Linux IOMMU Development
 help / color / mirror / Atom feed
From: stian@itx.no
To: sparclinux@vger.kernel.org
Cc: iommu@lists.linux.dev, m.szyprowski@samsung.com,
	Stian Halseth <stian@itx.no>
Subject: [PATCH] sparc: iommu/pci_sun4v: fix page alignment in dma mapping
Date: Tue, 17 Feb 2026 16:03:51 +0100	[thread overview]
Message-ID: <20260217150350.3541-2-stian@itx.no> (raw)
In-Reply-To: <7131d609-b083-4264-818a-21de38c1c3ed@samsung.com>

From: Stian Halseth <stian@itx.no>

'phys' may include an offset within the page, while 'base_paddr' is already page-aligned.
This caused incorrect DMA mapping in dma_4u_map_phys and dma_4v_map_phys.

Fix both functions by masking phys with IO_PAGE_MASK or subtracting the page offset.

Fixes: 38c0d0ebf520 ("sparc: Use physical address DMA mapping")

Reported-by: Stian Halseth <stian@itx.no>
Suggested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Stian Halseth <stian@itx.no>
---
 arch/sparc/kernel/iommu.c     | 2 ++
 arch/sparc/kernel/pci_sun4v.c | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/sparc/kernel/iommu.c b/arch/sparc/kernel/iommu.c
index 46ef88bc9c26..f3755a388ac7 100644
--- a/arch/sparc/kernel/iommu.c
+++ b/arch/sparc/kernel/iommu.c
@@ -312,6 +312,8 @@ static dma_addr_t dma_4u_map_phys(struct device *dev, phys_addr_t phys,
 	if (direction != DMA_TO_DEVICE)
 		iopte_protection |= IOPTE_WRITE;
 
+	phys -= offset_in_page(phys);
+
 	for (i = 0; i < npages; i++, base++, phys += IO_PAGE_SIZE)
 		iopte_val(*base) = iopte_protection | phys;
 
diff --git a/arch/sparc/kernel/pci_sun4v.c b/arch/sparc/kernel/pci_sun4v.c
index 791f0a76665f..2f30eeac4861 100644
--- a/arch/sparc/kernel/pci_sun4v.c
+++ b/arch/sparc/kernel/pci_sun4v.c
@@ -411,7 +411,7 @@ static dma_addr_t dma_4v_map_phys(struct device *dev, phys_addr_t phys,
 	iommu_batch_start(dev, prot, entry);
 
 	for (i = 0; i < npages; i++, phys += IO_PAGE_SIZE) {
-		long err = iommu_batch_add(phys, mask);
+		long err = iommu_batch_add(phys & IO_PAGE_MASK, mask);
 		if (unlikely(err < 0L))
 			goto iommu_map_fail;
 	}
-- 
2.53.0


  parent reply	other threads:[~2026-02-17 15:10 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20251015091313eucas1p2426ae40b579d7f3b4f29b46e0e788c0d@eucas1p2.samsung.com>
2025-10-15  9:12 ` [PATCH v5 00/14] Remove DMA map_page/map_resource and their unmap callbacks Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 01/14] dma-mapping: prepare dma_map_ops to conversion to physical address Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 02/14] dma-mapping: convert dummy ops to physical address mapping Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 03/14] ARM: dma-mapping: Reduce struct page exposure in arch_sync_dma*() Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 04/14] ARM: dma-mapping: Switch to physical address mapping callbacks Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 05/14] xen: swiotlb: " Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 06/14] dma-mapping: remove unused mapping resource callbacks Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 07/14] alpha: Convert mapping routine to rely on physical address Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 08/14] MIPS/jazzdma: Provide physical address directly Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 09/14] parisc: Convert DMA map_page to map_phys interface Leon Romanovsky
2025-12-17 12:17     ` [PATCH v5 09/14] parisc: Convert DMA map_page to map_phys interface [qemu test failure] Guenter Roeck
2025-12-18  8:40       ` Leon Romanovsky
2025-12-18 12:11         ` Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 10/14] powerpc: Convert to physical address DMA mapping Leon Romanovsky
2025-11-14  9:02     ` Christophe Leroy
2025-10-15  9:12   ` [PATCH v5 11/14] sparc: Use " Leon Romanovsky
2026-02-17 12:27     ` John Paul Adrian Glaubitz
2026-02-17 12:55       ` Marek Szyprowski
2026-02-17 13:15         ` Leon Romanovsky
2026-02-17 13:57         ` John Paul Adrian Glaubitz
2026-02-17 15:03         ` stian [this message]
2026-02-17 15:42           ` [PATCH] sparc: iommu/pci_sun4v: fix page alignment in dma mapping John Paul Adrian Glaubitz
2025-10-15  9:12   ` [PATCH v5 12/14] x86: Use physical address for DMA mapping Leon Romanovsky
2025-10-15  9:12   ` [PATCH v5 13/14] xen: swiotlb: Convert mapping routine to rely on physical address Leon Romanovsky
2025-10-15  9:13   ` [PATCH v5 14/14] dma-mapping: remove unused map_page callback Leon Romanovsky
2025-10-20 16:27   ` [PATCH v5 00/14] Remove DMA map_page/map_resource and their unmap callbacks Marek Szyprowski
2025-10-26  7:38     ` Leon Romanovsky
2025-10-29 10:00   ` Marek Szyprowski

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260217150350.3541-2-stian@itx.no \
    --to=stian@itx.no \
    --cc=iommu@lists.linux.dev \
    --cc=m.szyprowski@samsung.com \
    --cc=sparclinux@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox