RE: [PATCH v2 1/1] drm/amdkfd: Add TLB flush after MES queue eviction/suspension

AMD General

Sure Alex. I will check on this.

-----Original Message-----
From: Alex Deucher <[email protected]>
Sent: Friday, August 14, 2026 12:33 AM
To: Hosur, Priya <[email protected]>; Khatri, Sunil <[email protected]>
Cc: [email protected]; Liu, Shaoyun <[email protected]>; Lazar, Lijo <[email protected]>; Deucher, Alexander <[email protected]>; Limonciello, Mario <[email protected]>; Koenig, Christian <[email protected]>; Vishwakarma, Pratik <[email protected]>; Gopalakrishnan, Veerabadhran (Veera) <[email protected]>; Kuehling, Felix <[email protected]>
Subject: Re: [PATCH v2 1/1] drm/amdkfd: Add TLB flush after MES queue eviction/suspension

+ Sunil


On Thu, Aug 13, 2026 at 1:50 AM Priya Hosur <[email protected]> wrote:
>
> MES (Micro Engine Scheduler) does not perform heavy-weight TLB
> invalidation after unmapping queues, unlike HWS which does this
> automatically. This causes a race condition where in-flight DMA
> descriptors can access memory that has been unmapped, leading to page
> faults and GPU queue hangs during SVM page migration.
>
> The issue manifests as KFDSVMRangeTest.MultiThreadMigrationTest/1
> failures on gfx1151 (Ryzen AI MAX) with XNACK mode 1 enabled - the GPU
> compute queue hangs with packets submitted but never consumed.
>
> Add kfd_flush_tlb() with TLB_FLUSH_HEAVYWEIGHT in two MES code paths:
> 1. evict_process_queues_cpsch() - after MES removes queues 2.
> suspend_queues() - after MES suspends queues and mem_fence completes
>
> This ensures all in-flight memory accesses from unmapped queues are
> flushed before memory is freed or migrated.

Can you check amdgpu_userq.c as well?  I suspect it needs something similar when unmapping.

Thanks,

Alex

>
> Testing on gfx1151 shows this reduces failure rate from 100% to
> approximately 7-10%. The residual failures require further investigation.
>
> Change-Id: Ia83e6af7a66fe2bff56cb62c741728e9a18883e8
> Signed-off-by: Priya Hosur <[email protected]>
> Reviewed-by: Felix Kuehling <[email protected]>
> ---
>  .../gpu/drm/amd/amdkfd/kfd_device_queue_manager.c   | 13 ++++++++++++-
>  1 file changed, 12 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
> b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
> index a23384571193..71c60a42c581 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
> @@ -1450,6 +1450,14 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm,
>                 dqm_evict_mqd_bo(dqm, q);
>         }
>
> +       /*
> +        * Heavy-weight TLB flush after MES removes queues to ensure
> +        * in-flight memory accesses complete before memory is freed/migrated.
> +        * HWS does this automatically, MES does not.
> +        */
> +       if (dqm->dev->kfd->shared_resources.enable_mes)
> +               kfd_flush_tlb(pdd, TLB_FLUSH_HEAVYWEIGHT);
> +
>         if (!dqm->dev->kfd->shared_resources.enable_mes) {
>                 pdd->last_evict_timestamp = get_jiffies_64();
>                 retval = execute_queues_cpsch(dqm, @@ -3736,8 +3744,11
> @@ int suspend_queues(struct kfd_process *p,
>                 if (!per_device_suspended) {
>                         dqm_unlock(dqm);
>                         mutex_unlock(&p->event_mutex);
> -                       if (total_suspended)
> +                       if (total_suspended) {
>
> amdgpu_amdkfd_debug_mem_fence(dqm->dev->adev);
> +                               /* Heavy-weight TLB flush after MES suspends queues */
> +                               kfd_flush_tlb(pdd, TLB_FLUSH_HEAVYWEIGHT);
> +                       }
>                         continue;
>                 }
>
> --
> 2.43.0
>