAPR_CPU_CORE_INITFN(POWER9_v1.0, POWER9);
typedef struct SPAPRCoreInfo {
const char *name;
@@ -394,6 +395,10 @@ static const SPAPRCoreInfo spapr_cores[] = {
{ .name = "POWER8NVL_v1.0", .initfn = spapr_cpu_core_POWER8NVL_initfn },
{ .name = "POWER8NVL", .initfn = spapr_cpu_core_POWER8NVL_initfn },
+ /* POWER9 and aliases */
+ { .name = "POWER9_v1.0", .initfn = spapr_cpu_core_POWER9_initfn },
+ { .name = "POWER9", .initfn = spapr_cpu_core_POWER9_initfn },
+
{ .name = NULL }
};
diff --git a/target-ppc/cpu-models.c b/target-ppc/cpu-models.c
index 5209e63..901cf40 100644
--- a/target-ppc/cpu-models.c
+++ b/target-ppc/cpu-models.c
@@ -1147,6 +1147,10 @@
"POWER8NVL v1.0")
POWERPC_DEF("970_v2.2", CPU_POWERPC_970_v22, 970,
"PowerPC 970 v2.2")
+
+ POWERPC_DEF("POWER9_v1.0", CPU_POWERPC_POWER9_BASE, POWER9,
+ "POWER9 v1.0")
+
POWERPC_DEF("970fx_v1.0", CPU_POWERPC_970FX_v10, 970,
"PowerPC 970FX v1.0 (G5)")
POWERPC_DEF("970fx_v2.0", CPU_POWERPC_970FX_v20, 970,
@@ -1395,6 +1399,7 @@ PowerPCCPUAlias ppc_cpu_aliases[] = {
{ "POWER8E", "POWER8E_v2.1" },
{ "POWER8", "POWER8_v2.0" },
{ "POWER8NVL", "POWER8NVL_v1.0" },
+ { "POWER9", "POWER9_v1.0" },
{ "970", "970_v2.2" },
{ "970fx", "970fx_v3.1" },
{ "970mp", "970mp_v1.1" },
diff --git a/target-ppc/cpu-models.h b/target-ppc/cpu-models.h
index f21a44c..7d9e6a2 100644
--- a/target-ppc/cpu-models.h
+++ b/target-ppc/cpu-models.h
@@ -562,6 +562,7 @@ enum {
CPU_POWERPC_POWER8_v20 = 0x004D0200,
CPU_POWERPC_POWER8NVL_BASE = 0x004C0000,
CPU_POWERPC_POWER8NVL_v10 = 0x004C0100,
+ CPU_POWERPC_POWER9_BASE = 0x004E0000,
CPU_POWERPC_970_v22 = 0x00390202,
CPU_POWERPC_970FX_v10 = 0x00391100,
CPU_POWERPC_970FX_v20 = 0x003C0200,
diff --git a/target-ppc/cpu-qom.h b/target-ppc/cpu-qom.h
index 2864105..df2fb65 100644
--- a/target-ppc/cpu-qom.h
+++ b/target-ppc/cpu-qom.h
@@ -86,6 +86,13 @@ enum powerpc_mmu_t {
POWERPC_MMU_2_07 = POWERPC_MMU_64 | POWERPC_MMU_1TSEG
| POWERPC_MMU_64K
| POWERPC_MMU_AMR | 0x00000004,
+ /* for now , We can add radix later if needed */
+ /* POWERPC_MMU_3_00 = POWERPC_MMU_64 | POWERPC_MMU_1TSEG
+ * | POWERPC_MMU_AMR | 0x00000005,
+ */
+
+ POWERPC_MMU_3_00 = POWERPC_MMU_64 | POWERPC_MMU_AMR | 0x00000005,
+
/* Architecture 2.07 "degraded" (no 1T segments) */
POWERPC_MMU_2_07a = POWERPC_MMU_64 | POWERPC_MMU_AMR
| 0x00000004,
diff --git a/target-ppc/mmu_helper.c b/target-ppc/mmu_helper.c
index 3eb3cd7..dd0cf33 100644
--- a/target-ppc/mmu_helper.c
+++ b/target-ppc/mmu_helper.c
@@ -1935,13 +1935,14 @@ void ppc_tlb_invalidate_all(CPUPPCState *env)
case POWERPC_MMU_2_06a:
case POWERPC_MMU_2_07:
case POWERPC_MMU_2_07a:
+ case POWERPC_MMU_3_00:
#endif /* defined(TARGET_PPC64) */
env->tlb_need_flush = 0;
tlb_flush(CPU(cpu), 1);
break;
default:
/* XXX: TODO */
- cpu_abort(CPU(cpu), "Unknown MMU model\n");
+ cpu_abort(CPU(cpu), "Unknown MMU model %d\n", env->mmu_model);
break;
}
}
diff --git a/target-ppc/translate_init.c b/target-ppc/translate_init.c
index 7cb7842..5b100d0 100644
--- a/target-ppc/translate_init.c
+++ b/target-ppc/translate_init.c
@@ -7459,7 +7459,8 @@ enum BOOK3S_CPU_TYPE {
BOOK3S_CPU_POWER5PLUS,
BOOK3S_CPU_POWER6,
BOOK3S_CPU_POWER7,
- BOOK3S_CPU_POWER8
+ BOOK3S_CPU_POWER8,
+ BOOK3S_CPU_POWER9
};
static void gen_fscr_facility_check(DisasContext *ctx, int facility_sprn,
@@ -8241,6 +8242,7 @@ static void init_proc_book3s_64(CPUPPCState *env, int version)
break;
case BOOK3S_CPU_POWER7:
case BOOK3S_CPU_POWER8:
+ case BOOK3S_CPU_POWER9:
gen_spr_book3s_ids(env);
gen_spr_amr(env, version >= BOOK3S_CPU_POWER8);
gen_spr_book3s_purr(env);
@@ -8293,6 +8295,7 @@ static void init_proc_book3s_64(CPUPPCState *env, int version)
break;
case BOOK3S_CPU_POWER7:
case BOOK3S_CPU_POWER8:
+ case BOOK3S_CPU_POWER9:
default:
env->slb_nr = 32;
break;
@@ -8310,6 +8313,7 @@ static void init_proc_book3s_64(CPUPPCState *env, int version)
ppcPOWER7_irq_init(ppc_env_get_cpu(env));
break;
case BOOK3S_CPU_POWER8:
+ case BOOK3S_CPU_POWER9:
init_excp_POWER8(env);
ppcPOWER7_irq_init(ppc_env_get_cpu(env));
break;
@@ -8772,6 +8776,85 @@ POWERPC_FAMILY(POWER8)(ObjectClass *oc, void *data)
pcc->l1_icache_size = 0x8000;
pcc->interrupts_big_endian = ppc_cpu_interrupts_big_endian_lpcr;
}
+static void init_proc_POWER9(CPUPPCState *env)
+{
+ init_proc_book3s_64(env, BOOK3S_CPU_POWER9);
+}
+
+static bool ppc_pvr_match_power9(PowerPCCPUClass *pcc, uint32_t pvr)
+{
+ if ((pvr & CPU_POWERPC_POWER_SERVER_MASK) == CPU_POWERPC_POWER9_BASE) {
+ return true;
+ }
+ return false;
+}
+
+POWERPC_FAMILY(POWER9)(ObjectClass *oc, void *data)
+{
+ DeviceClass *dc = DEVICE_CLASS(oc);
+ PowerPCCPUClass *pcc = POWERPC_CPU_CLASS(oc);
+
+ dc->fw_name = "PowerPC,POWER9";
+ dc->desc = "POWER9";
+ dc->props = powerpc_servercpu_properties;
+ pcc->pvr_match = ppc_pvr_match_power9;
+ pcc->pcr_mask = PCR_COMPAT_2_05 | PCR_COMPAT_2_06 | PCR_COMPAT_2_07;
+ pcc->init_proc = init_proc_POWER9;
+ pcc->check_pow = check_pow_nocheck;
+ pcc->insns_flags = PPC_INSNS_BASE | PPC_ISEL | PPC_STRING | PPC_MFTB |
+ PPC_FLOAT | PPC_FLOAT_FSEL | PPC_FLOAT_FRES |
+ PPC_FLOAT_FSQRT | PPC_FLOAT_FRSQRTE |
+ PPC_FLOAT_FRSQRTES |
+ PPC_FLOAT_STFIWX |
+ PPC_FLOAT_EXT |
+ PPC_CACHE | PPC_CACHE_ICBI | PPC_CACHE_DCBZ |
+ PPC_MEM_SYNC | PPC_MEM_EIEIO |
+ PPC_MEM_TLBIE | PPC_MEM_TLBSYNC |
+ PPC_64B | PPC_64BX | PPC_ALTIVEC |
+ PPC_SEGMENT_64B | PPC_SLBI |
+ PPC_POPCNTB | PPC_POPCNTWD |
+ PPC_CILDST;
+ pcc->insns_flags2 = PPC2_VSX | PPC2_VSX207 | PPC2_DFP | PPC2_DBRX |
+ PPC2_PERM_ISA206 | PPC2_DIVE_ISA206 |
+ PPC2_ATOMIC_ISA206 | PPC2_FP_CVT_ISA206 |
+ PPC2_FP_TST_ISA206 | PPC2_BCTAR_ISA207 |
+ PPC2_LSQ_ISA207 | PPC2_ALTIVEC_207 |
+ PPC2_ISA205 | PPC2_ISA207S | PPC2_FP_CVT_S64 |
+ PPC2_TM | PPC2_PM_ISA206;
+ pcc->msr_mask = (1ull << MSR_SF) |
+ (1ull << MSR_TM) |
+ (1ull << MSR_VR) |
+ (1ull << MSR_VSX) |
+ (1ull << MSR_EE) |
+ (1ull << MSR_PR) |
+ (1ull << MSR_FP) |
+ (1ull << MSR_ME) |
+ (1ull << MSR_FE0) |
+ (1ull << MSR_SE) |
+ (1ull << MSR_DE) |
+ (1ull << MSR_FE1) |
+ (1ull << MSR_IR) |
+ (1ull << MSR_DR) |
+ (1ull << MSR_PMM) |
+ (1ull << MSR_RI) |
+ (1ull << MSR_LE);
+ pcc->mmu_model = POWERPC_MMU_3_00;
+#if defined(CONFIG_SOFTMMU)
+ pcc->handle_mmu_fault = ppc_hash64_handle_mmu_fault;
+ /* segment page size remain the same */
+ pcc->sps = &POWER7_POWER8_sps;
+#endif
+ pcc->excp_model = POWERPC_EXCP_POWER8;
+ pcc->bus_model = PPC_FLAGS_INPUT_POWER7;
+ pcc->bfd_mach = bfd_mach_ppc64;
+ pcc->flags = POWERPC_FLAG_VRE | POWERPC_FLAG_SE |
+ POWERPC_FLAG_BE | POWERPC_FLAG_PMM |
+ POWERPC_FLAG_BUS_CLK | POWERPC_FLAG_CFAR |
+ POWERPC_FLAG_VSX | POWERPC_FLAG_TM;
+ pcc->l1_dcache_size = 0x8000;
+ pcc->l1_icache_size = 0x8000;
+ pcc->interrupts_big_endian = ppc_cpu_interrupts_big_endian_lpcr;
+}
#if !defined(CONFIG_USER_ONLY)
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [RFC v2 01/13] target-ppc: Introduce Power9 family
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 01/13] target-ppc: Introduce Power9 family Nikunj A Dadhania
@ 2016-07-25 1:26 ` David Gibson
2016-07-25 8:48 ` Nikunj A Dadhania
0 siblings, 1 reply; 29+ messages in thread
From: David Gibson @ 2016-07-25 1:26 UTC (permalink / raw)
To: Nikunj A Dadhania; +Cc: qemu-ppc, rth, qemu-devel, bharata, aneesh.kumar
[-- Attachment #1: Type: text/plain, Size: 10435 bytes --]
On Sat, Jul 23, 2016 at 02:14:38PM +0530, Nikunj A Dadhania wrote:
> From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
>
> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
> [ rebased and added POWER9 alias and POWER9 SPAPR core ]
> Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
I can't really review this until the series is a bit more polished and
we have a better idea how the handling of the POWER9 MMU is going to
work. As it stands, this sets the P9 MMU type to something that only
makes sense in guest context, not host context.
