9:01 p.m.
Without this fairly bogus patch, I cannot compile 21.4 on Cygwin (at
least using gcc-3.3.3).
Comments? Recommendations?
- Vin
src/ChangeLog addition:
2005-01-24 Vin Shelton <acs(a)xemacs.org>
* sheap.c (more_static_core): Reformatted printf string so
gcc-3.3.3 will compile this on Cygwin.
21.4 source patch:
Diff command: cvs -q diff -u
Files affected: src/sheap.c
Index: src/sheap.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/sheap.c,v
retrieving revision 1.12
diff -a -u -u -r1.12 sheap.c
--- src/sheap.c 2001/04/12 18:24:19 1.12
+++ src/sheap.c 2005/01/24 20:58:38
＠＠ -82,17 +82,8 ＠＠
{
printf(
-"\nRequested %d bytes, static heap exhausted! base is %p, current ptr
-is %p. You have exhausted the static heap.
-
-If you are simply trying to compile, remove sheap-adjust.h
-and recompile from the top level. If this doesn't
-work then STATIC_HEAP_SLOP (defined in this file) is too small.
-
-If you want to run temacs, change SHEAP_ADJUSTMENT in sheap-adjust.h
-to 0 or a +ve number. Generally you should *not* try to run temacs
-with a static heap, you should dump first.\n", size,
-static_heap_base, static_heap_ptr);
+"\nRequested %d bytes, static heap exhausted! base is %p, current ptr\nis %p. You
have exhausted the static heap.\n\nIf you are simply trying to compile, remove
sheap-adjust.h\nand recompile from the top level. If this doesn't\nwork then
STATIC_HEAP_SLOP (defined in this file) is too small.\n\nIf you want to run temacs, change
SHEAP_ADJUSTMENT in sheap-adjust.h\nto 0 or a +ve number. Generally you should *not* try
to run temacs\nwith a static heap, you should dump first.\n",
+size, static_heap_base, static_heap_ptr);
exit(-1);
return 0;
9:10 p.m.
Vin Shelton <acs(a)xemacs.org> wrot:
Without this fairly bogus patch, I cannot compile 21.4 on Cygwin (at
least using gcc-3.3.3).
Comments? Recommendations?
It isn't bogus. Embedded newlines in strings was a GCC extension (i.e.,
it is NOT standard C). Support for embedded newlines was removed in gcc
3.3. Your patch is both correct and necessary.
To improve readability, though, you can use line continuations, like
this:
"\nRequested %d bytes, static heap exhausted! base is %p, current ptr\n\
is %p. You have exhausted the static heap.\n\
\n\
If you are simply trying to compile, remove sheap-adjust.h\n\
and recompile from the top level. If this doesn't\n\
work then STATIC_HEAP_SLOP (defined in this file) is too small.\n\
\n\
If you want to run temacs, change SHEAP_ADJUSTMENT in sheap-adjust.h\n\
to 0 or a +ve number. Generally you should *not* try to run temacs\n\
with a static heap, you should dump first.\n", size,
static_heap_base, static_heap_ptr);
--
Jerry James
http://www.ittc.ku.edu/~james/
1:11 a.m.
New subject: [AC21.4] sheap.c printf string patch for gcc-3.3.3 on Cygwin
Vin Shelton <acs(a)xemacs.org> writes:
Without this fairly bogus patch, I cannot compile 21.4 on Cygwin (at
least using gcc-3.3.3).
Comments? Recommendations?
- Vin
src/ChangeLog addition:
2005-01-24 Vin Shelton <acs(a)xemacs.org>
* sheap.c (more_static_core): Reformatted printf string so
gcc-3.3.3 will compile this on Cygwin.
21.4 source patch:
Diff command: cvs -q diff -u
Files affected: src/sheap.c
Index: src/sheap.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/sheap.c,v
retrieving revision 1.12
diff -a -u -u -r1.12 sheap.c
--- src/sheap.c 2001/04/12 18:24:19 1.12
+++ src/sheap.c 2005/01/24 20:58:38
＠＠ -82,17 +82,8 ＠＠
{
printf(
-"\nRequested %d bytes, static heap exhausted! base is %p, current ptr
-is %p. You have exhausted the static heap.
-
-If you are simply trying to compile, remove sheap-adjust.h
-and recompile from the top level. If this doesn't
-work then STATIC_HEAP_SLOP (defined in this file) is too small.
