about summary refs log tree commit diff
Commit message (Collapse)AuthorAgeFilesLines
* workaround gcc got-register-reload performance problems in mallocRich Felker2012-09-141-4/+8
| | | | | | | with this patch, the malloc in libc.so built with -Os is nearly the same speed as the one built with -O3. thus it solves the performance regression that resulted from removing the forced -O3 when building libc.so; now libc.so can be both small and fast.
* remove forced -O3 from shared library CFLAGSRich Felker2012-09-141-1/+1
| | | | | | | | | | | I originally added -O3 for shared libraries to counteract very bad behavior by GCC when building PIC code: it insists on reloading the GOT register in static functions that need it, even if the address of the function is never leaked from the translation unit and all local callers of the function have already loaded the GOT register. this measurably degrades performance in a few key areas like malloc. the inlining done at -O3 avoids the issue, but that's really not a good reason for overriding the user's choice of optimization level.
* use vfork if possible in posix_spawnRich Felker2012-09-141-1/+3
| | | | | | vfork is implemented as the fork syscall (with no atfork handlers run) on archs where it is not available, so this change does not introduce any change in behavior or regression for such archs.
* strsep is BSD|GNU, not GNU-only; it's originally from BSDRich Felker2012-09-131-1/+4
|
* add O_PATH/O_SEARCH support to fcntl.hRich Felker2012-09-135-1/+9
| | | | | | | I'm not 100% sure that Linux's O_PATH meets the POSIX requirements for O_SEARCH, but it seems very close if not perfect. and old kernels ignore it, so O_SEARCH will still work as desired as long as the caller has read permissions to the directory.
* improve mips syscall asm constraints to use immediates, if possibleRich Felker2012-09-111-12/+21
| | | | | | | | | | | | | by using the "ir" constraint (immediate or register) and the carefully constructed instruction addu $2,$0,%2 which can take either an immediate or a register for %2, the new inline asm admits maximal optimization with no register spillage to the stack when the compiler successfully performs constant propagration, but still works by allocating a register when the syscall number cannot be recognized as a constant. in the case of syscalls with 0-3 arguments it barely matters, but for 4-argument syscalls, using an immediate for the syscall number avoids creating a stack frame for the syscall wrapper function.
* eliminate assumption that mips syscall restart preserves r25Rich Felker2012-09-101-23/+12
| | | | | | | | | all past and current kernel versions have done so, but there seems to be no reason it's necessary and the sentiment from everyone I've asked has been that we should not rely on it. instead, use r7 (an argument register) which will necessarily be preserved upon syscall restart. however this only works for 0-3 argument syscalls, and we have to resort to the function call for 4-argument syscalls.
* asm for memmove on i386 and x86_64Rich Felker2012-09-102-0/+36
| | | | | | | for the sake of simplicity, I've only used rep movsb rather than breaking up the copy for using rep movsd/q. on all modern cpus, this seems to be fine, but if there are performance problems, there might be a need to go back and add support for rep movsd/q.
* fix another ppoll issue (missing sigset_t size argument)Rich Felker2012-09-101-1/+1
|
* reenable word-at-at-time copying in memmoveRich Felker2012-09-101-4/+27
| | | | | | | | | before restrict was added, memove called memcpy for forward copies and used a byte-at-a-time loop for reverse copies. this was changed to avoid invoking UB now that memcpy has an undefined copying order, making memmove considerably slower. performance is still rather bad, so I'll be adding asm soon.
* fix ppoll with null timeout argumentRich Felker2012-09-101-2/+2
|
* add LIBCC (compiler runtime) logic and override to configureRich Felker2012-09-101-0/+7
| | | | | | | | | this should both fix the issue with ARM needing -lgcc_eh (although that's really a bug in the libgcc build process that's causing considerable bloat, which should be fixed) and make it easier to build musl using clang/llvm in place of gcc. unfortunately I don't know a good way to detect and support pcc's -lpcc since it's not in pcc's default library search path...
