amd-pstate doesn't work since v5.18.11
Hello Mario.
The following commits were pulled into v5.18.11:
``` $ git log --oneline --no-merges v5.18.10..v5.18.11 | grep ACPI 2783414e6ef7 ACPI: CPPC: Don't require _OSC if X86_FEATURE_CPPC is supported 3068cfeca3b5 ACPI: CPPC: Only probe for _CPC if CPPC v2 is acked 8beb71759cc8 ACPI: bus: Set CPPC _OSC bits for all and when CPPC_LIB is supported 13bb696dd2f3 ACPI: CPPC: Check _OSC for flexible address space ```
and now this happens:
``` $ sudo modprobe amd-pstate shared_mem=1 modprobe: ERROR: could not insert 'amd_pstate': No such device ```
With v5.18.10 this worked just fine.
In your upstream commit 8b356e536e69f3a4d6778ae9f0858a1beadabb1f you write:
``` If there is additional breakage on the shared memory designs also missing this _OSC, additional follow up changes may be needed. ```
So the question is what else should be pulled into the stable tree to unbreak amd-pstate?
Thanks.
Signature · 1 line
Oleksandr Natalenko (post-factum)
[AMD Official Use Only - General]
Hi Oleksandr:
Quoted 42 lines
-----Original Message-----
From: Oleksandr Natalenko <[email protected]>
Sent: Wednesday, July 13, 2022 1:40 AM
To: Limonciello, Mario <[email protected]>; linux- [email protected]
Cc: Yuan, Perry <[email protected]>; Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: [REGRESSION] amd-pstate doesn't work since v5.18.11[CAUTION: External Email]
Hello Mario.
The following commits were pulled into v5.18.11:
``` $ git log --oneline --no-merges v5.18.10..v5.18.11 | grep ACPI 2783414e6ef7 ACPI: CPPC: Don't require _OSC if X86_FEATURE_CPPC is supported 3068cfeca3b5 ACPI: CPPC: Only probe for _CPC if CPPC v2 is acked 8beb71759cc8 ACPI: bus: Set CPPC _OSC bits for all and when CPPC_LIB is supported 13bb696dd2f3 ACPI: CPPC: Check _OSC for flexible address space ``` and now this happens: ``` $ sudo modprobe amd-pstate shared_mem=1 modprobe: ERROR: could not insert 'amd_pstate': No such device ```With v5.18.10 this worked just fine.
In your upstream commit
8b356e536e69f3a4d6778ae9f0858a1beadabb1fyou write:``` If there is additional breakage on the shared memory designs also missing this _OSC, additional follow up changes may be needed. ```
Quoted 8 lines
So the question is what else should be pulled into the stable tree to unbreak amd-pstate?
Thanks.
--
Oleksandr Natalenko (post-factum)
Could you share the lscpu output ?
Perry.
[Public]
+ Ray
Quoted 61 lines
-----Original Message-----
From: Yuan, Perry <[email protected]>
Sent: Tuesday, July 12, 2022 12:50
To: Oleksandr Natalenko <[email protected]>; Limonciello, Mario <[email protected]>; [email protected]
Cc: Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: RE: [REGRESSION] amd-pstate doesn't work since v5.18.11[AMD Official Use Only - General]
Hi Oleksandr:
Quoted 47 lines
-----Original Message-----
From: Oleksandr Natalenko <[email protected]>
Sent: Wednesday, July 13, 2022 1:40 AM
To: Limonciello, Mario <[email protected]>; linux- [email protected]
Cc: Yuan, Perry <[email protected]>; Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: [REGRESSION] amd-pstate doesn't work since v5.18.11[CAUTION: External Email]
Hello Mario.
