This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] aarch64: Fix ipc_perm definition for ILP32
- From: Szabolcs Nagy <szabolcs dot nagy at arm dot com>
- To: Yury Norov <ynorov at caviumnetworks dot com>
- Cc: nd at arm dot com, sellcey at cavium dot com, libc-alpha <libc-alpha at sourceware dot org>
- Date: Wed, 23 Aug 2017 12:02:19 +0100
- Subject: Re: [PATCH] aarch64: Fix ipc_perm definition for ILP32
- Authentication-results: sourceware.org; auth=none
- Authentication-results: spf=none (sender IP is ) smtp.mailfrom=Szabolcs dot Nagy at arm dot com;
- Nodisclaimer: True
- References: <1503423981.28672.31.camel@cavium.com> <599D50F6.6060009@arm.com> <20170823101023.uvc2ehejstemzxyz@yury-thinkpad> <599D5AEC.7090402@arm.com>
- Spamdiagnosticmetadata: NSPM
- Spamdiagnosticoutput: 1:99
On 23/08/17 11:37, Szabolcs Nagy wrote:
> we have to decide if mode_t is unsigned int or short on ilp32,
> changing just the ipc_perm struct in libc is nonconforming:
> http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/sys_ipc.h.html
>
> (there are some existing conformance issues like that in
> glibc/linux but we should try to avoid introducing new ones)
>
> i think the ilp32 linux uapi should typedef __kernel_mode_t
> to unsigned short, but i don't know the effect of that on
> the kernel, so please discuss this with the kernel folks.
>
hm it seems to me that a mode_t change would be
very intrusive..
can we keep the ipc_perm mode field unsigned int
and do endian fixup/zero pad in the syscall interface?