Merge branch 'ttm-fixes-3.13' of git://people.freedesktop.org/~thomash/linux into drm-fixes
The set_need_resched() removal fix and yet another fix in
ttm_bo_move_memcpy().
* 'ttm-fixes-3.13' of git://people.freedesktop.org/~thomash/linux:
drm/ttm: Remove set_need_resched from the ttm fault handler
drm/ttm: Don't move non-existing data
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 8d5a646..07e02c4 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -151,7 +151,7 @@
atomic_dec(&bo->glob->bo_count);
if (bo->resv == &bo->ttm_resv)
reservation_object_fini(&bo->ttm_resv);
-
+ mutex_destroy(&bo->wu_mutex);
if (bo->destroy)
bo->destroy(bo);
else {
@@ -1123,6 +1123,7 @@
INIT_LIST_HEAD(&bo->ddestroy);
INIT_LIST_HEAD(&bo->swap);
INIT_LIST_HEAD(&bo->io_reserve_lru);
+ mutex_init(&bo->wu_mutex);
bo->bdev = bdev;
bo->glob = bdev->glob;
bo->type = type;
@@ -1704,3 +1705,35 @@
;
}
EXPORT_SYMBOL(ttm_bo_swapout_all);
+
+/**
+ * ttm_bo_wait_unreserved - interruptible wait for a buffer object to become
+ * unreserved
+ *
+ * @bo: Pointer to buffer
+ */
+int ttm_bo_wait_unreserved(struct ttm_buffer_object *bo)
+{
+ int ret;
+
+ /*
+ * In the absense of a wait_unlocked API,
+ * Use the bo::wu_mutex to avoid triggering livelocks due to
+ * concurrent use of this function. Note that this use of
+ * bo::wu_mutex can go away if we change locking order to
+ * mmap_sem -> bo::reserve.
+ */
+ ret = mutex_lock_interruptible(&bo->wu_mutex);
+ if (unlikely(ret != 0))
+ return -ERESTARTSYS;
+ if (!ww_mutex_is_locked(&bo->resv->lock))
+ goto out_unlock;
+ ret = ttm_bo_reserve_nolru(bo, true, false, false, NULL);
+ if (unlikely(ret != 0))
+ goto out_unlock;
+ ww_mutex_unlock(&bo->resv->lock);
+
+out_unlock:
+ mutex_unlock(&bo->wu_mutex);
+ return ret;
+}
diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index 4834c46..15b86a9 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -350,10 +350,13 @@
goto out2;
/*
- * Move nonexistent data. NOP.
+ * Don't move nonexistent data. Clear destination instead.
*/
- if (old_iomap == NULL && ttm == NULL)
+ if (old_iomap == NULL &&
+ (ttm == NULL || ttm->state == tt_unpopulated)) {
+ memset_io(new_iomap, 0, new_mem->num_pages*PAGE_SIZE);
goto out2;
+ }
/*
* TTM might be null for moves within the same region.
diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
index ac617f3..b249ab9 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
@@ -107,13 +107,28 @@
/*
* Work around locking order reversal in fault / nopfn
* between mmap_sem and bo_reserve: Perform a trylock operation
- * for reserve, and if it fails, retry the fault after scheduling.
+ * for reserve, and if it fails, retry the fault after waiting
+ * for the buffer to become unreserved.
*/
-
- ret = ttm_bo_reserve(bo, true, true, false, 0);
+ ret = ttm_bo_reserve(bo, true, true, false, NULL);
if (unlikely(ret != 0)) {
- if (ret == -EBUSY)
- set_need_resched();
+ if (ret != -EBUSY)
+ return VM_FAULT_NOPAGE;
+
+ if (vmf->flags & FAULT_FLAG_ALLOW_RETRY) {
+ if (!(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) {
+ up_read(&vma->vm_mm->mmap_sem);
+ (void) ttm_bo_wait_unreserved(bo);
+ }
+
+ return VM_FAULT_RETRY;
+ }
+
+ /*
+ * If we'd want to change locking order to
+ * mmap_sem -> bo::reserve, we'd use a blocking reserve here
+ * instead of retrying the fault...
+ */
return VM_FAULT_NOPAGE;
}
@@ -123,7 +138,6 @@
case 0:
break;
case -EBUSY:
- set_need_resched();
case -ERESTARTSYS:
retval = VM_FAULT_NOPAGE;
goto out_unlock;
diff --git a/include/drm/ttm/ttm_bo_api.h b/include/drm/ttm/ttm_bo_api.h
index 751eaff..ee127ec 100644
--- a/include/drm/ttm/ttm_bo_api.h
+++ b/include/drm/ttm/ttm_bo_api.h
@@ -169,6 +169,7 @@
* @offset: The current GPU offset, which can have different meanings
* depending on the memory type. For SYSTEM type memory, it should be 0.
* @cur_placement: Hint of current placement.
+ * @wu_mutex: Wait unreserved mutex.
*
* Base class for TTM buffer object, that deals with data placement and CPU
* mappings. GPU mappings are really up to the driver, but for simpler GPUs
@@ -250,6 +251,7 @@
struct reservation_object *resv;
struct reservation_object ttm_resv;
+ struct mutex wu_mutex;
};
/**
@@ -702,5 +704,5 @@
size_t count, loff_t *f_pos, bool write);
extern void ttm_bo_swapout_all(struct ttm_bo_device *bdev);
-
+extern int ttm_bo_wait_unreserved(struct ttm_buffer_object *bo);
#endif