linux-kernel - Re: [PATCH] perf/x86/amd/uncore: fix error codes in amd_uncore

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: <ZSpZRM4ZEnNy98Rz@gmail.com>
Date:   Sat, 14 Oct 2023 11:03:00 +0200
From:   Ingo Molnar <mingo@...nel.org>
To:     Uros Bizjak <ubizjak@...il.com>
Cc:     Dan Carpenter <dan.carpenter@...aro.org>,
        Sandipan Das <sandipan.das@....com>,
        Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...hat.com>,
        Arnaldo Carvalho de Melo <acme@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Jiri Olsa <jolsa@...nel.org>,
        Namhyung Kim <namhyung@...nel.org>,
        Ian Rogers <irogers@...gle.com>,
        Adrian Hunter <adrian.hunter@...el.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Borislav Petkov <bp@...en8.de>,
        Dave Hansen <dave.hansen@...ux.intel.com>, x86@...nel.org,
        "H. Peter Anvin" <hpa@...or.com>, linux-perf-users@...r.kernel.org,
        linux-kernel@...r.kernel.org, kernel-janitors@...r.kernel.org
Subject: Re: [PATCH] perf/x86/amd/uncore: fix error codes in amd_uncore_init()


* Uros Bizjak <ubizjak@...il.com> wrote:

> On Fri, Oct 13, 2023 at 11:06 AM Ingo Molnar <mingo@...nel.org> wrote:
> >
> >
> > * Dan Carpenter <dan.carpenter@...aro.org> wrote:
> >
> > > On Fri, Oct 13, 2023 at 09:30:46AM +0200, Ingo Molnar wrote:
> > >
> > > > Ugh, why on Earth didn't GCC warn about this? The bad pattern is pretty
> > > > simple & obvious once pointed out ... compilers should have no trouble
> > > > realizing that 'ret' is returned uninitialized in some of these control
> > > > paths. Yet not a peep from the compiler ...
> > >
> > > We disabled that warning years ago (5?) because GCC had too many false
> > > positives.
> >
> > GCC had some pretty bogus notions about 'possible' uninitialized use that
> > encouraged some bad code patterns, but in this case there's readily
> > provable uninitialized use, that a compiler should warn about.
> >
> > Is it possible to disable just the unreliable, probabilistic part of GCC's
> > uninitialized variables warnings?
> 
> -Wno-maybe-uninitialized?

No combination of the relevant compiler options appears to be able to get 
GCC to notice this bug.

On top of tip:master, the patch below produces no compiler warnings with 
GCC 12.3.0:

  $ git revert 7543365739a4
  $ rm -f arch/x86/events/amd/uncore.o
  $ make V=1 W=1e arch/x86/events/amd/uncore.o

The "W=1e" incantation activates, with the patch below applied, among many 
other GCC options, the following options:

  -Wall
  -Wuninitialized
  -Wmaybe-uninitialized
  -Werror 

Which should have found this bug, right?

[ The V=1 helps double checking the compiler options. ]

Thanks,

	Ingo

 scripts/Makefile.extrawarn | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/scripts/Makefile.extrawarn b/scripts/Makefile.extrawarn
index 2fe6f2828d37..9d245fcff419 100644
--- a/scripts/Makefile.extrawarn
+++ b/scripts/Makefile.extrawarn
@@ -108,6 +108,8 @@ KBUILD_CFLAGS += $(call cc-option, -Wformat-overflow)
 KBUILD_CFLAGS += $(call cc-option, -Wformat-truncation)
 KBUILD_CFLAGS += $(call cc-option, -Wstringop-overflow)
 KBUILD_CFLAGS += $(call cc-option, -Wstringop-truncation)
+KBUILD_CFLAGS += $(call cc-option, -Wuninitialized)
+KBUILD_CFLAGS += $(call cc-option, -Wmaybe-uninitialized)
 
 KBUILD_CPPFLAGS += -Wundef
 KBUILD_CPPFLAGS += -DKBUILD_EXTRA_WARN1
@@ -176,7 +178,7 @@ KBUILD_CFLAGS += -Wno-shift-negative-value
 ifdef CONFIG_CC_IS_CLANG
 KBUILD_CFLAGS += -Wno-initializer-overrides
 else
-KBUILD_CFLAGS += -Wno-maybe-uninitialized
+#KBUILD_CFLAGS += -Wno-maybe-uninitialized
 endif
 
 endif