From: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
To: qemu-devel@nongnu.org, peter.maydell@linaro.org
Cc: jusual@redhat.com, stefanha@redhat.com, yuval.shaia.ml@gmail.com,
quintela@redhat.com
Subject: [Qemu-devel] [PATCH PULL 3/4] hw/rdma: Skip data-path mr_id translation
Date: Sat, 21 Mar 2020 21:13:10 +0200 [thread overview]
Message-ID: <20200321191311.31537-4-marcel.apfelbaum@gmail.com> (raw)
In-Reply-To: <20200321191311.31537-1-marcel.apfelbaum@gmail.com>
From: Yuval Shaia <yuval.shaia.ml@gmail.com>
With the change made in commit 68b89aee71 ("Utilize ibv_reg_mr_iova for
memory registration") the MR emulation is no longer needed in order to
translate the guest addresses into host addresses.
With that, the next obvious step is to skip entirely the processing in
data-path.
To accomplish this, return the backend's lkey to driver so we will not
need to do the emulated mr_id to backend mr_id translation in data-path.
The function build_host_sge_array is still called in data-path but only
for backward computability with statistics collection.
While there, as a cosmetic change to make the code cleaner - make one
copy of the function rdma_backend_create_mr and leave the redundant
guest_start argument in the legacy code.
Signed-off-by: Yuval Shaia <yuval.shaia.ml@gmail.com>
Reviewed-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Message-Id: <20200320143429.9490-3-yuval.shaia.ml@gmail.com>
Signed-off-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
---
hw/rdma/rdma_backend.c | 21 ++++++++++++++-------
hw/rdma/rdma_backend.h | 5 -----
hw/rdma/rdma_rm.c | 13 ++++++-------
3 files changed, 20 insertions(+), 19 deletions(-)
diff --git a/hw/rdma/rdma_backend.c b/hw/rdma/rdma_backend.c
index b7ffbef9c0..3dd39fe1a7 100644
--- a/hw/rdma/rdma_backend.c
+++ b/hw/rdma/rdma_backend.c
@@ -377,6 +377,7 @@ static void ah_cache_init(void)
destroy_ah_hash_key, destroy_ah_hast_data);
}
+#ifdef LEGACY_RDMA_REG_MR
static int build_host_sge_array(RdmaDeviceResources *rdma_dev_res,
struct ibv_sge *sge, uint8_t num_sge,
uint64_t *total_length)
@@ -391,9 +392,7 @@ static int build_host_sge_array(RdmaDeviceResources *rdma_dev_res,
return VENDOR_ERR_INVLKEY | sge[idx].lkey;
}
-#ifdef LEGACY_RDMA_REG_MR
sge[idx].addr = (uintptr_t)mr->virt + sge[idx].addr - mr->start;
-#endif
sge[idx].lkey = rdma_backend_mr_lkey(&mr->backend_mr);
*total_length += sge[idx].length;
@@ -401,6 +400,19 @@ static int build_host_sge_array(RdmaDeviceResources *rdma_dev_res,
return 0;
}
+#else
+static inline int build_host_sge_array(RdmaDeviceResources *rdma_dev_res,
+ struct ibv_sge *sge, uint8_t num_sge,
+ uint64_t *total_length)
+{
+ int idx;
+
+ for (idx = 0; idx < num_sge; idx++) {
+ *total_length += sge[idx].length;
+ }
+ return 0;
+}
+#endif
static void trace_mad_message(const char *title, char *buf, int len)
{
@@ -731,13 +743,8 @@ void rdma_backend_destroy_pd(RdmaBackendPD *pd)
}
}
-#ifdef LEGACY_RDMA_REG_MR
-int rdma_backend_create_mr(RdmaBackendMR *mr, RdmaBackendPD *pd, void *addr,
- size_t length, int access)
-#else
int rdma_backend_create_mr(RdmaBackendMR *mr, RdmaBackendPD *pd, void *addr,
size_t length, uint64_t guest_start, int access)
-#endif
{
#ifdef LEGACY_RDMA_REG_MR
mr->ibmr = ibv_reg_mr(pd->ibpd, addr, length, access);
diff --git a/hw/rdma/rdma_backend.h b/hw/rdma/rdma_backend.h
index 127f96e2d5..225af481e0 100644
--- a/hw/rdma/rdma_backend.h
+++ b/hw/rdma/rdma_backend.h
@@ -78,13 +78,8 @@ int rdma_backend_query_port(RdmaBackendDev *backend_dev,
int rdma_backend_create_pd(RdmaBackendDev *backend_dev, RdmaBackendPD *pd);
void rdma_backend_destroy_pd(RdmaBackendPD *pd);
-#ifdef LEGACY_RDMA_REG_MR
-int rdma_backend_create_mr(RdmaBackendMR *mr, RdmaBackendPD *pd, void *addr,
- size_t length, int access);
-#else
int rdma_backend_create_mr(RdmaBackendMR *mr, RdmaBackendPD *pd, void *addr,
size_t length, uint64_t guest_start, int access);
-#endif
void rdma_backend_destroy_mr(RdmaBackendMR *mr);
int rdma_backend_create_cq(RdmaBackendDev *backend_dev, RdmaBackendCQ *cq,
diff --git a/hw/rdma/rdma_rm.c b/hw/rdma/rdma_rm.c
index 1524dfaeaa..7e9ea283c9 100644
--- a/hw/rdma/rdma_rm.c
+++ b/hw/rdma/rdma_rm.c
@@ -227,21 +227,20 @@ int rdma_rm_alloc_mr(RdmaDeviceResources *dev_res, uint32_t pd_handle,
mr->length = guest_length;
mr->virt += (mr->start & (TARGET_PAGE_SIZE - 1));
-#ifdef LEGACY_RDMA_REG_MR
- ret = rdma_backend_create_mr(&mr->backend_mr, &pd->backend_pd, mr->virt,
- mr->length, access_flags);
-#else
ret = rdma_backend_create_mr(&mr->backend_mr, &pd->backend_pd, mr->virt,
mr->length, guest_start, access_flags);
-#endif
if (ret) {
ret = -EIO;
goto out_dealloc_mr;
}
+#ifdef LEGACY_RDMA_REG_MR
+ /* We keep mr_handle in lkey so send and recv get get mr ptr */
+ *lkey = *mr_handle;
+#else
+ *lkey = rdma_backend_mr_lkey(&mr->backend_mr);
+#endif
}
- /* We keep mr_handle in lkey so send and recv get get mr ptr */
- *lkey = *mr_handle;
*rkey = -1;
mr->pd_handle = pd_handle;
--
2.17.2
next prev parent reply other threads:[~2020-03-21 19:14 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-21 19:13 [Qemu-devel] [PATCH PULL 0/4] RDMA queue Marcel Apfelbaum
2020-03-21 19:13 ` [Qemu-devel] [PATCH PULL 1/4] hw/rdma/vmw/pvrdma_dev_ring: Replace strncpy with pstrcpy Marcel Apfelbaum
2020-03-21 19:13 ` [Qemu-devel] [PATCH PULL 2/4] hw/rdma: Cosmetic change - no need for two sge arrays Marcel Apfelbaum
2020-03-21 19:13 ` Marcel Apfelbaum [this message]
2020-03-21 19:13 ` [Qemu-devel] [PATCH PULL 4/4] hw/rdma: avoid suspicious strncpy() use Marcel Apfelbaum
2020-03-21 22:34 ` [Qemu-devel] [PATCH PULL 0/4] RDMA queue Peter Maydell
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=20200321191311.31537-4-marcel.apfelbaum@gmail.com \
--to=marcel.apfelbaum@gmail.com \
--cc=jusual@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=quintela@redhat.com \
--cc=stefanha@redhat.com \
--cc=yuval.shaia.ml@gmail.com \
/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).