Timeline
Mar 22, 2012:
- 9:57 PM Ticket #199 (libc: the ctype.h is_whatever doesn't handle EOF correctly / busted ...) closed by
 - fixed: Fixed in r3789 and r3891. Testcase in r3890. Note. Negative character …
 - 9:56 PM Changeset [3791] by
 - trunk: Backported ctype.h fix r3789. Fixes #199.
 - 9:54 PM Changeset [3790] by
 - ctype-1.c: Testcase for the EOF trouble.
 - 9:53 PM Changeset [3789] by
 - ctype.h: Correctly fend of queries reading EOF (-1). Fixes #199.
 - 9:13 PM Ticket #240 (libc: code set of setlocale() and nl_langinfo()) closed by
 - fixed: r3788 make setlocale default to the process code page when no codeset …
 - 9:12 PM Changeset [3788] by
 - setlocale.c: Changed getCodepage to default to the process codepage …
 - 8:49 PM Ticket #187 (strftime is broken for several locales) closed by
 - fixed: Fixed the en_GB date/time format string in r3787. Closing the defect …
 - 8:46 PM Changeset [3787] by
 - localeTimeDo: Fix bad date_time_fmt string for en_GB and others (hour …
 - 8:02 PM Ticket #186 (libc: check that stdout is properly flushed on exit) closed by
 - worksforme: Works for me. See r3786 for a simple testcase.
 - 8:01 PM Changeset [3786] by
 - Testcase for #186.
 - 7:38 PM Ticket #253 (libc: spawn() crash when starting a shell script) closed by
 - fixed: Found a memmove which moved too much, the script location/name length …
 - 7:36 PM Changeset [3785] by
 - spawnve.c: fixed memmove in the bang-hash-script path that moved too …
 - 7:36 PM Changeset [3784] by
 - 0.6/spawnve.c: fixed memmove in the bang-hash-script path that moved …
 - 7:10 PM Changeset [3783] by
 - 0.6/sharedpm.c: Disabled some racy/buggy code in libc_spmSelf.
 - 7:08 PM Changeset [3782] by
 - Forward ported r3780 and r3781 (nits).
 - 6:36 PM Changeset [3781] by
 - sharedpm: Don't assert if we get ERROR_ALREADY_POSTED (299).
 - 6:36 PM Changeset [3780] by
 - typo in comment
 - 6:28 PM Ticket #190 (Libc path rewriting is not working for fork()ed processes) closed by
 - wontfix: The reason why this is still broken (I believe I tried to fix it once …
 - 4:51 PM Ticket #211 (sigfpe in strtod) closed by
 - worksforme: Cannot reproduce. Tried two machines no SIGFPE.
 - 2:36 PM Ticket #193 (fnmatch() implementation is broken) closed by
 - fixed
 - 2:35 PM Changeset [3779] by
 - Drop the liberty fnmatch.h, adding the GNUisms to the emx header. …
 - 2:11 PM Ticket #248 (libc: Unix extended attributes can cause non-EAs aware applictions to fail) closed by
 - fixed: Dropped FEA_NEEDEA in r3778. Hope everyone will be happy now…
 - 2:10 PM Changeset [3778] by
 - libc-backend: Don't set FEA_NEEDEA - it causes weird problems here and …
 - 2:02 PM Ticket #205 (coreutils chmod does not work on files without EA) closed by
 - fixed: Implemented in r3777.
 - 2:01 PM Changeset [3777] by
 - chmod/fchmod: Add unix mode EA if not found and the FS supports it. …
 - 1:20 PM Ticket #247 (libc: feof(pipe_handle) keeps returning 0 after pipe is closed) closed by
 - fixed: Checked a little bit and seems ERROR_PIPE_NOT_CONNECTED is only used …
 - 1:19 PM Changeset [3776] by
 - read/fhOs2FileWrite: Translate ERROR_PIPE_NOT_CONNECTED when …
 
Mar 15, 2012:
- 10:34 PM Ticket #165 (libc: don't kill children when the parent exits) closed by
 - fixed: You're right, kLibC is still interring all children upon process …
 - 10:33 PM Changeset [3775] by
 - Leave decendants running. Fixes #165.
 - 9:34 PM Ticket #237 (libc: _fullpath() changes current drive) closed by
 - fixed: r3773 fixes _dt_split. Case hopefully closed.
 - 9:33 PM Changeset [3774] by
 - Backported r3773: dtsplit.c: Use _chdir_os2 instead of _chdir, since …
 - 9:32 PM Changeset [3773] by
 - dtsplit.c: Use _chdir_os2 instead of _chdir, since the former got the …
 - 9:25 PM Changeset [3772] by
 - Backported r3769: Reimplemented _fullpath to avoid changing the …
 - 9:06 PM Changeset [3771] by
 - Backported r3768: realpath and _realrealpath should not fail if the …
 - 9:02 PM Changeset [3770] by
 - src/kNIX: svn:eol-style and svn:keywords.
 - 9:00 PM Changeset [3769] by
 - Reimplemented _fullpath to avoid changing the current directory and …
 - 8:59 PM Changeset [3768] by
 - realpath and _realrealpath should not fail if the last component does …
 - 8:54 PM Changeset [3767] by
 - fullpath testcase (for #237).
 - 12:20 PM Ticket #241 (libc: build failure with newer gcc/binutils) closed by
 - fixed: Corrected this in r3765 and r3766.
 - 12:19 PM Changeset [3766] by
 - Backported r3765: FCW and FSW are 16-bit, not 32-bit, so use 'unsigned …
 - 12:16 PM Changeset [3765] by
 - FCW and FSW are 16-bit, not 32-bit, so use 'unsigned short' types …
 - 2:59 AM Ticket #234 (ld: Align .text by 16 bytes) closed by
 - fixed: Fixed in r3763 and r3764. Thanks for noticing.
 - 2:59 AM Changeset [3764] by
 - Backported r3763: ld: forgot to write out the text section padding for …
 - 2:58 AM Changeset [3763] by
 - ld: forgot to write out the text section padding for each object file.
 - 2:41 AM Ticket #219 (wcscmp() and wcsncmp() are broken) closed by
 - fixed: Addressed in r3761 (branch) and r3762 (trunk).
 - 2:40 AM Changeset [3762] by
 - wcsncmp,wcscmp: don't assume that wchar_t is the same as int. Would …
 - 2:38 AM Changeset [3761] by
 - wcsncmp,wcscmp: don't assume that wchar_t is the same as int. Would …
 