The following commits were pulled into v5.18.11:
``` $ git log --oneline --no-merges v5.18.10..v5.18.11 | grep ACPI 2783414e6ef7 ACPI: CPPC: Don't require _OSC if X86_FEATURE_CPPC is supported 3068cfeca3b5 ACPI: CPPC: Only probe for _CPC if CPPC v2 is acked 8beb71759cc8 ACPI: bus: Set CPPC _OSC bits for all and when CPPC_LIB is supported 13bb696dd2f3 ACPI: CPPC: Check _OSC for flexible address space ``` and now this happens: ``` $ sudo modprobe amd-pstate shared_mem=1 modprobe: ERROR: could not insert 'amd_pstate': No such device ```With v5.18.10 this worked just fine.
In your upstream commit
8b356e536e69f3a4d6778ae9f0858a1beadabb1fyou write:``` If there is additional breakage on the shared memory designs also missing this _OSC, additional follow up changes may be needed. ```So the question is what else should be pulled into the stable tree to unbreak amd-pstate?
Thanks.
Quoted 7 lines
Quoted 3 lines
--
Oleksandr Natalenko (post-factum)Could you share the lscpu output ?
Perry.
Thanks this is the sort of thing I was worried might happen as a result of requiring the _OSC. It was introduced as part of that commit 8beb71759cc8.
To solve it I think we need to add more things to cpc_supported_by_cpu
(https://github.com/torvalds/linux/blob/525496a030de4ae64bb9e1d6bfc88eec6f5fe6e2/arch/x86/kernel/acpi/cppc.c#L19)
The question is how do we safely detect the shared memory designs? These are a fixed quantity as newer designs /should/ be using the MSR.
I am tending to thing that unfortunately we need to have an allow-list of shared memory design here unless someone has other ideas.
Hello.
Quoted 70 lines · Mario
[Public]
+ Ray
Quoted 66 lines
-----Original Message-----
From: Yuan, Perry <[email protected]>
Sent: Tuesday, July 12, 2022 12:50
To: Oleksandr Natalenko <[email protected]>; Limonciello, Mario <[email protected]>; [email protected]
Cc: Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: RE: [REGRESSION] amd-pstate doesn't work since v5.18.11[AMD Official Use Only - General]
Hi Oleksandr:
Quoted 42 lines
-----Original Message-----
From: Oleksandr Natalenko <[email protected]>
Sent: Wednesday, July 13, 2022 1:40 AM
To: Limonciello, Mario <[email protected]>; linux- [email protected]
Cc: Yuan, Perry <[email protected]>; Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: [REGRESSION] amd-pstate doesn't work since v5.18.11[CAUTION: External Email]
Hello Mario.
The following commits were pulled into v5.18.11:
``` $ git log --oneline --no-merges v5.18.10..v5.18.11 | grep ACPI 2783414e6ef7 ACPI: CPPC: Don't require _OSC if X86_FEATURE_CPPC is supported 3068cfeca3b5 ACPI: CPPC: Only probe for _CPC if CPPC v2 is acked 8beb71759cc8 ACPI: bus: Set CPPC _OSC bits for all and when CPPC_LIB is supported 13bb696dd2f3 ACPI: CPPC: Check _OSC for flexible address space ``` and now this happens: ``` $ sudo modprobe amd-pstate shared_mem=1 modprobe: ERROR: could not insert 'amd_pstate': No such device ```With v5.18.10 this worked just fine.
In your upstream commit
8b356e536e69f3a4d6778ae9f0858a1beadabb1fyou write:``` If there is additional breakage on the shared memory designs also missing this _OSC, additional follow up changes may be needed. ```Quoted 8 lines
So the question is what else should be pulled into the stable tree to unbreak amd-pstate?
Thanks.
--
Oleksandr Natalenko (post-factum)Could you share the lscpu output ?
