[PATCH rdma-next v4 0/3] SG fix together with update to RDMA umem

[PATCH rdma-next v4 0/3] SG fix together with update to RDMA umem

[Maor Gottlieb]

From: Maor Gottlieb <maorg@nvidia.com>

Changelog:
v4:
 * Unify sg_free_table_entries with __sg_free_table
v3: https://lore.kernel.org/lkml/cover.1627551226.git.leonro@nvidia.com/
 * Rewrote to new API suggestion
 * Split for more patches
v2: https://lore.kernel.org/lkml/cover.1626605893.git.leonro@nvidia.com
 * Changed implementation of first patch, based on our discussion with
 * Christoph.
   https://lore.kernel.org/lkml/YNwaVTT0qmQdxaZz@infradead.org/
v1: https://lore.kernel.org/lkml/cover.1624955710.git.leonro@nvidia.com/
 * Fixed sg_page with a _dma_ API in the umem.c
v0: https://lore.kernel.org/lkml/cover.1624361199.git.leonro@nvidia.com

Maor Gottlieb (3):
  lib/scatterlist: Provide a dedicated function to support table append
  lib/scatterlist: Fix wrong update of orig_nents
  RDMA: Use the sg_table directly and remove the opencoded version from
    umem

 drivers/gpu/drm/drm_prime.c                 |  13 +-
 drivers/gpu/drm/i915/gem/i915_gem_userptr.c |  11 +-
 drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c  |  14 +-
 drivers/infiniband/core/umem.c              |  56 ++++---
 drivers/infiniband/core/umem_dmabuf.c       |   5 +-
 drivers/infiniband/hw/hns/hns_roce_db.c     |   4 +-
 drivers/infiniband/hw/irdma/verbs.c         |   2 +-
 drivers/infiniband/hw/mlx4/doorbell.c       |   3 +-
 drivers/infiniband/hw/mlx4/mr.c             |   4 +-
 drivers/infiniband/hw/mlx5/doorbell.c       |   3 +-
 drivers/infiniband/hw/mlx5/mr.c             |   3 +-
 drivers/infiniband/hw/qedr/verbs.c          |   2 +-
 drivers/infiniband/sw/rdmavt/mr.c           |   2 +-
 drivers/infiniband/sw/rxe/rxe_mr.c          |   2 +-
 include/linux/scatterlist.h                 |  56 +++++--
 include/rdma/ib_umem.h                      |  11 +-
 include/rdma/ib_verbs.h                     |  28 ++++
 lib/scatterlist.c                           | 155 ++++++++++++--------
 lib/sg_pool.c                               |   3 +-
 tools/testing/scatterlist/main.c            |  38 +++--
 20 files changed, 258 insertions(+), 157 deletions(-)

-- 
2.25.4

[PATCH rdma-next v4 1/3] lib/scatterlist: Provide a dedicated function to support table append

[Maor Gottlieb]

RDMA is the only in-kernel user that uses __sg_alloc_table_from_pages to
append pages dynamically. In the next patch. That mode will be extended
and that function will get more parameters. So separate it into a unique
function to make such change more clear.

Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 drivers/gpu/drm/drm_prime.c                 | 13 ++++---
 drivers/gpu/drm/i915/gem/i915_gem_userptr.c | 11 +++---
 drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c  | 14 +++-----
 drivers/infiniband/core/umem.c              |  4 +--
 include/linux/scatterlist.h                 | 39 ++++++++++++++++++---
 lib/scatterlist.c                           | 36 ++++++++++---------
 tools/testing/scatterlist/main.c            | 25 +++++++++----
 7 files changed, 90 insertions(+), 52 deletions(-)

diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
index 2a54f86856af..cf3278041f9c 100644
--- a/drivers/gpu/drm/drm_prime.c
+++ b/drivers/gpu/drm/drm_prime.c
@@ -807,8 +807,8 @@ struct sg_table *drm_prime_pages_to_sg(struct drm_device *dev,
 				       struct page **pages, unsigned int nr_pages)
 {
 	struct sg_table *sg;
-	struct scatterlist *sge;
 	size_t max_segment = 0;
+	int err;
 
 	sg = kmalloc(sizeof(struct sg_table), GFP_KERNEL);
 	if (!sg)
@@ -818,13 +818,12 @@ struct sg_table *drm_prime_pages_to_sg(struct drm_device *dev,
 		max_segment = dma_max_mapping_size(dev->dev);
 	if (max_segment == 0)
 		max_segment = UINT_MAX;
-	sge = __sg_alloc_table_from_pages(sg, pages, nr_pages, 0,
-					  nr_pages << PAGE_SHIFT,
-					  max_segment,
-					  NULL, 0, GFP_KERNEL);
-	if (IS_ERR(sge)) {
+	err = sg_alloc_table_from_pages_segment(sg, pages, nr_pages, 0,
+						nr_pages << PAGE_SHIFT,
+						max_segment, GFP_KERNEL);
+	if (err) {
 		kfree(sg);
-		sg = ERR_CAST(sge);
+		sg = ERR_PTR(err);
 	}
 	return sg;
 }
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
index 7487bab11f0b..458f797a9e1e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
@@ -133,7 +133,6 @@ static int i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 	unsigned int max_segment = i915_sg_segment_size();
 	struct sg_table *st;
 	unsigned int sg_page_sizes;
-	struct scatterlist *sg;
 	struct page **pvec;
 	int ret;
 
@@ -153,13 +152,11 @@ static int i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 	spin_unlock(&i915->mm.notifier_lock);
 
 alloc_table:
-	sg = __sg_alloc_table_from_pages(st, pvec, num_pages, 0,
-					 num_pages << PAGE_SHIFT, max_segment,
-					 NULL, 0, GFP_KERNEL);
-	if (IS_ERR(sg)) {
-		ret = PTR_ERR(sg);
+	ret = sg_alloc_table_from_pages_segment(st, pvec, num_pages, 0,
+						num_pages << PAGE_SHIFT,
+						max_segment, GFP_KERNEL);
+	if (ret)
 		goto err;
-	}
 
 	ret = i915_gem_gtt_prepare_pages(obj, st);
 	if (ret) {
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
index 0488042fb287..fc372d2e52a1 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
@@ -363,7 +363,6 @@ static int vmw_ttm_map_dma(struct vmw_ttm_tt *vmw_tt)
 	int ret = 0;
 	static size_t sgl_size;
 	static size_t sgt_size;
-	struct scatterlist *sg;
 
 	if (vmw_tt->mapped)
 		return 0;
@@ -386,15 +385,12 @@ static int vmw_ttm_map_dma(struct vmw_ttm_tt *vmw_tt)
 		if (unlikely(ret != 0))
 			return ret;
 
-		sg = __sg_alloc_table_from_pages(&vmw_tt->sgt, vsgt->pages,
-				vsgt->num_pages, 0,
-				(unsigned long) vsgt->num_pages << PAGE_SHIFT,
-				dma_get_max_seg_size(dev_priv->drm.dev),
-				NULL, 0, GFP_KERNEL);
-		if (IS_ERR(sg)) {
-			ret = PTR_ERR(sg);
+		ret = sg_alloc_table_from_pages_segment(
+			&vmw_tt->sgt, vsgt->pages, vsgt->num_pages, 0,
+			(unsigned long)vsgt->num_pages << PAGE_SHIFT,
+			dma_get_max_seg_size(dev_priv->drm.dev), GFP_KERNEL);
+		if (ret)
 			goto out_sg_alloc_fail;
-		}
 
 		if (vsgt->num_pages > vmw_tt->sgt.orig_nents) {
 			uint64_t over_alloc =
diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c
index 0eb40025075f..b741758e528f 100644
--- a/drivers/infiniband/core/umem.c
+++ b/drivers/infiniband/core/umem.c
@@ -226,8 +226,8 @@ struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr,
 
