On 10.03.2025 18:18, Jonathan Cavitt wrote:
> Add additional information to each VM so they can report up to the first
> 50 seen pagefaults. Only failed pagefaults are saved this way, as
> successful pagefaults should recover and not need to be reported to
> userspace.
>
> v2:
> - Free vm after use (Shuicheng)
> - Compress pf copy logic (Shuicheng)
> - Update fault_unsuccessful before storing (Shuicheng)
> - Fix old struct name in comments (Shuicheng)
> - Keep first 50 pagefaults instead of last 50 (Jianxun)
>
> v3:
> - Avoid unnecessary execution by checking MAX_PFS earlier (jcavitt)
> - Fix double-locking error (jcavitt)
> - Assert kmemdump is successful (Shuicheng)
>
> Signed-off-by: Jonathan Cavitt <jonathan.cavitt@xxxxxxxxx>
> Suggested-by: Matthew Brost <matthew.brost@xxxxxxxxx>
> CC: Shuicheng Lin <shuicheng.lin@xxxxxxxxx>
> CC: Zhang Jianxun <jianxun.zhang@xxxxxxxxx>
nit: s/CC/Cc
> ---
> drivers/gpu/drm/xe/xe_gt_pagefault.c | 29 +++++++++++++++++++++++++
> drivers/gpu/drm/xe/xe_vm.c | 32 ++++++++++++++++++++++++++++
> drivers/gpu/drm/xe/xe_vm.h | 6 ++++++
> drivers/gpu/drm/xe/xe_vm_types.h | 20 +++++++++++++++++
> 4 files changed, 87 insertions(+)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
> index c8a9058aa09f..7661b9ad1cd0 100644
> --- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
> +++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
> @@ -343,6 +343,34 @@ int xe_guc_pagefault_handler(struct xe_guc *guc, u32 *msg, u32 len)
> return full ? -ENOSPC : 0;
> }
>
> +static void save_pagefault_to_vm(struct xe_device *xe, struct xe_pagefault *pf)
> +{
> + struct xe_vm *vm;
> + struct xe_pagefault *store;
> +
> + vm = asid_to_vm(xe, pf->asid);
> + if (IS_ERR(vm))
> + return;
> +
> + spin_lock(&vm->pfs.lock);
> +
> + /**
this comment shouldn't be marked as kernel-doc
> + * Limit the number of pfs in the pf list to prevent memory overuse.
> + */
> + if (vm->pfs.len < MAX_PFS) {
> + store = kmemdup(pf, sizeof(*pf), GFP_KERNEL);
> + if (store)
> + xe_vm_add_pf_entry(vm, store);
> + else
> + drm_warn(&vm->xe->drm,
> + "Could not allocate memory for fault %i!",
> + vm->pfs.len);
> + }
> +
> + spin_unlock(&vm->pfs.lock);
> + xe_vm_put(vm);
> +}
> +
> #define USM_QUEUE_MAX_RUNTIME_MS 20
>
> static void pf_queue_work_func(struct work_struct *w)
> @@ -362,6 +390,7 @@ static void pf_queue_work_func(struct work_struct *w)
> if (unlikely(ret)) {
> print_pagefault(xe, &pf);
> pf.fault_unsuccessful = 1;
> + save_pagefault_to_vm(xe, &pf);
> drm_dbg(&xe->drm, "Fault response: Unsuccessful %d\n", ret);
> }
>
> diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
> index 22a26aff3a6e..ff79ba8d3794 100644
> --- a/drivers/gpu/drm/xe/xe_vm.c
> +++ b/drivers/gpu/drm/xe/xe_vm.c
> @@ -778,6 +778,33 @@ int xe_vm_userptr_check_repin(struct xe_vm *vm)
> list_empty_careful(&vm->userptr.invalidated)) ? 0 : -EAGAIN;
> }
>
missing kernel-doc for public functions (here and 'remove_pf_entries')
> +void xe_vm_add_pf_entry(struct xe_vm *vm, struct xe_pagefault *pf)
> +{
> + /** vm->pfs.lock must be held before calling this function */
this comment shouldn't be marked as kernel-doc
> + struct xe_vm_pf_entry *e = NULL;
> +
> + e = kzalloc(sizeof(*e), GFP_KERNEL);
> + xe_assert(vm->xe, e);
asserts should not be used as a replacement of proper error handling,
