From: Benjamin Marzinski <bmarzins@redhat.com>
To: Sarthak Kukreti <sarthakkukreti@google.com>
Cc: Alasdair Kergon <agk@redhat.com>,
Mike Snitzer <snitzer@kernel.org>,
Mikulas Patocka <mpatocka@redhat.com>,
Eric Biggers <ebiggers@kernel.org>,
Brian Geffon <bgeffon@google.com>,
David Stevens <stevensd@chromium.org>,
dm-devel@lists.linux.dev, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] dm-verity: export verify_wq to sysfs
Date: Tue, 17 Nov 2026 14:44:09 -0500 [thread overview]
Message-ID: <avyuiWKNFN19BSJB@redhat.com> (raw)
In-Reply-To: <20261116223301.1706170-1-sarthakkukreti@google.com>
On Mon, Nov 16, 2026 at 02:33:01PM -0800, Sarthak Kukreti wrote:
> Under heavy concurrent read bursts, dm-verity workers blocking on
> dm-bufio reads or locks can cause workqueue concurrency management to
> spawn hundreds of per-CPU kverityd threads across the system.
>
> Export verify_wq to sysfs via WQ_SYSFS so userspace can inspect and
> configure workqueue attributes (such as max_active) per dm-verity
> device under /sys/bus/workqueue/devices/kverityd-<dev>-<id>/.
>
> Co-developed-by: Brian Geffon <bgeffon@google.com>
> Signed-off-by: Brian Geffon <bgeffon@google.com>
> Co-developed-by: David Stevens <stevensd@chromium.org>
> Signed-off-by: David Stevens <stevensd@chromium.org>
> Signed-off-by: Sarthak Kukreti <sarthakkukreti@google.com>
Reviewed-by: Benjamin Marzinski <bmarzins@redhat.com>
> ---
> drivers/md/dm-verity-target.c | 22 ++++++++++++++++++----
> drivers/md/dm-verity.h | 1 +
> 2 files changed, 19 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c
> index 306ed1f6c118..730adf931494 100644
> --- a/drivers/md/dm-verity-target.c
> +++ b/drivers/md/dm-verity-target.c
> @@ -18,6 +18,7 @@
> #include "dm-verity-verify-sig.h"
> #include "dm-audit.h"
> #include <linux/hex.h>
> +#include <linux/idr.h>
> #include <linux/module.h>
> #include <linux/reboot.h>
> #include <linux/string.h>
> @@ -26,6 +27,8 @@
>
> #define DM_MSG_PREFIX "verity"
>
> +static DEFINE_IDA(workqueue_ida);
> +
> #define DM_VERITY_ENV_LENGTH 46
> #define DM_VERITY_ENV_VAR_NAME "DM_VERITY_ERR_BLOCK_NR"
>
> @@ -1065,6 +1068,8 @@ static void verity_dtr(struct dm_target *ti)
>
> if (v->verify_wq)
> destroy_workqueue(v->verify_wq);
> + if (v->workqueue_id)
> + ida_free(&workqueue_ida, v->workqueue_id);
>
> mempool_exit(&v->recheck_pool);
> if (v->io)
> @@ -1421,7 +1426,7 @@ static int verity_ctr(struct dm_target *ti, unsigned int argc, char **argv)
> struct dm_arg_set as;
> unsigned int num;
> unsigned long long num_ll;
> - int r;
> + int r, wq_id;
> int i;
> sector_t hash_position;
> char dummy;
> @@ -1637,6 +1642,14 @@ static int verity_ctr(struct dm_target *ti, unsigned int argc, char **argv)
> goto bad;
> }
>
> + wq_id = ida_alloc_min(&workqueue_ida, 1, GFP_KERNEL);
> + if (wq_id < 0) {
> + ti->error = "Cannot allocate workqueue id";
> + r = wq_id;
> + goto bad;
> + }
> + v->workqueue_id = wq_id;
> +
> /*
> * Using WQ_HIGHPRI improves throughput and completion latency by
> * reducing wait times when reading from a dm-verity device.
> @@ -1646,9 +1659,10 @@ static int verity_ctr(struct dm_target *ti, unsigned int argc, char **argv)
> * will fall-back to using it for error handling (or if the bufio cache
> * doesn't have required hashes).
> */
> - v->verify_wq = alloc_workqueue("kverityd",
> - WQ_MEM_RECLAIM | WQ_HIGHPRI | WQ_PERCPU,
> - 0);
> + v->verify_wq = alloc_workqueue("kverityd-%s-%d",
> + WQ_MEM_RECLAIM | WQ_HIGHPRI |
> + WQ_PERCPU | WQ_SYSFS, 0,
> + dm_table_device_name(ti->table), wq_id);
> if (!v->verify_wq) {
> ti->error = "Cannot allocate workqueue";
> r = -ENOMEM;
> diff --git a/drivers/md/dm-verity.h b/drivers/md/dm-verity.h
> index e104a651c657..53f894f259d3 100644
> --- a/drivers/md/dm-verity.h
> +++ b/drivers/md/dm-verity.h
> @@ -71,6 +71,7 @@ struct dm_verity {
> atomic_t corrupted_errs;/* Number of errors for corrupted blocks */
>
> struct workqueue_struct *verify_wq;
> + int workqueue_id;
>
> /* starting blocks for each tree level. 0 is the lowest level. */
> sector_t hash_level_block[DM_VERITY_MAX_LEVELS];
> --
> 2.56.0.679.gc527c65306-goog
prev parent reply other threads:[~2026-11-17 19:44 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-16 22:33 [PATCH] dm-verity: export verify_wq to sysfs Sarthak Kukreti
2026-11-17 19:44 ` Benjamin Marzinski [this message]
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=avyuiWKNFN19BSJB@redhat.com \
--to=bmarzins@redhat.com \
--cc=agk@redhat.com \
--cc=bgeffon@google.com \
--cc=dm-devel@lists.linux.dev \
--cc=ebiggers@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mpatocka@redhat.com \
--cc=sarthakkukreti@google.com \
--cc=snitzer@kernel.org \
--cc=stevensd@chromium.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.