linux-kernel - Re: [PATCH v4 6/6] acpi/ghes: Add a logic to inject ARM processor CPER

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: <87le1ija90.fsf@pond.sub.org>
Date: Wed, 31 Jul 2024 09:23:07 +0200
From: Markus Armbruster <armbru@...hat.com>
To: Mauro Carvalho Chehab <mchehab+huawei@...nel.org>
Cc: Jonathan Cameron <Jonathan.Cameron@...wei.com>,  Shiju Jose
 <shiju.jose@...wei.com>,  "Michael S. Tsirkin" <mst@...hat.com>,  Ani
 Sinha <anisinha@...hat.com>,  Dongjiu Geng <gengdongjiu1@...il.com>,  Eric
 Blake <eblake@...hat.com>,  Igor Mammedov <imammedo@...hat.com>,  Michael
 Roth <michael.roth@....com>,  Paolo Bonzini <pbonzini@...hat.com>,  Peter
 Maydell <peter.maydell@...aro.org>,  <linux-kernel@...r.kernel.org>,
  <qemu-arm@...gnu.org>,  <qemu-devel@...gnu.org>
Subject: Re: [PATCH v4 6/6] acpi/ghes: Add a logic to inject ARM processor CPER

Mauro Carvalho Chehab <mchehab+huawei@...nel.org> writes:

> Em Mon, 29 Jul 2024 17:31:09 +0100
> Jonathan Cameron <Jonathan.Cameron@...wei.com> escreveu:
>
>> On Mon, 29 Jul 2024 15:21:10 +0200
>> Mauro Carvalho Chehab <mchehab+huawei@...nel.org> wrote:
>
> ...
>
>> Markus suggested:
>> 
>> > A target-specific command like this one should be conditional.  Try
>> > this:
>> > 
>> >     { 'command': 'arm-inject-error',
>> >       'data': { 'errortypes': ['ArmProcessorErrorType'] },
>> >       'features': [ 'unstable' ],
>> >       'if': 'TARGET_ARM' }
>> >
>> > No need to provide a qmp_arm_inject_error() stub then.  
>> 
>> (I noticed because never knew you could do this.)
>> 
>> Probably crossed with your v4 posting.
>
> Tried it, but can't figure out how to properly set it up at meson.build,
> as it is basically producing build time errors during qapi file generation
> on non-ARM platforms. For instance:
>
> FAILED: libqemuutil.a.p/meson-generated_.._qapi_qapi-visit-arm-error-inject.c.o 

[...]

> In file included from qapi/qapi-visit-arm-error-inject.h:17,
>                  from qapi/qapi-visit-arm-error-inject.c:15:
> qapi/qapi-types-arm-error-inject.h:18:13: error: attempt to use poisoned "TARGET_ARM"
>    18 | #if defined(TARGET_ARM)
>       |             ^
> In file included from /new_devel/edac/qemu/include/exec/poison.h:7,
>                  from /new_devel/edac/qemu/include/qemu/osdep.h:38,
>                  from qapi/qapi-visit-arm-error-inject.c:13:
> ./config-poison.h:718:20: note: poisoned here
>
> Such error is created by two files generated from qapi, due
> to this change:
>
> 	diff --git a/qapi/meson.build b/qapi/meson.build
> 	index e7bc54e5d047..5927932c4be3 100644
> 	--- a/qapi/meson.build
> 	+++ b/qapi/meson.build
> 	@@ -24,2 +24,3 @@ endif
> 	 qapi_all_modules = [
> 	+  'arm-error-inject',
> 	   'authz',
>
> No idea how to fix it.

Uh, I neglected to point out an important detail.  Sorry about that!

The 'if' condition uses a symbol that is poisoned in target-independent
compiles.  Such conditions work only in target modules.  By convention,
these are named FOO-target.json.

So, you can either change new module's filename to end in -target.json,
or you stick the command into one of the two existing target modules,
machine-target.json and misc-target.json.

Speaking of modules: i'm not sure dedicating a module just to ARM error
injection is a good idea.  Perhaps we could have arm-target.json for
things that will only ever make sense on ARM.  Same for other targets.

But I recommend to first reach consensus on Igor's objection in review
of v3:

    Message-ID: <20240730131709.10e72c7d@...mmedo.users.ipa.redhat.com>
    https://lore.kernel.org/qemu-devel/20240730131709.10e72c7d@imammedo.users.ipa.redhat.com/