From: Thierry Reding <thierry.reding-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> To: Joerg Roedel <joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org> Cc: linux-tegra-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Dmitry Osipenko <digetx-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>, iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, Jon Hunter <jonathanh-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org> Subject: [PATCH 3/3] iommu/tegra-smmu: Fix page tables in > 4 GiB memory Date: Wed, 16 Oct 2019 13:50:26 +0200 [thread overview] Message-ID: <20191016115026.1768745-3-thierry.reding@gmail.com> (raw) In-Reply-To: <20191016115026.1768745-1-thierry.reding-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> From: Thierry Reding <treding-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org> Page tables that reside in physical memory beyond the 4 GiB boundary are currently not working properly. The reason is that when the physical address for page directory entries is read, it gets truncated at 32 bits and can cause crashes when passing that address to the DMA API. Fix this by first casting the PDE value to a dma_addr_t and then using the page frame number mask for the SMMU instance to mask out the invalid bits, which are typically used for mapping attributes, etc. Signed-off-by: Thierry Reding <treding-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org> --- drivers/iommu/tegra-smmu.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c index 9425d01a95ac..63a147b623e6 100644 --- a/drivers/iommu/tegra-smmu.c +++ b/drivers/iommu/tegra-smmu.c @@ -159,9 +159,9 @@ static bool smmu_dma_addr_valid(struct tegra_smmu *smmu, dma_addr_t addr) return (addr & smmu->pfn_mask) == addr; } -static dma_addr_t smmu_pde_to_dma(u32 pde) +static dma_addr_t smmu_pde_to_dma(struct tegra_smmu *smmu, u32 pde) { - return pde << 12; + return (dma_addr_t)(pde & smmu->pfn_mask) << 12; } static void smmu_flush_ptc_all(struct tegra_smmu *smmu) @@ -554,6 +554,7 @@ static u32 *tegra_smmu_pte_lookup(struct tegra_smmu_as *as, unsigned long iova, dma_addr_t *dmap) { unsigned int pd_index = iova_pd_index(iova); + struct tegra_smmu *smmu = as->smmu; struct page *pt_page; u32 *pd; @@ -562,7 +563,7 @@ static u32 *tegra_smmu_pte_lookup(struct tegra_smmu_as *as, unsigned long iova, return NULL; pd = page_address(as->pd); - *dmap = smmu_pde_to_dma(pd[pd_index]); + *dmap = smmu_pde_to_dma(smmu, pd[pd_index]); return tegra_smmu_pte_offset(pt_page, iova); } @@ -604,7 +605,7 @@ static u32 *as_get_pte(struct tegra_smmu_as *as, dma_addr_t iova, } else { u32 *pd = page_address(as->pd); - *dmap = smmu_pde_to_dma(pd[pde]); + *dmap = smmu_pde_to_dma(smmu, pd[pde]); } return tegra_smmu_pte_offset(as->pts[pde], iova); @@ -629,7 +630,7 @@ static void tegra_smmu_pte_put_use(struct tegra_smmu_as *as, unsigned long iova) if (--as->count[pde] == 0) { struct tegra_smmu *smmu = as->smmu; u32 *pd = page_address(as->pd); - dma_addr_t pte_dma = smmu_pde_to_dma(pd[pde]); + dma_addr_t pte_dma = smmu_pde_to_dma(smmu, pd[pde]); tegra_smmu_set_pde(as, iova, 0); -- 2.23.0
WARNING: multiple messages have this Message-ID (diff)