> ---
> hw/ppc/spapr_cpu_core.c | 5 +++
> target-ppc/cpu-models.c | 5 +++
> target-ppc/cpu-models.h | 1 +
> target-ppc/cpu-qom.h | 7 ++++
> target-ppc/mmu_helper.c | 3 +-
> target-ppc/translate_init.c | 85 ++++++++++++++++++++++++++++++++++++++++++++-
> 6 files changed, 104 insertions(+), 2 deletions(-)
>
> diff --git a/hw/ppc/spapr_cpu_core.c b/hw/ppc/spapr_cpu_core.c
> index 9347f07..766c515 100644
> --- a/hw/ppc/spapr_cpu_core.c
> +++ b/hw/ppc/spapr_cpu_core.c
> @@ -351,6 +351,7 @@ SPAPR_CPU_CORE_INITFN(POWER7+_v2.1, POWER7plus);
> SPAPR_CPU_CORE_INITFN(POWER8_v2.0, POWER8);
> SPAPR_CPU_CORE_INITFN(POWER8E_v2.1, POWER8E);
> SPAPR_CPU_CORE_INITFN(POWER8NVL_v1.0, POWER8NVL);
> +SPAPR_CPU_CORE_INITFN(POWER9_v1.0, POWER9);
>
> typedef struct SPAPRCoreInfo {
> const char *name;
> @@ -394,6 +395,10 @@ static const SPAPRCoreInfo spapr_cores[] = {
> { .name = "POWER8NVL_v1.0", .initfn = spapr_cpu_core_POWER8NVL_initfn },
> { .name = "POWER8NVL", .initfn = spapr_cpu_core_POWER8NVL_initfn },
>
> + /* POWER9 and aliases */
> + { .name = "POWER9_v1.0", .initfn = spapr_cpu_core_POWER9_initfn },
> + { .name = "POWER9", .initfn = spapr_cpu_core_POWER9_initfn },
> +
> { .name = NULL }
> };
>
> diff --git a/target-ppc/cpu-models.c b/target-ppc/cpu-models.c
> index 5209e63..901cf40 100644
> --- a/target-ppc/cpu-models.c
> +++ b/target-ppc/cpu-models.c
> @@ -1147,6 +1147,10 @@
> "POWER8NVL v1.0")
> POWERPC_DEF("970_v2.2", CPU_POWERPC_970_v22, 970,
> "PowerPC 970 v2.2")
> +
> + POWERPC_DEF("POWER9_v1.0", CPU_POWERPC_POWER9_BASE, POWER9,
> + "POWER9 v1.0")
> +
> POWERPC_DEF("970fx_v1.0", CPU_POWERPC_970FX_v10, 970,
> "PowerPC 970FX v1.0 (G5)")
> POWERPC_DEF("970fx_v2.0", CPU_POWERPC_970FX_v20, 970,
> @@ -1395,6 +1399,7 @@ PowerPCCPUAlias ppc_cpu_aliases[] = {
> { "POWER8E", "POWER8E_v2.1" },
> { "POWER8", "POWER8_v2.0" },
> { "POWER8NVL", "POWER8NVL_v1.0" },
> + { "POWER9", "POWER9_v1.0" },
> { "970", "970_v2.2" },
> { "970fx", "970fx_v3.1" },
> { "970mp", "970mp_v1.1" },
> diff --git a/target-ppc/cpu-models.h b/target-ppc/cpu-models.h
> index f21a44c..7d9e6a2 100644
> --- a/target-ppc/cpu-models.h
> +++ b/target-ppc/cpu-models.h
> @@ -562,6 +562,7 @@ enum {
> CPU_POWERPC_POWER8_v20 = 0x004D0200,
> CPU_POWERPC_POWER8NVL_BASE = 0x004C0000,
> CPU_POWERPC_POWER8NVL_v10 = 0x004C0100,
> + CPU_POWERPC_POWER9_BASE = 0x004E0000,
> CPU_POWERPC_970_v22 = 0x00390202,
> CPU_POWERPC_970FX_v10 = 0x00391100,
> CPU_POWERPC_970FX_v20 = 0x003C0200,
> diff --git a/target-ppc/cpu-qom.h b/target-ppc/cpu-qom.h
> index 2864105..df2fb65 100644
> --- a/target-ppc/cpu-qom.h
> +++ b/target-ppc/cpu-qom.h
> @@ -86,6 +86,13 @@ enum powerpc_mmu_t {
> POWERPC_MMU_2_07 = POWERPC_MMU_64 | POWERPC_MMU_1TSEG
> | POWERPC_MMU_64K
> | POWERPC_MMU_AMR | 0x00000004,
> + /* for now , We can add radix later if needed */
> + /* POWERPC_MMU_3_00 = POWERPC_MMU_64 | POWERPC_MMU_1TSEG
> + * | POWERPC_MMU_AMR | 0x00000005,
> + */
> +
> + POWERPC_MMU_3_00 = POWERPC_MMU_64 | POWERPC_MMU_AMR | 0x00000005,
> +
> /* Architecture 2.07 "degraded" (no 1T segments) */
> POWERPC_MMU_2_07a = POWERPC_MMU_64 | POWERPC_MMU_AMR
> | 0x00000004,
> diff --git a/target-ppc/mmu_helper.c b/target-ppc/mmu_helper.c
> index 3eb3cd7..dd0cf33 100644
> --- a/target-ppc/mmu_helper.c
> +++ b/target-ppc/mmu_helper.c
> @@ -1935,13 +1935,14 @@ void ppc_tlb_invalidate_all(CPUPPCState *env)
> case POWERPC_MMU_2_06a:
> case POWERPC_MMU_2_07:
> case POWERPC_MMU_2_07a:
> + case POWERPC_MMU_3_00:
> #endif /* defined(TARGET_PPC64) */
> env->tlb_need_flush = 0;
> tlb_flush(CPU(cpu), 1);
> break;
> default:
> /* XXX: TODO */
> - cpu_abort(CPU(cpu), "Unknown MMU model\n");
> + cpu_abort(CPU(cpu), "Unknown MMU model %d\n", env->mmu_model);
> break;
> }
> }
> diff --git a/target-ppc/translate_init.c b/target-ppc/translate_init.c
> index 7cb7842..5b100d0 100644
> --- a/target-ppc/translate_init.c
> +++ b/target-ppc/translate_init.c
> @@ -7459,7 +7459,8 @@ enum BOOK3S_CPU_TYPE {
> BOOK3S_CPU_POWER5PLUS,
> BOOK3S_CPU_POWER6,
> BOOK3S_CPU_POWER7,
> - BOOK3S_CPU_POWER8
> + BOOK3S_CPU_POWER8,
> + BOOK3S_CPU_POWER9
> };
>
> static void gen_fscr_facility_check(DisasContext *ctx, int facility_sprn,
> @@ -8241,6 +8242,7 @@ static void init_proc_book3s_64(CPUPPCState *env, int version)
> break;
> case BOOK3S_CPU_POWER7:
> case BOOK3S_CPU_POWER8:
> + case BOOK3S_CPU_POWER9:
> gen_spr_book3s_ids(env);
> gen_spr_amr(env, version >= BOOK3S_CPU_POWER8);
> gen_spr_book3s_purr(env);
> @@ -8293,6 +8295,7 @@ static void init_proc_book3s_64(CPUPPCState *env, int version)
> break;
> case BOOK3S_CPU_POWER7:
> case BOOK3S_CPU_POWER8:
> + case BOOK3S_CPU_POWER9:
> default:
> env->slb_nr = 32;
> break;
> @@ -8310,6 +8313,7 @@ static void init_proc_book3s_64(CPUPPCState *env, int version)
> ppcPOWER7_irq_init(ppc_env_get_cpu(env));
> break;
> case BOOK3S_CPU_POWER8:
> + case BOOK3S_CPU_POWER9:
> init_excp_POWER8(env);
> ppcPOWER7_irq_init(ppc_env_get_cpu(env));
> break;
> @@ -8772,6 +8776,85 @@ POWERPC_FAMILY(POWER8)(ObjectClass *oc, void *data)
> pcc->l1_icache_size = 0x8000;
> pcc->interrupts_big_endian = ppc_cpu_interrupts_big_endian_lpcr;
> }
> +static void init_proc_POWER9(CPUPPCState *env)
> +{
> + init_proc_book3s_64(env, BOOK3S_CPU_POWER9);
> +}
> +
> +static bool ppc_pvr_match_power9(PowerPCCPUClass *pcc, uint32_t pvr)
> +{
> + if ((pvr & CPU_POWERPC_POWER_SERVER_MASK) == CPU_POWERPC_POWER9_BASE) {
> + return true;
> + }
> + return false;
> +}
> +
> +POWERPC_FAMILY(POWER9)(ObjectClass *oc, void *data)
> +{
> + DeviceClass *dc = DEVICE_CLASS(oc);
> + PowerPCCPUClass *pcc = POWERPC_CPU_CLASS(oc);
> +
> + dc->fw_name = "PowerPC,POWER9";
> + dc->desc = "POWER9";
> + dc->props = powerpc_servercpu_properties;
> + pcc->pvr_match = ppc_pvr_match_power9;
> + pcc->pcr_mask = PCR_COMPAT_2_05 | PCR_COMPAT_2_06 | PCR_COMPAT_2_07;
> + pcc->init_proc = init_proc_POWER9;
> + pcc->check_pow = check_pow_nocheck;
> + pcc->insns_flags = PPC_INSNS_BASE | PPC_ISEL | PPC_STRING | PPC_MFTB |
> + PPC_FLOAT | PPC_FLOAT_FSEL | PPC_FLOAT_FRES |
> + PPC_FLOAT_FSQRT | PPC_FLOAT_FRSQRTE |
> + PPC_FLOAT_FRSQRTES |
> + PPC_FLOAT_STFIWX |
> + PPC_FLOAT_EXT |
> + PPC_CACHE | PPC_CACHE_ICBI | PPC_CACHE_DCBZ |
> + PPC_MEM_SYNC | PPC_MEM_EIEIO |
> + PPC_MEM_TLBIE | PPC_MEM_TLBSYNC |
> + PPC_64B | PPC_64BX | PPC_ALTIVEC |
> + PPC_SEGMENT_64B | PPC_SLBI |
> + PPC_POPCNTB | PPC_POPCNTWD |
> + PPC_CILDST;
> + pcc->insns_flags2 = PPC2_VSX | PPC2_VSX207 | PPC2_DFP | PPC2_DBRX |
> + PPC2_PERM_ISA206 | PPC2_DIVE_ISA206 |
> + PPC2_ATOMIC_ISA206 | PPC2_FP_CVT_ISA206 |
> + PPC2_FP_TST_ISA206 | PPC2_BCTAR_ISA207 |
> + PPC2_LSQ_ISA207 | PPC2_ALTIVEC_207 |
> + PPC2_ISA205 | PPC2_ISA207S | PPC2_FP_CVT_S64 |
> + PPC2_TM | PPC2_PM_ISA206;
> + pcc->msr_mask = (1ull << MSR_SF) |
> + (1ull << MSR_TM) |
> + (1ull << MSR_VR) |
> + (1ull << MSR_VSX) |
> + (1ull << MSR_EE) |
> + (1ull << MSR_PR) |
> + (1ull << MSR_FP) |
> + (1ull << MSR_ME) |
> + (1ull << MSR_FE0) |
> + (1ull << MSR_SE) |
> + (1ull << MSR_DE) |
> + (1ull << MSR_FE1) |
> + (1ull << MSR_IR) |
> + (1ull << MSR_DR) |
> + (1ull << MSR_PMM) |
> + (1ull << MSR_RI) |
> + (1ull << MSR_LE);
> + pcc->mmu_model = POWERPC_MMU_3_00;
> +#if defined(CONFIG_SOFTMMU)
> + pcc->handle_mmu_fault = ppc_hash64_handle_mmu_fault;
> + /* segment page size remain the same */
> + pcc->sps = &POWER7_POWER8_sps;
> +#endif
> + pcc->excp_model = POWERPC_EXCP_POWER8;
> + pcc->bus_model = PPC_FLAGS_INPUT_POWER7;
> + pcc->bfd_mach = bfd_mach_ppc64;
> + pcc->flags = POWERPC_FLAG_VRE | POWERPC_FLAG_SE |
> + POWERPC_FLAG_BE | POWERPC_FLAG_PMM |
> + POWERPC_FLAG_BUS_CLK | POWERPC_FLAG_CFAR |
> + POWERPC_FLAG_VSX | POWERPC_FLAG_TM;
> + pcc->l1_dcache_size = 0x8000;
> + pcc->l1_icache_size = 0x8000;
> + pcc->interrupts_big_endian = ppc_cpu_interrupts_big_endian_lpcr;
> +}
>
> #if !defined(CONFIG_USER_ONLY)
>
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [RFC v2 01/13] target-ppc: Introduce Power9 family
2016-07-25 1:26 ` David Gibson
@ 2016-07-25 8:48 ` Nikunj A Dadhania
0 siblings, 0 replies; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-25 8:48 UTC (permalink / raw)
To: David Gibson; +Cc: qemu-ppc, rth, qemu-devel, bharata, aneesh.kumar
David Gibson <david@gibson.dropbear.id.au> writes:
> [ Unknown signature status ]
> On Sat, Jul 23, 2016 at 02:14:38PM +0530, Nikunj A Dadhania wrote:
>> From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
>>
>> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
>> [ rebased and added POWER9 alias and POWER9 SPAPR core ]
>> Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
>
> I can't really review this until the series is a bit more polished and
> we have a better idea how the handling of the POWER9 MMU is going to
> work. As it stands, this sets the P9 MMU type to something that only
> makes sense in guest context, not host context.
Will send a patch without P9 MMU changes and have a FIXME to add it.