* add setdomainname syscall, fix getdomainname (previously a stub)Rich Felker2012-09-093-1/+18
|
* mincore syscall wrapperRich Felker2012-09-092-0/+9
|
* fix up lfs64 junk for preadv/pwritevRich Felker2012-09-093-2/+7
|
* add preadv/pwritev syscall wrappersRich Felker2012-09-093-0/+35
|
* fix typo introduced in poll.hRich Felker2012-09-091-1/+1
|
* add linux ppoll syscall wrapperRich Felker2012-09-092-0/+19
|
* reenable sync_file_range; should no longer break on mipsRich Felker2012-09-091-2/+2
|
* add 7-arg syscall support for mipsRich Felker2012-09-092-4/+8
| | | | | | | no syscalls actually use that many arguments; the issue is that some syscalls with 64-bit arguments have them ordered badly so that breaking them into aligned 32-bit half-arguments wastes slots with padding, and a 7th slot is needed for the last argument.
* inline syscall support for armRich Felker2012-09-091-0/+53
| | | | | | most pure-syscall-wrapper functions compile to the smallest/simplest code possible (save r7 ; load syscall # ; svc 0 ; restore r7 ; tail call to __syscall_ret).
* inline syscall support for mipsRich Felker2012-09-091-0/+57
| | | | | | | this drastically reduces the size of some functions which are purely syscall wrappers. disabled for clang due to known bugs satisfying register constraints.
* fix mips syscall_cp_asm code (saved register usage)Rich Felker2012-09-091-2/+2
|
* fix broken mips syscall asmRich Felker2012-09-091-2/+2
| | | | | | | this code was using $10 to save the syscall number, but $10 is not necessarily preserved by the kernel across syscalls. only mattered for syscalls that got interrupted by a signal and restarted. as far as i can tell, $25 is preserved by the kernel across syscalls.
* disable sync_file_range for nowRich Felker2012-09-081-2/+3
| | | | | | | something is wrong with the logic for the argument layout, resulting in compile errors on mips due to too many args to syscall... further information on how it's supposed to work will be needed before it can be reactivated.
* syscall organization overhaulRich Felker2012-09-0810-655/+420
| | | | | | | | | | | | now public syscall.h only exposes __NR_* and SYS_* constants and the variadic syscall function. no macros or inline functions, no __syscall_ret or other internal details, no 16-/32-bit legacy syscall renaming, etc. this logic has all been moved to src/internal/syscall.h with the arch-specific parts in arch/$(ARCH)/syscall_arch.h, and the amount of arch-specific stuff has been reduced to a minimum. changes still need to be reviewed/double-checked. minimal testing on i386 and mips has already been performed.
* add acct syscall source file, omitted in last syscalls commitRich Felker2012-09-081-0/+9
|
* add acct, accept4, setns, and dup3 syscalls (linux extensions)Rich Felker2012-09-088-0/+113
| | | | based on patch by Justin Cormack
* add IPPROTO_HOPOPTS to in.hRich Felker2012-09-081-0/+1
|
* add IPPROTO_MAX to in.hRich Felker2012-09-081-0/+1
|
* fix redundant _Noreturn def in err.hRich Felker2012-09-081-7/+1
| | | | not sure why this was missed in the earlier commit.
* remove all remaining redundant __restrict/__inline/_Noreturn defsRich Felker2012-09-0814-76/+14
|
* sysmacros major/minor: result should have type unsigned int, not dev_tRich Felker2012-09-081-2/+2
|
* add linux tee syscallRich Felker2012-09-082-0/+9
|
* add linux sync_file_range syscallRich Felker2012-09-082-0/+20
|
* move fallocate syscall wrapper to linux-specific syscalls dirRich Felker2012-09-081-0/+0
|
* add linux readahead syscallRich Felker2012-09-082-0/+9
|
* add fallocate (nonstandardized) functionRich Felker2012-09-082-0/+12
| | | | | | this is equivalent to posix_fallocate except that it has an extra mode/flags argument to control its behavior, and stores the error in errno rather than returning an error code.
