summaryrefslogtreecommitdiff
path: root/libexec
AgeCommit message (Collapse)Author
2006-09-01Pay attention to the new LFS_CLEANER_MUST_CLEAN flag in the Ifile.perseant
2006-08-30Only include libutil if !HOSTPROGchristos
2006-08-27Simplify multiple MACHINE_ARCH endian checks with single MACHINE_CPUsimonb
checks.
2006-08-26use efun(3)christos
2006-07-31Make filehandles opaque to userlandmartin
2006-07-13Fix alignement problems for fhandle_t, exposed by gcc4.1.martin
While touching all vptofh/fhtovp functions, get rid of VFS_MAXFIDSIZ, version the getfh(2) syscall and explicitly pass the size available in the filehandle from userland. Discussed on tech-kern, with lots of help from yamt (thanks!).
2006-07-07Place an upper bound on the number of segments to clean at one time, noperseant
matter how empty they are. Note that if two blocks have the same inode and block number, they sort the same (this should never happen, but if it does there's no reason to have qsort scramble the list). Add some diagnostic syslog messages for unusual cases.
2006-06-29Add (back) -Wl,-static to the link.skrll
2006-06-28Fix unaligned loads and stores generated by gcc4. Problem described insimonb
http://mail-index.NetBSD.org/tech-toolchain/2006/06/27/0001.html. This cleaner version of the patch from Christos - thanks!
2006-06-13Use .previous at the end of the NetBSD ident note so that we switchsimonb
back to the previous section. Fixes build problems on MIPS with gcc4 and -g.
2006-05-23in _rtld_bind() initialise new_value for GCC4.mrg
2006-05-22sprinkle some __noreturn__ attributes, saves gcc4 warnings elsewheredrochner
2006-05-21avoid more GCC4 uninitialised variable problems. (sh3 known, vax guessed.)mrg
2006-05-20appease gcc4 uninitialiased warningdan
ok mrg.
2006-05-20in _rtld_bind() initialise result to appease GCC4.mrg
2006-05-19Coverity CID 3766: Avoid null pointer derefchristos
2006-05-12Be more careful about checking return value of {m,re}alloc().perseant
In determining when to stop reading segments when counting bytes (-b flag), total the sizes of the blocks we're actually writing instead of assuming they are all full blocks: many could be fragments or inode blocks. This increases the number of segments per Ifile write, markedly improving the efficiency of the cleaner in the small file case.
2006-05-11sprinkle some -fno-strict-aliasing and -Wno-pointer-sign with GCC4.mrg
2006-05-11use a u_char[] when it is expected.mrg
2006-05-11avoid some char * vs. u_char * issues.mrg
2006-05-11end the argument list to exec*(3) with a NULL instead of a bare '0', asmrg
the latter isn't a pointer context in these varargs functions.
2006-05-10quell GCC 4.1 uninitialised variable warnings.mrg
XXX: we should audit the tree for which old ones are no longer needed after getting the older compilers out of the tree..
2006-05-09change (mostly) int to socklen_t. GCC 4 doesn't like that int andmrg
socklen_t are different signness.
2006-04-14New io statisticsblymn
2006-04-14When checking hidden cleaning costs, don't erroneously add the firstperseant
indirect block when considering the cleaning of block numbers less than NDADDR (which do not use indirect blocks). Also, note the loss of only half a block per segment to fragmentation when considering the benefit function, rather than a whole block.
2006-04-10modified man(1):chuck
- avoid double slashes when displaying man pages (got tired of '/usr/share/man//cat1/man.0'). - got rid of __P() while working on it. - incorporate some of my old notes explaining how manpath works into the comments of the code itself. - renamed some of the vars so that the code is consistent throughout (and hopefully clearer and easier to understand) - fixed relative man paths for multiple man pages (man did a chdir() on the first man page it had to format --- this broke any remaining relative path man pages left to process). save old directory and fchdir() back to it after formatting. - improved doc on "man -h" which does more than just whatis(1) [e.g. "man -h fopen" prints the required include files and the prototypes rather than just the one-liner you get with whatis(1)] - manconf.c now fills in the "len" length field in the TAG/ENTRY structures (man now uses len). revise man.conf file reading stuff to return error on failure in addentry/gettag (fka getlist) rather than just err()ing out. this allows man(1) to call cleanup and delete its tmp files rather than just leave them floating. revise other apps using this code (makewhatis, apropos, catman, whatis) to expect this. also remove __P on updated files.
2006-04-05Look for the inode pointers in the right place if lfs_bsize != lfs_fsize.perseant
2006-04-05Handle the case of inode blocks misidentified by the segment headersperseant
correctly, a problem reported by Kurt Schreiner; as well as similar error cases (realloc returning NULL or no cleanable segments found).
2006-04-03Remove empty lines.skrll
2006-04-01Coverity CID 2792: Fix memory leakschristos
2006-04-01Coverity CID 2846: Don't use LIST_FOREACH if you are going to free the cursorchristos
2006-04-01Coverity CID 2847: Don't use LIST_FOREACH if you are going to free thechristos
current pointer!
2006-04-01Document -i option, update date; at the request of wiz@.perseant
2006-03-30A new version of the cleaner. In general, works about as well as the oldperseant
cleaner, but with more legible code. Includes code for reading and writing to the raw disk device (so that an unmounted fs could be cleaned), for the use of a single daemon to clean multiple filesystems to save on resources, and for recording the old contents of cleaned segments to offline storage for regression testing of the LFS system as a whole; though these new features are not properly tested at this point.
2006-03-23Remove more references to kerberosIV.wiz
2006-03-21Coverity CID 2749: Fix another double free. Instead of removing that lastchristos
free() call, change the allocation policy to leave the responsibility for allocation/freeing the pathname to _rtld_map_object(), instead of having the caller allocate it and _rtld_map_object() free it. This simplifies the code a lot and it is more efficient.
2006-03-20Goodbye KerberosIVchristos
2006-03-18Coverity CID 923: Avoid NULL deref.christos
2006-03-18Fix some double frees / missing frees / use after freed allmatt
relating to _rtld_load_object. Fix Coverity CID 2601, 2600.
2006-03-17Remove unreachable code (res could never be NULL here).peter
Fixes Coverity CID 712. Ok by lukem@.
2006-03-17Set file to NULL after calling fclose().peter
Fixes Coverity CID 2669. Ok by lukem@.
2006-03-17Make sure that "su" is initialized before dereferencing it.peter
Fixes Coverity CID 1075. Ok by lukem@.
2006-03-17Fix three off-by-ones found by Coverity. CID 1444, 1445.elad
2006-03-04Remove the LD_BIND_NOW itemskrll
2006-03-04Add LD_BIND_NOW support for mips.skrll
This is the last arch to be done so remove the large ugly #if defined() stuff in reloc.c.
2006-02-25Fix some typos.wiz
2006-02-03Remove -Wl,-non_shared (same meaing as -static).rpaulo
2006-02-02Make the debug variable static, to accomodate static linking with libssh.a.he
2006-02-01debug -> ftpd_debugchristos
xstrdup -> ftpd_strdup
2006-01-12Bring across a change from FreeBSD from back in 1998:skrll
"Fix a bug that showed up when debugging dynamically linked programs. References from GDB to "printf" and various other functions would find the versions in the dynamic linker itself, rather than the versions in the program's libc. This fix moves the GDB link map entry for the dynamic linker to the end of the search list, where its symbols will be found only if they are not found anywhere else." "printf" isn't true for us, but various libc symbols are, e.g. "malloc". Fixes PR 32074 as noted by uwe@ OK'd by christos@