summaryrefslogtreecommitdiff
path: root/libc/stdlib/malloc
AgeCommit message (Collapse)Author
2007-04-24fix from Bernd Schmidt for realloc shrinkage bugMike Frysinger
2007-04-13fix up malloc debug buildingMike Frysinger
2007-04-02POSIX requires that errno be set whenever 0 is returned by malloc()Mike Frysinger
2007-01-29Richard Sandiford writes:Mike Frysinger
However, retesting on m68k showed up a problem that had appeared in uClibc since the last time I tried. Specifically, revision 15785 did: -#define HEAP_GRANULARITY (sizeof (HEAP_GRANULARITY_TYPE)) +#define HEAP_GRANULARITY (__alignof__ (HEAP_GRANULARITY_TYPE)) -#define MALLOC_ALIGNMENT (sizeof (double)) +#define MALLOC_ALIGNMENT (__alignof__ (double)) The problem is that (a) MALLOC_HEADER_SIZE == MALLOC_ALIGNMENT (b) the header contains a size value of type size_t (c) sizeof (size_t) is 4 on m68k, but... (d) __alignof__ (double) is only 2 (the largest alignment used on m68k) So we only allocate 2 bytes for the 4-byte header, and the least significant 2 bytes of the size are in the user's area rather than the header. The patch below fixes that problem by redefining MALLOC_HEADER_SIZE to: MAX (MALLOC_ALIGNMENT, sizeof (size_t)) (but without the help of the MAX macro ;)). However, we really would like to have word alignment on Coldfire. It makes a big performance difference, and because we have to allocate a 4-byte header anyway, what wastage there is will be confined to the end of the allocated block. Any wastage will also be limited to 2 bytes per allocation compared to the current alignment. I've therefore used the __aligned__ type attribute to create a double type that has at least sizeof (size_t) bytes of alignment. I've introduced a new __attribute_aligned__ macro for this. It might seem silly protecting against old or non-GNU compilers here, but the extra alignment is only an optimisation, and having the macro is more in the spirit of the other attribute code.
2006-12-11mostly revert the locking changes for 'malloc', as the implementationEric Andersen
does not easily lend itself to becoming complete pthread cancelation safe without first investing in some deep and serious thought... The other malloc implementations are pthread cancelation safe, and this one is mostly used for uClinux, where the lack is at least less likely to be a common problem.
2006-12-07Major cleanup of internal mutex locking. Be more consistant in how we doEric Andersen
things, and avoid potential deadlocks caused when a thread holding a uClibc internal lock get canceled and terminates without releasing the lock. This change also provides a single place, bits/uClibc_mutex.h, for thread libraries to modify to change all instances of internal locking.
2006-08-05merge fix from blackfin cvs:Mike Frysinger
bernds writes: Use __alignof__ instead of sizeof to get alignments. Eliminates some warnings about misalignments when malloc debugging is enabled.
2006-03-10Remove all non-constant libc_hidden_data_def(), it is too unreliable, sorry, ↵Peter S. Mazinger
most of global data relocations are back
2006-02-22check for a 0 size first, then check for a NULL pointerMike Frysinger
2006-02-18tweak the idea between having a MMU and actually using itMike Frysinger
2006-02-13libc-{a,so,multi}-y replaced by libc-y covering common objects both in ↵Peter S. Mazinger
libc.a/libc.so, the diffs go into libc-static-y/libc-shared-y exclusively, add IMA to libc, don't use any MSRC anymore
2006-01-29declare local prototype since we cant use system malloc.hMike Frysinger
2006-01-14hidden_def/hidden_proto: convert all users (I hope) termios split, add some ↵Peter S. Mazinger
missing headers, other jump relocs removed
2005-12-30fix path to heap.h as point out by thomas_chou in Bug 617Mike Frysinger
2005-12-16Convert all the rest, remove isxupper/isxlower, if someone objects, I'll add ↵Peter S. Mazinger
it back
2005-12-13__malloc_consolidate is only libc internal, enable use of __sbrkPeter S. Mazinger
2005-12-09Implement hidden poll, switch user to hidden *printf/*scanf/pollPeter S. Mazinger
2005-12-08mmap/mremap/socket/rewind gonePeter S. Mazinger
2005-12-08Use internal versionsPeter S. Mazinger
2005-12-06macro out the thread funcs in libc if threading is disabledMike Frysinger
2005-12-04More hiding, 300 leftPeter S. Mazinger
2005-12-03More hiding, including __mempcpyPeter S. Mazinger
2005-12-01Hide mostly used functionsPeter S. Mazinger
2005-11-21Remove TOPDIRPeter S. Mazinger
2005-11-03Enable IMAPeter S. Mazinger
2005-11-01Remove last unused references to libc-a-pic-, we use only the lists in ↵Peter S. Mazinger
libc-a-y for objects that go into static libs, changing their suffix to .os, of they should be PIC
2005-10-29Replace all Makefiles for new build infrastucturePeter S. Mazinger
2005-10-25All Makefile.in's. Only arm/i386/mips/powerpc/x86_64 are done, the other ↵Peter S. Mazinger
archs lack proper crt1. The Makefiles in extra/scripts are intended to be linked into each dir, where it is necessary to build locally.
