* Re: longlong.h: __udiv_qrnnd
@ 2000-04-12 14:02 Richard Kenner
2000-04-12 14:07 ` Richard Henderson
0 siblings, 1 reply; 6+ messages in thread
From: Richard Kenner @ 2000-04-12 14:02 UTC (permalink / raw)
To: rth; +Cc: gcc
*shrug* Does it matter? AFAIK we don't use anything from
longlong.h in gcc on Alpha. That gets used in like libgcc2's
__divdi3 etc, which never gets used.
Interesting you asked. ;-)
bitsizetype on Alpha is 128 bits, so we can get TImode arithmetic. I
added the proper #define's in longlong.h to make that be divti3 and
such for 64-bit hosts. Then I run into the above problem.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: longlong.h: __udiv_qrnnd
2000-04-12 14:02 longlong.h: __udiv_qrnnd Richard Kenner
@ 2000-04-12 14:07 ` Richard Henderson
0 siblings, 0 replies; 6+ messages in thread
From: Richard Henderson @ 2000-04-12 14:07 UTC (permalink / raw)
To: Richard Kenner; +Cc: gcc
On Wed, Apr 12, 2000 at 05:02:56PM -0400, Richard Kenner wrote:
> bitsizetype on Alpha is 128 bits, so we can get TImode arithmetic. I
> added the proper #define's in longlong.h to make that be divti3 and
> such for 64-bit hosts. Then I run into the above problem.
Ah. Well, perhaps we should add the appropriate routine to libgcc1?
Here is a copy for reference. It's been macro-ized for glibc, but
it's obvious how to rewrite the headers.
r~
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: longlong.h: __udiv_qrnnd
2000-04-12 13:54 Richard Kenner
@ 2000-04-12 13:57 ` Richard Henderson
0 siblings, 0 replies; 6+ messages in thread
From: Richard Henderson @ 2000-04-12 13:57 UTC (permalink / raw)
To: Richard Kenner; +Cc: gcc
On Wed, Apr 12, 2000 at 04:54:23PM -0400, Richard Kenner wrote:
> So, since we can't assume the user has glibc, we can't use that in
> GCC's longlong.h, right?
*shrug* Does it matter? AFAIK we don't use anything from
longlong.h in gcc on Alpha. That gets used in like libgcc2's
__divdi3 etc, which never gets used.
r~
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: longlong.h: __udiv_qrnnd
@ 2000-04-12 13:54 Richard Kenner
2000-04-12 13:57 ` Richard Henderson
0 siblings, 1 reply; 6+ messages in thread
From: Richard Kenner @ 2000-04-12 13:54 UTC (permalink / raw)
To: rth; +Cc: gcc
It's in glibc, and it was bundled with gmp, from whence the
current version of longlong.h is derived.
So, since we can't assume the user has glibc, we can't use that in
GCC's longlong.h, right?
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: longlong.h: __udiv_qrnnd
2000-04-12 5:33 Richard Kenner
@ 2000-04-12 13:47 ` Richard Henderson
0 siblings, 0 replies; 6+ messages in thread
From: Richard Henderson @ 2000-04-12 13:47 UTC (permalink / raw)
To: Richard Kenner; +Cc: gcc
On Wed, Apr 12, 2000 at 08:33:20AM -0400, Richard Kenner wrote:
> This function is defined for Alpha under some cases and there seems to
> be an assumption there's a library function of that name. I don't
> have it on my Alpha. Does anybody know what this is about?
It's in glibc, and it was bundled with gmp, from whence the
current version of longlong.h is derived.
r~
^ permalink raw reply [flat|nested] 6+ messages in thread
* longlong.h: __udiv_qrnnd
@ 2000-04-12 5:33 Richard Kenner
2000-04-12 13:47 ` Richard Henderson
0 siblings, 1 reply; 6+ messages in thread
From: Richard Kenner @ 2000-04-12 5:33 UTC (permalink / raw)
To: gcc
This function is defined for Alpha under some cases and there seems to
be an assumption there's a library function of that name. I don't
have it on my Alpha. Does anybody know what this is about?
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2000-04-12 14:07 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2000-04-12 14:02 longlong.h: __udiv_qrnnd Richard Kenner
2000-04-12 14:07 ` Richard Henderson
-- strict thread matches above, loose matches on Subject: below --
2000-04-12 13:54 Richard Kenner
2000-04-12 13:57 ` Richard Henderson
2000-04-12 5:33 Richard Kenner
2000-04-12 13:47 ` Richard Henderson
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).