 		cur_base += ret * PAGE_SIZE;
 		npages -= ret;
-		sg = __sg_alloc_table_from_pages(&umem->sg_head, page_list, ret,
-				0, ret << PAGE_SHIFT,
+		sg = sg_alloc_append_table_from_pages(&umem->sg_head, page_list,
+				ret, 0, ret << PAGE_SHIFT,
 				ib_dma_max_seg_size(device), sg, npages,
 				GFP_KERNEL);
 		umem->sg_nents = umem->sg_head.nents;
diff --git a/include/linux/scatterlist.h b/include/linux/scatterlist.h
index ecf87484814f..5c700f2a0d18 100644
--- a/include/linux/scatterlist.h
+++ b/include/linux/scatterlist.h
@@ -285,14 +285,45 @@ void sg_free_table(struct sg_table *);
 int __sg_alloc_table(struct sg_table *, unsigned int, unsigned int,
 		     struct scatterlist *, unsigned int, gfp_t, sg_alloc_fn *);
 int sg_alloc_table(struct sg_table *, unsigned int, gfp_t);
-struct scatterlist *__sg_alloc_table_from_pages(struct sg_table *sgt,
+struct scatterlist *sg_alloc_append_table_from_pages(struct sg_table *sgt,
 		struct page **pages, unsigned int n_pages, unsigned int offset,
 		unsigned long size, unsigned int max_segment,
 		struct scatterlist *prv, unsigned int left_pages,
 		gfp_t gfp_mask);
-int sg_alloc_table_from_pages(struct sg_table *sgt, struct page **pages,
-			      unsigned int n_pages, unsigned int offset,
-			      unsigned long size, gfp_t gfp_mask);
+int sg_alloc_table_from_pages_segment(struct sg_table *sgt, struct page **pages,
+				      unsigned int n_pages, unsigned int offset,
+				      unsigned long size,
+				      unsigned int max_segment, gfp_t gfp_mask);
+
+/**
+ * sg_alloc_table_from_pages - Allocate and initialize an sg table from
+ *			       an array of pages
+ * @sgt:	 The sg table header to use
+ * @pages:	 Pointer to an array of page pointers
+ * @n_pages:	 Number of pages in the pages array
+ * @offset:      Offset from start of the first page to the start of a buffer
+ * @size:        Number of valid bytes in the buffer (after offset)
+ * @gfp_mask:	 GFP allocation mask
+ *
+ *  Description:
+ *    Allocate and initialize an sg table from a list of pages. Contiguous
+ *    ranges of the pages are squashed into a single scatterlist node. A user
+ *    may provide an offset at a start and a size of valid data in a buffer
+ *    specified by the page array. The returned sg table is released by
+ *    sg_free_table.
+ *
+ * Returns:
+ *   0 on success, negative error on failure
+ */
+static inline int sg_alloc_table_from_pages(struct sg_table *sgt,
+					    struct page **pages,
+					    unsigned int n_pages,
+					    unsigned int offset,
+					    unsigned long size, gfp_t gfp_mask)
+{
+	return sg_alloc_table_from_pages_segment(sgt, pages, n_pages, offset,
+						 size, UINT_MAX, gfp_mask);
+}
 
 #ifdef CONFIG_SGL_ALLOC
 struct scatterlist *sgl_alloc_order(unsigned long long length,
diff --git a/lib/scatterlist.c b/lib/scatterlist.c
index 27efa6178153..611c63d4a958 100644
--- a/lib/scatterlist.c
+++ b/lib/scatterlist.c
@@ -397,7 +397,7 @@ static struct scatterlist *get_next_sg(struct sg_table *table,
 }
 
 /**
- * __sg_alloc_table_from_pages - Allocate and initialize an sg table from
+ * sg_alloc_append_table_from_pages - Allocate and initialize an sg table from
  *			         an array of pages
  * @sgt:	 The sg table header to use
  * @pages:	 Pointer to an array of page pointers
@@ -425,7 +425,7 @@ static struct scatterlist *get_next_sg(struct sg_table *table,
  *   If this function returns non-0 (eg failure), the caller must call
  *   sg_free_table() to cleanup any leftover allocations.
  */
-struct scatterlist *__sg_alloc_table_from_pages(struct sg_table *sgt,
+struct scatterlist *sg_alloc_append_table_from_pages(struct sg_table *sgt,
 		struct page **pages, unsigned int n_pages, unsigned int offset,
 		unsigned long size, unsigned int max_segment,
 		struct scatterlist *prv, unsigned int left_pages,
@@ -520,36 +520,40 @@ struct scatterlist *__sg_alloc_table_from_pages(struct sg_table *sgt,
 		sg_mark_end(s);
 	return s;
 }
-EXPORT_SYMBOL(__sg_alloc_table_from_pages);
+EXPORT_SYMBOL(sg_alloc_append_table_from_pages);
 
 /**
- * sg_alloc_table_from_pages - Allocate and initialize an sg table from
- *			       an array of pages
+ * sg_alloc_table_from_pages_segment - Allocate and initialize an sg table from
+ *                                     an array of pages and given maximum
+ *                                     segment.
  * @sgt:	 The sg table header to use
  * @pages:	 Pointer to an array of page pointers
  * @n_pages:	 Number of pages in the pages array
  * @offset:      Offset from start of the first page to the start of a buffer
  * @size:        Number of valid bytes in the buffer (after offset)
+ * @max_segment: Maximum size of a scatterlist element in bytes
  * @gfp_mask:	 GFP allocation mask
  *
  *  Description:
  *    Allocate and initialize an sg table from a list of pages. Contiguous
- *    ranges of the pages are squashed into a single scatterlist node. A user
- *    may provide an offset at a start and a size of valid data in a buffer
- *    specified by the page array. The returned sg table is released by
- *    sg_free_table.
+ *    ranges of the pages are squashed into a single scatterlist node up to the
+ *    maximum size specified in @max_segment. A user may provide an offset at a
+ *    start and a size of valid data in a buffer specified by the page array.
  *
- * Returns:
+ *    The returned sg table is released by sg_free_table.
+ *
+ *  Returns:
  *   0 on success, negative error on failure
  */
-int sg_alloc_table_from_pages(struct sg_table *sgt, struct page **pages,
-			      unsigned int n_pages, unsigned int offset,
-			      unsigned long size, gfp_t gfp_mask)
+int sg_alloc_table_from_pages_segment(struct sg_table *sgt, struct page **pages,
+				unsigned int n_pages, unsigned int offset,
+				unsigned long size, unsigned int max_segment,
+				gfp_t gfp_mask)
 {
-	return PTR_ERR_OR_ZERO(__sg_alloc_table_from_pages(sgt, pages, n_pages,
-			offset, size, UINT_MAX, NULL, 0, gfp_mask));
+	return PTR_ERR_OR_ZERO(sg_alloc_append_table_from_pages(sgt, pages,
+			n_pages, offset, size, max_segment, NULL, 0, gfp_mask));
 }
-EXPORT_SYMBOL(sg_alloc_table_from_pages);
+EXPORT_SYMBOL(sg_alloc_table_from_pages_segment);
 
 #ifdef CONFIG_SGL_ALLOC
 
diff --git a/tools/testing/scatterlist/main.c b/tools/testing/scatterlist/main.c
index 652254754b4c..c2ff9179c2cc 100644
--- a/tools/testing/scatterlist/main.c
+++ b/tools/testing/scatterlist/main.c
@@ -87,28 +87,39 @@ int main(void)
 		int left_pages = test->pfn_app ? test->num_pages : 0;
 		struct page *pages[MAX_PAGES];
 		struct sg_table st;
-		struct scatterlist *sg;
+		struct scatterlist *sg = NULL;
+		int ret;
 
 		set_pages(pages, test->pfn, test->num_pages);
 
-		sg = __sg_alloc_table_from_pages(&st, pages, test->num_pages, 0,
-				test->size, test->max_seg, NULL, left_pages, GFP_KERNEL);
-		assert(PTR_ERR_OR_ZERO(sg) == test->alloc_ret);
+		if (test->pfn_app) {
+			sg = sg_alloc_append_table_from_pages(
+				&st, pages, test->num_pages, 0, test->size,
+				test->max_seg, NULL, left_pages, GFP_KERNEL);
+			assert(PTR_ERR_OR_ZERO(sg) == test->alloc_ret);
+		} else {
+			ret = sg_alloc_table_from_pages_segment(
+				&st, pages, test->num_pages, 0, test->size,
+				test->max_seg, GFP_KERNEL);
+			assert(ret == test->alloc_ret);
+		}
 
 		if (test->alloc_ret)
 			continue;
 
 		if (test->pfn_app) {
 			set_pages(pages, test->pfn_app, test->num_pages);
-			sg = __sg_alloc_table_from_pages(&st, pages, test->num_pages, 0,
-					test->size, test->max_seg, sg, 0, GFP_KERNEL);
+			sg = sg_alloc_append_table_from_pages(
+				&st, pages, test->num_pages, 0, test->size,
+				test->max_seg, sg, 0, GFP_KERNEL);
 
 			assert(PTR_ERR_OR_ZERO(sg) == test->alloc_ret);
 		}
 
 		VALIDATE(st.nents == test->expected_segments, &st, test);
 		if (!test->pfn_app)
-			VALIDATE(st.orig_nents == test->expected_segments, &st, test);
+			VALIDATE(st.orig_nents == test->expected_segments, &st,
+				 test);
 
 		sg_free_table(&st);
 	}
-- 
2.25.4

[PATCH rdma-next v4 2/3] lib/scatterlist: Fix wrong update of orig_nents

[Maor Gottlieb]

orig_nents should represent the number of entries with pages,
but __sg_alloc_table_from_pages sets orig_nents as the number of
total entries in the table. This is wrong when the API is used for
dynamic allocation where not all the table entries are mapped with
pages. It wasn't observed until now, since RDMA umem who uses this
API in the dynamic form doesn't use orig_nents implicit or explicit
by the scatterlist APIs.

