mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH] nvmet: use appropriate types for authentication payload lengths
@ 2026-10-31  7:59 Sreeraj S Kurup
  2026-10-31  8:17 ` [PATCH v2] " Sreeraj S Kurup
  2026-11-03  7:36 ` [PATCH] " Hannes Reinecke
  0 siblings, 2 replies; 5+ messages in thread
From: Sreeraj S Kurup @ 2026-10-31  7:59 UTC (permalink / raw)
  To: hare, hch, sagi, kch; +Cc: linux-nvme, linux-kernel, Sreeraj S Kurup

The authentication payload length is obtained as a u32 value from the
NVMe command, but nvmet_auth_challenge() and nvmet_auth_success1()
accept it as a signed int.

Use u32 for the payload length parameters in both functions to keep
their types consistent with the value returned by
nvmet_auth_receive_data_len() and avoid an unnecessary conversion
from unsigned to signed.

Also, use size_t for data_size in nvmet_auth_challenge(), since the
calculation includes sizeof(), which returns size_t. This avoids
implicitly narrowing the calculated payload size to int.

These changes make the payload length handling consistent with the
underlying types and avoid potential signedness and narrowing issues.

Signed-off-by: Sreeraj S Kurup <sreekuttan2156239@gmail.com>
---
 drivers/nvme/target/fabrics-cmd-auth.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/nvme/target/fabrics-cmd-auth.c b/drivers/nvme/target/fabrics-cmd-auth.c
index 70136f6c22e8..99caad6fb87f 100644
--- a/drivers/nvme/target/fabrics-cmd-auth.c
+++ b/drivers/nvme/target/fabrics-cmd-auth.c
@@ -448,13 +448,13 @@ void nvmet_execute_auth_send(struct nvmet_req *req)
 	nvmet_req_complete(req, status);
 }
 
-static int nvmet_auth_challenge(struct nvmet_req *req, void *d, int al)
+static int nvmet_auth_challenge(struct nvmet_req *req, void *d, u32 al)
 {
 	struct nvmf_auth_dhchap_challenge_data *data = d;
 	struct nvmet_ctrl *ctrl = req->sq->ctrl;
 	int ret = 0;
 	int hash_len = nvme_auth_hmac_hash_len(ctrl->shash_id);
-	int data_size = sizeof(*data) + hash_len;
+	size_t data_size = sizeof(*data) + hash_len;
 
 	if (ctrl->dh_tfm)
 		data_size += ctrl->dh_keysize;
@@ -488,7 +488,7 @@ static int nvmet_auth_challenge(struct nvmet_req *req, void *d, int al)
 	return ret;
 }
 
-static int nvmet_auth_success1(struct nvmet_req *req, void *d, int al)
+static int nvmet_auth_success1(struct nvmet_req *req, void *d, u32 al)
 {
 	struct nvmf_auth_dhchap_success1_data *data = d;
 	struct nvmet_ctrl *ctrl = req->sq->ctrl;
-- 
2.55.0


^ permalink raw reply	[flat|nested] 5+ messages in thread

* [PATCH v2] nvmet: use appropriate types for authentication payload lengths
  2026-10-31  7:59 [PATCH] nvmet: use appropriate types for authentication payload lengths Sreeraj S Kurup
@ 2026-10-31  8:17 ` Sreeraj S Kurup
  2026-11-03  5:38   ` Christoph Hellwig
  2026-11-03 15:37   ` Keith Busch
  2026-11-03  7:36 ` [PATCH] " Hannes Reinecke
  1 sibling, 2 replies; 5+ messages in thread
From: Sreeraj S Kurup @ 2026-10-31  8:17 UTC (permalink / raw)
  To: sreekuttan2156239; +Cc: hare, hch, kch, linux-kernel, linux-nvme, sagi

The authentication payload length is obtained as a u32 value from the
NVMe command, but nvmet_auth_challenge() and nvmet_auth_success1()
accept it as a signed int.

Use u32 for the payload length parameters in both functions to keep
their types consistent with the value returned by
nvmet_auth_receive_data_len() and avoid an unnecessary conversion
from unsigned to signed.

