Permalink
Browse files

Revert "Disable vmalloc() direct reclaim"

This reverts commit 2092cf6.  The
use of the PF_MEMALLOC flag was always a hack to work around memory
reclaim deadlocks.  Those issues are believed to be resolved so this
workaround can be safely reverted.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
  • Loading branch information...
1 parent 617f79d commit 500e95c884f711883e094b48b58450ce2d80f2ea @behlendorf behlendorf committed Aug 18, 2012
Showing with 2 additions and 22 deletions.
  1. +2 −22 module/spl/spl-kmem.c
View
@@ -840,31 +840,11 @@ kv_alloc(spl_kmem_cache_t *skc, int size, int flags)
ASSERT(ISP2(size));
- if (skc->skc_flags & KMC_KMEM) {
+ if (skc->skc_flags & KMC_KMEM)
ptr = (void *)__get_free_pages(flags, get_order(size));
- } else {
- /*
- * As part of vmalloc() an __pte_alloc_kernel() allocation
- * may occur. This internal allocation does not honor the
- * gfp flags passed to vmalloc(). This means even when
- * vmalloc(GFP_NOFS) is called it is possible synchronous
- * reclaim will occur. This reclaim can trigger file IO
- * which can result in a deadlock. This issue can be avoided
- * by explicitly setting PF_MEMALLOC on the process to
- * subvert synchronous reclaim. The following bug has
- * been filed at kernel.org to track the issue.
- *
- * https://bugzilla.kernel.org/show_bug.cgi?id=30702
- */
- if (!(flags & __GFP_FS))
- current->flags |= PF_MEMALLOC;
-
+ else
ptr = __vmalloc(size, flags | __GFP_HIGHMEM, PAGE_KERNEL);
- if (!(flags & __GFP_FS))
- current->flags &= ~PF_MEMALLOC;
- }
-
/* Resulting allocated memory will be page aligned */
ASSERT(IS_P2ALIGNED(ptr, PAGE_SIZE));

0 comments on commit 500e95c

Please sign in to comment.