Here's my lscpu:
``` Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 43 bits physical, 48 bits virtual Byte Order: Little Endian CPU(s): 24 On-line CPU(s) list: 0-23 Vendor ID: AuthenticAMD Model name: AMD Ryzen 9 3900XT 12-Core Processor CPU family: 23 Model: 113 Thread(s) per core: 2 Core(s) per socket: 12 Socket(s): 1 Stepping: 0 Frequency boost: enabled CPU(s) scaling MHz: 59% CPU max MHz: 3800,0000 CPU min MHz: 2200,0000 BogoMIPS: 7589.71 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr rdpru wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl umip rdpid overflow_recov succor smca sev sev_es Virtualization: AMD-V L1d cache: 384 KiB (12 instances) L1i cache: 384 KiB (12 instances) L2 cache: 6 MiB (12 instances) L3 cache: 64 MiB (4 instances) NUMA node(s): 1 NUMA node0 CPU(s): 0-23 Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Retpolines, IBPB conditional, STIBP conditional, RSB filling Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected ```
Quoted 5 lines
Quoted 1 line
Perry.
Thanks this is the sort of thing I was worried might happen as a result of requiring the
_OSC. It was introduced as part of that commit8beb71759cc8.
Quoted 8 lines
To solve it I think we need to add more things to
cpc_supported_by_cpu
(https://github.com/torvalds/linux/blob/525496a030de4ae64bb9e1d6bfc88eec6f5fe6e2/arch/x86/kernel/acpi/cppc.c#L19)The question is how do we safely detect the shared memory designs? These are a fixed quantity as newer designs /should/ be using the MSR.
I am tending to thing that unfortunately we need to have an allow-list of shared memory design here unless someone has other ideas.
Happy to test any patches as needed.
Signature · 1 line
Oleksandr Natalenko (post-factum)
Quoted 125 lines · Oleksandr Natalenko
Hello.
Quoted 70 lines · Mario
[Public]
+ Ray
Quoted 66 lines
-----Original Message-----
From: Yuan, Perry <[email protected]>
Sent: Tuesday, July 12, 2022 12:50
To: Oleksandr Natalenko <[email protected]>; Limonciello, Mario <[email protected]>; [email protected]
Cc: Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: RE: [REGRESSION] amd-pstate doesn't work since v5.18.11[AMD Official Use Only - General]
Hi Oleksandr:
Quoted 42 lines
-----Original Message-----
From: Oleksandr Natalenko <[email protected]>
Sent: Wednesday, July 13, 2022 1:40 AM
To: Limonciello, Mario <[email protected]>; linux- [email protected]
Cc: Yuan, Perry <[email protected]>; Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: [REGRESSION] amd-pstate doesn't work since v5.18.11[CAUTION: External Email]
Hello Mario.
The following commits were pulled into v5.18.11:
``` $ git log --oneline --no-merges v5.18.10..v5.18.11 | grep ACPI 2783414e6ef7 ACPI: CPPC: Don't require _OSC if X86_FEATURE_CPPC is supported 3068cfeca3b5 ACPI: CPPC: Only probe for _CPC if CPPC v2 is acked 8beb71759cc8 ACPI: bus: Set CPPC _OSC bits for all and when CPPC_LIB is supported 13bb696dd2f3 ACPI: CPPC: Check _OSC for flexible address space ``` and now this happens: ``` $ sudo modprobe amd-pstate shared_mem=1 modprobe: ERROR: could not insert 'amd_pstate': No such device ```With v5.18.10 this worked just fine.
In your upstream commit
8b356e536e69f3a4d6778ae9f0858a1beadabb1fyou write:``` If there is additional breakage on the shared memory designs also missing this _OSC, additional follow up changes may be needed. ```Quoted 8 lines
So the question is what else should be pulled into the stable tree to unbreak amd-pstate?
Thanks.
--
Oleksandr Natalenko (post-factum)Could you share the lscpu output ?
Here's my
lscpu:``` Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 43 bits physical, 48 bits virtual Byte Order: Little Endian CPU(s): 24 On-line CPU(s) list: 0-23 Vendor ID: AuthenticAMD Model name: AMD Ryzen 9 3900XT 12-Core Processor CPU family: 23 Model: 113 Thread(s) per core: 2 Core(s) per socket: 12 Socket(s): 1 Stepping: 0 Frequency boost: enabled CPU(s) scaling MHz: 59% CPU max MHz: 3800,0000 CPU min MHz: 2200,0000 BogoMIPS: 7589.71 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr rdpru wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl umip rdpid overflow_recov succor smca sev sev_es Virtualization: AMD-V L1d cache: 384 KiB (12 instances) L1i cache: 384 KiB (12 instances) L2 cache: 6 MiB (12 instances) L3 cache: 64 MiB (4 instances) NUMA node(s): 1 NUMA node0 CPU(s): 0-23 Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Retpolines, IBPB conditional, STIBP conditional, RSB filling Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected ```Quoted 8 lines
Quoted 1 line
Perry.
