mbox series

[v2,0/2] x86/fpu: Extend kernel_fpu_begin_mask() for the In-Field Scan driver

Message ID 20240507235344.249103-1-chang.seok.bae@intel.com (mailing list archive)
Headers show
Series x86/fpu: Extend kernel_fpu_begin_mask() for the In-Field Scan driver | expand

Message

Chang S. Bae May 7, 2024, 11:53 p.m. UTC 
Following feedback from the previous posting [1], this revision comes
with the changelog updates for more clarity:

* The recent change in the SDM alone does not fully explain the
  underlying reasons. Provide a clearer explanation rather than simply
  echoing the SDM text.

* While it is feasible for administrators to isolate CPUs running the
  In-Field Scan tests from those running AMX workloads, this approach is
  considerably disruptive and conflicts with the goal of its live
  testing. Add this point to the changelog as well.

Thanks to Dave for highlighting these aspects [2]. Additionally, it is
worth noting that the IFS Technical Paper [3] was recently published,
which may offer further context on the IFS testing.

Thanks,
Chang

[1] V1: https://lore.kernel.org/lkml/20240430212508.105117-1-chang.seok.bae@intel.com/
[2] Feedback: https://lore.kernel.org/lkml/a41d7012-2eea-436e-9f7e-6a7702f7e2c2@intel.com/
[3] IFS Paper: https://www.intel.com/content/www/us/en/content-details/822279/finding-faulty-components-in-a-live-fleet-environment.html

Chang S. Bae (2):
  x86/fpu: Extend kernel_fpu_begin_mask() to initialize AMX state
  platform/x86/intel/ifs: Initialize AMX state for the scan test

 arch/x86/include/asm/fpu/api.h           |  1 +
 arch/x86/kernel/fpu/core.c               |  3 +++
 drivers/platform/x86/intel/ifs/ifs.h     | 15 +++++++++++++++
 drivers/platform/x86/intel/ifs/runtest.c |  6 ++++++
 4 files changed, 25 insertions(+)


base-commit: 7598293ab37c92025086de4b0ecd9474013a725f

Comments

Ilpo Järvinen May 8, 2024, 8:28 a.m. UTC | #1 
On Tue, 7 May 2024, Chang S. Bae wrote:

> Following feedback from the previous posting [1], this revision comes
> with the changelog updates for more clarity:
> 
> * The recent change in the SDM alone does not fully explain the
>   underlying reasons. Provide a clearer explanation rather than simply
>   echoing the SDM text.
> 
> * While it is feasible for administrators to isolate CPUs running the
>   In-Field Scan tests from those running AMX workloads, this approach is
>   considerably disruptive and conflicts with the goal of its live
>   testing. Add this point to the changelog as well.
> 
> Thanks to Dave for highlighting these aspects [2]. Additionally, it is
> worth noting that the IFS Technical Paper [3] was recently published,
> which may offer further context on the IFS testing.
> 
> Thanks,
> Chang
> 
> [1] V1: https://lore.kernel.org/lkml/20240430212508.105117-1-chang.seok.bae@intel.com/
> [2] Feedback: https://lore.kernel.org/lkml/a41d7012-2eea-436e-9f7e-6a7702f7e2c2@intel.com/
> [3] IFS Paper: https://www.intel.com/content/www/us/en/content-details/822279/finding-faulty-components-in-a-live-fleet-environment.html
> 
> Chang S. Bae (2):
>   x86/fpu: Extend kernel_fpu_begin_mask() to initialize AMX state
>   platform/x86/intel/ifs: Initialize AMX state for the scan test

For both patches:

Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>