Regards
Nikunj
^ permalink raw reply [flat|nested] 29+ messages in thread
* [Qemu-devel] [RFC v2 02/13] target-ppc: Introduce POWER ISA 3.0 flag
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 01/13] target-ppc: Introduce Power9 family Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 03/13] target-ppc: adding addpcis instruction Nikunj A Dadhania
` (10 subsequent siblings)
12 siblings, 0 replies; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
This flag will be used for POWER9 instructions.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
---
target-ppc/cpu.h | 5 ++++-
target-ppc/translate_init.c | 2 +-
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/target-ppc/cpu.h b/target-ppc/cpu.h
index 5fce1ff..c499315 100644
--- a/target-ppc/cpu.h
+++ b/target-ppc/cpu.h
@@ -2094,6 +2094,8 @@ enum {
PPC2_TM = 0x0000000000020000ULL,
/* Server PM instructgions (ISA 2.06, Book III) */
PPC2_PM_ISA206 = 0x0000000000040000ULL,
+ /* POWER ISA 3.0 */
+ PPC2_ISA300 = 0x0000000000080000ULL,
#define PPC_TCG_INSNS2 (PPC2_BOOKE206 | PPC2_VSX | PPC2_PRCNTL | PPC2_DBRX | \
PPC2_ISA205 | PPC2_VSX207 | PPC2_PERM_ISA206 | \
@@ -2101,7 +2103,8 @@ enum {
PPC2_FP_CVT_ISA206 | PPC2_FP_TST_ISA206 | \
PPC2_BCTAR_ISA207 | PPC2_LSQ_ISA207 | \
PPC2_ALTIVEC_207 | PPC2_ISA207S | PPC2_DFP | \
- PPC2_FP_CVT_S64 | PPC2_TM | PPC2_PM_ISA206)
+ PPC2_FP_CVT_S64 | PPC2_TM | PPC2_PM_ISA206 | \
+ PPC2_ISA300)
};
/*****************************************************************************/
diff --git a/target-ppc/translate_init.c b/target-ppc/translate_init.c
index 5b100d0..d207f68 100644
--- a/target-ppc/translate_init.c
+++ b/target-ppc/translate_init.c
@@ -8820,7 +8820,7 @@ POWERPC_FAMILY(POWER9)(ObjectClass *oc, void *data)
PPC2_FP_TST_ISA206 | PPC2_BCTAR_ISA207 |
PPC2_LSQ_ISA207 | PPC2_ALTIVEC_207 |
PPC2_ISA205 | PPC2_ISA207S | PPC2_FP_CVT_S64 |
- PPC2_TM | PPC2_PM_ISA206;
+ PPC2_TM | PPC2_PM_ISA206 | PPC2_ISA300;
pcc->msr_mask = (1ull << MSR_SF) |
(1ull << MSR_TM) |
(1ull << MSR_VR) |
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* [Qemu-devel] [RFC v2 03/13] target-ppc: adding addpcis instruction
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 01/13] target-ppc: Introduce Power9 family Nikunj A Dadhania
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 02/13] target-ppc: Introduce POWER ISA 3.0 flag Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 04/13] target-ppc: add cmprb instruction Nikunj A Dadhania
` (9 subsequent siblings)
12 siblings, 0 replies; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
ISA 3.0 instruction for adding immediate value shifted with next
instruction address and return the result in the target register.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
---
target-ppc/translate.c | 26 ++++++++++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 92030b6..ca246ea 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -432,6 +432,20 @@ static inline uint32_t name(uint32_t opcode) \
return (((opcode >> (shift1)) & ((1 << (nb1)) - 1)) << nb2) | \
((opcode >> (shift2)) & ((1 << (nb2)) - 1)); \
}
+
+#define EXTRACT_HELPER_DXFORM(name, \
+ d0_bits, shift_op_d0, shift_d0, \
+ d1_bits, shift_op_d1, shift_d1, \
+ d2_bits, shift_op_d2, shift_d2) \
+static inline int16_t name(uint32_t opcode) \
+{ \
+ return \
+ (((opcode >> (shift_op_d0)) & ((1 << (d0_bits)) - 1)) << (shift_d0)) | \
+ (((opcode >> (shift_op_d1)) & ((1 << (d1_bits)) - 1)) << (shift_d1)) | \
+ (((opcode >> (shift_op_d2)) & ((1 << (d2_bits)) - 1)) << (shift_d2)); \
+}
+
+
/* Opcode part 1 */
EXTRACT_HELPER(opc1, 26, 6);
/* Opcode part 2 */
@@ -501,6 +515,9 @@ EXTRACT_HELPER(FPL, 25, 1);
EXTRACT_HELPER(FPFLM, 17, 8);
EXTRACT_HELPER(FPW, 16, 1);
+/* addpcis */
+EXTRACT_HELPER_DXFORM(DX, 10, 6, 6, 5, 16, 1, 1, 0, 0)
+
/*** Jump target decoding ***/
/* Immediate address */
static inline target_ulong LI(uint32_t opcode)
@@ -984,6 +1001,14 @@ static void gen_addis(DisasContext *ctx)
}
}
+/* addpcis */
+static void gen_addpcis(DisasContext *ctx)
+{
+ target_long d = DX(ctx->opcode);
+
+ tcg_gen_movi_tl(cpu_gpr[rD(ctx->opcode)], ctx->nip + (d << 16));
+}
+
static inline void gen_op_arith_divw(DisasContext *ctx, TCGv ret, TCGv arg1,
TCGv arg2, int sign, int compute_ov)
{
@@ -9877,6 +9902,7 @@ GEN_HANDLER(addi, 0x0E, 0xFF, 0xFF, 0x00000000, PPC_INTEGER),
GEN_HANDLER(addic, 0x0C, 0xFF, 0xFF, 0x00000000, PPC_INTEGER),
GEN_HANDLER2(addic_, "addic.", 0x0D, 0xFF, 0xFF, 0x00000000, PPC_INTEGER),
GEN_HANDLER(addis, 0x0F, 0xFF, 0xFF, 0x00000000, PPC_INTEGER),
+GEN_HANDLER_E(addpcis, 0x13, 0x2, 0xFF, 0x00000000, PPC_NONE, PPC2_ISA300),
GEN_HANDLER(mulhw, 0x1F, 0x0B, 0x02, 0x00000400, PPC_INTEGER),
GEN_HANDLER(mulhwu, 0x1F, 0x0B, 0x00, 0x00000400, PPC_INTEGER),
GEN_HANDLER(mullw, 0x1F, 0x0B, 0x07, 0x00000000, PPC_INTEGER),
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* [Qemu-devel] [RFC v2 04/13] target-ppc: add cmprb instruction
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
` (2 preceding siblings ...)
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 03/13] target-ppc: adding addpcis instruction Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-23 16:06 ` Richard Henderson
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 05/13] target-ppc: add modulo word operations Nikunj A Dadhania
` (8 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
ISA 3.0 Compare Ranged Byte instruction useful for
isupper/islower/isaplha kind of operation.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
target-ppc/translate.c | 38 ++++++++++++++++++++++++++++++++++++++
1 file changed, 38 insertions(+)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index ca246ea..7e349e8 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -817,6 +817,43 @@ static void gen_cmpli(DisasContext *ctx)
}
}
+/* cmprb - range comparison: isupper, isaplha, islower*/
+static void gen_cmprb(DisasContext *ctx)
+{
+ TCGv_i32 src1 = tcg_temp_new_i32();
+ TCGv_i32 src2 = tcg_temp_new_i32();
+ TCGv_i32 src2lo = tcg_temp_new_i32();
+ TCGv_i32 src2hi = tcg_temp_new_i32();
+ TCGv_i32 crf = cpu_crf[crfD(ctx->opcode)];
+
+ tcg_gen_trunc_tl_i32(src1, cpu_gpr[rA(ctx->opcode)]);
+ tcg_gen_trunc_tl_i32(src2, cpu_gpr[rB(ctx->opcode)]);
+
+ tcg_gen_ext8u_i32(src2lo, src2);
+ tcg_gen_shri_i32(src2, src2, 8);
+ tcg_gen_ext8u_i32(src2hi, src2);
+
+ tcg_gen_setcond_i32(TCG_COND_LEU, src2lo, src2lo, src1);
+ tcg_gen_setcond_i32(TCG_COND_LEU, src2hi, src1, src2hi);
+ tcg_gen_and_i32(crf, src2lo, src2hi);
+
+ if (ctx->opcode & 0x00200000) {
+ tcg_gen_shri_i32(src2, src2, 8);
+ tcg_gen_ext8u_i32(src2lo, src2);
+ tcg_gen_shri_i32(src2, src2, 8);
+ tcg_gen_ext8u_i32(src2hi, src2);
+ tcg_gen_setcond_i32(TCG_COND_LEU, src2lo, src2lo, src1);
+ tcg_gen_setcond_i32(TCG_COND_LEU, src2hi, src1, src2hi);
+ tcg_gen_and_i32(src2lo, src2lo, src2hi);
+ tcg_gen_or_i32(crf, crf, src2lo);
+ }
+ tcg_gen_shli_i32(crf, crf, CRF_GT);
+ tcg_temp_free_i32(src1);
+ tcg_temp_free_i32(src2);
+ tcg_temp_free_i32(src2lo);
+ tcg_temp_free_i32(src2hi);
+}
+
/* isel (PowerPC 2.03 specification) */
static void gen_isel(DisasContext *ctx)
{
@@ -9897,6 +9934,7 @@ GEN_HANDLER(cmpi, 0x0B, 0xFF, 0xFF, 0x00400000, PPC_INTEGER),
GEN_HANDLER(cmpl, 0x1F, 0x00, 0x01, 0x00400000, PPC_INTEGER),
GEN_HANDLER(cmpli, 0x0A, 0xFF, 0xFF, 0x00400000, PPC_INTEGER),
GEN_HANDLER_E(cmpb, 0x1F, 0x1C, 0x0F, 0x00000001, PPC_NONE, PPC2_ISA205),
+GEN_HANDLER_E(cmprb, 0x1F, 0x00, 0x06, 0x00400001, PPC_NONE, PPC2_ISA300),
GEN_HANDLER(isel, 0x1F, 0x0F, 0xFF, 0x00000001, PPC_ISEL),
GEN_HANDLER(addi, 0x0E, 0xFF, 0xFF, 0x00000000, PPC_INTEGER),
GEN_HANDLER(addic, 0x0C, 0xFF, 0xFF, 0x00000000, PPC_INTEGER),
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* [Qemu-devel] [RFC v2 05/13] target-ppc: add modulo word operations
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
` (3 preceding siblings ...)