Fix it by changing the append API to track the SG append table
state and have an API to free the append table according to the
total number of entries in the table.
Now all APIs set orig_nents as number of enries with pages.

Fixes: 07da1223ec93 ("lib/scatterlist: Add support in dynamic allocation of SG table from pages")
Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 drivers/infiniband/core/umem.c   |  34 +++++----
 include/linux/scatterlist.h      |  19 +++--
 include/rdma/ib_umem.h           |   1 +
 lib/scatterlist.c                | 127 +++++++++++++++++++------------
 lib/sg_pool.c                    |   3 +-
 tools/testing/scatterlist/main.c |  45 ++++++-----
 6 files changed, 137 insertions(+), 92 deletions(-)

diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c
index b741758e528f..42481e7a72e8 100644
--- a/drivers/infiniband/core/umem.c
+++ b/drivers/infiniband/core/umem.c
@@ -59,7 +59,7 @@ static void __ib_umem_release(struct ib_device *dev, struct ib_umem *umem, int d
 		unpin_user_page_range_dirty_lock(sg_page(sg),
 			DIV_ROUND_UP(sg->length, PAGE_SIZE), make_dirty);
 
-	sg_free_table(&umem->sg_head);
+	sg_free_append_table(&umem->sgt_append);
 }
 
 /**
@@ -155,8 +155,7 @@ struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr,
 	unsigned long dma_attr = 0;
 	struct mm_struct *mm;
 	unsigned long npages;
-	int ret;
-	struct scatterlist *sg = NULL;
+	int pinned, ret;
 	unsigned int gup_flags = FOLL_WRITE;
 
 	/*
@@ -216,28 +215,33 @@ struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr,
 
 	while (npages) {
 		cond_resched();
-		ret = pin_user_pages_fast(cur_base,
+		pinned = pin_user_pages_fast(cur_base,
 					  min_t(unsigned long, npages,
 						PAGE_SIZE /
 						sizeof(struct page *)),
 					  gup_flags | FOLL_LONGTERM, page_list);
-		if (ret < 0)
+		if (pinned < 0) {
+			ret = pinned;
 			goto umem_release;
+		}
 
-		cur_base += ret * PAGE_SIZE;
-		npages -= ret;
-		sg = sg_alloc_append_table_from_pages(&umem->sg_head, page_list,
-				ret, 0, ret << PAGE_SHIFT,
-				ib_dma_max_seg_size(device), sg, npages,
-				GFP_KERNEL);
-		umem->sg_nents = umem->sg_head.nents;
-		if (IS_ERR(sg)) {
-			unpin_user_pages_dirty_lock(page_list, ret, 0);
-			ret = PTR_ERR(sg);
+		cur_base += pinned * PAGE_SIZE;
+		npages -= pinned;
+		ret = sg_alloc_append_table_from_pages(
+			&umem->sgt_append, page_list, pinned, 0,
+			pinned << PAGE_SHIFT, ib_dma_max_seg_size(device),
+			npages, GFP_KERNEL);
+		umem->sg_nents = umem->sgt_append.sgt.nents;
+		if (ret) {
+			memcpy(&umem->sg_head.sgl, &umem->sgt_append.sgt,
+			       sizeof(umem->sgt_append.sgt));
+			unpin_user_pages_dirty_lock(page_list, pinned, 0);
 			goto umem_release;
 		}
 	}
 
+	memcpy(&umem->sg_head.sgl, &umem->sgt_append.sgt,
+	       sizeof(umem->sgt_append.sgt));
 	if (access & IB_ACCESS_RELAXED_ORDERING)
 		dma_attr |= DMA_ATTR_WEAK_ORDERING;
 
diff --git a/include/linux/scatterlist.h b/include/linux/scatterlist.h
index 5c700f2a0d18..266754a55327 100644
--- a/include/linux/scatterlist.h
+++ b/include/linux/scatterlist.h
@@ -39,6 +39,12 @@ struct sg_table {
 	unsigned int orig_nents;	/* original size of list */
 };
 
+struct sg_append_table {
+	struct sg_table sgt;		/* The scatter list table */
+	struct scatterlist *prv;	/* last populated sge in the table */
+	unsigned int total_nents;	/* Total entries in the table */
+};
+
 /*
  * Notes on SG table design.
  *
@@ -280,16 +286,17 @@ typedef struct scatterlist *(sg_alloc_fn)(unsigned int, gfp_t);
 typedef void (sg_free_fn)(struct scatterlist *, unsigned int);
 
 void __sg_free_table(struct sg_table *, unsigned int, unsigned int,
-		     sg_free_fn *);
+		     sg_free_fn *, unsigned int);
 void sg_free_table(struct sg_table *);
+void sg_free_append_table(struct sg_append_table *sgt);
 int __sg_alloc_table(struct sg_table *, unsigned int, unsigned int,
 		     struct scatterlist *, unsigned int, gfp_t, sg_alloc_fn *);
 int sg_alloc_table(struct sg_table *, unsigned int, gfp_t);
-struct scatterlist *sg_alloc_append_table_from_pages(struct sg_table *sgt,
-		struct page **pages, unsigned int n_pages, unsigned int offset,
-		unsigned long size, unsigned int max_segment,
-		struct scatterlist *prv, unsigned int left_pages,
-		gfp_t gfp_mask);
+int sg_alloc_append_table_from_pages(struct sg_append_table *sgt,
+				     struct page **pages, unsigned int n_pages,
+				     unsigned int offset, unsigned long size,
+				     unsigned int max_segment,
+				     unsigned int left_pages, gfp_t gfp_mask);
 int sg_alloc_table_from_pages_segment(struct sg_table *sgt, struct page **pages,
 				      unsigned int n_pages, unsigned int offset,
 				      unsigned long size,
diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h
index 676c57f5ca80..33cb23b2ee3c 100644
--- a/include/rdma/ib_umem.h
+++ b/include/rdma/ib_umem.h
@@ -26,6 +26,7 @@ struct ib_umem {
 	u32 is_odp : 1;
 	u32 is_dmabuf : 1;
 	struct work_struct	work;
+	struct sg_append_table  sgt_append;
 	struct sg_table sg_head;
 	int             nmap;
 	unsigned int    sg_nents;
diff --git a/lib/scatterlist.c b/lib/scatterlist.c
index 611c63d4a958..b587f90cca28 100644
--- a/lib/scatterlist.c
+++ b/lib/scatterlist.c
@@ -182,6 +182,7 @@ static void sg_kfree(struct scatterlist *sg, unsigned int nents)
  * @nents_first_chunk: Number of entries int the (preallocated) first
  * 	scatterlist chunk, 0 means no such preallocated first chunk
  * @free_fn:	Free function
+ * @num_ents:	Number of entries in the table
  *
  *  Description:
  *    Free an sg table previously allocated and setup with
@@ -190,7 +191,8 @@ static void sg_kfree(struct scatterlist *sg, unsigned int nents)
  *
  **/
 void __sg_free_table(struct sg_table *table, unsigned int max_ents,
-		     unsigned int nents_first_chunk, sg_free_fn *free_fn)
+		     unsigned int nents_first_chunk, sg_free_fn *free_fn,
+		     unsigned int num_ents)
 {
 	struct scatterlist *sgl, *next;
 	unsigned curr_max_ents = nents_first_chunk ?: max_ents;
@@ -199,8 +201,8 @@ void __sg_free_table(struct sg_table *table, unsigned int max_ents,
 		return;
 
 	sgl = table->sgl;
-	while (table->orig_nents) {
-		unsigned int alloc_size = table->orig_nents;
+	while (num_ents) {
+		unsigned int alloc_size = num_ents;
 		unsigned int sg_size;
 
 		/*
@@ -218,7 +220,7 @@ void __sg_free_table(struct sg_table *table, unsigned int max_ents,
 			next = NULL;
 		}
 
-		table->orig_nents -= sg_size;
+		num_ents -= sg_size;
 		if (nents_first_chunk)
 			nents_first_chunk = 0;
 		else
@@ -231,6 +233,19 @@ void __sg_free_table(struct sg_table *table, unsigned int max_ents,
 }
 EXPORT_SYMBOL(__sg_free_table);
 
