Add support for Hygon Dhyana Family 18h processor
Pu Wen
maintainerv6 cover Add support for Hygon Dhyana Family 18h processor
As a new x86 CPU Vendor, Chengdu Haiguang IC Design Co., Ltd (Hygon) is a Joint Venture between AMD and Haiguang Information Technology Co., Ltd., and aims at providing high performance x86 processor for China server market.
The first generation Hygon's processor(Dhyana) originates from AMD technology and shares most of the architecture with AMD's family 17h, but with different CPU Vendor ID("HygonGenuine")/PCIE Device Vendor ID (0x1D94)/Family series number (Family 18h).
To enable the support of Linux kernel to Hygon's CPU, we added a new vendor type (X86_VENDOR_HYGON, with value of 9) in arch/x86/include/ asm/processor.h, and shared most of kernel support codes with AMD family 17h.
As Hygon will negotiate with AMD to make sure that only Hygon will use family 18h, so try to minimize code modification and share most codes with AMD under this consideration.
This patch series have been applied and tested successfully on Hygon Dhyana SoC silicon. Also tested on AMD EPYC (Family 17h) processor, it works fine and makes no harm to the existing codes.
v5->v6:
- Rebased on 4.19-rc3 and tested against it.
- Add Reviewed-by from Borislav Petkov for cacheinfo, smpboot, alternative and kvm.
- Rework the patch subjects and patch descriptions.
- Rework vendor checking for some patches to minimize the code modification.
v4->v5:
- Rebased on 4.19-rc1 and tested against it.
- Add Reviewed-by from Boris Ostrovsky for Xen.
- Rework EDAC patch without vendor checking for minimal modification.
v3->v4:
- Rebased on 4.18.3 and tested against it.
- Merge patchs 05/17 perfctr and 10/17 events in v3 to patch 05/16 PMU for better patch function group.
- Add hygon_get_topology_early() in patch 01/16.
- Rework vendor checking and refine coding style.
- Add Acked-by from Bjorn Helgaas for pci.
- Add Acked-by from Rafael J. Wysocki for cpufreq and acpi.
v2->v3:
- Rebased on 4.18-rc8 and tested against it.
- Rework vendor checking codes to improve consistency.
v1->v2:
- Rebased on 4.18-rc6 and tested against it.
- Split the patchset to small series of patches.
- Rework patch descriptions.
- Create a separated arch/x86/kernel/cpu/hygon.c for Dhyana CPU initialization to reduce long-term maintenance effort.
Pu Wen (16):
x86/cpu: Create Hygon Dhyana architecture support file x86/cpu: Get cache info and setup cache cpumap for Hygon Dhyana x86/cpu/mtrr: Support TOP_MEM2 and get MTRR number x86/smpboot: SMP init nodelay and not flush caches before sleep perf/x86: Add Hygon Dhyana support to PMU infrastructure x86/alternative: Init ideal_nops for Hygon Dhyana x86/pci: Add Hygon Dhyana support to PCI and north bridge x86/apic: Add Hygon Dhyana support to APIC
x86/bugs: Add mitigation to spectre and no meltdown for Hygon Dhyana x86/mce: Add Hygon Dhyana support to MCE infrastructure x86/kvm: Add Hygon Dhyana support to KVM infrastructure x86/xen: Add Hygon Dhyana support to Xen
ACPI, x86: Add Hygon Dhyana support
cpufreq, x86: Add Hygon Dhyana support
EDAC, amd64: Add Hygon Dhyana support
cpupower, x86: Add Hygon Dhyana support
MAINTAINERS | 6 + arch/x86/Kconfig.cpu | 14 + arch/x86/events/amd/core.c | 4 + arch/x86/events/amd/uncore.c | 20 +- arch/x86/events/core.c | 4 + arch/x86/include/asm/cacheinfo.h | 1 + arch/x86/include/asm/kvm_emulate.h | 4 + arch/x86/include/asm/mce.h | 2 + arch/x86/include/asm/processor.h | 3 +- arch/x86/include/asm/virtext.h | 5 +- arch/x86/kernel/alternative.c | 4 + arch/x86/kernel/amd_nb.c | 47 ++- arch/x86/kernel/apic/apic.c | 13 +- arch/x86/kernel/apic/probe_32.c | 1 + arch/x86/kernel/cpu/Makefile | 1 + arch/x86/kernel/cpu/bugs.c | 6 +- arch/x86/kernel/cpu/cacheinfo.c | 31 +- arch/x86/kernel/cpu/common.c | 1 + arch/x86/kernel/cpu/cpu.h | 1 + arch/x86/kernel/cpu/hygon.c | 411 +++++++++++++++++++++ arch/x86/kernel/cpu/mcheck/mce-severity.c | 3 +- arch/x86/kernel/cpu/mcheck/mce.c | 20 +- arch/x86/kernel/cpu/mtrr/cleanup.c | 3 +- arch/x86/kernel/cpu/mtrr/mtrr.c | 2 +- arch/x86/kernel/cpu/perfctr-watchdog.c | 2 + arch/x86/kernel/smpboot.c | 4 +- arch/x86/kvm/emulate.c | 11 +- arch/x86/pci/amd_bus.c | 6 +- arch/x86/xen/pmu.c | 12 +- drivers/acpi/acpi_pad.c | 1 + drivers/acpi/processor_idle.c | 1 + drivers/cpufreq/acpi-cpufreq.c | 5 + drivers/cpufreq/amd_freq_sensitivity.c | 9 +- drivers/edac/amd64_edac.c | 8 +- drivers/edac/mce_amd.c | 4 +- include/linux/pci_ids.h | 2 + tools/power/cpupower/utils/cpufreq-info.c | 6 +- tools/power/cpupower/utils/helpers/amd.c | 4 +- tools/power/cpupower/utils/helpers/cpuid.c | 8 +- tools/power/cpupower/utils/helpers/helpers.h | 2 +- tools/power/cpupower/utils/helpers/misc.c | 2 +- .../cpupower/utils/idle_monitor/mperf_monitor.c | 3 +- 42 files changed, 646 insertions(+), 51 deletions(-) create mode 100644 arch/x86/kernel/cpu/hygon.c
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 1/16 x86/cpu: Create Hygon Dhyana architecture support file
Add x86 architecture support for new processor Hygon Dhyana Family 18h. Rework to create a separated file(arch/x86/kernel/cpu/hygon.c) from the AMD one(arch/x86/kernel/cpu/amd.c) to initialize Dhyana. In this way we can remove old AMD architecture support codes and generate a clear initialization flow for Hygon Dhyana CPU. It also reduce the long-term maintenance effort. Add Maintainer information for hygon.c in accordance.
To identify Hygon Dhyana CPU, add a new vendor type X86_VENDOR_HYGON(9) for system recognition.
To enable the config for Hygon Dhyana CPU, add a separated Kconfig entry(CPU_SUP_HYGON) in kernel config setup. And select CPU_SUP_AMD to solve the dependency issues in kernel.
--- MAINTAINERS | 6 + arch/x86/Kconfig.cpu | 14 ++ arch/x86/include/asm/processor.h | 3 +- arch/x86/kernel/cpu/Makefile | 1 + arch/x86/kernel/cpu/hygon.c | 408 +++++++++++++++++++++++++++++++++++++++ 5 files changed, 431 insertions(+), 1 deletion(-) create mode 100644 arch/x86/kernel/cpu/hygon.c
MAINTAINERS +6 -0
@@ -6760,6 +6760,12 @@ S: Maintained F: mm/memory-failure.c F: mm/hwpoison-inject.c +HYGON PROCESSOR SUPPORT +M: Pu Wen <[email protected]> +L: [email protected] +S: Maintained +F: arch/x86/kernel/cpu/hygon.c + Hyper-V CORE AND DRIVERS M: "K. Y. Srinivasan" <[email protected]> M: Haiyang Zhang <[email protected]>
arch/x86/Kconfig.cpu +14 -0
@@ -426,6 +426,20 @@ config CPU_SUP_AMD If unsure, say N. +config CPU_SUP_HYGON + default y + bool "Support Hygon processors" if PROCESSOR_SELECT + select CPU_SUP_AMD + help + This enables detection, tunings and quirks for Hygon processors + + You need this enabled if you want your kernel to run on an + Hygon CPU. Disabling this option on other types of CPUs + makes the kernel a tiny bit smaller. Disabling it on an Hygon + CPU might render the kernel unbootable. + + If unsure, say N. + config CPU_SUP_CENTAUR default y bool "Support Centaur processors" if PROCESSOR_SELECT
arch/x86/include/asm/processor.h +2 -1
@@ -155,7 +155,8 @@ enum cpuid_regs_idx {#define X86_VENDOR_CENTAUR 5#define X86_VENDOR_TRANSMETA 7#define X86_VENDOR_NSC 8 -#define X86_VENDOR_NUM 9 +#define X86_VENDOR_HYGON 9 +#define X86_VENDOR_NUM 10#define X86_VENDOR_UNKNOWN 0xff
arch/x86/kernel/cpu/Makefile +1 -0
@@ -30,6 +30,7 @@ obj-$(CONFIG_X86_FEATURE_NAMES) += capflags.o powerflags.o obj-$(CONFIG_CPU_SUP_INTEL) += intel.o intel_pconfig.o obj-$(CONFIG_CPU_SUP_AMD) += amd.o +obj-$(CONFIG_CPU_SUP_HYGON) += hygon.o obj-$(CONFIG_CPU_SUP_CYRIX_32) += cyrix.o obj-$(CONFIG_CPU_SUP_CENTAUR) += centaur.o obj-$(CONFIG_CPU_SUP_TRANSMETA_32) += transmeta.o
arch/x86/kernel/cpu/hygon.c new file +408 -0
408 lines not on this page · raw message
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 2/16 x86/cpu: Get cache info and setup cache cpumap for Hygon Dhyana
In reply to Pu WenThe Hygon Dhyana CPU has topology extensions bit in CPUID. With this bit, the kernel can get the cache information. So add support in cpuid4_cache_lookup_regs() to get the correct cache size.
The Hygon Dhyana CPU also discovers num_cache_leaves via CPUID leaf 0x8000001d, so add support to it in find_num_cache_leaves().
Also add cacheinfo_hygon_init_llc_id() and init_hygon_cacheinfo() functions to initialize Dhyana cache info. Setup cache cpumap in the same way as AMD does.
--- arch/x86/include/asm/cacheinfo.h | 1 + arch/x86/kernel/cpu/cacheinfo.c | 31 +++++++++++++++++++++++++++++-- arch/x86/kernel/cpu/cpu.h | 1 + arch/x86/kernel/cpu/hygon.c | 3 +++ 4 files changed, 34 insertions(+), 2 deletions(-)
arch/x86/include/asm/cacheinfo.h +1 -0
@@ -3,5 +3,6 @@#define _ASM_X86_CACHEINFO_Hvoid cacheinfo_amd_init_llc_id (struct cpuinfo_x86 *c,int cpu,u8 node_id); +void cacheinfo_hygon_init_llc_id (struct cpuinfo_x86 *c,int cpu,u8 node_id);#endif /* _ASM_X86_CACHEINFO_H */
arch/x86/kernel/cpu/cacheinfo.c +29 -2
@@ -602,6 +602,10 @@ cpuid4_cache_lookup_regs(int index, struct _cpuid4_info_regs *this_leaf)else amd_cpuid4 (index, &eax, &ebx, &ecx);amd_init_l3_cache (this_leaf, index); + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { +cpuid_count (0x8000001d , index, &eax.full, + &ebx.full, &ecx.full, &edx); +amd_init_l3_cache (this_leaf, index); }else {cpuid_count (4 , index, &eax.full, &ebx.full, &ecx.full, &edx); }
@@ -625,7 +629,8 @@ static int find_num_cache_leaves(struct cpuinfo_x86 *c)union _cpuid4_leaf_eax cache_eax;int i =-1 ; -if (c->x86_vendor ==X86_VENDOR_AMD ) +if (c->x86_vendor ==X86_VENDOR_AMD || + c->x86_vendor ==X86_VENDOR_HYGON ) op =0x8000001d ;else op =4 ;
@@ -678,6 +683,22 @@ void cacheinfo_amd_init_llc_id(struct cpuinfo_x86 *c, int cpu, u8 node_id) } } +void cacheinfo_hygon_init_llc_id (struct cpuinfo_x86 *c,int cpu,u8 node_id) +{ +/* +* We may have multiple LLCs if L3 caches exist, so check if we +* have an L3 cache by looking at the L3 cache CPUID leaf. +*/ +if (!cpuid_edx (0x80000006 )) +return ; + +/* +* LLC is at the core complex level. +* Core complex ID is ApicId[3] for these processors. +*/ +per_cpu (cpu_llc_id, cpu) = c->apicid >>3 ; +} +void init_amd_cacheinfo (struct cpuinfo_x86 *c) {
@@ -691,6 +712,11 @@ void init_amd_cacheinfo(struct cpuinfo_x86 *c) } } +void init_hygon_cacheinfo (struct cpuinfo_x86 *c) +{ + num_cache_leaves =find_num_cache_leaves (c); +} +void init_intel_cacheinfo (struct cpuinfo_x86 *c) {/* Cache sizes */
@@ -913,7 +939,8 @@ static void __cache_cpumap_setup(unsigned int cpu, int index,int index_msb, i;struct cpuinfo_x86 *c = &cpu_data (cpu); -if (c->x86_vendor ==X86_VENDOR_AMD ) { +if (c->x86_vendor ==X86_VENDOR_AMD || + c->x86_vendor ==X86_VENDOR_HYGON ) {if (__cache_amd_cpumap_setup (cpu, index, base))return ; }
arch/x86/kernel/cpu/cpu.h +1 -0
@@ -54,6 +54,7 @@ extern u32 get_scattered_cpuid_leaf(unsigned int level, enum cpuid_regs_idx reg);extern void init_intel_cacheinfo (struct cpuinfo_x86 *c);extern void init_amd_cacheinfo (struct cpuinfo_x86 *c); +extern void init_hygon_cacheinfo (struct cpuinfo_x86 *c);extern void detect_num_cpu_cores (struct cpuinfo_x86 *c);extern int detect_extended_topology_early (struct cpuinfo_x86 *c);
arch/x86/kernel/cpu/hygon.c +3 -0
@@ -90,6 +90,7 @@ static void hygon_get_topology(struct cpuinfo_x86 *c)if (!err) c->x86_coreid_bits =get_count_order (c->x86_max_cores); +cacheinfo_hygon_init_llc_id (c, cpu, node_id); }else if (cpu_has (c,X86_FEATURE_NODEID_MSR )) {u64 value;
@@ -324,6 +325,8 @@ static void init_hygon(struct cpuinfo_x86 *c)hygon_get_topology (c);srat_detect_node (c); +init_hygon_cacheinfo (c); +if (cpu_has (c,X86_FEATURE_XMM2 )) {unsigned long long val;int ret;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 3/16 x86/cpu/mtrr: Support TOP_MEM2 and get MTRR number
In reply to Pu WenThe Hygon Dhyana CPU have a special magic MSR way to force WB for memory >4GB, and support TOP_MEM2. Therefore, it is necessary to add Hygon Dhyana support in amd_special_default_mtrr().
The number of variable MTRRs for Hygon is 2 as AMD's.
--- arch/x86/kernel/cpu/mtrr/cleanup.c | 3 ++- arch/x86/kernel/cpu/mtrr/mtrr.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-)
arch/x86/kernel/cpu/mtrr/cleanup.c +2 -1
@@ -831,7 +831,8 @@ int __init amd_special_default_mtrr(void) {u32 l, h; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return 0 ;if (boot_cpu_data.x86 <0xf )return 0 ;
arch/x86/kernel/cpu/mtrr/mtrr.c +1 -1
@@ -127,7 +127,7 @@ static void __init set_num_var_ranges(void)if (use_intel ())rdmsr (MSR_MTRRcap, config, dummy); -else if (is_cpu (AMD )) +else if (is_cpu (AMD ) ||is_cpu (HYGON )) config =2 ;else if (is_cpu (CYRIX ) ||is_cpu (CENTAUR )) config =8 ;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 4/16 x86/smpboot: SMP init nodelay and not flush caches before sleep
In reply to Pu WenThe Hygon Dhyana CPU use no delay in smp_quirk_init_udelay(), and return in mwait_play_dead() as AMD does.
--- arch/x86/kernel/smpboot.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
arch/x86/kernel/smpboot.c +3 -1
@@ -676,6 +676,7 @@ static void __init smp_quirk_init_udelay(void)/* if modern processor, use no delay */ if (((boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) && (boot_cpu_data.x86 ==6 )) || + ((boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) && (boot_cpu_data.x86 >=0x18 )) || ((boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) && (boot_cpu_data.x86 >=0xF ))) { init_udelay =0 ;return ;
@@ -1592,7 +1593,8 @@ static inline void mwait_play_dead(void)void *mwait_ptr;int i; -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON )return ;if (!this_cpu_has (X86_FEATURE_MWAIT ))return ;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 5/16 perf/x86: Add Hygon Dhyana support to PMU infrastructure
In reply to Pu WenThe PMU architecture for Hygon Dhyana CPU is similar to the AMD Family 17h one. To support Hygon Dhyana PMU, call amd_pmu_init() to share AMD PMU initialization flow, and change the PMU name to "HYGON".
The Hygon Dhyana CPU support both legacy and extension PMC MSRs(perf counter registers and event selection registers), so add Hygon Dhyana support to get bit offset in the similar way as AMD does.
--- arch/x86/events/amd/core.c | 4 ++++ arch/x86/events/amd/uncore.c | 20 +++++++++++++------- arch/x86/events/core.c | 4 ++++ arch/x86/kernel/cpu/perfctr-watchdog.c | 2 ++ 4 files changed, 23 insertions(+), 7 deletions(-)
arch/x86/events/amd/core.c +4 -0
@@ -669,6 +669,10 @@ static int __init amd_core_pmu_init(void)* We fallback to using default amd_get_event_constraints. */ break ; +case 0x18 : +pr_cont ("Fam18h " ); +/* Using default amd_get_event_constraints. */ +break ;default :pr_err ("core perfctr but no constraints; unknown hardware!\n" );return -ENODEV ;
arch/x86/events/amd/uncore.c +13 -7
@@ -507,17 +507,19 @@ static int __init amd_uncore_init(void) {int ret = -ENODEV ; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return -ENODEV ;if (!boot_cpu_has (X86_FEATURE_TOPOEXT ))return -ENODEV ; -if (boot_cpu_data.x86 ==0x17 ) { +if (boot_cpu_data.x86 ==0x17 || boot_cpu_data.x86 ==0x18 ) {/* -* For F17h, the Northbridge counters are repurposed as Data -* Fabric counters. Also, L3 counters are supported too. The PMUs -* are exported based on family as either L2 or L3 and NB or DF. +* For F17h or F18h, the Northbridge counters are +* repurposed as Data Fabric counters. Also, L3 +* counters are supported too. The PMUs are exported +* based on family as either L2 or L3 and NB or DF. */ num_counters_nb =NUM_COUNTERS_NB ; num_counters_llc =NUM_COUNTERS_L3 ;
@@ -547,7 +549,9 @@ static int __init amd_uncore_init(void)if (ret) goto fail_nb; -pr_info ("AMD NB counters detected\n" ); +pr_info ("%s NB counters detected\n" , + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ? +"HYGON" :"AMD" ); ret =0 ; }
@@ -561,7 +565,9 @@ static int __init amd_uncore_init(void)if (ret) goto fail_llc; -pr_info ("AMD LLC counters detected\n" ); +pr_info ("%s LLC counters detected\n" , + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ? +"HYGON" :"AMD" ); ret =0 ; }
arch/x86/events/core.c +4 -0
@@ -1776,6 +1776,10 @@ static int __init init_hw_perf_events(void)case X86_VENDOR_AMD : err =amd_pmu_init ();break ; +case X86_VENDOR_HYGON : + err =amd_pmu_init (); + x86_pmu.name ="HYGON" ; +break ;default : err = -ENOTSUPP ; }
arch/x86/kernel/cpu/perfctr-watchdog.c +2 -0
@@ -46,6 +46,7 @@ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr) {/* returns the bit offset of the performance counter register */ switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :if (msr >=MSR_F15H_PERF_CTR )return (msr -MSR_F15H_PERF_CTR ) >>1 ;
@@ -74,6 +75,7 @@ static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr) {/* returns the bit offset of the event selection register */ switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :if (msr >=MSR_F15H_PERF_CTL )return (msr -MSR_F15H_PERF_CTL ) >>1 ;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 6/16 x86/alternative: Init ideal_nops for Hygon Dhyana
In reply to Pu WenThe ideal_nops for Hygon Dhyana CPU should be p6_nops.
--- arch/x86/kernel/alternative.c | 4 ++++ 1 file changed, 4 insertions(+)
arch/x86/kernel/alternative.c +4 -0
@@ -222,6 +222,10 @@ void __init arch_init_ideal_nops(void) }break ; +case X86_VENDOR_HYGON : + ideal_nops = p6_nops; +return ; +case X86_VENDOR_AMD :if (boot_cpu_data.x86 >0xf ) { ideal_nops = p6_nops;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 7/16 x86/pci: Add Hygon Dhyana support to PCI and north bridge
In reply to Pu WenAs Hygon registered its PCI Vendor ID as a new one 0x1d94, and there are PCI Devices 0x1450/0x1463/0x1464 for Host bridge on Hygon Dhyana platforms, so add Hygon Dhyana support to the PCI and north bridge subsystem by using the code path of AMD family 17h.
--- arch/x86/kernel/amd_nb.c | 47 +++++++++++++++++++++++++++++++++++++++++------ arch/x86/pci/amd_bus.c | 6 ++++-- include/linux/pci_ids.h | 2 ++ 3 files changed, 47 insertions(+), 8 deletions(-)
arch/x86/kernel/amd_nb.c +41 -6
@@ -61,6 +61,21 @@ static const struct pci_device_id amd_nb_link_ids[] = { {} }; +static const struct pci_device_id hygon_root_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_ROOT ) }, + {} +}; + +const struct pci_device_id hygon_nb_misc_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_DF_F3 ) }, + {} +}; + +static const struct pci_device_id hygon_nb_link_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_DF_F4 ) }, + {} +}; +const struct amd_nb_bus_dev_range amd_nb_bus_dev_ranges[]__initconst = { {0x00 ,0x18 ,0x20 }, {0xff ,0x00 ,0x20 },
@@ -197,12 +212,25 @@ int amd_cache_northbridges(void)u16 i =0 ;struct amd_northbridge *nb;struct pci_dev *root, *misc, *link; +const struct pci_device_id *root_ids =NULL ; +const struct pci_device_id *misc_ids =NULL ; +const struct pci_device_id *link_ids =NULL ; + +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { + root_ids = amd_root_ids; + misc_ids = amd_nb_misc_ids; + link_ids = amd_nb_link_ids; + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + root_ids = hygon_root_ids; + misc_ids = hygon_nb_misc_ids; + link_ids = hygon_nb_link_ids; + }if (amd_northbridges.num)return 0 ; misc =NULL ; -while ((misc =next_northbridge (misc, amd_nb_misc_ids)) !=NULL ) +while ((misc =next_northbridge (misc, misc_ids)) !=NULL ) i++;if (!i)
@@ -218,11 +246,11 @@ int amd_cache_northbridges(void) link = misc = root =NULL ;for (i =0 ; i != amd_northbridges.num; i++) {node_to_amd_nb (i)->root = root = -next_northbridge (root, amd_root_ids); +next_northbridge (root, root_ids);node_to_amd_nb (i)->misc = misc = -next_northbridge (misc, amd_nb_misc_ids); +next_northbridge (misc, misc_ids);node_to_amd_nb (i)->link = link = -next_northbridge (link, amd_nb_link_ids); +next_northbridge (link, link_ids); }if (amd_gart_present ())
@@ -263,9 +291,15 @@ bool __init early_is_amd_nb(u32 device) {const struct pci_device_id *id;u32 vendor = device &0xffff ; +const struct pci_device_id *misc_ids =NULL ; + +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) + misc_ids = amd_nb_misc_ids; +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + misc_ids = hygon_nb_misc_ids; device >>=16 ; -for (id = amd_nb_misc_ids; id->vendor; id++) +for (id = misc_ids; id->vendor; id++)if (vendor == id->vendor && device == id->device)return true;return false;
@@ -277,7 +311,8 @@ struct resource *amd_get_mmconfig_range(struct resource *res)u64 base, msr;unsigned int segn_busn_bits; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return NULL ;/* assume all cpus from fam10h have mmconfig */
arch/x86/pci/amd_bus.c +4 -2
@@ -93,7 +93,8 @@ static int __init early_root_info_init(void) vendor = id &0xffff ; device = (id>>16 ) &0xffff ; -if (vendor !=PCI_VENDOR_ID_AMD ) +if (vendor !=PCI_VENDOR_ID_AMD && + vendor !=PCI_VENDOR_ID_HYGON )continue ;if (hb_probes[i].device == device) {
@@ -390,7 +391,8 @@ static int __init pci_io_ecs_init(void)static int __init amd_postcore_init (void ) { -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return 0 ;early_root_info_init ();
include/linux/pci_ids.h +2 -0
@@ -2561,6 +2561,8 @@#define PCI_VENDOR_ID_AMAZON 0x1d0f +#define PCI_VENDOR_ID_HYGON 0x1d94 +#define PCI_VENDOR_ID_TEKRAM 0x1de1#define PCI_DEVICE_ID_TEKRAM_DC290 0xdc29
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 8/16 x86/apic: Add Hygon Dhyana support to APIC
In reply to Pu WenAdd Hygon Dhyana support to the APIC subsystem as it use modern APIC. When running on 32 bit mode, bigsmp should be enabled if there are more than 8 cores online.
