linux-kernel - Re: [PATCH] firmware_loader: Replace kmap() with kmap_local

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: <1731607.VLH7GnMWUR@opensuse>
Date:   Fri, 15 Jul 2022 01:54:15 +0200
From:   "Fabio M. De Francesco" <fmdefrancesco@...il.com>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Luis Chamberlain <mcgrof@...nel.org>
Cc:     Russ Weight <russell.h.weight@...el.com>,
        "Rafael J. Wysocki" <rafael@...nel.org>,
        Nick Terrell <terrelln@...com>, linux-kernel@...r.kernel.org,
        Ira Weiny <ira.weiny@...el.com>
Subject: Re: [PATCH] firmware_loader: Replace kmap() with kmap_local_page()

On giovedì 14 luglio 2022 22:40:32 CEST Luis Chamberlain wrote:
> On Thu, Jul 14, 2022 at 04:50:42PM +0200, Greg Kroah-Hartman wrote:
> > On Mon, Jul 11, 2022 at 08:52:49PM +0200, Fabio M. De Francesco wrote:
> > > 
> > > I see that the outputs of selftests, regardless of running a 5.19.0-
rc6  
> > > kernel with or without my changes, show always the same error:
> > > 
> > > "not ok 1 selftests: firmware: fw_run_tests.sh # TIMEOUT 165 
seconds".".
> > > 
> > > I ran those tests on a QEMU/KVM 32-bits VM, booting a vanilla 5.19.0-
rc6 
> > > kernel with HIGHMEM64GB enabled. 
> > > 
> > > As said, outputs don't change with or without my patch. Instead it 
changes 
> > > with the latest openSUSE stock kernel (5.18.9-2-pae):
> > > 
> > > "ok 1 selftests: firmware: fw_run_tests.sh".
> > > 
> > > Unfortunately, I'm not familiar with kernel selftests. Any ideas 
about what 
> > > could have made this tests fail? Is it expected? 
> > > 
> > > If not, I can try and figure out why these outputs are not what they 
should 
> > > be (the second version of my patch can wait the time it takes).
> > 
> > No idea, sorry, Luis might know more.
> 
> The selftests should have a config file, ensure that stuff is all
> enabled. Can you verify if this stuff was enabled on your config:
> 
> cat tools/testing/selftests/firmware/config 
> CONFIG_TEST_FIRMWARE=y
> CONFIG_FW_LOADER=y
> CONFIG_FW_LOADER_USER_HELPER=y
> CONFIG_IKCONFIG=y
> CONFIG_IKCONFIG_PROC=y
> CONFIG_FW_UPLOAD=y
> 
>   Luis
> 
You've been helpful to tell me how to check that all the necessary options 
were enabled. Actually they were _all_ enabled.

Soon after checking, I started to look at something else...

Being not familiar at all with selftests I had not realized that I was 
using the tests built for an older version of the kernel. 

I made some confusion between cross-compiling for x86-32 in a separate 
directory from sources, installing my custom 5.19.0-rc6 kernel in the VM, 
and running the selftests built for an older 5.18.0-rc[something].

Everything is working perfectly now (with and without my patch)[1]:
"ok 1 selftests: firmware: fw_run_tests.sh".

I'm sorry for my lack of experience :-(

Thank you so much,

Fabio

[1] https://lore.kernel.org/lkml/20220714235030.12732-1-fmdefrancesco@gmail.com/