[PATCH] ublk: avoid teardown retry loop on xarray allocation failure

__ublk_shmem_remove_ranges() removes matching maple tree ranges in
batches, but first stores each range into a temporary xarray so that the
pages can be unpinned after dropping the maple tree lock.

That temporary xarray is filled under the maple tree lock with
xa_store(..., GFP_ATOMIC). If the store fails before mas_erase(), the
current range is left in the tree and the helper returns false. The
outer ublk_shmem_remove_ranges() loop then immediately retries the same
range. While the atomic allocation keeps failing, the teardown path has
no forward progress.

The issue can be reproduced with radix_tree_node failslab injection after
a SHMEM_ZC buffer has already been registered:

  # Kernel config:
  #   CONFIG_BLK_DEV_UBLK=y
  #   CONFIG_DEBUG_FS=y
  #   CONFIG_FAULT_INJECTION=y
  #   CONFIG_FAULT_INJECTION_DEBUG_FS=y
  #   CONFIG_FAILSLAB=y

  echo 10 > /proc/sys/vm/nr_hugepages
  mkdir -p /tmp/htlb
  mount -t hugetlbfs none /tmp/htlb
  fallocate -l 4M /tmp/htlb/ublk_buf

  dev_id=$(kublk add -t null --shmem_zc \
		--htlb /tmp/htlb/ublk_buf |
	   awk -F '[ :]' '/dev id/ {print $3}')

  echo 1 > /sys/kernel/slab/radix_tree_node/failslab
  echo Y > /sys/kernel/debug/failslab/cache-filter
  echo Y > /sys/kernel/debug/failslab/ignore-gfp-wait
  echo 1 > /sys/kernel/debug/failslab/interval
  echo -1 > /sys/kernel/debug/failslab/times
  echo 100 > /sys/kernel/debug/failslab/probability

  kublk del -n "$dev_id"

On the unfixed kernel the delete command was still running after 3
seconds. Disabling failslab made it return. The fault-injection stack
showed:

  should_failslab
  kmem_cache_alloc_lru_noprof
  __xas_nomem
  __xa_store
  xa_store
  __ublk_shmem_remove_ranges
  ublk_cdev_rel
  ublk_ctrl_del_dev

Remove the allocation from the teardown loop. Keep the existing batch
limit, but collect {base_pfn, nr_pages} pairs in a fixed-size stack array.
Once a matching range is found, the range is erased from the maple tree
before dropping the lock, so each successful scan makes progress without
depending on any GFP_ATOMIC allocation.

With the same failslab settings, the fixed kernel completed
"kublk del -n $dev_id" successfully in about 45 ms.

Fixes: 309e02dccf64 ("ublk: avoid unpinning pages under maple tree spinlock")
Signed-off-by: Yao Sang <[email protected]>
---
 drivers/block/ublk_drv.c | 31 ++++++++++++++-----------------
 1 file changed, 14 insertions(+), 17 deletions(-)

diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c
index 66f10a49509f..00a661146243 100644
--- a/drivers/block/ublk_drv.c
+++ b/drivers/block/ublk_drv.c
@@ -5519,39 +5519,36 @@ static void ublk_unpin_range_pages(unsigned long base_pfn,
 
 /*
  * Inner loop: erase up to UBLK_REMOVE_BATCH matching ranges under
- * mas_lock, collecting them into an xarray. Then drop the lock and
- * unpin pages + free ranges outside spinlock context.
+ * mas_lock, collecting the page ranges in a fixed-size array. Then
+ * drop the lock and unpin pages + free ranges outside spinlock context.
  *
  * Returns true if the tree walk completed, false if more ranges remain.
- * Xarray key is the base PFN, value encodes nr_pages via xa_mk_value().
  */
 #define UBLK_REMOVE_BATCH	64
 
+struct ublk_unpin_range {
+	unsigned long base_pfn;
+	unsigned long nr_pages;
+};
+
 static bool __ublk_shmem_remove_ranges(struct ublk_device *ub,
 					int buf_index, int *ret)
 {
 	MA_STATE(mas, &ub->buf_tree, 0, ULONG_MAX);
 	struct ublk_buf_range *range;
-	struct xarray to_unpin;
-	unsigned long idx;
+	struct ublk_unpin_range to_unpin[UBLK_REMOVE_BATCH];
 	unsigned int count = 0;
+	unsigned int i;
 	bool done = false;
-	void *entry;
-
-	xa_init(&to_unpin);
 
 	mas_lock(&mas);
 	mas_for_each(&mas, range, ULONG_MAX) {
-		unsigned long nr;
-
 		if (buf_index >= 0 && range->buf_index != buf_index)
 			continue;
 
 		*ret = 0;
-		nr = mas.last - mas.index + 1;
-		if (xa_err(xa_store(&to_unpin, mas.index,
-				    xa_mk_value(nr), GFP_ATOMIC)))
-			goto unlock;
+		to_unpin[count].base_pfn = mas.index;
+		to_unpin[count].nr_pages = mas.last - mas.index + 1;
 		mas_erase(&mas);
 		kfree(range);
 		if (++count >= UBLK_REMOVE_BATCH)
@@ -5561,9 +5558,9 @@ static bool __ublk_shmem_remove_ranges(struct ublk_device *ub,
 unlock:
 	mas_unlock(&mas);
 
-	xa_for_each(&to_unpin, idx, entry)
-		ublk_unpin_range_pages(idx, xa_to_value(entry));
-	xa_destroy(&to_unpin);
+	for (i = 0; i < count; i++)
+		ublk_unpin_range_pages(to_unpin[i].base_pfn,
+				       to_unpin[i].nr_pages);
 
 	return done;
 }
-- 
2.25.1