shinys000114/linux
1
0
Fork 0
mirror of https://github.com/hardkernel/linux.git synced 2026-03-26 04:20:23 +09:00
Code Issues Packages Projects Releases Wiki Activity

net/tls: Perform immediate device ctx cleanup when possible

Browse Source 
[ Upstream commit 113671b255 ]

TLS context destructor can be run in atomic context. Cleanup operations
for device-offloaded contexts could require access and interaction with
the device callbacks, which might sleep. Hence, the cleanup of such
contexts must be deferred and completed inside an async work.

For all others, this is not necessary, as cleanup is atomic. Invoke
cleanup immediately for them, avoiding queueing redundant gc work.

Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 6b47808f22 ("net: tls: avoid discarding data on record close")
Signed-off-by: Sasha Levin <sashal@kernel.org>
This commit is contained in:
Tariq Toukan
2022-07-27 12:43:41 +03:00
committed by Greg Kroah-Hartman
parent 51222e1c77
commit 2d93157b7e
1 changed files with 18 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
net/tls/tls_device.c
View File

@@ -95,19 +95,29 @@ static void tls_device_gc_task(struct work_struct *work)
static void tls_device_queue_ctx_destruction(struct tls_context *ctx)
{
unsigned long flags;
bool async_cleanup;
spin_lock_irqsave(&tls_device_lock, flags);
if (unlikely(!refcount_dec_and_test(&ctx->refcount)))
goto unlock;
if (unlikely(!refcount_dec_and_test(&ctx->refcount))) {
spin_unlock_irqrestore(&tls_device_lock, flags);
return;
}
list_move_tail(&ctx->list, &tls_device_gc_list);
async_cleanup = ctx->netdev && ctx->tx_conf == TLS_HW;
if (async_cleanup) {
list_move_tail(&ctx->list, &tls_device_gc_list);
/* schedule_work inside the spinlock
* to make sure tls_device_down waits for that work.
*/
schedule_work(&tls_device_gc_work);
unlock:
/* schedule_work inside the spinlock
* to make sure tls_device_down waits for that work.
*/
schedule_work(&tls_device_gc_work);
} else {
list_del(&ctx->list);
}
spin_unlock_irqrestore(&tls_device_lock, flags);
if (!async_cleanup)
tls_device_free_ctx(ctx);
}
/* We assume that the socket is already connected */