+/**
+ * sg_free_append_table - Free a previously allocated append sg table.
+ * @table:	 The mapped sg append table header
+ *
+ **/
+void sg_free_append_table(struct sg_append_table *table)
+{
+	__sg_free_table(&table->sgt, SG_MAX_SINGLE_ALLOC, false, sg_kfree,
+			table->total_nents);
+}
+EXPORT_SYMBOL(sg_free_append_table);
+
+
 /**
  * sg_free_table - Free a previously allocated sg table
  * @table:	The mapped sg table header
@@ -238,7 +253,8 @@ EXPORT_SYMBOL(__sg_free_table);
  **/
 void sg_free_table(struct sg_table *table)
 {
-	__sg_free_table(table, SG_MAX_SINGLE_ALLOC, false, sg_kfree);
+	__sg_free_table(table, SG_MAX_SINGLE_ALLOC, false, sg_kfree,
+			table->orig_nents);
 }
 EXPORT_SYMBOL(sg_free_table);
 
@@ -359,13 +375,14 @@ int sg_alloc_table(struct sg_table *table, unsigned int nents, gfp_t gfp_mask)
 	ret = __sg_alloc_table(table, nents, SG_MAX_SINGLE_ALLOC,
 			       NULL, 0, gfp_mask, sg_kmalloc);
 	if (unlikely(ret))
-		__sg_free_table(table, SG_MAX_SINGLE_ALLOC, 0, sg_kfree);
+		__sg_free_table(table, SG_MAX_SINGLE_ALLOC, 0, sg_kfree,
+				table->orig_nents);
 
 	return ret;
 }
 EXPORT_SYMBOL(sg_alloc_table);
 
-static struct scatterlist *get_next_sg(struct sg_table *table,
+static struct scatterlist *get_next_sg(struct sg_append_table *table,
 				       struct scatterlist *cur,
 				       unsigned long needed_sges,
 				       gfp_t gfp_mask)
@@ -386,54 +403,52 @@ static struct scatterlist *get_next_sg(struct sg_table *table,
 		return ERR_PTR(-ENOMEM);
 	sg_init_table(new_sg, alloc_size);
 	if (cur) {
+		table->total_nents += alloc_size - 1;
 		__sg_chain(next_sg, new_sg);
-		table->orig_nents += alloc_size - 1;
 	} else {
-		table->sgl = new_sg;
-		table->orig_nents = alloc_size;
-		table->nents = 0;
+		table->sgt.sgl = new_sg;
+		table->total_nents = alloc_size;
 	}
 	return new_sg;
 }
 
 /**
- * sg_alloc_append_table_from_pages - Allocate and initialize an sg table from
- *			         an array of pages
- * @sgt:	 The sg table header to use
- * @pages:	 Pointer to an array of page pointers
- * @n_pages:	 Number of pages in the pages array
+ * sg_alloc_append_table_from_pages - Allocate and initialize an append sg
+ *                                    table from an array of pages
+ * @sgt_append:  The sg append table to use
+ * @pages:       Pointer to an array of page pointers
+ * @n_pages:     Number of pages in the pages array
  * @offset:      Offset from start of the first page to the start of a buffer
  * @size:        Number of valid bytes in the buffer (after offset)
  * @max_segment: Maximum size of a scatterlist element in bytes
- * @prv:	 Last populated sge in sgt
  * @left_pages:  Left pages caller have to set after this call
  * @gfp_mask:	 GFP allocation mask
  *
  * Description:
- *    If @prv is NULL, allocate and initialize an sg table from a list of pages,
- *    else reuse the scatterlist passed in at @prv.
- *    Contiguous ranges of the pages are squashed into a single scatterlist
- *    entry up to the maximum size specified in @max_segment.  A user may
- *    provide an offset at a start and a size of valid data in a buffer
- *    specified by the page array.
+ *    In the first call it allocate and initialize an sg table from a list of
+ *    pages, else reuse the scatterlist from sgt_append. Contiguous ranges of
+ *    the pages are squashed into a single scatterlist entry up to the maximum
+ *    size specified in @max_segment.  A user may provide an offset at a start
+ *    and a size of valid data in a buffer specified by the page array. The
+ *    returned sg table is released by sg_free_append_table
  *
  * Returns:
- *   Last SGE in sgt on success, PTR_ERR on otherwise.
- *   The allocation in @sgt must be released by sg_free_table.
+ *   0 on success, negative error on failure
  *
  * Notes:
  *   If this function returns non-0 (eg failure), the caller must call
- *   sg_free_table() to cleanup any leftover allocations.
+ *   sg_free_append_table() to cleanup any leftover allocations.
+ *
+ *   In the fist call, sgt_append must by initialized.
  */
-struct scatterlist *sg_alloc_append_table_from_pages(struct sg_table *sgt,
+int sg_alloc_append_table_from_pages(struct sg_append_table *sgt_append,
 		struct page **pages, unsigned int n_pages, unsigned int offset,
 		unsigned long size, unsigned int max_segment,
-		struct scatterlist *prv, unsigned int left_pages,
-		gfp_t gfp_mask)
+		unsigned int left_pages, gfp_t gfp_mask)
 {
 	unsigned int chunks, cur_page, seg_len, i, prv_len = 0;
 	unsigned int added_nents = 0;
-	struct scatterlist *s = prv;
+	struct scatterlist *s = sgt_append->prv;
 
 	/*
 	 * The algorithm below requires max_segment to be aligned to PAGE_SIZE
@@ -441,25 +456,26 @@ struct scatterlist *sg_alloc_append_table_from_pages(struct sg_table *sgt,
 	 */
 	max_segment = ALIGN_DOWN(max_segment, PAGE_SIZE);
 	if (WARN_ON(max_segment < PAGE_SIZE))
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
-	if (IS_ENABLED(CONFIG_ARCH_NO_SG_CHAIN) && prv)
-		return ERR_PTR(-EOPNOTSUPP);
+	if (IS_ENABLED(CONFIG_ARCH_NO_SG_CHAIN) && sgt_append->prv)
+		return -EOPNOTSUPP;
 
-	if (prv) {
-		unsigned long paddr = (page_to_pfn(sg_page(prv)) * PAGE_SIZE +
-				       prv->offset + prv->length) /
-				      PAGE_SIZE;
+	if (sgt_append->prv) {
+		unsigned long paddr =
+			(page_to_pfn(sg_page(sgt_append->prv)) * PAGE_SIZE +
+			 sgt_append->prv->offset + sgt_append->prv->length) /
+			PAGE_SIZE;
 
 		if (WARN_ON(offset))
-			return ERR_PTR(-EINVAL);
+			return -EINVAL;
 
 		/* Merge contiguous pages into the last SG */
-		prv_len = prv->length;
+		prv_len = sgt_append->prv->length;
 		while (n_pages && page_to_pfn(pages[0]) == paddr) {
-			if (prv->length + PAGE_SIZE > max_segment)
+			if (sgt_append->prv->length + PAGE_SIZE > max_segment)
 				break;
-			prv->length += PAGE_SIZE;
+			sgt_append->prv->length += PAGE_SIZE;
 			paddr++;
 			pages++;
 			n_pages--;
@@ -496,15 +512,16 @@ struct scatterlist *sg_alloc_append_table_from_pages(struct sg_table *sgt,
 		}
 
 		/* Pass how many chunks might be left */
-		s = get_next_sg(sgt, s, chunks - i + left_pages, gfp_mask);
+		s = get_next_sg(sgt_append, s, chunks - i + left_pages,
+				gfp_mask);
 		if (IS_ERR(s)) {
 			/*
 			 * Adjust entry length to be as before function was
 			 * called.
 			 */
-			if (prv)
-				prv->length = prv_len;
-			return s;
+			if (sgt_append->prv)
+				sgt_append->prv->length = prv_len;
+			return PTR_ERR(s);
 		}
 		chunk_size = ((j - cur_page) << PAGE_SHIFT) - offset;
 		sg_set_page(s, pages[cur_page],
@@ -514,11 +531,13 @@ struct scatterlist *sg_alloc_append_table_from_pages(struct sg_table *sgt,
 		offset = 0;
 		cur_page = j;
 	}
-	sgt->nents += added_nents;
+	sgt_append->sgt.nents += added_nents;
+	sgt_append->sgt.orig_nents = sgt_append->sgt.nents;
+	sgt_append->prv = s;
 out:
 	if (!left_pages)
 		sg_mark_end(s);
-	return s;
+	return 0;
 }
 EXPORT_SYMBOL(sg_alloc_append_table_from_pages);
 