-
-If you want to run temacs, change SHEAP_ADJUSTMENT in sheap-adjust.h
-to 0 or a +ve number. Generally you should *not* try to run temacs
-with a static heap, you should dump first.\n", size,
-static_heap_base, static_heap_ptr);
+"\nRequested %d bytes, static heap exhausted! base is %p, current ptr\nis %p. You
have exhausted the static heap.\n\nIf you are simply trying to compile, remove
sheap-adjust.h\nand recompile from the top level. If this doesn't\nwork then
STATIC_HEAP_SLOP (defined in this file) is too small.\n\nIf you want to run temacs, change
SHEAP_ADJUSTMENT in sheap-adjust.h\nto 0 or a +ve number. Generally you should *not* try
to run temacs\nwith a static heap, you should dump first.\n",
+size, static_heap_base, static_heap_ptr);
exit(-1);
return 0;
APPROVE COMMIT 21.4
I actually committed the James variation of the patch. Patch will
appear in XEmacs 21.4.17, "Jumbo Shrimp".
- Vin
5:06 a.m.
New subject: [AC21.4] sheap.c printf string patch for gcc-3.3.3 on Cygwin
-----Original Message-----
From: xemacs-patches-bounces(a)xemacs.org
[mailto:xemacs-patches-bounces＠xemacs.org] On Behalf Of Vin Shelton
Sent: Monday, January 24, 2005 7:11 PM
To: xemacs-patches(a)xemacs.org
Cc: Dr. Volker Zell
Subject: [AC21.4] sheap.c printf string patch for gcc-3.3.3 on Cygwin
Vin Shelton <acs(a)xemacs.org> writes:
> Without this fairly bogus patch, I cannot compile 21.4 on
Cygwin (at
> least using gcc-3.3.3).
>
> Comments? Recommendations?
>
> - Vin
>
>
> src/ChangeLog addition:
>
> 2005-01-24 Vin Shelton <acs(a)xemacs.org>
>
> * sheap.c (more_static_core): Reformatted printf string so
> gcc-3.3.3 will compile this on Cygwin.
>
>
> 21.4 source patch:
> Diff command: cvs -q diff -u
> Files affected: src/sheap.c
>
> Index: src/sheap.c
> ===================================================================
> RCS file: /pack/xemacscvs/XEmacs/xemacs/src/sheap.c,v
> retrieving revision 1.12
> diff -a -u -u -r1.12 sheap.c
> --- src/sheap.c 2001/04/12 18:24:19 1.12
> +++ src/sheap.c 2005/01/24 20:58:38
> ＠＠ -82,17 +82,8 ＠＠
> {
> printf(
>
> -"\nRequested %d bytes, static heap exhausted! base is %p, current
> ptr -is %p. You have exhausted the static heap.
> -
> -If you are simply trying to compile, remove sheap-adjust.h -and
> recompile from the top level. If this doesn't -work then
> STATIC_HEAP_SLOP (defined in this file) is too small.
> -
> -If you want to run temacs, change SHEAP_ADJUSTMENT in
sheap-adjust.h
> -to 0 or a +ve number. Generally you should *not* try to run temacs
> -with a static heap, you should dump first.\n", size,
> -static_heap_base, static_heap_ptr);
> +"\nRequested %d bytes, static heap exhausted! base is %p, current
> +ptr\nis %p. You have exhausted the static heap.\n\nIf you
are simply
> +trying to compile, remove sheap-adjust.h\nand recompile
from the top
> +level. If this doesn't\nwork then STATIC_HEAP_SLOP
(defined in this
> +file) is too small.\n\nIf you want to run temacs, change
> +SHEAP_ADJUSTMENT in sheap-adjust.h\nto 0 or a +ve number.
Generally
> +you should *not* try to run temacs\nwith a static heap, you should
> +dump first.\n", size, static_heap_base, static_heap_ptr);
>
> exit(-1);
> return 0;
APPROVE COMMIT 21.4
I actually committed the James variation of the patch. Patch
will appear in XEmacs 21.4.17, "Jumbo Shrimp".
- Vin
Vin -- it would be better to commit the version that I sent, not the version
that you have currently. Using multiple "" strings is the ANSI-correct way
to do things.
1:34 a.m.
New subject: [AC21.4] sheap.c printf string patch for gcc-3.3.3 on Cygwin
"Ben Wing" <ben(a)666.com> writes:
> APPROVE COMMIT 21.4
>
> I actually committed the James variation of the patch. Patch
> will appear in XEmacs 21.4.17, "Jumbo Shrimp".