From: Thierry Reding <thierry.reding@gmail.com> To: Joerg Roedel <joro@8bytes.org> Cc: linux-tegra@vger.kernel.org, Dmitry Osipenko <digetx@gmail.com>, iommu@lists.linux-foundation.org, Jon Hunter <jonathanh@nvidia.com> Subject: [PATCH 3/3] iommu/tegra-smmu: Fix page tables in > 4 GiB memory Date: Wed, 16 Oct 2019 13:50:26 +0200 [thread overview] Message-ID: <20191016115026.1768745-3-thierry.reding@gmail.com> (raw) In-Reply-To: <20191016115026.1768745-1-thierry.reding@gmail.com> From: Thierry Reding <treding@nvidia.com> Page tables that reside in physical memory beyond the 4 GiB boundary are currently not working properly. The reason is that when the physical address for page directory entries is read, it gets truncated at 32 bits and can cause crashes when passing that address to the DMA API. Fix this by first casting the PDE value to a dma_addr_t and then using the page frame number mask for the SMMU instance to mask out the invalid bits, which are typically used for mapping attributes, etc. Signed-off-by: Thierry Reding <treding@nvidia.com> --- drivers/iommu/tegra-smmu.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c index 9425d01a95ac..63a147b623e6 100644 --- a/drivers/iommu/tegra-smmu.c +++ b/drivers/iommu/tegra-smmu.c @@ -159,9 +159,9 @@ static bool smmu_dma_addr_valid(struct tegra_smmu *smmu, dma_addr_t addr) return (addr & smmu->pfn_mask) == addr; } -static dma_addr_t smmu_pde_to_dma(u32 pde) +static dma_addr_t smmu_pde_to_dma(struct tegra_smmu *smmu, u32 pde) { - return pde << 12; + return (dma_addr_t)(pde & smmu->pfn_mask) << 12; } static void smmu_flush_ptc_all(struct tegra_smmu *smmu) @@ -554,6 +554,7 @@ static u32 *tegra_smmu_pte_lookup(struct tegra_smmu_as *as, unsigned long iova, dma_addr_t *dmap) { unsigned int pd_index = iova_pd_index(iova); + struct tegra_smmu *smmu = as->smmu; struct page *pt_page; u32 *pd; @@ -562,7 +563,7 @@ static u32 *tegra_smmu_pte_lookup(struct tegra_smmu_as *as, unsigned long iova, return NULL; pd = page_address(as->pd); - *dmap = smmu_pde_to_dma(pd[pd_index]); + *dmap = smmu_pde_to_dma(smmu, pd[pd_index]); return tegra_smmu_pte_offset(pt_page, iova); } @@ -604,7 +605,7 @@ static u32 *as_get_pte(struct tegra_smmu_as *as, dma_addr_t iova, } else { u32 *pd = page_address(as->pd); - *dmap = smmu_pde_to_dma(pd[pde]); + *dmap = smmu_pde_to_dma(smmu, pd[pde]); } return tegra_smmu_pte_offset(as->pts[pde], iova); @@ -629,7 +630,7 @@ static void tegra_smmu_pte_put_use(struct tegra_smmu_as *as, unsigned long iova) if (--as->count[pde] == 0) { struct tegra_smmu *smmu = as->smmu; u32 *pd = page_address(as->pd); - dma_addr_t pte_dma = smmu_pde_to_dma(pd[pde]); + dma_addr_t pte_dma = smmu_pde_to_dma(smmu, pd[pde]); tegra_smmu_set_pde(as, iova, 0); -- 2.23.0 _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
next prev parent reply other threads:[~2019-10-16 11:50 UTC|newest] Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top 2019-10-16 11:50 [PATCH 1/3] iommu/tegra-smmu: Use non-secure register for flushing Thierry Reding 2019-10-16 11:50 ` Thierry Reding [not found] ` <20191016115026.1768745-1-thierry.reding-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 2019-10-16 11:50 ` [PATCH 2/3] iommu/tegra-smmu: Fix client enablement order Thierry Reding 2019-10-16 11:50 ` Thierry Reding 2019-10-16 11:50 ` Thierry Reding [this message] 2019-10-16 11:50 ` [PATCH 3/3] iommu/tegra-smmu: Fix page tables in > 4 GiB memory Thierry Reding 2019-10-18 9:46 ` [PATCH 1/3] iommu/tegra-smmu: Use non-secure register for flushing Joerg Roedel 2019-10-18 9:46 ` Joerg Roedel
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=20191016115026.1768745-3-thierry.reding@gmail.com \ --to=thierry.reding-re5jqeeqqe8avxtiumwx3w@public.gmane.org \ --cc=digetx-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \ --cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \ --cc=jonathanh-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org \ --cc=joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org \ --cc=linux-tegra-u79uwXL29TY76Z2rM5mHXA@public.gmane.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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.