@@ -550,8 +569,18 @@ int sg_alloc_table_from_pages_segment(struct sg_table *sgt, struct page **pages,
 				unsigned long size, unsigned int max_segment,
 				gfp_t gfp_mask)
 {
-	return PTR_ERR_OR_ZERO(sg_alloc_append_table_from_pages(sgt, pages,
-			n_pages, offset, size, max_segment, NULL, 0, gfp_mask));
+	struct sg_append_table append = {};
+	int err;
+
+	err = sg_alloc_append_table_from_pages(&append, pages, n_pages, offset,
+					       size, max_segment, 0, gfp_mask);
+	if (err) {
+		sg_free_append_table(&append);
+		return err;
+	}
+	memcpy(sgt, &append.sgt, sizeof(*sgt));
+	WARN_ON(append.total_nents != sgt->orig_nents);
+	return 0;
 }
 EXPORT_SYMBOL(sg_alloc_table_from_pages_segment);
 
diff --git a/lib/sg_pool.c b/lib/sg_pool.c
index db29e5c1f790..a0b1a52cd6f7 100644
--- a/lib/sg_pool.c
+++ b/lib/sg_pool.c
@@ -90,7 +90,8 @@ void sg_free_table_chained(struct sg_table *table,
 	if (nents_first_chunk == 1)
 		nents_first_chunk = 0;
 
-	__sg_free_table(table, SG_CHUNK_SIZE, nents_first_chunk, sg_pool_free);
+	__sg_free_table(table, SG_CHUNK_SIZE, nents_first_chunk, sg_pool_free,
+			table->orig_nents);
 }
 EXPORT_SYMBOL_GPL(sg_free_table_chained);
 
diff --git a/tools/testing/scatterlist/main.c b/tools/testing/scatterlist/main.c
index c2ff9179c2cc..08465a701cd5 100644
--- a/tools/testing/scatterlist/main.c
+++ b/tools/testing/scatterlist/main.c
@@ -85,43 +85,46 @@ int main(void)
 
 	for (i = 0, test = tests; test->expected_segments; test++, i++) {
 		int left_pages = test->pfn_app ? test->num_pages : 0;
+		struct sg_append_table append = {};
 		struct page *pages[MAX_PAGES];
-		struct sg_table st;
-		struct scatterlist *sg = NULL;
 		int ret;
 
 		set_pages(pages, test->pfn, test->num_pages);
 
-		if (test->pfn_app) {
-			sg = sg_alloc_append_table_from_pages(
-				&st, pages, test->num_pages, 0, test->size,
-				test->max_seg, NULL, left_pages, GFP_KERNEL);
-			assert(PTR_ERR_OR_ZERO(sg) == test->alloc_ret);
-		} else {
+		if (test->pfn_app)
+			ret = sg_alloc_append_table_from_pages(
+				&append, pages, test->num_pages, 0, test->size,
+				test->max_seg, left_pages, GFP_KERNEL);
+		else
 			ret = sg_alloc_table_from_pages_segment(
-				&st, pages, test->num_pages, 0, test->size,
-				test->max_seg, GFP_KERNEL);
-			assert(ret == test->alloc_ret);
-		}
+				&append.sgt, pages, test->num_pages, 0,
+				test->size, test->max_seg, GFP_KERNEL);
+
+		assert(ret == test->alloc_ret);
 
 		if (test->alloc_ret)
 			continue;
 
 		if (test->pfn_app) {
 			set_pages(pages, test->pfn_app, test->num_pages);
-			sg = sg_alloc_append_table_from_pages(
-				&st, pages, test->num_pages, 0, test->size,
-				test->max_seg, sg, 0, GFP_KERNEL);
+			ret = sg_alloc_append_table_from_pages(
+				&append, pages, test->num_pages, 0, test->size,
+				test->max_seg, 0, GFP_KERNEL);
 
-			assert(PTR_ERR_OR_ZERO(sg) == test->alloc_ret);
+			assert(ret == test->alloc_ret);
 		}
 
-		VALIDATE(st.nents == test->expected_segments, &st, test);
+		VALIDATE(append.sgt.nents == test->expected_segments,
+			 &append.sgt, test);
 		if (!test->pfn_app)
-			VALIDATE(st.orig_nents == test->expected_segments, &st,
-				 test);
-
-		sg_free_table(&st);
+			VALIDATE(append.sgt.orig_nents ==
+					 test->expected_segments,
+				 &append.sgt, test);
+
+		if (test->pfn_app)
+			sg_free_append_table(&append);
+		else
+			sg_free_table(&append.sgt);
 	}
 
 	assert(i == (sizeof(tests) / sizeof(tests[0])) - 1);
-- 
2.25.4

[PATCH rdma-next v4 3/3] RDMA: Use the sg_table directly and remove the opencoded version from umem

[Maor Gottlieb]

This allows using the normal sg_table APIs and makes all the code
cleaner. Remove sgt, nents and nmapd from ib_umem.

Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
---
 drivers/infiniband/core/umem.c          | 32 +++++++++----------------
 drivers/infiniband/core/umem_dmabuf.c   |  5 ++--
 drivers/infiniband/hw/hns/hns_roce_db.c |  4 ++--
 drivers/infiniband/hw/irdma/verbs.c     |  2 +-
 drivers/infiniband/hw/mlx4/doorbell.c   |  3 ++-
 drivers/infiniband/hw/mlx4/mr.c         |  4 ++--
 drivers/infiniband/hw/mlx5/doorbell.c   |  3 ++-
 drivers/infiniband/hw/mlx5/mr.c         |  3 ++-
 drivers/infiniband/hw/qedr/verbs.c      |  2 +-
 drivers/infiniband/sw/rdmavt/mr.c       |  2 +-
 drivers/infiniband/sw/rxe/rxe_mr.c      |  2 +-
 include/rdma/ib_umem.h                  | 10 ++++----
 include/rdma/ib_verbs.h                 | 28 ++++++++++++++++++++++
 13 files changed, 59 insertions(+), 41 deletions(-)

diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c
index 42481e7a72e8..86d479772fbc 100644
--- a/drivers/infiniband/core/umem.c
+++ b/drivers/infiniband/core/umem.c
@@ -51,11 +51,11 @@ static void __ib_umem_release(struct ib_device *dev, struct ib_umem *umem, int d
 	struct scatterlist *sg;
 	unsigned int i;
 
-	if (umem->nmap > 0)
-		ib_dma_unmap_sg(dev, umem->sg_head.sgl, umem->sg_nents,
-				DMA_BIDIRECTIONAL);
+	if (dirty)
+		ib_dma_unmap_sgtable_attrs(dev, &umem->sgt_append.sgt,
+					   DMA_BIDIRECTIONAL, 0);
 
-	for_each_sg(umem->sg_head.sgl, sg, umem->sg_nents, i)
+	for_each_sgtable_sg(&umem->sgt_append.sgt, sg, i)
 		unpin_user_page_range_dirty_lock(sg_page(sg),
 			DIV_ROUND_UP(sg->length, PAGE_SIZE), make_dirty);
 
@@ -111,7 +111,7 @@ unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem,
 	/* offset into first SGL */
 	pgoff = umem->address & ~PAGE_MASK;
 
-	for_each_sg(umem->sg_head.sgl, sg, umem->nmap, i) {
+	for_each_sgtable_dma_sg(&umem->sgt_append.sgt, sg, i) {
 		/* Walk SGL and reduce max page size if VA/PA bits differ
 		 * for any address.
 		 */
@@ -121,7 +121,7 @@ unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem,
 		 * the maximum possible page size as the low bits of the iova
 		 * must be zero when starting the next chunk.
 		 */
-		if (i != (umem->nmap - 1))
+		if (i != (umem->sgt_append.sgt.nents - 1))
 			mask |= va;
 		pgoff = 0;
 	}
@@ -231,30 +231,19 @@ struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr,
 			&umem->sgt_append, page_list, pinned, 0,
 			pinned << PAGE_SHIFT, ib_dma_max_seg_size(device),
 			npages, GFP_KERNEL);
-		umem->sg_nents = umem->sgt_append.sgt.nents;
 		if (ret) {
-			memcpy(&umem->sg_head.sgl, &umem->sgt_append.sgt,
-			       sizeof(umem->sgt_append.sgt));
 			unpin_user_pages_dirty_lock(page_list, pinned, 0);
 			goto umem_release;
 		}
 	}
 
-	memcpy(&umem->sg_head.sgl, &umem->sgt_append.sgt,
-	       sizeof(umem->sgt_append.sgt));
 	if (access & IB_ACCESS_RELAXED_ORDERING)
 		dma_attr |= DMA_ATTR_WEAK_ORDERING;
 
-	umem->nmap =
-		ib_dma_map_sg_attrs(device, umem->sg_head.sgl, umem->sg_nents,
-				    DMA_BIDIRECTIONAL, dma_attr);
-
-	if (!umem->nmap) {
-		ret = -ENOMEM;
+	ret = ib_dma_map_sgtable_attrs(device, &umem->sgt_append.sgt,
+				       DMA_BIDIRECTIONAL, dma_attr);
+	if (ret)
 		goto umem_release;
-	}
-
-	ret = 0;
 	goto out;
 
 umem_release:
@@ -314,7 +303,8 @@ int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offset,
 		return -EINVAL;
 	}
 
-	ret = sg_pcopy_to_buffer(umem->sg_head.sgl, umem->sg_nents, dst, length,
+	ret = sg_pcopy_to_buffer(umem->sgt_append.sgt.sgl,
+				 umem->sgt_append.sgt.orig_nents, dst, length,
 				 offset + ib_umem_offset(umem));
 
 	if (ret < 0)
diff --git a/drivers/infiniband/core/umem_dmabuf.c b/drivers/infiniband/core/umem_dmabuf.c
index c6e875619fac..e824baf4640d 100644
--- a/drivers/infiniband/core/umem_dmabuf.c
+++ b/drivers/infiniband/core/umem_dmabuf.c
@@ -55,9 +55,8 @@ int ib_umem_dmabuf_map_pages(struct ib_umem_dmabuf *umem_dmabuf)
 		cur += sg_dma_len(sg);
 	}
 
-	umem_dmabuf->umem.sg_head.sgl = umem_dmabuf->first_sg;
-	umem_dmabuf->umem.sg_head.nents = nmap;
-	umem_dmabuf->umem.nmap = nmap;
+	umem_dmabuf->umem.sgt_append.sgt.sgl = umem_dmabuf->first_sg;
+	umem_dmabuf->umem.sgt_append.sgt.nents = nmap;
 	umem_dmabuf->sgt = sgt;
 
 wait_fence:
diff --git a/drivers/infiniband/hw/hns/hns_roce_db.c b/drivers/infiniband/hw/hns/hns_roce_db.c
index d40ea3d87260..751470c7a2ce 100644
--- a/drivers/infiniband/hw/hns/hns_roce_db.c
+++ b/drivers/infiniband/hw/hns/hns_roce_db.c
@@ -42,8 +42,8 @@ int hns_roce_db_map_user(struct hns_roce_ucontext *context, unsigned long virt,
 
 found:
 	offset = virt - page_addr;
-	db->dma = sg_dma_address(page->umem->sg_head.sgl) + offset;
-	db->virt_addr = sg_virt(page->umem->sg_head.sgl) + offset;
+	db->dma = sg_dma_address(page->umem->sgt_append.sgt.sgl) + offset;
+	db->virt_addr = sg_virt(page->umem->sgt_append.sgt.sgl) + offset;
 	db->u.user_page = page;
 	refcount_inc(&page->refcount);
 
diff --git a/drivers/infiniband/hw/irdma/verbs.c b/drivers/infiniband/hw/irdma/verbs.c
index 9b8c451e2426..e8089da1b770 100644
--- a/drivers/infiniband/hw/irdma/verbs.c
+++ b/drivers/infiniband/hw/irdma/verbs.c
@@ -2231,7 +2231,7 @@ static void irdma_copy_user_pgaddrs(struct irdma_mr *iwmr, u64 *pbl,
 	pinfo = (level == PBLE_LEVEL_1) ? NULL : palloc->level2.leaf;
 
 	if (iwmr->type == IRDMA_MEMREG_TYPE_QP)
-		iwpbl->qp_mr.sq_page = sg_page(region->sg_head.sgl);
+		iwpbl->qp_mr.sq_page = sg_page(region->sgt_append.sgt.sgl);
 
 	rdma_umem_for_each_dma_block(region, &biter, iwmr->page_size) {
 		*pbl = rdma_block_iter_dma_address(&biter);
diff --git a/drivers/infiniband/hw/mlx4/doorbell.c b/drivers/infiniband/hw/mlx4/doorbell.c
index d41f03ccb0e1..9bbd695a9fd5 100644
--- a/drivers/infiniband/hw/mlx4/doorbell.c
+++ b/drivers/infiniband/hw/mlx4/doorbell.c
@@ -75,7 +75,8 @@ int mlx4_ib_db_map_user(struct ib_udata *udata, unsigned long virt,
 	list_add(&page->list, &context->db_page_list);
 
 found:
-	db->dma = sg_dma_address(page->umem->sg_head.sgl) + (virt & ~PAGE_MASK);
+	db->dma = sg_dma_address(page->umem->sgt_append.sgt.sgl) +
+		  (virt & ~PAGE_MASK);
 	db->u.user_page = page;
 	++page->refcnt;
 
diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c
index 50becc0e4b62..04a67b481608 100644
--- a/drivers/infiniband/hw/mlx4/mr.c
+++ b/drivers/infiniband/hw/mlx4/mr.c
@@ -200,7 +200,7 @@ int mlx4_ib_umem_write_mtt(struct mlx4_ib_dev *dev, struct mlx4_mtt *mtt,
 	mtt_shift = mtt->page_shift;
 	mtt_size = 1ULL << mtt_shift;
 
-	for_each_sg(umem->sg_head.sgl, sg, umem->nmap, i) {
+	for_each_sgtable_dma_sg(&umem->sgt_append.sgt, sg, i) {
 		if (cur_start_addr + len == sg_dma_address(sg)) {
 			/* still the same block */
 			len += sg_dma_len(sg);
@@ -273,7 +273,7 @@ int mlx4_ib_umem_calc_optimal_mtt_size(struct ib_umem *umem, u64 start_va,
 
 	*num_of_mtts = ib_umem_num_dma_blocks(umem, PAGE_SIZE);
 
-	for_each_sg(umem->sg_head.sgl, sg, umem->nmap, i) {
+	for_each_sgtable_dma_sg(&umem->sgt_append.sgt, sg, i) {
 		/*
 		 * Initialization - save the first chunk start as the
 		 * current_block_start - block means contiguous pages.
diff --git a/drivers/infiniband/hw/mlx5/doorbell.c b/drivers/infiniband/hw/mlx5/doorbell.c
index 9ca2e61807ec..6398e2f48579 100644
--- a/drivers/infiniband/hw/mlx5/doorbell.c
+++ b/drivers/infiniband/hw/mlx5/doorbell.c
@@ -78,7 +78,8 @@ int mlx5_ib_db_map_user(struct mlx5_ib_ucontext *context, unsigned long virt,
 	list_add(&page->list, &context->db_page_list);
 
 found:
-	db->dma = sg_dma_address(page->umem->sg_head.sgl) + (virt & ~PAGE_MASK);
+	db->dma = sg_dma_address(page->umem->sgt_append.sgt.sgl) +
+		  (virt & ~PAGE_MASK);
 	db->u.user_page = page;
 	++page->refcnt;
 
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 3263851ea574..d1eb49dbca15 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -1226,7 +1226,8 @@ int mlx5_ib_update_mr_pas(struct mlx5_ib_mr *mr, unsigned int flags)
 	orig_sg_length = sg.length;
 
 	cur_mtt = mtt;
-	rdma_for_each_block (mr->umem->sg_head.sgl, &biter, mr->umem->nmap,
+	rdma_for_each_block (mr->umem->sgt_append.sgt.sgl, &biter,
+			     mr->umem->sgt_append.sgt.nents,
 			     BIT(mr->page_shift)) {
 		if (cur_mtt == (void *)mtt + sg.length) {
 			dma_sync_single_for_device(ddev, sg.addr, sg.length,
diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c
index 16d4c0228d76..fc081c5629ee 100644
--- a/drivers/infiniband/hw/qedr/verbs.c
+++ b/drivers/infiniband/hw/qedr/verbs.c
@@ -1481,7 +1481,7 @@ static int qedr_init_srq_user_params(struct ib_udata *udata,
 		return PTR_ERR(srq->prod_umem);
 	}
 
-	sg = srq->prod_umem->sg_head.sgl;
+	sg = srq->prod_umem->sgt_append.sgt.sgl;
 	srq->hw_srq.phy_prod_pair_addr = sg_dma_address(sg);
 
 	return 0;
diff --git a/drivers/infiniband/sw/rdmavt/mr.c b/drivers/infiniband/sw/rdmavt/mr.c
index 34b7af6ab9c2..dfb99a56d952 100644
--- a/drivers/infiniband/sw/rdmavt/mr.c
+++ b/drivers/infiniband/sw/rdmavt/mr.c
@@ -410,7 +410,7 @@ struct ib_mr *rvt_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 	mr->mr.page_shift = PAGE_SHIFT;
 	m = 0;
 	n = 0;
-	for_each_sg_page (umem->sg_head.sgl, &sg_iter, umem->nmap, 0) {
+	for_each_sgtable_page (&umem->sgt_append.sgt, &sg_iter, 0) {
 		void *vaddr;
 
 		vaddr = page_address(sg_page_iter_page(&sg_iter));
diff --git a/drivers/infiniband/sw/rxe/rxe_mr.c b/drivers/infiniband/sw/rxe/rxe_mr.c
index 1ee5bd8291e5..16de1bc6db6e 100644
--- a/drivers/infiniband/sw/rxe/rxe_mr.c
+++ b/drivers/infiniband/sw/rxe/rxe_mr.c
@@ -141,7 +141,7 @@ int rxe_mr_init_user(struct rxe_pd *pd, u64 start, u64 length, u64 iova,
 	if (length > 0) {
 		buf = map[0]->buf;
 