--- arch/x86/kernel/apic/apic.c | 13 +++++++++++-- arch/x86/kernel/apic/probe_32.c | 1 + 2 files changed, 12 insertions(+), 2 deletions(-)
arch/x86/kernel/apic/apic.c +11 -2
@@ -224,6 +224,11 @@ static int modern_apic(void)if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD && boot_cpu_data.x86 >=0xf )return 1 ; + +/* Hygon systems use modern APIC */ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) +return 1 ; +return lapic_get_version () >=0x14 ; }
@@ -1211,9 +1216,11 @@ void __init sync_Arb_IDs(void) {/* * Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 And not -* needed on AMD. +* needed on AMD or Hygon. */ -if (modern_apic () || boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (modern_apic () || + boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON )return ;/*
@@ -1912,6 +1919,8 @@ static int __init detect_init_APIC(void) (boot_cpu_data.x86 >= 15))break ; goto no_apic; +case X86_VENDOR_HYGON : +break ;case X86_VENDOR_INTEL :if (boot_cpu_data.x86 ==6 || boot_cpu_data.x86 ==15 || (boot_cpu_data.x86 ==5 &&boot_cpu_has (X86_FEATURE_APIC )))
arch/x86/kernel/apic/probe_32.c +1 -0
@@ -185,6 +185,7 @@ void __init default_setup_apic_routing(void)break ; }/* If P4 and above fall through */ +case X86_VENDOR_HYGON :case X86_VENDOR_AMD : def_to_bigsmp =1 ; }
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 9/16 x86/bugs: Add mitigation to spectre and no meltdown for Hygon Dhyana
In reply to Pu WenThe Hygon Dhyana CPU has the same speculative execution as AMD family 17h, so share AMD spectre mitigation code for Hygon Dhyana.
Also Hygon Dhyana is not affected by meltdown vulnerability as AMD, so add the exception for Hygon Dhyana.
--- arch/x86/kernel/cpu/bugs.c | 6 ++++-- arch/x86/kernel/cpu/common.c | 1 + 2 files changed, 5 insertions(+), 2 deletions(-)
arch/x86/kernel/cpu/bugs.c +4 -2
@@ -312,8 +312,9 @@ static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void) }if (cmd ==SPECTRE_V2_CMD_RETPOLINE_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON && boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) { -pr_err ("retpoline,amd selected but CPU is not AMD. Switching to AUTO select\n" ); +pr_err ("retpoline,amd selected but CPU is not AMD or Hygon. Switching to AUTO select\n" );return SPECTRE_V2_CMD_AUTO ; }
@@ -371,7 +372,8 @@ static void __init spectre_v2_select_mitigation(void)return ; retpoline_auto: -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { retpoline_amd:if (!boot_cpu_has (X86_FEATURE_LFENCE_RDTSC )) {pr_err ("Spectre mitigation: LFENCE not serializing, switching to generic retpoline\n" );
arch/x86/kernel/cpu/common.c +1 -0
@@ -963,6 +963,7 @@ static const __initconst struct x86_cpu_id cpu_no_speculation[] = {static const __initconst struct x86_cpu_id cpu_no_meltdown[] = { {X86_VENDOR_AMD }, + {X86_VENDOR_HYGON }, {} };
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 10/16 x86/mce: Add Hygon Dhyana support to MCE infrastructure
In reply to Pu WenThe machine check architecture for Hygon Dhyana CPU is similar to the AMD family 17h one. To make MCE working on Hygon platforms, add vendor checking for Hygon Dhyana to share the code path of AMD family 17h.
--- arch/x86/include/asm/mce.h | 2 ++ arch/x86/kernel/cpu/mcheck/mce-severity.c | 3 ++- arch/x86/kernel/cpu/mcheck/mce.c | 20 +++++++++++++++----- 3 files changed, 19 insertions(+), 6 deletions(-)
arch/x86/include/asm/mce.h +2 -0
@@ -214,6 +214,8 @@ static inline void mce_amd_feature_init(struct cpuinfo_x86 *c) { }static inline int umc_normaddr_to_sysaddr (u64 norm_addr,u16 nid,u8 umc,u64 *sys_addr) {return -EINVAL ; };#endif +static inline void mce_hygon_feature_init (struct cpuinfo_x86 *c) {return mce_amd_feature_init (c); } +int mce_available (struct cpuinfo_x86 *c);bool mce_is_memory_error (struct mce *m);
arch/x86/kernel/cpu/mcheck/mce-severity.c +2 -1
@@ -336,7 +336,8 @@ int (*mce_severity)(struct mce *m, int tolerant, char **msg, bool is_excp) =void __init mcheck_vendor_init_severity (void ) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) mce_severity = mce_severity_amd; }
arch/x86/kernel/cpu/mcheck/mce.c +15 -5
@@ -270,7 +270,7 @@ static void print_mce(struct mce *m) {__print_mce (m); -if (m->cpuvendor !=X86_VENDOR_AMD ) +if (m->cpuvendor !=X86_VENDOR_AMD && m->cpuvendor !=X86_VENDOR_HYGON )pr_emerg_ratelimited (HW_ERR "Run the above through 'mcelog --ascii'\n" ); }
@@ -508,9 +508,9 @@ static int mce_usable_address(struct mce *m)bool mce_is_memory_error (struct mce *m) { -if (m->cpuvendor ==X86_VENDOR_AMD ) { +if (m->cpuvendor ==X86_VENDOR_AMD || + m->cpuvendor ==X86_VENDOR_HYGON ) {return amd_mce_is_memory_error (m); - }else if (m->cpuvendor ==X86_VENDOR_INTEL ) {/* * Intel SDM Volume 3B - 15.9.2 Compound Error Codes
@@ -539,6 +539,9 @@ static bool mce_is_correctable(struct mce *m)if (m->cpuvendor ==X86_VENDOR_AMD && m->status &MCI_STATUS_DEFERRED )return false; +if (m->cpuvendor ==X86_VENDOR_HYGON && m->status &MCI_STATUS_DEFERRED ) +return false; +if (m->status &MCI_STATUS_UC )return false;
@@ -1705,7 +1708,7 @@ static int __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c)*/ static void __mcheck_cpu_init_early (struct cpuinfo_x86 *c) { -if (c->x86_vendor ==X86_VENDOR_AMD ) { +if (c->x86_vendor ==X86_VENDOR_AMD || c->x86_vendor ==X86_VENDOR_HYGON ) { mce_flags.overflow_recov = !!cpu_has (c,X86_FEATURE_OVERFLOW_RECOV ); mce_flags.succor = !!cpu_has (c,X86_FEATURE_SUCCOR ); mce_flags.smca = !!cpu_has (c,X86_FEATURE_SMCA );
@@ -1746,6 +1749,11 @@ static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)mce_amd_feature_init (c);break ; } + +case X86_VENDOR_HYGON : +mce_hygon_feature_init (c); +break ; +case X86_VENDOR_CENTAUR :mce_centaur_feature_init (c);break ;
@@ -1971,12 +1979,14 @@ static void mce_disable_error_reporting(void)static void vendor_disable_error_reporting (void ) {/* -* Don't clear on Intel or AMD CPUs. Some of these MSRs are socket-wide. +* Don't clear on Intel or AMD or Hygon CPUs. Some of these MSRs +* are socket-wide. * Disabling them for just a single offlined CPU is bad, since it will * inhibit reporting for all shared resources on the socket like the * last level cache (LLC), the integrated memory controller (iMC), etc. */ if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON || boot_cpu_data.x86_vendor ==X86_VENDOR_AMD )return ;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 11/16 x86/kvm: Add Hygon Dhyana support to KVM infrastructure
In reply to Pu WenThe Hygon Dhyana CPU has the SVM feature as AMD family 17h does. So enable the KVM infrastructure support to it.
--- arch/x86/include/asm/kvm_emulate.h | 4 ++++ arch/x86/include/asm/virtext.h | 5 +++-- arch/x86/kvm/emulate.c | 11 ++++++++++- 3 files changed, 17 insertions(+), 3 deletions(-)
arch/x86/include/asm/kvm_emulate.h +4 -0
@@ -364,6 +364,10 @@ struct x86_emulate_ctxt {#define X86EMUL_CPUID_VENDOR_AMDisbetterI_ecx 0x21726574#define X86EMUL_CPUID_VENDOR_AMDisbetterI_edx 0x74656273 +#define X86EMUL_CPUID_VENDOR_HygonGenuine_ebx 0x6f677948 +#define X86EMUL_CPUID_VENDOR_HygonGenuine_ecx 0x656e6975 +#define X86EMUL_CPUID_VENDOR_HygonGenuine_edx 0x6e65476e +#define X86EMUL_CPUID_VENDOR_GenuineIntel_ebx 0x756e6547#define X86EMUL_CPUID_VENDOR_GenuineIntel_ecx 0x6c65746e#define X86EMUL_CPUID_VENDOR_GenuineIntel_edx 0x49656e69
arch/x86/include/asm/virtext.h +3 -2
@@ -83,9 +83,10 @@ static inline void cpu_emergency_vmxoff(void)*/ static inline int cpu_has_svm (const char **msg) { -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON ) {if (msg) - *msg ="not amd" ; + *msg ="not amd or hygon" ;return 0 ; }
arch/x86/kvm/emulate.c +10 -1
@@ -2711,7 +2711,16 @@ static bool em_syscall_is_enabled(struct x86_emulate_ctxt *ctxt) edx == X86EMUL_CPUID_VENDOR_AMDisbetterI_edx)return true; -/* default: (not Intel, not AMD), apply Intel's stricter rules... */ +/* Hygon ("HygonGenuine") */ +if (ebx == X86EMUL_CPUID_VENDOR_HygonGenuine_ebx && + ecx == X86EMUL_CPUID_VENDOR_HygonGenuine_ecx && + edx == X86EMUL_CPUID_VENDOR_HygonGenuine_edx) +return true; + +/* +* default: (not Intel, not AMD, not Hygon), apply Intel's +* stricter rules... +*/ return false; }
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 12/16 x86/xen: Add Hygon Dhyana support to Xen
In reply to Pu WenTo make Xen works functionally on Hygon platforms, reuse AMD's Xen support code path for Hygon Dhyana CPU.
There are six core performance events counters per thread, so there are six MSRs for these counters(0-5). Also there are four legacy PMC MSRs, they are alias of the counters(0-3).
In this version of kernel Hygon use the lagacy and safe versions of MSR access. It works fine when VPMU enabled in Xen on Hygon platforms by testing with perf.
--- arch/x86/xen/pmu.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
arch/x86/xen/pmu.c +9 -3
@@ -90,6 +90,12 @@ static void xen_pmu_arch_init(void) k7_counters_mirrored =0 ;break ; } + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + amd_num_counters =F10H_NUM_COUNTERS ; + amd_counters_base =MSR_K7_PERFCTR0 ; + amd_ctrls_base =MSR_K7_EVNTSEL0 ; + amd_msr_step =1 ; + k7_counters_mirrored =0 ; }else {uint32_t eax, ebx, ecx, edx;
@@ -285,7 +291,7 @@ static bool xen_amd_pmu_emulate(unsigned int msr, u64 *val, bool is_read)bool pmu_msr_read (unsigned int msr,uint64_t *val,int *err) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL ) {if (is_amd_pmu_msr (msr)) {if (!xen_amd_pmu_emulate (msr, val,1 )) *val =native_read_msr_safe (msr, err);
@@ -308,7 +314,7 @@ bool pmu_msr_write(unsigned int msr, uint32_t low, uint32_t high, int *err) {uint64_t val = ((uint64_t )high <<32 ) | low; -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL ) {if (is_amd_pmu_msr (msr)) {if (!xen_amd_pmu_emulate (msr, &val,0 )) *err =native_write_msr_safe (msr, low, high);
@@ -379,7 +385,7 @@ static unsigned long long xen_intel_read_pmc(int counter)unsigned long long xen_read_pmc (int counter) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL )return xen_amd_read_pmc (counter);else return xen_intel_read_pmc (counter);
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 12/16 x86/xen: Add Hygon Dhyana support to Xen
In reply to Pu WenTo make Xen works functionally on Hygon platforms, reuse AMD's Xen support code path for Hygon Dhyana CPU.
There are six core performance events counters per thread, so there are six MSRs for these counters(0-5). Also there are four legacy PMC MSRs, they are alias of the counters(0-3).
In this version of kernel Hygon use the lagacy and safe versions of MSR access. It works fine when VPMU enabled in Xen on Hygon platforms by testing with perf.
--- arch/x86/xen/pmu.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
arch/x86/xen/pmu.c +9 -3
@@ -90,6 +90,12 @@ static void xen_pmu_arch_init(void) k7_counters_mirrored =0 ;break ; } + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + amd_num_counters =F10H_NUM_COUNTERS ; + amd_counters_base =MSR_K7_PERFCTR0 ; + amd_ctrls_base =MSR_K7_EVNTSEL0 ; + amd_msr_step =1 ; + k7_counters_mirrored =0 ; }else {uint32_t eax, ebx, ecx, edx;
@@ -285,7 +291,7 @@ static bool xen_amd_pmu_emulate(unsigned int msr, u64 *val, bool is_read)bool pmu_msr_read (unsigned int msr,uint64_t *val,int *err) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL ) {if (is_amd_pmu_msr (msr)) {if (!xen_amd_pmu_emulate (msr, val,1 )) *val =native_read_msr_safe (msr, err);
@@ -308,7 +314,7 @@ bool pmu_msr_write(unsigned int msr, uint32_t low, uint32_t high, int *err) {uint64_t val = ((uint64_t )high <<32 ) | low; -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL ) {if (is_amd_pmu_msr (msr)) {if (!xen_amd_pmu_emulate (msr, &val,0 )) *err =native_write_msr_safe (msr, low, high);
@@ -379,7 +385,7 @@ static unsigned long long xen_intel_read_pmc(int counter)unsigned long long xen_read_pmc (int counter) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL )return xen_amd_read_pmc (counter);else return xen_intel_read_pmc (counter);
Signature · 7 lines
2.7.4 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
Pu Wen
maintainerPATCH v6 13/16 ACPI, x86: Add Hygon Dhyana support
In reply to Pu WenThe Hygon Dhyana CPU has NONSTOP TSC feature, so enable the ACPI driver support to it.
--- drivers/acpi/acpi_pad.c | 1 + drivers/acpi/processor_idle.c | 1 + 2 files changed, 2 insertions(+)
drivers/acpi/acpi_pad.c +1 -0
@@ -70,6 +70,7 @@ static void power_saving_mwait_init(void)#if defined(CONFIG_X86 )switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :case X86_VENDOR_INTEL :/*
drivers/acpi/processor_idle.c +1 -0
@@ -205,6 +205,7 @@ static void lapic_timer_state_broadcast(struct acpi_processor *pr,static void tsc_check_state (int state) {switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :case X86_VENDOR_INTEL :case X86_VENDOR_CENTAUR :
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 14/16 cpufreq, x86: Add Hygon Dhyana support
In reply to Pu WenAs the Hygon Dhyana CPU support ACPI P-States feature, and there is SMBus device(PCI device ID 0x790b) on Hygon platform, so add Hygon Dhyana support to the cpufreq driver by using the code path of AMD family 17h.
--- drivers/cpufreq/acpi-cpufreq.c | 5 +++++ drivers/cpufreq/amd_freq_sensitivity.c | 9 +++++++-- 2 files changed, 12 insertions(+), 2 deletions(-)
drivers/cpufreq/acpi-cpufreq.c +5 -0
@@ -61,6 +61,7 @@ enum {#define INTEL_MSR_RANGE (0xffff)#define AMD_MSR_RANGE (0x7) +#define HYGON_MSR_RANGE (0x7)#define MSR_K7_HWCR_CPB_DIS (1ULL << 25)
@@ -95,6 +96,7 @@ static bool boost_state(unsigned int cpu)rdmsr_on_cpu (cpu,MSR_IA32_MISC_ENABLE , &lo, &hi); msr = lo | ((u64 )hi <<32 );return !(msr &MSR_IA32_MISC_ENABLE_TURBO_DISABLE ); +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :rdmsr_on_cpu (cpu,MSR_K7_HWCR , &lo, &hi); msr = lo | ((u64 )hi <<32 );
@@ -113,6 +115,7 @@ static int boost_set_msr(bool enable) msr_addr =MSR_IA32_MISC_ENABLE ; msr_mask =MSR_IA32_MISC_ENABLE_TURBO_DISABLE ;break ; +case X86_VENDOR_HYGON :case X86_VENDOR_AMD : msr_addr =MSR_K7_HWCR ; msr_mask =MSR_K7_HWCR_CPB_DIS ;
@@ -225,6 +228,8 @@ static unsigned extract_msr(struct cpufreq_policy *policy, u32 msr)if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) msr &=AMD_MSR_RANGE ; +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + msr &=HYGON_MSR_RANGE ;else msr &=INTEL_MSR_RANGE ;
drivers/cpufreq/amd_freq_sensitivity.c +7 -2
@@ -111,11 +111,16 @@ static int __init amd_freq_sensitivity_init(void) {u64 val;struct pci_dev *pcidev; +unsigned int pci_vendor; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) + pci_vendor =PCI_VENDOR_ID_AMD ; +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + pci_vendor =PCI_VENDOR_ID_HYGON ; +else return -ENODEV ; - pcidev =pci_get_device (PCI_VENDOR_ID_AMD , + pcidev =pci_get_device (pci_vendor,PCI_DEVICE_ID_AMD_KERNCZ_SMBUS ,NULL );if (!pcidev) {
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 15/16 EDAC, amd64: Add Hygon Dhyana support
In reply to Pu WenTo make AMD64 EDAC and MCE drivers working on Hygon platforms, add support for Hygon Dhyana CPU by using the code path of AMD family 17h.
--- drivers/edac/amd64_edac.c | 8 +++++++- drivers/edac/mce_amd.c | 4 +++- 2 files changed, 10 insertions(+), 2 deletions(-)
drivers/edac/amd64_edac.c +7 -1
@@ -211,7 +211,7 @@ static int __set_scrub_rate(struct amd64_pvt *pvt, u32 new_bw, u32 min_rate) scrubval = scrubrates[i].scrubval; -if (pvt->fam ==0x17 ) { +if (pvt->fam ==0x17 || pvt->fam ==0x18 ) {__f17h_set_scrubval (pvt, scrubval); }else if (pvt->fam ==0x15 && pvt->model ==0x60 ) {f15h_select_dct (pvt,0 );
@@ -264,6 +264,7 @@ static int get_scrub_rate(struct mem_ctl_info *mci)break ;case 0x17 : +case 0x18 :amd64_read_pci_cfg (pvt->F6,F17H_SCR_BASE_ADDR , &scrubval);if (scrubval &BIT (0 )) {amd64_read_pci_cfg (pvt->F6,F17H_SCR_LIMIT_ADDR , &scrubval);
@@ -1044,6 +1045,7 @@ static void determine_memory_type(struct amd64_pvt *pvt) goto ddr3;case 0x17 : +case 0x18 :if ((pvt->umc[0 ].dimm_cfg | pvt->umc[1 ].dimm_cfg) &BIT (5 )) pvt->dram_type =MEM_LRDDR4 ;else if ((pvt->umc[0 ].dimm_cfg | pvt->umc[1 ].dimm_cfg) &BIT (4 ))
@@ -3188,8 +3190,11 @@ static struct amd64_family_type *per_family_init(struct amd64_pvt *pvt)break ;case 0x17 : +case 0x18 : fam_type = &family_types[F17_CPUS ]; pvt->ops = &family_types[F17_CPUS ].ops; +if (pvt->fam ==0x18 ) + family_types[F17_CPUS ].ctl_name ="F18h" ;break ;default :
@@ -3428,6 +3433,7 @@ static const struct x86_cpu_id amd64_cpuids[] = { { X86_VENDOR_AMD, 0x15, X86_MODEL_ANY, X86_FEATURE_ANY, 0 }, { X86_VENDOR_AMD, 0x16, X86_MODEL_ANY, X86_FEATURE_ANY, 0 }, { X86_VENDOR_AMD, 0x17, X86_MODEL_ANY, X86_FEATURE_ANY, 0 }, + { X86_VENDOR_HYGON, 0x18, X86_MODEL_ANY, X86_FEATURE_ANY, 0 }, { } };MODULE_DEVICE_TABLE (x86cpu, amd64_cpuids);
drivers/edac/mce_amd.c +3 -1
@@ -1059,7 +1059,8 @@ static int __init mce_amd_init(void) {struct cpuinfo_x86 *c = &boot_cpu_data; -if (c->x86_vendor !=X86_VENDOR_AMD ) +if (c->x86_vendor !=X86_VENDOR_AMD && + c->x86_vendor !=X86_VENDOR_HYGON )return -ENODEV ; fam_ops =kzalloc (sizeof (struct amd_decoder_ops ),GFP_KERNEL );
@@ -1113,6 +1114,7 @@ static int __init mce_amd_init(void)break ;case 0x17 : +case 0x18 : xec_mask =0x3f ;if (!boot_cpu_has (X86_FEATURE_SMCA )) {printk (KERN_WARNING "Decoding supported only on Scalable MCA processors.\n" );
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v6 16/16 cpupower, x86: Add Hygon Dhyana support
In reply to Pu WenTool cpupower is useful to get CPU frequency information and monitor power stats on Hygon Dhyana platforms. So add Hygon Dhyana support to it by checking vendor and family to share the code path of AMD family 17h.
