This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [RFC 2.0] Implementing hwcap2
- From: David Miller <davem at davemloft dot net>
- To: munroesj at us dot ibm dot com, munroesj at linux dot vnet dot ibm dot com
- Cc: rth at twiddle dot net, roland at hack dot frob dot com, rsa at us dot ibm dot com, libc-alpha at sourceware dot org
- Date: Tue, 09 Apr 2013 12:44:07 -0400 (EDT)
- Subject: Re: [RFC 2.0] Implementing hwcap2
- References: <20130408220507 dot 5699D2C088 at topped-with-meat dot com> <5163FA57 dot 5020506 at twiddle dot net> <1365518878 dot 30724 dot 57 dot camel at spokane1 dot rchland dot ibm dot com>
From: Steven Munroe <munroesj@linux.vnet.ibm.com>
Date: Tue, 09 Apr 2013 09:47:58 -0500
> On Tue, 2013-04-09 at 06:24 -0500, Richard Henderson wrote:
>> On 2013-04-08 17:05, Roland McGrath wrote:
>> > The calling convention for IFUNC resolvers is machine-specific.
>> > Thus, it's up to each machine maintainer to decide what it should be
>> > for his machine. Given how rare IFUNC use still is, I'd say it's
>> > also up to each machine maintainer to decide whether or not an
>> > incompatible change to the convention is acceptable at this stage.
>>
>> While I suppose that's a valid position, my position is that merely
>> adding a second 32-bit argument to the ifunc resolver maintains
>> compatability across all machines, and that is good for everyone.
>
> I agree. there exist several cases where we pass more parameters (main()
> and the auxv for example) then most programs use or care about. It is a
> simple solution that works for every one and does not harm to any
> application that only uses the initial parameter set.
I'm good with this kind of approach too.