This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
| Other format: | [Raw text] | |
Richard Henderson <rth@redhat.com> writes:
> On Wed, Feb 25, 2004 at 02:14:15PM -0800, Zack Weinberg wrote:
>> I was thinking of the other way round ...
>>
>> static void __attribute((regparm(1))) foo(int x);
>> void bar()
>> {
>> foo(x);
>> }
>> static void foo(int x)
>> {
>> printf("%d", x);
>> }
>
> Granted, that works. If we can reliably detect things in
> this direction (earliest declaration must contain regparm),
> then I'd be in amenable to that.
The question is, can we detect this?
It does work and should not generate an error but currently generates
one and therefore glibc fails to build on x86 platforms where this
construct is used in the dynamic linker.
Andreas
--
Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj
SuSE Linux AG, Maxfeldstr. 5, 90409 Nürnberg, Germany
GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126
Attachment:
pgp00000.pgp
Description: PGP signature
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |