| History log of /src/lib/libc/stdlib/malloc.c |
| Revision | | Date | Author | Comments |
| 1.60 |
| 15-May-2020 |
joerg | 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.
|
| 1.59 |
| 13-Jan-2017 |
christos | use getpagesize(3).
|
| 1.58 |
| 12-Jan-2017 |
christos | Avoid sysconf: __sysconf -> sysctlgetmibinfo -> strtoimax -> locale, etc.
|
| 1.57 |
| 12-Jan-2017 |
christos | make this smaller.
|
| 1.56 |
| 18-Sep-2014 |
christos | branches: 1.56.2;
make more descriptors that we open as close-on-exec
|
| 1.55 |
| 30-Dec-2012 |
dholland | typo in comment, from Abhinav Upadhyay in PR 47385
|
| 1.54 |
| 18-May-2011 |
christos | branches: 1.54.4; 1.54.10;
cleanup saving and restoring errno, and make it similar.
|
| 1.53 |
| 13-May-2011 |
christos | don't let readlink trash errno.;
|
| 1.52 |
| 03-Feb-2008 |
christos | fix posix_memalign; we are not going to support alignments > pagesize in
this implementation.
|
| 1.51 |
| 12-Dec-2007 |
macallan | make this compile again
|
| 1.50 |
| 04-Dec-2007 |
christos | move decls for _malloc_{pre,post}fork() to extern.h
|
| 1.49 |
| 28-Nov-2007 |
christos | 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
|
| 1.48 |
| 24-Nov-2006 |
christos | branches: 1.48.8;
fix spelling of accidentally; from Zapher
|
| 1.47 |
| 08-Oct-2006 |
elad | 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
|
| 1.46 |
| 24-Dec-2005 |
perry | __inline__ -> inline
|
| 1.45 |
| 29-Nov-2005 |
christos | WARNS=4
|
| 1.44 |
| 12-Jun-2005 |
lukem | Add missing __RCSID()
|
| 1.43 |
| 05-Oct-2003 |
junyoung | 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.
|
| 1.42 |
| 23-Jul-2003 |
itojun | call malloc_init() on first call to free/realloc, to make malloc.conf to
take effect. in response to PR 22223
|
| 1.41 |
| 18-Jan-2003 |
thorpej | Merge the nathanw_sa branch.
|
| 1.40 |
| 09-Dec-2002 |
chris | Add a couple of missing THREAD_UNLOCK's, for the error exit path.
No functional change on Mainline, however SA branch probably needs this.
|
| 1.39 |
| 11-Nov-2002 |
thorpej | Fix signed/unsigned comparison warnings.
|
| 1.38 |
| 06-May-2001 |
christos | 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.
|
| 1.37 |
| 03-May-2001 |
christos | 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.
|
| 1.36 |
| 19-Feb-2001 |
cgd | branches: 1.36.2;
convert from __progname to getprogname()
|
| 1.35 |
| 05-Jan-2001 |
jdolecek | we now have utrace, support it here
|
| 1.34 |
| 20-Dec-2000 |
christos | fix sbrk casts
|
| 1.33 |
| 06-Jul-2000 |
christos | avoid losing precision warnings
|
| 1.32 |
| 20-May-2000 |
simonb | Disable malloc's hints to the kernel - now that the madvise() syscall
actually does something, this can be expensive. From discussion on
current-users.
Note that the man page is already correct - we imported malloc.c from
FreeBSD before they made the same change but malloc.3 after...
|
| 1.31 |
| 07-Apr-2000 |
enami | Fix the bug introduced in rev 1.28 and actually caused failure on alpha;
In the function malloc_make_chunks, don't define the local variable ``l'',
which is used to hold the actual size of pginfo, as size_t. Define it as
a signed integer as before. We expect it to become zero or negative value,
and we know that it is much smaller than pagesize and it is small enough to
fit in signed interger.
|
| 1.30 |
| 07-Apr-2000 |
enami | Backout the previous backing out.
|
| 1.29 |
| 31-Jan-2000 |
thorpej | Back out the previous 2 revs. The delint'ing of this file caused
something to break horribly on the Alpha.
|
| 1.28 |
| 23-Jan-2000 |
mycroft | Delint.
|
| 1.27 |
| 23-Jan-2000 |
mycroft | Partially delint, and a few microoptimizations.
|
| 1.26 |
| 10-Sep-1999 |
kleink | Restore the behaviour of not setting errno to ENOMEM when allocating 0
units of storage and returning a null pointer in System V mode; this was
broken by the `fix' in rev. 1.24. Also, as it is stated in ISO C that
such operation does not constitute an allocation failure, do not abort()
even if the `X' option is set.
