[PATCH] net/iavf: fix dead store on reset path

[PATCH] net/iavf: fix dead store on reset path

[Ciara Loftus]

When dev_start fails during reset recovery, the code defers the start
to the next link-up event and resets ret to 0 to avoid treating this
as an error. This is redundant since ret is unconditionally
overwritten by the return value of iavf_post_reset_reconfig()
immediately after. Remove the dead assignment.

Fixes: 9aac2219dc2a ("net/iavf: defer device start when PF VSI not ready")

Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
---
 drivers/net/intel/iavf/iavf_ethdev.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/intel/iavf/iavf_ethdev.c b/drivers/net/intel/iavf/iavf_ethdev.c
index a8daa507bb..7915aef348 100644
--- a/drivers/net/intel/iavf/iavf_ethdev.c
+++ b/drivers/net/intel/iavf/iavf_ethdev.c
@@ -3513,7 +3513,6 @@ iavf_handle_hw_reset(struct rte_eth_dev *dev, bool vf_initiated_reset)
 				    ret);
 			vf->start_pending = true;
 			dev->data->dev_started = 0;
-			ret = 0;
 		}
 	}
 
-- 
2.43.0

Re: [PATCH] net/iavf: fix dead store on reset path

[Burakov, Anatoly]

On 9/25/2026 10:58 AM, Ciara Loftus wrote:
> When dev_start fails during reset recovery, the code defers the start
> to the next link-up event and resets ret to 0 to avoid treating this
> as an error. This is redundant since ret is unconditionally
> overwritten by the return value of iavf_post_reset_reconfig()
> immediately after. Remove the dead assignment.
> 
> Fixes: 9aac2219dc2a ("net/iavf: defer device start when PF VSI not ready")
> 
> Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
> ---

I wouldn't really qualify this as a bug fix except maybe static analysis 
tools might flag it so perhaps having a Fixes tag is worth it.

Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>

-- 
Thanks,
Anatoly

RE: [PATCH] net/iavf: fix dead store on reset path

[Loftus, Ciara]

> Subject: Re: [PATCH] net/iavf: fix dead store on reset path
> 
> On 9/25/2026 10:58 AM, Ciara Loftus wrote:
> > When dev_start fails during reset recovery, the code defers the start
> > to the next link-up event and resets ret to 0 to avoid treating this
> > as an error. This is redundant since ret is unconditionally
> > overwritten by the return value of iavf_post_reset_reconfig()
> > immediately after. Remove the dead assignment.
> >
> > Fixes: 9aac2219dc2a ("net/iavf: defer device start when PF VSI not ready")
> >
> > Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
> > ---
> 
> I wouldn't really qualify this as a bug fix except maybe static analysis
> tools might flag it so perhaps having a Fixes tag is worth it.

Apologies this was indeed found by static analysis and should have
included the following tag, if it can be added on apply:

Coverity issue: 506150

> 
> Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
> 
> --
> Thanks,
> Anatoly

Re: [PATCH] net/iavf: fix dead store on reset path

[Bruce Richardson]

On Fri, Sep 25, 2026 at 03:33:08PM +0100, Loftus, Ciara wrote:
> > Subject: Re: [PATCH] net/iavf: fix dead store on reset path
> > 
> > On 9/25/2026 10:58 AM, Ciara Loftus wrote:
> > > When dev_start fails during reset recovery, the code defers the start
> > > to the next link-up event and resets ret to 0 to avoid treating this
> > > as an error. This is redundant since ret is unconditionally
> > > overwritten by the return value of iavf_post_reset_reconfig()
> > > immediately after. Remove the dead assignment.
> > >
> > > Fixes: 9aac2219dc2a ("net/iavf: defer device start when PF VSI not ready")
> > >
> > > Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
> > > ---
> > 
> > I wouldn't really qualify this as a bug fix except maybe static analysis
> > tools might flag it so perhaps having a Fixes tag is worth it.
> 
> Apologies this was indeed found by static analysis and should have
> included the following tag, if it can be added on apply:
> 
> Coverity issue: 506150
> 
> > 
> > Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
> > 
Added the coverity issue tag - because otherwise I was wondering why we
were bothering with this - and applied patch to dpdk-next-net-intel

Thanks,
/Bruce