From: Leon Romanovsky <leon@kernel.org>
To: Siva Reddy Kallam <siva.kallam@broadcom.com>
Cc: jgg@nvidia.com, linux-rdma@vger.kernel.org,
usman.ansari@broadcom.com, Simon Horman <horms@kernel.org>,
kernel test robot <lkp@intel.com>,
Dan Carpenter <error27@gmail.com>
Subject: Re: [PATCH] RDMA/bng_re: Unwind bng_re_dev_init properly and remove unnecessary rdev check
Date: Sun, 11 Jan 2026 15:22:55 +0200 [thread overview]
Message-ID: <20260111132255.GA14378@unreal> (raw)
In-Reply-To: <20260107091607.104468-1-siva.kallam@broadcom.com>
On Wed, Jan 07, 2026 at 09:16:07AM +0000, Siva Reddy Kallam wrote:
> Fix below smatch warnings:
> drivers/infiniband/hw/bng_re/bng_dev.c:113
> bng_re_net_ring_free() warn: variable dereferenced before check 'rdev'
> (see line 107)
> drivers/infiniband/hw/bng_re/bng_dev.c:270
> bng_re_dev_init() warn: missing unwind goto?
Please provide commit message.
>
> Fixes: 4f830cd8d7fe ("RDMA/bng_re: Add infrastructure for enabling Firmware channel")
> Reported-by: Simon Horman <horms@kernel.org>
> Reported-by: kernel test robot <lkp@intel.com>
> Reported-by: Dan Carpenter <error27@gmail.com>
> Closes: https://lore.kernel.org/r/202601010413.sWadrQel-lkp@intel.com/
> Signed-off-by: Siva Reddy Kallam <siva.kallam@broadcom.com>
> Reviewed-by: Usman Ansari <usman.ansari@broadcom.com>
> ---
> drivers/infiniband/hw/bng_re/bng_dev.c | 33 +++++++++++++-------------
> 1 file changed, 16 insertions(+), 17 deletions(-)
>
> diff --git a/drivers/infiniband/hw/bng_re/bng_dev.c b/drivers/infiniband/hw/bng_re/bng_dev.c
> index d8f8d7f7075f..e2dd2c8eb6d2 100644
> --- a/drivers/infiniband/hw/bng_re/bng_dev.c
> +++ b/drivers/infiniband/hw/bng_re/bng_dev.c
> @@ -124,9 +124,6 @@ static int bng_re_net_ring_free(struct bng_re_dev *rdev,
> struct bnge_fw_msg fw_msg = {};
> int rc = -EINVAL;
>
> - if (!rdev)
You have other places with impossible "if (rdev)" check in this path which you should
delete as well.
> - return rc;
> -
> if (!aux_dev)
You should remove this check too.
> return rc;
>
> @@ -303,7 +300,7 @@ static int bng_re_dev_init(struct bng_re_dev *rdev)
> if (rc) {
> ibdev_err(&rdev->ibdev,
> "Failed to register with netedev: %#x\n", rc);
> - return -EINVAL;
> + goto reg_netdev_fail;
> }
>
> set_bit(BNG_RE_FLAG_NETDEV_REGISTERED, &rdev->flags);
> @@ -312,19 +309,16 @@ static int bng_re_dev_init(struct bng_re_dev *rdev)
> ibdev_err(&rdev->ibdev,
> "RoCE requires minimum 2 MSI-X vectors, but only %d reserved\n",
> rdev->aux_dev->auxr_info->msix_requested);
> - bnge_unregister_dev(rdev->aux_dev);
> - clear_bit(BNG_RE_FLAG_NETDEV_REGISTERED, &rdev->flags);
> - return -EINVAL;
> + rc = -EINVAL;
> + goto msix_ctx_fail;
> }
> ibdev_dbg(&rdev->ibdev, "Got %d MSI-X vectors\n",
> rdev->aux_dev->auxr_info->msix_requested);
>
> rc = bng_re_setup_chip_ctx(rdev);
> if (rc) {
> - bnge_unregister_dev(rdev->aux_dev);
> - clear_bit(BNG_RE_FLAG_NETDEV_REGISTERED, &rdev->flags);
> ibdev_err(&rdev->ibdev, "Failed to get chip context\n");
> - return -EINVAL;
> + goto msix_ctx_fail;
> }
>
> bng_re_query_hwrm_version(rdev);
> @@ -333,16 +327,14 @@ static int bng_re_dev_init(struct bng_re_dev *rdev)
> if (rc) {
> ibdev_err(&rdev->ibdev,
> "Failed to allocate RCFW Channel: %#x\n", rc);
> - goto fail;
> + goto alloc_fw_chl_fail;
> }
>
> /* Allocate nq record memory */
> rdev->nqr = kzalloc(sizeof(*rdev->nqr), GFP_KERNEL);
> if (!rdev->nqr) {
> - bng_re_destroy_chip_ctx(rdev);
> - bnge_unregister_dev(rdev->aux_dev);
> - clear_bit(BNG_RE_FLAG_NETDEV_REGISTERED, &rdev->flags);
> - return -ENOMEM;
> + rc = -ENOMEM;
> + goto nq_alloc_fail;
> }
>
> rdev->nqr->num_msix = rdev->aux_dev->auxr_info->msix_requested;
> @@ -411,9 +403,16 @@ static int bng_re_dev_init(struct bng_re_dev *rdev)
> free_ring:
> bng_re_net_ring_free(rdev, rdev->rcfw.creq.ring_id, type);
> free_rcfw:
> + kfree(rdev->nqr);
> + rdev->nqr = NULL;
Why do you need to set NULL here?
> +nq_alloc_fail:
> bng_re_free_rcfw_channel(&rdev->rcfw);
> -fail:
> - bng_re_dev_uninit(rdev);
> +alloc_fw_chl_fail:
> + bng_re_destroy_chip_ctx(rdev);
> +msix_ctx_fail:
> + bnge_unregister_dev(rdev->aux_dev);
> + clear_bit(BNG_RE_FLAG_NETDEV_REGISTERED, &rdev->flags);
> +reg_netdev_fail:
> return rc;
> }
>
> --
> 2.25.1
>
>
next prev parent reply other threads:[~2026-01-11 13:23 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-07 9:16 [PATCH] RDMA/bng_re: Unwind bng_re_dev_init properly and remove unnecessary rdev check Siva Reddy Kallam
2026-01-11 13:22 ` Leon Romanovsky [this message]
2026-01-12 9:14 ` Siva Reddy Kallam
2026-01-12 9:45 ` Leon Romanovsky
2026-01-12 10:15 ` Siva Reddy Kallam
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=20260111132255.GA14378@unreal \
--to=leon@kernel.org \
--cc=error27@gmail.com \
--cc=horms@kernel.org \
--cc=jgg@nvidia.com \
--cc=linux-rdma@vger.kernel.org \
--cc=lkp@intel.com \
--cc=siva.kallam@broadcom.com \
--cc=usman.ansari@broadcom.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