2005-10-12Rewrote almost all Makefiles: do not use strip $(x),y; run strip on all ↵Peter S. Mazinger
objects at once; use :=//$</$^; use CRT_SRC/CRT_OBJ/SCRT_OBJ/CSRC/COBJ/SSRC/SOBJ/MSRC/MOBJ where no more is needed, if only CSRC is present use OBJS directly instead of COBJ; CTOR_TARGETS are created directly in lib; remove unused/unneeded parts. Hope I haven't broken too much.
2005-01-25merge parallel build supportMike Frysinger
2004-08-16The variable used to store pagesize is not the same as theEric Andersen
_dl_pagesize variable in ldso, so avoid aliasing. -Erik
2004-08-13Fixup PAGE_SIZE problemsEric Andersen
-Erik
2004-07-15Use MAP_PRIVATE whenever __ARCH_HAS_MMU__ is set.Eric Andersen
Only use MAP_SHARED when mmu-less.
2004-01-16s/UCLIBC_HAS_MMU/ARCH_HAS_MMU/gEric Andersen
2003-12-30Rework malloc. The new default implementation is based on dlmalloc from DougEric Andersen
Lea. It is about 2x faster than the old malloc-930716, and behave itself much better -- it will properly release memory back to the system, and it uses a combination of brk() for small allocations and mmap() for larger allocations. -Erik
2003-12-27Fix a long-standing bug with pthreads. A couple of linuxthreads filesManuel Novoa III
were including libc-lock.h which had a bunch of weak pragmas. Also, uClibc supplied a number of no-op weak thread functions even though many weren't needed. This combined result was that sometimes the functional versions of thread functions in pthread would not override the weaks in libc. While fixing this, I also prepended double-underscore to all necessary weak thread funcs in uClibc, and removed all unused weaks. I did a test build, but haven't tested this since these changes are a backport from my working tree. I did test the changes there and no longer need to explicitly add -lpthread in the perl build for perl to pass its thread self tests.
2003-10-18Peter Kjellerstedt writes:Eric Andersen
rm.patch: * Define $(RM) as rm -f in Rules.mak and test/Rules.mak (this is the same definition as gmake uses by default). * Change all occurrences of rm and rm -f into $(RM).
2003-10-16Code formatting cleanup.Miles Bader
2003-10-16Don't set errno if for zero SIZE returns.Miles Bader
Code formatting cleanup.
2003-09-22More fiddling with static free-areas: make sure both their size _and_Miles Bader
their alignment are correct.
2003-09-19Make sure we don't allocate too little space for static free-areasMiles Bader
because of our fiddling with alignment (because doing so is VERY BAD).
2003-09-19Add dependencies on header files.Miles Bader
2003-09-17Make sure static heaps are aligned correctly.Miles Bader
2003-09-06'extern inline' doesn't work... Use 'static inline'Eric Andersen
2003-09-06add missing prototypeEric Andersen
2003-09-06Fix errno values. Fix MALLOC_GLIBC_COMPAT handling in malloc/malloc.c,Eric Andersen
which was reversed. Provide more consistancy between implementations. Handle it when people do stupid things like malloc(-1);
2003-08-08Add in a MALLOC_GLIBC_COMPAT option to let people decide if theyEric Andersen
want glibc style malloc(0) behavior
2003-04-02(free_to_heap): Unlock the heap before returning in theMiles Bader
__UCLIBC_UCLINUX_BROKEN_MUNMAP__ case. Fix from YYang1@Winbond.com.tw.
2003-01-23Update malloc behavior on malloc(0) to be consistant withEric Andersen
malloc-930716 behavior, i.e. return a NULL.
2002-11-21Debugging tweaks.Miles Bader