summaryrefslogtreecommitdiff
path: root/devel/unproven-pthreads
AgeCommit message (Collapse)AuthorFilesLines
2001-02-25Cleanup MKDIR usage => INSTALL_*_DIRhubertf1-2/+2
XXX need to teach pkglint to be more picky about this
2001-02-16Update to new COMMENT style: COMMENT var in Makefile instead of pkg/COMMENT.wiz2-2/+2
2001-01-10Make the PIC relocations correct on the powerpc.briggs2-8/+8
2001-01-09Add support for the PowerPC.briggs9-3/+649
2001-01-04The way that shared objects were handled in the PLISTs and bsd.pkg.mk wasagc1-1/+3
out of date - it was based on a.out OBJECT_FMT, and added entries in the generated PLISTs to reflect the symlinks that ELF packages uses. It also tried to be clever, and removed and recreated any symbolic links that were created, which has resulted in some fun, especially with packages which use dlopen(3) to load modules. Some recent changes to our ld.so to bring it more into line with other Operating Systems also exposed some cracks. + Modify bsd.pkg.mk and its shared object handling, so that PLISTs now contain the ELF symlinks. + Don't mess about with file system entries when handling shared objects in bsd.pkg.mk, since it's likely that libtool and the BSD *.mk processing will have got it right, and have a much better idea than we do. + Modify PLISTs to contain "ELF symlinks" + On a.out platforms, delete any "ELF symlinks" from the generated PLISTs + On ELF platforms, no extra processing needs to be done in bsd.pkg.mk + Modify print-PLIST target in bsd.pkg.mk to add dummy symlink entries on a.out platforms + Update the documentation in Packages.txt With many thanks to Thomas Klausner for keeping me honest with this.
2000-12-28Add alpha support to unproven-pthreads.wiz8-4/+1001
By Olaf Seibert <rhialto@polderland.nl>, posted on port-alpha on 2000/08/08. Compiles fine, otherwise untested.
2000-11-29Add similar patch to the one in mit-pthreads, fixing the return value fromagc2-1/+30
detzcode() from long to int. From Tom Thai.
2000-09-11Modify order of compile flags of pgcc, as requested in pkg/9921.wiz3-6/+14
Approved by explorer. Step to 0.17nb1.
2000-07-10bump to 0.17 -- fixes a resolver bug, and works with 1.5{,A,B} as well.explorer3-9/+6
2000-05-24bump to 0.16, which includes Jason Thorpe's mutex changesexplorer4-7/+10
2000-04-18re-release 0.15 -- minor typo in a source fileexplorer1-2/+2
2000-04-18bump up to 0.15 to implement p{read,write}{,v}explorer2-4/+4
2000-04-14bump to 0.14 -- handle signals a little better. Bugs are still there, but ↵explorer2-4/+4
at least they _work_ now.
2000-04-07bump to 0.13 -- fixes sigaction() brokennessexplorer2-4/+4
2000-03-27bump to 0.12 which fixes a locking bug in dup2()explorer5-30/+9
2000-02-22add CONFLICTS. installs the same files as mit-pthreadsdmcmahill1-1/+2
2000-02-22patch configure script to correctly recognise netbsd-1.4.* in additiondmcmahill3-0/+26
to netbsd-1.4[A-Z]
2000-01-27upgrade to 0.11, and enable on m68k since I've been told it works there, too.explorer2-6/+6
2000-01-18update to 0.10 to include NetBSD-current's netdb.h, with ipv6 stuff. Iexplorer2-4/+4
really need to take the various getXbyY_r() functions and put them in the main source tree, and ditch this file.
2000-01-05Strip trailing '.', and/or leading '(a|an) 'abs1-1/+1
2000-01-02replaced some commands by their ${COMMAND} counterpartswiz1-4/+4
1999-12-28replaced some commands by their ${COMMAND} counterpartswiz1-4/+4
1999-12-11version 0.9: fixes sendmsg() and recvmsg(), and fixes an errno tracking problemexplorer2-4/+4
1999-12-02up to 0.8 now, whee. bind9 builds without warnings now.explorer2-4/+4
1999-11-24move to 0.7. This silences some rather annoying errno/sys_errlist noise.explorer2-10/+4
1999-11-12upgrade to 0.6, which should work on NetBSD-1.4.1/arm32explorer4-9/+10
1999-11-12Import unproven-pthreads, the mit-pthreads based cleanup I'm hacking on.explorer5-0/+81
Why am I using MIT-Pthreads? Because all the alternatives seem to have very low level problems. PTL2 has a locking problem of some sort that I cannot track down, and the author insists that lock = PTHREAD_MUTEX_INIT; is always to be legal, so he uses a pointer for locks. This means pthread_lock() can break in out of memory situations, and therefore so can pthread_once(), and if you're using that to protect logging, and you need to report a memory depletion, you're screwed. --Michael