[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <Zwjjfpb0hSSE2Ap4@pathway.suse.cz>
Date: Fri, 11 Oct 2024 10:36:14 +0200
From: Petr Mladek <pmladek@...e.com>
To: zhang warden <zhangwarden@...il.com>
Cc: Josh Poimboeuf <jpoimboe@...nel.org>,
Marcos Paulo de Souza <mpdesouza@...e.com>,
Miroslav Benes <mbenes@...e.cz>, Jiri Kosina <jikos@...nel.org>,
Joe Lawrence <joe.lawrence@...hat.com>,
live-patching@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/1] selftests: livepatch: add test cases of stack_order
sysfs interface
On Fri 2024-10-11 09:51:07, zhang warden wrote:
>
>
> > On Oct 10, 2024, at 23:51, Josh Poimboeuf <jpoimboe@...nel.org> wrote:
> >
> > Maybe add a replace=[true|false] module parameter.
> >
>
> How to do it?
> Isn't the way we build modules using make?
> How to set this replace value?
You could find inspiration in
tools/testing/selftests/livepatch/test-livepatch.sh, see
load_lp $MOD_REPLACE replace=0
vs.
load_lp $MOD_REPLACE replace=1
You could more or less copy the "multiple livepatches" and
"atomic replace livepatch" self-tests. The new tests
would check the "stack_order" sysfs value instread of checking
whether /proc/cmdline and /proc/meminfo are livepatched.
Best Regards,
Petr
Powered by blists - more mailing lists