[PATCH v7 8/8] drm/gpusvm: Use hmm_range_fault_unlocked_timeout() for range faults

[Stanislav Kinsburskii]

Several GPU SVM paths take mmap_read_lock() only to call hmm_range_fault(),
then retry -EBUSY until HMM_RANGE_DEFAULT_TIMEOUT expires. Those paths use
MMU interval notifiers whose mm matches the mm that was locked for the HMM
fault.

Use hmm_range_fault_unlocked_timeout() for those faults and pass the
remaining retry budget to HMM. The helper owns mmap_lock acquisition and
refreshes range->notifier_seq internally for each retry, while GPU SVM
keeps its existing driver-lock validation with mmu_interval_read_retry()
after a successful fault.

Leave drm_gpusvm_check_pages() on hmm_range_fault() because that path is
called with the mmap lock already held by its caller.

Signed-off-by: Stanislav Kinsburskii <skinsburs...@gmail.com>
---
 drivers/gpu/drm/drm_gpusvm.c |   52 ++++++------------------------------------
 1 file changed, 7 insertions(+), 45 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c
index 958cb605aedd..2601b793428c 100644
--- a/drivers/gpu/drm/drm_gpusvm.c
+++ b/drivers/gpu/drm/drm_gpusvm.c
@@ -788,22 +788,8 @@ enum drm_gpusvm_scan_result drm_gpusvm_scan_mm(struct 
drm_gpusvm_range *range,
        hmm_range.hmm_pfns = pfns;
 
 retry:
-       hmm_range.notifier_seq = mmu_interval_read_begin(notifier);
-       mmap_read_lock(range->gpusvm->mm);
-
-       while (true) {
-               err = hmm_range_fault(&hmm_range);
-               if (err == -EBUSY) {
-                       if (time_after(jiffies, timeout))
-                               break;
-
-                       hmm_range.notifier_seq =
-                               mmu_interval_read_begin(notifier);
-                       continue;
-               }
-               break;
-       }
-       mmap_read_unlock(range->gpusvm->mm);
+       err = hmm_range_fault_unlocked_timeout(&hmm_range,
+                               max_t(long, timeout - jiffies, 1));
        if (err)
                goto err_free;
 
@@ -1439,21 +1425,8 @@ int drm_gpusvm_get_pages(struct drm_gpusvm *gpusvm,
        }
 
        hmm_range.hmm_pfns = pfns;
-       while (true) {
-               mmap_read_lock(mm);
-               err = hmm_range_fault(&hmm_range);
-               mmap_read_unlock(mm);
-
-               if (err == -EBUSY) {
-                       if (time_after(jiffies, timeout))
-                               break;
-
-                       hmm_range.notifier_seq =
-                               mmu_interval_read_begin(notifier);
-                       continue;
-               }
-               break;
-       }
+       err = hmm_range_fault_unlocked_timeout(&hmm_range,
+                               max_t(long, timeout - jiffies, 1));
        mmput(mm);
        if (err)
                goto err_free;
@@ -1736,24 +1709,13 @@ int drm_gpusvm_range_evict(struct drm_gpusvm *gpusvm,
                return -ENOMEM;
 
        hmm_range.hmm_pfns = pfns;
-       while (!time_after(jiffies, timeout)) {
-               hmm_range.notifier_seq = mmu_interval_read_begin(notifier);
-               if (time_after(jiffies, timeout)) {
-                       err = -ETIME;
-                       break;
-               }
-
-               mmap_read_lock(mm);
-               err = hmm_range_fault(&hmm_range);
-               mmap_read_unlock(mm);
-               if (err != -EBUSY)
-                       break;
-       }
+       err = hmm_range_fault_unlocked_timeout(&hmm_range,
+                               max_t(long, timeout - jiffies, 1));
 
        kvfree(pfns);
        mmput(mm);
 
-       return err;
+       return err == -EBUSY ? -ETIME : err;
 }
 EXPORT_SYMBOL_GPL(drm_gpusvm_range_evict);