On Thu, Jun 06, 2019 at 09:00:42AM -0700, Nathan Chancellor wrote:
> On Thu, Jun 06, 2019 at 02:53:23PM +0200, Pavel Machek wrote:
> > Hi!
> >
> > > [ Upstream commit faf5a744f4f8d76e7c03912b5cd381ac8045f6ec ]
> > >
> > > clang -Wuninitialized incorrectly sees a variable being used without
> > > initialization:
> > >
> > > drivers/scsi/lpfc/lpfc_nvme.c:2102:37: error: variable 'localport' is uninitialized when used here
> > > [-Werror,-Wuninitialized]
> > > lport = (struct lpfc_nvme_lport *)localport->private;
> > > ^~~~~~~~~
> > > drivers/scsi/lpfc/lpfc_nvme.c:2059:38: note: initialize the variable 'localport' to silence this warning
> > > struct nvme_fc_local_port *localport;
> > > ^
> > > = NULL
> > > 1 error generated.
> > >
> > > This is clearly in dead code, as the condition leading up to it is always
> > > false when CONFIG_NVME_FC is disabled, and the variable is always
> > > initialized when nvme_fc_register_localport() got called successfully.
> > >
> > > Change the preprocessor conditional to the equivalent C construct, which
> > > makes the code more readable and gets rid of the warning.
> >
> > Unfortunately, this missed "else" branch where the code was freeing
> > the memory with kfree(cstat)... so this introduces a memory leak.
> >
> > Best regards,
> > Pavel
>
> For the record, this is not a problem with the upstream commit (not
> saying you thought that or not, I just want to be clear).
>
> Looks like commit 4c47efc140fa ("scsi: lpfc: Move SCSI and NVME Stats to
> hardware queue structures") "resolved" this by not making it an issue in
> the first place. I think the simpler fix is this.
>
> Thanks for pointing it out!
>
> diff --git a/drivers/scsi/lpfc/lpfc_nvme.c b/drivers/scsi/lpfc/lpfc_nvme.c
> index 099f70798fdd..645ffb5332b4 100644
> --- a/drivers/scsi/lpfc/lpfc_nvme.c
> +++ b/drivers/scsi/lpfc/lpfc_nvme.c
> @@ -2477,14 +2477,14 @@ lpfc_nvme_create_localport(struct lpfc_vport *vport)
> lpfc_nvme_template.max_sgl_segments = phba->cfg_nvme_seg_cnt + 1;
> lpfc_nvme_template.max_hw_queues = phba->cfg_nvme_io_channel;
>
> + if (!IS_ENABLED(CONFIG_NVME_FC))
> + return ret;
> +
> cstat = kmalloc((sizeof(struct lpfc_nvme_ctrl_stat) *
> phba->cfg_nvme_io_channel), GFP_KERNEL);
> if (!cstat)
> return -ENOMEM;
>
> - if (!IS_ENABLED(CONFIG_NVME_FC))
> - return ret;
> -
> /* localport is allocated from the stack, but the registration
> * call allocates heap memory as well as the private area.
> */
>
Can you send this as a real patch that I can queue up?
thanks,
greg k-h
