mbox series

[0/8,SRU,j:linux,j:linux-oem-5.17,f:linux-oem-5.14] Fix AMD-PState driver for Genoa CPU

Message ID 20221128164755.368178-1-paolo.pisati@canonical.com
Headers show
Series Fix AMD-PState driver for Genoa CPU | expand

Message

Paolo Pisati Nov. 28, 2022, 4:47 p.m. UTC
BugLink: https://bugs.launchpad.net/bugs/1998106

[Impact]

During AMD Genoa development, the datacenter team identified performance
regressions that were caused by the amd-pstate driver. A first round of fixes
already landed upstream (v6.1-rc7) in the amd-pstate driver, and should be
backported to all the affected kernels.

[Fix]

Backport the upstream amd-pstate fixes:

https://lore.kernel.org/linux-pm/20221117073541.3350600-1-Perry.Yuan@amd.com/T/#m4616d857ca472937d1e1d31131ddc3261fa17b2a

[Regression potential]

Clean cherry-picks of upstream fixes, impact a single driver (amd-pstate) and
the new mode (passive) is disabled by default/opt-in. Cherry-picked three
additional documentation patches to make the entire set apply cleanly.

Jan Engelhardt (1):
  Documentation: amd-pstate: grammar and sentence structure updates

Jinzhou Su (1):
  Documentation: amd-pstate: add tracer tool introduction

Meng Li (1):
  Documentation: amd-pstate: Add unit test introduction

Perry Yuan (4):
  cpufreq: amd-pstate: change amd-pstate driver to be built-in type
  cpufreq: amd-pstate: add amd-pstate driver parameter for mode
    selection
  Documentation: amd-pstate: add driver working mode introduction
  Documentation: add amd-pstate kernel command line options

Wyes Karny (1):
  cpufreq: amd-pstate: cpufreq: amd-pstate: reset MSR_AMD_PERF_CTL
    register at init

 .../admin-guide/kernel-parameters.txt         |  11 +
 Documentation/admin-guide/pm/amd-pstate.rst   | 249 ++++++++++++------
 drivers/cpufreq/Kconfig.x86                   |   2 +-
 drivers/cpufreq/amd-pstate.c                  |  49 ++--
 4 files changed, 219 insertions(+), 92 deletions(-)

Comments

Timo Aaltonen Nov. 29, 2022, 3 p.m. UTC | #1
Paolo Pisati kirjoitti 28.11.2022 klo 18.47:
> BugLink: https://bugs.launchpad.net/bugs/1998106
> 
> [Impact]
> 
> During AMD Genoa development, the datacenter team identified performance
> regressions that were caused by the amd-pstate driver. A first round of fixes
> already landed upstream (v6.1-rc7) in the amd-pstate driver, and should be
> backported to all the affected kernels.
> 
> [Fix]
> 
> Backport the upstream amd-pstate fixes:
> 
> https://lore.kernel.org/linux-pm/20221117073541.3350600-1-Perry.Yuan@amd.com/T/#m4616d857ca472937d1e1d31131ddc3261fa17b2a
> 
> [Regression potential]
> 
> Clean cherry-picks of upstream fixes, impact a single driver (amd-pstate) and
> the new mode (passive) is disabled by default/opt-in. Cherry-picked three
> additional documentation patches to make the entire set apply cleanly.
> 
> Jan Engelhardt (1):
>    Documentation: amd-pstate: grammar and sentence structure updates
> 
> Jinzhou Su (1):
>    Documentation: amd-pstate: add tracer tool introduction
> 
> Meng Li (1):
>    Documentation: amd-pstate: Add unit test introduction
> 
> Perry Yuan (4):
>    cpufreq: amd-pstate: change amd-pstate driver to be built-in type
>    cpufreq: amd-pstate: add amd-pstate driver parameter for mode
>      selection
>    Documentation: amd-pstate: add driver working mode introduction
>    Documentation: add amd-pstate kernel command line options
> 
> Wyes Karny (1):
>    cpufreq: amd-pstate: cpufreq: amd-pstate: reset MSR_AMD_PERF_CTL
>      register at init
> 
>   .../admin-guide/kernel-parameters.txt         |  11 +
>   Documentation/admin-guide/pm/amd-pstate.rst   | 249 ++++++++++++------
>   drivers/cpufreq/Kconfig.x86                   |   2 +-
>   drivers/cpufreq/amd-pstate.c                  |  49 ++--
>   4 files changed, 219 insertions(+), 92 deletions(-)
> 

