IB/core: Add "type 2" memory windows support

This patch enhances the IB core support for Memory Windows (MWs).

MWs allow an application to have better/flexible control over remote
access to memory.

Two types of MWs are supported, with the second type having two flavors:

    Type 1  - associated with PD only
    Type 2A - associated with QPN only
    Type 2B - associated with PD and QPN

Applications can allocate a MW once, and then repeatedly bind the MW
to different ranges in MRs that are associated to the same PD. Type 1
windows are bound through a verb, while type 2 windows are bound by
posting a work request.

The 32-bit memory key is composed of a 24-bit index and an 8-bit
key. The key is changed with each bind, thus allowing more control
over the peer's use of the memory key.

The changes introduced are the following:

* add memory window type enum and a corresponding parameter to ib_alloc_mw.
* type 2 memory window bind work request support.
* create a struct that contains the common part of the bind verb struct
  ibv_mw_bind and the bind work request into a single struct.
* add the ib_inc_rkey helper function to advance the tag part of an rkey.

Consumer interface details:

* new device capability flags IB_DEVICE_MEM_WINDOW_TYPE_2A and
  IB_DEVICE_MEM_WINDOW_TYPE_2B are added to indicate device support
  for these features.

  Devices can set either IB_DEVICE_MEM_WINDOW_TYPE_2A or
  IB_DEVICE_MEM_WINDOW_TYPE_2B if it supports type 2A or type 2B
  memory windows. It can set neither to indicate it doesn't support
  type 2 windows at all.

* modify existing provides and consumers code to the new param of
  ib_alloc_mw and the ib_mw_bind_info structure

Signed-off-by: Haggai Eran <haggaie@mellanox.com>
Signed-off-by: Shani Michaeli <shanim@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>

1 files changed, 64 insertions, 9 deletions

diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 46bc045bbe15..98cc4b29fc5b 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -115,6 +115,8 @@ enum ib_device_cap_flags {
         IB_DEVICE_XRC                   = (1<<20),
         IB_DEVICE_MEM_MGT_EXTENSIONS    = (1<<21),
         IB_DEVICE_BLOCK_MULTICAST_LOOPBACK = (1<<22),
+        IB_DEVICE_MEM_WINDOW_TYPE_2A    = (1<<23),
+        IB_DEVICE_MEM_WINDOW_TYPE_2B    = (1<<24)
 };
 
 enum ib_atomic_cap {
@@ -715,6 +717,11 @@ enum ib_mig_state {
         IB_MIG_ARMED
 };
 
+enum ib_mw_type {
+        IB_MW_TYPE_1 = 1,
+        IB_MW_TYPE_2 = 2
+};
+
 struct ib_qp_attr {
         enum ib_qp_state        qp_state;
         enum ib_qp_state        cur_qp_state;
@@ -758,6 +765,7 @@ enum ib_wr_opcode {
         IB_WR_FAST_REG_MR,
         IB_WR_MASKED_ATOMIC_CMP_AND_SWP,
         IB_WR_MASKED_ATOMIC_FETCH_AND_ADD,
+        IB_WR_BIND_MW,
 };
 
 enum ib_send_flags {
@@ -780,6 +788,23 @@ struct ib_fast_reg_page_list {
         unsigned int            max_page_list_len;
 };
 
+/**
+ * struct ib_mw_bind_info - Parameters for a memory window bind operation.
+ * @mr: A memory region to bind the memory window to.
+ * @addr: The address where the memory window should begin.
+ * @length: The length of the memory window, in bytes.
+ * @mw_access_flags: Access flags from enum ib_access_flags for the window.
+ *
+ * This struct contains the shared parameters for type 1 and type 2
+ * memory window bind operations.
+ */
+struct ib_mw_bind_info {
+        struct ib_mr   *mr;
+        u64             addr;
+        u64             length;
+        int             mw_access_flags;
+};
+
 struct ib_send_wr {
         struct ib_send_wr      *next;
         u64                     wr_id;
@@ -823,6 +848,12 @@ struct ib_send_wr {
                         int                             access_flags;
                         u32                             rkey;
                 } fast_reg;
+                struct {
+                        struct ib_mw            *mw;
+                        /* The new rkey for the memory window. */
+                        u32                      rkey;
+                        struct ib_mw_bind_info   bind_info;
+                } bind_mw;
         } wr;
         u32                     xrc_remote_srq_num;     /* XRC TGT QPs only */
 };
@@ -839,7 +870,8 @@ enum ib_access_flags {
         IB_ACCESS_REMOTE_WRITE  = (1<<1),
         IB_ACCESS_REMOTE_READ   = (1<<2),
         IB_ACCESS_REMOTE_ATOMIC = (1<<3),
-        IB_ACCESS_MW_BIND       = (1<<4)
+        IB_ACCESS_MW_BIND       = (1<<4),
+        IB_ZERO_BASED           = (1<<5)
 };
 