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 04/13] target-ppc: add cmprb instruction Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-24 1:24 ` Richard Henderson
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 06/13] target-ppc: add modulo dword operations Nikunj A Dadhania
` (7 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
Adding following instructions:
moduw: Modulo Unsigned Word
modsw: Modulo Signed Word
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
target-ppc/helper.h | 2 ++
target-ppc/int_helper.c | 15 +++++++++++++++
target-ppc/translate.c | 19 +++++++++++++++++++
3 files changed, 36 insertions(+)
diff --git a/target-ppc/helper.h b/target-ppc/helper.h
index 1f5cfd0..76072fd 100644
--- a/target-ppc/helper.h
+++ b/target-ppc/helper.h
@@ -41,6 +41,8 @@ DEF_HELPER_FLAGS_1(cntlzw, TCG_CALL_NO_RWG_SE, tl, tl)
DEF_HELPER_FLAGS_1(popcntb, TCG_CALL_NO_RWG_SE, tl, tl)
DEF_HELPER_FLAGS_1(popcntw, TCG_CALL_NO_RWG_SE, tl, tl)
DEF_HELPER_FLAGS_2(cmpb, TCG_CALL_NO_RWG_SE, tl, tl, tl)
+DEF_HELPER_FLAGS_2(modsw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
+DEF_HELPER_FLAGS_2(moduw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
DEF_HELPER_3(sraw, tl, env, tl, tl)
#if defined(TARGET_PPC64)
DEF_HELPER_FLAGS_1(cntlzd, TCG_CALL_NO_RWG_SE, tl, tl)
diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
index 7445376..631e0b4 100644
--- a/target-ppc/int_helper.c
+++ b/target-ppc/int_helper.c
@@ -139,6 +139,21 @@ uint64_t helper_divde(CPUPPCState *env, uint64_t rau, uint64_t rbu, uint32_t oe)
#endif
+uint32_t helper_modsw(uint32_t rau, uint32_t rbu)
+{
+ int32_t ra = (int32_t) rau;
+ int32_t rb = (int32_t) rbu;
+
+ if ((rb == 0) || (ra == INT32_MIN && rb == -1)) {
+ return 0;
+ }
+ return ra % rb;
+}
+
+uint32_t helper_moduw(uint32_t ra, uint32_t rb)
+{
+ return rb ? ra % rb : 0;
+}
target_ulong helper_cntlzw(target_ulong t)
{
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 7e349e8..4348efd 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -1175,6 +1175,23 @@ GEN_DIVE(divde, divde, 0);
GEN_DIVE(divdeo, divde, 1);
#endif
+#define GEN_INT_ARITH_MODW(name, opc3, sign) \
+static void glue(gen_, name)(DisasContext *ctx) \
+{ \
+ TCGv_i32 t0 = tcg_temp_new_i32(); \
+ TCGv_i32 t1 = tcg_temp_new_i32(); \
+ \
+ tcg_gen_trunc_tl_i32(t0, cpu_gpr[rA(ctx->opcode)]); \
+ tcg_gen_trunc_tl_i32(t1, cpu_gpr[rB(ctx->opcode)]); \
+ gen_helper_##name(t0, t0 , t1); \
+ tcg_gen_extu_i32_tl(cpu_gpr[rD(ctx->opcode)], t0); \
+ tcg_temp_free_i32(t0); \
+ tcg_temp_free_i32(t1); \
+}
+
+GEN_INT_ARITH_MODW(moduw, 0x08, 0);
+GEN_INT_ARITH_MODW(modsw, 0x18, 1);
+
/* mulhw mulhw. */
static void gen_mulhw(DisasContext *ctx)
{
@@ -10241,6 +10258,8 @@ GEN_HANDLER_E(divwe, 0x1F, 0x0B, 0x0D, 0, PPC_NONE, PPC2_DIVE_ISA206),
GEN_HANDLER_E(divweo, 0x1F, 0x0B, 0x1D, 0, PPC_NONE, PPC2_DIVE_ISA206),
GEN_HANDLER_E(divweu, 0x1F, 0x0B, 0x0C, 0, PPC_NONE, PPC2_DIVE_ISA206),
GEN_HANDLER_E(divweuo, 0x1F, 0x0B, 0x1C, 0, PPC_NONE, PPC2_DIVE_ISA206),
+GEN_HANDLER_E(modsw, 0x1F, 0x0B, 0x18, 0x00000001, PPC_NONE, PPC2_ISA300),
+GEN_HANDLER_E(moduw, 0x1F, 0x0B, 0x08, 0x00000001, PPC_NONE, PPC2_ISA300),
#if defined(TARGET_PPC64)
#undef GEN_INT_ARITH_DIVD
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [RFC v2 05/13] target-ppc: add modulo word operations
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 05/13] target-ppc: add modulo word operations Nikunj A Dadhania
@ 2016-07-24 1:24 ` Richard Henderson
2016-07-25 5:37 ` Nikunj A Dadhania
` (2 more replies)
0 siblings, 3 replies; 29+ messages in thread
From: Richard Henderson @ 2016-07-24 1:24 UTC (permalink / raw)
To: Nikunj A Dadhania, qemu-ppc, david; +Cc: qemu-devel, bharata, aneesh.kumar
On 07/23/2016 02:14 PM, Nikunj A Dadhania wrote:
> Adding following instructions:
>
> moduw: Modulo Unsigned Word
> modsw: Modulo Signed Word
>
> Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
> ---
> target-ppc/helper.h | 2 ++
> target-ppc/int_helper.c | 15 +++++++++++++++
> target-ppc/translate.c | 19 +++++++++++++++++++
> 3 files changed, 36 insertions(+)
>
> diff --git a/target-ppc/helper.h b/target-ppc/helper.h
> index 1f5cfd0..76072fd 100644
> --- a/target-ppc/helper.h
> +++ b/target-ppc/helper.h
> @@ -41,6 +41,8 @@ DEF_HELPER_FLAGS_1(cntlzw, TCG_CALL_NO_RWG_SE, tl, tl)
> DEF_HELPER_FLAGS_1(popcntb, TCG_CALL_NO_RWG_SE, tl, tl)
> DEF_HELPER_FLAGS_1(popcntw, TCG_CALL_NO_RWG_SE, tl, tl)
> DEF_HELPER_FLAGS_2(cmpb, TCG_CALL_NO_RWG_SE, tl, tl, tl)
> +DEF_HELPER_FLAGS_2(modsw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
> +DEF_HELPER_FLAGS_2(moduw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
> DEF_HELPER_3(sraw, tl, env, tl, tl)
> #if defined(TARGET_PPC64)
> DEF_HELPER_FLAGS_1(cntlzd, TCG_CALL_NO_RWG_SE, tl, tl)
> diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
> index 7445376..631e0b4 100644
> --- a/target-ppc/int_helper.c
> +++ b/target-ppc/int_helper.c
> @@ -139,6 +139,21 @@ uint64_t helper_divde(CPUPPCState *env, uint64_t rau, uint64_t rbu, uint32_t oe)
>
> #endif
>
> +uint32_t helper_modsw(uint32_t rau, uint32_t rbu)
> +{
> + int32_t ra = (int32_t) rau;
> + int32_t rb = (int32_t) rbu;
> +
> + if ((rb == 0) || (ra == INT32_MIN && rb == -1)) {
> + return 0;
> + }
> + return ra % rb;
> +}
> +
> +uint32_t helper_moduw(uint32_t ra, uint32_t rb)
> +{
> + return rb ? ra % rb : 0;
> +}
I think, like you, I got distracted by the current div implementation in ppc.
I've just re-read the spec and seen the "undefined" language. Which of course
gives us much more freedom.
With this freedom, we can do the division inline, without branches. Please see
target-mips/translate.c, gen_r6_muldiv.
Basically, we check for the offending cases and modify the divisor prior to the
division. For unsigned:
a / (b == 0 ? 1 : b)
For signed:
a / ((a == INT_MAX & b == -1) | (b == 0) ? : b)
r~
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [RFC v2 05/13] target-ppc: add modulo word operations
2016-07-24 1:24 ` Richard Henderson
@ 2016-07-25 5:37 ` Nikunj A Dadhania
2016-07-25 6:07 ` [Qemu-devel] [Qemu-ppc] " Nikunj A Dadhania
2016-07-25 11:14 ` [Qemu-devel] " Nikunj A Dadhania
2016-07-25 16:31 ` Nikunj A Dadhania
2 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-25 5:37 UTC (permalink / raw)
To: Richard Henderson, qemu-ppc, david; +Cc: qemu-devel, bharata, aneesh.kumar
Richard Henderson <rth@twiddle.net> writes:
> On 07/23/2016 02:14 PM, Nikunj A Dadhania wrote:
>> Adding following instructions:
>>
>> moduw: Modulo Unsigned Word
>> modsw: Modulo Signed Word
>>
>> Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
>> ---
>> target-ppc/helper.h | 2 ++
>> target-ppc/int_helper.c | 15 +++++++++++++++
>> target-ppc/translate.c | 19 +++++++++++++++++++
>> 3 files changed, 36 insertions(+)
>>
>> diff --git a/target-ppc/helper.h b/target-ppc/helper.h
>> index 1f5cfd0..76072fd 100644
>> --- a/target-ppc/helper.h
>> +++ b/target-ppc/helper.h
>> @@ -41,6 +41,8 @@ DEF_HELPER_FLAGS_1(cntlzw, TCG_CALL_NO_RWG_SE, tl, tl)
>> DEF_HELPER_FLAGS_1(popcntb, TCG_CALL_NO_RWG_SE, tl, tl)
>> DEF_HELPER_FLAGS_1(popcntw, TCG_CALL_NO_RWG_SE, tl, tl)
>> DEF_HELPER_FLAGS_2(cmpb, TCG_CALL_NO_RWG_SE, tl, tl, tl)
>> +DEF_HELPER_FLAGS_2(modsw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
>> +DEF_HELPER_FLAGS_2(moduw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
>> DEF_HELPER_3(sraw, tl, env, tl, tl)
>> #if defined(TARGET_PPC64)
>> DEF_HELPER_FLAGS_1(cntlzd, TCG_CALL_NO_RWG_SE, tl, tl)
>> diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
>> index 7445376..631e0b4 100644
>> --- a/target-ppc/int_helper.c
>> +++ b/target-ppc/int_helper.c
>> @@ -139,6 +139,21 @@ uint64_t helper_divde(CPUPPCState *env, uint64_t rau, uint64_t rbu, uint32_t oe)
>>
>> #endif
>>
>> +uint32_t helper_modsw(uint32_t rau, uint32_t rbu)
>> +{
>> + int32_t ra = (int32_t) rau;
>> + int32_t rb = (int32_t) rbu;
>> +
>> + if ((rb == 0) || (ra == INT32_MIN && rb == -1)) {
>> + return 0;
>> + }
>> + return ra % rb;
>> +}
>> +
>> +uint32_t helper_moduw(uint32_t ra, uint32_t rb)
>> +{
>> + return rb ? ra % rb : 0;
>> +}
>
> I think, like you, I got distracted by the current div implementation in ppc.
> I've just re-read the spec and seen the "undefined" language. Which of course
> gives us much more freedom.
Right, I too thought of the same but didn't do it in this series.
Current div implementation is pretty complicated.
> With this freedom, we can do the division inline, without branches. Please see
> target-mips/translate.c, gen_r6_muldiv.
>
> Basically, we check for the offending cases and modify the divisor prior to the
> division. For unsigned:
>
> a / (b == 0 ? 1 : b)
>
> For signed:
>
> a / ((a == INT_MAX & b == -1) | (b == 0) ? : b)
Sure, will add it in this series.
Regards
Nikunj
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [Qemu-ppc] [RFC v2 05/13] target-ppc: add modulo word operations
2016-07-25 5:37 ` Nikunj A Dadhania
@ 2016-07-25 6:07 ` Nikunj A Dadhania
0 siblings, 0 replies; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-25 6:07 UTC (permalink / raw)
To: Richard Henderson, qemu-ppc, david; +Cc: qemu-devel, bharata, aneesh.kumar
Nikunj A Dadhania <nikunj@linux.vnet.ibm.com> writes:
> Richard Henderson <rth@twiddle.net> writes:
>
>> On 07/23/2016 02:14 PM, Nikunj A Dadhania wrote:
>> +uint32_t helper_modsw(uint32_t rau, uint32_t rbu)
>>> +{
>>> + int32_t ra = (int32_t) rau;
>>> + int32_t rb = (int32_t) rbu;
>>> +
>>> + if ((rb == 0) || (ra == INT32_MIN && rb == -1)) {
>>> + return 0;
>>> + }
>>> + return ra % rb;
>>> +}
>>> +
>>> +uint32_t helper_moduw(uint32_t ra, uint32_t rb)
>>> +{
>>> + return rb ? ra % rb : 0;
>>> +}
>>
>> I think, like you, I got distracted by the current div implementation in ppc.
>> I've just re-read the spec and seen the "undefined" language. Which of course
>> gives us much more freedom.
>
> Right, I too thought of the same but didn't do it in this series.
> Current div implementation is pretty complicated.
To be precise gen_op_arith_div[d,w]() implementation.
>> With this freedom, we can do the division inline, without branches. Please see
>> target-mips/translate.c, gen_r6_muldiv.
>>
>> Basically, we check for the offending cases and modify the divisor prior to the
>> division. For unsigned:
>>
>> a / (b == 0 ? 1 : b)
>>
>> For signed:
>>
>> a / ((a == INT_MAX & b == -1) | (b == 0) ? : b)
>
> Sure, will add it in this series.
>
> Regards
> Nikunj
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [Qemu-devel] [RFC v2 05/13] target-ppc: add modulo word operations
2016-07-24 1:24 ` Richard Henderson
2016-07-25 5:37 ` Nikunj A Dadhania
@ 2016-07-25 11:14 ` Nikunj A Dadhania
2016-07-25 14:34 ` Richard Henderson
2016-07-25 16:31 ` Nikunj A Dadhania
2 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-25 11:14 UTC (permalink / raw)
To: Richard Henderson, qemu-ppc, david; +Cc: qemu-devel, bharata, aneesh.kumar
Richard Henderson <rth@twiddle.net> writes:
> On 07/23/2016 02:14 PM, Nikunj A Dadhania wrote:
>> Adding following instructions:
>>
>> moduw: Modulo Unsigned Word
>> modsw: Modulo Signed Word
>>
>> Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
>> ---
>> target-ppc/helper.h | 2 ++
>> target-ppc/int_helper.c | 15 +++++++++++++++
>> target-ppc/translate.c | 19 +++++++++++++++++++
>> 3 files changed, 36 insertions(+)
>>
>> diff --git a/target-ppc/helper.h b/target-ppc/helper.h
>> index 1f5cfd0..76072fd 100644
>> --- a/target-ppc/helper.h
>> +++ b/target-ppc/helper.h
>> @@ -41,6 +41,8 @@ DEF_HELPER_FLAGS_1(cntlzw, TCG_CALL_NO_RWG_SE, tl, tl)
>> DEF_HELPER_FLAGS_1(popcntb, TCG_CALL_NO_RWG_SE, tl, tl)
>> DEF_HELPER_FLAGS_1(popcntw, TCG_CALL_NO_RWG_SE, tl, tl)
>> DEF_HELPER_FLAGS_2(cmpb, TCG_CALL_NO_RWG_SE, tl, tl, tl)
>> +DEF_HELPER_FLAGS_2(modsw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
>> +DEF_HELPER_FLAGS_2(moduw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
>> DEF_HELPER_3(sraw, tl, env, tl, tl)
>> #if defined(TARGET_PPC64)
>> DEF_HELPER_FLAGS_1(cntlzd, TCG_CALL_NO_RWG_SE, tl, tl)
>> diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
>> index 7445376..631e0b4 100644
>> --- a/target-ppc/int_helper.c
>> +++ b/target-ppc/int_helper.c
>> @@ -139,6 +139,21 @@ uint64_t helper_divde(CPUPPCState *env, uint64_t rau, uint64_t rbu, uint32_t oe)
>>
>> #endif
>>
>> +uint32_t helper_modsw(uint32_t rau, uint32_t rbu)
>> +{
>> + int32_t ra = (int32_t) rau;
>> + int32_t rb = (int32_t) rbu;
>> +
>> + if ((rb == 0) || (ra == INT32_MIN && rb == -1)) {
>> + return 0;
>> + }
>> + return ra % rb;
>> +}
>> +
>> +uint32_t helper_moduw(uint32_t ra, uint32_t rb)
>> +{
>> + return rb ? ra % rb : 0;
>> +}
>
> I think, like you, I got distracted by the current div implementation in ppc.
> I've just re-read the spec and seen the "undefined" language. Which of course
> gives us much more freedom.