applied to oem-5.14, 5.17, thanks
Tim Gardner Nov. 29, 2022, 6:45 p.m. UTC | #2
On 11/28/22 9:47 AM, Paolo Pisati wrote:
> BugLink: https://bugs.launchpad.net/bugs/1998106
> 
> [Impact]
> 
> During AMD Genoa development, the datacenter team identified performance
> regressions that were caused by the amd-pstate driver. A first round of fixes
> already landed upstream (v6.1-rc7) in the amd-pstate driver, and should be
> backported to all the affected kernels.
> 
> [Fix]
> 
> Backport the upstream amd-pstate fixes:
> 
> https://lore.kernel.org/linux-pm/20221117073541.3350600-1-Perry.Yuan@amd.com/T/#m4616d857ca472937d1e1d31131ddc3261fa17b2a
> 
> [Regression potential]
> 
> Clean cherry-picks of upstream fixes, impact a single driver (amd-pstate) and
> the new mode (passive) is disabled by default/opt-in. Cherry-picked three
> additional documentation patches to make the entire set apply cleanly.
> 
> Jan Engelhardt (1):
>    Documentation: amd-pstate: grammar and sentence structure updates
> 
> Jinzhou Su (1):
>    Documentation: amd-pstate: add tracer tool introduction
> 
> Meng Li (1):
>    Documentation: amd-pstate: Add unit test introduction
> 
> Perry Yuan (4):
>    cpufreq: amd-pstate: change amd-pstate driver to be built-in type
>    cpufreq: amd-pstate: add amd-pstate driver parameter for mode
>      selection
>    Documentation: amd-pstate: add driver working mode introduction
>    Documentation: add amd-pstate kernel command line options
> 
> Wyes Karny (1):
>    cpufreq: amd-pstate: cpufreq: amd-pstate: reset MSR_AMD_PERF_CTL
>      register at init
> 
>   .../admin-guide/kernel-parameters.txt         |  11 +
>   Documentation/admin-guide/pm/amd-pstate.rst   | 249 ++++++++++++------
>   drivers/cpufreq/Kconfig.x86                   |   2 +-
>   drivers/cpufreq/amd-pstate.c                  |  49 ++--
>   4 files changed, 219 insertions(+), 92 deletions(-)
> 
Acked-by: Tim Gardner <tim.gardner@canonical.com>
Cory Todd Nov. 30, 2022, 4:16 p.m. UTC | #3
On Mon, Nov 28, 2022 at 05:47:47PM +0100, Paolo Pisati wrote:
> BugLink: https://bugs.launchpad.net/bugs/1998106
> 
> [Impact]
> 
> During AMD Genoa development, the datacenter team identified performance
> regressions that were caused by the amd-pstate driver. A first round of fixes
> already landed upstream (v6.1-rc7) in the amd-pstate driver, and should be
> backported to all the affected kernels.
> 
> [Fix]
> 
> Backport the upstream amd-pstate fixes:
> 
> https://lore.kernel.org/linux-pm/20221117073541.3350600-1-Perry.Yuan@amd.com/T/#m4616d857ca472937d1e1d31131ddc3261fa17b2a
> 
> [Regression potential]
> 
> Clean cherry-picks of upstream fixes, impact a single driver (amd-pstate) and
> the new mode (passive) is disabled by default/opt-in. Cherry-picked three
> additional documentation patches to make the entire set apply cleanly.
> 
> Jan Engelhardt (1):
>   Documentation: amd-pstate: grammar and sentence structure updates
> 
> Jinzhou Su (1):
>   Documentation: amd-pstate: add tracer tool introduction
> 
> Meng Li (1):
>   Documentation: amd-pstate: Add unit test introduction
> 
> Perry Yuan (4):
>   cpufreq: amd-pstate: change amd-pstate driver to be built-in type
>   cpufreq: amd-pstate: add amd-pstate driver parameter for mode
>     selection
>   Documentation: amd-pstate: add driver working mode introduction
>   Documentation: add amd-pstate kernel command line options
> 
> Wyes Karny (1):
>   cpufreq: amd-pstate: cpufreq: amd-pstate: reset MSR_AMD_PERF_CTL
>     register at init
> 
>  .../admin-guide/kernel-parameters.txt         |  11 +
>  Documentation/admin-guide/pm/amd-pstate.rst   | 249 ++++++++++++------
>  drivers/cpufreq/Kconfig.x86                   |   2 +-
>  drivers/cpufreq/amd-pstate.c                  |  49 ++--
>  4 files changed, 219 insertions(+), 92 deletions(-)
> 

Acked-by: Cory Todd <cory.todd@canonical.com>
Luke Nowakowski-Krijger Dec. 1, 2022, 11:34 p.m. UTC | #4
Applied to jammy:linux master-next,

Thanks :)

- Luke

On Mon, Nov 28, 2022 at 8:48 AM Paolo Pisati <paolo.pisati@canonical.com>
wrote:

> BugLink: https://bugs.launchpad.net/bugs/1998106
>
> [Impact]
>
> During AMD Genoa development, the datacenter team identified performance
> regressions that were caused by the amd-pstate driver. A first round of
> fixes
> already landed upstream (v6.1-rc7) in the amd-pstate driver, and should be
> backported to all the affected kernels.
>
> [Fix]
>
> Backport the upstream amd-pstate fixes:
>
>
> https://lore.kernel.org/linux-pm/20221117073541.3350600-1-Perry.Yuan@amd.com/T/#m4616d857ca472937d1e1d31131ddc3261fa17b2a
>
> [Regression potential]
>
> Clean cherry-picks of upstream fixes, impact a single driver (amd-pstate)
> and
> the new mode (passive) is disabled by default/opt-in. Cherry-picked three
> additional documentation patches to make the entire set apply cleanly.
>
> Jan Engelhardt (1):
>   Documentation: amd-pstate: grammar and sentence structure updates
>
> Jinzhou Su (1):
>   Documentation: amd-pstate: add tracer tool introduction
>
> Meng Li (1):
>   Documentation: amd-pstate: Add unit test introduction
>
> Perry Yuan (4):
>   cpufreq: amd-pstate: change amd-pstate driver to be built-in type
>   cpufreq: amd-pstate: add amd-pstate driver parameter for mode
>     selection
>   Documentation: amd-pstate: add driver working mode introduction
>   Documentation: add amd-pstate kernel command line options
>
> Wyes Karny (1):
>   cpufreq: amd-pstate: cpufreq: amd-pstate: reset MSR_AMD_PERF_CTL
>     register at init
>
>  .../admin-guide/kernel-parameters.txt         |  11 +
>  Documentation/admin-guide/pm/amd-pstate.rst   | 249 ++++++++++++------
>  drivers/cpufreq/Kconfig.x86                   |   2 +-
>  drivers/cpufreq/amd-pstate.c                  |  49 ++--
>  4 files changed, 219 insertions(+), 92 deletions(-)
>
> --
> 2.34.1
>
>
> --
> kernel-team mailing list
> kernel-team@lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/kernel-team
>