Also, use size_t for data_size in nvmet_auth_challenge(), since the
calculation includes sizeof(), which returns size_t. This avoids
implicitly narrowing the calculated payload size to int.

These changes make the payload length handling consistent with the
underlying types and avoid potential signedness and narrowing issues.

Signed-off-by: Sreeraj S Kurup <sreekuttan2156239@gmail.com>
---
v2:
 - Update the debug format specifiers in "nvmet_auth_challenge()" to match the 
   updated parameter types. Use "%u" for the "u32" payload length and "%zu" for 
   the "size_t" data size to avoid format type mismatches and ensure correct debug output.

 drivers/nvme/target/fabrics-cmd-auth.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/nvme/target/fabrics-cmd-auth.c b/drivers/nvme/target/fabrics-cmd-auth.c
index 70136f6c22e8..c90917d26c98 100644
--- a/drivers/nvme/target/fabrics-cmd-auth.c
+++ b/drivers/nvme/target/fabrics-cmd-auth.c
@@ -448,18 +448,18 @@ void nvmet_execute_auth_send(struct nvmet_req *req)
 	nvmet_req_complete(req, status);
 }
 
-static int nvmet_auth_challenge(struct nvmet_req *req, void *d, int al)
+static int nvmet_auth_challenge(struct nvmet_req *req, void *d, u32 al)
 {
 	struct nvmf_auth_dhchap_challenge_data *data = d;
 	struct nvmet_ctrl *ctrl = req->sq->ctrl;
 	int ret = 0;
 	int hash_len = nvme_auth_hmac_hash_len(ctrl->shash_id);
-	int data_size = sizeof(*data) + hash_len;
+	size_t data_size = sizeof(*data) + hash_len;
 
 	if (ctrl->dh_tfm)
 		data_size += ctrl->dh_keysize;
 	if (al < data_size) {
-		pr_debug("%s: buffer too small (al %d need %d)\n", __func__,
+		pr_debug("%s: buffer too small (al %u need %zu)\n", __func__,
 			 al, data_size);
 		return -EINVAL;
 	}
@@ -488,7 +488,7 @@ static int nvmet_auth_challenge(struct nvmet_req *req, void *d, int al)
 	return ret;
 }
 
-static int nvmet_auth_success1(struct nvmet_req *req, void *d, int al)
+static int nvmet_auth_success1(struct nvmet_req *req, void *d, u32 al)
 {
 	struct nvmf_auth_dhchap_success1_data *data = d;
 	struct nvmet_ctrl *ctrl = req->sq->ctrl;
-- 
2.55.0


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH v2] nvmet: use appropriate types for authentication payload lengths
  2026-10-31  8:17 ` [PATCH v2] " Sreeraj S Kurup
@ 2026-11-03  5:38   ` Christoph Hellwig
  2026-11-03 15:37   ` Keith Busch
  1 sibling, 0 replies; 5+ messages in thread
From: Christoph Hellwig @ 2026-11-03  5:38 UTC (permalink / raw)
  To: Sreeraj S Kurup; +Cc: hare, hch, kch, linux-kernel, linux-nvme, sagi

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

But please don't resend new versions in the same thread.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] nvmet: use appropriate types for authentication payload lengths
  2026-10-31  7:59 [PATCH] nvmet: use appropriate types for authentication payload lengths Sreeraj S Kurup
  2026-10-31  8:17 ` [PATCH v2] " Sreeraj S Kurup
@ 2026-11-03  7:36 ` Hannes Reinecke
  1 sibling, 0 replies; 5+ messages in thread
From: Hannes Reinecke @ 2026-11-03  7:36 UTC (permalink / raw)
  To: Sreeraj S Kurup, hch, sagi, kch; +Cc: linux-nvme, linux-kernel

