linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Georgi Djakov <quic_c_gdjako@quicinc.com>
To: <will@kernel.org>, <robin.murphy@arm.com>
Cc: <joro@8bytes.org>, <isaacm@codeaurora.org>,
	<baolu.lu@linux.intel.com>, <pratikp@codeaurora.org>,
	<iommu@lists.linux-foundation.org>,
	<linux-arm-kernel@lists.infradead.org>,
	<linux-kernel@vger.kernel.org>, <djakov@kernel.org>
Subject: [PATCH v7 07/15] iommu: Hook up '->unmap_pages' driver callback
Date: Wed, 16 Jun 2021 06:38:48 -0700	[thread overview]
Message-ID: <1623850736-389584-8-git-send-email-quic_c_gdjako@quicinc.com> (raw)
In-Reply-To: <1623850736-389584-1-git-send-email-quic_c_gdjako@quicinc.com>

From: Will Deacon <will@kernel.org>

Extend iommu_pgsize() to populate an optional 'count' parameter so that
we can direct unmapping operation to the ->unmap_pages callback if it
has been provided by the driver.

Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
Signed-off-by: Georgi Djakov <quic_c_gdjako@quicinc.com>
---
 drivers/iommu/iommu.c | 59 +++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 50 insertions(+), 9 deletions(-)

diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 80e14c139d40..725622c7e603 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -2376,11 +2376,11 @@ phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
 EXPORT_SYMBOL_GPL(iommu_iova_to_phys);
 
 static size_t iommu_pgsize(struct iommu_domain *domain, unsigned long iova,
-			   phys_addr_t paddr, size_t size)
+			   phys_addr_t paddr, size_t size, size_t *count)
 {
-	unsigned int pgsize_idx;
+	unsigned int pgsize_idx, pgsize_idx_next;
 	unsigned long pgsizes;
-	size_t pgsize;
+	size_t offset, pgsize, pgsize_next;
 	unsigned long addr_merge = paddr | iova;
 
 	/* Page sizes supported by the hardware and small enough for @size */
@@ -2396,7 +2396,36 @@ static size_t iommu_pgsize(struct iommu_domain *domain, unsigned long iova,
 	/* Pick the biggest page size remaining */
 	pgsize_idx = __fls(pgsizes);
 	pgsize = BIT(pgsize_idx);
+	if (!count)
+		return pgsize;
 
+	/* Find the next biggest support page size, if it exists */
+	pgsizes = domain->pgsize_bitmap & ~GENMASK(pgsize_idx, 0);
+	if (!pgsizes)
+		goto out_set_count;
+
+	pgsize_idx_next = __ffs(pgsizes);
+	pgsize_next = BIT(pgsize_idx_next);
+
+	/*
+	 * There's no point trying a bigger page size unless the virtual
+	 * and physical addresses are similarly offset within the larger page.
+	 */
+	if ((iova ^ paddr) & (pgsize_next - 1))
+		goto out_set_count;
+
+	/* Calculate the offset to the next page size alignment boundary */
+	offset = pgsize_next - (addr_merge & (pgsize_next - 1));
+
+	/*
+	 * If size is big enough to accommodate the larger page, reduce
+	 * the number of smaller pages.
+	 */
+	if (offset + pgsize_next <= size)
+		size = offset;
+
+out_set_count:
+	*count = size >> pgsize_idx;
 	return pgsize;
 }
 
@@ -2434,7 +2463,7 @@ static int __iommu_map(struct iommu_domain *domain, unsigned long iova,
 	pr_debug("map: iova 0x%lx pa %pa size 0x%zx\n", iova, &paddr, size);
 
 	while (size) {
-		size_t pgsize = iommu_pgsize(domain, iova, paddr, size);
+		size_t pgsize = iommu_pgsize(domain, iova, paddr, size, NULL);
 
 		pr_debug("mapping: iova 0x%lx pa %pa pgsize 0x%zx\n",
 			 iova, &paddr, pgsize);
@@ -2485,6 +2514,19 @@ int iommu_map_atomic(struct iommu_domain *domain, unsigned long iova,
 }
 EXPORT_SYMBOL_GPL(iommu_map_atomic);
 
+static size_t __iommu_unmap_pages(struct iommu_domain *domain,
+				  unsigned long iova, size_t size,
+				  struct iommu_iotlb_gather *iotlb_gather)
+{
+	const struct iommu_ops *ops = domain->ops;
+	size_t pgsize, count;
+
+	pgsize = iommu_pgsize(domain, iova, iova, size, &count);
+	return ops->unmap_pages ?
+	       ops->unmap_pages(domain, iova, pgsize, count, iotlb_gather) :
+	       ops->unmap(domain, iova, pgsize, iotlb_gather);
+}
+
 static size_t __iommu_unmap(struct iommu_domain *domain,
 			    unsigned long iova, size_t size,
 			    struct iommu_iotlb_gather *iotlb_gather)
@@ -2494,7 +2536,7 @@ static size_t __iommu_unmap(struct iommu_domain *domain,
 	unsigned long orig_iova = iova;
 	unsigned int min_pagesz;
 