>
> With this freedom, we can do the division inline, without branches. Please see
> target-mips/translate.c, gen_r6_muldiv.
>
> Basically, we check for the offending cases and modify the divisor prior to the
> division. For unsigned:
>
> a / (b == 0 ? 1 : b)
Modulo case: a % (b == 0 ? 1 : b)
tcg_gen_trunc_tl_i32(t0, cpu_gpr[rA(ctx->opcode)]);
tcg_gen_trunc_tl_i32(t1, cpu_gpr[rB(ctx->opcode)]);
tcg_gen_setcondi_i32(TCG_COND_EQ, t2, t1, 0);
tcg_gen_movi_i32(t3, 0);
tcg_gen_movcond_i32(TCG_COND_NE, t1, t2, t3, t2, t1);
tcg_gen_remu_i32(t3, t0, t1);
tcg_gen_extu_i32_tl(cpu_gpr[rD(ctx->opcode)], t3);
> For signed:
>
> a / ((a == INT_MAX & b == -1) | (b == 0) ? : b)
Modulo case: a % ((a == INT_MAX & b == -1) | (b == 0) ? 1 : b)
tcg_gen_trunc_tl_i32(t0, cpu_gpr[rA(ctx->opcode)]);
tcg_gen_trunc_tl_i32(t1, cpu_gpr[rB(ctx->opcode)]);
tcg_gen_setcondi_i32(TCG_COND_EQ, t2, t0, INT_MIN);
tcg_gen_setcondi_i32(TCG_COND_EQ, t3, t1, -1);
tcg_gen_and_i32(t2, t2, t3);
tcg_gen_setcondi_i32(TCG_COND_EQ, t3, t1, 0);
tcg_gen_or_i32(t2, t2, t3);
tcg_gen_movi_i32(t3, 0);
tcg_gen_movcond_i32(TCG_COND_NE, t1, t2, t3, t2, t1);
tcg_gen_rem_i32(t3, t0, t1);
tcg_gen_extu_i32_tl(cpu_gpr[rD(ctx->opcode)], t3);
I think you were suggesting something like above?
For "div[wd]o." we will have further cases to implement overflow.
Regards,
Nikunj
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [RFC v2 05/13] target-ppc: add modulo word operations
2016-07-25 11:14 ` [Qemu-devel] " Nikunj A Dadhania
@ 2016-07-25 14:34 ` Richard Henderson
0 siblings, 0 replies; 29+ messages in thread
From: Richard Henderson @ 2016-07-25 14:34 UTC (permalink / raw)
To: Nikunj A Dadhania, qemu-ppc, david; +Cc: qemu-devel, bharata, aneesh.kumar
On 07/25/2016 04:44 PM, Nikunj A Dadhania wrote:
> Modulo case: a % (b == 0 ? 1 : b)
>
> tcg_gen_trunc_tl_i32(t0, cpu_gpr[rA(ctx->opcode)]);
> tcg_gen_trunc_tl_i32(t1, cpu_gpr[rB(ctx->opcode)]);
> tcg_gen_setcondi_i32(TCG_COND_EQ, t2, t1, 0);
> tcg_gen_movi_i32(t3, 0);
> tcg_gen_movcond_i32(TCG_COND_NE, t1, t2, t3, t2, t1);
No setcond for unsigned; you can do the whole thing with movcond.
t2 = tcg_const_i32(1);
t3 = tcg_const_i32(0);
tcg_gen_movcond_i32(TCG_COND_EQ, t1, t1, t3, t2, t1);
But otherwise, yes.
> For "div[wd]o." we will have further cases to implement overflow.
For divwo, you *would* actually do the setcond as above, because that would be
exactly the overflow condition that you're supposed to compute.
The signed case you pasted looks correct, and the T2 input to the movcond is
also exactly the overflow condition.
r~
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [Qemu-devel] [RFC v2 05/13] target-ppc: add modulo word operations
2016-07-24 1:24 ` Richard Henderson
2016-07-25 5:37 ` Nikunj A Dadhania
2016-07-25 11:14 ` [Qemu-devel] " Nikunj A Dadhania
@ 2016-07-25 16:31 ` Nikunj A Dadhania
2 siblings, 0 replies; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-25 16:31 UTC (permalink / raw)
To: Richard Henderson, qemu-ppc, david; +Cc: qemu-devel, bharata, aneesh.kumar
Richard Henderson <rth@twiddle.net> writes:
> On 07/23/2016 02:14 PM, Nikunj A Dadhania wrote:
[...]
> Basically, we check for the offending cases and modify the divisor prior to the
> division. For unsigned:
>
> a / (b == 0 ? 1 : b)
>
> For signed:
>
> a / ((a == INT_MAX & b == -1) | (b == 0) ? : b)
So when we change the divisor to 1, undefined result is set as "a". Hope
that should be fine and doesnt break anything :-)
Regards,
Nikunj
^ permalink raw reply [flat|nested] 29+ messages in thread
* [Qemu-devel] [RFC v2 06/13] target-ppc: add modulo dword operations
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
` (4 preceding siblings ...)
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 05/13] target-ppc: add modulo word operations Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 07/13] target-ppc: add cnttzd[.] instruction Nikunj A Dadhania
` (6 subsequent siblings)
12 siblings, 0 replies; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
Adding following instructions for ISA3.0 support
modud: Modulo Unsigned Dword
modsd: Modulo Signed Dword
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
target-ppc/helper.h | 2 ++
target-ppc/int_helper.c | 16 ++++++++++++++++
target-ppc/translate.c | 15 +++++++++++++++
3 files changed, 33 insertions(+)
diff --git a/target-ppc/helper.h b/target-ppc/helper.h
index 76072fd..26a0930 100644
--- a/target-ppc/helper.h
+++ b/target-ppc/helper.h
@@ -45,6 +45,8 @@ DEF_HELPER_FLAGS_2(modsw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
DEF_HELPER_FLAGS_2(moduw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
DEF_HELPER_3(sraw, tl, env, tl, tl)
#if defined(TARGET_PPC64)
+DEF_HELPER_FLAGS_2(modsd, TCG_CALL_NO_RWG_SE, i64, i64, i64)
+DEF_HELPER_FLAGS_2(modud, TCG_CALL_NO_RWG_SE, i64, i64, i64)
DEF_HELPER_FLAGS_1(cntlzd, TCG_CALL_NO_RWG_SE, tl, tl)
DEF_HELPER_FLAGS_1(popcntd, TCG_CALL_NO_RWG_SE, tl, tl)
DEF_HELPER_FLAGS_2(bpermd, TCG_CALL_NO_RWG_SE, i64, i64, i64)
diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
index 631e0b4..e95572b 100644
--- a/target-ppc/int_helper.c
+++ b/target-ppc/int_helper.c
@@ -161,6 +161,22 @@ target_ulong helper_cntlzw(target_ulong t)
}
#if defined(TARGET_PPC64)
+uint64_t helper_modsd(uint64_t rau, uint64_t rbu)
+{
+ int64_t ra = (int64_t)rau;
+ int64_t rb = (int64_t)rbu;
+
+ if ((rb == 0) || (ra == INT32_MIN && rb == -1)) {
+ return 0;
+ }
+ return ra % rb;
+}
+
+uint64_t helper_modud(uint64_t ra, uint64_t rb)
+{
+ return rb ? ra % rb : 0;
+}
+
target_ulong helper_cntlzd(target_ulong t)
{
return clz64(t);
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 4348efd..38a117c 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -1192,6 +1192,19 @@ static void glue(gen_, name)(DisasContext *ctx) \
GEN_INT_ARITH_MODW(moduw, 0x08, 0);
GEN_INT_ARITH_MODW(modsw, 0x18, 1);
+#if defined(TARGET_PPC64)
+#define GEN_INT_ARITH_MODD(name, opc3, sign) \
+ static void glue(gen_, name)(DisasContext *ctx) \
+{ \
+ gen_helper_##name(cpu_gpr[rD(ctx->opcode)], \
+ cpu_gpr[rA(ctx->opcode)], \
+ cpu_gpr[rB(ctx->opcode)]); \
+}
+
+GEN_INT_ARITH_MODD(modud, 0x08, 0);
+GEN_INT_ARITH_MODD(modsd, 0x18, 1);
+#endif
+
/* mulhw mulhw. */
static void gen_mulhw(DisasContext *ctx)
{
@@ -10274,6 +10287,8 @@ GEN_HANDLER_E(divdeu, 0x1F, 0x09, 0x0C, 0, PPC_NONE, PPC2_DIVE_ISA206),
GEN_HANDLER_E(divdeuo, 0x1F, 0x09, 0x1C, 0, PPC_NONE, PPC2_DIVE_ISA206),
GEN_HANDLER_E(divde, 0x1F, 0x09, 0x0D, 0, PPC_NONE, PPC2_DIVE_ISA206),
GEN_HANDLER_E(divdeo, 0x1F, 0x09, 0x1D, 0, PPC_NONE, PPC2_DIVE_ISA206),
+GEN_HANDLER_E(modsd, 0x1F, 0x09, 0x18, 0x00000001, PPC_NONE, PPC2_ISA300),
+GEN_HANDLER_E(modud, 0x1F, 0x09, 0x08, 0x00000001, PPC_NONE, PPC2_ISA300),
#undef GEN_INT_ARITH_MUL_HELPER
#define GEN_INT_ARITH_MUL_HELPER(name, opc3) \
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* [Qemu-devel] [RFC v2 07/13] target-ppc: add cnttzd[.] instruction
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
` (5 preceding siblings ...)
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 06/13] target-ppc: add modulo dword operations Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-24 1:26 ` Richard Henderson
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 08/13] target-ppc: add cnttzw[.] instruction Nikunj A Dadhania
` (5 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth
Cc: qemu-devel, nikunj, bharata, aneesh.kumar, Sandipan Das
From: Sandipan Das <sandipandas1990@gmail.com>
Add ISA3.0 Count trailing zeros double word
Signed-off-by: Sandipan Das <sandipandas1990@gmail.com>
[ added ISA300 flag ]
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
target-ppc/helper.h | 1 +
target-ppc/int_helper.c | 5 +++++
target-ppc/translate.c | 10 ++++++++++
3 files changed, 16 insertions(+)
diff --git a/target-ppc/helper.h b/target-ppc/helper.h
index 26a0930..8a3eb5d 100644
--- a/target-ppc/helper.h
+++ b/target-ppc/helper.h
@@ -48,6 +48,7 @@ DEF_HELPER_3(sraw, tl, env, tl, tl)
DEF_HELPER_FLAGS_2(modsd, TCG_CALL_NO_RWG_SE, i64, i64, i64)
DEF_HELPER_FLAGS_2(modud, TCG_CALL_NO_RWG_SE, i64, i64, i64)
DEF_HELPER_FLAGS_1(cntlzd, TCG_CALL_NO_RWG_SE, tl, tl)
+DEF_HELPER_FLAGS_1(cnttzd, TCG_CALL_NO_RWG_SE, tl, tl)
DEF_HELPER_FLAGS_1(popcntd, TCG_CALL_NO_RWG_SE, tl, tl)
DEF_HELPER_FLAGS_2(bpermd, TCG_CALL_NO_RWG_SE, i64, i64, i64)
DEF_HELPER_3(srad, tl, env, tl, tl)
diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
index e95572b..1d02e8a 100644
--- a/target-ppc/int_helper.c
+++ b/target-ppc/int_helper.c
@@ -181,6 +181,11 @@ target_ulong helper_cntlzd(target_ulong t)
{
return clz64(t);
}
+
+target_ulong helper_cnttzd(target_ulong t)
+{
+ return ctz64(t);
+}
#endif
#if defined(TARGET_PPC64)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 38a117c..1fc1922 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -1760,6 +1760,15 @@ static void gen_cntlzd(DisasContext *ctx)
if (unlikely(Rc(ctx->opcode) != 0))
gen_set_Rc0(ctx, cpu_gpr[rA(ctx->opcode)]);
}
+
+/* cnttzd */
+static void gen_cnttzd(DisasContext *ctx)
+{
+ gen_helper_cnttzd(cpu_gpr[rA(ctx->opcode)], cpu_gpr[rS(ctx->opcode)]);
+ if (unlikely(Rc(ctx->opcode) != 0)) {
+ gen_set_Rc0(ctx, cpu_gpr[rA(ctx->opcode)]);
+ }
+}
#endif
/*** Integer rotate ***/
@@ -9997,6 +10006,7 @@ GEN_HANDLER_E(prtyw, 0x1F, 0x1A, 0x04, 0x0000F801, PPC_NONE, PPC2_ISA205),
#if defined(TARGET_PPC64)
GEN_HANDLER(popcntd, 0x1F, 0x1A, 0x0F, 0x0000F801, PPC_POPCNTWD),
GEN_HANDLER(cntlzd, 0x1F, 0x1A, 0x01, 0x00000000, PPC_64B),
+GEN_HANDLER_E(cnttzd, 0x1F, 0x1A, 0x11, 0x00000000, PPC_NONE, PPC2_ISA300),
GEN_HANDLER_E(prtyd, 0x1F, 0x1A, 0x05, 0x0000F801, PPC_NONE, PPC2_ISA205),
GEN_HANDLER_E(bpermd, 0x1F, 0x1C, 0x07, 0x00000001, PPC_NONE, PPC2_PERM_ISA206),
#endif
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [RFC v2 07/13] target-ppc: add cnttzd[.] instruction
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 07/13] target-ppc: add cnttzd[.] instruction Nikunj A Dadhania
@ 2016-07-24 1:26 ` Richard Henderson
0 siblings, 0 replies; 29+ messages in thread
From: Richard Henderson @ 2016-07-24 1:26 UTC (permalink / raw)
To: Nikunj A Dadhania, qemu-ppc, david
Cc: qemu-devel, bharata, aneesh.kumar, Sandipan Das
On 07/23/2016 02:14 PM, Nikunj A Dadhania wrote:
> From: Sandipan Das <sandipandas1990@gmail.com>
>
> Add ISA3.0 Count trailing zeros double word
>
> Signed-off-by: Sandipan Das <sandipandas1990@gmail.com>
> [ added ISA300 flag ]
> Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
> ---
> target-ppc/helper.h | 1 +
> target-ppc/int_helper.c | 5 +++++
> target-ppc/translate.c | 10 ++++++++++
> 3 files changed, 16 insertions(+)
Reviewed-by: Richard Henderson <rth@twiddle.net>
r~
^ permalink raw reply [flat|nested] 29+ messages in thread
* [Qemu-devel] [RFC v2 08/13] target-ppc: add cnttzw[.] instruction
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
` (6 preceding siblings ...)
