* segfault in guile 1.6.7
@ 2005-03-24 17:11 Alexey Voinov
2005-03-25 20:37 ` Kevin Ryde
2005-05-22 18:59 ` Marius Vollmer
0 siblings, 2 replies; 7+ messages in thread
From: Alexey Voinov @ 2005-03-24 17:11 UTC (permalink / raw)
[-- Attachment #1.1.1: Type: text/plain, Size: 463 bytes --]
hi.
There's a bug in guile srfi-13 implementation.
Steps to reproduce are:
$ guile
guile> (use-modules (srfi srfi-13))
guile> (string-index (string #\a) char-numeric?)
#f
guile> (string-index (string (integer->char 209)) char-numeric?)
zsh: 2760 segmentation fault guile
I haven't checked if this bug is still present in cvs HEAD.
Patch attached.
--
Best Regards!
Alexey Voinov
voins@voins.program.ru
voins@altlinux.ru
[-- Attachment #1.1.2: guile-1.6.7-alt-unsigned.patch --]
[-- Type: text/plain, Size: 7439 bytes --]
--- guile-1.6.7/srfi/srfi-13.c.orig 2004-12-14 08:21:39 +0300
+++ guile-1.6.7/srfi/srfi-13.c 2005-03-23 21:57:37 +0300
@@ -61,7 +61,7 @@
"successful application of @var{pred}.")
#define FUNC_NAME s_scm_string_any
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM res;
@@ -108,7 +108,7 @@
"application of @var{pred} to the last character of @var{s}.")
#define FUNC_NAME s_scm_string_every
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM res;
@@ -189,7 +189,7 @@
"Convert the string @var{str} into a list of characters.")
#define FUNC_NAME s_scm_string_to_listS
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM result = SCM_EOL;
@@ -645,7 +645,7 @@
"trimmed.")
#define FUNC_NAME s_scm_string_trim
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s, cstr,
@@ -720,7 +720,7 @@
"trimmed.")
#define FUNC_NAME s_scm_string_trim_right
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s, cstr,
@@ -795,7 +795,7 @@
"trimmed.")
#define FUNC_NAME s_scm_string_trim_both
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s, cstr,
@@ -906,7 +906,7 @@
"@var{i} is the first position that does not match.")
#define FUNC_NAME s_scm_string_compare
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -949,7 +949,7 @@
"character comparison is done case-insensitively.")
#define FUNC_NAME s_scm_string_compare_ci
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -987,7 +987,7 @@
"value otherwise.")
#define FUNC_NAME s_scm_string_eq
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1022,7 +1022,7 @@
"value otherwise.")
#define FUNC_NAME s_scm_string_neq
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1057,7 +1057,7 @@
"true value otherwise.")
#define FUNC_NAME s_scm_string_lt
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1092,7 +1092,7 @@
"true value otherwise.")
#define FUNC_NAME s_scm_string_gt
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1127,7 +1127,7 @@
"value otherwise.")
#define FUNC_NAME s_scm_string_le
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1162,7 +1162,7 @@
"otherwise.")
#define FUNC_NAME s_scm_string_ge
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1198,7 +1198,7 @@
"case-insensitively.")
#define FUNC_NAME s_scm_string_ci_eq
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1234,7 +1234,7 @@
"case-insensitively.")
#define FUNC_NAME s_scm_string_ci_neq
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1270,7 +1270,7 @@
"case-insensitively.")
#define FUNC_NAME s_scm_string_ci_lt
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1306,7 +1306,7 @@
"case-insensitively.")
#define FUNC_NAME s_scm_string_ci_gt
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1342,7 +1342,7 @@
"case-insensitively.")
#define FUNC_NAME s_scm_string_ci_le
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1378,7 +1378,7 @@
"case-insensitively.")