--- tools/power/cpupower/utils/cpufreq-info.c | 6 ++++-- tools/power/cpupower/utils/helpers/amd.c | 4 ++-- tools/power/cpupower/utils/helpers/cpuid.c | 8 +++++--- tools/power/cpupower/utils/helpers/helpers.h | 2 +- tools/power/cpupower/utils/helpers/misc.c | 2 +- tools/power/cpupower/utils/idle_monitor/mperf_monitor.c | 3 ++- 6 files changed, 15 insertions(+), 10 deletions(-)
tools/power/cpupower/utils/cpufreq-info.c +4 -2
@@ -170,6 +170,7 @@ static int get_boost_mode(unsigned int cpu)unsigned long pstates[MAX_HW_PSTATES ] = {0 ,};if (cpupower_cpu_info.vendor !=X86_VENDOR_AMD && + cpupower_cpu_info.vendor !=X86_VENDOR_HYGON && cpupower_cpu_info.vendor !=X86_VENDOR_INTEL )return 0 ;
@@ -190,8 +191,9 @@ static int get_boost_mode(unsigned int cpu)printf (_ (" Supported: %s\n" ), support ?_ ("yes" ) :_ ("no" ));printf (_ (" Active: %s\n" ), active ?_ ("yes" ) :_ ("no" )); -if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD && - cpupower_cpu_info.family >=0x10 ) { +if ((cpupower_cpu_info.vendor ==X86_VENDOR_AMD && + cpupower_cpu_info.family >=0x10 ) || + cpupower_cpu_info.vendor ==X86_VENDOR_HYGON ) { ret =decode_pstates (cpu, cpupower_cpu_info.family, b_states, pstates, &pstate_no);if (ret)
tools/power/cpupower/utils/helpers/amd.c +2 -2
@@ -45,7 +45,7 @@ static int get_did(int family, union msr_pstate pstate)if (family ==0x12 ) t = pstate.val &0xf ; -else if (family ==0x17 ) +else if (family ==0x17 || family ==0x18 ) t = pstate.fam17h_bits.did;else t = pstate.bits.did;
@@ -59,7 +59,7 @@ static int get_cof(int family, union msr_pstate pstate)int fid, did, cof; did =get_did (family, pstate); -if (family ==0x17 ) { +if (family ==0x17 || family ==0x18 ) { fid = pstate.fam17h_bits.fid; cof =200 * fid / did; }else {
tools/power/cpupower/utils/helpers/cpuid.c +5 -3
@@ -8,7 +8,7 @@#include "helpers/helpers.h" static const char *cpu_vendor_table[X86_VENDOR_MAX ] = { -"Unknown" ,"GenuineIntel" ,"AuthenticAMD" , +"Unknown" ,"GenuineIntel" ,"AuthenticAMD" ,"HygonGenuine" , };#if defined(__i386__) || defined(__x86_64__)
@@ -109,6 +109,7 @@ int get_cpu_info(struct cpupower_cpu_info *cpu_info)fclose (fp);/* Get some useful CPU capabilities from cpuid */ if (cpu_info->vendor !=X86_VENDOR_AMD && + cpu_info->vendor !=X86_VENDOR_HYGON && cpu_info->vendor !=X86_VENDOR_INTEL )return ret;
@@ -124,8 +125,9 @@ int get_cpu_info(struct cpupower_cpu_info *cpu_info)if (cpuid_level >=6 && (cpuid_ecx (6 ) &0x1 )) cpu_info->caps |=CPUPOWER_CAP_APERF ; -/* AMD Boost state enable/disable register */ -if (cpu_info->vendor ==X86_VENDOR_AMD ) { +/* AMD or Hygon Boost state enable/disable register */ +if (cpu_info->vendor ==X86_VENDOR_AMD || + cpu_info->vendor ==X86_VENDOR_HYGON ) {if (ext_cpuid_level >=0x80000007 && (cpuid_edx (0x80000007 ) & (1 <<9 ))) cpu_info->caps |=CPUPOWER_CAP_AMD_CBP ;
tools/power/cpupower/utils/helpers/helpers.h +1 -1
@@ -61,7 +61,7 @@ extern int be_verbose;/* cpuid and cpuinfo helpers **************************/ enum cpupower_cpu_vendor {X86_VENDOR_UNKNOWN =0 ,X86_VENDOR_INTEL , -X86_VENDOR_AMD ,X86_VENDOR_MAX }; +X86_VENDOR_AMD ,X86_VENDOR_HYGON ,X86_VENDOR_MAX };#define CPUPOWER_CAP_INV_TSC 0x00000001#define CPUPOWER_CAP_APERF 0x00000002
tools/power/cpupower/utils/helpers/misc.c +1 -1
@@ -26,7 +26,7 @@ int cpufreq_has_boost_support(unsigned int cpu, int *support, int *active,* has Hardware determined variable increments instead. */ -if (cpu_info.family ==0x17 ) { +if (cpu_info.family ==0x17 || cpu_info.family ==0x18 ) {if (!read_msr (cpu,MSR_AMD_HWCR , &val)) {if (!(val &CPUPOWER_AMD_CPBDIS )) *active =1 ;
tools/power/cpupower/utils/idle_monitor/mperf_monitor.c +2 -1
@@ -241,7 +241,8 @@ static int init_maxfreq_mode(void)if (!(cpupower_cpu_info.caps &CPUPOWER_CAP_INV_TSC )) goto use_sysfs; -if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD ) { +if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD || + cpupower_cpu_info.vendor ==X86_VENDOR_HYGON ) {/* MSR_AMD_HWCR tells us whether TSC runs at P0/mperf * freq. * A test whether hwcr is accessable/available would be:
Signature · 1 line
2.7.4
Re: [PATCH v6 01/16] x86/cpu: Create Hygon Dhyana architecture support file
In reply to Pu WenQuoted 1 line · Pu Wen
Add x86 architecture support for new processor Hygon Dhyana Family
"... for a new processor: Hygon... "
Quoted 1 line
18h.
Instead of all that:
Quoted 7 lines
Rework to create a separated file(arch/x86/kernel/cpu/hygon.c) from the AMD one(arch/x86/kernel/cpu/amd.c) to initialize Dhyana.
In this way we can remove old AMD architecture support codes and generate a clear initialization flow for Hygon Dhyana CPU. It also reduce the long-term maintenance effort.
... simply say: "Carve out initialization code needed by Dhyana into a separate compilation unit."
Quoted 2 lines
Add Maintainer information
forhygon.cin accordance.
That sentence is not needed - that fact is visible in the diff.
Quoted 3 lines
To identify Hygon Dhyana CPU, add a new vendor type
X86_VENDOR_HYGON(9)for system recognition.
No need for that "(9)" - also visible from the diff and an implementation detail.
Quoted 3 lines
To enable the config for Hygon Dhyana CPU, add a separated
Kconfigentry(CPU_SUP_HYGON)in kernel config setup.
That sentence is not needed for the same reason as above.
Quoted 2 lines
And select
CPU_SUP_AMD
to solve the dependency issues in kernel.
No, this should say: "Since Dhyana uses AMD functionality to a large degree, select CPU_SUP_AMD which provides that functionality."
Quoted 10 lines
Signed-off-by Pu Wen <[email protected]>--- MAINTAINERS | 6 + arch/x86/Kconfig.cpu | 14 ++ arch/x86/include/asm/processor.h | 3 +- arch/x86/kernel/cpu/Makefile | 1 + arch/x86/kernel/cpu/hygon.c | 408 +++++++++++++++++++++++++++++++++++++++ 5 files changed, 431 insertions(+), 1 deletion(-) create mode 100644 arch/x86/kernel/cpu/hygon.c
With that taken care of:
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Re: [PATCH v6 03/16] x86/cpu/mtrr: Support TOP_MEM2 and get MTRR number
In reply to Pu WenQuoted 1 line · Pu Wen
The Hygon Dhyana CPU have a special magic MSR way to force WB for
From the last review round:
Also, it is "The ... CPU has a special..."
Please take your time and incorporate *all* review feedback - no need to *rush* a new revision out and drop review feedback.
Quoted 10 lines
memory >4GB, and support
TOP_MEM2. Therefore, it is necessary to add Hygon Dhyana support inamd_special_default_mtrr().The number of variable MTRRs for Hygon is 2 as AMD's.
Signed-off-by Pu Wen <[email protected]>--- arch/x86/kernel/cpu/mtrr/cleanup.c | 3 ++- arch/x86/kernel/cpu/mtrr/mtrr.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-)
With that fixed:
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Re: [PATCH v6 05/16] perf/x86: Add Hygon Dhyana support to PMU infrastructure
In reply to Pu WenQuoted 5 lines · Pu Wen
The PMU architecture for Hygon Dhyana CPU is similar to the AMD Family 17h one. To support Hygon Dhyana PMU, call
amd_pmu_init()to share AMD PMU initialization flow, and change the PMU name to "HYGON".The Hygon Dhyana CPU support both legacy and extension PMC MSRs(perf
I don't know but for some reason, you are writing "Hygon Dhyana CPU" as being plural. But it is singular:
"The Hygon Dhyna CPU supports both ..."
^ |||
Quoted 2 lines
counter registers and event selection registers), so add Hygon Dhyana support to get bit offset in the similar way as AMD does.
"to get bit offset"?
Quoted 7 lines
Signed-off-by Pu Wen <[email protected]>--- arch/x86/events/amd/core.c | 4 ++++ arch/x86/events/amd/uncore.c | 20 +++++++++++++------- arch/x86/events/core.c | 4 ++++ arch/x86/kernel/cpu/perfctr-watchdog.c | 2 ++ 4 files changed, 23 insertions(+), 7 deletions(-)
With that addressed:
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Pu Wen
maintainerRe: [PATCH v6 01/16] x86/cpu: Create Hygon Dhyana architecture support file
In reply to Borislav PetkovQuoted 7 lines · Borislav Petkov
That sentence is not needed for the same reason as above.
Quoted 2 lines
And select
CPU_SUP_AMD
to solve the dependency issues in kernel.No, this should say: "Since Dhyana uses AMD functionality to a large degree, select
CPU_SUP_AMDwhich provides that functionality."
Thanks for the detailed correction. Will improve the description and remove unneeded sentences in next version of patch.
Signature · 2 lines
Regards, Pu Wen
Pu Wen
maintainerRe: [PATCH v6 03/16] x86/cpu/mtrr: Support TOP_MEM2 and get MTRR number
In reply to Borislav PetkovQuoted 9 lines · Borislav Petkov
Quoted 1 line · Pu Wen
The Hygon Dhyana CPU have a special magic MSR way to force WB for
From the last review round:
Also, it is "The ... CPU has a special..."
Please take your time and incorporate *all* review feedback - no need to *rush* a new revision out and drop review feedback.
Sorry for not taking much care of the usage between singular and plural. Will fix those typos.
Signature · 2 lines
Regards, Pu Wen
Pu Wen
maintainerRe: [PATCH v6 05/16] perf/x86: Add Hygon Dhyana support to PMU infrastructure
In reply to Borislav PetkovQuoted 8 lines · Borislav Petkov
Quoted 1 line
The Hygon Dhyana CPU support both legacy and extension PMC MSRs(perf
I don't know but for some reason, you are writing "Hygon Dhyana CPU" as being plural. But it is singular:
"The Hygon Dhyna CPU supports both ..."
^ |||
Will take more care of the usage between singular and plural and fix the typo.
Quoted 4 lines
Quoted 2 lines
counter registers and event selection registers), so add Hygon Dhyana support to get bit offset in the similar way as AMD does.
"to get bit offset"?
These words are used with the tendency to explain what the patch do, actually they are not needed and will be removed.
Signature · 2 lines
Regards, Pu Wen
Re: [PATCH v6 07/16] x86/pci: Add Hygon Dhyana support to PCI and north bridge
In reply to Pu WenQuoted 5 lines · Pu Wen
As Hygon registered its PCI Vendor ID as a new one
0x1d94, and there are PCI Devices 0x1450/0x1463/0x1464 for Host bridge on Hygon Dhyana platforms, so add Hygon Dhyana support to the PCI and north bridge subsystem by using the code path of AMD family 17h.
Quoted 7 lines
Signed-off-by Pu Wen <[email protected]>--- arch/x86/kernel/amd_nb.c | 47 +++++++++++++++++++++++++++++++++++++++++------ arch/x86/pci/amd_bus.c | 6 ++++-- include/linux/pci_ids.h | 2 ++ 3 files changed, 47 insertions(+), 8 deletions(-)
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Re: [PATCH v6 08/16] x86/apic: Add Hygon Dhyana support to APIC
In reply to Pu WenQuoted 27 lines · Pu Wen
Add Hygon Dhyana support to the APIC subsystem as it use modern APIC. When running on 32 bit mode, bigsmp should be enabled if there are more than 8 cores online.
Signed-off-by Pu Wen <[email protected]>--- arch/x86/kernel/apic/apic.c | 13 +++++++++++-- arch/x86/kernel/apic/probe_32.c | 1 + 2 files changed, 12 insertions(+), 2 deletions(-)arch/x86/kernel/apic/apic.c +5
-0@@ -224,6 +224,11 @@ static int modern_apic(void)if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD && boot_cpu_data.x86 >=0xf )return 1 ; + +/* Hygon systems use modern APIC */ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) +return 1 ; +return lapic_get_version () >=0x14 ; }
arch/x86/kernel/apic/apic.c void __init sync_Arb_IDs(void) · line 1223
Quoted 10 lines
arch/x86/kernel/apic/apic.c +4
-2@@ -1211,9 +1216,11 @@ void __init sync_Arb_IDs(void) {/* * Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 And not -* needed on AMD. +* needed on AMD or Hygon. */ -if (modern_apic () || boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (modern_apic () || + boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON )
This change is not needed, after the above hunk in modern_apic(), right?
With it removed:
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Borislav Petkov
maintainerRe: [PATCH v6 09/16] x86/bugs: Add mitigation to spectre and no meltdown for Hygon Dhyana
In reply to Pu WenQuoted 17 lines · Pu Wen
The Hygon Dhyana CPU has the same speculative execution as AMD family 17h, so share AMD spectre mitigation code for Hygon Dhyana.
Also Hygon Dhyana is not affected by meltdown vulnerability as AMD, so add the exception for Hygon Dhyana.
Signed-off-by Pu Wen <[email protected]>--- arch/x86/kernel/cpu/bugs.c | 6 ++++-- arch/x86/kernel/cpu/common.c | 1 + 2 files changed, 5 insertions(+), 2 deletions(-)arch/x86/kernel/cpu/bugs.c +0
-0
arch/x86/kernel/cpu/bugs.c static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void) · line 317
Quoted 7 lines
arch/x86/kernel/cpu/bugs.c +2
-1@@ -312,8 +312,9 @@ static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void) }if (cmd ==SPECTRE_V2_CMD_RETPOLINE_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON && boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) { -pr_err ("retpoline,amd selected but CPU is not AMD. Switching to AUTO select\n" ); +pr_err ("retpoline,amd selected but CPU is not AMD or Hygon. Switching to AUTO select\n" );
Didn't we say, no user-visible changes pls?
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Re: [PATCH v6 10/16] x86/mce: Add Hygon Dhyana support to MCE infrastructure
In reply to Pu WenQuoted 10 lines · Pu Wen
The machine check architecture for Hygon Dhyana CPU is similar to the AMD family 17h one. To make MCE working on Hygon platforms, add vendor checking for Hygon Dhyana to share the code path of AMD family 17h.
Signed-off-by Pu Wen <[email protected]>--- arch/x86/include/asm/mce.h | 2 ++ arch/x86/kernel/cpu/mcheck/mce-severity.c | 3 ++- arch/x86/kernel/cpu/mcheck/mce.c | 20 +++++++++++++++----- 3 files changed, 19 insertions(+), 6 deletions(-)
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Borislav Petkov
maintainerRe: [PATCH v6 15/16] EDAC, amd64: Add Hygon Dhyana support
In reply to Pu WenQuoted 9 lines · Pu Wen
To make AMD64 EDAC and MCE drivers working on Hygon platforms, add support for Hygon Dhyana CPU by using the code path of AMD family 17h.
Signed-off-by Pu Wen <[email protected]>--- drivers/edac/amd64_edac.c | 8 +++++++- drivers/edac/mce_amd.c | 4 +++- 2 files changed, 10 insertions(+), 2 deletions(-)
This one conflicts with
8960de4a5ca7 ("EDAC, amd64: Add Family 17h, models 10h-2fh support")
in my tree.
tglx, what do you suggest?
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Pu Wen
maintainerRe: [PATCH v6 08/16] x86/apic: Add Hygon Dhyana support to APIC
In reply to Borislav PetkovQuoted 5 lines · Borislav Petkov
Quoted 5 lines
arch/x86/kernel/apic/apic.c +0
-1@@ -1211,9 +1216,11 @@ void __init sync_Arb_IDs(void) {/* * Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 And not -* needed on AMD.
Quoted 8 lines
arch/x86/kernel/apic/apic.c void __init sync_Arb_IDs(void) · line 1223
Quoted 6 lines
arch/x86/kernel/apic/apic.c +4
-1+* needed on AMD or Hygon. */ -if (modern_apic () || boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (modern_apic () || + boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON )This change is not needed, after the above hunk in
modern_apic(), right?
Yes, you are right. Since the Hygon Dhyana CPU always uses modern APIC, so the function modern_apic will always return true for Hygon Dhyana. I will remove this change in next version of patch.
Quoted 4 lines
With it removed:
Reviewed-by Borislav Petkov <[email protected]>
Signature · 2 lines
Regards, Pu Wen
Pu Wen
maintainerRe: [PATCH v6 09/16] x86/bugs: Add mitigation to spectre and no meltdown for Hygon Dhyana
In reply to Borislav PetkovQuoted 7 lines · Borislav Petkov
arch/x86/kernel/cpu/bugs.c static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void) · line 317
Quoted 5 lines
if (cmd ==
SPECTRE_V2_CMD_RETPOLINE_AMD&&arch/x86/kernel/cpu/bugs.c +2
-1+ boot_cpu_data.x86_vendor != X86_VENDOR_HYGON && boot_cpu_data.x86_vendor != X86_VENDOR_AMD) { -pr_err ("retpoline,amd selected but CPU is not AMD. Switching to AUTO select\n" ); +pr_err ("retpoline,amd selected but CPU is not AMD or Hygon. Switching to AUTO select\n" );Didn't we say, no user-visible changes pls?
Sorry for the incomplete modification. Will revert the pr_err string.
Signature · 2 lines
Regards, Pu Wen
Re: [PATCH v6 07/16] x86/pci: Add Hygon Dhyana support to PCI and north bridge
In reply to Pu WenQuoted 46 lines · Pu Wen
As Hygon registered its PCI Vendor ID as a new one
0x1d94, and there are PCI Devices 0x1450/0x1463/0x1464 for Host bridge on Hygon Dhyana platforms, so add Hygon Dhyana support to the PCI and north bridge subsystem by using the code path of AMD family 17h.Signed-off-by Pu Wen <[email protected]>--- arch/x86/kernel/amd_nb.c | 47 +++++++++++++++++++++++++++++++++++++++++------ arch/x86/pci/amd_bus.c | 6 ++++-- include/linux/pci_ids.h | 2 ++ 3 files changed, 47 insertions(+), 8 deletions(-)arch/x86/kernel/amd_nb.c +18
-0@@ -61,6 +61,21 @@ static const struct pci_device_id amd_nb_link_ids[] = { {} }; +static const struct pci_device_id hygon_root_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_ROOT ) }, + {} +}; + +const struct pci_device_id hygon_nb_misc_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_DF_F3 ) }, + {} +}; + +static const struct pci_device_id hygon_nb_link_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_DF_F4 ) }, + {} +}; +const struct amd_nb_bus_dev_range amd_nb_bus_dev_ranges[]__initconst = { {0x00 ,0x18 ,0x20 }, {0xff ,0x00 ,0x20 },@@ -197,12 +212,25 @@ int amd_cache_northbridges(void)u16 i =0 ;struct amd_northbridge *nb;struct pci_dev *root, *misc, *link; +const struct pci_device_id *root_ids =NULL ; +const struct pci_device_id *misc_ids =NULL ; +const struct pci_device_id *link_ids =NULL ;
arch/x86/kernel/amd_nb.c int amd_cache_northbridges(void) · line 227
Quoted 10 lines
arch/x86/kernel/amd_nb.c +10
-0+ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { + root_ids = amd_root_ids; + misc_ids = amd_nb_misc_ids; + link_ids = amd_nb_link_ids; + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + root_ids = hygon_root_ids; + misc_ids = hygon_nb_misc_ids; + link_ids = hygon_nb_link_ids; + }
To be compatible with "before this patch" you should probably do:
if (boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) {
root_ids = hygon_root_ids;
misc_ids = hygon_nb_misc_ids;
link_ids = hygon_nb_link_ids;
} else {
root_ids = amd_root_ids;
misc_ids = amd_nb_misc_ids;
link_ids = amd_nb_link_ids;
}That way they are always the AMD values if not your chip.
Quoted 26 lines
arch/x86/kernel/amd_nb.c +4
-4if (amd_northbridges.num)return 0 ; misc =NULL ; -while ((misc =next_northbridge (misc, amd_nb_misc_ids)) !=NULL ) +while ((misc =next_northbridge (misc, misc_ids)) !=NULL ) i++;if (!i)@@ -218,11 +246,11 @@ int amd_cache_northbridges(void) link = misc = root =NULL ;for (i =0 ; i != amd_northbridges.num; i++) {node_to_amd_nb (i)->root = root = -next_northbridge (root, amd_root_ids); +next_northbridge (root, root_ids);node_to_amd_nb (i)->misc = misc = -next_northbridge (misc, amd_nb_misc_ids); +next_northbridge (misc, misc_ids);node_to_amd_nb (i)->link = link = -next_northbridge (link, amd_nb_link_ids); +next_northbridge (link, link_ids); }if (amd_gart_present ())
arch/x86/kernel/amd_nb.c bool __init early_is_amd_nb(u32 device) · line 299
Quoted 9 lines
arch/x86/kernel/amd_nb.c +6
-0@@ -263,9 +291,15 @@ bool __init early_is_amd_nb(u32 device) {const struct pci_device_id *id;u32 vendor = device &0xffff ; +const struct pci_device_id *misc_ids =NULL ; + +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) + misc_ids = amd_nb_misc_ids; +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + misc_ids = hygon_nb_misc_ids;
Same comment as above. This will probably eliminate the PANIC that that was reported by LKP.
Thanks,
Tom
Quoted 55 lines
arch/x86/kernel/amd_nb.c +3
-2device >>=16 ; -for (id = amd_nb_misc_ids; id->vendor; id++) +for (id = misc_ids; id->vendor; id++)if (vendor == id->vendor && device == id->device)return true;return false;@@ -277,7 +311,8 @@ struct resource *amd_get_mmconfig_range(struct resource *res)u64 base, msr;unsigned int segn_busn_bits; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return NULL ;/* assume all cpus from fam10h have mmconfig */ arch/x86/pci/amd_bus.c +4
-2@@ -93,7 +93,8 @@ static int __init early_root_info_init(void) vendor = id &0xffff ; device = (id>>16 ) &0xffff ; -if (vendor !=PCI_VENDOR_ID_AMD ) +if (vendor !=PCI_VENDOR_ID_AMD && + vendor !=PCI_VENDOR_ID_HYGON )continue ;if (hb_probes[i].device == device) {@@ -390,7 +391,8 @@ static int __init pci_io_ecs_init(void)static int __init amd_postcore_init (void ) { -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return 0 ;early_root_info_init ();include/linux/pci_ids.h +2
-0@@ -2561,6 +2561,8 @@#define PCI_VENDOR_ID_AMAZON 0x1d0f +#define PCI_VENDOR_ID_HYGON 0x1d94 +#define PCI_VENDOR_ID_TEKRAM 0x1de1#define PCI_DEVICE_ID_TEKRAM_DC290 0xdc29
Re: [PATCH v6 07/16] x86/pci: Add Hygon Dhyana support to PCI and north bridge
Quoted 13 lines · Thomas
To be compatible with "before this patch" you should probably do:
if (boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) { root_ids = hygon_root_ids; misc_ids = hygon_nb_misc_ids; link_ids = hygon_nb_link_ids; } else { root_ids = amd_root_ids; misc_ids = amd_nb_misc_ids; link_ids = amd_nb_link_ids; }
Quoted 1 line
That way they are always the AMD values if not your chip.
And please can we have a comment why this is called on anything which is not AMD or HYGON. It's all but obvious and clearly unexpected.
Thanks,
tglx
Pu Wen
maintainerRe: [PATCH v6 07/16] x86/pci: Add Hygon Dhyana support to PCI and north bridge
In reply to Lendacky, ThomasQuoted 32 lines · Thomas
Quoted 7 lines
arch/x86/kernel/amd_nb.c +3
-0@@ -197,12 +212,25 @@ int amd_cache_northbridges(void)u16 i =0 ;struct amd_northbridge *nb;struct pci_dev *root, *misc, *link; +const struct pci_device_id *root_ids =NULL ; +const struct pci_device_id *misc_ids =NULL ; +const struct pci_device_id *link_ids =NULL ;arch/x86/kernel/amd_nb.c int amd_cache_northbridges(void) · line 227
Quoted 10 lines
arch/x86/kernel/amd_nb.c +10
-0+ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { + root_ids = amd_root_ids; + misc_ids = amd_nb_misc_ids; + link_ids = amd_nb_link_ids; + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + root_ids = hygon_root_ids; + misc_ids = hygon_nb_misc_ids; + link_ids = hygon_nb_link_ids; + }To be compatible with "before this patch" you should probably do:
if (boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) { root_ids = hygon_root_ids; misc_ids = hygon_nb_misc_ids; link_ids = hygon_nb_link_ids; } else { root_ids = amd_root_ids; misc_ids = amd_nb_misc_ids; link_ids = amd_nb_link_ids; }That way they are always the AMD values if not your chip.
...