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 07/13] target-ppc: add cnttzd[.] instruction Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-24 1:26 ` Richard Henderson
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 09/13] target-ppc: add cmpeqb instruction Nikunj A Dadhania
` (4 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
Add ISA3.0: Count trailing zeros word instruction.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
target-ppc/helper.h | 1 +
target-ppc/int_helper.c | 5 +++++
target-ppc/translate.c | 11 +++++++++++
3 files changed, 17 insertions(+)
diff --git a/target-ppc/helper.h b/target-ppc/helper.h
index 8a3eb5d..3b3dc36 100644
--- a/target-ppc/helper.h
+++ b/target-ppc/helper.h
@@ -38,6 +38,7 @@ DEF_HELPER_4(divweu, tl, env, tl, tl, i32)
DEF_HELPER_4(divwe, tl, env, tl, tl, i32)
DEF_HELPER_FLAGS_1(cntlzw, TCG_CALL_NO_RWG_SE, tl, tl)
+DEF_HELPER_FLAGS_1(cnttzw, TCG_CALL_NO_RWG_SE, tl, tl)
DEF_HELPER_FLAGS_1(popcntb, TCG_CALL_NO_RWG_SE, tl, tl)
DEF_HELPER_FLAGS_1(popcntw, TCG_CALL_NO_RWG_SE, tl, tl)
DEF_HELPER_FLAGS_2(cmpb, TCG_CALL_NO_RWG_SE, tl, tl, tl)
diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
index 1d02e8a..ac47230 100644
--- a/target-ppc/int_helper.c
+++ b/target-ppc/int_helper.c
@@ -160,6 +160,11 @@ target_ulong helper_cntlzw(target_ulong t)
return clz32(t);
}
+target_ulong helper_cnttzw(target_ulong t)
+{
+ return ctz32(t);
+}
+
#if defined(TARGET_PPC64)
uint64_t helper_modsd(uint64_t rau, uint64_t rbu)
{
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 1fc1922..c28ddff 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -1520,6 +1520,16 @@ static void gen_cntlzw(DisasContext *ctx)
if (unlikely(Rc(ctx->opcode) != 0))
gen_set_Rc0(ctx, cpu_gpr[rA(ctx->opcode)]);
}
+
+/* cnttzw */
+static void gen_cnttzw(DisasContext *ctx)
+{
+ gen_helper_cnttzw(cpu_gpr[rA(ctx->opcode)], cpu_gpr[rS(ctx->opcode)]);
+ if (unlikely(Rc(ctx->opcode) != 0)) {
+ gen_set_Rc0(ctx, cpu_gpr[rA(ctx->opcode)]);
+ }
+}
+
/* eqv & eqv. */
GEN_LOGICAL2(eqv, tcg_gen_eqv_tl, 0x08, PPC_INTEGER);
/* extsb & extsb. */
@@ -9994,6 +10004,7 @@ GEN_HANDLER(subfic, 0x08, 0xFF, 0xFF, 0x00000000, PPC_INTEGER),
GEN_HANDLER2(andi_, "andi.", 0x1C, 0xFF, 0xFF, 0x00000000, PPC_INTEGER),
GEN_HANDLER2(andis_, "andis.", 0x1D, 0xFF, 0xFF, 0x00000000, PPC_INTEGER),
GEN_HANDLER(cntlzw, 0x1F, 0x1A, 0x00, 0x00000000, PPC_INTEGER),
+GEN_HANDLER_E(cnttzw, 0x1F, 0x1A, 0x10, 0x00000000, PPC_NONE, PPC2_ISA300),
GEN_HANDLER(or, 0x1F, 0x1C, 0x0D, 0x00000000, PPC_INTEGER),
GEN_HANDLER(xor, 0x1F, 0x1C, 0x09, 0x00000000, PPC_INTEGER),
GEN_HANDLER(ori, 0x18, 0xFF, 0xFF, 0x00000000, PPC_INTEGER),
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* [Qemu-devel] [RFC v2 09/13] target-ppc: add cmpeqb instruction
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
` (7 preceding siblings ...)
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 08/13] target-ppc: add cnttzw[.] instruction Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-24 1:31 ` Richard Henderson
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 10/13] target-ppc: add setb instruction Nikunj A Dadhania
` (3 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
Search a byte in the stream of 8bytes provided in the register
Suggested-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
target-ppc/helper.h | 1 +
target-ppc/int_helper.c | 22 ++++++++++++++++++++++
target-ppc/translate.c | 12 ++++++++++++
3 files changed, 35 insertions(+)
diff --git a/target-ppc/helper.h b/target-ppc/helper.h
index 3b3dc36..bd6f8f1 100644
--- a/target-ppc/helper.h
+++ b/target-ppc/helper.h
@@ -46,6 +46,7 @@ DEF_HELPER_FLAGS_2(modsw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
DEF_HELPER_FLAGS_2(moduw, TCG_CALL_NO_RWG_SE, i32, i32, i32)
DEF_HELPER_3(sraw, tl, env, tl, tl)
#if defined(TARGET_PPC64)
+DEF_HELPER_FLAGS_2(cmpeqb, TCG_CALL_NO_RWG_SE, i32, tl, tl)
DEF_HELPER_FLAGS_2(modsd, TCG_CALL_NO_RWG_SE, i64, i64, i64)
DEF_HELPER_FLAGS_2(modud, TCG_CALL_NO_RWG_SE, i64, i64, i64)
DEF_HELPER_FLAGS_1(cntlzd, TCG_CALL_NO_RWG_SE, tl, tl)
diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
index ac47230..280603f 100644
--- a/target-ppc/int_helper.c
+++ b/target-ppc/int_helper.c
@@ -166,6 +166,28 @@ target_ulong helper_cnttzw(target_ulong t)
}
#if defined(TARGET_PPC64)
+/* if x = 0xab, returns 0xababababababababa */
+#define pattern(x) (((x) & 0xff) * (~(target_ulong)0 / 0xff))
+
+/* substract 1 from each byte, and with inverse, check if MSB is set at each
+ * byte.
+ * i.e. ((0x00 - 0x01) & ~(0x00)) & 0x80
+ * (0xFF & 0xFF) & 0x80 = 0x80 (zero found)
+ */
+#define haszero(v) (((v) - pattern(0x01)) & ~(v) & pattern(0x80))
+
+/* When you XOR the pattern and there is a match, that byte will be zero */
+#define hasvalue(x, n) (haszero((x) ^ pattern(n)))
+
+uint32_t helper_cmpeqb(target_ulong ra, target_ulong rb)
+{
+ return hasvalue(rb, ra) ? 1 << CRF_GT : 0;
+}
+
+#undef pattern
+#undef haszero
+#undef hasvalue
+
uint64_t helper_modsd(uint64_t rau, uint64_t rbu)
{
int64_t ra = (int64_t)rau;
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index c28ddff..0d999d4 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -854,6 +854,15 @@ static void gen_cmprb(DisasContext *ctx)
tcg_temp_free_i32(src2hi);
}
+#if defined(TARGET_PPC64)
+/* cmpeqb */
+static void gen_cmpeqb(DisasContext *ctx)
+{
+ gen_helper_cmpeqb(cpu_crf[crfD(ctx->opcode)], cpu_gpr[rA(ctx->opcode)],
+ cpu_gpr[rB(ctx->opcode)]);
+}
+#endif
+
/* isel (PowerPC 2.03 specification) */
static void gen_isel(DisasContext *ctx)
{
@@ -9982,6 +9991,9 @@ GEN_HANDLER(cmp, 0x1F, 0x00, 0x00, 0x00400000, PPC_INTEGER),
GEN_HANDLER(cmpi, 0x0B, 0xFF, 0xFF, 0x00400000, PPC_INTEGER),
GEN_HANDLER(cmpl, 0x1F, 0x00, 0x01, 0x00400000, PPC_INTEGER),
GEN_HANDLER(cmpli, 0x0A, 0xFF, 0xFF, 0x00400000, PPC_INTEGER),
+#if defined(TARGET_PPC64)
+GEN_HANDLER_E(cmpeqb, 0x1F, 0x00, 0x07, 0x00600000, PPC_NONE, PPC2_ISA300),
+#endif
GEN_HANDLER_E(cmpb, 0x1F, 0x1C, 0x0F, 0x00000001, PPC_NONE, PPC2_ISA205),
GEN_HANDLER_E(cmprb, 0x1F, 0x00, 0x06, 0x00400001, PPC_NONE, PPC2_ISA300),
GEN_HANDLER(isel, 0x1F, 0x0F, 0xFF, 0x00000001, PPC_ISEL),
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [RFC v2 09/13] target-ppc: add cmpeqb instruction
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 09/13] target-ppc: add cmpeqb instruction Nikunj A Dadhania
@ 2016-07-24 1:31 ` Richard Henderson
0 siblings, 0 replies; 29+ messages in thread
From: Richard Henderson @ 2016-07-24 1:31 UTC (permalink / raw)
To: Nikunj A Dadhania, qemu-ppc, david; +Cc: qemu-devel, bharata, aneesh.kumar
On 07/23/2016 02:14 PM, Nikunj A Dadhania wrote:
> Search a byte in the stream of 8bytes provided in the register
>
> Suggested-by: Richard Henderson <rth@twiddle.net>
> Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
> ---
> target-ppc/helper.h | 1 +
> target-ppc/int_helper.c | 22 ++++++++++++++++++++++
> target-ppc/translate.c | 12 ++++++++++++
> 3 files changed, 35 insertions(+)
Reviewed-by: Richard Henderson <rth@twiddle.net>
r~
^ permalink raw reply [flat|nested] 29+ messages in thread
* [Qemu-devel] [RFC v2 10/13] target-ppc: add setb instruction
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
` (8 preceding siblings ...)
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 09/13] target-ppc: add cmpeqb instruction Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-24 1:35 ` Richard Henderson
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 11/13] target-ppc: add maddld instruction Nikunj A Dadhania
` (2 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth
Cc: qemu-devel, nikunj, bharata, aneesh.kumar, Vivek Andrew Sha
From: Vivek Andrew Sha <vivekandrewsha@gmail.com>
The CR number is provided in the opcode as - BFA (11:13)
Returns:
-1 if bit 0 of CR field is set
1 if bit 1 of CR field is set
0 otherwise.
Signed-off-by: Vivek Andrew Sha <vivekandrewsha@gmail.com>
[ reworded commit, used 32bit ops as crf is 32bits ]
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
target-ppc/translate.c | 22 ++++++++++++++++++++++
1 file changed, 22 insertions(+)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 0d999d4..ba40fe8 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -4804,6 +4804,27 @@ static void gen_mtspr(DisasContext *ctx)
}
}
+#if defined(TARGET_PPC64)
+/* setb */
+static void gen_setb(DisasContext *ctx)
+{
+ TCGv_i32 t0 = tcg_temp_new_i32();
+ TCGv_i32 t8 = tcg_temp_new_i32();
+ TCGv_i32 tm1 = tcg_temp_new_i32();
+ int crf = crfS(ctx->opcode);
+
+ tcg_gen_setcondi_i32(TCG_COND_GEU, t0, cpu_crf[crf], 4);
+ tcg_gen_movi_i32(t8, 8);
+ tcg_gen_movi_i32(tm1, -1);
+ tcg_gen_movcond_i32(TCG_COND_GEU, t0, cpu_crf[crf], t8, tm1, t0);
+ tcg_gen_ext_i32_tl(cpu_gpr[rD(ctx->opcode)], t0);
+
+ tcg_temp_free_i32(t0);
+ tcg_temp_free_i32(t8);
+ tcg_temp_free_i32(tm1);
+}
+#endif
+
/*** Cache management ***/
/* dcbf */
@@ -10122,6 +10143,7 @@ GEN_HANDLER(mftb, 0x1F, 0x13, 0x0B, 0x00000001, PPC_MFTB),
GEN_HANDLER(mtcrf, 0x1F, 0x10, 0x04, 0x00000801, PPC_MISC),
#if defined(TARGET_PPC64)
GEN_HANDLER(mtmsrd, 0x1F, 0x12, 0x05, 0x001EF801, PPC_64B),
+GEN_HANDLER_E(setb, 0x1F, 0x00, 0x04, 0x0003F801, PPC_NONE, PPC2_ISA300),
#endif
GEN_HANDLER(mtmsr, 0x1F, 0x12, 0x04, 0x001EF801, PPC_MISC),
GEN_HANDLER(mtspr, 0x1F, 0x13, 0x0E, 0x00000000, PPC_MISC),
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [RFC v2 10/13] target-ppc: add setb instruction
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 10/13] target-ppc: add setb instruction Nikunj A Dadhania
@ 2016-07-24 1:35 ` Richard Henderson
0 siblings, 0 replies; 29+ messages in thread
From: Richard Henderson @ 2016-07-24 1:35 UTC (permalink / raw)
To: Nikunj A Dadhania, qemu-ppc, david
Cc: qemu-devel, bharata, aneesh.kumar, Vivek Andrew Sha
On 07/23/2016 02:14 PM, Nikunj A Dadhania wrote:
> From: Vivek Andrew Sha <vivekandrewsha@gmail.com>
>
> The CR number is provided in the opcode as - BFA (11:13)
>
> Returns:
> -1 if bit 0 of CR field is set
> 1 if bit 1 of CR field is set
> 0 otherwise.