Thanks this is the sort of thing I was worried might happen as a result of requiring the
_OSC. It was introduced as part of that commit8beb71759cc8.To solve it I think we need to add more things to
cpc_supported_by_cpu
(https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Ftorvalds%2Flinux%2Fblob%2F525496a030de4ae64bb9e1d6bfc88eec6f5fe6e2%2Farch%2Fx86%2Fkernel%2Facpi%2Fcppc.c%23L19&data=05%7C01%7CMario.Limonciello%40amd.com%7C96addaab0edc4e22779908da642f84ac%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637932453099304670%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=4KHD3UUlfDJEmpTpqDCmuV1x%2F7nj%2F0iuhwdnhJqtQeU%3D&reserved=0)
Quoted 8 lines
Quoted 5 lines
The question is how do we safely detect the shared memory designs? These are a fixed quantity as newer designs /should/ be using the MSR.
I am tending to thing that unfortunately we need to have an allow-list of shared memory design here unless someone has other ideas.
Happy to test any patches as needed.
See if this helps out:
arch/x86/kernel/acpi/cppc.c +7 -0
@@ -16,6 +16,13 @@ bool cpc_supported_by_cpu(void)switch (boot_cpu_data.x86_vendor) {case X86_VENDOR_AMD :case X86_VENDOR_HYGON : +if (boot_cpu_data.x86 ==0x19 && + ((boot_cpu_data.x86_model >=0x00 && boot_cpu_data.x86_model <=0x0f ) || + (boot_cpu_data.x86_model >=0x20 && boot_cpu_data.x86_model <=0x2f ))) +return true; +else if (boot_cpu_data.x86 ==0x17 && + boot_cpu_data.x86_model >=0x70 && boot_cpu_data.x86_model <=0x7f ) +return true;return boot_cpu_has (X86_FEATURE_CPPC ); }return false;
If that works and no one has a better idea how to do it for these systems I'll send out a proper proper patch tomorrow.
Hello.
Quoted 19 lines · Mario
See if this helps out:
arch/x86/kernel/acpi/cppc.c +2
-0@@ -16,6 +16,13 @@ bool cpc_supported_by_cpu(void)switch (boot_cpu_data.x86_vendor) {case X86_VENDOR_AMD :case X86_VENDOR_HYGON : + if (boot_cpu_data.x86 == 0x19 && + ((boot_cpu_data.x86_model >= 0x00 &&boot_cpu_data.x86_model <=
0x0f) ||
+ (boot_cpu_data.x86_model >=0x20&&
boot_cpu_data.x86_model <=0x2f)))arch/x86/kernel/acpi/cppc.c +3
-0+return true; + else if (boot_cpu_data.x86 == 0x17 && + boot_cpu_data.x86_model >= 0x70 &&boot_cpu_data.x86_model <=
0x7f)
Quoted 7 lines
+ return true; return boot_cpu_has(X86_FEATURE_CPPC); } return false;If that works and no one has a better idea how to do it for these systems I'll send out a proper proper patch tomorrow.
This patch (also mirrored here [1]) works for me, thanks.
[1] https://codeberg.org/pf-kernel/linux/commit/cc5c57b8c349c85b81b44cf5fe2d2577b947a7de
Signature · 1 line
Oleksandr Natalenko (post-factum)
[AMD Official Use Only - General]
Hi Mario.