and since alloc could fail we risk a crash in the next line
> + e->pf = pf;
> +
> + list_add_tail(&e->list, &vm->pfs.list);
> + vm->pfs.len++;
> +}
> +
> +void xe_vm_remove_pf_entries(struct xe_vm *vm)
> +{
> + struct xe_vm_pf_entry *e, *tmp;
> +
> + spin_lock(&vm->pfs.lock);
> + list_for_each_entry_safe(e, tmp, &vm->pfs.list, list) {
> + list_del(&e->list);
> + kfree(e->pf);
> + kfree(e);
> + }
> + vm->pfs.len = 0;
> + spin_unlock(&vm->pfs.lock);
> +}
> +
> static int xe_vma_ops_alloc(struct xe_vma_ops *vops, bool array_of_binds)
> {
> int i;
> @@ -1660,6 +1687,9 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags)
> init_rwsem(&vm->userptr.notifier_lock);
> spin_lock_init(&vm->userptr.invalidated_lock);
>
> + INIT_LIST_HEAD(&vm->pfs.list);
> + spin_lock_init(&vm->pfs.lock);
> +
> ttm_lru_bulk_move_init(&vm->lru_bulk_move);
>
> INIT_WORK(&vm->destroy_work, vm_destroy_work_func);
> @@ -1930,6 +1960,8 @@ void xe_vm_close_and_put(struct xe_vm *vm)
> }
> up_write(&xe->usm.lock);
>
> + xe_vm_remove_pf_entries(vm);
> +
> for_each_tile(tile, xe, id)
> xe_range_fence_tree_fini(&vm->rftree[id]);
>
> diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
> index 0ef811fc2bde..fe068e4448a6 100644
> --- a/drivers/gpu/drm/xe/xe_vm.h
> +++ b/drivers/gpu/drm/xe/xe_vm.h
> @@ -12,6 +12,8 @@
> #include "xe_map.h"
> #include "xe_vm_types.h"
>
> +#define MAX_PFS 50
add kernel-doc, also likely needs to be renamed since its too generic,
and maybe it can be defined in .c file as it looks like internal?
> +
> struct drm_device;
> struct drm_printer;
> struct drm_file;
> @@ -257,6 +259,10 @@ int xe_vma_userptr_pin_pages(struct xe_userptr_vma *uvma);
>
> int xe_vma_userptr_check_repin(struct xe_userptr_vma *uvma);
>
> +void xe_vm_add_pf_entry(struct xe_vm *vm, struct xe_pagefault *pf);
> +
> +void xe_vm_remove_pf_entries(struct xe_vm *vm);
> +
> bool xe_vm_validate_should_retry(struct drm_exec *exec, int err, ktime_t *end);
>
> int xe_vm_lock_vma(struct drm_exec *exec, struct xe_vma *vma);
> diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
> index 84fa41b9fa20..3a2329f54fb1 100644
> --- a/drivers/gpu/drm/xe/xe_vm_types.h
> +++ b/drivers/gpu/drm/xe/xe_vm_types.h
> @@ -19,6 +19,7 @@
> #include "xe_range_fence.h"
>
> struct xe_bo;
> +struct xe_pagefault;
> struct xe_svm_range;
> struct xe_sync_entry;
> struct xe_user_fence;
> @@ -142,6 +143,13 @@ struct xe_userptr_vma {
>
> struct xe_device;
>
missing kernel-doc for the xe_vm_pf_entry
> +struct xe_vm_pf_entry {
> + /** @pf: observed pagefault */
> + struct xe_pagefault *pf;
> + /** @list: link into @xe_vm.pfs.list */
> + struct list_head list;
> +};
> +
> struct xe_vm {
> /** @gpuvm: base GPUVM used to track VMAs */
> struct drm_gpuvm gpuvm;
> @@ -305,6 +313,18 @@ struct xe_vm {
> bool capture_once;
> } error_capture;
>
> + /**
> + * @pfs: List of all pagefaults associated with this VM
> + */
> + struct {
> + /** @lock: lock protecting @pfs.list */
this doc must include inner struct name, so it should be:
/** @pfs.lock: ....
> + spinlock_t lock;
> + /** @list: list of xe_exec_queue_ban_entry entries */
> + struct list_head list;
> + /** @len: length of @pfs.list */
> + unsigned int len;
> + } pfs;
> +
> /**
> * @tlb_flush_seqno: Required TLB flush seqno for the next exec.
> * protected by the vm resv.
