linux-kernel - Re: [PATCH v7 1/3] scanf: remove redundant debug logs

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: <CAJ-ks9=WA0LSzgiLQfH+MZO9JcuAJgTFAL2EWf7UStt1NjG4jg@mail.gmail.com>
Date: Tue, 11 Feb 2025 12:50:17 -0500
From: Tamir Duberstein <tamird@...il.com>
To: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Cc: David Gow <davidgow@...gle.com>, Petr Mladek <pmladek@...e.com>, 
	Steven Rostedt <rostedt@...dmis.org>, Rasmus Villemoes <linux@...musvillemoes.dk>, 
	Sergey Senozhatsky <senozhatsky@...omium.org>, Andrew Morton <akpm@...ux-foundation.org>, 
	Shuah Khan <shuah@...nel.org>, Geert Uytterhoeven <geert@...ux-m68k.org>, linux-kernel@...r.kernel.org, 
	linux-kselftest@...r.kernel.org
Subject: Re: [PATCH v7 1/3] scanf: remove redundant debug logs

On Tue, Feb 11, 2025 at 12:16 PM Andy Shevchenko
<andriy.shevchenko@...ux.intel.com> wrote:
>
> On Tue, Feb 11, 2025 at 11:02:59AM -0500, Tamir Duberstein wrote:
> > On Tue, Feb 11, 2025 at 10:58 AM Andy Shevchenko
> > <andriy.shevchenko@...ux.intel.com> wrote:
> > > On Tue, Feb 11, 2025 at 10:50:33AM -0500, Tamir Duberstein wrote:
> > > > On Tue, Feb 11, 2025 at 10:42 AM Andy Shevchenko
> > > > <andriy.shevchenko@...ux.intel.com> wrote:
> > > > > On Tue, Feb 11, 2025 at 10:13:37AM -0500, Tamir Duberstein wrote:
> > > > > > The test already prints the same information on failure; remove
> > > > > > redundant pr_debug() logs.
>
> ...
>
> > > > > > -     pr_debug("\"%s\", \"%s\" ->\n", str, fmt);                              \
> > > > >
> > > > > What *if* the n_args == 0 here?
> > > >
> > > > Then there's no assertion in this block, so the test cannot possibly fail here.
> > >
> > > Correct, but I'm talking about this in a scope of the removed debug print.
> > > I.o.w. how would we even know that this was the case?
> > >
> > > (I'm not objecting removal, what I want from you is to have a descriptive and
> > >  explanatory commit message that's answers to "why is this needed?" and "why is
> > >  it safe to do?")
> >
> > The true answer to "why is this needed" is Petr requested it in
> > https://lore.kernel.org/all/Z6s2eqh0jkYHntUL@pathway.suse.cz/ (again,
> > lore is having issues):
> >
> > On Tue, Feb 11, 2025 at 6:37 AM Petr Mladek <pmladek@...e.com> wrote:
>
> [...]
>
> > > But when thinking more about it. I think that even pr_debug() is not
> > > the right solution.
> > >
> > > IMHO, we really want to print these details only when the test fails.
> > >
> > > Best Regards,
> > > Petr
> >
> > The commit message already answers "why is it safe to do":
>
> Not really. It answers that "why is it safe to do when test case fails?".
>
> > > The test already prints the same information on failure; remove
> > > redundant pr_debug() logs.
> >
> > Perhaps what you're asking for is an assertion to be added if n_args
> > == 0? I think that would make sense. Does it belong in this series?
>
> I don't know if it's possible case. I don't know if we need an assertion.
> Please, research.

Such an assertion is not necessary. `_check_numbers_template` is
called from `check_{ull,ll,ulong,long,uint,int,ushort,short,uchar,char}`
which are in turn called from `_test`:

> if (ret != n_args) {
>   KUNIT_FAIL(test, "vsscanf(\"%s\", \"%s\", ...) returned %d expected %d", string, fmt, ret, n_args);
> } else {
>   (*fn)(test, check_data, string, fmt, n_args, ap); // <-- `fn` is `check_*`.
> }

So `n_args` comes from the test expectation, and it's already checked
against the return value of `vsscanf`.