-		for_each_sg_page(umem->sg_head.sgl, &sg_iter, umem->nmap, 0) {
+		for_each_sgtable_page (&umem->sgt_append.sgt, &sg_iter, 0) {
 			if (num_buf >= RXE_BUF_PER_MAP) {
 				map++;
 				buf = map[0]->buf;
diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h
index 33cb23b2ee3c..62f8e0604207 100644
--- a/include/rdma/ib_umem.h
+++ b/include/rdma/ib_umem.h
@@ -27,9 +27,6 @@ struct ib_umem {
 	u32 is_dmabuf : 1;
 	struct work_struct	work;
 	struct sg_append_table  sgt_append;
-	struct sg_table sg_head;
-	int             nmap;
-	unsigned int    sg_nents;
 };
 
 struct ib_umem_dmabuf {
@@ -57,7 +54,7 @@ static inline int ib_umem_offset(struct ib_umem *umem)
 static inline unsigned long ib_umem_dma_offset(struct ib_umem *umem,
 					       unsigned long pgsz)
 {
-	return (sg_dma_address(umem->sg_head.sgl) + ib_umem_offset(umem)) &
+	return (sg_dma_address(umem->sgt_append.sgt.sgl) + ib_umem_offset(umem)) &
 	       (pgsz - 1);
 }
 
@@ -78,7 +75,8 @@ static inline void __rdma_umem_block_iter_start(struct ib_block_iter *biter,
 						struct ib_umem *umem,
 						unsigned long pgsz)
 {
-	__rdma_block_iter_start(biter, umem->sg_head.sgl, umem->nmap, pgsz);
+	__rdma_block_iter_start(biter, umem->sgt_append.sgt.sgl,
+				umem->sgt_append.sgt.nents, pgsz);
 }
 
 /**
@@ -129,7 +127,7 @@ static inline unsigned long ib_umem_find_best_pgoff(struct ib_umem *umem,
 						    unsigned long pgsz_bitmap,
 						    u64 pgoff_bitmask)
 {
-	struct scatterlist *sg = umem->sg_head.sgl;
+	struct scatterlist *sg = umem->sgt_append.sgt.sgl;
 	dma_addr_t dma_addr;
 
 	dma_addr = sg_dma_address(sg) + (umem->address & ~PAGE_MASK);
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index aa7806335cba..4b50d9a3018a 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -4085,6 +4085,34 @@ static inline void ib_dma_unmap_sg_attrs(struct ib_device *dev,
 				   dma_attrs);
 }
 
+/**
+ * ib_dma_map_sgtable_attrs - Map a scatter/gather table to DMA addresses
+ * @dev: The device for which the DMA addresses are to be created
+ * @sg: The sg_table object describing the buffer
+ * @direction: The direction of the DMA
+ * @attrs: Optional DMA attributes for the map operation
+ */
+static inline int ib_dma_map_sgtable_attrs(struct ib_device *dev,
+					   struct sg_table *sgt,
+					   enum dma_data_direction direction,
+					   unsigned long dma_attrs)
+{
+	if (ib_uses_virt_dma(dev)) {
+		ib_dma_virt_map_sg(dev, sgt->sgl, sgt->orig_nents);
+		return 0;
+	}
+	return dma_map_sgtable(dev->dma_device, sgt, direction, dma_attrs);
+}
+
+static inline void ib_dma_unmap_sgtable_attrs(struct ib_device *dev,
+					      struct sg_table *sgt,
+					      enum dma_data_direction direction,
+					      unsigned long dma_attrs)
+{
+	if (!ib_uses_virt_dma(dev))
+		dma_unmap_sgtable(dev->dma_device, sgt, direction, dma_attrs);
+}
+
 /**
  * ib_dma_map_sg - Map a scatter/gather list to DMA addresses
  * @dev: The device for which the DMA addresses are to be created
-- 
2.25.4

Re: [PATCH rdma-next v4 2/3] lib/scatterlist: Fix wrong update of orig_nents

[Jason Gunthorpe]

On Tue, Aug 24, 2021 at 05:25:30PM +0300, Maor Gottlieb wrote:
> @@ -514,11 +531,13 @@ struct scatterlist *sg_alloc_append_table_from_pages(struct sg_table *sgt,
>  		offset = 0;
>  		cur_page = j;
>  	}
> -	sgt->nents += added_nents;
> +	sgt_append->sgt.nents += added_nents;
> +	sgt_append->sgt.orig_nents = sgt_append->sgt.nents;
> +	sgt_append->prv = s;

Why is nents being touched here?

Shouldn't it just be

    sgt_append->sgt.orig_nents += added_nents; 
    sgt_append->prv = s;

?

Let me know I can fix it

Jason

Re: [PATCH rdma-next v4 2/3] lib/scatterlist: Fix wrong update of orig_nents

[Maor Gottlieb]

On 8/24/2021 10:12 PM, Jason Gunthorpe wrote:
> On Tue, Aug 24, 2021 at 05:25:30PM +0300, Maor Gottlieb wrote:
>> @@ -514,11 +531,13 @@ struct scatterlist *sg_alloc_append_table_from_pages(struct sg_table *sgt,
>>   		offset = 0;
>>   		cur_page = j;
>>   	}
>> -	sgt->nents += added_nents;
>> +	sgt_append->sgt.nents += added_nents;
>> +	sgt_append->sgt.orig_nents = sgt_append->sgt.nents;
>> +	sgt_append->prv = s;
> Why is nents being touched here?
>
> Shouldn't it just be
>
>      sgt_append->sgt.orig_nents += added_nents;
>      sgt_append->prv = s;
>
> ?

In general yes, but all the sg_alloc_* functions update both orig_nents 
and nents (to same value). May be drivers that pass nents instead of 
orig_nents to dma_map_sg* (like umem did before this patch set). So I 
prefer to set it here as well, also not to break 
sg_alloc_table_from_pages_segment which use the same path. This kind of 
change could be done very carefully next cycle for all sg_alloc_*.

>
> Let me know I can fix it
>
> Jason

Re: [PATCH rdma-next v4 2/3] lib/scatterlist: Fix wrong update of orig_nents

[Jason Gunthorpe]

On Wed, Aug 25, 2021 at 07:59:27AM +0300, Maor Gottlieb wrote:
> 
> On 8/24/2021 10:12 PM, Jason Gunthorpe wrote:
> > On Tue, Aug 24, 2021 at 05:25:30PM +0300, Maor Gottlieb wrote:
> > > @@ -514,11 +531,13 @@ struct scatterlist *sg_alloc_append_table_from_pages(struct sg_table *sgt,
> > >   		offset = 0;
> > >   		cur_page = j;
> > >   	}
> > > -	sgt->nents += added_nents;
> > > +	sgt_append->sgt.nents += added_nents;
> > > +	sgt_append->sgt.orig_nents = sgt_append->sgt.nents;
> > > +	sgt_append->prv = s;
> > Why is nents being touched here?
> > 
> > Shouldn't it just be
> > 
> >      sgt_append->sgt.orig_nents += added_nents;
> >      sgt_append->prv = s;
> > 
> > ?
> 
> In general yes, but all the sg_alloc_* functions update both
> orig_nents and nents (to same value). May be drivers that pass nents
> instead of orig_nents to dma_map_sg* (like umem did before this
> patch set). So I prefer to set it here as well, also not to break
> sg_alloc_table_from_pages_segment which use the same path. This kind
> of change could be done very carefully next cycle for all
> sg_alloc_*.