-	if (unlikely(ops->unmap == NULL ||
+	if (unlikely(!(ops->unmap || ops->unmap_pages) ||
 		     domain->pgsize_bitmap == 0UL))
 		return 0;
 
@@ -2522,10 +2564,9 @@ static size_t __iommu_unmap(struct iommu_domain *domain,
 	 * or we hit an area that isn't mapped.
 	 */
 	while (unmapped < size) {
-		size_t pgsize;
-
-		pgsize = iommu_pgsize(domain, iova, iova, size - unmapped);
-		unmapped_page = ops->unmap(domain, iova, pgsize, iotlb_gather);
+		unmapped_page = __iommu_unmap_pages(domain, iova,
+						    size - unmapped,
+						    iotlb_gather);
 		if (!unmapped_page)
 			break;
 

  parent reply	other threads:[~2021-06-16 13:39 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-16 13:38 [PATCH v7 00/15] Optimizing iommu_[map/unmap] performance Georgi Djakov
2021-06-16 13:38 ` [PATCH v7 01/15] iommu/io-pgtable: Introduce unmap_pages() as a page table op Georgi Djakov
2021-06-16 13:38 ` [PATCH v7 02/15] iommu: Add an unmap_pages() op for IOMMU drivers Georgi Djakov
2021-06-16 13:38 ` [PATCH v7 03/15] iommu/io-pgtable: Introduce map_pages() as a page table op Georgi Djakov
2021-06-16 13:38 ` [PATCH v7 04/15] iommu: Add a map_pages() op for IOMMU drivers Georgi Djakov
2021-06-16 13:38 ` [PATCH v7 05/15] iommu: Use bitmap to calculate page size in iommu_pgsize() Georgi Djakov
2021-06-17  7:16   ` Lu Baolu
2021-06-16 13:38 ` [PATCH v7 06/15] iommu: Split 'addr_merge' argument to iommu_pgsize() into separate parts Georgi Djakov
2021-06-17  7:17   ` Lu Baolu
2021-06-16 13:38 ` Georgi Djakov [this message]
2021-06-17  7:18   ` [PATCH v7 07/15] iommu: Hook up '->unmap_pages' driver callback Lu Baolu
2021-06-16 13:38 ` [PATCH v7 08/15] iommu: Add support for the map_pages() callback Georgi Djakov
2021-06-17  7:18   ` Lu Baolu
2021-06-16 13:38 ` [PATCH v7 09/15] iommu/io-pgtable-arm: Prepare PTE methods for handling multiple entries Georgi Djakov
2021-06-16 13:38 ` [PATCH v7 10/15] iommu/io-pgtable-arm: Implement arm_lpae_unmap_pages() Georgi Djakov
2021-07-15  9:31   ` Kunkun Jiang
2021-06-16 13:38 ` [PATCH v7 11/15] iommu/io-pgtable-arm: Implement arm_lpae_map_pages() Georgi Djakov
2021-06-16 13:38 ` [PATCH v7 12/15] iommu/io-pgtable-arm-v7s: Implement arm_v7s_unmap_pages() Georgi Djakov
2021-06-16 13:38 ` [PATCH v7 13/15] iommu/io-pgtable-arm-v7s: Implement arm_v7s_map_pages() Georgi Djakov
2021-06-16 13:38 ` [PATCH v7 14/15] iommu/arm-smmu: Implement the unmap_pages() IOMMU driver callback Georgi Djakov
2021-06-16 13:38 ` [PATCH v7 15/15] iommu/arm-smmu: Implement the map_pages() " Georgi Djakov
2021-07-14 14:24 ` [PATCH v7 00/15] Optimizing iommu_[map/unmap] performance Georgi Djakov
2021-07-15  1:23   ` Lu Baolu
2021-07-15  1:51     ` chenxiang (M)
2021-07-26 10:37 ` 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=1623850736-389584-8-git-send-email-quic_c_gdjako@quicinc.com \
    --to=quic_c_gdjako@quicinc.com \
    --cc=baolu.lu@linux.intel.com \
    --cc=djakov@kernel.org \
    --cc=iommu@lists.linux-foundation.org \
    --cc=isaacm@codeaurora.org \
    --cc=joro@8bytes.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pratikp@codeaurora.org \
    --cc=robin.murphy@arm.com \
    --cc=will@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;
as well as URLs for NNTP newsgroup(s).