patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Xueming Li <xuemingl@nvidia.com>
To: Maxime Coquelin <maxime.coquelin@redhat.com>
Cc: <xuemingl@nvidia.com>, Yu Jiang <yux.jiang@intel.com>,
	David Marchand <david.marchand@redhat.com>,
	dpdk stable <stable@dpdk.org>
Subject: patch 'vhost: restrict set max queue pair API to VDUSE' has been queued to stable release 23.11.3
Date: Mon, 11 Nov 2024 14:27:37 +0800	[thread overview]
Message-ID: <20241111062847.216344-52-xuemingl@nvidia.com> (raw)
In-Reply-To: <20241111062847.216344-1-xuemingl@nvidia.com>

Hi,

FYI, your patch has been queued to stable release 23.11.3

Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
It will be pushed if I get no objections before 11/30/24. So please
shout if anyone has objections.

Also note that after the patch there's a diff of the upstream commit vs the
patch applied to the branch. This will indicate if there was any rebasing
needed to apply to the stable branch. If there were code changes for rebasing
(ie: not only metadata diffs), please double check that the rebase was
correctly done.

Queued patches are on a temporary branch at:
https://git.dpdk.org/dpdk-stable/log/?h=23.11-staging

This queued commit can be viewed at:
https://git.dpdk.org/dpdk-stable/commit/?h=23.11-staging&id=e1bd966815e01d4bbac59b413ec35e5ebd0a416c

Thanks.

Xueming Li <xuemingl@nvidia.com>

---
From e1bd966815e01d4bbac59b413ec35e5ebd0a416c Mon Sep 17 00:00:00 2001
From: Maxime Coquelin <maxime.coquelin@redhat.com>
Date: Thu, 3 Oct 2024 10:11:10 +0200
Subject: [PATCH] vhost: restrict set max queue pair API to VDUSE
Cc: Xueming Li <xuemingl@nvidia.com>

[ upstream commit e1808999d36bb2e136a649f4651f36030aa468f1 ]

In order to avoid breaking Vhost-user live-migration, we want the
rte_vhost_driver_set_max_queue_num API to only be effective with
VDUSE.

Furthermore, this API is only really needed for VDUSE where the
device number of queues is defined by the backend. For Vhost-user,
this is defined by the frontend (e.g. QEMU), so the advantages of
restricting more the maximum number of queue pairs is limited to
a small memory gain (a handful of pointers).

Fixes: 4aa1f88ac13d ("vhost: add API to set max queue pairs")

Reported-by: Yu Jiang <yux.jiang@intel.com>
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: David Marchand <david.marchand@redhat.com>
---
 lib/vhost/rte_vhost.h |  2 ++
 lib/vhost/socket.c    | 11 +++++++++++
 2 files changed, 13 insertions(+)

diff --git a/lib/vhost/rte_vhost.h b/lib/vhost/rte_vhost.h
index db92f05344..c6dba67a67 100644
--- a/lib/vhost/rte_vhost.h
+++ b/lib/vhost/rte_vhost.h
@@ -613,6 +613,8 @@ rte_vhost_driver_get_queue_num(const char *path, uint32_t *queue_num);
  * @b EXPERIMENTAL: this API may change, or be removed, without prior notice.
  *
  * Set the maximum number of queue pairs supported by the device.
+ * The value set is ignored for Vhost-user backends. It is only taken into
+ * account with VDUSE backends.
  *
  * @param path
  *  The vhost-user socket file path
diff --git a/lib/vhost/socket.c b/lib/vhost/socket.c
index 0b95c54c5b..ffb8518e74 100644
--- a/lib/vhost/socket.c
+++ b/lib/vhost/socket.c
@@ -865,6 +865,17 @@ rte_vhost_driver_set_max_queue_num(const char *path, uint32_t max_queue_pairs)
 		goto unlock_exit;
 	}

+	/*
+	 * This is only useful for VDUSE for which number of virtqueues is set
+	 * by the backend. For Vhost-user, the number of virtqueues is defined
+	 * by the frontend.
+	 */
+	if (!vsocket->is_vduse) {
+		VHOST_LOG_CONFIG(path, DEBUG, "Keeping %u max queue pairs for Vhost-user backend\n",
+				 VHOST_MAX_QUEUE_PAIRS);
+		goto unlock_exit;
+	}
+
 	vsocket->max_queue_pairs = max_queue_pairs;

 unlock_exit:
--
2.34.1

---
  Diff of the applied patch vs upstream commit (please double-check if non-empty:
---
--- -	2024-11-11 14:23:07.569998190 +0800
+++ 0051-vhost-restrict-set-max-queue-pair-API-to-VDUSE.patch	2024-11-11 14:23:05.122192840 +0800
@@ -1 +1 @@
-From e1808999d36bb2e136a649f4651f36030aa468f1 Mon Sep 17 00:00:00 2001
+From e1bd966815e01d4bbac59b413ec35e5ebd0a416c Mon Sep 17 00:00:00 2001
@@ -4,0 +5,3 @@
+Cc: Xueming Li <xuemingl@nvidia.com>
+
+[ upstream commit e1808999d36bb2e136a649f4651f36030aa468f1 ]
@@ -17 +19,0 @@
-Cc: stable@dpdk.org
@@ -24,2 +26,2 @@
- lib/vhost/socket.c    | 12 ++++++++++++
- 2 files changed, 14 insertions(+)
+ lib/vhost/socket.c    | 11 +++++++++++
+ 2 files changed, 13 insertions(+)
@@ -28 +30 @@
-index c7a5f56df8..1a91a00f02 100644
+index db92f05344..c6dba67a67 100644
@@ -31 +33 @@
-@@ -614,6 +614,8 @@ rte_vhost_driver_get_queue_num(const char *path, uint32_t *queue_num);
+@@ -613,6 +613,8 @@ rte_vhost_driver_get_queue_num(const char *path, uint32_t *queue_num);
@@ -41 +43 @@
-index a75728a2e4..d29d15494c 100644
+index 0b95c54c5b..ffb8518e74 100644
@@ -44 +46 @@
-@@ -860,6 +860,18 @@ rte_vhost_driver_set_max_queue_num(const char *path, uint32_t max_queue_pairs)
+@@ -865,6 +865,17 @@ rte_vhost_driver_set_max_queue_num(const char *path, uint32_t max_queue_pairs)
@@ -54,3 +56,2 @@
-+		VHOST_CONFIG_LOG(path, DEBUG,
-+				"Keeping %u max queue pairs for Vhost-user backend",
-+				VHOST_MAX_QUEUE_PAIRS);
++		VHOST_LOG_CONFIG(path, DEBUG, "Keeping %u max queue pairs for Vhost-user backend\n",
++				 VHOST_MAX_QUEUE_PAIRS);

  parent reply	other threads:[~2024-11-11  6:34 UTC|newest]

Thread overview: 128+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-11-11  6:26 patch " Xueming Li
2024-11-11  6:26 ` patch 'bus/vdev: revert fix devargs in secondary process' " Xueming Li
2024-11-11  6:26 ` patch 'log: add a per line log helper' " Xueming Li
2024-11-12  9:02   ` David Marchand
2024-11-12 11:35     ` Xueming Li
2024-11-12 12:47       ` David Marchand
2024-11-12 13:56         ` Xueming Li
2024-11-12 14:09           ` David Marchand
2024-11-12 14:11             ` Xueming Li
2024-11-11  6:26 ` patch 'drivers: remove redundant newline from logs' " Xueming Li
2024-11-11  6:26 ` patch 'eal/x86: fix 32-bit write combining store' " Xueming Li
2024-11-11  6:26 ` patch 'test/event: fix schedule type' " Xueming Li
2024-11-11  6:26 ` patch 'test/event: fix target event queue' " Xueming Li
2024-11-11  6:26 ` patch 'examples/eventdev: fix queue crash with generic pipeline' " Xueming Li
2024-11-11  6:26 ` patch 'crypto/dpaa2_sec: fix memory leak' " Xueming Li
2024-11-11  6:26 ` patch 'common/dpaax/caamflib: fix PDCP SNOW-ZUC watchdog' " Xueming Li
2024-11-11  6:26 ` patch 'dev: fix callback lookup when unregistering device' " Xueming Li
2024-11-11  6:26 ` patch 'crypto/scheduler: fix session size computation' " Xueming Li
2024-11-11  6:26 ` patch 'examples/ipsec-secgw: fix dequeue count from cryptodev' " Xueming Li
2024-11-11  6:26 ` patch 'bpf: fix free function mismatch if convert fails' " Xueming Li
2024-11-11  6:27 ` patch 'baseband/la12xx: fix use after free in modem config' " Xueming Li
2024-11-11  6:27 ` patch 'common/qat: fix use after free in device probe' " Xueming Li
2024-11-11  6:27 ` patch 'common/idpf: fix use after free in mailbox init' " Xueming Li
2024-11-11  6:27 ` patch 'crypto/bcmfs: fix free function mismatch' " Xueming Li
2024-11-11  6:27 ` patch 'dma/idxd: fix free function mismatch in device probe' " Xueming Li
2024-11-11  6:27 ` patch 'event/cnxk: fix free function mismatch in port config' " Xueming Li
2024-11-11  6:27 ` patch 'net/cnxk: fix use after free in mempool create' " Xueming Li
2024-11-11  6:27 ` patch 'net/cpfl: fix invalid free in JSON parser' " Xueming Li
2024-11-11  6:27 ` patch 'net/e1000: fix use after free in filter flush' " Xueming Li
2024-11-11  6:27 ` patch 'net/nfp: fix double free in flow destroy' " Xueming Li
2024-11-11  6:27 ` patch 'net/sfc: fix use after free in debug logs' " Xueming Li
2024-11-11  6:27 ` patch 'raw/ifpga/base: fix use after free' " Xueming Li
2024-11-11  6:27 ` patch 'raw/ifpga: fix free function mismatch in interrupt config' " Xueming Li
2024-11-11  6:27 ` patch 'examples/vhost: fix free function mismatch' " Xueming Li
2024-11-11  6:27 ` patch 'net/nfb: fix use after free' " Xueming Li
2024-11-11  6:27 ` patch 'power: enable CPPC' " Xueming Li
2024-11-11  6:27 ` patch 'fib6: add runtime checks in AVX512 lookup' " Xueming Li
2024-11-11  6:27 ` patch 'pcapng: fix handling of chained mbufs' " Xueming Li
2024-11-11  6:27 ` patch 'app/dumpcap: fix handling of jumbo frames' " Xueming Li
2024-11-11  6:27 ` patch 'ml/cnxk: fix handling of TVM model I/O' " Xueming Li
2024-11-11  6:27 ` patch 'net/cnxk: fix Rx timestamp handling for VF' " Xueming Li
2024-11-11  6:27 ` patch 'net/cnxk: fix Rx offloads to handle timestamp' " Xueming Li
2024-11-11  6:27 ` patch 'event/cnxk: fix Rx timestamp handling' " Xueming Li
2024-11-11  6:27 ` patch 'common/cnxk: fix MAC address change with active VF' " Xueming Li
2024-11-11  6:27 ` patch 'common/cnxk: fix inline CTX write' " Xueming Li
2024-11-11  6:27 ` patch 'common/cnxk: fix CPT HW word size for outbound SA' " Xueming Li
2024-11-11  6:27 ` patch 'net/cnxk: fix OOP handling for inbound packets' " Xueming Li
2024-11-11  6:27 ` patch 'event/cnxk: fix OOP handling in event mode' " Xueming Li
2024-11-11  6:27 ` patch 'common/cnxk: fix base log level' " Xueming Li
2024-11-11  6:27 ` patch 'common/cnxk: fix IRQ reconfiguration' " Xueming Li
2024-11-11  6:27 ` patch 'baseband/acc: fix access to deallocated mem' " Xueming Li
2024-11-11  6:27 ` patch 'baseband/acc: fix soft output bypass RM' " Xueming Li
2024-11-11  6:27 ` patch 'vhost: fix offset while mapping log base address' " Xueming Li
2024-11-11  6:27 ` patch 'vdpa: update used flags in used ring relay' " Xueming Li
2024-11-11  6:27 ` patch 'vdpa/nfp: fix hardware initialization' " Xueming Li
2024-11-11  6:27 ` patch 'vdpa/nfp: fix reconfiguration' " Xueming Li
2024-11-11  6:27 ` patch 'net/virtio-user: reset used index counter' " Xueming Li
2024-11-11  6:27 ` Xueming Li [this message]
2024-11-11  6:27 ` patch 'fib: fix AVX512 lookup' " Xueming Li
2024-11-11  6:27 ` patch 'net/e1000: fix link status crash in secondary process' " Xueming Li
2024-11-11  6:27 ` patch 'net/cpfl: add checks for flow action types' " Xueming Li
2024-11-11  6:27 ` patch 'net/iavf: fix crash when link is unstable' " Xueming Li
2024-11-11  6:27 ` patch 'net/cpfl: fix parsing protocol ID mask field' " Xueming Li
2024-11-11  6:27 ` patch 'net/ice/base: fix link speed for 200G' " Xueming Li
2024-11-11  6:27 ` patch 'net/ice/base: fix iteration of TLVs in Preserved Fields Area' " Xueming Li
2024-11-11  6:27 ` patch 'net/ixgbe/base: fix unchecked return value' " Xueming Li
2024-11-11  6:27 ` patch 'net/i40e/base: fix setting flags in init function' " Xueming Li
2024-11-11  6:27 ` patch 'net/i40e/base: fix misleading debug logs and comments' " Xueming Li
2024-11-11  6:27 ` patch 'net/i40e/base: add missing X710TL device check' " Xueming Li
2024-11-11  6:27 ` patch 'net/i40e/base: fix blinking X722 with X557 PHY' " Xueming Li
2024-11-11  6:27 ` patch 'net/i40e/base: fix DDP loading with reserved track ID' " Xueming Li
2024-11-11  6:27 ` patch 'net/i40e/base: fix repeated register dumps' " Xueming Li
2024-11-11  6:27 ` patch 'net/i40e/base: fix unchecked return value' " Xueming Li
2024-11-11  6:27 ` patch 'net/i40e/base: fix loop bounds' " Xueming Li
2024-11-11  6:27 ` patch 'net/iavf: delay VF reset command' " Xueming Li
2024-11-11  6:27 ` patch 'net/i40e: fix AVX-512 pointer copy on 32-bit' " Xueming Li
2024-11-11  6:27 ` patch 'net/ice: " Xueming Li
2024-11-11  6:27 ` patch 'net/iavf: " Xueming Li
2024-11-11  6:27 ` patch 'common/idpf: " Xueming Li
2024-11-11  6:27 ` patch 'net/gve: fix queue setup and stop' " Xueming Li
2024-11-11  6:28 ` patch 'net/gve: fix Tx for chained mbuf' " Xueming Li
2024-11-11  6:28 ` patch 'net/tap: avoid memcpy with null argument' " Xueming Li
2024-11-11  6:28 ` patch 'app/testpmd: remove unnecessary cast' " Xueming Li
2024-11-11  6:28 ` patch 'net/pcap: set live interface as non-blocking' " Xueming Li
2024-11-11  6:28 ` patch 'net/mana: support rdma-core via pkg-config' " Xueming Li
2024-11-11  6:28 ` patch 'net/ena: revert redefining memcpy' " Xueming Li
2024-11-11  6:28 ` patch 'net/hns3: remove some basic address dump' " Xueming Li
2024-11-11  6:28 ` patch 'net/hns3: fix dump counter of registers' " Xueming Li
2024-11-11  6:28 ` patch 'ethdev: fix overflow in descriptor count' " Xueming Li
2024-11-11  6:28 ` patch 'bus/dpaa: fix PFDRs leaks due to FQRNIs' " Xueming Li
2024-11-11  6:28 ` patch 'net/dpaa: fix typecasting channel ID' " Xueming Li
2024-11-11  6:28 ` patch 'bus/dpaa: fix VSP for 1G fm1-mac9 and 10' " Xueming Li
2024-11-11  6:28 ` patch 'bus/dpaa: fix the fman details status' " Xueming Li
2024-11-11  6:28 ` patch 'net/dpaa: fix reallocate mbuf handling' " Xueming Li
2024-11-11  6:28 ` patch 'net/gve: fix mbuf allocation memory leak for DQ Rx' " Xueming Li
2024-11-11  6:28 ` patch 'net/gve: always attempt Rx refill on DQ' " Xueming Li
2024-11-11  6:28 ` patch 'net/nfp: fix type declaration of some variables' " Xueming Li
2024-11-11  6:28 ` patch 'net/nfp: fix representor port link status update' " Xueming Li
2024-11-11  6:28 ` patch 'net/gve: fix refill logic causing memory corruption' " Xueming Li
2024-11-11  6:28 ` patch 'net/gve: add IO memory barriers before reading descriptors' " Xueming Li
2024-11-11  6:28 ` patch 'net/memif: fix buffer overflow in zero copy Rx' " Xueming Li
2024-11-11  6:28 ` patch 'net/tap: restrict maximum number of MP FDs' " Xueming Li
2024-11-11  6:28 ` patch 'ethdev: verify queue ID in Tx done cleanup' " Xueming Li
2024-11-11  6:28 ` patch 'net/hns3: verify reset type from firmware' " Xueming Li
2024-11-11  6:28 ` patch 'net/nfp: fix link change return value' " Xueming Li
2024-11-11  6:28 ` patch 'net/nfp: fix pause frame setting check' " Xueming Li
2024-11-11  6:28 ` patch 'net/pcap: fix blocking Rx' " Xueming Li
2024-11-11  6:28 ` patch 'net/ice/base: add bounds check' " Xueming Li
2024-11-11  6:28 ` patch 'net/ice/base: fix VLAN replay after reset' " Xueming Li
2024-11-11  6:28 ` patch 'net/iavf: preserve MAC address with i40e PF Linux driver' " Xueming Li
2024-11-11  6:28 ` patch 'net/mlx5: workaround list management of Rx queue control' " Xueming Li
2024-11-11  6:28 ` patch 'net/mlx5/hws: fix flex item as tunnel header' " Xueming Li
2024-11-11  6:28 ` patch 'net/mlx5: add flex item query for tunnel mode' " Xueming Li
2024-11-11  6:28 ` patch 'net/mlx5: fix flex item " Xueming Li
2024-11-11  6:28 ` patch 'net/mlx5: fix number of supported flex parsers' " Xueming Li
2024-11-11  6:28 ` patch 'app/testpmd: remove flex item init command leftover' " Xueming Li
2024-11-11  6:28 ` patch 'net/mlx5: fix next protocol validation after flex item' " Xueming Li
2024-11-11  6:28 ` patch 'net/mlx5: fix non full word sample fields in " Xueming Li
2024-11-11  6:28 ` patch 'net/mlx5: fix flex item header length field translation' " Xueming Li
2024-11-11  6:28 ` patch 'build: remove version check on compiler links function' " Xueming Li
2024-11-11  6:28 ` patch 'hash: fix thash LFSR initialization' " Xueming Li
2024-11-11  6:28 ` patch 'net/nfp: notify flower firmware about PF speed' " Xueming Li
2024-11-11  6:28 ` patch 'net/nfp: do not set IPv6 flag in transport mode' " Xueming Li
2024-11-11  6:28 ` patch 'dmadev: fix potential null pointer access' " Xueming Li
2024-11-11  6:28 ` patch 'net/gve/base: fix build with Fedora Rawhide' " Xueming Li
2024-11-11  6:28 ` patch 'power: fix mapped lcore ID' " Xueming Li
2024-11-11  6:28 ` patch 'net/ionic: fix build with Fedora Rawhide' " Xueming Li
2024-11-11  6:28 ` patch '' " Xueming Li

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=20241111062847.216344-52-xuemingl@nvidia.com \
    --to=xuemingl@nvidia.com \
    --cc=david.marchand@redhat.com \
    --cc=maxime.coquelin@redhat.com \
    --cc=stable@dpdk.org \
    --cc=yux.jiang@intel.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).