Amusingly enough the SVID, Fourth Edition, specifies the `unique pointer'
return behaviour for this kind of allocation, so this is kind of mis-named.
|
| 1.25 |
| 22-Aug-1999 |
kleink | Need "namespace.h".
|
| 1.24 |
| 08-Jul-1999 |
thorpej | Make sure errno is set to ENOMEM if we return NULL.
|
| 1.23 |
| 05-Jul-1999 |
thorpej | index -> ldx (and similar in some related variable names). Avoids a lint
warning.
|
| 1.22 |
| 05-Jul-1999 |
thorpej | Oops, back out 1.21. Christos and I apparently stepped on each other.
|
| 1.21 |
| 05-Jul-1999 |
christos | s/index/idx/
s/EXTRA_SANITY/MALLOC_EXTRA_SANITY/
const poisoning
use sysconf to get the pagesize
|
| 1.20 |
| 05-Jul-1999 |
thorpej | Compute page size related parameters at run-time.
|
| 1.19 |
| 05-Jul-1999 |
thorpej | EXTRA_SANITY -> MALLOC_EXTRA_SANITY, consistently.
|
| 1.18 |
| 04-Jul-1999 |
thorpej | Only define MADV_FREE do MADV_DONTNEED if it's not already defined
by <sys/mman.h>.
|
| 1.17 |
| 26-Jun-1999 |
tls | The 'new malloc' (phk's malloc, from FreeBSD) is now our only malloc.
|
| 1.16 |
| 29-Jan-1999 |
kleink | Protect access of state with a mutex. Miscellaneous notes:
* still needs an implementation-independent ("reentrant.h") method for
disabling cancellation,
* break a leg to leave realloc() `storage compaction' as was for non-reentrant
code,
* use of stdio in assertion code is unsafe.
|
| 1.15 |
| 15-Nov-1998 |
christos | delint
|
| 1.14 |
| 26-Aug-1998 |
perry | make memcpy into memmove in some cases
|
| 1.13 |
| 10-Aug-1998 |
perry | bzero->memset, bcopy->memcpy, bcmp->memcmp
|
| 1.12 |
| 03-Feb-1998 |
perry | remove obsolete register declarations
|
| 1.11 |
| 30-Jan-1998 |
perry | update to lite-2
|
| 1.10 |
| 21-Jul-1997 |
jtc | If port provides __weak_alias(), provide an Standard C and POSIX pure
identifier namespace by renaming non standard functions and variables
such that they have a leading underscore. The library will use those
names internally. Weak aliases are used to provide the original names
to the API.
This is only the first part of this change. It is most of the functions
which are implemented in C for all NetBSD ports. Subsequent changes are
to add the same support to the remaining C files, to assembly files, and
to the automagically generated assembly source used for system calls.
When all of the above is done, ports with weak alias support should add
a definition for __weak_alias to <sys/cdefs.h>.
|
| 1.9 |
| 13-Jul-1997 |
christos | Add local.h for local prototypes.
Fix namespace issues for strtoq and strtouq
Fix gcc warnings.
Fix RCSID's
|
| 1.8 |
| 07-Apr-1997 |
christos | PR/3448: Bill Sommerfeld: Change "p" to __STRING(p)
|
| 1.7 |
| 20-Dec-1996 |
cgd | don't confuse long and int; pointed out by lint
|
| 1.6 |
| 17-Jan-1996 |
jtc | branches: 1.6.2;
The C standard says (about realloc) that "If size is zero and ptr is
not a null pointer, the object it points to is freed." Reported by
Peter Seebach in PR #1806.
|
| 1.5 |
| 28-Dec-1995 |
thorpej | New-style RCS ids.
|
| 1.4 |
| 19-Oct-1994 |
cgd | branches: 1.4.2;
casting types.
|
| 1.3 |
| 26-Aug-1993 |
jtc | Declare rcsid strings so they are stored in text segment.
|
| 1.2 |
| 01-Aug-1993 |
mycroft | Add RCS identifiers.
|
| 1.1 |
| 21-Mar-1993 |
cgd | branches: 1.1.1;
Initial revision
|
| 1.1.1.2 |
| 30-Jan-1998 |
perry | import lite-2
|
| 1.1.1.1 |
| 21-Mar-1993 |
cgd | initial import of 386bsd-0.1 sources
|
| 1.4.2.1 |
| 02-May-1995 |
jtc | #include "namespace.h"
|
| 1.6.2.1 |
| 20-Sep-1996 |
jtc | snapshot namespace cleanup: net, etc.
|
| 1.36.2.6 |
| 10-Dec-2002 |
thorpej | Sync with HEAD.
|
| 1.36.2.5 |
| 25-Mar-2002 |
nathanw | Bind libc's pthread desires to _libc_pthread_*, rather than pthread_*, to
prevent accidental use of pthread libraries that aren't actually integrated
with libc.
|
| 1.36.2.4 |
| 11-Mar-2002 |
nathanw | Use an ordinary mutex for malloc now that internal spinlocks aren't exported.
|
| 1.36.2.3 |
| 01-Jan-2002 |
wdk | pt_spin_t has been renamed to pthread_spin_t
|
| 1.36.2.2 |
| 08-Oct-2001 |
nathanw | Catch up to -current.
|
| 1.36.2.1 |
| 08-Aug-2001 |
nathanw | Add THREAD_LOCK stubs that use the library's spinlocks.
(XXX why spinlocks instead of a normal mutex?)
|
| 1.48.8.2 |
| 23-Mar-2008 |
matt | sync with HEAD
|
| 1.48.8.1 |
| 09-Jan-2008 |
matt | sync with HEAD
|
| 1.54.10.1 |
| 25-Feb-2013 |
tls | resync with head
|
| 1.54.4.1 |
| 23-Jan-2013 |
yamt | sync with head
|
| 1.56.2.1 |
| 20-Mar-2017 |
pgoyette | Sync with HEAD
|