Mar 14, 2012:
- 11:24 AM Ticket #182 (iconv() and path separator) closed by
 - fixed: Fixed comment.
 - 11:24 AM Changeset [3760] by
 - iconv.c: comment typo.
 - 11:24 AM Changeset [3759] by
 - iconv.c: comment typo.
 
Mar 9, 2012:
- 9:33 PM Ticket #256 (LIBC stream fmutex panic) created by
 - An attempt to terminate a multi-threaded console program with Ctrl-C …
 
Mar 5, 2012:
- 12:42 PM Ticket #182 (iconv() and path separator) reopened by
 - Comment is wrong. WON is assigned to an ASCII code of a backward …
 - 12:37 PM Ticket #249 (LIBC extended attributes can cause non-EAs aware applictions to fail) closed by
 - duplicate: Duplicate of #248.
 - 12:37 PM Ticket #234 (ld: Align .text by 16 bytes) reopened by
 - r3749 and r3750 causes the executable to crash. I attach a testcase. …
 - 12:26 PM Ticket #255 (libc064 gives SIGSEGV when closing some processes due to being linked ...) closed by
 - fixed: I've change my local build environment to use WLINK instead of ILINK. …
 - 12:23 PM Ticket #242 (sbrk/_sbrk conflicting declarations) closed by
 - invalid: I fail to see what needs fixing in libc here. unistd.h and stdlib.h …
 - 10:53 AM Ticket #208 (Remark 'todo' stuffs in 'time.h') closed by
 - wontfix
 - 2:27 AM Ticket #236 (sys/socket.h: Use _ALIGN instead of ALIGN for CMSG_NXTHDR) closed by
 - fixed: Thanks for spotting this. Done in r3757 and r3758.
 - 2:26 AM Changeset [3758] by
 - Backported r3757: sys/socket.h: Use _ALIGN instead of ALIGN because of …
 - 2:25 AM Changeset [3757] by
 - sys/socket.h: Use _ALIGN instead of ALIGN because of …
 - 2:16 AM Ticket #227 (wcrtomb() broken) closed by
 - invalid: What happens is that when setting LC_ALL, LC_CTYPE or LANG to 'de_CH' …
 - 2:10 AM Changeset [3756] by
 - 227-wctomb-de_ch.c: Testcase for #227.
 
Mar 4, 2012:
- 10:36 PM Ticket #182 (iconv() and path separator) closed by
 - fixed: Changed iconv to not clear CVTTYPE_PATH in r3754. Backported to 0.6 in …
 - 10:34 PM Changeset [3755] by
 - Backported r3754: iconv.c: Assume input is PATHs so that samba works …
 - 10:33 PM Changeset [3754] by
 - iconv.c: Assume input is PATHs so that samba works for Koreans.
 - 10:25 PM Ticket #254 (weakld: weakld incorrectly reports duplicate symbols) closed by
 - fixed: r3751 tries to address the problem. It is a relatively risky fix as …
 - 10:23 PM Changeset [3753] by
 - Backported r3751: weakld.c: Put exported names in a different symbol …
 - 10:22 PM Changeset [3752] by
 - Testcase for #254.
 - 10:17 PM Changeset [3751] by
 - weakld.c: Put exported names in a different symbol table since they …
 - 8:03 PM Ticket #234 (ld: Align .text by 16 bytes) closed by
 - fixed: Addressed in r3749 and r3750. Case closed.
 - 8:02 PM Changeset [3750] by
 - 0.6/ld: Backported r3749 which makes ld align .text on a SSE friendly …
 - 8:00 PM Changeset [3749] by
 - ld: Align text section on 16-byte boundrary for SSE constants.
 - 7:52 PM Ticket #233 (getsockname() and getpeername() fail when using high memory) closed by
 - fixed
 - 7:51 PM Changeset [3748] by
 - trunk: forward ported r3745, r3746 and r3747 - high memory regression …
 - 7:45 PM Changeset [3747] by
 - socketpair: support osfd (return sockets descriptors) pointing to high …
 - 7:41 PM Changeset [3746] by
 - highmemsocket.cpp: Fixed two bad tests causing connect() and others to …
 - 7:40 PM Changeset [3745] by
 - getpeername.c: fix regression from #233.
 
Feb 29, 2012:
- 10:49 PM Changeset [3744] by
 - Preparing for 0.6.5…
 - 10:48 PM Changeset [3743] by
 - Preparing for 0.6.5…
 
Dec 28, 2011:
- 9:34 PM Ticket #255 (libc064 gives SIGSEGV when closing some processes due to being linked ...) created by
 - Many applications I've built (httpd, lyx, postgres to name 3) are …
 
  Note:
 See   TracTimeline
 for information about the timeline view.
    