linux-kernel - Re: [PATCH] scsi: Use vzalloc instead of vmalloc/memset

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] [day] [month] [year] [list]

Message-ID: <alpine.DEB.2.20.1711080741261.1984@hadrien>
Date:   Wed, 8 Nov 2017 07:43:14 +0800 (CST)
From:   Julia Lawall <julia.lawall@...6.fr>
To:     Himanshu Jha <himanshujha199640@...il.com>
cc:     "Luis R. Rodriguez" <mcgrof@...nel.org>,
        anil.gurumurthy@...gic.com, sudarsana.kalluru@...gic.com,
        jejb@...ux.vnet.ibm.com, martin.petersen@...cle.com,
        qla2xxx-upstream@...gic.com, kartilak@...co.com,
        sebaddel@...co.com, linux-scsi@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] scsi: Use vzalloc instead of vmalloc/memset



On Wed, 8 Nov 2017, Himanshu Jha wrote:

> On Tue, Nov 07, 2017 at 08:51:36PM +0100, Luis R. Rodriguez wrote:
> > On Sun, Nov 05, 2017 at 03:26:26AM +0530, Himanshu Jha wrote:
> > > Use vzalloc instead of vmalloc/memset to allocate memory filled with 0
> > > value.
> > >
> > > Done using Coccinelle.
> > > Semantic patch used :
> > >
> > > @@
> > > expression x,a;
> > > statement S;
> > > @@
> > >
> > > - x = vmalloc(a);
> > > + x = vzalloc(a);
> > >   if (x == NULL || ...) S
> > > - memset(x, 0, a);
> >
> > How many false positives do you get? Have you identified any?
> > If not you should consider adding this SmPL rule to:
> >
> > scripts/coccinelle/api/
> >
> > Some maintainers may ask you for the SmPL rule to be upstream first,
> > not all though. So its good practice for you to strive for this.
> > Another reason for it to go upstream is then other maintainers
> > can / should be running coccicheck against their subsystem to avoid
> > stupid regressions.
> >
> > You may want to explain for patches like these that they have been
> > tested by 0-day without any issues found.
> >
> > Also add the tag:
> >
> > Generated-by: Coccinelle SmPL
> >
> > > Signed-off-by: Himanshu Jha <himanshujha199640@...il.com>
> > > ---
> > >  drivers/scsi/bfa/bfad.c            | 3 +--
> > >  drivers/scsi/bfa/bfad_debugfs.c    | 8 ++------
> > >  drivers/scsi/qla2xxx/qla_bsg.c     | 3 +--
> > >  drivers/scsi/qla2xxx/tcm_qla2xxx.c | 5 +----
> > >  drivers/scsi/scsi_debug.c          | 6 ++----
> > >  drivers/scsi/snic/snic_trc.c       | 3 +--
> > >  6 files changed, 8 insertions(+), 20 deletions(-)
> >
> > Split this up per driver, and resend by using ./scripts/get_maintainer.pl
> > foo.patch and ensuring the right folks get the email.  Right now you
> > just spammed tons of people and the changes may be preferred to go
> > upstream atomically per driver, always assume this first.

Depending on the subsystem, you may get similar pushback if you send one
patch per file - "why send so many patches for such a small change when
they are all going through my tree".  So consider grouping the patches by
set of maintainers.

julia

> >
> > Other than this, feel free to add to each of the patches you created:
> >
>
> Thanks for the feeedback! I will resend the patch with the necessary
> changes.
>
> > Acked-by: Luis R. Rodriguez <mcgrof@...nel.org>
>
>
> Thanks
> Himanshu Jha
>