Quoted 5 lines
Quoted 5 lines
arch/x86/kernel/amd_nb.c +1
-0@@ -263,9 +291,15 @@ bool __init early_is_amd_nb(u32 device) {const struct pci_device_id *id;u32 vendor = device &0xffff ; +const struct pci_device_id *misc_ids =NULL ;
Quoted 8 lines
arch/x86/kernel/amd_nb.c bool __init early_is_amd_nb(u32 device) · line 299
Quoted 5 lines
arch/x86/kernel/amd_nb.c +5
-0+ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) + misc_ids = amd_nb_misc_ids; +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + misc_ids = hygon_nb_misc_ids;Same comment as above. This will probably eliminate the PANIC that that was reported by LKP.
Yes, it's this modification that caused the PANIC reported by LKP. Because the misc_ids will be NULL if running on Intel chip.
If change the vendor checking like this:
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_HYGON)
+ misc_ids = hygon_nb_misc_ids;
+ else
+ misc_ids = amd_nb_misc_ids;
The PANIC will be eliminated.
The PANIC will also be eliminated by adding AMD vendor checking in amd_gart_present:
+ if(boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
+ return false;
To prevent the function return true on non AMD platform.
Thanks,
Pu Wen
Pu Wen
maintainerv8 cover Add support for Hygon Dhyana Family 18h processor
As a new x86 CPU Vendor, Chengdu Haiguang IC Design Co., Ltd (Hygon) is a Joint Venture between AMD and Haiguang Information Technology Co., Ltd., and aims at providing high performance x86 processor for China server market.
The first generation Hygon's processor(Dhyana) originates from AMD technology and shares most of the architecture with AMD's family 17h, but with different CPU Vendor ID("HygonGenuine")/PCIE Device Vendor ID (0x1D94)/Family series number (Family 18h).
To enable the support of Linux kernel to Hygon's CPU, we added a new vendor type (X86_VENDOR_HYGON, with value of 9) in arch/x86/include/ asm/processor.h, and shared most of kernel support codes with AMD family 17h.
As Hygon will negotiate with AMD to make sure that only Hygon will use family 18h, so try to minimize code modification and share most codes with AMD under this consideration.
This patch series have been applied and tested successfully on Hygon Dhyana SoC silicon. Also tested on AMD EPYC (Family 17h) processor, it works fine and makes no harm to the existing codes.
v7->v8:
- Fix the issue that patch x86/pci may cause LKP boot test failure on some Intel platform.
v6->v7:
- Rebased on 4.19-rc4 and tested against it.
- Add Reviewed-by from Borislav Petkov for patches x86/cpu,
x86/cpu/mtrr, perf/x86, x86/pci, x86/apic, x86/mce.
- Refine the patch descriptions and fix some typos.
- Remove unneeded vendor checking for patch x86/apic.
- Restore the user-visible pr_err string in patch x86/bugs.
v5->v6:
- Rebased on 4.19-rc3 and tested against it.
- Add Reviewed-by from Borislav Petkov for cacheinfo, smpboot, alternative and kvm.
- Rework the patch subjects and patch descriptions.
- Rework vendor checking for some patches to minimize the code modification.
v4->v5:
- Rebased on 4.19-rc1 and tested against it.
- Add Reviewed-by from Boris Ostrovsky for Xen.
- Rework EDAC patch without vendor checking for minimal modification.
v3->v4:
- Rebased on 4.18.3 and tested against it.
- Merge patches 05/17 perfctr and 10/17 events in v3 to patch 05/16 PMU for better patch function group.
- Add hygon_get_topology_early() in patch 01/16.
- Rework vendor checking and refine coding style.
- Add Acked-by from Bjorn Helgaas for pci.
- Add Acked-by from Rafael J. Wysocki for cpufreq and acpi.
v2->v3:
- Rebased on 4.18-rc8 and tested against it.
- Rework vendor checking codes to improve consistency.
v1->v2:
- Rebased on 4.18-rc6 and tested against it.
- Split the patch set to small series of patches.
- Rework patch descriptions.
- Create a separated arch/x86/kernel/cpu/hygon.c for Dhyana CPU initialization to reduce long-term maintenance effort.
Pu Wen (16):
x86/cpu: Create Hygon Dhyana architecture support file x86/cpu: Get cache info and setup cache cpumap for Hygon Dhyana x86/cpu/mtrr: Support TOP_MEM2 and get MTRR number x86/smpboot: SMP init no delay and not flush caches before sleep perf/x86: Add Hygon Dhyana support to PMU infrastructure x86/alternative: Init ideal_nops for Hygon Dhyana x86/pci: Add Hygon Dhyana support to PCI and north bridge x86/apic: Add Hygon Dhyana support to APIC
x86/bugs: Add mitigation to spectre and no meltdown for Hygon Dhyana x86/mce: Add Hygon Dhyana support to MCE infrastructure x86/kvm: Add Hygon Dhyana support to KVM infrastructure x86/xen: Add Hygon Dhyana support to Xen
ACPI, x86: Add Hygon Dhyana support
cpufreq, x86: Add Hygon Dhyana support
EDAC, amd64: Add Hygon Dhyana support
cpupower, x86: Add Hygon Dhyana support
MAINTAINERS | 6 + arch/x86/Kconfig.cpu | 14 + arch/x86/events/amd/core.c | 4 + arch/x86/events/amd/uncore.c | 20 +- arch/x86/events/core.c | 4 + arch/x86/include/asm/amd_nb.h | 3 + arch/x86/include/asm/cacheinfo.h | 1 + arch/x86/include/asm/kvm_emulate.h | 4 + arch/x86/include/asm/mce.h | 2 + arch/x86/include/asm/processor.h | 3 +- arch/x86/include/asm/virtext.h | 5 +- arch/x86/kernel/alternative.c | 4 + arch/x86/kernel/amd_nb.c | 47 ++- arch/x86/kernel/apic/apic.c | 7 + arch/x86/kernel/apic/probe_32.c | 1 + arch/x86/kernel/cpu/Makefile | 1 + arch/x86/kernel/cpu/bugs.c | 4 +- arch/x86/kernel/cpu/cacheinfo.c | 31 +- arch/x86/kernel/cpu/common.c | 1 + arch/x86/kernel/cpu/cpu.h | 1 + arch/x86/kernel/cpu/hygon.c | 411 +++++++++++++++++++++ arch/x86/kernel/cpu/mcheck/mce-severity.c | 3 +- arch/x86/kernel/cpu/mcheck/mce.c | 20 +- arch/x86/kernel/cpu/mtrr/cleanup.c | 3 +- arch/x86/kernel/cpu/mtrr/mtrr.c | 2 +- arch/x86/kernel/cpu/perfctr-watchdog.c | 2 + arch/x86/kernel/smpboot.c | 4 +- arch/x86/kvm/emulate.c | 11 +- arch/x86/pci/amd_bus.c | 6 +- arch/x86/xen/pmu.c | 12 +- drivers/acpi/acpi_pad.c | 1 + drivers/acpi/processor_idle.c | 1 + drivers/cpufreq/acpi-cpufreq.c | 5 + drivers/cpufreq/amd_freq_sensitivity.c | 9 +- drivers/edac/amd64_edac.c | 8 +- drivers/edac/mce_amd.c | 4 +- include/linux/pci_ids.h | 2 + tools/power/cpupower/utils/cpufreq-info.c | 6 +- tools/power/cpupower/utils/helpers/amd.c | 4 +- tools/power/cpupower/utils/helpers/cpuid.c | 8 +- tools/power/cpupower/utils/helpers/helpers.h | 2 +- tools/power/cpupower/utils/helpers/misc.c | 2 +- .../cpupower/utils/idle_monitor/mperf_monitor.c | 3 +- 43 files changed, 644 insertions(+), 48 deletions(-) create mode 100644 arch/x86/kernel/cpu/hygon.c
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 1/16 x86/cpu: Create Hygon Dhyana architecture support file
Add x86 architecture support for a new processor: Hygon Dhyana Family 18h. Carve out initialization code needed by Dhyana into a separate compilation unit.
To identify Hygon Dhyana CPU, add a new vendor type X86_VENDOR_HYGON for system recognition.
Since Dhyana uses AMD functionality to a large degree, select CPU_SUP_AMD which provides that functionality.
--- MAINTAINERS | 6 + arch/x86/Kconfig.cpu | 14 ++ arch/x86/include/asm/processor.h | 3 +- arch/x86/kernel/cpu/Makefile | 1 + arch/x86/kernel/cpu/hygon.c | 408 +++++++++++++++++++++++++++++++++++++++ 5 files changed, 431 insertions(+), 1 deletion(-) create mode 100644 arch/x86/kernel/cpu/hygon.c
MAINTAINERS +6 -0
@@ -6762,6 +6762,12 @@ S: Maintained F: mm/memory-failure.c F: mm/hwpoison-inject.c +HYGON PROCESSOR SUPPORT +M: Pu Wen <[email protected]> +L: [email protected] +S: Maintained +F: arch/x86/kernel/cpu/hygon.c + Hyper-V CORE AND DRIVERS M: "K. Y. Srinivasan" <[email protected]> M: Haiyang Zhang <[email protected]>
arch/x86/Kconfig.cpu +14 -0
@@ -426,6 +426,20 @@ config CPU_SUP_AMD If unsure, say N. +config CPU_SUP_HYGON + default y + bool "Support Hygon processors" if PROCESSOR_SELECT + select CPU_SUP_AMD + help + This enables detection, tunings and quirks for Hygon processors + + You need this enabled if you want your kernel to run on an + Hygon CPU. Disabling this option on other types of CPUs + makes the kernel a tiny bit smaller. Disabling it on an Hygon + CPU might render the kernel unbootable. + + If unsure, say N. + config CPU_SUP_CENTAUR default y bool "Support Centaur processors" if PROCESSOR_SELECT
arch/x86/include/asm/processor.h +2 -1
@@ -155,7 +155,8 @@ enum cpuid_regs_idx {#define X86_VENDOR_CENTAUR 5#define X86_VENDOR_TRANSMETA 7#define X86_VENDOR_NSC 8 -#define X86_VENDOR_NUM 9 +#define X86_VENDOR_HYGON 9 +#define X86_VENDOR_NUM 10#define X86_VENDOR_UNKNOWN 0xff
arch/x86/kernel/cpu/Makefile +1 -0
@@ -30,6 +30,7 @@ obj-$(CONFIG_X86_FEATURE_NAMES) += capflags.o powerflags.o obj-$(CONFIG_CPU_SUP_INTEL) += intel.o intel_pconfig.o obj-$(CONFIG_CPU_SUP_AMD) += amd.o +obj-$(CONFIG_CPU_SUP_HYGON) += hygon.o obj-$(CONFIG_CPU_SUP_CYRIX_32) += cyrix.o obj-$(CONFIG_CPU_SUP_CENTAUR) += centaur.o obj-$(CONFIG_CPU_SUP_TRANSMETA_32) += transmeta.o
arch/x86/kernel/cpu/hygon.c new file +408 -0
408 lines not on this page · raw message
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 2/16 x86/cpu: Get cache info and setup cache cpumap for Hygon Dhyana
In reply to Pu WenThe Hygon Dhyana CPU has topology extensions bit in CPUID. With this bit, the kernel can get the cache information. So add support in cpuid4_cache_lookup_regs() to get the correct cache size.
The Hygon Dhyana CPU also discovers num_cache_leaves via CPUID leaf 0x8000001d, so add support to it in find_num_cache_leaves().
Also add cacheinfo_hygon_init_llc_id() and init_hygon_cacheinfo() functions to initialize Dhyana cache info. Setup cache cpumap in the same way as AMD does.
--- arch/x86/include/asm/cacheinfo.h | 1 + arch/x86/kernel/cpu/cacheinfo.c | 31 +++++++++++++++++++++++++++++-- arch/x86/kernel/cpu/cpu.h | 1 + arch/x86/kernel/cpu/hygon.c | 3 +++ 4 files changed, 34 insertions(+), 2 deletions(-)
arch/x86/include/asm/cacheinfo.h +1 -0
@@ -3,5 +3,6 @@#define _ASM_X86_CACHEINFO_Hvoid cacheinfo_amd_init_llc_id (struct cpuinfo_x86 *c,int cpu,u8 node_id); +void cacheinfo_hygon_init_llc_id (struct cpuinfo_x86 *c,int cpu,u8 node_id);#endif /* _ASM_X86_CACHEINFO_H */
arch/x86/kernel/cpu/cacheinfo.c +29 -2
@@ -602,6 +602,10 @@ cpuid4_cache_lookup_regs(int index, struct _cpuid4_info_regs *this_leaf)else amd_cpuid4 (index, &eax, &ebx, &ecx);amd_init_l3_cache (this_leaf, index); + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { +cpuid_count (0x8000001d , index, &eax.full, + &ebx.full, &ecx.full, &edx); +amd_init_l3_cache (this_leaf, index); }else {cpuid_count (4 , index, &eax.full, &ebx.full, &ecx.full, &edx); }
@@ -625,7 +629,8 @@ static int find_num_cache_leaves(struct cpuinfo_x86 *c)union _cpuid4_leaf_eax cache_eax;int i =-1 ; -if (c->x86_vendor ==X86_VENDOR_AMD ) +if (c->x86_vendor ==X86_VENDOR_AMD || + c->x86_vendor ==X86_VENDOR_HYGON ) op =0x8000001d ;else op =4 ;
@@ -678,6 +683,22 @@ void cacheinfo_amd_init_llc_id(struct cpuinfo_x86 *c, int cpu, u8 node_id) } } +void cacheinfo_hygon_init_llc_id (struct cpuinfo_x86 *c,int cpu,u8 node_id) +{ +/* +* We may have multiple LLCs if L3 caches exist, so check if we +* have an L3 cache by looking at the L3 cache CPUID leaf. +*/ +if (!cpuid_edx (0x80000006 )) +return ; + +/* +* LLC is at the core complex level. +* Core complex ID is ApicId[3] for these processors. +*/ +per_cpu (cpu_llc_id, cpu) = c->apicid >>3 ; +} +void init_amd_cacheinfo (struct cpuinfo_x86 *c) {
@@ -691,6 +712,11 @@ void init_amd_cacheinfo(struct cpuinfo_x86 *c) } } +void init_hygon_cacheinfo (struct cpuinfo_x86 *c) +{ + num_cache_leaves =find_num_cache_leaves (c); +} +void init_intel_cacheinfo (struct cpuinfo_x86 *c) {/* Cache sizes */
@@ -913,7 +939,8 @@ static void __cache_cpumap_setup(unsigned int cpu, int index,int index_msb, i;struct cpuinfo_x86 *c = &cpu_data (cpu); -if (c->x86_vendor ==X86_VENDOR_AMD ) { +if (c->x86_vendor ==X86_VENDOR_AMD || + c->x86_vendor ==X86_VENDOR_HYGON ) {if (__cache_amd_cpumap_setup (cpu, index, base))return ; }
arch/x86/kernel/cpu/cpu.h +1 -0
@@ -54,6 +54,7 @@ extern u32 get_scattered_cpuid_leaf(unsigned int level, enum cpuid_regs_idx reg);extern void init_intel_cacheinfo (struct cpuinfo_x86 *c);extern void init_amd_cacheinfo (struct cpuinfo_x86 *c); +extern void init_hygon_cacheinfo (struct cpuinfo_x86 *c);extern void detect_num_cpu_cores (struct cpuinfo_x86 *c);extern int detect_extended_topology_early (struct cpuinfo_x86 *c);
arch/x86/kernel/cpu/hygon.c +3 -0
@@ -90,6 +90,7 @@ static void hygon_get_topology(struct cpuinfo_x86 *c)if (!err) c->x86_coreid_bits =get_count_order (c->x86_max_cores); +cacheinfo_hygon_init_llc_id (c, cpu, node_id); }else if (cpu_has (c,X86_FEATURE_NODEID_MSR )) {u64 value;
@@ -324,6 +325,8 @@ static void init_hygon(struct cpuinfo_x86 *c)hygon_get_topology (c);srat_detect_node (c); +init_hygon_cacheinfo (c); +if (cpu_has (c,X86_FEATURE_XMM2 )) {unsigned long long val;int ret;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 3/16 x86/cpu/mtrr: Support TOP_MEM2 and get MTRR number
In reply to Pu WenThe Hygon Dhyana CPU has a special magic MSR way to force WB for memory >4GB, and support TOP_MEM2. Therefore, it is necessary to add Hygon Dhyana support in amd_special_default_mtrr().
The number of variable MTRRs for Hygon is 2 as AMD's.
--- arch/x86/kernel/cpu/mtrr/cleanup.c | 3 ++- arch/x86/kernel/cpu/mtrr/mtrr.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-)
arch/x86/kernel/cpu/mtrr/cleanup.c +2 -1
@@ -831,7 +831,8 @@ int __init amd_special_default_mtrr(void) {u32 l, h; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return 0 ;if (boot_cpu_data.x86 <0xf )return 0 ;
arch/x86/kernel/cpu/mtrr/mtrr.c +1 -1
@@ -127,7 +127,7 @@ static void __init set_num_var_ranges(void)if (use_intel ())rdmsr (MSR_MTRRcap, config, dummy); -else if (is_cpu (AMD )) +else if (is_cpu (AMD ) ||is_cpu (HYGON )) config =2 ;else if (is_cpu (CYRIX ) ||is_cpu (CENTAUR )) config =8 ;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 4/16 x86/smpboot: SMP init no delay and not flush caches before sleep
In reply to Pu WenThe Hygon Dhyana CPU uses no delay in smp_quirk_init_udelay(), and returns in mwait_play_dead() as AMD does.
--- arch/x86/kernel/smpboot.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
arch/x86/kernel/smpboot.c +3 -1
@@ -676,6 +676,7 @@ static void __init smp_quirk_init_udelay(void)/* if modern processor, use no delay */ if (((boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) && (boot_cpu_data.x86 ==6 )) || + ((boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) && (boot_cpu_data.x86 >=0x18 )) || ((boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) && (boot_cpu_data.x86 >=0xF ))) { init_udelay =0 ;return ;
@@ -1592,7 +1593,8 @@ static inline void mwait_play_dead(void)void *mwait_ptr;int i; -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON )return ;if (!this_cpu_has (X86_FEATURE_MWAIT ))return ;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 5/16 perf/x86: Add Hygon Dhyana support to PMU infrastructure
In reply to Pu WenThe PMU architecture for Hygon Dhyana CPU is similar to the AMD Family 17h one. To support Hygon Dhyana PMU, call amd_pmu_init() to share AMD PMU initialization flow, and change the PMU name to "HYGON".
The Hygon Dhyana CPU supports both legacy and extension PMC MSRs(perf counter registers and event selection registers), so add Hygon Dhyana support in the similar way as AMD does.
--- arch/x86/events/amd/core.c | 4 ++++ arch/x86/events/amd/uncore.c | 20 +++++++++++++------- arch/x86/events/core.c | 4 ++++ arch/x86/kernel/cpu/perfctr-watchdog.c | 2 ++ 4 files changed, 23 insertions(+), 7 deletions(-)
arch/x86/events/amd/core.c +4 -0
@@ -669,6 +669,10 @@ static int __init amd_core_pmu_init(void)* We fallback to using default amd_get_event_constraints. */ break ; +case 0x18 : +pr_cont ("Fam18h " ); +/* Using default amd_get_event_constraints. */ +break ;default :pr_err ("core perfctr but no constraints; unknown hardware!\n" );return -ENODEV ;
arch/x86/events/amd/uncore.c +13 -7
@@ -507,17 +507,19 @@ static int __init amd_uncore_init(void) {int ret = -ENODEV ; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return -ENODEV ;if (!boot_cpu_has (X86_FEATURE_TOPOEXT ))return -ENODEV ; -if (boot_cpu_data.x86 ==0x17 ) { +if (boot_cpu_data.x86 ==0x17 || boot_cpu_data.x86 ==0x18 ) {/* -* For F17h, the Northbridge counters are repurposed as Data -* Fabric counters. Also, L3 counters are supported too. The PMUs -* are exported based on family as either L2 or L3 and NB or DF. +* For F17h or F18h, the Northbridge counters are +* repurposed as Data Fabric counters. Also, L3 +* counters are supported too. The PMUs are exported +* based on family as either L2 or L3 and NB or DF. */ num_counters_nb =NUM_COUNTERS_NB ; num_counters_llc =NUM_COUNTERS_L3 ;
@@ -547,7 +549,9 @@ static int __init amd_uncore_init(void)if (ret) goto fail_nb; -pr_info ("AMD NB counters detected\n" ); +pr_info ("%s NB counters detected\n" , + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ? +"HYGON" :"AMD" ); ret =0 ; }
@@ -561,7 +565,9 @@ static int __init amd_uncore_init(void)if (ret) goto fail_llc; -pr_info ("AMD LLC counters detected\n" ); +pr_info ("%s LLC counters detected\n" , + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ? +"HYGON" :"AMD" ); ret =0 ; }
arch/x86/events/core.c +4 -0
@@ -1776,6 +1776,10 @@ static int __init init_hw_perf_events(void)case X86_VENDOR_AMD : err =amd_pmu_init ();break ; +case X86_VENDOR_HYGON : + err =amd_pmu_init (); + x86_pmu.name ="HYGON" ; +break ;default : err = -ENOTSUPP ; }
arch/x86/kernel/cpu/perfctr-watchdog.c +2 -0
@@ -46,6 +46,7 @@ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr) {/* returns the bit offset of the performance counter register */ switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :if (msr >=MSR_F15H_PERF_CTR )return (msr -MSR_F15H_PERF_CTR ) >>1 ;
@@ -74,6 +75,7 @@ static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr) {/* returns the bit offset of the event selection register */ switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :if (msr >=MSR_F15H_PERF_CTL )return (msr -MSR_F15H_PERF_CTL ) >>1 ;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 6/16 x86/alternative: Init ideal_nops for Hygon Dhyana
In reply to Pu WenThe ideal_nops for Hygon Dhyana CPU should be p6_nops.
--- arch/x86/kernel/alternative.c | 4 ++++ 1 file changed, 4 insertions(+)
arch/x86/kernel/alternative.c +4 -0
@@ -222,6 +222,10 @@ void __init arch_init_ideal_nops(void) }break ; +case X86_VENDOR_HYGON : + ideal_nops = p6_nops; +return ; +case X86_VENDOR_AMD :if (boot_cpu_data.x86 >0xf ) { ideal_nops = p6_nops;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 7/16 x86/pci: Add Hygon Dhyana support to PCI and north bridge
In reply to Pu WenAs Hygon registered its PCI Vendor ID as a new one 0x1d94, and there are PCI Devices 0x1450/0x1463/0x1464 for Host bridge on Hygon Dhyana platform, so add Hygon Dhyana support to the PCI and north bridge subsystem by using the code path of AMD family 17h.
To prevent further checking PCI device ids which cannot happen both on Hygon and Intel platform, the function amd_gart_present should return if it's not a AMD CPU.
--- arch/x86/include/asm/amd_nb.h | 3 +++ arch/x86/kernel/amd_nb.c | 47 +++++++++++++++++++++++++++++++++++++------ arch/x86/pci/amd_bus.c | 6 ++++-- include/linux/pci_ids.h | 2 ++ 4 files changed, 50 insertions(+), 8 deletions(-)
arch/x86/include/asm/amd_nb.h +3 -0
@@ -103,6 +103,9 @@ static inline u16 amd_pci_dev_to_node_id(struct pci_dev *pdev)static inline bool amd_gart_present (void ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +return false; +/* GART present only on Fam15h, upto model 0fh */ if (boot_cpu_data.x86 ==0xf || boot_cpu_data.x86 ==0x10 || (boot_cpu_data.x86 ==0x15 && boot_cpu_data.x86_model <0x10 ))
arch/x86/kernel/amd_nb.c +41 -6
86 lines not on this page · raw message
arch/x86/pci/amd_bus.c +4 -2
@@ -93,7 +93,8 @@ static int __init early_root_info_init(void) vendor = id &0xffff ; device = (id>>16 ) &0xffff ; -if (vendor !=PCI_VENDOR_ID_AMD ) +if (vendor !=PCI_VENDOR_ID_AMD && + vendor !=PCI_VENDOR_ID_HYGON )continue ;if (hb_probes[i].device == device) {
@@ -390,7 +391,8 @@ static int __init pci_io_ecs_init(void)static int __init amd_postcore_init (void ) { -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return 0 ;early_root_info_init ();
include/linux/pci_ids.h +2 -0
@@ -2561,6 +2561,8 @@#define PCI_VENDOR_ID_AMAZON 0x1d0f +#define PCI_VENDOR_ID_HYGON 0x1d94 +#define PCI_VENDOR_ID_TEKRAM 0x1de1#define PCI_DEVICE_ID_TEKRAM_DC290 0xdc29
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 8/16 x86/apic: Add Hygon Dhyana support to APIC
In reply to Pu WenAdd Hygon Dhyana support to the APIC subsystem as it uses modern APIC. When running on 32 bit mode, bigsmp should be enabled if there are more than 8 cores online.
