linux-kernel - Re: [PATCH 17/20] misc: genwqe: card_base: Do not pass unused argument 'fatal

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <235f96392b5a2f06b4410e9532b1322b@linux.vnet.ibm.com>
Date:   Tue, 30 Jun 2020 11:54:24 +0200
From:   haver <haver@...ux.vnet.ibm.com>
To:     Lee Jones <lee.jones@...aro.org>
Cc:     arnd@...db.de, gregkh@...uxfoundation.org,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        Michael Jung <mijung@....net>,
        Michael Ruettger <michael@...ra.de>,
        Frank Haverkamp <haver@...ux.ibm.com>,
        Joerg-Stephan Vogt <jsvogt@...ibm.com>
Subject: Re: [PATCH 17/20] misc: genwqe: card_base: Do not pass unused
 argument 'fatal_err'

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@....net>
>> > Cc: Michael Ruettger <michael@...ra.de>
>> > Cc: Frank Haverkamp <haver@...ux.ibm.com>
>> > Cc: Joerg-Stephan Vogt <jsvogt@...ibm.com>
>> > Signed-off-by: Lee Jones <lee.jones@...aro.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.

Regards

Frank