Quoted 202 lines
-----Original Message-----
From: Limonciello, Mario <[email protected]>
Sent: Wednesday, July 13, 2022 4:07 AM
To: Oleksandr Natalenko <[email protected]>; Yuan, Perry <[email protected]>; [email protected]; Huang, Ray <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: Re: [REGRESSION] amd-pstate doesn't work since v5.18.11Quoted 56 lines · Oleksandr Natalenko
Hello.
Quoted 53 lines · Mario
[Public]
+ Ray
Quoted 49 lines
-----Original Message-----
From: Yuan, Perry <[email protected]>
Sent: Tuesday, July 12, 2022 12:50
To: Oleksandr Natalenko <[email protected]>; Limonciello, Mario <[email protected]>; [email protected]
Cc: Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: RE: [REGRESSION] amd-pstate doesn't work since v5.18.11[AMD Official Use Only - General]
Hi Oleksandr:
Quoted 35 lines
-----Original Message-----
From: Oleksandr Natalenko <[email protected]>
Sent: Wednesday, July 13, 2022 1:40 AM
To: Limonciello, Mario <[email protected]>; linux- [email protected]
Cc: Yuan, Perry <[email protected]>; Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: [REGRESSION] amd-pstate doesn't work since v5.18.11[CAUTION: External Email]
Hello Mario.
The following commits were pulled into v5.18.11:
``` $ git log --oneline --no-merges v5.18.10..v5.18.11 | grep ACPI 2783414e6ef7 ACPI: CPPC: Don't require _OSC if X86_FEATURE_CPPC is supported 3068cfeca3b5 ACPI: CPPC: Only probe for _CPC if CPPC v2 is acked 8beb71759cc8 ACPI: bus: Set CPPC _OSC bits for all and when CPPC_LIB is supported 13bb696dd2f3 ACPI: CPPC: Check _OSC for flexible address space ``` and now this happens: ``` $ sudo modprobe amd-pstate shared_mem=1 modprobe: ERROR: could not insert 'amd_pstate': No such device ```With v5.18.10 this worked just fine.
In your upstream commit
8b356e536e69f3a4d6778ae9f0858a1beadabb1fQuoted 33 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
you write:
``` If there is additional breakage on the shared memory designs also missing this _OSC, additional follow up changes may be needed. ```So the question is what else should be pulled into the stable tree to unbreak amd-pstate?
Quoted 7 lines
Quoted 7 lines
Quoted 5 lines
Thanks.
--
Oleksandr Natalenko (post-factum)Could you share the lscpu output ?
Here's my
lscpu:```
Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 43 bits physical, 48 bits virtual Byte Order: Little Endian CPU(s): 24 On-line CPU(s) list: 0-23 Vendor ID: AuthenticAMD Model name: AMD Ryzen 9 3900XT 12-Core Processor CPU family: 23 Model: 113 Thread(s) per core: 2 Core(s) per socket: 12Quoted 8 lines
Socket(s): 1 Stepping: 0 Frequency boost: enabled CPU(s) scaling MHz: 59% CPU max MHz: 3800,0000 CPU min MHz: 2200,0000 BogoMIPS: 7589.71 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pgemca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext
fxsr_optpdpe1gb rdtscp lmconstant_tscrep_goodnoplnonstop_tsccpuidextd_apicidaperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16sse4_1sse4_2x2apic movbe popcnt aes xsave avx f16c rdrandlahf_lmcmp_legacysvm extapiccr8_legacyabm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoextperfctr_coreperfctr_nbbpextperfctr_llcmwaitx cpbcat_l3cdp_l3hw_pstatessbd mba ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqmrdt_ardseed adx smap clflushopt clwbsha_nixsaveopt xsavec xgetbv1 xsavescqm_llccqm_occup_llccqm_mbm_totalcqm_mbm_localclzero irperf xsaveerptr rdpru wbnoinvd arat npt lbrvsvm_locknrip_savetsc_scalevmcb_cleanflushbyasid decodeassists pausefilter pfthreshold avicv_vmsave_vmloadvgifv_spec_ctrlumip rdpidoverflow_recovsuccor smca sevsev_esQuoted 5 lines
Virtualization: AMD-V L1d cache: 384 KiB (12 instances) L1i cache: 384 KiB (12 instances) L2 cache: 6 MiB (12 instances) L3 cache: 64 MiB (4 instances)Quoted 8 lines
NUMA node(s): 1 NUMA node0 CPU(s): 0-23 Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypassdisabled via prctl
Quoted 1 line
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and
__userpointer sanitizationQuoted 1 line
Vulnerability Spectre v2: Mitigation; Retpolines, IBPB conditional, STIBP
conditional, RSB filling
Quoted 9 lines
Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected```
Quoted 4 lines
Quoted 1 line
Perry.
