[v6,07/11] dmapool: ignore init_on_free when DMAPOOL_DEBUG enabled

Commit Message

Tony Battersby June 7, 2022, 6:44 p.m. UTC

There are two cases:

1) In the normal case that the memory is being freed correctly, then
DMAPOOL_DEBUG will memset the memory anyway, so speed thing up by
avoiding a double-memset of the same memory.

2) In the abnormal case that DMAPOOL_DEBUG detects that a driver passes
incorrect parameters to dma_pool_free() (e.g. double-free, invalid
free, mismatched vaddr/dma, etc.), then that is a kernel bug, and we
don't want to clear the passed-in possibly-invalid memory pointer
because we can't be sure that the memory is really free.  So don't clear
it just because init_on_free=1.

Signed-off-by: Tony Battersby <[e-mail]>
---
 mm/dmapool.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

Patch

diff --git a/mm/dmapool.c b/mm/dmapool.c
index facdb3571976..44038089a41a 100644
--- a/mm/dmapool.c
+++ b/mm/dmapool.c
@@ -406,8 +406,6 @@  void dma_pool_free(struct dma_pool *pool, void *vaddr, dma_addr_t dma)
 	}
 
 	offset = vaddr - page->vaddr;
-	if (want_init_on_free())
-		memset(vaddr, 0, pool->size);
 #ifdef	DMAPOOL_DEBUG
 	if ((dma - page->dma) != offset) {
 		spin_unlock_irqrestore(&pool->lock, flags);
@@ -452,6 +450,9 @@  void dma_pool_free(struct dma_pool *pool, void *vaddr, dma_addr_t dma)
 			goto freelist_corrupt;
 	}
 	memset(vaddr, POOL_POISON_FREED, pool->size);
+#else
+	if (want_init_on_free())
+		memset(vaddr, 0, pool->size);
 #endif
 
 	page->in_use--;