* fix broken fallocate syscall in posix_fallocateRich Felker2012-09-081-1/+1
| | | | | the syscall takes an extra flag argument which should be zero to meet the POSIX requirements.
* add timerfd interfaces (untested)Rich Felker2012-09-082-0/+35
|
* add stdnoreturn.h (C11)Rich Felker2012-09-081-0/+5
| | | | features.h contains the fallback logic for pre-C11 compilers
* TCP_* is in the reserved namespace for tcp.h; make use of thatRich Felker2012-09-071-3/+4
|
* remove unneeded judgemental commentary from ftw.hRich Felker2012-09-071-4/+0
|
* default features: make musl usable without feature test macrosRich Felker2012-09-0748-129/+102
| | | | | | | | | | the old behavior of exposing nothing except plain ISO C can be obtained by defining __STRICT_ANSI__ or using a compiler option (such as -std=c99) that predefines it. the new default featureset is POSIX with XSI plus _BSD_SOURCE. any explicit feature test macros will inhibit the default. installation docs have also been updated to reflect this change.
* add clang-compatible thread-pointer code for mipsRich Felker2012-09-071-0/+4
| | | | | | | | | | clang does not presently support the "v" constraint we want to use to get the result from $3, and trying to use register...__asm__("$3") to do the same invokes serious compiler bugs. so for now, i'm working around the issue with an extra temp register and putting $3 in the clobber list instead of using it as output. when the bugs in clang are fixed, this issue should be revisited to generate smaller/faster code like what gcc gets.
* cleanup src/linux and src/misc trees, etc.Rich Felker2012-09-0745-98/+74
| | | | | | | | | | | | previously, it was pretty much random which one of these trees a given function appeared in. they have now been organized into: src/linux: non-POSIX linux syscalls (possibly shard with other nixen) src/legacy: various obsolete/legacy functions, mostly wrappers src/misc: still mostly uncategorized; some misc POSIX, some nonstd src/crypt: crypt hash functions further cleanup will be done later.
* fix constraint violation in ftwRich Felker2012-09-061-1/+4
| | | | void* does not implicitly convert to function pointer types.
* provide loff_t for splice syscallRich Felker2012-09-061-0/+1
| | | | | | | | | | | so far, this is the only actual use of loff_t i've found. some software, including glib, assumes loff_t must exist if splice exists; this is a reasonable assumption since the official prototype for splice uses loff_t, as it always works with 64-bit offsets regardless of the selected libc off_t size. i'm using #define for now rather than a typedef to make it easy to define in other headers if necessary (like the LFS64 ugliness), but it may be necessary to add it to alltypes.h eventually if other functions end up needing it.
* further use of _Noreturn, for non-plain-C functionsRich Felker2012-09-0611-19/+47
| | | | | | | | | | | | | | | | | | | note that POSIX does not specify these functions as _Noreturn, because POSIX is aligned with C99, not the new C11 standard. when POSIX is eventually updated to C11, it will almost surely give these functions the _Noreturn attribute. for now, the actual _Noreturn keyword is not used anyway when compiling with a c99 compiler, which is what POSIX requires; the GCC __attribute__ is used instead if it's available, however. in a few places, I've added infinite for loops at the end of _Noreturn functions to silence compiler warnings. presumably __buildin_unreachable could achieve the same thing, but it would only work on newer GCCs and would not be portable. the loops should have near-zero code size cost anyway. like the previous _Noreturn commit, this one is based on patches contributed by philomath.
* fix invalid implicit pointer conversion in gnulib-compat functionsRich Felker2012-09-061-1/+1
|
generated by cgit-pink 1.4.1 (git 2.36.1) at 2024-11-06 23:44:58 +0000