Thanks this is the sort of thing I was worried might happen as a result of requiring the
_OSC. It was introduced as part of that commit
8beb71759cc8.Quoted 5 lines
Quoted 5 lines
To solve it I think we need to add more things to
cpc_supported_by_cpu
(https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Ftorvalds%2Flinux%2Fblob%2F525496a030de4ae64bb9e1d6bfc8 8eec
Quoted 1 line
Quoted 1 line
6f5fe6e2%2Farch%2Fx86%2Fkernel%2Facpi%2Fcppc.c%23L19&data=05 %7C01
Quoted 1 line
Quoted 1 line
%7CMario.Limonciello%40amd.com%7C96addaab0edc4e22779908da642f
84ac%7C3
Quoted 1 line
Quoted 1 line
dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637932453099304670 %7CUnknow
Quoted 1 line
Quoted 1 line
n%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1ha WwiL
Quoted 1 line
Quoted 1 line
CJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=4KHD3UUlfDJEmpTpqDC muV1x%2F7n
Quoted 10 lines
Quoted 7 lines
j%2F0iuhwdnhJqtQeU%3D&reserved=0)
The question is how do we safely detect the shared memory designs? These are a fixed quantity as newer designs /should/ be using the MSR.
I am tending to thing that unfortunately we need to have an allow-list of shared memory design here unless someone has other ideas.
Happy to test any patches as needed.
See if this helps out:
arch/x86/kernel/acpi/cppc.c index renamed from arch/x86/kernel/acpi/cppc.c +0
-0
734b96454896..88a81e6b9228100644arch/x86/kernel/acpi/cppc.c +2
-0@@ -16,6 +16,13 @@ bool cpc_supported_by_cpu(void)switch (boot_cpu_data.x86_vendor) {case X86_VENDOR_AMD :case X86_VENDOR_HYGON : + if (boot_cpu_data.x86 == 0x19 && + ((boot_cpu_data.x86_model >= 0x00 &&boot_cpu_data.x86_model <=
0x0f) ||
+ (boot_cpu_data.x86_model >=0x20&&
boot_cpu_data.x86_model <=0x2f)))arch/x86/kernel/acpi/cppc.c +3
-0+return true; + else if (boot_cpu_data.x86 == 0x17 && + boot_cpu_data.x86_model >= 0x70 &&boot_cpu_data.x86_model <=
0x7f)
Quoted 7 lines
+ return true; return boot_cpu_has(X86_FEATURE_CPPC); } return false;If that works and no one has a better idea how to do it for these systems I'll send out a proper proper patch tomorrow.
This could be a short-term solution, I would prefer to add CPU Ids check and we can maintain that list for all the model info including MSRs and Shared mem types.
Meanwhile I have the similar issues concern for the coming EPP driver, some systems don`t support EPP and we cannot identify that without IDs list.
Perry.
Quoted 211 lines · Perry
[AMD Official Use Only - General]
Hi Mario.
Quoted 207 lines
-----Original Message-----
From: Limonciello, Mario <[email protected]>
Sent: Wednesday, July 13, 2022 4:07 AM
To: Oleksandr Natalenko <[email protected]>; Yuan, Perry <[email protected]>; [email protected]; Huang, Ray <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: Re: [REGRESSION] amd-pstate doesn't work since v5.18.11Quoted 56 lines · Oleksandr Natalenko
Hello.
