AMD General Looks like it should work. The only time we call ptl_control is from disable_request or disable_release, so we should be able to exit early in both of those cases, covering the existing flow and also not sending invalid dmesg warnings. Thanks!
Reviewed-by: Kent Russell <[email protected]> > -----Original Message----- > From: Huang, JinHuiEric <[email protected]> > Sent: June 19, 2026 1:30 PM > To: [email protected] > Cc: Russell, Kent <[email protected]>; [email protected]; Huang, > JinHuiEric <[email protected]> > Subject: [PATCH] drm/amdkfd: avoid PTL confused warning 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 move hw_supported check earlier to > avoid it. > > Signed-off-by: Eric Huang <[email protected]> > --- > drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 9 ++++++--- > 1 file changed, 6 insertions(+), 3 deletions(-) > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c > b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c > index 575b13b557af..ab9e53dc8deb 100644 > --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c > @@ -1783,9 +1783,6 @@ static int kfd_ptl_control(struct kfd_process_device > *pdd, bool enable) > uint32_t ptl_state = enable ? 1 : 0; > int ret; > > - if (!ptl->hw_supported) > - return -EOPNOTSUPP; > - > if (!pdd->dev->kfd2kgd || !pdd->dev->kfd2kgd->ptl_ctrl) > return -EOPNOTSUPP; > > @@ -1804,6 +1801,9 @@ int kfd_ptl_disable_request(struct kfd_process_device > *pdd, > struct amdgpu_ptl *ptl = &adev->psp.ptl; > int ret = 0; > > + if (!ptl->hw_supported) > + return -EOPNOTSUPP; > + > mutex_lock(&ptl->mutex); > > if (pdd->ptl_disable_req) > @@ -1833,6 +1833,9 @@ int kfd_ptl_disable_release(struct kfd_process_device > *pdd, > struct amdgpu_ptl *ptl = &adev->psp.ptl; > int ret = 0; > > + if (!ptl->hw_supported) > + return -EOPNOTSUPP; > + > mutex_lock(&ptl->mutex); > > if (!pdd->ptl_disable_req) > -- > 2.34.1
