IB/uverbs: Export ib_umem_get()/ib_umem_release() to modules

Export ib_umem_get()/ib_umem_release() and put low-level drivers in
control of when to call ib_umem_get() to pin and DMA map userspace,
rather than always calling it in ib_uverbs_reg_mr() before calling the
low-level driver's reg_user_mr method.

Also move these functions to be in the ib_core module instead of
ib_uverbs, so that driver modules using them do not depend on
ib_uverbs.

This has a number of advantages:
 - It is better design from the standpoint of making generic code a
   library that can be used or overridden by device-specific code as
   the details of specific devices dictate.
 - Drivers that do not need to pin userspace memory regions do not
   need to take the performance hit of calling ib_mem_get().  For
   example, although I have not tried to implement it in this patch,
   the ipath driver should be able to avoid pinning memory and just
   use copy_{to,from}_user() to access userspace memory regions.
 - Buffers that need special mapping treatment can be identified by
   the low-level driver.  For example, it may be possible to solve
   some Altix-specific memory ordering issues with mthca CQs in
   userspace by mapping CQ buffers with extra flags.
 - Drivers that need to pin and DMA map userspace memory for things
   other than memory regions can use ib_umem_get() directly, instead
   of hacks using extra parameters to their reg_phys_mr method.  For
   example, the mlx4 driver that is pending being merged needs to pin
   and DMA map QP and CQ buffers, but it does not need to create a
   memory key for these buffers.  So the cleanest solution is for mlx4
   to call ib_umem_get() in the create_qp and create_cq methods.

Signed-off-by: Roland Dreier <rolandd@cisco.com>

1 files changed, 27 insertions, 11 deletions

diff --git a/drivers/infiniband/hw/ipath/ipath_mr.c b/drivers/infiniband/hw/ipath/ipath_mr.c
index 31e70732e369..bdeef8d4f279 100644
--- a/drivers/infiniband/hw/ipath/ipath_mr.c
+++ b/drivers/infiniband/hw/ipath/ipath_mr.c
@@ -31,6 +31,7 @@
  * SOFTWARE.
  */
 
+#include <rdma/ib_umem.h>
 #include <rdma/ib_pack.h>
 #include <rdma/ib_smi.h>
 
@@ -147,6 +148,7 @@ struct ib_mr *ipath_reg_phys_mr(struct ib_pd *pd,
         mr->mr.offset = 0;
         mr->mr.access_flags = acc;
         mr->mr.max_segs = num_phys_buf;
+        mr->umem = NULL;
 
         m = 0;
         n = 0;
@@ -170,46 +172,56 @@ bail:
 /**
  * ipath_reg_user_mr - register a userspace memory region
  * @pd: protection domain for this memory region
- * @region: the user memory region
+ * @start: starting userspace address
+ * @length: length of region to register
+ * @virt_addr: virtual address to use (from HCA's point of view)
  * @mr_access_flags: access flags for this memory region
  * @udata: unused by the InfiniPath driver
  *
  * Returns the memory region on success, otherwise returns an errno.
  */
-struct ib_mr *ipath_reg_user_mr(struct ib_pd *pd, struct ib_umem *region,
-                                int mr_access_flags, struct ib_udata *udata)
+struct ib_mr *ipath_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
+                                u64 virt_addr, int mr_access_flags,
+                                struct ib_udata *udata)
 {
         struct ipath_mr *mr;
+        struct ib_umem *umem;
         struct ib_umem_chunk *chunk;
         int n, m, i;
         struct ib_mr *ret;
 
-        if (region->length == 0) {
+        if (length == 0) {
                 ret = ERR_PTR(-EINVAL);
                 goto bail;
         }
 
+        umem = ib_umem_get(pd->uobject->context, start, length, mr_access_flags);
+        if (IS_ERR(umem))
+                return (void *) umem;
+
         n = 0;
-        list_for_each_entry(chunk, &region->chunk_list, list)
+        list_for_each_entry(chunk, &umem->chunk_list, list)
                 n += chunk->nents;
 
         mr = alloc_mr(n, &to_idev(pd->device)->lk_table);
         if (!mr) {
                 ret = ERR_PTR(-ENOMEM);
+                ib_umem_release(umem);
                 goto bail;
         }
 
         mr->mr.pd = pd;
-        mr->mr.user_base = region->user_base;
-        mr->mr.iova = region->virt_base;
-        mr->mr.length = region->length;
-        mr->mr.offset = region->offset;
+        mr->mr.user_base = start;
+        mr->mr.iova = virt_addr;
+        mr->mr.length = length;
+        mr->mr.offset = umem->offset;
         mr->mr.access_flags = mr_access_flags;
         mr->mr.max_segs = n;
+        mr->umem = umem;
 
         m = 0;
         n = 0;
-        list_for_each_entry(chunk, &region->chunk_list, list) {
+        list_for_each_entry(chunk, &umem->chunk_list, list) {
                 for (i = 0; i < chunk->nents; i++) {
                         void *vaddr;
 
@@ -219,7 +231,7 @@ struct ib_mr *ipath_reg_user_mr(struct ib_pd *pd, struct ib_umem *region,
                                 goto bail;
                         }
                         mr->mr.map[m]->segs[n].vaddr = vaddr;
-                        mr->mr.map[m]->segs[n].length = region->page_size;
+                        mr->mr.map[m]->segs[n].length = umem->page_size;
                         n++;
                         if (n == IPATH_SEGSZ) {
                                 m++;
@@ -253,6 +265,10 @@ int ipath_dereg_mr(struct ib_mr *ibmr)
                 i--;
                 kfree(mr->mr.map[i]);
         }
+
+        if (mr->umem)
+                ib_umem_release(mr->umem);
+
         kfree(mr);
         return 0;
 }