Quoted 53 lines · Mario
[Public]
+ Ray
Quoted 49 lines
-----Original Message-----
From: Yuan, Perry <[email protected]>
Sent: Tuesday, July 12, 2022 12:50
To: Oleksandr Natalenko <[email protected]>; Limonciello, Mario <[email protected]>; [email protected]
Cc: Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: RE: [REGRESSION] amd-pstate doesn't work since v5.18.11[AMD Official Use Only - General]
Hi Oleksandr:
Quoted 35 lines
-----Original Message-----
From: Oleksandr Natalenko <[email protected]>
Sent: Wednesday, July 13, 2022 1:40 AM
To: Limonciello, Mario <[email protected]>; linux- [email protected]
Cc: Yuan, Perry <[email protected]>; Rafael J. Wysocki <[email protected]>; Sasha Levin <[email protected]>; [email protected]; H. Peter Anvin <[email protected]>; Greg Kroah-Hartman <[email protected]>
Subject: [REGRESSION] amd-pstate doesn't work since v5.18.11[CAUTION: External Email]
Hello Mario.
The following commits were pulled into v5.18.11:
``` $ git log --oneline --no-merges v5.18.10..v5.18.11 | grep ACPI 2783414e6ef7 ACPI: CPPC: Don't require _OSC if X86_FEATURE_CPPC is supported 3068cfeca3b5 ACPI: CPPC: Only probe for _CPC if CPPC v2 is acked 8beb71759cc8 ACPI: bus: Set CPPC _OSC bits for all and when CPPC_LIB is supported 13bb696dd2f3 ACPI: CPPC: Check _OSC for flexible address space ``` and now this happens: ``` $ sudo modprobe amd-pstate shared_mem=1 modprobe: ERROR: could not insert 'amd_pstate': No such device ```With v5.18.10 this worked just fine.
In your upstream commit
8b356e536e69f3a4d6778ae9f0858a1beadabb1fQuoted 33 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
you write:
``` If there is additional breakage on the shared memory designs also missing this _OSC, additional follow up changes may be needed. ```So the question is what else should be pulled into the stable tree to unbreak amd-pstate?
Quoted 7 lines
Quoted 7 lines
Quoted 5 lines
Thanks.
--
Oleksandr Natalenko (post-factum)Could you share the lscpu output ?
Here's my
lscpu:```
Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 43 bits physical, 48 bits virtual Byte Order: Little Endian CPU(s): 24 On-line CPU(s) list: 0-23 Vendor ID: AuthenticAMD Model name: AMD Ryzen 9 3900XT 12-Core Processor CPU family: 23 Model: 113 Thread(s) per core: 2 Core(s) per socket: 12Quoted 8 lines
Socket(s): 1 Stepping: 0 Frequency boost: enabled CPU(s) scaling MHz: 59% CPU max MHz: 3800,0000 CPU min MHz: 2200,0000 BogoMIPS: 7589.71 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pgemca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext
fxsr_optpdpe1gb rdtscp lmconstant_tscrep_goodnoplnonstop_tsccpuidextd_apicidaperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16sse4_1sse4_2x2apic movbe popcnt aes xsave avx f16c rdrandlahf_lmcmp_legacysvm extapiccr8_legacyabm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoextperfctr_coreperfctr_nbbpextperfctr_llcmwaitx cpbcat_l3cdp_l3hw_pstatessbd mba ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqmrdt_ardseed adx smap clflushopt clwbsha_nixsaveopt xsavec xgetbv1 xsavescqm_llccqm_occup_llccqm_mbm_totalcqm_mbm_localclzero irperf xsaveerptr rdpru wbnoinvd arat npt lbrvsvm_locknrip_savetsc_scalevmcb_cleanflushbyasid decodeassists pausefilter pfthreshold avicv_vmsave_vmloadvgifv_spec_ctrlumip rdpidoverflow_recovsuccor smca sevsev_esQuoted 5 lines
Virtualization: AMD-V L1d cache: 384 KiB (12 instances) L1i cache: 384 KiB (12 instances) L2 cache: 6 MiB (12 instances) L3 cache: 64 MiB (4 instances)Quoted 8 lines
NUMA node(s): 1 NUMA node0 CPU(s): 0-23 Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypassdisabled via prctl
Quoted 1 line
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and
__userpointer sanitizationQuoted 1 line
Vulnerability Spectre v2: Mitigation; Retpolines, IBPB conditional, STIBP
conditional, RSB filling
Quoted 9 lines
Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected```
Quoted 4 lines
Quoted 1 line
Perry.