>
> Signed-off-by: Vivek Andrew Sha <vivekandrewsha@gmail.com>
> [ reworded commit, used 32bit ops as crf is 32bits ]
> Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
> ---
> target-ppc/translate.c | 22 ++++++++++++++++++++++
> 1 file changed, 22 insertions(+)
Reviewed-by: Richard Henderson <rth@twiddle.net>
r~
^ permalink raw reply [flat|nested] 29+ messages in thread
* [Qemu-devel] [RFC v2 11/13] target-ppc: add maddld instruction
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
` (9 preceding siblings ...)
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 10/13] target-ppc: add setb instruction Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-24 1:36 ` Richard Henderson
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 12/13] target-ppc: add maddhd and maddhdu instruction Nikunj A Dadhania
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 13/13] target-ppc: introduce opc4 for Expanded Opcode Nikunj A Dadhania
12 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
maddld: Multiply-Add Low Doubleword
Multiplies two 64-bit registers (RA * RB), adds third register(RC) to
the result(quadword) and returns the lower dword in the target
register(RT).
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
target-ppc/translate.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index ba40fe8..ebce3e8 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -7677,6 +7677,17 @@ GEN_VAFORM_PAIRED(vmsumshm, vmsumshs, 20)
GEN_VAFORM_PAIRED(vsel, vperm, 21)
GEN_VAFORM_PAIRED(vmaddfp, vnmsubfp, 23)
+#if defined(TARGET_PPC64)
+static void gen_maddld(DisasContext *ctx)
+{
+ TCGv_i64 t1 = tcg_temp_new_i64();
+
+ tcg_gen_mul_i64(t1, cpu_gpr[rA(ctx->opcode)], cpu_gpr[rB(ctx->opcode)]);
+ tcg_gen_add_i64(cpu_gpr[rD(ctx->opcode)], t1, cpu_gpr[rC(ctx->opcode)]);
+ tcg_temp_free_i64(t1);
+}
+#endif /* defined(TARGET_PPC64) */
+
GEN_VXFORM_NOA(vclzb, 1, 28)
GEN_VXFORM_NOA(vclzh, 1, 29)
GEN_VXFORM_NOA(vclzw, 1, 30)
@@ -10292,6 +10303,9 @@ GEN_HANDLER(lvsr, 0x1f, 0x06, 0x01, 0x00000001, PPC_ALTIVEC),
GEN_HANDLER(mfvscr, 0x04, 0x2, 0x18, 0x001ff800, PPC_ALTIVEC),
GEN_HANDLER(mtvscr, 0x04, 0x2, 0x19, 0x03ff0000, PPC_ALTIVEC),
GEN_HANDLER(vmladduhm, 0x04, 0x11, 0xFF, 0x00000000, PPC_ALTIVEC),
+#if defined(TARGET_PPC64)
+GEN_HANDLER_E(maddld, 0x04, 0x19, 0xFF, 0x00000000, PPC_NONE, PPC2_ISA300),
+#endif
GEN_HANDLER2(evsel0, "evsel", 0x04, 0x1c, 0x09, 0x00000000, PPC_SPE),
GEN_HANDLER2(evsel1, "evsel", 0x04, 0x1d, 0x09, 0x00000000, PPC_SPE),
GEN_HANDLER2(evsel2, "evsel", 0x04, 0x1e, 0x09, 0x00000000, PPC_SPE),
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [RFC v2 11/13] target-ppc: add maddld instruction
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 11/13] target-ppc: add maddld instruction Nikunj A Dadhania
@ 2016-07-24 1:36 ` Richard Henderson
0 siblings, 0 replies; 29+ messages in thread
From: Richard Henderson @ 2016-07-24 1:36 UTC (permalink / raw)
To: Nikunj A Dadhania, qemu-ppc, david; +Cc: qemu-devel, bharata, aneesh.kumar
On 07/23/2016 02:14 PM, Nikunj A Dadhania wrote:
> maddld: Multiply-Add Low Doubleword
>
> Multiplies two 64-bit registers (RA * RB), adds third register(RC) to
> the result(quadword) and returns the lower dword in the target
> register(RT).
>
> Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
> ---
> target-ppc/translate.c | 14 ++++++++++++++
> 1 file changed, 14 insertions(+)
Reviewed-by: Richard Henderson <rth@twiddle.net>
r~
^ permalink raw reply [flat|nested] 29+ messages in thread
* [Qemu-devel] [RFC v2 12/13] target-ppc: add maddhd and maddhdu instruction
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
` (10 preceding siblings ...)
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 11/13] target-ppc: add maddld instruction Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
2016-07-24 1:36 ` Richard Henderson
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 13/13] target-ppc: introduce opc4 for Expanded Opcode Nikunj A Dadhania
12 siblings, 1 reply; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
maddhd: Multiply-Add High Doubleword
maddhdu: Multiply-Add High Doubleword Unsigned
Above two instruction are dual form and differ by 1 bit
(31st bit)
Multiplies two 64-bit registers (RA * RB), adds third register(RC) to
the result(quadword) and returns the higher dword in the target
register(RT).
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
target-ppc/translate.c | 25 +++++++++++++++++++++++++
1 file changed, 25 insertions(+)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index ebce3e8..99203e8 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -7686,6 +7686,29 @@ static void gen_maddld(DisasContext *ctx)
tcg_gen_add_i64(cpu_gpr[rD(ctx->opcode)], t1, cpu_gpr[rC(ctx->opcode)]);
tcg_temp_free_i64(t1);
}
+
+/* maddhd maddhdu */
+static void gen_maddhd_maddhdu(DisasContext *ctx)
+{
+ TCGv_i64 lo = tcg_temp_new_i64();
+ TCGv_i64 hi = tcg_temp_new_i64();
+ TCGv_i64 t1 = tcg_temp_new_i64();
+
+ if (Rc(ctx->opcode)) {
+ tcg_gen_mulu2_i64(lo, hi, cpu_gpr[rA(ctx->opcode)],
+ cpu_gpr[rB(ctx->opcode)]);
+ tcg_gen_movi_i64(t1, 0);
+ } else {
+ tcg_gen_muls2_i64(lo, hi, cpu_gpr[rA(ctx->opcode)],
+ cpu_gpr[rB(ctx->opcode)]);
+ tcg_gen_sari_i64(t1, cpu_gpr[rC(ctx->opcode)], 63);
+ }
+ tcg_gen_add2_i64(t1, cpu_gpr[rD(ctx->opcode)], lo, hi,
+ cpu_gpr[rC(ctx->opcode)], t1);
+ tcg_temp_free_i64(lo);
+ tcg_temp_free_i64(hi);
+ tcg_temp_free_i64(t1);
+}
#endif /* defined(TARGET_PPC64) */
GEN_VXFORM_NOA(vclzb, 1, 28)
@@ -10304,6 +10327,8 @@ GEN_HANDLER(mfvscr, 0x04, 0x2, 0x18, 0x001ff800, PPC_ALTIVEC),
GEN_HANDLER(mtvscr, 0x04, 0x2, 0x19, 0x03ff0000, PPC_ALTIVEC),
GEN_HANDLER(vmladduhm, 0x04, 0x11, 0xFF, 0x00000000, PPC_ALTIVEC),
#if defined(TARGET_PPC64)
+GEN_HANDLER_E(maddhd_maddhdu, 0x04, 0x18, 0xFF, 0x00000000, PPC_NONE,
+ PPC2_ISA300),
GEN_HANDLER_E(maddld, 0x04, 0x19, 0xFF, 0x00000000, PPC_NONE, PPC2_ISA300),
#endif
GEN_HANDLER2(evsel0, "evsel", 0x04, 0x1c, 0x09, 0x00000000, PPC_SPE),
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [Qemu-devel] [RFC v2 12/13] target-ppc: add maddhd and maddhdu instruction
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 12/13] target-ppc: add maddhd and maddhdu instruction Nikunj A Dadhania
@ 2016-07-24 1:36 ` Richard Henderson
0 siblings, 0 replies; 29+ messages in thread
From: Richard Henderson @ 2016-07-24 1:36 UTC (permalink / raw)
To: Nikunj A Dadhania, qemu-ppc, david; +Cc: qemu-devel, bharata, aneesh.kumar
On 07/23/2016 02:14 PM, Nikunj A Dadhania wrote:
> maddhd: Multiply-Add High Doubleword
> maddhdu: Multiply-Add High Doubleword Unsigned
>
> Above two instruction are dual form and differ by 1 bit
> (31st bit)
>
> Multiplies two 64-bit registers (RA * RB), adds third register(RC) to
> the result(quadword) and returns the higher dword in the target
> register(RT).
>
> Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
> ---
> target-ppc/translate.c | 25 +++++++++++++++++++++++++
> 1 file changed, 25 insertions(+)
Reviewed-by: Richard Henderson <rth@twiddle.net>
r~
^ permalink raw reply [flat|nested] 29+ messages in thread
* [Qemu-devel] [RFC v2 13/13] target-ppc: introduce opc4 for Expanded Opcode
2016-07-23 8:44 [Qemu-devel] [RFC v2 00/13] POWER9 TCG enablements - part1 Nikunj A Dadhania
` (11 preceding siblings ...)
2016-07-23 8:44 ` [Qemu-devel] [RFC v2 12/13] target-ppc: add maddhd and maddhdu instruction Nikunj A Dadhania
@ 2016-07-23 8:44 ` Nikunj A Dadhania
12 siblings, 0 replies; 29+ messages in thread
From: Nikunj A Dadhania @ 2016-07-23 8:44 UTC (permalink / raw)
To: qemu-ppc, david, rth; +Cc: qemu-devel, nikunj, bharata, aneesh.kumar
ISA 3.0 has introduced EO - Expanded Opcode. Introduce third level
indirect opcode table and corresponding parsing routines.