#define FUNC_NAME s_scm_string_ci_ge
{
- char * cstr1, * cstr2;
+ unsigned char * cstr1, * cstr2;
int cstart1, cend1, cstart2, cend2;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s1, cstr1,
@@ -1658,7 +1658,7 @@
"@end itemize")
#define FUNC_NAME s_scm_string_indexS
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s, cstr,
@@ -1717,7 +1717,7 @@
"@end itemize")
#define FUNC_NAME s_scm_string_index_right
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s, cstr,
@@ -1777,7 +1777,7 @@
"@end itemize")
#define FUNC_NAME s_scm_string_skip
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s, cstr,
@@ -1837,7 +1837,7 @@
"@end itemize")
#define FUNC_NAME s_scm_string_skip_right
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM_VALIDATE_SUBSTRING_SPEC_COPY (1, s, cstr,
@@ -1896,7 +1896,7 @@
"@end itemize")
#define FUNC_NAME s_scm_string_count
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
int count = 0;
@@ -2442,7 +2442,7 @@
"string elements is not specified.")
#define FUNC_NAME s_scm_string_map
{
- char * cstr, *p;
+ unsigned char * cstr, *p;
int cstart, cend;
SCM result;
@@ -2473,7 +2473,7 @@
"modified in-place, the return value is not specified.")
#define FUNC_NAME s_scm_string_map_x
{
- char * cstr, *p;
+ unsigned char * cstr, *p;
int cstart, cend;
SCM_VALIDATE_PROC (1, proc);
@@ -2502,7 +2502,7 @@
"result of @var{kons}' application.")
#define FUNC_NAME s_scm_string_fold
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM result;
@@ -2529,7 +2529,7 @@
"result of @var{kons}' application.")
#define FUNC_NAME s_scm_string_fold_right
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM result;
@@ -2678,7 +2678,7 @@
"return value is not specified.")
#define FUNC_NAME s_scm_string_for_each
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM_VALIDATE_PROC (1, proc);
@@ -2914,7 +2914,7 @@
"character set, it is tested for membership.")
#define FUNC_NAME s_scm_string_filter
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM result;
int idx;
@@ -2980,7 +2980,7 @@
"character set, it is tested for membership.")
#define FUNC_NAME s_scm_string_delete
{
- char * cstr;
+ unsigned char * cstr;
int cstart, cend;
SCM result;
int idx;
[-- Attachment #1.2: Type: application/pgp-signature, Size: 189 bytes --]
[-- Attachment #2: Type: text/plain, Size: 143 bytes --]
_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: segfault in guile 1.6.7
2005-03-24 17:11 segfault in guile 1.6.7 Alexey Voinov
@ 2005-03-25 20:37 ` Kevin Ryde
2005-03-25 20:47 ` Kevin Ryde
2005-05-22 18:59 ` Marius Vollmer
1 sibling, 1 reply; 7+ messages in thread
From: Kevin Ryde @ 2005-03-25 20:37 UTC (permalink / raw)
Cc: guile-devel
[-- Attachment #1: Type: text/plain, Size: 426 bytes --]
Alexey Voinov <voins@voins.program.ru> writes:
>
> guile> (string-index (string (integer->char 209)) char-numeric?)
> zsh: 2760 segmentation fault guile
Thanks.
> I haven't checked if this bug is still present in cvs HEAD.
Marius made a change that squashes it.
> Patch attached.
I'll bring across what's in the head, which is to have SCM_MAKE_CHAR
cast to an unsigned char, instead of demanding that from all callers.
[-- Attachment #2: chars.h.unsigned.diff --]
[-- Type: text/plain, Size: 406 bytes --]
--- chars.h.~1.11.4.4.~ 2004-04-25 09:25:32.000000000 +1000
+++ chars.h 2005-03-26 07:36:06.574079280 +1100
@@ -51,7 +51,7 @@
*/
#define SCM_CHARP(x) (SCM_ITAG8(x) == scm_tc8_char)
#define SCM_CHAR(x) ((unsigned int)SCM_ITAG8_DATA(x))
-#define SCM_MAKE_CHAR(x) SCM_MAKE_ITAG8((scm_t_bits) x, scm_tc8_char)
+#define SCM_MAKE_CHAR(x) SCM_MAKE_ITAG8((scm_t_bits) (unsigned char) (x), scm_tc8_char)
\f
[-- Attachment #3: Type: text/plain, Size: 143 bytes --]
_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: segfault in guile 1.6.7
2005-03-25 20:37 ` Kevin Ryde
@ 2005-03-25 20:47 ` Kevin Ryde
0 siblings, 0 replies; 7+ messages in thread
From: Kevin Ryde @ 2005-03-25 20:47 UTC (permalink / raw)
Cc: guile-devel
I wrote:
>
> Marius
Actually it was Han-Wen Nienhuys.
