summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2003-11-08For now, disable the iconv utilityEric Andersen
2003-11-08Joakim Tjernlund writes:Eric Andersen
> Very interesting. Do you have any suggestions for how > we could fix our powerpc shared library loader Removing those instr. comes with a very big performance penalty. To flush the dcache you will have read up to 8KB dummy data and to invalidate the icache you will have to execute up to 16KB nops. I don't know of any other way from user space. hmm, actually I think it will work reliable to perform a store to the same page(s) as the dcbst/icbi will act on. That way you will make the DTLB Error happen(if any) prior to the dcbst/icbi. The worst thing that can happen then is a regular DTLB Miss and that works for dcbst/icbi. You will have to lookout for if dcbst/icbi crosses a page boundary. Then you will have to perform a store to both pages. Jocke # And again later writes: Hi again I think I know what the problem is. The PPC_DCBST;PPC_SYNC;PPC_ICBI;PPC_ISYNC sequence is executed even if no modification has been done i some cases: _dl_linux_resolver(), the last else has no store for insns[0]. these is a insns[1] = OPCODE_B(delta - 4) that does not have a PPC_DCBST. _dl_do_lazy_reloc(), for R_PPC_NONE there is no store. for R_PPC_JMP_SLOT there is a insns[1] = OPCODE_B(delta)that does not have a PPC_DCBST. _dl_do_reloc(), for R_PPC_COPY there is no store. for R_PPC_JMP_SLOT there is a reloc_addr[1] = OPCODE_B(delta) that does not have a PPC_DCBST. _dl_init_got(), I THINK that the PPC_DCBST(plt); PPC_DCBST(plt+4); PPC_DCBST(plt+8); PPC_SYNC; PPC_ICBI(plt); PPC_ICBI(plt+4); PPC_ICBI(plt+8); PPC_ISYNC; is off a bit. The address range does not match the sum of the plt[] and tramp[] address range. Jocke # And then later added the comment: I think that the tramp[] part should be included in the PPC_DCBST/PPC_ICBI sequence. Then you have to add entries for plt+12 and plt+16. If the tramp[] part should be excluded, then all is well. Jocke
2003-11-08Fix up several errors related to filename length and errno thatEric Andersen
showed up while running the latest LTP testsuite. -Erik
2003-11-08Add some basic cvs usage infoEric Andersen
2003-11-07Sigh. I forgot to define INSTALL which, understandable,Eric Andersen
broke the install target
2003-11-07Add clock_gettime() submitted by Justus Pendleton <uc@ryoohki.net>.Manuel Novoa III
2003-11-07remove some junk I added while debuggingEric Andersen
2003-11-07It is remotely possible the utils might even compile this timeEric Andersen
around...
2003-11-06Begin converting the client utilsEric Andersen
2003-11-06Revert change 1.26, which breaks everything. :-(Eric Andersen
2003-11-06Sigh... I wasn't checking if the gid field in a group line actuallyManuel Novoa III
contained a digit. Also adjust a comment.
2003-11-06s,(RUNTIME_PREFIX)/,(RUNTIME_PREFIX),gEric Andersen
2003-11-06s,(DEVEL_PREFIX)/,(DEVEL_PREFIX),gEric Andersen
2003-11-06Add a utility macro.Manuel Novoa III
2003-11-06Oops... don't need recursive mutexes.Manuel Novoa III
2003-11-06Rewrite the pwd.h, grp.h, and shadow.h functions (except lckpwdf/ulckpwdf).Manuel Novoa III
2003-11-06Add in aliases provided by the non arch specific implementationsEric Andersen
2003-11-06George Thanos writes:Eric Andersen
Dear Erik, We downloded uClibc lattest version from the CVS. Still there are some minor problems with extra/Configs/Config.e1 You have actually set ARCH_HAS_C_SYMBOL_PREFIX to NO which is not correct for our architecture. Please apply the patch that will fix the problem. Best Regards, - George P.S. Patch also removes some irritating comments we have added in the past.
2003-11-06A few debugging cleanupsEric Andersen
2003-11-06The curse of cut-n-pasteEric Andersen
2003-11-06Add some initial x86 string optimizations. These make no attempt to use niftyEric Andersen
things like mmx/3dnow/etc. These are not inline, and will therefore not be as fast as modifying the headers to use inlines (and cannot therefore do tricky things when dealing with const memory). But they should (I hope!) be faster than their generic equivalents.... More importantly, these should provide a good example for others to follow when adding arch specific optimizations. -Erik
2003-11-05Remove erroneous trailing semicolon.Manuel Novoa III
2003-11-05Fix a misnamed arrayEric Andersen
2003-11-05tiny cleanupEric Andersen
2003-11-05Rework the powerpc lib loader to better match the other architecturesEric Andersen
2003-11-05Rip the guts out of the dynamically generated sysconf.c, and live with a simpleEric Andersen
static version. This will need further work later on, but should do the job for the time being,
2003-11-05last but certinaly not least, kill off initfini.cEric Andersen
2003-11-05Add quick hacks to create fake crti.o and crtn.o files. These architecturesEric Andersen
need to be fixed properly. I tried, but I was unable to build a cross toolchain for each of these (using stock binutils and gcc) so it is someone else's problem to fix them now.
2003-11-05Kill initfini.awkEric Andersen
2003-11-05Continue the conversion to using per-arch crti.S and crtn.SEric Andersen
2003-11-05Continue the conversion to using per-arch crti.S and crtn.SEric Andersen
2003-11-05Continue the conversion to using per-arch crti.S and crtn.SEric Andersen
2003-11-05v850 appers to be little endian onlyEric Andersen
2003-11-05Continue the conversion to using per-arch crti.S and crtn.SEric Andersen
2003-11-05duh. the alpha has an fpuEric Andersen
2003-11-05Yet more messing with selecting sane default configsEric Andersen
2003-11-05Yet more config system updatesEric Andersen
2003-11-05Continue the conversion to using per-arch crti.S and crtn.SEric Andersen
2003-11-05Fix a buglet that shows up on powerpcEric Andersen
2003-11-05m68k is always big endian afaikEric Andersen
2003-11-05powerpc is always big endianEric Andersen
2003-11-05Continue the conversion to using per-arch crti.S and crtn.SEric Andersen
2003-11-05Patch from Peter S. Mazinger to hunt for awk in additionalEric Andersen
places....
2003-11-05Arthur Shipkowski, art ! videon-central ! com, writes:Eric Andersen
I've noticed a few people have posted over the last year about problems compiling programs that use vfork when pthreads are involved. Some detective work turned up that ptfork.c aliases vfork to fork and then tries to call the original fork as __libc_fork. This patch removes the aliasing when there is no MMU present, and uses the same call semantics to call __libc_vfork. I then added a symbol to the m68k vfork.S to allow vfork to be called as __libc_vfork. The same bug exists in the uClibc CVS, and with a possible tweak this patch should go through there as well. Obviously, all other platforms need __libc_vfork as a workable means to call vfork in order for this to work for them. Let me know if there are any problems with this patch. Art Shipkowski Videon Central Software Engineer (814)235-1111 x307
2003-11-05Force cris to be little endian. Afaik, there is no bigEric Andersen
endian cris architecture.
2003-11-05Continue the conversion to using per-arch crti.S and crtn.SEric Andersen
2003-11-05OopsEric Andersen
2003-11-05Continue the conversion to using per-arch crti.S and crtn.SEric Andersen
2003-11-05Do not provide a default for endianness. This needs to be selectedEric Andersen
on a per-arch basis, or left to the user to choose.
2003-11-05Kill off "mipsel" and just use "mips"Eric Andersen