headers
Mike Stump | 1 Nov 2002 01:00
Picon
Favicon

Re: [basic-improvements] VxWorks 0/n: purge existing

On Monday, October 28, 2002, at 09:09 PM, Zack Weinberg wrote:
> This is the first of a series of patches which will eventually bring
> GCC 3.x's support for VxWorks up to snuff.  It deletes all the
> existing VxWorks support.

:-)

Bye bye, so long, I hate to say good bye.  [ sing that ]
Permalink | Reply | 0 comments |
headers
Richard Kenner | 1 Nov 2002 01:05
Picon

Re: Patch: stab info for const fields

    If you look again at the bit you snipped, the implementation I
    suggested should be totally language-independent (in the
    language-independent parts of the compiler).

Yes, the *implementation* is, but the *definition* is not, so that when
people are looking at a tree, they can't know whether it's valid or not
without knowlege of the language in question and I think that's a bad idea.
Permalink | Reply | 0 comments |
headers
Zack Weinberg | 1 Nov 2002 01:12

Re: [basic-improvements] Remove __gthread_key_dtor

On Thu, Oct 31, 2002 at 02:31:21PM -0800, Eric Christopher wrote:
> 
> re mutexes: Sony only has mutexes and not a real thread system at all.
> So, we get to "invent" things via locking and mutexes.

Understood.  In fact, looking over your code, I just wrote some very
similar stuff for VxWorks (which has threads, but neither once_t nor
thread-specific data as POSIX defines it).

> Now, I may have a problem with my implementation based on rth's email,
> but here's the dtor function as it currently exists:
> 
> static inline int
> __gthread_key_dtor (__gthread_key_t key, void *ptr)
> {
>   /* Just reset the key value to zero.  */
>   if (ptr)
>     return __gthread_setspecific (key, 0);
>   else
>     return 0;
> }

This implementation is identical with several of the ones I calmly
wiped out of existence as "does nothing useful".  The thread
destruction logic is (should be) responsible for doing this.

I see a number of problems with your fake TSD implementation, but
nothing that requires you to have a __gthread_key_dtor routine.  See
below.
(Continue reading)

Permalink | Reply | 3 comments |
headers
Gabriel Dos Reis | 1 Nov 2002 01:21

Re: Patch: stab info for const fields

kenner <at> vlsi1.ultra.nyu.edu (Richard Kenner) writes:

|     If you look again at the bit you snipped, the implementation I
|     suggested should be totally language-independent (in the
|     language-independent parts of the compiler).
| 
| Yes, the *implementation* is, but the *definition* is not, so that when
| people are looking at a tree, they can't know whether it's valid or not
| without knowlege of the language in question and I think that's a bad idea.

I strongly agree with Kenner on this point.  There ought to be a
precise definition.  I support option (1) and won't oppose to (2).
But (3) or (3a) aren't workable, IMO.

-- Gaby
Permalink | Reply | 0 comments |
headers
Richard Henderson | 1 Nov 2002 02:06
Picon
Favicon

Re: Using SCRATCH Register for a mov op,

On Wed, Oct 30, 2002 at 05:45:57PM +0100, Pierre Mallard wrote:
> I'd like to use scratch reg for operation such as 
> reg X = mem(reg X) generated during the RELOAD PASS.
> with X a number of a hard reg same for both side of
> expression

You'll have to use a PARALLEL with a (clobber (match_scratch)).

r~
Permalink | Reply | 0 comments |
headers
David O'Brien | 1 Nov 2002 02:07
Picon
Favicon

[PATCH] reformatting patch for FreeBSD/sparc64

I find this version easier to read and would like to commit it.

2002-10-31  David O'Brien  <obrien <at> FreeBSD.org>

	* config/sparc/freebsd.h (TRANSFER_FROM_TRAMPOLINE): Reformat.

Index: config/sparc/freebsd.h
===================================================================
RCS file: /cvs/gcc/egcs/gcc/config/sparc/freebsd.h,v
retrieving revision 1.13
diff -u -r1.13 freebsd.h
--- config/sparc/freebsd.h	31 Oct 2002 09:08:11 -0000	1.13
+++ config/sparc/freebsd.h	1 Nov 2002 01:06:32 -0000
 <at>  <at>  -97,35 +97,29  <at>  <at> 
 #define SPARC_DEFAULT_CMODEL	CM_MEDLOW

 #define TRANSFER_FROM_TRAMPOLINE					\
-static int need_enable_exec_stack;					\
+  static int need_enable_exec_stack;					\
+  static void check_enabling(void) __attribute__ ((constructor));	\
+  static void check_enabling(void)					\
+  {									\
+    extern int sysctlbyname(const char *, void *, size_t *, void *, size_t);\
+    int prot = 0;							\
+    size_t len = sizeof(prot);						\
 									\
-static void check_enabling(void) __attribute__ ((constructor));		\
-static void check_enabling(void)					\
-{									\
-  extern int sysctlbyname(const char *, void *, size_t *, void *, size_t);\
(Continue reading)

Permalink | Reply | 0 comments |
headers
Richard Henderson | 1 Nov 2002 02:34
Picon
Favicon

Re: apsi bug

On Thu, Oct 31, 2002 at 06:25:31PM +0100, Jan Hubicka wrote:
>         if (optimize
>   	  && FLOAT_MODE_P (GET_MODE (x))
> + 	  && !VECTOR_MODE_P (GET_MODE (x))

We should have 

#define SCALAR_FLOAT_MODE_P(MODE) 
  (GET_MODE_CLASS (MODE) == MODE_FLOAT)

and use that.

r~
Permalink | Reply | 1 comment |
headers
Richard Henderson | 1 Nov 2002 02:39
Picon
Favicon

Re: IA64 Inline division patch

On Thu, Oct 31, 2002 at 12:58:45PM -0800, Steve Ellcey wrote:
> 	* config/ia64/ia64.h (MASK_INLINE_DIV_LAT): Remove.
> 	(MASK_INLINE_DIV_THR): Remove.
> 	(TARGET_INLINE_DIV_LAT): Remove.
> 	(TARGET_INLINE_DIV_THR): Remove.
> 	(TARGET_INLINE_DIV): Remove.
> 	(MASK_INLINE_FLOAT_DIV_LAT): New macro.
> 	(MASK_INLINE_FLOAT_DIV_THR): New macro.
> 	(MASK_INLINE_INT_DIV_LAT): New macro.
> 	(MASK_INLINE_INT_DIV_THR): New macro.
> 	(TARGET_INLINE_FLOAT_DIV_LAT): New macro.
> 	(TARGET_INLINE_FLOAT_DIV_THR): New macro.
> 	(TARGET_INLINE_INT_DIV_LAT): New macro.
> 	(TARGET_INLINE_INT_DIV_THR): New macro.
> 	(TARGET_INLINE_FLOAT_DIV): New macro.
> 	(TARGET_INLINE_INT_DIV): New macro.
> 	* config/ia64/ia64.md (divsi3): Change to use new macros.
> 	(modsi3): Ditto.
> 	(udivsi3): Ditto.
> 	(umodsi3): Ditto.
> 	(divsi3_internal): Ditto.
> 	(divdi3): Ditto.
> 	(moddi3): Ditto.
> 	(udivdi3): Ditto.
> 	(umoddi3): Ditto.
> 	(divdi3_internal_lat): Ditto.
> 	(divdi3_internal_thr): Ditto.
> 	(divsf3): Ditto.
> 	(divsf3_internal_lat): Ditto.
> 	(divsf3_internal_thr): Ditto.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Richard Henderson | 1 Nov 2002 02:41
Picon
Favicon

Re: Patch: fix SPEC regressions

On Thu, Oct 31, 2002 at 03:22:19PM -0800, Dale Johannesen wrote:
> I'll rewrite it simplified if you prefer.

Thanks.

> Are you comfortable with the algorithm?

Yes.  

r~
Permalink | Reply | 2 comments |
headers
David O'Brien | 1 Nov 2002 03:26
Picon
Favicon

[PATCH] fix Emacs building on FreeBSD/sparc64

I'd like to commit this to both mainline and 3.2-branch.

2002-10-31  David O'Brien  <obrien <at> FreeBSD.org>

	* config/sparc/freebsd.h (CPP_CPU64_DEFAULT_SPEC): Define __arch64__.
	Add comment.

Index: freebsd.h
===================================================================
RCS file: /cvs/gcc/egcs/gcc/config/sparc/freebsd.h,v
retrieving revision 1.13
diff -u -r1.13 freebsd.h
--- freebsd.h	31 Oct 2002 09:08:11 -0000	1.13
+++ freebsd.h	1 Nov 2002 02:25:11 -0000
 <at>  <at>  -18,11 +18,13  <at>  <at> 
 along with GNU CC; see the file COPYING.  If not, write to
 the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.  */

-/* FreeBSD needs's the platform name (sparc64) defined.  */
+/* FreeBSD needs's the platform name (sparc64) defined.
+   Emacs needs to know if the arch is 64 or 32-bits.  */

 #undef  CPP_CPU64_DEFAULT_SPEC
-#define CPP_CPU64_DEFAULT_SPEC "-D__sparc64__ -D__sparc_v9__"
+#define CPP_CPU64_DEFAULT_SPEC "-D__sparc64__ -D__sparc_v9__ -D__arch64__"

+/* Because we include sparc/sysv4.h.  */
 #undef  CPP_PREDEFINES
 #define CPP_PREDEFINES FBSD_CPP_PREDEFINES
(Continue reading)

Permalink | Reply | 1 comment |

Gmane