 struct ib_phys_buf {
@@ -862,13 +894,16 @@ enum ib_mr_rereg_flags {
         IB_MR_REREG_ACCESS      = (1<<2)
 };
 
+/**
+ * struct ib_mw_bind - Parameters for a type 1 memory window bind operation.
+ * @wr_id:      Work request id.
+ * @send_flags: Flags from ib_send_flags enum.
+ * @bind_info:  More parameters of the bind operation.
+ */
 struct ib_mw_bind {
-        struct ib_mr   *mr;
-        u64             wr_id;
-        u64             addr;
-        u32             length;
-        int             send_flags;
-        int             mw_access_flags;
+        u64                    wr_id;
+        int                    send_flags;
+        struct ib_mw_bind_info bind_info;
 };
 
 struct ib_fmr_attr {
@@ -991,6 +1026,7 @@ struct ib_mw {
         struct ib_pd            *pd;
         struct ib_uobject       *uobject;
         u32                     rkey;
+        enum ib_mw_type         type;
 };
 
 struct ib_fmr {
@@ -1202,7 +1238,8 @@ struct ib_device {
                                                     int num_phys_buf,
                                                     int mr_access_flags,
                                                     u64 *iova_start);
-        struct ib_mw *             (*alloc_mw)(struct ib_pd *pd);
+        struct ib_mw *             (*alloc_mw)(struct ib_pd *pd,
+                                               enum ib_mw_type type);
         int                        (*bind_mw)(struct ib_qp *qp,
                                               struct ib_mw *mw,
                                               struct ib_mw_bind *mw_bind);
@@ -2019,6 +2056,8 @@ int ib_query_mr(struct ib_mr *mr, struct ib_mr_attr *mr_attr);
  * ib_dereg_mr - Deregisters a memory region and removes it from the
  *   HCA translation table.
  * @mr: The memory region to deregister.
+ *
+ * This function can fail, if the memory region has memory windows bound to it.
  */
 int ib_dereg_mr(struct ib_mr *mr);
 
@@ -2071,10 +2110,22 @@ static inline void ib_update_fast_reg_key(struct ib_mr *mr, u8 newkey)
 }
 
 /**
+ * ib_inc_rkey - increments the key portion of the given rkey. Can be used
+ * for calculating a new rkey for type 2 memory windows.
+ * @rkey - the rkey to increment.
+ */
+static inline u32 ib_inc_rkey(u32 rkey)
+{
+        const u32 mask = 0x000000ff;
+        return ((rkey + 1) & mask) | (rkey & ~mask);
+}
+
+/**
  * ib_alloc_mw - Allocates a memory window.
  * @pd: The protection domain associated with the memory window.
+ * @type: The type of the memory window (1 or 2).
  */
-struct ib_mw *ib_alloc_mw(struct ib_pd *pd);
+struct ib_mw *ib_alloc_mw(struct ib_pd *pd, enum ib_mw_type type);
 
 /**
  * ib_bind_mw - Posts a work request to the send queue of the specified
@@ -2084,6 +2135,10 @@ struct ib_mw *ib_alloc_mw(struct ib_pd *pd);
  * @mw: The memory window to bind.
  * @mw_bind: Specifies information about the memory window, including
  *   its address range, remote access rights, and associated memory region.
+ *
+ * If there is no immediate error, the function will update the rkey member
+ * of the mw parameter to its new value. The bind operation can still fail
+ * asynchronously.
  */
 static inline int ib_bind_mw(struct ib_qp *qp,
                              struct ib_mw *mw,