Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 30 Jun 2020 15:00:38 +0100
From:   Lee Jones <lee.jones@linaro.org>
To:     haver <haver@linux.vnet.ibm.com>
Cc:     arnd@arndb.de, gregkh@linuxfoundation.org,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
        Michael Jung <mijung@gmx.net>,
        Michael Ruettger <michael@ibmra.de>,
        Frank Haverkamp <haver@linux.ibm.com>,
        Joerg-Stephan Vogt <jsvogt@de.ibm.com>
Subject: Re: [PATCH 17/20] misc: genwqe: card_base: Do not pass unused
 argument 'fatal_err'
Message-ID: <20200630140038.GN1179328@dell>
References: <20200629140442.1043957-1-lee.jones@linaro.org>
 <20200629140442.1043957-18-lee.jones@linaro.org>
 <016718c7218cf811897cc32cf9476222@linux.vnet.ibm.com>
 <20200630091020.GJ1179328@dell>
 <235f96392b5a2f06b4410e9532b1322b@linux.vnet.ibm.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <235f96392b5a2f06b4410e9532b1322b@linux.vnet.ibm.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, 30 Jun 2020, haver wrote:

> On 2020-06-30 11:10, Lee Jones wrote:
> > On Tue, 30 Jun 2020, haver wrote:
> > 
> > > On 2020-06-29 16:04, Lee Jones wrote:
> > > > 'fatal_err' is taken as an argument to a static function which is only
> > > > invoked once.  During this invocation 'fatal_err' is not set.  There
> > > > doesn't appear to be a good reason to keep it around.
> > > >
> > > > Also fixes the following W=1 kernel build warning:
> > > >
> > > >  drivers/misc/genwqe/card_base.c:588: warning: Function parameter or
> > > > member 'fatal_err' not described in 'genwqe_recover_card'
> > > >
> > > > Cc: Michael Jung <mijung@gmx.net>
> > > > Cc: Michael Ruettger <michael@ibmra.de>
> > > > Cc: Frank Haverkamp <haver@linux.ibm.com>
> > > > Cc: Joerg-Stephan Vogt <jsvogt@de.ibm.com>
> > > > Signed-off-by: Lee Jones <lee.jones@linaro.org>
> > > > ---
> > > >  drivers/misc/genwqe/card_base.c | 18 +++---------------
> > > >  1 file changed, 3 insertions(+), 15 deletions(-)
> > > >
> > > > diff --git a/drivers/misc/genwqe/card_base.c
> > > > b/drivers/misc/genwqe/card_base.c
> > > > index bceebf49de2d5..809a6f46f6de3 100644
> > > > --- a/drivers/misc/genwqe/card_base.c
> > > > +++ b/drivers/misc/genwqe/card_base.c
> > > > @@ -569,30 +569,18 @@ static int genwqe_stop(struct genwqe_dev *cd)
> > > >
> > > >  /**
> > > >   * genwqe_recover_card() - Try to recover the card if it is possible
> > > > - *
> > > > - * If fatal_err is set no register access is possible anymore. It is
> > > > - * likely that genwqe_start fails in that situation. Proper error
> > > > - * handling is required in this case.
> > > > + * @cd: GenWQE device information
> > > >   *
> > > >   * genwqe_bus_reset() will cause the pci code to call genwqe_remove()
> > > >   * and later genwqe_probe() for all virtual functions.
> > > >   */
> > > > -static int genwqe_recover_card(struct genwqe_dev *cd, int fatal_err)
> > > > +static int genwqe_recover_card(struct genwqe_dev *cd)
> > > >  {
> > > >  	int rc;
> > > >  	struct pci_dev *pci_dev = cd->pci_dev;
> > > >
> > > >  	genwqe_stop(cd);
> > > >
> > > > -	/*
> > > > -	 * Make sure chip is not reloaded to maintain FFDC. Write SLU
> > > > -	 * Reset Register, CPLDReset field to 0.
> > > > -	 */
> > > > -	if (!fatal_err) {
> > > > -		cd->softreset = 0x70ull;
> > > > -		__genwqe_writeq(cd, IO_SLC_CFGREG_SOFTRESET, cd->softreset);
> > > > -	}
> > > > -
> > > >  	rc = genwqe_bus_reset(cd);
> > > >  	if (rc != 0) {
> > > >  		dev_err(&pci_dev->dev,
> > > > @@ -958,7 +946,7 @@ static int genwqe_health_thread(void *data)
> > > >
> > > >  			cd->card_state = GENWQE_CARD_FATAL_ERROR;
> > > >
> > > > -			rc = genwqe_recover_card(cd, 0);
> > > > +			rc = genwqe_recover_card(cd);
> > > >  			if (rc < 0) {
> > > >  				/* FIXME Card is unusable and needs unbind! */
> > > >  				goto fatal_error;
> > > 
> > > I think this one I want to keep. Since fatal_err is 0, !fatal_error
> > > is 1 and
> > > the register write is actually executed.
> > 
> > Ah yes, good catch.
> > 
> > What if we *always* did instead then?
> 
> I knew you would ask that ;-).
> 
> > 
> > > I also want to keep the parameter even though it is only used with
> > > 0. The
> > > register bit causes a less drastic recovery, but if we would
> > > discover that
> > > that is not working ok, we rather discard the debug data we get in
> > > this case
> > > instead of letting the recovery fail. I think it does not hurt to
> > > keep it.
> > 
> > I'm not 100% against it, but it is dead code.
> > 
> > > Maybe you can add a comment?
> > 
> > If you really want to keep it, I can just update the kerneldoc
> > instead.
> 
> I prefer that option. I want to indicate that there are two possible ways to
> recover on a problem. If we delete the currently not exploited parameter,
> this information gets lost.

That's fine.  Will fix.

-- 
Lee Jones [李琼斯]
Senior Technical Lead - Developer Services
Linaro.org │ Open source software for Arm SoCs
Follow Linaro: Facebook | Twitter | Blog