Hook up proper fork lock handling for malloc:
- lock all relevant mutexes just before fork
- unlock all mutexes just after fork in the parent
- full reinit non-spinlocks in the child
This is not usin

Hook up proper fork lock handling for malloc:
- lock all relevant mutexes just before fork
- unlock all mutexes just after fork in the parent
- full reinit non-spinlocks in the child
This is not using the normal pthread_atfork interface to ensure order of
operation, malloc is used as implementation detail too often.

show more ...

use getpagesize(3).

Avoid sysconf: __sysconf -> sysctlgetmibinfo -> strtoimax -> locale, etc.

make this smaller.

make more descriptors that we open as close-on-exec

typo in comment, from Abhinav Upadhyay in PR 47385

cleanup saving and restoring errno, and make it similar.

don't let readlink trash errno.;

fix posix_memalign; we are not going to support alignments > pagesize in
this implementation.

make this compile again

move decls for _malloc_{pre,post}fork() to extern.h

Sync phk malloc.c with the latest one from FreeBSD. This gives us the
same api as jemalloc by:

- adding posix_memalign()
- moving calloc in the same file
- renaming malloc_options -> _ma

Sync phk malloc.c with the latest one from FreeBSD. This gives us the
same api as jemalloc by:

- adding posix_memalign()
- moving calloc in the same file
- renaming malloc_options -> _malloc_options

show more ...

fix spelling of accidentally; from Zapher

PR/34602: sushant: Bug in malloc implementation contains dead code in
free_pages()

Applied patch from Mindaugas <unex at linija org> (in #NetBSD-code), thanks

__inline__ -> inline

WARNS=4

Add missing __RCSID()

Style fixes to reduce diff size against the FreeBSD version, notably:
- Use NULL where appropriate.
- No need to pull in stdint.h.
- Get rid of "register" keywords in the hope that the compiler wi

Style fixes to reduce diff size against the FreeBSD version, notably:
- Use NULL where appropriate.
- No need to pull in stdint.h.
- Get rid of "register" keywords in the hope that the compiler will
do a better job for us. :-)
- Every message ends with a period.
- MMAP() macro is supposed to return MAP_FAILED rather than -1 if it
fails.
- De-__P.

show more ...

call malloc_init() on first call to free/realloc, to make malloc.conf to
take effect. in response to PR 22223

Merge the nathanw_sa branch.

Add a couple of missing THREAD_UNLOCK's, for the error exit path.

No functional change on Mainline, however SA branch probably needs this.

Fix signed/unsigned comparison warnings.

More fixes:

1. use uintptr_t instead of u_long
2. check for overflow in map_pages and malloc_pages
3. bring in fixes from FreeBSD [int -> size_t, and a missing THREAD_UNLOCK]
4. rewrite map_pages to

More fixes:

1. use uintptr_t instead of u_long
2. check for overflow in map_pages and malloc_pages
3. bring in fixes from FreeBSD [int -> size_t, and a missing THREAD_UNLOCK]
4. rewrite map_pages to use sbrk() only to grow memory (avoids extra syscall
and elides bug in brk(2) that ross is fixing)
5. restore the break point to its original value if the mmap(2) for the page
directory or the alignment sbrk breaks.

reviewed by: chuq and ross
tested by: make build and reboot
Now memtest nearly works; unfortunately there is no way currently to lower
the break point as we free, so memtest keeps trying to reduce memory when
mlock() fails and that does not work.

show more ...

PR/12810: Chris ?: malloc core-dumps when given large number as the argument.
This is because integer overflow occurs in the computation of the size of
the page directory array. We now detect that, a

PR/12810: Chris ?: malloc core-dumps when given large number as the argument.
This is because integer overflow occurs in the computation of the size of
the page directory array. We now detect that, and return ENOMEM.

show more ...

convert from __progname to getprogname()