AMD General
Shouldn't we be putting an
if (ptl->hw_supported) {
Call in the kfd_profiler_ptl_control function instead? Instead of papering over
the issue, we should only try to do the function if ptl is enabled.
Kent
> -----Original Message-----
> From: amd-gfx <[email protected]> On Behalf Of Eric
> Huang
> Sent: June 19, 2026 12:46 PM
> To: [email protected]
> Cc: Huang, JinHuiEric <[email protected]>
> Subject: [PATCH] drm/amdkfd: change a PTL warning message to debug message
>
> PTL is a special feature for gfxv9.4.4, but the warning is
> always appearing on other ASICs when rocprof is running, it
> causes confusion, so reduce the message's level to solve it.
>
> Signed-off-by: Eric Huang <[email protected]>
> ---
> drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> index 575b13b557af..566955adb58f 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> @@ -1813,7 +1813,7 @@ int kfd_ptl_disable_request(struct kfd_process_device
> *pdd,
> ret = kfd_ptl_control(pdd, false);
> if (ret) {
> atomic_dec(&ptl->disable_ref);
> - dev_warn(pdd->dev->adev->dev,
> + dev_dbg(pdd->dev->adev->dev,
> "failed to disable PTL\n");
> goto out;
> }
> --
> 2.34.1