Ok, interesting

Jason

Re: [PATCH rdma-next v4 0/3] SG fix together with update to RDMA umem

[Leon Romanovsky]

On Tue, Aug 24, 2021 at 05:25:28PM +0300, Maor Gottlieb wrote:
> From: Maor Gottlieb <maorg@nvidia.com>
> 
> Changelog:
> v4:
>  * Unify sg_free_table_entries with __sg_free_table
> v3: https://lore.kernel.org/lkml/cover.1627551226.git.leonro@nvidia.com/
>  * Rewrote to new API suggestion
>  * Split for more patches
> v2: https://lore.kernel.org/lkml/cover.1626605893.git.leonro@nvidia.com
>  * Changed implementation of first patch, based on our discussion with
>  * Christoph.
>    https://lore.kernel.org/lkml/YNwaVTT0qmQdxaZz@infradead.org/
> v1: https://lore.kernel.org/lkml/cover.1624955710.git.leonro@nvidia.com/
>  * Fixed sg_page with a _dma_ API in the umem.c
> v0: https://lore.kernel.org/lkml/cover.1624361199.git.leonro@nvidia.com
> 
> Maor Gottlieb (3):
>   lib/scatterlist: Provide a dedicated function to support table append
>   lib/scatterlist: Fix wrong update of orig_nents
>   RDMA: Use the sg_table directly and remove the opencoded version from
>     umem
> 
>  drivers/gpu/drm/drm_prime.c                 |  13 +-
>  drivers/gpu/drm/i915/gem/i915_gem_userptr.c |  11 +-
>  drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c  |  14 +-
>  drivers/infiniband/core/umem.c              |  56 ++++---
>  drivers/infiniband/core/umem_dmabuf.c       |   5 +-
>  drivers/infiniband/hw/hns/hns_roce_db.c     |   4 +-
>  drivers/infiniband/hw/irdma/verbs.c         |   2 +-
>  drivers/infiniband/hw/mlx4/doorbell.c       |   3 +-
>  drivers/infiniband/hw/mlx4/mr.c             |   4 +-
>  drivers/infiniband/hw/mlx5/doorbell.c       |   3 +-
>  drivers/infiniband/hw/mlx5/mr.c             |   3 +-
>  drivers/infiniband/hw/qedr/verbs.c          |   2 +-
>  drivers/infiniband/sw/rdmavt/mr.c           |   2 +-
>  drivers/infiniband/sw/rxe/rxe_mr.c          |   2 +-
>  include/linux/scatterlist.h                 |  56 +++++--
>  include/rdma/ib_umem.h                      |  11 +-
>  include/rdma/ib_verbs.h                     |  28 ++++
>  lib/scatterlist.c                           | 155 ++++++++++++--------
>  lib/sg_pool.c                               |   3 +-
>  tools/testing/scatterlist/main.c            |  38 +++--
>  20 files changed, 258 insertions(+), 157 deletions(-)

Jason,

Did you add these patches to the -next? I can't find them.

Thanks

> 
> -- 
> 2.25.4
> 

Re: [PATCH rdma-next v4 0/3] SG fix together with update to RDMA umem

[Jason Gunthorpe]

On Mon, Aug 30, 2021 at 11:21:00AM +0300, Leon Romanovsky wrote:
> On Tue, Aug 24, 2021 at 05:25:28PM +0300, Maor Gottlieb wrote:
> > From: Maor Gottlieb <maorg@nvidia.com>
> > 
> > Changelog:
> > v4:
> >  * Unify sg_free_table_entries with __sg_free_table
> > v3: https://lore.kernel.org/lkml/cover.1627551226.git.leonro@nvidia.com/
> >  * Rewrote to new API suggestion
> >  * Split for more patches
> > v2: https://lore.kernel.org/lkml/cover.1626605893.git.leonro@nvidia.com
> >  * Changed implementation of first patch, based on our discussion with
> >  * Christoph.
> >    https://lore.kernel.org/lkml/YNwaVTT0qmQdxaZz@infradead.org/
> > v1: https://lore.kernel.org/lkml/cover.1624955710.git.leonro@nvidia.com/
> >  * Fixed sg_page with a _dma_ API in the umem.c
> > v0: https://lore.kernel.org/lkml/cover.1624361199.git.leonro@nvidia.com
> > 
> > Maor Gottlieb (3):
> >   lib/scatterlist: Provide a dedicated function to support table append
> >   lib/scatterlist: Fix wrong update of orig_nents
> >   RDMA: Use the sg_table directly and remove the opencoded version from
> >     umem
> > 
> >  drivers/gpu/drm/drm_prime.c                 |  13 +-
> >  drivers/gpu/drm/i915/gem/i915_gem_userptr.c |  11 +-
> >  drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c  |  14 +-
> >  drivers/infiniband/core/umem.c              |  56 ++++---
> >  drivers/infiniband/core/umem_dmabuf.c       |   5 +-
> >  drivers/infiniband/hw/hns/hns_roce_db.c     |   4 +-
> >  drivers/infiniband/hw/irdma/verbs.c         |   2 +-
> >  drivers/infiniband/hw/mlx4/doorbell.c       |   3 +-
> >  drivers/infiniband/hw/mlx4/mr.c             |   4 +-
> >  drivers/infiniband/hw/mlx5/doorbell.c       |   3 +-
> >  drivers/infiniband/hw/mlx5/mr.c             |   3 +-
> >  drivers/infiniband/hw/qedr/verbs.c          |   2 +-
> >  drivers/infiniband/sw/rdmavt/mr.c           |   2 +-
> >  drivers/infiniband/sw/rxe/rxe_mr.c          |   2 +-
> >  include/linux/scatterlist.h                 |  56 +++++--
> >  include/rdma/ib_umem.h                      |  11 +-
> >  include/rdma/ib_verbs.h                     |  28 ++++
> >  lib/scatterlist.c                           | 155 ++++++++++++--------
> >  lib/sg_pool.c                               |   3 +-
> >  tools/testing/scatterlist/main.c            |  38 +++--
> >  20 files changed, 258 insertions(+), 157 deletions(-)
> 
> Jason,
> 
> Did you add these patches to the -next? I can't find them.

They sat in linux-next for awhile outside the rdma-next tree

All synced up now

Jason

Re: [PATCH rdma-next v4 0/3] SG fix together with update to RDMA umem

[Leon Romanovsky]

On Mon, Aug 30, 2021 at 10:31:28AM -0300, Jason Gunthorpe wrote:
> On Mon, Aug 30, 2021 at 11:21:00AM +0300, Leon Romanovsky wrote:
> > On Tue, Aug 24, 2021 at 05:25:28PM +0300, Maor Gottlieb wrote:
> > > From: Maor Gottlieb <maorg@nvidia.com>
> > > 
> > > Changelog:
> > > v4:
> > >  * Unify sg_free_table_entries with __sg_free_table
> > > v3: https://lore.kernel.org/lkml/cover.1627551226.git.leonro@nvidia.com/
> > >  * Rewrote to new API suggestion
> > >  * Split for more patches
> > > v2: https://lore.kernel.org/lkml/cover.1626605893.git.leonro@nvidia.com
> > >  * Changed implementation of first patch, based on our discussion with
> > >  * Christoph.
> > >    https://lore.kernel.org/lkml/YNwaVTT0qmQdxaZz@infradead.org/
> > > v1: https://lore.kernel.org/lkml/cover.1624955710.git.leonro@nvidia.com/
> > >  * Fixed sg_page with a _dma_ API in the umem.c
> > > v0: https://lore.kernel.org/lkml/cover.1624361199.git.leonro@nvidia.com
> > > 
> > > Maor Gottlieb (3):
> > >   lib/scatterlist: Provide a dedicated function to support table append
> > >   lib/scatterlist: Fix wrong update of orig_nents
> > >   RDMA: Use the sg_table directly and remove the opencoded version from
> > >     umem
> > > 
> > >  drivers/gpu/drm/drm_prime.c                 |  13 +-
> > >  drivers/gpu/drm/i915/gem/i915_gem_userptr.c |  11 +-
> > >  drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c  |  14 +-
> > >  drivers/infiniband/core/umem.c              |  56 ++++---
> > >  drivers/infiniband/core/umem_dmabuf.c       |   5 +-
> > >  drivers/infiniband/hw/hns/hns_roce_db.c     |   4 +-
> > >  drivers/infiniband/hw/irdma/verbs.c         |   2 +-
> > >  drivers/infiniband/hw/mlx4/doorbell.c       |   3 +-
> > >  drivers/infiniband/hw/mlx4/mr.c             |   4 +-
> > >  drivers/infiniband/hw/mlx5/doorbell.c       |   3 +-
> > >  drivers/infiniband/hw/mlx5/mr.c             |   3 +-
> > >  drivers/infiniband/hw/qedr/verbs.c          |   2 +-
> > >  drivers/infiniband/sw/rdmavt/mr.c           |   2 +-
> > >  drivers/infiniband/sw/rxe/rxe_mr.c          |   2 +-
> > >  include/linux/scatterlist.h                 |  56 +++++--
> > >  include/rdma/ib_umem.h                      |  11 +-
> > >  include/rdma/ib_verbs.h                     |  28 ++++
> > >  lib/scatterlist.c                           | 155 ++++++++++++--------
> > >  lib/sg_pool.c                               |   3 +-
> > >  tools/testing/scatterlist/main.c            |  38 +++--
> > >  20 files changed, 258 insertions(+), 157 deletions(-)
> > 
> > Jason,
> > 
> > Did you add these patches to the -next? I can't find them.
> 
> They sat in linux-next for awhile outside the rdma-next tree
> 
> All synced up now

Thanks

> 
> Jason