This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH v3 00/17] Consolidate Linux sysvipc implementation
- From: Yury Norov <ynorov at caviumnetworks dot com>
- To: Arnd Bergmann <arnd at arndb dot de>
- Cc: <libc-alpha at sourceware dot org>, Steve Ellcey <sellcey at caviumnetworks dot com>, Adhemerval Zanella <adhemerval dot zanella at linaro dot org>
- Date: Fri, 6 Jan 2017 02:27:18 +0530
- Subject: Re: [PATCH v3 00/17] Consolidate Linux sysvipc implementation
- Authentication-results: sourceware.org; auth=none
- Authentication-results: spf=none (sender IP is ) smtp.mailfrom=Yuri dot Norov at caviumnetworks dot com;
- References: <fc666312-8077-f458-7fef-f5e79df0afcd@linaro.org> <1481308253.29811.77.camel@caviumnetworks.com> <10763878.Gf8T87Klc7@wuerfel>
- Spamdiagnosticmetadata: NSPM
- Spamdiagnosticoutput: 1:99
On Fri, Dec 09, 2016 at 09:04:18PM +0100, Arnd Bergmann wrote:
> On Friday, December 9, 2016 10:30:53 AM CET Steve Ellcey wrote:
> >
> > Adhemerval,
> >
> > It looks like you may have to tweek your change to test-skeleton.c in
> > this patch. The FAIL_EXIT macro was moved from test-skeleton.c
> > to support/check.h so I assume your new FAIL_UNSUPPORTED macro should
> > be moved there as well.
> >
> > I have been doing aarch64 and x86 builds here with your patch and have
> > not had any problems except with my new aarch64 ILP32 code. There I
> > had to change the definition of __IPC_64 in the new header
> > file sysdeps/unix/sysv/linux/aarch64/ipc_priv.h from:
> >
> > #define __IPC_64 0x0
> >
> > to:
> >
> > #ifdef __LP64__
> > # define __IPC_64 0x0
> > #else /* __ILP32 */
> > # define __IPC_64 0x100
> > #endif
>
> Huh?
>
> I would hope that this is not really needed, it looks like
> a kernel bug. The old style IPC has been deprecated for so
> long that nothing should need to pass the 0x100 flag, unless
> it dates back to the last century.
>
> Do we need to override the version parsing in the kernel?
Hi Arnd,
It's not completely clear what you do mean here. I think you mean that
function that parses flags should be changed, but Steve noticed that
it may also mean that you want to replace compat sys_msg{ctl,rcv,snd}
in generic unistd,h with non-compat versions. If so, it should be
probably done before taking aarch64/ilp32.
Yury