--- arch/x86/kernel/apic/apic.c | 7 +++++++ arch/x86/kernel/apic/probe_32.c | 1 + 2 files changed, 8 insertions(+)
arch/x86/kernel/apic/apic.c +7 -0
@@ -224,6 +224,11 @@ static int modern_apic(void)if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD && boot_cpu_data.x86 >=0xf )return 1 ; + +/* Hygon systems use modern APIC */ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) +return 1 ; +return lapic_get_version () >=0x14 ; }
@@ -1912,6 +1917,8 @@ static int __init detect_init_APIC(void) (boot_cpu_data.x86 >= 15))break ; goto no_apic; +case X86_VENDOR_HYGON : +break ;case X86_VENDOR_INTEL :if (boot_cpu_data.x86 ==6 || boot_cpu_data.x86 ==15 || (boot_cpu_data.x86 ==5 &&boot_cpu_has (X86_FEATURE_APIC )))
arch/x86/kernel/apic/probe_32.c +1 -0
@@ -185,6 +185,7 @@ void __init default_setup_apic_routing(void)break ; }/* If P4 and above fall through */ +case X86_VENDOR_HYGON :case X86_VENDOR_AMD : def_to_bigsmp =1 ; }
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 9/16 x86/bugs: Add mitigation to spectre and no meltdown for Hygon Dhyana
In reply to Pu WenThe Hygon Dhyana CPU has the same speculative execution as AMD family 17h, so share AMD spectre mitigation code for Hygon Dhyana.
Also Hygon Dhyana is not affected by meltdown vulnerability, so add exception for it.
--- arch/x86/kernel/cpu/bugs.c | 4 +++- arch/x86/kernel/cpu/common.c | 1 + 2 files changed, 4 insertions(+), 1 deletion(-)
arch/x86/kernel/cpu/bugs.c +3 -1
@@ -312,6 +312,7 @@ static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void) }if (cmd ==SPECTRE_V2_CMD_RETPOLINE_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON && boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) {pr_err ("retpoline,amd selected but CPU is not AMD. Switching to AUTO select\n" );return SPECTRE_V2_CMD_AUTO ;
@@ -371,7 +372,8 @@ static void __init spectre_v2_select_mitigation(void)return ; retpoline_auto: -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { retpoline_amd:if (!boot_cpu_has (X86_FEATURE_LFENCE_RDTSC )) {pr_err ("Spectre mitigation: LFENCE not serializing, switching to generic retpoline\n" );
arch/x86/kernel/cpu/common.c +1 -0
@@ -963,6 +963,7 @@ static const __initconst struct x86_cpu_id cpu_no_speculation[] = {static const __initconst struct x86_cpu_id cpu_no_meltdown[] = { {X86_VENDOR_AMD }, + {X86_VENDOR_HYGON }, {} };
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 10/16 x86/mce: Add Hygon Dhyana support to MCE infrastructure
In reply to Pu WenThe machine check architecture for Hygon Dhyana CPU is similar to the AMD family 17h one. To make MCE working on Hygon platform, add vendor checking for Hygon Dhyana to share the code path of AMD family 17h.
--- arch/x86/include/asm/mce.h | 2 ++ arch/x86/kernel/cpu/mcheck/mce-severity.c | 3 ++- arch/x86/kernel/cpu/mcheck/mce.c | 20 +++++++++++++++----- 3 files changed, 19 insertions(+), 6 deletions(-)
arch/x86/include/asm/mce.h +2 -0
@@ -214,6 +214,8 @@ static inline void mce_amd_feature_init(struct cpuinfo_x86 *c) { }static inline int umc_normaddr_to_sysaddr (u64 norm_addr,u16 nid,u8 umc,u64 *sys_addr) {return -EINVAL ; };#endif +static inline void mce_hygon_feature_init (struct cpuinfo_x86 *c) {return mce_amd_feature_init (c); } +int mce_available (struct cpuinfo_x86 *c);bool mce_is_memory_error (struct mce *m);
arch/x86/kernel/cpu/mcheck/mce-severity.c +2 -1
@@ -336,7 +336,8 @@ int (*mce_severity)(struct mce *m, int tolerant, char **msg, bool is_excp) =void __init mcheck_vendor_init_severity (void ) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) mce_severity = mce_severity_amd; }
arch/x86/kernel/cpu/mcheck/mce.c +15 -5
@@ -270,7 +270,7 @@ static void print_mce(struct mce *m) {__print_mce (m); -if (m->cpuvendor !=X86_VENDOR_AMD ) +if (m->cpuvendor !=X86_VENDOR_AMD && m->cpuvendor !=X86_VENDOR_HYGON )pr_emerg_ratelimited (HW_ERR "Run the above through 'mcelog --ascii'\n" ); }
@@ -508,9 +508,9 @@ static int mce_usable_address(struct mce *m)bool mce_is_memory_error (struct mce *m) { -if (m->cpuvendor ==X86_VENDOR_AMD ) { +if (m->cpuvendor ==X86_VENDOR_AMD || + m->cpuvendor ==X86_VENDOR_HYGON ) {return amd_mce_is_memory_error (m); - }else if (m->cpuvendor ==X86_VENDOR_INTEL ) {/* * Intel SDM Volume 3B - 15.9.2 Compound Error Codes
@@ -539,6 +539,9 @@ static bool mce_is_correctable(struct mce *m)if (m->cpuvendor ==X86_VENDOR_AMD && m->status &MCI_STATUS_DEFERRED )return false; +if (m->cpuvendor ==X86_VENDOR_HYGON && m->status &MCI_STATUS_DEFERRED ) +return false; +if (m->status &MCI_STATUS_UC )return false;
@@ -1705,7 +1708,7 @@ static int __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c)*/ static void __mcheck_cpu_init_early (struct cpuinfo_x86 *c) { -if (c->x86_vendor ==X86_VENDOR_AMD ) { +if (c->x86_vendor ==X86_VENDOR_AMD || c->x86_vendor ==X86_VENDOR_HYGON ) { mce_flags.overflow_recov = !!cpu_has (c,X86_FEATURE_OVERFLOW_RECOV ); mce_flags.succor = !!cpu_has (c,X86_FEATURE_SUCCOR ); mce_flags.smca = !!cpu_has (c,X86_FEATURE_SMCA );
@@ -1746,6 +1749,11 @@ static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)mce_amd_feature_init (c);break ; } + +case X86_VENDOR_HYGON : +mce_hygon_feature_init (c); +break ; +case X86_VENDOR_CENTAUR :mce_centaur_feature_init (c);break ;
@@ -1971,12 +1979,14 @@ static void mce_disable_error_reporting(void)static void vendor_disable_error_reporting (void ) {/* -* Don't clear on Intel or AMD CPUs. Some of these MSRs are socket-wide. +* Don't clear on Intel or AMD or Hygon CPUs. Some of these MSRs +* are socket-wide. * Disabling them for just a single offlined CPU is bad, since it will * inhibit reporting for all shared resources on the socket like the * last level cache (LLC), the integrated memory controller (iMC), etc. */ if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON || boot_cpu_data.x86_vendor ==X86_VENDOR_AMD )return ;
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 11/16 x86/kvm: Add Hygon Dhyana support to KVM infrastructure
In reply to Pu WenThe Hygon Dhyana CPU has the SVM feature as AMD family 17h does. So enable the KVM infrastructure support to it.
--- arch/x86/include/asm/kvm_emulate.h | 4 ++++ arch/x86/include/asm/virtext.h | 5 +++-- arch/x86/kvm/emulate.c | 11 ++++++++++- 3 files changed, 17 insertions(+), 3 deletions(-)
arch/x86/include/asm/kvm_emulate.h +4 -0
@@ -364,6 +364,10 @@ struct x86_emulate_ctxt {#define X86EMUL_CPUID_VENDOR_AMDisbetterI_ecx 0x21726574#define X86EMUL_CPUID_VENDOR_AMDisbetterI_edx 0x74656273 +#define X86EMUL_CPUID_VENDOR_HygonGenuine_ebx 0x6f677948 +#define X86EMUL_CPUID_VENDOR_HygonGenuine_ecx 0x656e6975 +#define X86EMUL_CPUID_VENDOR_HygonGenuine_edx 0x6e65476e +#define X86EMUL_CPUID_VENDOR_GenuineIntel_ebx 0x756e6547#define X86EMUL_CPUID_VENDOR_GenuineIntel_ecx 0x6c65746e#define X86EMUL_CPUID_VENDOR_GenuineIntel_edx 0x49656e69
arch/x86/include/asm/virtext.h +3 -2
@@ -83,9 +83,10 @@ static inline void cpu_emergency_vmxoff(void)*/ static inline int cpu_has_svm (const char **msg) { -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON ) {if (msg) - *msg ="not amd" ; + *msg ="not amd or hygon" ;return 0 ; }
arch/x86/kvm/emulate.c +10 -1
@@ -2711,7 +2711,16 @@ static bool em_syscall_is_enabled(struct x86_emulate_ctxt *ctxt) edx == X86EMUL_CPUID_VENDOR_AMDisbetterI_edx)return true; -/* default: (not Intel, not AMD), apply Intel's stricter rules... */ +/* Hygon ("HygonGenuine") */ +if (ebx == X86EMUL_CPUID_VENDOR_HygonGenuine_ebx && + ecx == X86EMUL_CPUID_VENDOR_HygonGenuine_ecx && + edx == X86EMUL_CPUID_VENDOR_HygonGenuine_edx) +return true; + +/* +* default: (not Intel, not AMD, not Hygon), apply Intel's +* stricter rules... +*/ return false; }
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 12/16 x86/xen: Add Hygon Dhyana support to Xen
In reply to Pu WenTo make Xen works functionally on Hygon platform, reuse AMD's Xen support code path for Hygon Dhyana CPU.
There are six core performance events counters per thread, so there are six MSRs for these counters(0-5). Also there are four legacy PMC MSRs, they are alias of the counters(0-3).
In this version of kernel Hygon use the legacy and safe version of MSR access. It works fine when VPMU enabled in Xen on Hygon platform by testing with perf.
--- arch/x86/xen/pmu.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
arch/x86/xen/pmu.c +9 -3
@@ -90,6 +90,12 @@ static void xen_pmu_arch_init(void) k7_counters_mirrored =0 ;break ; } + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + amd_num_counters =F10H_NUM_COUNTERS ; + amd_counters_base =MSR_K7_PERFCTR0 ; + amd_ctrls_base =MSR_K7_EVNTSEL0 ; + amd_msr_step =1 ; + k7_counters_mirrored =0 ; }else {uint32_t eax, ebx, ecx, edx;
@@ -285,7 +291,7 @@ static bool xen_amd_pmu_emulate(unsigned int msr, u64 *val, bool is_read)bool pmu_msr_read (unsigned int msr,uint64_t *val,int *err) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL ) {if (is_amd_pmu_msr (msr)) {if (!xen_amd_pmu_emulate (msr, val,1 )) *val =native_read_msr_safe (msr, err);
@@ -308,7 +314,7 @@ bool pmu_msr_write(unsigned int msr, uint32_t low, uint32_t high, int *err) {uint64_t val = ((uint64_t )high <<32 ) | low; -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL ) {if (is_amd_pmu_msr (msr)) {if (!xen_amd_pmu_emulate (msr, &val,0 )) *err =native_write_msr_safe (msr, low, high);
@@ -379,7 +385,7 @@ static unsigned long long xen_intel_read_pmc(int counter)unsigned long long xen_read_pmc (int counter) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL )return xen_amd_read_pmc (counter);else return xen_intel_read_pmc (counter);
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 12/16 x86/xen: Add Hygon Dhyana support to Xen
In reply to Pu WenTo make Xen works functionally on Hygon platform, reuse AMD's Xen support code path for Hygon Dhyana CPU.
There are six core performance events counters per thread, so there are six MSRs for these counters(0-5). Also there are four legacy PMC MSRs, they are alias of the counters(0-3).
In this version of kernel Hygon use the legacy and safe version of MSR access. It works fine when VPMU enabled in Xen on Hygon platform by testing with perf.
--- arch/x86/xen/pmu.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
arch/x86/xen/pmu.c +9 -3
@@ -90,6 +90,12 @@ static void xen_pmu_arch_init(void) k7_counters_mirrored =0 ;break ; } + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + amd_num_counters =F10H_NUM_COUNTERS ; + amd_counters_base =MSR_K7_PERFCTR0 ; + amd_ctrls_base =MSR_K7_EVNTSEL0 ; + amd_msr_step =1 ; + k7_counters_mirrored =0 ; }else {uint32_t eax, ebx, ecx, edx;
@@ -285,7 +291,7 @@ static bool xen_amd_pmu_emulate(unsigned int msr, u64 *val, bool is_read)bool pmu_msr_read (unsigned int msr,uint64_t *val,int *err) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL ) {if (is_amd_pmu_msr (msr)) {if (!xen_amd_pmu_emulate (msr, val,1 )) *val =native_read_msr_safe (msr, err);
@@ -308,7 +314,7 @@ bool pmu_msr_write(unsigned int msr, uint32_t low, uint32_t high, int *err) {uint64_t val = ((uint64_t )high <<32 ) | low; -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL ) {if (is_amd_pmu_msr (msr)) {if (!xen_amd_pmu_emulate (msr, &val,0 )) *err =native_write_msr_safe (msr, low, high);
@@ -379,7 +385,7 @@ static unsigned long long xen_intel_read_pmc(int counter)unsigned long long xen_read_pmc (int counter) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL )return xen_amd_read_pmc (counter);else return xen_intel_read_pmc (counter);
Signature · 7 lines
2.7.4 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
Pu Wen
maintainerPATCH v8 13/16 ACPI, x86: Add Hygon Dhyana support
In reply to Pu WenThe Hygon Dhyana CPU has NONSTOP TSC feature, so enable the ACPI driver support to it.
--- drivers/acpi/acpi_pad.c | 1 + drivers/acpi/processor_idle.c | 1 + 2 files changed, 2 insertions(+)
drivers/acpi/acpi_pad.c +1 -0
@@ -70,6 +70,7 @@ static void power_saving_mwait_init(void)#if defined(CONFIG_X86 )switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :case X86_VENDOR_INTEL :/*
drivers/acpi/processor_idle.c +1 -0
@@ -205,6 +205,7 @@ static void lapic_timer_state_broadcast(struct acpi_processor *pr,static void tsc_check_state (int state) {switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :case X86_VENDOR_INTEL :case X86_VENDOR_CENTAUR :
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 14/16 cpufreq, x86: Add Hygon Dhyana support
In reply to Pu WenThe Hygon Dhyana CPU supports ACPI P-States feature, and there is SMBus device(PCI device ID 0x790b) on Hygon platform, so add Hygon Dhyana support to the cpufreq driver by using the code path of AMD family 17h.
--- drivers/cpufreq/acpi-cpufreq.c | 5 +++++ drivers/cpufreq/amd_freq_sensitivity.c | 9 +++++++-- 2 files changed, 12 insertions(+), 2 deletions(-)
drivers/cpufreq/acpi-cpufreq.c +5 -0
@@ -61,6 +61,7 @@ enum {#define INTEL_MSR_RANGE (0xffff)#define AMD_MSR_RANGE (0x7) +#define HYGON_MSR_RANGE (0x7)#define MSR_K7_HWCR_CPB_DIS (1ULL << 25)
@@ -95,6 +96,7 @@ static bool boost_state(unsigned int cpu)rdmsr_on_cpu (cpu,MSR_IA32_MISC_ENABLE , &lo, &hi); msr = lo | ((u64 )hi <<32 );return !(msr &MSR_IA32_MISC_ENABLE_TURBO_DISABLE ); +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :rdmsr_on_cpu (cpu,MSR_K7_HWCR , &lo, &hi); msr = lo | ((u64 )hi <<32 );
@@ -113,6 +115,7 @@ static int boost_set_msr(bool enable) msr_addr =MSR_IA32_MISC_ENABLE ; msr_mask =MSR_IA32_MISC_ENABLE_TURBO_DISABLE ;break ; +case X86_VENDOR_HYGON :case X86_VENDOR_AMD : msr_addr =MSR_K7_HWCR ; msr_mask =MSR_K7_HWCR_CPB_DIS ;
@@ -225,6 +228,8 @@ static unsigned extract_msr(struct cpufreq_policy *policy, u32 msr)if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) msr &=AMD_MSR_RANGE ; +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + msr &=HYGON_MSR_RANGE ;else msr &=INTEL_MSR_RANGE ;
drivers/cpufreq/amd_freq_sensitivity.c +7 -2
@@ -111,11 +111,16 @@ static int __init amd_freq_sensitivity_init(void) {u64 val;struct pci_dev *pcidev; +unsigned int pci_vendor; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) + pci_vendor =PCI_VENDOR_ID_AMD ; +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + pci_vendor =PCI_VENDOR_ID_HYGON ; +else return -ENODEV ; - pcidev =pci_get_device (PCI_VENDOR_ID_AMD , + pcidev =pci_get_device (pci_vendor,PCI_DEVICE_ID_AMD_KERNCZ_SMBUS ,NULL );if (!pcidev) {
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 15/16 EDAC, amd64: Add Hygon Dhyana support
In reply to Pu WenTo make AMD64 EDAC and MCE drivers working on Hygon platform, add support for Hygon Dhyana CPU by using the code path of AMD family 17h.
--- drivers/edac/amd64_edac.c | 8 +++++++- drivers/edac/mce_amd.c | 4 +++- 2 files changed, 10 insertions(+), 2 deletions(-)
drivers/edac/amd64_edac.c +7 -1
@@ -211,7 +211,7 @@ static int __set_scrub_rate(struct amd64_pvt *pvt, u32 new_bw, u32 min_rate) scrubval = scrubrates[i].scrubval; -if (pvt->fam ==0x17 ) { +if (pvt->fam ==0x17 || pvt->fam ==0x18 ) {__f17h_set_scrubval (pvt, scrubval); }else if (pvt->fam ==0x15 && pvt->model ==0x60 ) {f15h_select_dct (pvt,0 );
@@ -264,6 +264,7 @@ static int get_scrub_rate(struct mem_ctl_info *mci)break ;case 0x17 : +case 0x18 :amd64_read_pci_cfg (pvt->F6,F17H_SCR_BASE_ADDR , &scrubval);if (scrubval &BIT (0 )) {amd64_read_pci_cfg (pvt->F6,F17H_SCR_LIMIT_ADDR , &scrubval);
@@ -1044,6 +1045,7 @@ static void determine_memory_type(struct amd64_pvt *pvt) goto ddr3;case 0x17 : +case 0x18 :if ((pvt->umc[0 ].dimm_cfg | pvt->umc[1 ].dimm_cfg) &BIT (5 )) pvt->dram_type =MEM_LRDDR4 ;else if ((pvt->umc[0 ].dimm_cfg | pvt->umc[1 ].dimm_cfg) &BIT (4 ))
@@ -3188,8 +3190,11 @@ static struct amd64_family_type *per_family_init(struct amd64_pvt *pvt)break ;case 0x17 : +case 0x18 : fam_type = &family_types[F17_CPUS ]; pvt->ops = &family_types[F17_CPUS ].ops; +if (pvt->fam ==0x18 ) + family_types[F17_CPUS ].ctl_name ="F18h" ;break ;default :
@@ -3428,6 +3433,7 @@ static const struct x86_cpu_id amd64_cpuids[] = { { X86_VENDOR_AMD, 0x15, X86_MODEL_ANY, X86_FEATURE_ANY, 0 }, { X86_VENDOR_AMD, 0x16, X86_MODEL_ANY, X86_FEATURE_ANY, 0 }, { X86_VENDOR_AMD, 0x17, X86_MODEL_ANY, X86_FEATURE_ANY, 0 }, + { X86_VENDOR_HYGON, 0x18, X86_MODEL_ANY, X86_FEATURE_ANY, 0 }, { } };MODULE_DEVICE_TABLE (x86cpu, amd64_cpuids);
drivers/edac/mce_amd.c +3 -1
@@ -1059,7 +1059,8 @@ static int __init mce_amd_init(void) {struct cpuinfo_x86 *c = &boot_cpu_data; -if (c->x86_vendor !=X86_VENDOR_AMD ) +if (c->x86_vendor !=X86_VENDOR_AMD && + c->x86_vendor !=X86_VENDOR_HYGON )return -ENODEV ; fam_ops =kzalloc (sizeof (struct amd_decoder_ops ),GFP_KERNEL );
@@ -1113,6 +1114,7 @@ static int __init mce_amd_init(void)break ;case 0x17 : +case 0x18 : xec_mask =0x3f ;if (!boot_cpu_has (X86_FEATURE_SMCA )) {printk (KERN_WARNING "Decoding supported only on Scalable MCA processors.\n" );
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH v8 16/16 cpupower, x86: Add Hygon Dhyana support
In reply to Pu WenTool cpupower is useful to get CPU frequency information and monitor power stats on Hygon Dhyana platform. So add Hygon Dhyana support to it by checking vendor and family to share the code path of AMD family 17h.