On 10/31/26 8:59 AM, Sreeraj S Kurup wrote:
> The authentication payload length is obtained as a u32 value from the
> NVMe command, but nvmet_auth_challenge() and nvmet_auth_success1()
> accept it as a signed int.
> 
> Use u32 for the payload length parameters in both functions to keep
> their types consistent with the value returned by
> nvmet_auth_receive_data_len() and avoid an unnecessary conversion
> from unsigned to signed.
> 
> Also, use size_t for data_size in nvmet_auth_challenge(), since the
> calculation includes sizeof(), which returns size_t. This avoids
> implicitly narrowing the calculated payload size to int.
> 
> These changes make the payload length handling consistent with the
> underlying types and avoid potential signedness and narrowing issues.
> 
> Signed-off-by: Sreeraj S Kurup <sreekuttan2156239@gmail.com>
> ---
>   drivers/nvme/target/fabrics-cmd-auth.c | 6 +++---
>   1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/nvme/target/fabrics-cmd-auth.c b/drivers/nvme/target/fabrics-cmd-auth.c
> index 70136f6c22e8..99caad6fb87f 100644
> --- a/drivers/nvme/target/fabrics-cmd-auth.c
> +++ b/drivers/nvme/target/fabrics-cmd-auth.c
> @@ -448,13 +448,13 @@ void nvmet_execute_auth_send(struct nvmet_req *req)
>   	nvmet_req_complete(req, status);
>   }
>   
> -static int nvmet_auth_challenge(struct nvmet_req *req, void *d, int al)
> +static int nvmet_auth_challenge(struct nvmet_req *req, void *d, u32 al)
>   {
>   	struct nvmf_auth_dhchap_challenge_data *data = d;
>   	struct nvmet_ctrl *ctrl = req->sq->ctrl;
>   	int ret = 0;
>   	int hash_len = nvme_auth_hmac_hash_len(ctrl->shash_id);
> -	int data_size = sizeof(*data) + hash_len;
> +	size_t data_size = sizeof(*data) + hash_len;
>   
>   	if (ctrl->dh_tfm)
>   		data_size += ctrl->dh_keysize;
> @@ -488,7 +488,7 @@ static int nvmet_auth_challenge(struct nvmet_req *req, void *d, int al)
>   	return ret;
>   }
>   
> -static int nvmet_auth_success1(struct nvmet_req *req, void *d, int al)
> +static int nvmet_auth_success1(struct nvmet_req *req, void *d, u32 al)
>   {
>   	struct nvmf_auth_dhchap_success1_data *data = d;
>   	struct nvmet_ctrl *ctrl = req->sq->ctrl;

Reviewed-by: Hannes Reinecke <hare@kernel.org>

Cheers,

Hannes
-- 
Dr. Hannes Reinecke                  Kernel Storage Architect
hare@suse.de                                +49 911 74053 688
SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg
HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH v2] nvmet: use appropriate types for authentication payload lengths
  2026-10-31  8:17 ` [PATCH v2] " Sreeraj S Kurup
  2026-11-03  5:38   ` Christoph Hellwig
@ 2026-11-03 15:37   ` Keith Busch
  1 sibling, 0 replies; 5+ messages in thread
From: Keith Busch @ 2026-11-03 15:37 UTC (permalink / raw)
  To: Sreeraj S Kurup; +Cc: hare, hch, kch, linux-kernel, linux-nvme, sagi

On Sat, Oct 31, 2026 at 08:17:51AM +0000, Sreeraj S Kurup wrote:
> The authentication payload length is obtained as a u32 value from the
> NVMe command, but nvmet_auth_challenge() and nvmet_auth_success1()
> accept it as a signed int.
> 
> Use u32 for the payload length parameters in both functions to keep
> their types consistent with the value returned by
> nvmet_auth_receive_data_len() and avoid an unnecessary conversion
> from unsigned to signed.
> 
> Also, use size_t for data_size in nvmet_auth_challenge(), since the
> calculation includes sizeof(), which returns size_t. This avoids
> implicitly narrowing the calculated payload size to int.
> 
> These changes make the payload length handling consistent with the
> underlying types and avoid potential signedness and narrowing issues.

Thanks, applied to nvme-7.4.

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2026-11-03 15:38 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-10-31  7:59 [PATCH] nvmet: use appropriate types for authentication payload lengths Sreeraj S Kurup
2026-10-31  8:17 ` [PATCH v2] " Sreeraj S Kurup
2026-11-03  5:38   ` Christoph Hellwig
2026-11-03 15:37   ` Keith Busch
2026-11-03  7:36 ` [PATCH] " Hannes Reinecke

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®