EO (11:12) Expanded opcode field
Formats: XX1
EO (11:15) Expanded opcode field
Formats: VX, X, XX2
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
---
target-ppc/translate.c | 82 ++++++++++++++++++++--------
target-ppc/translate_init.c | 126 ++++++++++++++++++++++++++++++++------------
2 files changed, 154 insertions(+), 54 deletions(-)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 99203e8..912be16 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -367,12 +367,13 @@ GEN_OPCODE2(name, onam, opc1, opc2, opc3, inval, type, PPC_NONE)
#define GEN_HANDLER2_E(name, onam, opc1, opc2, opc3, inval, type, type2) \
GEN_OPCODE2(name, onam, opc1, opc2, opc3, inval, type, type2)
+#define GEN_HANDLER_E_2(name, opc1, opc2, opc3, opc4, inval, type, type2) \
+GEN_OPCODE3(name, opc1, opc2, opc3, opc4, inval, type, type2)
+
typedef struct opcode_t {
- unsigned char opc1, opc2, opc3;
+ unsigned char opc1, opc2, opc3, opc4;
#if HOST_LONG_BITS == 64 /* Explicitly align to 64 bits */
- unsigned char pad[5];
-#else
- unsigned char pad[1];
+ unsigned char pad[4];
#endif
opc_handler_t handler;
const char *oname;
@@ -452,6 +453,8 @@ EXTRACT_HELPER(opc1, 26, 6);
EXTRACT_HELPER(opc2, 1, 5);
/* Opcode part 3 */
EXTRACT_HELPER(opc3, 6, 5);
+/* Opcode part 4 */
+EXTRACT_HELPER(opc4, 16, 5);
/* Update Cr0 flags */
EXTRACT_HELPER(Rc, 0, 1);
/* Update Cr6 flags (Altivec) */
@@ -589,7 +592,7 @@ EXTRACT_HELPER(SP, 19, 2);
.opc1 = op1, \
.opc2 = op2, \
.opc3 = op3, \
- .pad = { 0, }, \
+ .opc4 = 0xff, \
.handler = { \
.inval1 = invl, \
.type = _typ, \
@@ -604,7 +607,7 @@ EXTRACT_HELPER(SP, 19, 2);
.opc1 = op1, \
.opc2 = op2, \
.opc3 = op3, \
- .pad = { 0, }, \
+ .opc4 = 0xff, \
.handler = { \
.inval1 = invl1, \
.inval2 = invl2, \
@@ -620,7 +623,7 @@ EXTRACT_HELPER(SP, 19, 2);
.opc1 = op1, \
.opc2 = op2, \
.opc3 = op3, \
- .pad = { 0, }, \
+ .opc4 = 0xff, \
.handler = { \
.inval1 = invl, \
.type = _typ, \
@@ -630,13 +633,28 @@ EXTRACT_HELPER(SP, 19, 2);
}, \
.oname = onam, \
}
+#define GEN_OPCODE3(name, op1, op2, op3, op4, invl, _typ, _typ2) \
+{ \
+ .opc1 = op1, \
+ .opc2 = op2, \
+ .opc3 = op3, \
+ .opc4 = op4, \
+ .handler = { \
+ .inval1 = invl, \
+ .type = _typ, \
+ .type2 = _typ2, \
+ .handler = &gen_##name, \
+ .oname = stringify(name), \
+ }, \
+ .oname = stringify(name), \
+}
#else
#define GEN_OPCODE(name, op1, op2, op3, invl, _typ, _typ2) \
{ \
.opc1 = op1, \
.opc2 = op2, \
.opc3 = op3, \
- .pad = { 0, }, \
+ .opc4 = 0xff, \
.handler = { \
.inval1 = invl, \
.type = _typ, \
@@ -650,7 +668,7 @@ EXTRACT_HELPER(SP, 19, 2);
.opc1 = op1, \
.opc2 = op2, \
.opc3 = op3, \
- .pad = { 0, }, \
+ .opc4 = 0xff, \
.handler = { \
.inval1 = invl1, \
.inval2 = invl2, \
@@ -665,7 +683,7 @@ EXTRACT_HELPER(SP, 19, 2);
.opc1 = op1, \
.opc2 = op2, \
.opc3 = op3, \
- .pad = { 0, }, \
+ .opc4 = 0xff, \
.handler = { \
.inval1 = invl, \
.type = _typ, \
@@ -674,6 +692,20 @@ EXTRACT_HELPER(SP, 19, 2);
}, \
.oname = onam, \
}
+#define GEN_OPCODE3(name, op1, op2, op3, op4, invl, _typ, _typ2) \
+{ \
+ .opc1 = op1, \
+ .opc2 = op2, \
+ .opc3 = op3, \
+ .opc4 = op4, \
+ .handler = { \
+ .inval1 = invl, \
+ .type = _typ, \
+ .type2 = _typ2, \
+ .handler = &gen_##name, \
+ }, \
+ .oname = stringify(name), \
+}
#endif
/* SPR load/store helpers */
@@ -11842,9 +11874,10 @@ void gen_intermediate_code(CPUPPCState *env, struct TranslationBlock *tb)
} else {
ctx.opcode = cpu_ldl_code(env, ctx.nip);
}
- LOG_DISAS("translate opcode %08x (%02x %02x %02x) (%s)\n",
- ctx.opcode, opc1(ctx.opcode), opc2(ctx.opcode),
- opc3(ctx.opcode), ctx.le_mode ? "little" : "big");
+ LOG_DISAS("translate opcode %08x (%02x %02x %02x %02x) (%s)\n",
+ ctx.opcode, opc1(ctx.opcode), opc2(ctx.opcode),
+ opc3(ctx.opcode), opc4(ctx.opcode),
+ ctx.le_mode ? "little" : "big");
ctx.nip += 4;
table = env->opcodes;
handler = table[opc1(ctx.opcode)];
@@ -11854,14 +11887,20 @@ void gen_intermediate_code(CPUPPCState *env, struct TranslationBlock *tb)
if (is_indirect_opcode(handler)) {
table = ind_table(handler);
handler = table[opc3(ctx.opcode)];
+ if (is_indirect_opcode(handler)) {
+ table = ind_table(handler);
+ handler = table[opc4(ctx.opcode)];
+ }
}
}
/* Is opcode *REALLY* valid ? */
if (unlikely(handler->handler == &gen_invalid)) {
qemu_log_mask(LOG_GUEST_ERROR, "invalid/unsupported opcode: "
- "%02x - %02x - %02x (%08x) " TARGET_FMT_lx " %d\n",
+ "%02x - %02x - %02x - %02x (%08x) "
+ TARGET_FMT_lx " %d\n",
opc1(ctx.opcode), opc2(ctx.opcode),
- opc3(ctx.opcode), ctx.opcode, ctx.nip - 4, (int)msr_ir);
+ opc3(ctx.opcode), opc4(ctx.opcode),
+ ctx.opcode, ctx.nip - 4, (int)msr_ir);
} else {
uint32_t inval;
@@ -11873,9 +11912,10 @@ void gen_intermediate_code(CPUPPCState *env, struct TranslationBlock *tb)
if (unlikely((ctx.opcode & inval) != 0)) {
qemu_log_mask(LOG_GUEST_ERROR, "invalid bits: %08x for opcode: "
- "%02x - %02x - %02x (%08x) " TARGET_FMT_lx "\n",
- ctx.opcode & inval, opc1(ctx.opcode),
- opc2(ctx.opcode), opc3(ctx.opcode),
+ "%02x - %02x - %02x - %02x (%08x) "
+ TARGET_FMT_lx "\n", ctx.opcode & inval,
+ opc1(ctx.opcode), opc2(ctx.opcode),
+ opc3(ctx.opcode), opc4(ctx.opcode),
ctx.opcode, ctx.nip - 4);
gen_inval_exception(ctxp, POWERPC_EXCP_INVAL_INVAL);
break;
@@ -11902,9 +11942,9 @@ void gen_intermediate_code(CPUPPCState *env, struct TranslationBlock *tb)
break;
}
if (tcg_check_temp_count()) {
- fprintf(stderr, "Opcode %02x %02x %02x (%08x) leaked temporaries\n",
- opc1(ctx.opcode), opc2(ctx.opcode), opc3(ctx.opcode),
- ctx.opcode);
+ fprintf(stderr, "Opcode %02x %02x %02x %02x (%08x) leaked "
+ "temporaries\n", opc1(ctx.opcode), opc2(ctx.opcode),
+ opc3(ctx.opcode), opc4(ctx.opcode), ctx.opcode);
exit(1);
}
}
diff --git a/target-ppc/translate_init.c b/target-ppc/translate_init.c
index d207f68..52c3bcb 100644
--- a/target-ppc/translate_init.c
+++ b/target-ppc/translate_init.c
@@ -9252,13 +9252,47 @@ static int register_dblind_insn (opc_handler_t **ppc_opcodes,
return 0;
}
+static int register_trplind_insn(opc_handler_t **ppc_opcodes,
+ unsigned char idx1, unsigned char idx2,
+ unsigned char idx3, unsigned char idx4,
+ opc_handler_t *handler)
+{
+ opc_handler_t **table;
+
+ if (register_ind_in_table(ppc_opcodes, idx1, idx2, NULL) < 0) {
+ printf("*** ERROR: unable to join indirect table idx "
+ "[%02x-%02x]\n", idx1, idx2);
+ return -1;
+ }
+ table = ind_table(ppc_opcodes[idx1]);
+ if (register_ind_in_table(table, idx2, idx3, NULL) < 0) {
+ printf("*** ERROR: unable to join 2nd-level indirect table idx "
+ "[%02x-%02x-%02x]\n", idx1, idx2, idx3);
+ return -1;
+ }
+ table = ind_table(table[idx2]);
+ if (register_ind_in_table(table, idx3, idx4, handler) < 0) {
+ printf("*** ERROR: unable to insert opcode "
+ "[%02x-%02x-%02x-%02x]\n", idx1, idx2, idx3, idx4);
+ return -1;
+ }
+ return 0;
+}
static int register_insn (opc_handler_t **ppc_opcodes, opcode_t *insn)
{
if (insn->opc2 != 0xFF) {
if (insn->opc3 != 0xFF) {
- if (register_dblind_insn(ppc_opcodes, insn->opc1, insn->opc2,
- insn->opc3, &insn->handler) < 0)
- return -1;
+ if (insn->opc4 != 0xFF) {
+ if (register_trplind_insn(ppc_opcodes, insn->opc1, insn->opc2,
+ insn->opc3, insn->opc4,
+ &insn->handler) < 0) {
+ return -1;
+ }
+ } else {
+ if (register_dblind_insn(ppc_opcodes, insn->opc1, insn->opc2,
+ insn->opc3, &insn->handler) < 0)
+ return -1;
+ }
} else {
if (register_ind_insn(ppc_opcodes, insn->opc1,
insn->opc2, &insn->handler) < 0)
@@ -9334,7 +9368,7 @@ static void dump_ppc_insns (CPUPPCState *env)
{
opc_handler_t **table, *handler;
const char *p, *q;
- uint8_t opc1, opc2, opc3;
+ uint8_t opc1, opc2, opc3, opc4;
printf("Instructions set:\n");
/* opc1 is 6 bits long */
@@ -9354,34 +9388,50 @@ static void dump_ppc_insns (CPUPPCState *env)
for (opc3 = 0; opc3 < PPC_CPU_INDIRECT_OPCODES_LEN;
opc3++) {
handler = table[opc3];
- if (handler->handler != &gen_invalid) {
- /* Special hack to properly dump SPE insns */
- p = strchr(handler->oname, '_');
- if (p == NULL) {
- printf("INSN: %02x %02x %02x (%02d %04d) : "
- "%s\n",
- opc1, opc2, opc3, opc1,
- (opc3 << 5) | opc2,
- handler->oname);
- } else {
- q = "speundef";
- if ((p - handler->oname) != strlen(q) ||
- memcmp(handler->oname, q, strlen(q)) != 0) {
- /* First instruction */
- printf("INSN: %02x %02x %02x (%02d %04d) : "
- "%.*s\n",
- opc1, opc2 << 1, opc3, opc1,
- (opc3 << 6) | (opc2 << 1),
- (int)(p - handler->oname),
+ if (is_indirect_opcode(handler)) {
+ table = ind_table(handler);
+ /* opc4 is 5 bits long */
+ for (opc4 = 0; opc4 < PPC_CPU_INDIRECT_OPCODES_LEN;
+ opc4++) {
+ handler = table[opc4];
+ if (handler->handler != &gen_invalid) {
+ printf("INSN: %02x %02x %02x %02x -- "
+ "(%02d %04d %02d) : %s\n",
+ opc1, opc2, opc3, opc4,
+ opc1, (opc3 << 5) | opc2, opc4,
handler->oname);
}
- if (strcmp(p + 1, q) != 0) {
- /* Second instruction */
+ }
+ } else {
+ if (handler->handler != &gen_invalid) {
+ /* Special hack to properly dump SPE insns */
+ p = strchr(handler->oname, '_');
+ if (p == NULL) {
printf("INSN: %02x %02x %02x (%02d %04d) : "
"%s\n",
- opc1, (opc2 << 1) | 1, opc3, opc1,
- (opc3 << 6) | (opc2 << 1) | 1,
- p + 1);
+ opc1, opc2, opc3, opc1,
+ (opc3 << 5) | opc2,
+ handler->oname);
+ } else {
+ q = "speundef";
+ if ((p - handler->oname) != strlen(q) ||
+ memcmp(handler->oname, q, strlen(q)) != 0) {
+ /* First instruction */
+ printf("INSN: %02x %02x %02x"
+ "(%02d %04d) : %.*s\n",
+ opc1, opc2 << 1, opc3, opc1,
+ (opc3 << 6) | (opc2 << 1),
+ (int)(p - handler->oname),
+ handler->oname);
+ }
+ if (strcmp(p + 1, q) != 0) {
+ /* Second instruction */
+ printf("INSN: %02x %02x %02x "
+ "(%02d %04d) : %s\n", opc1,
+ (opc2 << 1) | 1, opc3, opc1,
+ (opc3 << 6) | (opc2 << 1) | 1,
+ p + 1);
+ }
}
}
}
@@ -9857,8 +9907,8 @@ static void ppc_cpu_unrealizefn(DeviceState *dev, Error **errp)
{
PowerPCCPU *cpu = POWERPC_CPU(dev);
CPUPPCState *env = &cpu->env;
- opc_handler_t **table;
- int i, j;
+ opc_handler_t **table, **table_2;
+ int i, j, k;
cpu_exec_exit(CPU(dev));
@@ -9869,10 +9919,20 @@ static void ppc_cpu_unrealizefn(DeviceState *dev, Error **errp)
if (is_indirect_opcode(env->opcodes[i])) {
table = ind_table(env->opcodes[i]);
for (j = 0; j < PPC_CPU_INDIRECT_OPCODES_LEN; j++) {
- if (table[j] != &invalid_handler &&
- is_indirect_opcode(table[j])) {
+ if (table[j] == &invalid_handler) {
+ continue;
+ }
+ if (is_indirect_opcode(table[j])) {
+ table_2 = ind_table(table[j]);
+ for (k = 0; k < PPC_CPU_INDIRECT_OPCODES_LEN; k++) {
+ if (table_2[k] != &invalid_handler &&
+ is_indirect_opcode(table_2[k])) {
+ g_free((opc_handler_t *)((uintptr_t)table_2[k] &
+ ~PPC_INDIRECT));
+ }
+ }
g_free((opc_handler_t *)((uintptr_t)table[j] &
- ~PPC_INDIRECT));
+ ~PPC_INDIRECT));
}
}
g_free((opc_handler_t *)((uintptr_t)env->opcodes[i] &
--
2.7.4
^ permalink raw reply related [flat|nested] 29+ messages in thread