--- tools/power/cpupower/utils/cpufreq-info.c | 6 ++++-- tools/power/cpupower/utils/helpers/amd.c | 4 ++-- tools/power/cpupower/utils/helpers/cpuid.c | 8 +++++--- tools/power/cpupower/utils/helpers/helpers.h | 2 +- tools/power/cpupower/utils/helpers/misc.c | 2 +- tools/power/cpupower/utils/idle_monitor/mperf_monitor.c | 3 ++- 6 files changed, 15 insertions(+), 10 deletions(-)
tools/power/cpupower/utils/cpufreq-info.c +4 -2
@@ -170,6 +170,7 @@ static int get_boost_mode(unsigned int cpu)unsigned long pstates[MAX_HW_PSTATES ] = {0 ,};if (cpupower_cpu_info.vendor !=X86_VENDOR_AMD && + cpupower_cpu_info.vendor !=X86_VENDOR_HYGON && cpupower_cpu_info.vendor !=X86_VENDOR_INTEL )return 0 ;
@@ -190,8 +191,9 @@ static int get_boost_mode(unsigned int cpu)printf (_ (" Supported: %s\n" ), support ?_ ("yes" ) :_ ("no" ));printf (_ (" Active: %s\n" ), active ?_ ("yes" ) :_ ("no" )); -if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD && - cpupower_cpu_info.family >=0x10 ) { +if ((cpupower_cpu_info.vendor ==X86_VENDOR_AMD && + cpupower_cpu_info.family >=0x10 ) || + cpupower_cpu_info.vendor ==X86_VENDOR_HYGON ) { ret =decode_pstates (cpu, cpupower_cpu_info.family, b_states, pstates, &pstate_no);if (ret)
tools/power/cpupower/utils/helpers/amd.c +2 -2
@@ -45,7 +45,7 @@ static int get_did(int family, union msr_pstate pstate)if (family ==0x12 ) t = pstate.val &0xf ; -else if (family ==0x17 ) +else if (family ==0x17 || family ==0x18 ) t = pstate.fam17h_bits.did;else t = pstate.bits.did;
@@ -59,7 +59,7 @@ static int get_cof(int family, union msr_pstate pstate)int fid, did, cof; did =get_did (family, pstate); -if (family ==0x17 ) { +if (family ==0x17 || family ==0x18 ) { fid = pstate.fam17h_bits.fid; cof =200 * fid / did; }else {
tools/power/cpupower/utils/helpers/cpuid.c +5 -3
@@ -8,7 +8,7 @@#include "helpers/helpers.h" static const char *cpu_vendor_table[X86_VENDOR_MAX ] = { -"Unknown" ,"GenuineIntel" ,"AuthenticAMD" , +"Unknown" ,"GenuineIntel" ,"AuthenticAMD" ,"HygonGenuine" , };#if defined(__i386__) || defined(__x86_64__)
@@ -109,6 +109,7 @@ int get_cpu_info(struct cpupower_cpu_info *cpu_info)fclose (fp);/* Get some useful CPU capabilities from cpuid */ if (cpu_info->vendor !=X86_VENDOR_AMD && + cpu_info->vendor !=X86_VENDOR_HYGON && cpu_info->vendor !=X86_VENDOR_INTEL )return ret;
@@ -124,8 +125,9 @@ int get_cpu_info(struct cpupower_cpu_info *cpu_info)if (cpuid_level >=6 && (cpuid_ecx (6 ) &0x1 )) cpu_info->caps |=CPUPOWER_CAP_APERF ; -/* AMD Boost state enable/disable register */ -if (cpu_info->vendor ==X86_VENDOR_AMD ) { +/* AMD or Hygon Boost state enable/disable register */ +if (cpu_info->vendor ==X86_VENDOR_AMD || + cpu_info->vendor ==X86_VENDOR_HYGON ) {if (ext_cpuid_level >=0x80000007 && (cpuid_edx (0x80000007 ) & (1 <<9 ))) cpu_info->caps |=CPUPOWER_CAP_AMD_CBP ;
tools/power/cpupower/utils/helpers/helpers.h +1 -1
@@ -61,7 +61,7 @@ extern int be_verbose;/* cpuid and cpuinfo helpers **************************/ enum cpupower_cpu_vendor {X86_VENDOR_UNKNOWN =0 ,X86_VENDOR_INTEL , -X86_VENDOR_AMD ,X86_VENDOR_MAX }; +X86_VENDOR_AMD ,X86_VENDOR_HYGON ,X86_VENDOR_MAX };#define CPUPOWER_CAP_INV_TSC 0x00000001#define CPUPOWER_CAP_APERF 0x00000002
tools/power/cpupower/utils/helpers/misc.c +1 -1
@@ -26,7 +26,7 @@ int cpufreq_has_boost_support(unsigned int cpu, int *support, int *active,* has Hardware determined variable increments instead. */ -if (cpu_info.family ==0x17 ) { +if (cpu_info.family ==0x17 || cpu_info.family ==0x18 ) {if (!read_msr (cpu,MSR_AMD_HWCR , &val)) {if (!(val &CPUPOWER_AMD_CPBDIS )) *active =1 ;
tools/power/cpupower/utils/idle_monitor/mperf_monitor.c +2 -1
@@ -241,7 +241,8 @@ static int init_maxfreq_mode(void)if (!(cpupower_cpu_info.caps &CPUPOWER_CAP_INV_TSC )) goto use_sysfs; -if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD ) { +if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD || + cpupower_cpu_info.vendor ==X86_VENDOR_HYGON ) {/* MSR_AMD_HWCR tells us whether TSC runs at P0/mperf * freq. * A test whether hwcr is accessable/available would be:
Signature · 1 line
2.7.4
Borislav Petkov
maintainerRe: [PATCH v8 07/16] x86/pci: Add Hygon Dhyana support to PCI and north bridge
In reply to Pu WenQuoted 5 lines · Pu Wen
As Hygon registered its PCI Vendor ID as a new one
0x1d94, and there are PCI Devices 0x1450/0x1463/0x1464 for Host bridge on Hygon Dhyana platform, so add Hygon Dhyana support to the PCI and north bridge subsystem by using the code path of AMD family 17h.
Quoted 7 lines
To prevent further checking PCI device ids which cannot happen both on Hygon and Intel platform, the function
amd_gart_presentshould return if it's not a AMD CPU.Signed-off-by Pu Wen <[email protected]>Reviewed-by Borislav Petkov <[email protected]>
A note for your future submissions: if you rework a patch in a non-trivial way, you *must* drop Reviewed-by/Acked-by tags because they are not valid anymore.
In this case I guess Bjorn's tag can stay because he's acking the pci_ids.h addition but mine definitely must go.
Thx.
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Pu Wen
maintainerRe: [PATCH v8 07/16] x86/pci: Add Hygon Dhyana support to PCI and north bridge
Quoted 10 lines · Borislav Petkov
Quoted 3 lines
Signed-off-by Pu Wen <[email protected]>Reviewed-by Borislav Petkov <[email protected]>A note for your future submissions: if you rework a patch in a non-trivial way, you *must* drop Reviewed-by/Acked-by tags because they are not valid anymore.
In this case I guess Bjorn's tag can stay because he's acking the
pci_ids.haddition but mine definitely must go.
Thanks for the note. I'm sorry for not removing the Reviewed-by tag. Is this revision still OK by you? :)
Signature · 2 lines
Regards, Pu Wen
Borislav Petkov
maintainerRe: [PATCH v8 07/16] x86/pci: Add Hygon Dhyana support to PCI and north bridge
In reply to Pu WenQuoted 24 lines · Pu Wen
As Hygon registered its PCI Vendor ID as a new one
0x1d94, and there are PCI Devices 0x1450/0x1463/0x1464 for Host bridge on Hygon Dhyana platform, so add Hygon Dhyana support to the PCI and north bridge subsystem by using the code path of AMD family 17h.To prevent further checking PCI device ids which cannot happen both on Hygon and Intel platform, the function
amd_gart_presentshould return if it's not a AMD CPU.Signed-off-by Pu Wen <[email protected]>Reviewed-by Borislav Petkov <[email protected]>--- arch/x86/include/asm/amd_nb.h | 3 +++ arch/x86/kernel/amd_nb.c | 47 +++++++++++++++++++++++++++++++++++++------ arch/x86/pci/amd_bus.c | 6 ++++-- include/linux/pci_ids.h | 2 ++ 4 files changed, 50 insertions(+), 8 deletions(-)arch/x86/include/asm/amd_nb.h +0
-0
arch/x86/include/asm/amd_nb.h static inline u16 amd_pci_dev_to_node_id(struct pci_dev *pdev) · line 108
Quoted 6 lines
arch/x86/include/asm/amd_nb.h +3
-0@@ -103,6 +103,9 @@ static inline u16 amd_pci_dev_to_node_id(struct pci_dev *pdev)static inline bool amd_gart_present (void ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +return false; +
What is that for?
Hygon doesn't have F15h so that function will return false there too...
... or is that because the qemu script you got from the 0day bot guys uses -cpu kvm64 which is family 0xf:
[ 0.214353] smpboot: CPU0: AMD Common KVM processor (family: 0xf, model: 0x6, stepping: 0x1)
?
and that makes amd_gart_present() say yes.
In that case, please make a *prepatch* which adds the vendor check to both amd_gart_present() and early_is_amd_nb() and send it as a reply to this message.
*Then*, do this patch ontop and also as a reply.
Quoted 36 lines
arch/x86/include/asm/amd_nb.h +0
-0/* GART present only on Fam15h, upto model 0fh */ if (boot_cpu_data.x86 ==0xf || boot_cpu_data.x86 ==0x10 || (boot_cpu_data.x86 ==0x15 && boot_cpu_data.x86_model <0x10 ))arch/x86/kernel/amd_nb.c +18
-0@@ -61,6 +61,21 @@ static const struct pci_device_id amd_nb_link_ids[] = { {} }; +static const struct pci_device_id hygon_root_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_ROOT ) }, + {} +}; + +const struct pci_device_id hygon_nb_misc_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_DF_F3 ) }, + {} +}; + +static const struct pci_device_id hygon_nb_link_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_DF_F4 ) }, + {} +}; +const struct amd_nb_bus_dev_range amd_nb_bus_dev_ranges[]__initconst = { {0x00 ,0x18 ,0x20 }, {0xff ,0x00 ,0x20 },@@ -197,12 +212,25 @@ int amd_cache_northbridges(void)u16 i =0 ;struct amd_northbridge *nb;struct pci_dev *root, *misc, *link; +const struct pci_device_id *root_ids =NULL ; +const struct pci_device_id *misc_ids =NULL ; +const struct pci_device_id *link_ids =NULL ;
arch/x86/kernel/amd_nb.c int amd_cache_northbridges(void) · line 227
Quoted 10 lines
arch/x86/kernel/amd_nb.c +10
-0+ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + root_ids = hygon_root_ids; + misc_ids = hygon_nb_misc_ids; + link_ids = hygon_nb_link_ids; + }else { + root_ids = amd_root_ids; + misc_ids = amd_nb_misc_ids; + link_ids = amd_nb_link_ids; + }
Also, you can make this assignment differently:
const struct pci_device_id *root_ids = amd_root_ids;
const struct pci_device_id *misc_ids = amd_nb_misc_ids;
const struct pci_device_id *link_ids = amd_nb_link_ids;
if (boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) {
root_ids = hygon_root_ids;
misc_ids = hygon_nb_misc_ids;
link_ids = hygon_nb_link_ids;
}This way the change is obvious and it is only for Hygon without affecting the other vendors.
Ditto for the other assignment.
Thx.
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Pu Wen
maintainerRe: [PATCH v8 07/16] x86/pci: Add Hygon Dhyana support to PCI and north bridge
Quoted 14 lines · Borislav Petkov
arch/x86/include/asm/amd_nb.h static inline u16 amd_pci_dev_to_node_id(struct pci_dev *pdev) · line 108
Quoted 11 lines
arch/x86/include/asm/amd_nb.h +3
-0@@ -103,6 +103,9 @@ static inline u16 amd_pci_dev_to_node_id(struct pci_dev *pdev)static inline bool amd_gart_present (void ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +return false; +What is that for?
Quoted 1 line
Hygon doesn't have F15h so that function will return false there too...
This modification is used to solve the 0day PANIC occurred on Intel platform. The function will always return false when running on Hygon Dhyana platform, but will return true on Intel platform with CPU family 0xf/0x10/0x15.
Then the Intel CPU will run through early_gart_iommu_check and call early_is_amd_nb, but misc_ids for Intel in patch v7 Is NULL so there will be a PANIC:
[ 0.839894] RIP: 0010:early_is_amd_nb+0x30/0x4b ... [ 0.976358] Call Trace: [ 0.980238] ? early_gart_iommu_check+0xef/0x2c5 [ 0.987493] ? setup_arch+0x4fa/0xc67 [ 0.993231] ? printk+0x52/0x6e [ 0.998157] ? start_kernel+0x6e/0x4dc [ 1.004044] ? load_ucode_bsp+0x42/0x12e [ 1.010145] ? secondary_startup_64+0xa4/0xb0
Quoted 11 lines
... or is that because the qemu script you got from the 0day bot guys uses -cpu kvm64 which is family
0xf:[ 0.214353] smpboot: CPU0: AMD Common KVM processor (family: 0xf, model: 0x6, stepping: 0x1)?
and that makes
amd_gart_present()say yes.
In that case, please make a *prepatch* which adds the vendor check to bothamd_gart_present()andearly_is_amd_nb()and send it as a reply to this message.
OK, I'll make a separate prepatch for this change.
Quoted 2 lines
*Then*, do this patch ontop and also as a reply.
Do you mean do the follow change on top of the prepatch?
Quoted 31 lines
Quoted 7 lines
arch/x86/include/asm/amd_nb.h +3
-0@@ -197,12 +212,25 @@ int amd_cache_northbridges(void)u16 i =0 ;struct amd_northbridge *nb;struct pci_dev *root, *misc, *link; +const struct pci_device_id *root_ids =NULL ; +const struct pci_device_id *misc_ids =NULL ; +const struct pci_device_id *link_ids =NULL ;arch/x86/include/asm/amd_nb.h int amd_cache_northbridges(void) · line 227
Quoted 10 lines
arch/x86/include/asm/amd_nb.h +10
-0+ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + root_ids = hygon_root_ids; + misc_ids = hygon_nb_misc_ids; + link_ids = hygon_nb_link_ids; + }else { + root_ids = amd_root_ids; + misc_ids = amd_nb_misc_ids; + link_ids = amd_nb_link_ids; + }Also, you can make this assignment differently:
const struct pci_device_id *root_ids = amd_root_ids; const struct pci_device_id *misc_ids = amd_nb_misc_ids; const struct pci_device_id *link_ids = amd_nb_link_ids; if (boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) { root_ids = hygon_root_ids; misc_ids = hygon_nb_misc_ids; link_ids = hygon_nb_link_ids; }
Quoted 4 lines
This way the change is obvious and it is only for Hygon without affecting the other vendors.
Ditto for the other assignment.
All right, I will change these in the next version patch set.
Thanks,
Pu Wen
Borislav Petkov
maintainerRe: [PATCH v8 07/16] x86/pci: Add Hygon Dhyana support to PCI and north bridge
Quoted 1 line · Pu Wen
All right, I will change these in the next version patch set.
Just send two patches as a reply to this thread here:
1. add the vendor check to both amd_gart_present() and early_is_amd_nb()
2. redo 7/16 ontop of the first patch.
I don't think you need to resend the whole patchset as the rest is still good.
Right?
Thx.
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Pu Wen
maintainerRe: [PATCH v8 07/16] x86/pci: Add Hygon Dhyana support to PCI and north bridge
Quoted 10 lines · Borislav Petkov
Quoted 1 line · Pu Wen
All right, I will change these in the next version patch set.
Just send two patches as a reply to this thread here:
1. add the vendor check to both
amd_gart_present()andearly_is_amd_nb()
2. redo 7/16 ontop of the first patch.I don't think you need to resend the whole patchset as the rest is still good.
OK, I got the point. :)
I'll just send the two patches.
Thx.
Signature · 2 lines
Regards, Pu Wen
Pu Wen
maintainerPATCH 1/2 x86/amd_nb: Add vendor checking for strict function access
In reply to Borislav PetkovTo prevent further PCI device ids checking in early_gart_iommu_check which cannot happen both on Hygon and Intel platform, the function amd_gart_present should return if it's not a AMD CPU.
Also only AMD or Hygon CPU should use the functionality provided by early_is_amd_nb.
--- arch/x86/include/asm/amd_nb.h | 3 +++ arch/x86/kernel/amd_nb.c | 4 ++++ 2 files changed, 7 insertions(+)
arch/x86/include/asm/amd_nb.h +3 -0
@@ -103,6 +103,9 @@ static inline u16 amd_pci_dev_to_node_id(struct pci_dev *pdev)static inline bool amd_gart_present (void ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +return false; +/* GART present only on Fam15h, upto model 0fh */ if (boot_cpu_data.x86 ==0xf || boot_cpu_data.x86 ==0x10 || (boot_cpu_data.x86 ==0x15 && boot_cpu_data.x86_model <0x10 ))
arch/x86/kernel/amd_nb.c +4 -0
@@ -264,6 +264,10 @@ bool __init early_is_amd_nb(u32 device)const struct pci_device_id *id;u32 vendor = device &0xffff ; +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON ) +return false; + device >>=16 ;for (id = amd_nb_misc_ids; id->vendor; id++)if (vendor == id->vendor && device == id->device)
Signature · 1 line
2.7.4
Pu Wen
maintainerPATCH 2/2 x86/pci: Add Hygon Dhyana support to PCI and north bridge
In reply to Borislav PetkovAs Hygon registered its PCI Vendor ID as a new one 0x1d94, and there are PCI Devices 0x1450/0x1463/0x1464 for Host bridge on Hygon Dhyana platform, so add Hygon Dhyana support to the PCI and north bridge subsystem by using the code path of AMD family 17h.
--- arch/x86/kernel/amd_nb.c | 41 +++++++++++++++++++++++++++++++++++------ arch/x86/pci/amd_bus.c | 6 ++++-- include/linux/pci_ids.h | 2 ++ 3 files changed, 41 insertions(+), 8 deletions(-)
arch/x86/kernel/amd_nb.c +35 -6
@@ -61,6 +61,21 @@ static const struct pci_device_id amd_nb_link_ids[] = { {} }; +static const struct pci_device_id hygon_root_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_ROOT ) }, + {} +}; + +const struct pci_device_id hygon_nb_misc_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_DF_F3 ) }, + {} +}; + +static const struct pci_device_id hygon_nb_link_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_DF_F4 ) }, + {} +}; +const struct amd_nb_bus_dev_range amd_nb_bus_dev_ranges[]__initconst = { {0x00 ,0x18 ,0x20 }, {0xff ,0x00 ,0x20 },
@@ -197,12 +212,21 @@ int amd_cache_northbridges(void)u16 i =0 ;struct amd_northbridge *nb;struct pci_dev *root, *misc, *link; +const struct pci_device_id *root_ids = amd_root_ids; +const struct pci_device_id *misc_ids = amd_nb_misc_ids; +const struct pci_device_id *link_ids = amd_nb_link_ids; + +if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + root_ids = hygon_root_ids; + misc_ids = hygon_nb_misc_ids; + link_ids = hygon_nb_link_ids; + }if (amd_northbridges.num)return 0 ; misc =NULL ; -while ((misc =next_northbridge (misc, amd_nb_misc_ids)) !=NULL ) +while ((misc =next_northbridge (misc, misc_ids)) !=NULL ) i++;if (!i)
@@ -218,11 +242,11 @@ int amd_cache_northbridges(void) link = misc = root =NULL ;for (i =0 ; i != amd_northbridges.num; i++) {node_to_amd_nb (i)->root = root = -next_northbridge (root, amd_root_ids); +next_northbridge (root, root_ids);node_to_amd_nb (i)->misc = misc = -next_northbridge (misc, amd_nb_misc_ids); +next_northbridge (misc, misc_ids);node_to_amd_nb (i)->link = link = -next_northbridge (link, amd_nb_link_ids); +next_northbridge (link, link_ids); }if (amd_gart_present ())
@@ -263,13 +287,17 @@ bool __init early_is_amd_nb(u32 device) {const struct pci_device_id *id;u32 vendor = device &0xffff ; +const struct pci_device_id *misc_ids = amd_nb_misc_ids;if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return false; +if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + misc_ids = hygon_nb_misc_ids; + device >>=16 ; -for (id = amd_nb_misc_ids; id->vendor; id++) +for (id = misc_ids; id->vendor; id++)if (vendor == id->vendor && device == id->device)return true;return false;
@@ -281,7 +309,8 @@ struct resource *amd_get_mmconfig_range(struct resource *res)u64 base, msr;unsigned int segn_busn_bits; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return NULL ;/* assume all cpus from fam10h have mmconfig */
arch/x86/pci/amd_bus.c +4 -2
@@ -93,7 +93,8 @@ static int __init early_root_info_init(void) vendor = id &0xffff ; device = (id>>16 ) &0xffff ; -if (vendor !=PCI_VENDOR_ID_AMD ) +if (vendor !=PCI_VENDOR_ID_AMD && + vendor !=PCI_VENDOR_ID_HYGON )continue ;if (hb_probes[i].device == device) {
@@ -390,7 +391,8 @@ static int __init pci_io_ecs_init(void)static int __init amd_postcore_init (void ) { -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return 0 ;early_root_info_init ();
include/linux/pci_ids.h +2 -0
@@ -2561,6 +2561,8 @@#define PCI_VENDOR_ID_AMAZON 0x1d0f +#define PCI_VENDOR_ID_HYGON 0x1d94 +#define PCI_VENDOR_ID_TEKRAM 0x1de1#define PCI_DEVICE_ID_TEKRAM_DC290 0xdc29
Signature · 1 line
2.7.4
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/cpu: Create Hygon Dhyana architecture support file
In reply to Pu Wenx86/cpu: Create Hygon Dhyana architecture support file
Commit-ID: c9661c1e80b609cd038db7c908e061f0535804ef Gitweb: https://git.kernel.org/tip/c9661c1e80b609cd038db7c908e061f0535804ef Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:33:12 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 16:14:05 +0200
x86/cpu: Create Hygon Dhyana architecture support file
Add x86 architecture support for a new processor: Hygon Dhyana Family 18h. Carve out initialization code needed by Dhyana into a separate compilation unit.
To identify Hygon Dhyana CPU, add a new vendor type X86_VENDOR_HYGON.
Since Dhyana uses AMD functionality to a large degree, select CPU_SUP_AMD which provides that functionality.
[ bp: drop explicit license statement as it has an SPDX tag already. ]
--- MAINTAINERS | 6 + arch/x86/Kconfig.cpu | 14 ++ arch/x86/include/asm/processor.h | 3 +- arch/x86/kernel/cpu/Makefile | 1 + arch/x86/kernel/cpu/hygon.c | 405 +++++++++++++++++++++++++++++++++++++++ 5 files changed, 428 insertions(+), 1 deletion(-)
MAINTAINERS +6 -0
@@ -6762,6 +6762,12 @@ S: Maintained F: mm/memory-failure.c F: mm/hwpoison-inject.c +HYGON PROCESSOR SUPPORT +M: Pu Wen <[email protected]> +L: [email protected] +S: Maintained +F: arch/x86/kernel/cpu/hygon.c + Hyper-V CORE AND DRIVERS M: "K. Y. Srinivasan" <[email protected]> M: Haiyang Zhang <[email protected]>
arch/x86/Kconfig.cpu +14 -0
@@ -426,6 +426,20 @@ config CPU_SUP_AMD If unsure, say N. +config CPU_SUP_HYGON + default y + bool "Support Hygon processors" if PROCESSOR_SELECT + select CPU_SUP_AMD + help + This enables detection, tunings and quirks for Hygon processors + + You need this enabled if you want your kernel to run on an + Hygon CPU. Disabling this option on other types of CPUs + makes the kernel a tiny bit smaller. Disabling it on an Hygon + CPU might render the kernel unbootable. + + If unsure, say N. + config CPU_SUP_CENTAUR default y bool "Support Centaur processors" if PROCESSOR_SELECT
arch/x86/include/asm/processor.h +2 -1
@@ -155,7 +155,8 @@ enum cpuid_regs_idx {#define X86_VENDOR_CENTAUR 5#define X86_VENDOR_TRANSMETA 7#define X86_VENDOR_NSC 8 -#define X86_VENDOR_NUM 9 +#define X86_VENDOR_HYGON 9 +#define X86_VENDOR_NUM 10#define X86_VENDOR_UNKNOWN 0xff
arch/x86/kernel/cpu/Makefile +1 -0
@@ -30,6 +30,7 @@ obj-$(CONFIG_X86_FEATURE_NAMES) += capflags.o powerflags.o obj-$(CONFIG_CPU_SUP_INTEL) += intel.o intel_pconfig.o obj-$(CONFIG_CPU_SUP_AMD) += amd.o +obj-$(CONFIG_CPU_SUP_HYGON) += hygon.o obj-$(CONFIG_CPU_SUP_CYRIX_32) += cyrix.o obj-$(CONFIG_CPU_SUP_CENTAUR) += centaur.o obj-$(CONFIG_CPU_SUP_TRANSMETA_32) += transmeta.o
arch/x86/kernel/cpu/hygon.c new file +405 -0
405 lines not on this page · raw message
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/cpu: Get cache info and setup cache cpumap for Hygon Dhyana
In reply to Pu Wenx86/cpu: Get cache info and setup cache cpumap for Hygon Dhyana
Commit-ID: d4f7423efdd1419b17524d090ff9ff4024bcf09b Gitweb: https://git.kernel.org/tip/d4f7423efdd1419b17524d090ff9ff4024bcf09b Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:33:44 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:57 +0200
x86/cpu: Get cache info and setup cache cpumap for Hygon Dhyana
The Hygon Dhyana CPU has a topology extensions bit in CPUID. With this bit, the kernel can get the cache information. So add support in cpuid4_cache_lookup_regs() to get the correct cache size.
The Hygon Dhyana CPU also discovers num_cache_leaves via CPUID leaf 0x8000001d, so add support to it in find_num_cache_leaves().
Also add cacheinfo_hygon_init_llc_id() and init_hygon_cacheinfo() functions to initialize Dhyana cache info. Setup cache cpumap in the same way as AMD does.
