From: Saeed Mahameed <saeed@kernel.org>
To: "David S. Miller" <davem@davemloft.net>,
Jakub Kicinski <kuba@kernel.org>
Cc: netdev@vger.kernel.org, Moshe Shemesh <moshe@nvidia.com>,
Eran Ben Elisha <eranbe@nvidia.com>,
Saeed Mahameed <saeedm@nvidia.com>
Subject: [net 10/11] net/mlx5: Set command entry semaphore up once got index free
Date: Thu, 6 Jan 2022 16:58:30 -0800 [thread overview]
Message-ID: <20220107005831.78909-11-saeed@kernel.org> (raw)
In-Reply-To: <20220107005831.78909-1-saeed@kernel.org>
From: Moshe Shemesh <moshe@nvidia.com>
Avoid a race where command work handler may fail to allocate command
entry index, by holding the command semaphore down till command entry
index is being freed.
Fixes: 410bd754cd73 ("net/mlx5: Add retry mechanism to the command entry index allocation")
Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 15 ++++++---------
1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index a46284ca5172..f588503157d0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -148,8 +148,12 @@ static void cmd_ent_put(struct mlx5_cmd_work_ent *ent)
if (!refcount_dec_and_test(&ent->refcnt))
return;
- if (ent->idx >= 0)
- cmd_free_index(ent->cmd, ent->idx);
+ if (ent->idx >= 0) {
+ struct mlx5_cmd *cmd = ent->cmd;
+
+ cmd_free_index(cmd, ent->idx);
+ up(ent->page_queue ? &cmd->pages_sem : &cmd->sem);
+ }
cmd_free_ent(ent);
}
@@ -1602,8 +1606,6 @@ static void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool force
vector = vec & 0xffffffff;
for (i = 0; i < (1 << cmd->log_sz); i++) {
if (test_bit(i, &vector)) {
- struct semaphore *sem;
-
ent = cmd->ent_arr[i];
/* if we already completed the command, ignore it */
@@ -1626,10 +1628,6 @@ static void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool force
dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR)
cmd_ent_put(ent);
- if (ent->page_queue)
- sem = &cmd->pages_sem;
- else
- sem = &cmd->sem;
ent->ts2 = ktime_get_ns();
memcpy(ent->out->first.data, ent->lay->out, sizeof(ent->lay->out));
dump_command(dev, ent, 0);
@@ -1683,7 +1681,6 @@ static void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool force
*/
complete(&ent->done);
}
- up(sem);
}
}
}
--
2.33.1
next prev parent reply other threads:[~2022-01-07 0:59 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-01-07 0:58 [pull request][net 00/11] mlx5 fixes 2022-01-06 Saeed Mahameed
2022-01-07 0:58 ` [net 01/11] net/mlx5e: Fix page DMA map/unmap attributes Saeed Mahameed
2022-01-07 11:20 ` patchwork-bot+netdevbpf
2022-01-07 0:58 ` [net 02/11] net/mlx5e: Fix nullptr on deleting mirroring rule Saeed Mahameed
2022-01-07 0:58 ` [net 03/11] net/mlx5e: Fix wrong usage of fib_info_nh when routes with nexthop objects are used Saeed Mahameed
2022-01-07 0:58 ` [net 04/11] net/mlx5e: Don't block routes with nexthop objects in SW Saeed Mahameed
2022-01-07 0:58 ` [net 05/11] Revert "net/mlx5e: Block offload of outer header csum for UDP tunnels" Saeed Mahameed
2022-01-07 0:58 ` [net 06/11] Revert "net/mlx5e: Block offload of outer header csum for GRE tunnel" Saeed Mahameed
2022-01-07 0:58 ` [net 07/11] net/mlx5e: Fix matching on modified inner ip_ecn bits Saeed Mahameed
2022-01-07 0:58 ` [net 08/11] net/mlx5: Fix access to sf_dev_table on allocation failure Saeed Mahameed
2022-01-07 0:58 ` [net 09/11] net/mlx5e: Sync VXLAN udp ports during uplink representor profile change Saeed Mahameed
2022-01-07 0:58 ` Saeed Mahameed [this message]
2022-01-07 0:58 ` [net 11/11] Revert "net/mlx5: Add retry mechanism to the command entry index allocation" Saeed Mahameed
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=20220107005831.78909-11-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).