|SUCCESS| crypto/scheduler: fix typo

dpdklab at iol.unh.edu dpdklab at iol.unh.edu
Thu Apr 16 18:10:19 CEST 2026


Test-Label: ai-code-review
Test-Status: SUCCESS
https://dpdk.org/patch/163120

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-04-16

Commit: 6812b9bf470e ("crypto/scheduler: use unified session")
Author: Radu Nicolau <radu.nicolau at intel.com>
Subject: crypto/scheduler: fix typo

This patch corrects a typo where pending_deq_ops was incorrectly used
instead of pending_enq_ops in two locations within the mc_scheduler_worker
function.

> diff --git a/drivers/crypto/scheduler/scheduler_multicore.c b/drivers/crypto/scheduler/scheduler_multicore.c
> index a21b522f9f..70f8a25b70 100644
> --- a/drivers/crypto/scheduler/scheduler_multicore.c
> +++ b/drivers/crypto/scheduler/scheduler_multicore.c
> @@ -191,11 +191,11 @@ mc_scheduler_worker(struct rte_cryptodev *dev)
>  					worker->qp_id,
>  					&enq_ops[pending_enq_ops_idx],
>  					pending_enq_ops);
> -			if (processed_ops < pending_deq_ops)
> +			if (processed_ops < pending_enq_ops)

Does this comparison check if all enqueue operations were processed? The
variable name change suggests this should compare against pending_enq_ops
to determine if there are unprocessed enqueue operations remaining.

>  				scheduler_retrieve_sessions(
>  					&enq_ops[pending_enq_ops_idx +
>  						processed_ops],
> -					pending_deq_ops - processed_ops);
> +					pending_enq_ops - processed_ops);

Is the session retrieval count correctly calculated now? Previously this
used pending_deq_ops which appears unrelated to the enqueue operations
being processed. Does this fix ensure that sessions are retrieved for the
correct number of unprocessed enqueue operations?

>  			pending_enq_ops -= processed_ops;
>  			pending_enq_ops_idx += processed_ops;
>  			inflight_ops += processed_ops;

The patch correctly addresses the typo. However, reviewing the broader
context of mc_scheduler_worker would be helpful to confirm that
pending_deq_ops is never meant to be used in this code section, and that
all references to enqueue operations use the correct variable.


More information about the test-report mailing list