Commit c49825fa authored Sep 24, 2010 by Malahal Naineni Committed by Jens Axboe Sep 24, 2010

block: set the bounce_pfn to the actual DMA limit rather than to max memory



The bounce_pfn of the request queue in 64 bit systems is set to the
current max_low_pfn. Adding more memory later makes this incorrect.
Memory allocated beyond this boot time max_low_pfn appear to require
bounce buffers (bounce buffers are actually not allocated but used in
calculating segments that may result in "over max segments limit"
errors).

Signed-off-by: Malahal Naineni <malahal@us.ibm.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>

parent 4b197769

block/blk-settings.c

+1 −3

Original line number	Diff line number	Diff line
		@@ -214,16 +214,14 @@ void blk_queue_bounce_limit(struct request_queue *q, u64 dma_mask)
		*/
		if (b_pfn < (min_t(u64, 0xffffffffUL, BLK_BOUNCE_HIGH) >> PAGE_SHIFT))
		dma = 1;
		q->limits.bounce_pfn = max_low_pfn;
		#else
		if (b_pfn < blk_max_low_pfn)
		dma = 1;
		q->limits.bounce_pfn = b_pfn;
		#endif
		q->limits.bounce_pfn = b_pfn;
		if (dma) {
		init_emergency_isa_pool();
		q->bounce_gfp = GFP_NOIO \| GFP_DMA;
		q->limits.bounce_pfn = b_pfn;
		}
		}
		EXPORT_SYMBOL(blk_queue_bounce_limit);

Admin message