linux-kernel - [PATCH] swiotlb: Suppress "Out of SW-IOMMU" errors for NO

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Wed,  2 May 2018 11:05:01 +0100
From:   Chris Wilson <chris@...is-wilson.co.uk>
To:     iommu@...ts.linux-foundation.org
Cc:     linux-kernel@...r.kernel.org, intel-gfx@...ts.freedesktop.org,
        Chris Wilson <chris@...is-wilson.co.uk>,
        Christian König <christian.koenig@....com>,
        Mike Galbraith <efault@....de>,
        Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
        Christoph Hellwig <hch@....de>
Subject: [PATCH] swiotlb: Suppress "Out of SW-IOMMU" errors for NO_WARN

This extends the warning suppression from commit d0bc0c2a31c9 ("swiotlb:
suppress warning when __GFP_NOWARN is set") to suppress the warnings
when DMA_ATTR_NO_WARN is given by caller. In such cases the caller wants
to handle the error themselves.

Fixes: d0bc0c2a31c9 ("swiotlb: suppress warning when __GFP_NOWARN is set")
Cc: Christian König <christian.koenig@....com>
Cc: Mike Galbraith <efault@....de>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
Cc: Christoph Hellwig <hch@....de>
---
 lib/swiotlb.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index fece57566d45..2bfb936c5708 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -768,11 +768,14 @@ static bool swiotlb_free_buffer(struct device *dev, size_t size,
 
 static void
 swiotlb_full(struct device *dev, size_t size, enum dma_data_direction dir,
-	     int do_panic)
+	     unsigned long attrs, int do_panic)
 {
 	if (swiotlb_force == SWIOTLB_NO_FORCE)
 		return;
 
+	if (attrs & DMA_ATTR_NO_WARN)
+		return;
+
 	/*
 	 * Ran out of IOMMU space for this operation. This is very bad.
 	 * Unfortunately the drivers cannot handle this operation properly.
@@ -823,7 +826,7 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
 	/* Oh well, have to allocate and map a bounce buffer. */
 	map = map_single(dev, phys, size, dir, attrs);
 	if (map == SWIOTLB_MAP_ERROR) {
-		swiotlb_full(dev, size, dir, 1);
+		swiotlb_full(dev, size, dir, attrs, 1);
 		return __phys_to_dma(dev, io_tlb_overflow_buffer);
 	}
 
@@ -959,7 +962,7 @@ swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems,
 			if (map == SWIOTLB_MAP_ERROR) {
 				/* Don't panic here, we expect map_sg users
 				   to do proper error handling. */
-				swiotlb_full(hwdev, sg->length, dir, 0);
+				swiotlb_full(hwdev, sg->length, dir, attrs, 0);
 				attrs |= DMA_ATTR_SKIP_CPU_SYNC;
 				swiotlb_unmap_sg_attrs(hwdev, sgl, i, dir,
 						       attrs);
-- 
2.17.0