[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <877dwx3u9y.fsf@x220.int.ebiederm.org>
Date: Wed, 27 May 2020 17:08:57 -0500
From: ebiederm@...ssion.com (Eric W. Biederman)
To: Arnd Bergmann <arnd@...db.de>
Cc: Alexander Viro <viro@...iv.linux.org.uk>,
Kees Cook <keescook@...omium.org>,
Valdis Klētnieks <valdis.kletnieks@...edu>,
Greg Ungerer <gerg@...ux-m68k.org>,
linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] binfmt_elf_fdpic: fix execfd build regression
Arnd Bergmann <arnd@...db.de> writes:
> The change to bprm->have_execfd was incomplete, leading
> to a build failure:
>
> fs/binfmt_elf_fdpic.c: In function 'create_elf_fdpic_tables':
> fs/binfmt_elf_fdpic.c:591:27: error: 'BINPRM_FLAGS_EXECFD' undeclared
>
> Change the last user of BINPRM_FLAGS_EXECFD in a corresponding
> way.
>
> Reported-by: Valdis Klētnieks <valdis.kletnieks@...edu>
> Fixes: b8a61c9e7b4a ("exec: Generic execfd support")
> Signed-off-by: Arnd Bergmann <arnd@...db.de>
> ---
> I have no idea whether this is right, I only looked briefly at
> the commit that introduced the problem.
It is correct and my fault.
Is there an easy to build-test configuration that includes
binfmt_elf_fdpic?
I have this sense that it might be smart to unify binfmt_elf
and binftm_elf_fdpic to the extent possible, and that will take build
tests.
Eric
> ---
> fs/binfmt_elf_fdpic.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
> index bba3ad555b94..aaf332d32326 100644
> --- a/fs/binfmt_elf_fdpic.c
> +++ b/fs/binfmt_elf_fdpic.c
> @@ -588,7 +588,7 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm,
> nitems = 1 + DLINFO_ITEMS + (k_platform ? 1 : 0) +
> (k_base_platform ? 1 : 0) + AT_VECTOR_SIZE_ARCH;
>
> - if (bprm->interp_flags & BINPRM_FLAGS_EXECFD)
> + if (bprm->have_execfd)
> nitems++;
>
> csp = sp;
Powered by blists - more mailing lists