Thanks this is the sort of thing I was worried might happen as a result of requiring the
_OSC. It was introduced as part of that commit
8beb71759cc8.Quoted 5 lines
Quoted 5 lines
To solve it I think we need to add more things to
cpc_supported_by_cpu
(https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Ftorvalds%2Flinux%2Fblob%2F525496a030de4ae64bb9e1d6bfc8 8eec
Quoted 1 line
Quoted 1 line
6f5fe6e2%2Farch%2Fx86%2Fkernel%2Facpi%2Fcppc.c%23L19&data=05 %7C01
Quoted 1 line
Quoted 1 line
%7CMario.Limonciello%40amd.com%7C96addaab0edc4e22779908da642f
84ac%7C3
Quoted 1 line
Quoted 1 line
dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637932453099304670 %7CUnknow
Quoted 1 line
Quoted 1 line
n%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1ha WwiL
Quoted 1 line
Quoted 1 line
CJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=4KHD3UUlfDJEmpTpqDC muV1x%2F7n
Quoted 10 lines
Quoted 7 lines
j%2F0iuhwdnhJqtQeU%3D&reserved=0)
The question is how do we safely detect the shared memory designs? These are a fixed quantity as newer designs /should/ be using the MSR.
I am tending to thing that unfortunately we need to have an allow-list of shared memory design here unless someone has other ideas.
Happy to test any patches as needed.
See if this helps out:
arch/x86/kernel/acpi/cppc.c index renamed from arch/x86/kernel/acpi/cppc.c +0
-0
734b96454896..88a81e6b9228100644arch/x86/kernel/acpi/cppc.c +2
-0@@ -16,6 +16,13 @@ bool cpc_supported_by_cpu(void)switch (boot_cpu_data.x86_vendor) {case X86_VENDOR_AMD :case X86_VENDOR_HYGON : + if (boot_cpu_data.x86 == 0x19 && + ((boot_cpu_data.x86_model >= 0x00 &&boot_cpu_data.x86_model <=
0x0f) ||
+ (boot_cpu_data.x86_model >=0x20&&
boot_cpu_data.x86_model <=0x2f)))arch/x86/kernel/acpi/cppc.c +3
-0+return true; + else if (boot_cpu_data.x86 == 0x17 && + boot_cpu_data.x86_model >= 0x70 &&boot_cpu_data.x86_model <=
0x7f)+ return true; return boot_cpu_has(X86_FEATURE_CPPC); } return false;
Quoted 5 lines
Quoted 2 lines
If that works and no one has a better idea how to do it for these systems I'll send out a proper proper patch tomorrow.
This could be a short-term solution, I would prefer to add CPU Ids check and we can maintain that list for all the model info including MSRs and Shared mem types.
What's longer term solution when it comes to shared mem? It seems like it's either a list of IDs or a heuristic. Given it's a fixed list and new designs take MSR, I would think the list of IDs is preferable.
Furthermore; I would argue that if there was another design introduced for some reason that takes shared mem instead of MSR it should be using _OSC to indicate CPPCv2 support not this list. This list only needs to exist because the requirement for CPPC support in the _OSC is very recent to the kernel.
Regarding MSR -boot_cpu_has(X86_FEATURE_CPPC) indicates the MSR support. There shouldn't be any need to maintain a list in this _OSC override check here.
Quoted 2 lines
Meanwhile I have the similar issues concern for the coming EPP driver, some systems don`t support EPP and we cannot identify that without IDs list.
That will be localized into the EPP driver source at least.
Quoted 3 lines
Perry.