linux-kernel - Re: [PATCH v3 0/4] Add binder state and statistics to binderfs

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]

Date:   Wed, 4 Sep 2019 17:12:13 +0200
From:   Christian Brauner <christian.brauner@...ntu.com>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     Joel Fernandes <joel@...lfernandes.org>,
        Hridya Valsaraju <hridya@...gle.com>,
        "open list:ANDROID DRIVERS" <devel@...verdev.osuosl.org>,
        kernel-team <kernel-team@...roid.com>,
        Todd Kjos <tkjos@...roid.com>,
        LKML <linux-kernel@...r.kernel.org>,
        Arve Hjønnevåg <arve@...roid.com>,
        Martijn Coenen <maco@...roid.com>
Subject: Re: [PATCH v3 0/4] Add binder state and statistics to binderfs

On Wed, Sep 04, 2019 at 04:49:03PM +0200, Greg Kroah-Hartman wrote:
> On Wed, Sep 04, 2019 at 10:20:32AM -0400, Joel Fernandes wrote:
> > On September 4, 2019 7:19:35 AM EDT, Christian Brauner
> > <christian.brauner@...ntu.com> wrote:
> > >On Tue, Sep 03, 2019 at 09:16:51AM -0700, Hridya Valsaraju wrote:
> > >> Currently, the only way to access binder state and
> > >> statistics is through debugfs. We need a way to
> > >> access the same even when debugfs is not mounted.
> > >> These patches add a mount option to make this
> > >> information available in binderfs without affecting
> > >> its presence in debugfs. The following debugfs nodes
> > >> will be made available in a binderfs instance when
> > >> mounted with the mount option 'stats=global' or 'stats=local'.
> > >>
> > >>  /sys/kernel/debug/binder/failed_transaction_log
> > >>  /sys/kernel/debug/binder/proc
> > >>  /sys/kernel/debug/binder/state
> > >>  /sys/kernel/debug/binder/stats
> > >>  /sys/kernel/debug/binder/transaction_log
> > >>  /sys/kernel/debug/binder/transactions
> > >
> > >Acked-by: Christian Brauner <christian.brauner@...ntu.com>
> > >
> > >Btw, I think your counting is off-by-one. :) We usually count the
> > >initial send of a series as 0 and the first rework of that series as
> > >v1.
> > >I think you counted your initial send as v1 and the first rework as v2.
> > 
> > Which is fine. I have done it both ways. Is this a rule written somewhere?
> 
> No where, I can count both ways, it's not a big deal :)

It isn't documented (as many things we still do are) and it's not a big
deal. But most people seem to be counting revisions starting from 0 it
seems. I went looking for previous version to link to in the patch cover
letter and was confused because I was missing a v1. :)

Anyway, I'm happy that Hridya landed this! It was fun helping her the
last couple of weeks on- and off-list. Thanks for getting this done! I
hope we'll see even more patches in the future. :)

Christian