>
> - Vin
>
Vin -- it would be better to commit the version that I sent, not the version
that you have currently. Using multiple "" strings is the ANSI-correct way
to do things.
I disagree - the patch I committed is perfectly valid and could
potentially be accepted by more compilers.
- Vin
12:14 a.m.
New subject: [AC21.4] sheap.c printf string patch for gcc-3.3.3 on Cygwin
-----Original Message-----
From: Vin Shelton [mailto:acs＠xemacs.org]
Sent: Wednesday, January 26, 2005 7:35 PM
To: Ben Wing
Cc: xemacs-patches(a)xemacs.org; 'Dr. Volker Zell'
Subject: Re: [AC21.4] sheap.c printf string patch for
gcc-3.3.3 on Cygwin
"Ben Wing" <ben(a)666.com> writes:
>> APPROVE COMMIT 21.4
>>
>> I actually committed the James variation of the patch. Patch
>> will appear in XEmacs 21.4.17, "Jumbo Shrimp".
>>
>> - Vin
>>
>
> Vin -- it would be better to commit the version that I
sent, not the
> version that you have currently. Using multiple "" strings is the
> ANSI-correct way to do things.
I disagree - the patch I committed is perfectly valid and
could potentially be accepted by more compilers.
Only if they are non-ANSI, which we don't support. In the past we've had
problems with compilers choking on long strings with embedded backslashes,
so "potentially" it could go the other way as well.
3:59 a.m.
New subject: [AC21.4] sheap.c printf string patch for gcc-3.3.3 on Cygwin
I'm turning into a standards geek. Shoot me now.
"Ben Wing" <ben(a)666.com> wrote:
> -----Original Message-----
> From: Vin Shelton [mailto:acs＠xemacs.org]
> Sent: Wednesday, January 26, 2005 7:35 PM
> To: Ben Wing
> Cc: xemacs-patches(a)xemacs.org; 'Dr. Volker Zell'
> Subject: Re: [AC21.4] sheap.c printf string patch for
> gcc-3.3.3 on Cygwin
>
>
> "Ben Wing" <ben(a)666.com> writes:
>
>
> >> APPROVE COMMIT 21.4
> >>
> >> I actually committed the James variation of the patch. Patch
> >> will appear in XEmacs 21.4.17, "Jumbo Shrimp".
> >>
> >> - Vin
> >>
> >
> > Vin -- it would be better to commit the version that I
> sent, not the
> > version that you have currently. Using multiple "" strings is the
> > ANSI-correct way to do things.
>
> I disagree - the patch I committed is perfectly valid and
> could potentially be accepted by more compilers.
Only if they are non-ANSI, which we don't support. In the past we've had
problems with compilers choking on long strings with embedded backslashes,
so "potentially" it could go the other way as well.
Do you happen to remember which compilers had trouble with this
construct? I don't know that it really matters, since we certainly want
to avoid constructs that trigger compiler bugs when can reasonably do
so. I'm just curious.
As far as ANSI goes, though, section 5.1.1.2 of the draft standard [1]
(I'm too cheap to buy the final version) says that the preprocessor is
to delete every backslash-newline pair before the compiler proper even
sees the code. Also, the C99 Rationale [2] in section 6.4.5, String
literals (page 55), says:
A string can be continued across multiple lines by using the
backslash-newline line continuation, but this requires that the
continuation of the string start in the first position of the next
line. To permit more flexible layout, and to solve some preprocessing
problems (see paragraph 6.10.3), the C89 Committee introduced string
literal concatenation. Two string literals in a row are pasted
together, with no null character in the middle, to make one combined
string literal. This addition to the C language allows a programmer
to extend a string literal beyond the end of a physical line without
having to use the backslash-newline mechanism and thereby destroying
the indentation scheme of the program.
So string pasting is the preferred method, but backslash-newline is also
allowed by the standard (along with its drawbacks). Since I'm the
trouble-maker that suggested backslash-newline in the first place, let
me say that I've changed my mind. Vin, we ought to do it Ben's way.
Footnotes:
[1] http://www.open-std.org/jtc1/sc22/wg14/www/docs/n843.htm
[2] http://www.open-std.org/jtc1/sc22/wg14/www/C99RationaleV5.10.pdf
--
Jerry James
http://www.ittc.ku.edu/~james/
7022
days inactive
7026
days old
6 comments
3 participants
participants (3)
-
Ben Wing -
Jerry James -
Vin Shelton