_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: segfault in guile 1.6.7
2005-03-24 17:11 segfault in guile 1.6.7 Alexey Voinov
2005-03-25 20:37 ` Kevin Ryde
@ 2005-05-22 18:59 ` Marius Vollmer
2005-05-23 1:34 ` Kevin Ryde
1 sibling, 1 reply; 7+ messages in thread
From: Marius Vollmer @ 2005-05-22 18:59 UTC (permalink / raw)
Cc: guile-devel
Alexey Voinov <voins@voins.program.ru> writes:
> There's a bug in guile srfi-13 implementation.
Thanks! I applied it.
> I haven't checked if this bug is still present in cvs HEAD.
It should be fixed in CVS HEAD already.
--
GPG: D5D4E405 - 2F9B BCCC 8527 692A 04E3 331E FAF8 226A D5D4 E405
_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: segfault in guile 1.6.7
2005-05-22 18:59 ` Marius Vollmer
@ 2005-05-23 1:34 ` Kevin Ryde
2005-06-06 19:20 ` Marius Vollmer
0 siblings, 1 reply; 7+ messages in thread
From: Kevin Ryde @ 2005-05-23 1:34 UTC (permalink / raw)
Cc: Alexey Voinov, guile-devel
Marius Vollmer <mvo@zagadka.de> writes:
>
> Thanks! I applied it.
I already made a fix (from the head by Han-Wen) in chars.h which
covers this, by stopping signed values getting into the value field of
a char SCM (there's 24 or 56 bits there). Should fix applications
passing "char" to SCM_MAKE_CHAR too.
_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: segfault in guile 1.6.7
2005-05-23 1:34 ` Kevin Ryde
@ 2005-06-06 19:20 ` Marius Vollmer
2005-06-06 22:07 ` Kevin Ryde
0 siblings, 1 reply; 7+ messages in thread
From: Marius Vollmer @ 2005-06-06 19:20 UTC (permalink / raw)
Cc: guile-devel
Kevin Ryde <user42@zip.com.au> writes:
> Marius Vollmer <mvo@zagadka.de> writes:
>>
>> Thanks! I applied it.
>
> I already made a fix (from the head by Han-Wen) in chars.h which
> covers this, by stopping signed values getting into the value field of
> a char SCM (there's 24 or 56 bits there).
Oops. Is there any harm in having it fixed in both places?
--
GPG: D5D4E405 - 2F9B BCCC 8527 692A 04E3 331E FAF8 226A D5D4 E405
_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: segfault in guile 1.6.7
2005-06-06 19:20 ` Marius Vollmer
@ 2005-06-06 22:07 ` Kevin Ryde
0 siblings, 0 replies; 7+ messages in thread
From: Kevin Ryde @ 2005-06-06 22:07 UTC (permalink / raw)
Cc: Alexey Voinov, guile-devel
Marius Vollmer <mvo@zagadka.de> writes:
>
> Oops. Is there any harm in having it fixed in both places?
Shouldn't be.
_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2005-06-06 22:07 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-03-24 17:11 segfault in guile 1.6.7 Alexey Voinov
2005-03-25 20:37 ` Kevin Ryde
2005-03-25 20:47 ` Kevin Ryde
2005-05-22 18:59 ` Marius Vollmer
2005-05-23 1:34 ` Kevin Ryde
2005-06-06 19:20 ` Marius Vollmer
2005-06-06 22:07 ` Kevin Ryde
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).