--- arch/x86/include/asm/cacheinfo.h | 1 + arch/x86/kernel/cpu/cacheinfo.c | 31 +++++++++++++++++++++++++++++-- arch/x86/kernel/cpu/cpu.h | 1 + arch/x86/kernel/cpu/hygon.c | 3 +++ 4 files changed, 34 insertions(+), 2 deletions(-)
arch/x86/include/asm/cacheinfo.h +1 -0
@@ -3,5 +3,6 @@#define _ASM_X86_CACHEINFO_Hvoid cacheinfo_amd_init_llc_id (struct cpuinfo_x86 *c,int cpu,u8 node_id); +void cacheinfo_hygon_init_llc_id (struct cpuinfo_x86 *c,int cpu,u8 node_id);#endif /* _ASM_X86_CACHEINFO_H */
arch/x86/kernel/cpu/cacheinfo.c +29 -2
@@ -602,6 +602,10 @@ cpuid4_cache_lookup_regs(int index, struct _cpuid4_info_regs *this_leaf)else amd_cpuid4 (index, &eax, &ebx, &ecx);amd_init_l3_cache (this_leaf, index); + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { +cpuid_count (0x8000001d , index, &eax.full, + &ebx.full, &ecx.full, &edx); +amd_init_l3_cache (this_leaf, index); }else {cpuid_count (4 , index, &eax.full, &ebx.full, &ecx.full, &edx); }
@@ -625,7 +629,8 @@ static int find_num_cache_leaves(struct cpuinfo_x86 *c)union _cpuid4_leaf_eax cache_eax;int i =-1 ; -if (c->x86_vendor ==X86_VENDOR_AMD ) +if (c->x86_vendor ==X86_VENDOR_AMD || + c->x86_vendor ==X86_VENDOR_HYGON ) op =0x8000001d ;else op =4 ;
@@ -678,6 +683,22 @@ void cacheinfo_amd_init_llc_id(struct cpuinfo_x86 *c, int cpu, u8 node_id) } } +void cacheinfo_hygon_init_llc_id (struct cpuinfo_x86 *c,int cpu,u8 node_id) +{ +/* +* We may have multiple LLCs if L3 caches exist, so check if we +* have an L3 cache by looking at the L3 cache CPUID leaf. +*/ +if (!cpuid_edx (0x80000006 )) +return ; + +/* +* LLC is at the core complex level. +* Core complex ID is ApicId[3] for these processors. +*/ +per_cpu (cpu_llc_id, cpu) = c->apicid >>3 ; +} +void init_amd_cacheinfo (struct cpuinfo_x86 *c) {
@@ -691,6 +712,11 @@ void init_amd_cacheinfo(struct cpuinfo_x86 *c) } } +void init_hygon_cacheinfo (struct cpuinfo_x86 *c) +{ + num_cache_leaves =find_num_cache_leaves (c); +} +void init_intel_cacheinfo (struct cpuinfo_x86 *c) {/* Cache sizes */
@@ -913,7 +939,8 @@ static void __cache_cpumap_setup(unsigned int cpu, int index,int index_msb, i;struct cpuinfo_x86 *c = &cpu_data (cpu); -if (c->x86_vendor ==X86_VENDOR_AMD ) { +if (c->x86_vendor ==X86_VENDOR_AMD || + c->x86_vendor ==X86_VENDOR_HYGON ) {if (__cache_amd_cpumap_setup (cpu, index, base))return ; }
arch/x86/kernel/cpu/cpu.h +1 -0
@@ -54,6 +54,7 @@ extern u32 get_scattered_cpuid_leaf(unsigned int level, enum cpuid_regs_idx reg);extern void init_intel_cacheinfo (struct cpuinfo_x86 *c);extern void init_amd_cacheinfo (struct cpuinfo_x86 *c); +extern void init_hygon_cacheinfo (struct cpuinfo_x86 *c);extern void detect_num_cpu_cores (struct cpuinfo_x86 *c);extern int detect_extended_topology_early (struct cpuinfo_x86 *c);
arch/x86/kernel/cpu/hygon.c +3 -0
@@ -87,6 +87,7 @@ static void hygon_get_topology(struct cpuinfo_x86 *c)if (!err) c->x86_coreid_bits =get_count_order (c->x86_max_cores); +cacheinfo_hygon_init_llc_id (c, cpu, node_id); }else if (cpu_has (c,X86_FEATURE_NODEID_MSR )) {u64 value;
@@ -321,6 +322,8 @@ static void init_hygon(struct cpuinfo_x86 *c)hygon_get_topology (c);srat_detect_node (c); +init_hygon_cacheinfo (c); +if (cpu_has (c,X86_FEATURE_XMM2 )) {unsigned long long val;int ret;
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/cpu/mtrr: Support TOP_MEM2 and get MTRR number
In reply to Pu Wenx86/cpu/mtrr: Support TOP_MEM2 and get MTRR number
Commit-ID: 39dc6f154dac134e4612827cb5283934c1862cb8 Gitweb: https://git.kernel.org/tip/39dc6f154dac134e4612827cb5283934c1862cb8 Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:34:16 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:57 +0200
x86/cpu/mtrr: Support TOP_MEM2 and get MTRR number
The Hygon Dhyana CPU has a special MSR way to force WB for memory >4GB, and support TOP_MEM2. Therefore, it is necessary to add Hygon Dhyana support in amd_special_default_mtrr().
The number of variable MTRRs for Hygon is 2 as AMD's.
--- arch/x86/kernel/cpu/mtrr/cleanup.c | 3 ++- arch/x86/kernel/cpu/mtrr/mtrr.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-)
arch/x86/kernel/cpu/mtrr/cleanup.c +2 -1
@@ -831,7 +831,8 @@ int __init amd_special_default_mtrr(void) {u32 l, h; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return 0 ;if (boot_cpu_data.x86 <0xf )return 0 ;
arch/x86/kernel/cpu/mtrr/mtrr.c +1 -1
@@ -127,7 +127,7 @@ static void __init set_num_var_ranges(void)if (use_intel ())rdmsr (MSR_MTRRcap, config, dummy); -else if (is_cpu (AMD )) +else if (is_cpu (AMD ) ||is_cpu (HYGON )) config =2 ;else if (is_cpu (CYRIX ) ||is_cpu (CENTAUR )) config =8 ;
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/smpboot: Do not use BSP INIT delay and MWAIT to idle on Dhyana
In reply to Pu Wenx86/smpboot: Do not use BSP INIT delay and MWAIT to idle on Dhyana
Commit-ID: 0b13bec787dccca96f8c431da732657ae01baf9a Gitweb: https://git.kernel.org/tip/0b13bec787dccca96f8c431da732657ae01baf9a Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:34:32 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:57 +0200
x86/smpboot: Do not use BSP INIT delay and MWAIT to idle on Dhyana
The Hygon Dhyana CPU uses no delay in smp_quirk_init_udelay(), and does HLT on idle just like AMD does.
--- arch/x86/kernel/smpboot.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
arch/x86/kernel/smpboot.c +3 -1
@@ -676,6 +676,7 @@ static void __init smp_quirk_init_udelay(void)/* if modern processor, use no delay */ if (((boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) && (boot_cpu_data.x86 ==6 )) || + ((boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) && (boot_cpu_data.x86 >=0x18 )) || ((boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) && (boot_cpu_data.x86 >=0xF ))) { init_udelay =0 ;return ;
@@ -1592,7 +1593,8 @@ static inline void mwait_play_dead(void)void *mwait_ptr;int i; -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON )return ;if (!this_cpu_has (X86_FEATURE_MWAIT ))return ;
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/events: Add Hygon Dhyana support to PMU infrastructure
In reply to Pu Wenx86/events: Add Hygon Dhyana support to PMU infrastructure
Commit-ID: 6d0ef316b9f8ea03fa867debda70b2f11a0b9736 Gitweb: https://git.kernel.org/tip/6d0ef316b9f8ea03fa867debda70b2f11a0b9736 Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:34:47 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:57 +0200
x86/events: Add Hygon Dhyana support to PMU infrastructure
The PMU architecture for the Hygon Dhyana CPU is similar to the AMD Family 17h one. To support it, call amd_pmu_init() to share the AMD PMU initialization flow, and change the PMU name to "HYGON".
The Hygon Dhyana CPU supports both legacy and extension PMC MSRs (perf counter registers and event selection registers), so add Hygon Dhyana support in the similar way as AMD does.
--- arch/x86/events/amd/core.c | 4 ++++ arch/x86/events/amd/uncore.c | 20 +++++++++++++------- arch/x86/events/core.c | 4 ++++ arch/x86/kernel/cpu/perfctr-watchdog.c | 2 ++ 4 files changed, 23 insertions(+), 7 deletions(-)
arch/x86/events/amd/core.c +4 -0
@@ -669,6 +669,10 @@ static int __init amd_core_pmu_init(void)* We fallback to using default amd_get_event_constraints. */ break ; +case 0x18 : +pr_cont ("Fam18h " ); +/* Using default amd_get_event_constraints. */ +break ;default :pr_err ("core perfctr but no constraints; unknown hardware!\n" );return -ENODEV ;
arch/x86/events/amd/uncore.c +13 -7
@@ -507,17 +507,19 @@ static int __init amd_uncore_init(void) {int ret = -ENODEV ; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return -ENODEV ;if (!boot_cpu_has (X86_FEATURE_TOPOEXT ))return -ENODEV ; -if (boot_cpu_data.x86 ==0x17 ) { +if (boot_cpu_data.x86 ==0x17 || boot_cpu_data.x86 ==0x18 ) {/* -* For F17h, the Northbridge counters are repurposed as Data -* Fabric counters. Also, L3 counters are supported too. The PMUs -* are exported based on family as either L2 or L3 and NB or DF. +* For F17h or F18h, the Northbridge counters are +* repurposed as Data Fabric counters. Also, L3 +* counters are supported too. The PMUs are exported +* based on family as either L2 or L3 and NB or DF. */ num_counters_nb =NUM_COUNTERS_NB ; num_counters_llc =NUM_COUNTERS_L3 ;
@@ -547,7 +549,9 @@ static int __init amd_uncore_init(void)if (ret) goto fail_nb; -pr_info ("AMD NB counters detected\n" ); +pr_info ("%s NB counters detected\n" , + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ? +"HYGON" :"AMD" ); ret =0 ; }
@@ -561,7 +565,9 @@ static int __init amd_uncore_init(void)if (ret) goto fail_llc; -pr_info ("AMD LLC counters detected\n" ); +pr_info ("%s LLC counters detected\n" , + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ? +"HYGON" :"AMD" ); ret =0 ; }
arch/x86/events/core.c +4 -0
@@ -1776,6 +1776,10 @@ static int __init init_hw_perf_events(void)case X86_VENDOR_AMD : err =amd_pmu_init ();break ; +case X86_VENDOR_HYGON : + err =amd_pmu_init (); + x86_pmu.name ="HYGON" ; +break ;default : err = -ENOTSUPP ; }
arch/x86/kernel/cpu/perfctr-watchdog.c +2 -0
@@ -46,6 +46,7 @@ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr) {/* returns the bit offset of the performance counter register */ switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :if (msr >=MSR_F15H_PERF_CTR )return (msr -MSR_F15H_PERF_CTR ) >>1 ;
@@ -74,6 +75,7 @@ static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr) {/* returns the bit offset of the event selection register */ switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :if (msr >=MSR_F15H_PERF_CTL )return (msr -MSR_F15H_PERF_CTL ) >>1 ;
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/alternative: Init ideal_nops for Hygon Dhyana
In reply to Pu Wenx86/alternative: Init ideal_nops for Hygon Dhyana
Commit-ID: c3fecca457c1aa1c1a2f81bfe68393af244a263e Gitweb: https://git.kernel.org/tip/c3fecca457c1aa1c1a2f81bfe68393af244a263e Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:35:01 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:58 +0200
x86/alternative: Init ideal_nops for Hygon Dhyana
The ideal_nops for Hygon Dhyana CPU should be p6_nops.
--- arch/x86/kernel/alternative.c | 4 ++++ 1 file changed, 4 insertions(+)
arch/x86/kernel/alternative.c +4 -0
@@ -222,6 +222,10 @@ void __init arch_init_ideal_nops(void) }break ; +case X86_VENDOR_HYGON : + ideal_nops = p6_nops; +return ; +case X86_VENDOR_AMD :if (boot_cpu_data.x86 >0xf ) { ideal_nops = p6_nops;
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/amd_nb: Check vendor in AMD-only functions
In reply to Pu Wenx86/amd_nb: Check vendor in AMD-only functions
Commit-ID: b7a5cb4f220e78490735b2b984ad29b7d8e612a9 Gitweb: https://git.kernel.org/tip/b7a5cb4f220e78490735b2b984ad29b7d8e612a9 Author: Pu Wen <[email protected]> AuthorDate: Tue, 25 Sep 2018 22:45:01 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:58 +0200
x86/amd_nb: Check vendor in AMD-only functions
Exit early in functions which are meant to run on AMD only but which get run on different vendor (VMs, etc).
[ bp: rewrite commit message. ]
--- arch/x86/include/asm/amd_nb.h | 3 +++ arch/x86/kernel/amd_nb.c | 4 ++++ 2 files changed, 7 insertions(+)
arch/x86/include/asm/amd_nb.h +3 -0
@@ -103,6 +103,9 @@ static inline u16 amd_pci_dev_to_node_id(struct pci_dev *pdev)static inline bool amd_gart_present (void ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +return false; +/* GART present only on Fam15h, upto model 0fh */ if (boot_cpu_data.x86 ==0xf || boot_cpu_data.x86 ==0x10 || (boot_cpu_data.x86 ==0x15 && boot_cpu_data.x86_model <0x10 ))
arch/x86/kernel/amd_nb.c +4 -0
@@ -264,6 +264,10 @@ bool __init early_is_amd_nb(u32 device)const struct pci_device_id *id;u32 vendor = device &0xffff ; +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON ) +return false; + device >>=16 ;for (id = amd_nb_misc_ids; id->vendor; id++)if (vendor == id->vendor && device == id->device)
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/pci, x86/amd_nb: Add Hygon Dhyana support to PCI and northbridge
In reply to Pu Wenx86/pci, x86/amd_nb: Add Hygon Dhyana support to PCI and northbridge
Commit-ID: c6babb5806b77c6ca7078c3487bb0a29704a4e38 Gitweb: https://git.kernel.org/tip/c6babb5806b77c6ca7078c3487bb0a29704a4e38 Author: Pu Wen <[email protected]> AuthorDate: Tue, 25 Sep 2018 22:46:11 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:58 +0200
x86/pci, x86/amd_nb: Add Hygon Dhyana support to PCI and northbridge
Hygon's PCI vendor ID is 0x1d94, and there are PCI devices 0x1450/0x1463/0x1464 for the host bridge on the Hygon Dhyana platform. Add Hygon Dhyana support to the PCI and northbridge subsystems by using the code path of AMD family 17h.
[ bp: Massage commit message, sort local vars into reverse xmas tree
order and move the amd_northbridges.num check up. ]
--- arch/x86/kernel/amd_nb.c | 45 +++++++++++++++++++++++++++++++++++++-------- arch/x86/pci/amd_bus.c | 6 ++++-- include/linux/pci_ids.h | 2 ++ 3 files changed, 43 insertions(+), 10 deletions(-)
arch/x86/kernel/amd_nb.c +37 -8
@@ -61,6 +61,21 @@ static const struct pci_device_id amd_nb_link_ids[] = { {} }; +static const struct pci_device_id hygon_root_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_ROOT ) }, + {} +}; + +const struct pci_device_id hygon_nb_misc_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_DF_F3 ) }, + {} +}; + +static const struct pci_device_id hygon_nb_link_ids[] = { + {PCI_DEVICE (PCI_VENDOR_ID_HYGON ,PCI_DEVICE_ID_AMD_17H_DF_F4 ) }, + {} +}; +const struct amd_nb_bus_dev_range amd_nb_bus_dev_ranges[]__initconst = { {0x00 ,0x18 ,0x20 }, {0xff ,0x00 ,0x20 },
@@ -194,15 +209,24 @@ EXPORT_SYMBOL_GPL(amd_df_indirect_read);int amd_cache_northbridges (void ) { -u16 i =0 ; -struct amd_northbridge *nb; +const struct pci_device_id *misc_ids = amd_nb_misc_ids; +const struct pci_device_id *link_ids = amd_nb_link_ids; +const struct pci_device_id *root_ids = amd_root_ids;struct pci_dev *root, *misc, *link; +struct amd_northbridge *nb; +u16 i =0 ;if (amd_northbridges.num)return 0 ; +if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + root_ids = hygon_root_ids; + misc_ids = hygon_nb_misc_ids; + link_ids = hygon_nb_link_ids; + } + misc =NULL ; -while ((misc =next_northbridge (misc, amd_nb_misc_ids)) !=NULL ) +while ((misc =next_northbridge (misc, misc_ids)) !=NULL ) i++;if (!i)
@@ -218,11 +242,11 @@ int amd_cache_northbridges(void) link = misc = root =NULL ;for (i =0 ; i != amd_northbridges.num; i++) {node_to_amd_nb (i)->root = root = -next_northbridge (root, amd_root_ids); +next_northbridge (root, root_ids);node_to_amd_nb (i)->misc = misc = -next_northbridge (misc, amd_nb_misc_ids); +next_northbridge (misc, misc_ids);node_to_amd_nb (i)->link = link = -next_northbridge (link, amd_nb_link_ids); +next_northbridge (link, link_ids); }if (amd_gart_present ())
@@ -261,6 +285,7 @@ EXPORT_SYMBOL_GPL(amd_cache_northbridges);*/ bool __init early_is_amd_nb (u32 device) { +const struct pci_device_id *misc_ids = amd_nb_misc_ids;const struct pci_device_id *id;u32 vendor = device &0xffff ;
@@ -268,8 +293,11 @@ bool __init early_is_amd_nb(u32 device) boot_cpu_data.x86_vendor != X86_VENDOR_HYGON)return false; +if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + misc_ids = hygon_nb_misc_ids; + device >>=16 ; -for (id = amd_nb_misc_ids; id->vendor; id++) +for (id = misc_ids; id->vendor; id++)if (vendor == id->vendor && device == id->device)return true;return false;
@@ -281,7 +309,8 @@ struct resource *amd_get_mmconfig_range(struct resource *res)u64 base, msr;unsigned int segn_busn_bits; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return NULL ;/* assume all cpus from fam10h have mmconfig */
arch/x86/pci/amd_bus.c +4 -2
@@ -93,7 +93,8 @@ static int __init early_root_info_init(void) vendor = id &0xffff ; device = (id>>16 ) &0xffff ; -if (vendor !=PCI_VENDOR_ID_AMD ) +if (vendor !=PCI_VENDOR_ID_AMD && + vendor !=PCI_VENDOR_ID_HYGON )continue ;if (hb_probes[i].device == device) {
@@ -390,7 +391,8 @@ static int __init pci_io_ecs_init(void)static int __init amd_postcore_init (void ) { -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON )return 0 ;early_root_info_init ();
include/linux/pci_ids.h +2 -0
@@ -2561,6 +2561,8 @@#define PCI_VENDOR_ID_AMAZON 0x1d0f +#define PCI_VENDOR_ID_HYGON 0x1d94 +#define PCI_VENDOR_ID_TEKRAM 0x1de1#define PCI_DEVICE_ID_TEKRAM_DC290 0xdc29
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/apic: Add Hygon Dhyana support
In reply to Pu Wenx86/apic: Add Hygon Dhyana support
Commit-ID: da33dfef404174b0b452f4d2a9a9e00801794f3a Gitweb: https://git.kernel.org/tip/da33dfef404174b0b452f4d2a9a9e00801794f3a Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:35:28 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:58 +0200
x86/apic: Add Hygon Dhyana support
Add Hygon Dhyana support to the APIC subsystem. When running in 32 bit mode, bigsmp should be enabled if there are more than 8 cores online.
--- arch/x86/kernel/apic/apic.c | 7 +++++++ arch/x86/kernel/apic/probe_32.c | 1 + 2 files changed, 8 insertions(+)
arch/x86/kernel/apic/apic.c +7 -0
@@ -224,6 +224,11 @@ static int modern_apic(void)if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD && boot_cpu_data.x86 >=0xf )return 1 ; + +/* Hygon systems use modern APIC */ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) +return 1 ; +return lapic_get_version () >=0x14 ; }
@@ -1912,6 +1917,8 @@ static int __init detect_init_APIC(void) (boot_cpu_data.x86 >= 15))break ; goto no_apic; +case X86_VENDOR_HYGON : +break ;case X86_VENDOR_INTEL :if (boot_cpu_data.x86 ==6 || boot_cpu_data.x86 ==15 || (boot_cpu_data.x86 ==5 &&boot_cpu_has (X86_FEATURE_APIC )))
arch/x86/kernel/apic/probe_32.c +1 -0
@@ -185,6 +185,7 @@ void __init default_setup_apic_routing(void)break ; }/* If P4 and above fall through */ +case X86_VENDOR_HYGON :case X86_VENDOR_AMD : def_to_bigsmp =1 ; }
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/bugs: Add Hygon Dhyana to the respective mitigation machinery
In reply to Pu Wenx86/bugs: Add Hygon Dhyana to the respective mitigation machinery
Commit-ID: 1a576b23d63794f39a247fb31056eecccbf9a287 Gitweb: https://git.kernel.org/tip/1a576b23d63794f39a247fb31056eecccbf9a287 Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:35:50 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:59 +0200
x86/bugs: Add Hygon Dhyana to the respective mitigation machinery
The Hygon Dhyana CPU has the same speculative execution as AMD family 17h, so share AMD spectre mitigation code with Hygon Dhyana.
Also Hygon Dhyana is not affected by meltdown, so add exception for it.
--- arch/x86/kernel/cpu/bugs.c | 4 +++- arch/x86/kernel/cpu/common.c | 1 + 2 files changed, 4 insertions(+), 1 deletion(-)
arch/x86/kernel/cpu/bugs.c +3 -1
@@ -312,6 +312,7 @@ static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void) }if (cmd ==SPECTRE_V2_CMD_RETPOLINE_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON && boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) {pr_err ("retpoline,amd selected but CPU is not AMD. Switching to AUTO select\n" );return SPECTRE_V2_CMD_AUTO ;
@@ -371,7 +372,8 @@ static void __init spectre_v2_select_mitigation(void)return ; retpoline_auto: -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { retpoline_amd:if (!boot_cpu_has (X86_FEATURE_LFENCE_RDTSC )) {pr_err ("Spectre mitigation: LFENCE not serializing, switching to generic retpoline\n" );
arch/x86/kernel/cpu/common.c +1 -0
@@ -963,6 +963,7 @@ static const __initconst struct x86_cpu_id cpu_no_speculation[] = {static const __initconst struct x86_cpu_id cpu_no_meltdown[] = { {X86_VENDOR_AMD }, + {X86_VENDOR_HYGON }, {} };
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/mce: Add Hygon Dhyana support to the MCA infrastructure
In reply to Pu Wenx86/mce: Add Hygon Dhyana support to the MCA infrastructure
Commit-ID: ac78bd72355d0da64c073c12927264d4ff19b886 Gitweb: https://git.kernel.org/tip/ac78bd72355d0da64c073c12927264d4ff19b886 Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:36:04 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:59 +0200
x86/mce: Add Hygon Dhyana support to the MCA infrastructure
The machine check architecture for Hygon Dhyana CPU is similar to the AMD family 17h one. Add vendor checking for Hygon Dhyana to share the code path of AMD family 17h.
--- arch/x86/include/asm/mce.h | 2 ++ arch/x86/kernel/cpu/mcheck/mce-severity.c | 3 ++- arch/x86/kernel/cpu/mcheck/mce.c | 20 +++++++++++++++----- 3 files changed, 19 insertions(+), 6 deletions(-)
arch/x86/include/asm/mce.h +2 -0
@@ -214,6 +214,8 @@ static inline void mce_amd_feature_init(struct cpuinfo_x86 *c) { }static inline int umc_normaddr_to_sysaddr (u64 norm_addr,u16 nid,u8 umc,u64 *sys_addr) {return -EINVAL ; };#endif +static inline void mce_hygon_feature_init (struct cpuinfo_x86 *c) {return mce_amd_feature_init (c); } +int mce_available (struct cpuinfo_x86 *c);bool mce_is_memory_error (struct mce *m);
arch/x86/kernel/cpu/mcheck/mce-severity.c +2 -1
@@ -336,7 +336,8 @@ int (*mce_severity)(struct mce *m, int tolerant, char **msg, bool is_excp) =void __init mcheck_vendor_init_severity (void ) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) mce_severity = mce_severity_amd; }
arch/x86/kernel/cpu/mcheck/mce.c +15 -5
@@ -270,7 +270,7 @@ static void print_mce(struct mce *m) {__print_mce (m); -if (m->cpuvendor !=X86_VENDOR_AMD ) +if (m->cpuvendor !=X86_VENDOR_AMD && m->cpuvendor !=X86_VENDOR_HYGON )pr_emerg_ratelimited (HW_ERR "Run the above through 'mcelog --ascii'\n" ); }
@@ -508,9 +508,9 @@ static int mce_usable_address(struct mce *m)bool mce_is_memory_error (struct mce *m) { -if (m->cpuvendor ==X86_VENDOR_AMD ) { +if (m->cpuvendor ==X86_VENDOR_AMD || + m->cpuvendor ==X86_VENDOR_HYGON ) {return amd_mce_is_memory_error (m); - }else if (m->cpuvendor ==X86_VENDOR_INTEL ) {/* * Intel SDM Volume 3B - 15.9.2 Compound Error Codes
@@ -539,6 +539,9 @@ static bool mce_is_correctable(struct mce *m)if (m->cpuvendor ==X86_VENDOR_AMD && m->status &MCI_STATUS_DEFERRED )return false; +if (m->cpuvendor ==X86_VENDOR_HYGON && m->status &MCI_STATUS_DEFERRED ) +return false; +if (m->status &MCI_STATUS_UC )return false;
@@ -1705,7 +1708,7 @@ static int __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c)*/ static void __mcheck_cpu_init_early (struct cpuinfo_x86 *c) { -if (c->x86_vendor ==X86_VENDOR_AMD ) { +if (c->x86_vendor ==X86_VENDOR_AMD || c->x86_vendor ==X86_VENDOR_HYGON ) { mce_flags.overflow_recov = !!cpu_has (c,X86_FEATURE_OVERFLOW_RECOV ); mce_flags.succor = !!cpu_has (c,X86_FEATURE_SUCCOR ); mce_flags.smca = !!cpu_has (c,X86_FEATURE_SMCA );
@@ -1746,6 +1749,11 @@ static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)mce_amd_feature_init (c);break ; } + +case X86_VENDOR_HYGON : +mce_hygon_feature_init (c); +break ; +case X86_VENDOR_CENTAUR :mce_centaur_feature_init (c);break ;
@@ -1971,12 +1979,14 @@ static void mce_disable_error_reporting(void)static void vendor_disable_error_reporting (void ) {/* -* Don't clear on Intel or AMD CPUs. Some of these MSRs are socket-wide. +* Don't clear on Intel or AMD or Hygon CPUs. Some of these MSRs +* are socket-wide. * Disabling them for just a single offlined CPU is bad, since it will * inhibit reporting for all shared resources on the socket like the * last level cache (LLC), the integrated memory controller (iMC), etc. */ if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL || + boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON || boot_cpu_data.x86_vendor ==X86_VENDOR_AMD )return ;
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/kvm: Add Hygon Dhyana support to KVM
In reply to Pu Wenx86/kvm: Add Hygon Dhyana support to KVM
Commit-ID: b8f4abb652146ddde04ab6e2a80e8cde27ff4470 Gitweb: https://git.kernel.org/tip/b8f4abb652146ddde04ab6e2a80e8cde27ff4470 Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:36:31 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:59 +0200
x86/kvm: Add Hygon Dhyana support to KVM
The Hygon Dhyana CPU has the SVM feature as AMD family 17h does. So enable the KVM infrastructure support to it.
--- arch/x86/include/asm/kvm_emulate.h | 4 ++++ arch/x86/include/asm/virtext.h | 5 +++-- arch/x86/kvm/emulate.c | 11 ++++++++++- 3 files changed, 17 insertions(+), 3 deletions(-)
arch/x86/include/asm/kvm_emulate.h +4 -0
@@ -364,6 +364,10 @@ struct x86_emulate_ctxt {#define X86EMUL_CPUID_VENDOR_AMDisbetterI_ecx 0x21726574#define X86EMUL_CPUID_VENDOR_AMDisbetterI_edx 0x74656273 +#define X86EMUL_CPUID_VENDOR_HygonGenuine_ebx 0x6f677948 +#define X86EMUL_CPUID_VENDOR_HygonGenuine_ecx 0x656e6975 +#define X86EMUL_CPUID_VENDOR_HygonGenuine_edx 0x6e65476e +#define X86EMUL_CPUID_VENDOR_GenuineIntel_ebx 0x756e6547#define X86EMUL_CPUID_VENDOR_GenuineIntel_ecx 0x6c65746e#define X86EMUL_CPUID_VENDOR_GenuineIntel_edx 0x49656e69
arch/x86/include/asm/virtext.h +3 -2
@@ -83,9 +83,10 @@ static inline void cpu_emergency_vmxoff(void)*/ static inline int cpu_has_svm (const char **msg) { -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD && + boot_cpu_data.x86_vendor !=X86_VENDOR_HYGON ) {if (msg) - *msg ="not amd" ; + *msg ="not amd or hygon" ;return 0 ; }
arch/x86/kvm/emulate.c +10 -1
@@ -2711,7 +2711,16 @@ static bool em_syscall_is_enabled(struct x86_emulate_ctxt *ctxt) edx == X86EMUL_CPUID_VENDOR_AMDisbetterI_edx)return true; -/* default: (not Intel, not AMD), apply Intel's stricter rules... */ +/* Hygon ("HygonGenuine") */ +if (ebx == X86EMUL_CPUID_VENDOR_HygonGenuine_ebx && + ecx == X86EMUL_CPUID_VENDOR_HygonGenuine_ecx && + edx == X86EMUL_CPUID_VENDOR_HygonGenuine_edx) +return true; + +/* +* default: (not Intel, not AMD, not Hygon), apply Intel's +* stricter rules... +*/ return false; }
tip-bot for Pu Wen
merge bot[tip:x86/cpu] x86/xen: Add Hygon Dhyana support to Xen
In reply to Pu Wenx86/xen: Add Hygon Dhyana support to Xen
Commit-ID: 4044240365e85ef7ae43a6dc454669b57853124c Gitweb: https://git.kernel.org/tip/4044240365e85ef7ae43a6dc454669b57853124c Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:36:46 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:28:59 +0200
x86/xen: Add Hygon Dhyana support to Xen
To make Xen work on the Hygon platform, reuse AMD's Xen support code path for Hygon Dhyana CPU.
There are six core performance events counters per thread, so there are six MSRs for these counters. Also there are four legacy PMC MSRs, they are aliases of the counters.
In this version, use the legacy and safe version of MSR access. Tested successfully with VPMU enabled in Xen on Hygon platform by testing with perf.
--- arch/x86/xen/pmu.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
arch/x86/xen/pmu.c +9 -3
@@ -90,6 +90,12 @@ static void xen_pmu_arch_init(void) k7_counters_mirrored =0 ;break ; } + }else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) { + amd_num_counters =F10H_NUM_COUNTERS ; + amd_counters_base =MSR_K7_PERFCTR0 ; + amd_ctrls_base =MSR_K7_EVNTSEL0 ; + amd_msr_step =1 ; + k7_counters_mirrored =0 ; }else {uint32_t eax, ebx, ecx, edx;
@@ -285,7 +291,7 @@ static bool xen_amd_pmu_emulate(unsigned int msr, u64 *val, bool is_read)bool pmu_msr_read (unsigned int msr,uint64_t *val,int *err) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL ) {if (is_amd_pmu_msr (msr)) {if (!xen_amd_pmu_emulate (msr, val,1 )) *val =native_read_msr_safe (msr, err);
@@ -308,7 +314,7 @@ bool pmu_msr_write(unsigned int msr, uint32_t low, uint32_t high, int *err) {uint64_t val = ((uint64_t )high <<32 ) | low; -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) { +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL ) {if (is_amd_pmu_msr (msr)) {if (!xen_amd_pmu_emulate (msr, &val,0 )) *err =native_write_msr_safe (msr, low, high);
@@ -379,7 +385,7 @@ static unsigned long long xen_intel_read_pmc(int counter)unsigned long long xen_read_pmc (int counter) { -if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL )return xen_amd_read_pmc (counter);else return xen_intel_read_pmc (counter);
tip-bot for Pu Wen
merge bot[tip:x86/cpu] ACPI: Add Hygon Dhyana support
In reply to Pu WenACPI: Add Hygon Dhyana support
Commit-ID: 7377ed4bd56e6cc1ddbb63f03626fc5b92d3d6fe Gitweb: https://git.kernel.org/tip/7377ed4bd56e6cc1ddbb63f03626fc5b92d3d6fe Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:37:05 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:29:00 +0200
ACPI: Add Hygon Dhyana support
The Hygon Dhyana CPU has NONSTOP TSC feature, so enable the ACPI driver support to it.
--- drivers/acpi/acpi_pad.c | 1 + drivers/acpi/processor_idle.c | 1 + 2 files changed, 2 insertions(+)
drivers/acpi/acpi_pad.c +1 -0
@@ -70,6 +70,7 @@ static void power_saving_mwait_init(void)#if defined(CONFIG_X86 )switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :case X86_VENDOR_INTEL :/*
drivers/acpi/processor_idle.c +1 -0
@@ -205,6 +205,7 @@ static void lapic_timer_state_broadcast(struct acpi_processor *pr,static void tsc_check_state (int state) {switch (boot_cpu_data.x86_vendor) { +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :case X86_VENDOR_INTEL :case X86_VENDOR_CENTAUR :
tip-bot for Pu Wen
merge bot[tip:x86/cpu] cpufreq: Add Hygon Dhyana support
In reply to Pu Wencpufreq: Add Hygon Dhyana support
Commit-ID: cc9690cfc7a36873b219d569049e10f073dd22e4 Gitweb: https://git.kernel.org/tip/cc9690cfc7a36873b219d569049e10f073dd22e4 Author: Pu Wen <[email protected]> AuthorDate: Sun, 23 Sep 2018 17:37:38 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 27 Sep 2018 18:29:00 +0200
cpufreq: Add Hygon Dhyana support
The Hygon Dhyana CPU supports ACPI P-States, and there is SMBus device (PCI device ID 0x790b) on the Hygon platform. Add Hygon Dhyana support to the cpufreq driver by using the code path of AMD family 17h.
--- drivers/cpufreq/acpi-cpufreq.c | 5 +++++ drivers/cpufreq/amd_freq_sensitivity.c | 9 +++++++-- 2 files changed, 12 insertions(+), 2 deletions(-)
drivers/cpufreq/acpi-cpufreq.c +5 -0
@@ -61,6 +61,7 @@ enum {#define INTEL_MSR_RANGE (0xffff)#define AMD_MSR_RANGE (0x7) +#define HYGON_MSR_RANGE (0x7)#define MSR_K7_HWCR_CPB_DIS (1ULL << 25)
@@ -95,6 +96,7 @@ static bool boost_state(unsigned int cpu)rdmsr_on_cpu (cpu,MSR_IA32_MISC_ENABLE , &lo, &hi); msr = lo | ((u64 )hi <<32 );return !(msr &MSR_IA32_MISC_ENABLE_TURBO_DISABLE ); +case X86_VENDOR_HYGON :case X86_VENDOR_AMD :rdmsr_on_cpu (cpu,MSR_K7_HWCR , &lo, &hi); msr = lo | ((u64 )hi <<32 );
@@ -113,6 +115,7 @@ static int boost_set_msr(bool enable) msr_addr =MSR_IA32_MISC_ENABLE ; msr_mask =MSR_IA32_MISC_ENABLE_TURBO_DISABLE ;break ; +case X86_VENDOR_HYGON :case X86_VENDOR_AMD : msr_addr =MSR_K7_HWCR ; msr_mask =MSR_K7_HWCR_CPB_DIS ;
@@ -225,6 +228,8 @@ static unsigned extract_msr(struct cpufreq_policy *policy, u32 msr)if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) msr &=AMD_MSR_RANGE ; +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + msr &=HYGON_MSR_RANGE ;else msr &=INTEL_MSR_RANGE ;
drivers/cpufreq/amd_freq_sensitivity.c +7 -2
@@ -111,11 +111,16 @@ static int __init amd_freq_sensitivity_init(void) {u64 val;struct pci_dev *pcidev; +unsigned int pci_vendor; -if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) + pci_vendor =PCI_VENDOR_ID_AMD ; +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_HYGON ) + pci_vendor =PCI_VENDOR_ID_HYGON ; +else return -ENODEV ; - pcidev =pci_get_device (PCI_VENDOR_ID_AMD , + pcidev =pci_get_device (pci_vendor,PCI_DEVICE_ID_AMD_KERNCZ_SMBUS ,NULL );if (!pcidev) {
Re: [PATCH v6 16/16] cpupower, x86: Add Hygon Dhyana support
In reply to Pu WenQuoted 6 lines · Pu Wen
Tool cpupower is useful to get CPU frequency information and monitor power stats on Hygon Dhyana platforms. So add Hygon Dhyana support to it by checking vendor and family to share the code path of AMD family 17h.
Signed-off-by Pu Wen <[email protected]>
Looks good to me. I am assuming it will go into x86 with the rest of the series.
thanks,
-- Shuah
Pu Wen
maintainerPATCH RESEND v8 16/16 cpupower: Add Hygon Dhyana support
Tool cpupower is useful to get CPU frequency information and monitor power stats on Hygon Dhyana platform. So add Hygon Dhyana support to it by checking vendor and family to share the code path of AMD family 17h.
--- tools/power/cpupower/utils/cpufreq-info.c | 6 ++++-- tools/power/cpupower/utils/helpers/amd.c | 4 ++-- tools/power/cpupower/utils/helpers/cpuid.c | 8 +++++--- tools/power/cpupower/utils/helpers/helpers.h | 2 +- tools/power/cpupower/utils/helpers/misc.c | 2 +- tools/power/cpupower/utils/idle_monitor/mperf_monitor.c | 3 ++- 6 files changed, 15 insertions(+), 10 deletions(-)
tools/power/cpupower/utils/cpufreq-info.c +4 -2
@@ -170,6 +170,7 @@ static int get_boost_mode(unsigned int cpu)unsigned long pstates[MAX_HW_PSTATES ] = {0 ,};if (cpupower_cpu_info.vendor !=X86_VENDOR_AMD && + cpupower_cpu_info.vendor !=X86_VENDOR_HYGON && cpupower_cpu_info.vendor !=X86_VENDOR_INTEL )return 0 ;
@@ -190,8 +191,9 @@ static int get_boost_mode(unsigned int cpu)printf (_ (" Supported: %s\n" ), support ?_ ("yes" ) :_ ("no" ));printf (_ (" Active: %s\n" ), active ?_ ("yes" ) :_ ("no" )); -if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD && - cpupower_cpu_info.family >=0x10 ) { +if ((cpupower_cpu_info.vendor ==X86_VENDOR_AMD && + cpupower_cpu_info.family >=0x10 ) || + cpupower_cpu_info.vendor ==X86_VENDOR_HYGON ) { ret =decode_pstates (cpu, cpupower_cpu_info.family, b_states, pstates, &pstate_no);if (ret)
tools/power/cpupower/utils/helpers/amd.c +2 -2
@@ -45,7 +45,7 @@ static int get_did(int family, union msr_pstate pstate)if (family ==0x12 ) t = pstate.val &0xf ; -else if (family ==0x17 ) +else if (family ==0x17 || family ==0x18 ) t = pstate.fam17h_bits.did;else t = pstate.bits.did;
@@ -59,7 +59,7 @@ static int get_cof(int family, union msr_pstate pstate)int fid, did, cof; did =get_did (family, pstate); -if (family ==0x17 ) { +if (family ==0x17 || family ==0x18 ) { fid = pstate.fam17h_bits.fid; cof =200 * fid / did; }else {
tools/power/cpupower/utils/helpers/cpuid.c +5 -3
@@ -8,7 +8,7 @@#include "helpers/helpers.h" static const char *cpu_vendor_table[X86_VENDOR_MAX ] = { -"Unknown" ,"GenuineIntel" ,"AuthenticAMD" , +"Unknown" ,"GenuineIntel" ,"AuthenticAMD" ,"HygonGenuine" , };#if defined(__i386__) || defined(__x86_64__)
@@ -109,6 +109,7 @@ int get_cpu_info(struct cpupower_cpu_info *cpu_info)fclose (fp);/* Get some useful CPU capabilities from cpuid */ if (cpu_info->vendor !=X86_VENDOR_AMD && + cpu_info->vendor !=X86_VENDOR_HYGON && cpu_info->vendor !=X86_VENDOR_INTEL )return ret;
@@ -124,8 +125,9 @@ int get_cpu_info(struct cpupower_cpu_info *cpu_info)if (cpuid_level >=6 && (cpuid_ecx (6 ) &0x1 )) cpu_info->caps |=CPUPOWER_CAP_APERF ; -/* AMD Boost state enable/disable register */ -if (cpu_info->vendor ==X86_VENDOR_AMD ) { +/* AMD or Hygon Boost state enable/disable register */ +if (cpu_info->vendor ==X86_VENDOR_AMD || + cpu_info->vendor ==X86_VENDOR_HYGON ) {if (ext_cpuid_level >=0x80000007 && (cpuid_edx (0x80000007 ) & (1 <<9 ))) cpu_info->caps |=CPUPOWER_CAP_AMD_CBP ;
tools/power/cpupower/utils/helpers/helpers.h +1 -1
@@ -61,7 +61,7 @@ extern int be_verbose;/* cpuid and cpuinfo helpers **************************/ enum cpupower_cpu_vendor {X86_VENDOR_UNKNOWN =0 ,X86_VENDOR_INTEL , -X86_VENDOR_AMD ,X86_VENDOR_MAX }; +X86_VENDOR_AMD ,X86_VENDOR_HYGON ,X86_VENDOR_MAX };#define CPUPOWER_CAP_INV_TSC 0x00000001#define CPUPOWER_CAP_APERF 0x00000002
tools/power/cpupower/utils/helpers/misc.c +1 -1
@@ -26,7 +26,7 @@ int cpufreq_has_boost_support(unsigned int cpu, int *support, int *active,* has Hardware determined variable increments instead. */ -if (cpu_info.family ==0x17 ) { +if (cpu_info.family ==0x17 || cpu_info.family ==0x18 ) {if (!read_msr (cpu,MSR_AMD_HWCR , &val)) {if (!(val &CPUPOWER_AMD_CPBDIS )) *active =1 ;
tools/power/cpupower/utils/idle_monitor/mperf_monitor.c +2 -1
@@ -241,7 +241,8 @@ static int init_maxfreq_mode(void)if (!(cpupower_cpu_info.caps &CPUPOWER_CAP_INV_TSC )) goto use_sysfs; -if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD ) { +if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD || + cpupower_cpu_info.vendor ==X86_VENDOR_HYGON ) {/* MSR_AMD_HWCR tells us whether TSC runs at P0/mperf * freq. * A test whether hwcr is accessable/available would be:
Signature · 1 line
2.7.4
tip-bot for Pu Wen
merge bot[tip:x86/cpu] tools/cpupower: Add Hygon Dhyana support
tools/cpupower: Add Hygon Dhyana support
Commit-ID: 995d5f64b62f20f05b8e0972f07ec4d6c23333c9 Gitweb: https://git.kernel.org/tip/995d5f64b62f20f05b8e0972f07ec4d6c23333c9 Author: Pu Wen <[email protected]> AuthorDate: Thu, 4 Oct 2018 09:21:43 +0800 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 4 Oct 2018 09:57:25 +0200
tools/cpupower: Add Hygon Dhyana support
The tool cpupower is useful to get CPU frequency information and monitor power stats on the Hygon Dhyana platform. So add Hygon Dhyana support to it by checking vendor and family to share the code path of AMD family 17h.
--- tools/power/cpupower/utils/cpufreq-info.c | 6 ++++-- tools/power/cpupower/utils/helpers/amd.c | 4 ++-- tools/power/cpupower/utils/helpers/cpuid.c | 8 +++++--- tools/power/cpupower/utils/helpers/helpers.h | 2 +- tools/power/cpupower/utils/helpers/misc.c | 2 +- tools/power/cpupower/utils/idle_monitor/mperf_monitor.c | 3 ++- 6 files changed, 15 insertions(+), 10 deletions(-)
tools/power/cpupower/utils/cpufreq-info.c +4 -2
@@ -170,6 +170,7 @@ static int get_boost_mode(unsigned int cpu)unsigned long pstates[MAX_HW_PSTATES ] = {0 ,};if (cpupower_cpu_info.vendor !=X86_VENDOR_AMD && + cpupower_cpu_info.vendor !=X86_VENDOR_HYGON && cpupower_cpu_info.vendor !=X86_VENDOR_INTEL )return 0 ;
@@ -190,8 +191,9 @@ static int get_boost_mode(unsigned int cpu)printf (_ (" Supported: %s\n" ), support ?_ ("yes" ) :_ ("no" ));printf (_ (" Active: %s\n" ), active ?_ ("yes" ) :_ ("no" )); -if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD && - cpupower_cpu_info.family >=0x10 ) { +if ((cpupower_cpu_info.vendor ==X86_VENDOR_AMD && + cpupower_cpu_info.family >=0x10 ) || + cpupower_cpu_info.vendor ==X86_VENDOR_HYGON ) { ret =decode_pstates (cpu, cpupower_cpu_info.family, b_states, pstates, &pstate_no);if (ret)
tools/power/cpupower/utils/helpers/amd.c +2 -2
@@ -45,7 +45,7 @@ static int get_did(int family, union msr_pstate pstate)if (family ==0x12 ) t = pstate.val &0xf ; -else if (family ==0x17 ) +else if (family ==0x17 || family ==0x18 ) t = pstate.fam17h_bits.did;else t = pstate.bits.did;
@@ -59,7 +59,7 @@ static int get_cof(int family, union msr_pstate pstate)int fid, did, cof; did =get_did (family, pstate); -if (family ==0x17 ) { +if (family ==0x17 || family ==0x18 ) { fid = pstate.fam17h_bits.fid; cof =200 * fid / did; }else {
tools/power/cpupower/utils/helpers/cpuid.c +5 -3
@@ -8,7 +8,7 @@#include "helpers/helpers.h" static const char *cpu_vendor_table[X86_VENDOR_MAX ] = { -"Unknown" ,"GenuineIntel" ,"AuthenticAMD" , +"Unknown" ,"GenuineIntel" ,"AuthenticAMD" ,"HygonGenuine" , };#if defined(__i386__) || defined(__x86_64__)
@@ -109,6 +109,7 @@ out:fclose (fp);/* Get some useful CPU capabilities from cpuid */ if (cpu_info->vendor !=X86_VENDOR_AMD && + cpu_info->vendor !=X86_VENDOR_HYGON && cpu_info->vendor !=X86_VENDOR_INTEL )return ret;
@@ -124,8 +125,9 @@ out:if (cpuid_level >=6 && (cpuid_ecx (6 ) &0x1 )) cpu_info->caps |=CPUPOWER_CAP_APERF ; -/* AMD Boost state enable/disable register */ -if (cpu_info->vendor ==X86_VENDOR_AMD ) { +/* AMD or Hygon Boost state enable/disable register */ +if (cpu_info->vendor ==X86_VENDOR_AMD || + cpu_info->vendor ==X86_VENDOR_HYGON ) {if (ext_cpuid_level >=0x80000007 && (cpuid_edx (0x80000007 ) & (1 <<9 ))) cpu_info->caps |=CPUPOWER_CAP_AMD_CBP ;
tools/power/cpupower/utils/helpers/helpers.h +1 -1
@@ -61,7 +61,7 @@ extern int be_verbose;/* cpuid and cpuinfo helpers **************************/ enum cpupower_cpu_vendor {X86_VENDOR_UNKNOWN =0 ,X86_VENDOR_INTEL , -X86_VENDOR_AMD ,X86_VENDOR_MAX }; +X86_VENDOR_AMD ,X86_VENDOR_HYGON ,X86_VENDOR_MAX };#define CPUPOWER_CAP_INV_TSC 0x00000001#define CPUPOWER_CAP_APERF 0x00000002
tools/power/cpupower/utils/helpers/misc.c +1 -1
@@ -26,7 +26,7 @@ int cpufreq_has_boost_support(unsigned int cpu, int *support, int *active,* has Hardware determined variable increments instead. */ -if (cpu_info.family ==0x17 ) { +if (cpu_info.family ==0x17 || cpu_info.family ==0x18 ) {if (!read_msr (cpu,MSR_AMD_HWCR , &val)) {if (!(val &CPUPOWER_AMD_CPBDIS )) *active =1 ;
tools/power/cpupower/utils/idle_monitor/mperf_monitor.c +2 -1
@@ -241,7 +241,8 @@ static int init_maxfreq_mode(void)if (!(cpupower_cpu_info.caps &CPUPOWER_CAP_INV_TSC )) goto use_sysfs; -if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD ) { +if (cpupower_cpu_info.vendor ==X86_VENDOR_AMD || + cpupower_cpu_info.vendor ==X86_VENDOR_HYGON ) {/* MSR_AMD_HWCR tells us whether TSC runs at P0/mperf * freq. * A test whether hwcr is accessable/available would be: