From: saeed@kernel.org
To: "David S. Miller" <davem@davemloft.net>,
Jakub Kicinski <kuba@kernel.org>
Cc: netdev@vger.kernel.org, Eran Ben Elisha <eranbe@nvidia.com>,
Saeed Mahameed <saeedm@nvidia.com>,
Moshe Shemesh <moshe@nvidia.com>
Subject: [net 05/15] net/mlx5: Add retry mechanism to the command entry index allocation
Date: Wed, 30 Sep 2020 19:05:06 -0700 [thread overview]
Message-ID: <20201001020516.41217-6-saeed@kernel.org> (raw)
In-Reply-To: <20201001020516.41217-1-saeed@kernel.org>
From: Eran Ben Elisha <eranbe@nvidia.com>
It is possible that new command entry index allocation will temporarily
fail. The new command holds the semaphore, so it means that a free entry
should be ready soon. Add one second retry mechanism before returning an
error.
Patch "net/mlx5: Avoid possible free of command entry while timeout comp
handler" increase the possibility to bump into this temporarily failure
as it delays the entry index release for non-callback commands.
Fixes: e126ba97dba9 ("mlx5: Add driver for Mellanox Connect-IB adapters")
Signed-off-by: Eran Ben Elisha <eranbe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
---
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 21 ++++++++++++++++++-
1 file changed, 20 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index 65ae6ef2039e..4b54c9241fd7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -883,6 +883,25 @@ static bool opcode_allowed(struct mlx5_cmd *cmd, u16 opcode)
return cmd->allowed_opcode == opcode;
}
+static int cmd_alloc_index_retry(struct mlx5_cmd *cmd)
+{
+ unsigned long alloc_end = jiffies + msecs_to_jiffies(1000);
+ int idx;
+
+retry:
+ idx = cmd_alloc_index(cmd);
+ if (idx < 0 && time_before(jiffies, alloc_end)) {
+ /* Index allocation can fail on heavy load of commands. This is a temporary
+ * situation as the current command already holds the semaphore, meaning that
+ * another command completion is being handled and it is expected to release
+ * the entry index soon.
+ */
+ cond_resched();
+ goto retry;
+ }
+ return idx;
+}
+
static void cmd_work_handler(struct work_struct *work)
{
struct mlx5_cmd_work_ent *ent = container_of(work, struct mlx5_cmd_work_ent, work);
@@ -900,7 +919,7 @@ static void cmd_work_handler(struct work_struct *work)
sem = ent->page_queue ? &cmd->pages_sem : &cmd->sem;
down(sem);
if (!ent->page_queue) {
- alloc_ret = cmd_alloc_index(cmd);
+ alloc_ret = cmd_alloc_index_retry(cmd);
if (alloc_ret < 0) {
mlx5_core_err_rl(dev, "failed to allocate command entry\n");
if (ent->callback) {
--
2.26.2
next prev parent reply other threads:[~2020-10-01 2:05 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-01 2:05 [pull request][net 00/15] mlx5 fixes 2020-09-30 saeed
2020-10-01 2:05 ` [net 01/15] net/mlx5: Don't allow health work when device is uninitialized saeed
2020-10-01 19:21 ` David Miller
2020-10-01 19:41 ` Saeed Mahameed
2020-10-01 2:05 ` [net 02/15] net/mlx5: Fix a race when moving command interface to polling mode saeed
2020-10-01 2:05 ` [net 03/15] net/mlx5: Avoid possible free of command entry while timeout comp handler saeed
2020-10-01 2:05 ` [net 04/15] net/mlx5: poll cmd EQ in case of command timeout saeed
2020-10-01 2:05 ` saeed [this message]
2020-10-01 2:05 ` [net 06/15] net/mlx5: cmdif, Avoid skipping reclaim pages if FW is not accessible saeed
2020-10-01 2:05 ` [net 07/15] net/mlx5: Fix request_irqs error flow saeed
2020-10-01 2:05 ` [net 08/15] net/mlx5e: Fix error path for RQ alloc saeed
2020-10-01 2:05 ` [net 09/15] net/mlx5e: Add resiliency in Striding RQ mode for packets larger than MTU saeed
2020-10-01 2:05 ` [net 10/15] net/mlx5e: CT, Fix coverity issue saeed
2020-10-01 2:05 ` [net 11/15] net/mlx5e: Fix driver's declaration to support GRE offload saeed
2020-10-01 2:05 ` [net 12/15] net/mlx5e: Fix return status when setting unsupported FEC mode saeed
2020-10-01 2:05 ` [net 13/15] net/mlx5e: Fix VLAN cleanup flow saeed
2020-10-01 2:05 ` [net 14/15] net/mlx5e: Fix VLAN create flow saeed
2020-10-01 2:05 ` [net 15/15] net/mlx5e: Fix race condition on nhe->n pointer in neigh update saeed
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=20201001020516.41217-6-saeed@kernel.org \
--to=saeed@kernel.org \
--cc=davem@davemloft.net \
--cc=eranbe@nvidia.com \
--cc=kuba@kernel.org \
--cc=moshe@nvidia.com \
--cc=netdev@vger.kernel.org \
--cc=saeedm@nvidia.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).