Attachment 298053 Details for Bug 433554 – Don't perform unnecessarily swiotlb copies

[patch] Don't perform unnecessarily swiotlb copies

linux-2.6.18-xen-swiotlb-check-bounce-range.patch (text/plain), 3.12 KB, created by Stephen Tweedie on 2008-03-14 14:27:53 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Stephen Tweedie

Created: 2008-03-14 14:27:53 UTC

Size: 3.12 KB

patch

obsolete

>commit a571047896444f92cf6a489943e7f595b084e4ab
>Author: Stephen Tweedie <sct@redhat.com>
>Date:   Thu Mar 13 17:49:28 2008 +0000
>
>    xen dma: avoid unnecessarily SWIOTLB bounce buffering.
>    
>    On Xen kernels, BIOVEC_PHYS_MERGEABLE permits merging of disk IOs that
>    span multiple pages, provided that the pages are both pseudophysically-
>    AND machine-contiguous ---
>    
>    	(((bvec_to_phys((vec1)) + (vec1)->bv_len) == bvec_to_phys((vec2))) && \
>    	 ((bvec_to_pseudophys((vec1)) + (vec1)->bv_len) == \
>    	  bvec_to_pseudophys((vec2))))
>    
>    However, this best-effort merging of adjacent pages can occur in
>    regions of dom0 memory which just happen, by virtue of having been
>    initially set up that way, to be machine-contiguous.  Such pages
>    which occur outside of a range created by xen_create_contiguous_
>    region won't be seen as contiguous by range_straddles_page_boundary(),
>    so the pci-dma-xen.c code for dma_map_sg() will send these regions
>    to the swiotlb for bounce buffering.
>    
>    In RHEL-5.1 this did not happen, because we did not have the check
>    for range_straddles_page_boundary() in that code.  Now that that check
>    has been added, these SG ranges --- which ARE machine contiguous and
>    which can perfectly well be sent to a dma engine --- are being bounce-
>    buffered in the swiotlb instead, causing a performance overhead and
>    potentially leading to early swiotlb exhaustion.
>    
>    This patch adds a new check, check_pages_physically_contiguous(),
>    to the swiotlb_map_sg() code to capture these ranges and map them
>    directly via virt_to_bus() mapping rather than through the swiotlb.
>    
>    Signed-off-by: Stephen Tweedie <sct@redhat.com>
>
>diff --git a/arch/i386/kernel/swiotlb.c b/arch/i386/kernel/swiotlb.c
>index ae9e3b7..68a6b33 100644
>--- a/arch/i386/kernel/swiotlb.c
>+++ b/arch/i386/kernel/swiotlb.c
>@@ -516,6 +516,25 @@ swiotlb_sync_single_for_device(struct device *hwdev, dma_addr_t dev_addr,
> 		sync_single(hwdev, bus_to_virt(dev_addr), size, dir);
> }
> 
>+static int check_pages_physically_contiguous(struct page *page, 
>+					     unsigned int offset,
>+					     unsigned int length)
>+{
>+	unsigned long pfn, next_mfn;
>+	int i;
>+	int nr_pages;
>+	
>+	pfn = page_to_pfn(page);
>+	next_mfn = pfn_to_mfn(pfn);
>+	nr_pages = (offset + length + PAGE_SIZE-1) >> PAGE_SHIFT;
>+	
>+	for (i = 1; i < nr_pages; i++) {
>+		if (pfn_to_mfn(++pfn) != ++next_mfn) 
>+			return 0;
>+	}
>+	return 1;
>+}
>+
> /*
>  * Map a set of buffers described by scatterlist in streaming mode for DMA.
>  * This is the scatter-gather version of the above swiotlb_map_single
>@@ -545,8 +564,10 @@ swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nelems,
> 
> 	for (i = 0; i < nelems; i++, sg++) {
> 		dev_addr = SG_ENT_PHYS_ADDRESS(sg);
>-		if (range_straddles_page_boundary(page_to_pseudophys(sg->page)
>-						  + sg->offset, sg->length)
>+		if ((range_straddles_page_boundary(page_to_pseudophys(sg->page)
>+						   + sg->offset, sg->length) &&
>+		     !check_pages_physically_contiguous(sg->page, sg->offset, 
>+							sg->length))
> 		    || address_needs_mapping(hwdev, dev_addr)) {
> 			buffer.page   = sg->page;
> 			buffer.offset = sg->offset;

Actions: View | Diff

Attachments on bug 433554: 295354 | 295355 | 295526 | 298053 | 299461