leitner
|
d020e6d88b
|
another gcc 6 warning fix
compile on osx again
remove asm version of i?mult64 (was needed only for clang because it
didn't have __uint128_t, but it now has it)
|
9 years ago |
|
leitner
|
d26b8082d8
|
shut up new gcc 6 warnings
|
9 years ago |
|
leitner
|
05e4dcc94a
|
stupid sign extension
|
9 years ago |
|
leitner
|
624412ce03
|
fix two issues found by clang's static analyser
|
9 years ago |
|
leitner
|
83712b7f23
|
attempt to support FreeBSD
|
9 years ago |
|
leitner
|
755ebfcb53
|
-fomit-leaf-frame-pointer was wrong even for gcc m(
|
9 years ago |
|
leitner
|
ffc648cc14
|
typo
|
9 years ago |
|
leitner
|
2208bdff68
|
gcc -> clang
|
9 years ago |
|
leitner
|
a1f1986247
|
include <foo> -> include "foo"
|
9 years ago |
|
leitner
|
48592a2400
|
-I.
|
9 years ago |
|
leitner
|
09c7eac41f
|
#include <> to ""
|
9 years ago |
|
leitner
|
dace178f25
|
add base64url support (forgot header)
compile byte/* with -O3
|
9 years ago |
|
leitner
|
7f829a36bc
|
add base64url support
|
9 years ago |
|
leitner
|
9ed7f82dc8
|
ar -> $(CROSS)ar
|
9 years ago |
|
leitner
|
ed507aad14
|
support $DESTDIR for make install targets
|
9 years ago |
|
leitner
|
fb6c39be63
|
special case buffer_get_token with token length 1 through memccpy (almost 4x speedup)
|
9 years ago |
|
leitner
|
9cfa19396e
|
special case buffer_get_token with token length 1 through memccpy
|
9 years ago |
|
leitner
|
9b2377c4be
|
fix operator precedence
|
9 years ago |
|
leitner
|
80a9933ff1
|
use static inline instead of extern inline
|
9 years ago |
|
leitner
|
774a4b6714
|
add unistd.h
|
9 years ago |
|
leitner
|
685ebdfb80
|
oops, did not actually flush the buffer
|
9 years ago |
|
leitner
|
a0cde0fee9
|
slight optimization for buffer_flush
|
9 years ago |
|
leitner
|
189d8bad1d
|
darwin compat
|
9 years ago |
|
leitner
|
f74f9b9b25
|
add socket_udp4b and socket_udp6b
|
9 years ago |
|
leitner
|
32a312b8f1
|
add marshaling code for protocol buffers
|
9 years ago |
|
leitner
|
6859c4b6ca
|
add netstring formatter
fix scan_netstring to actually correctly return the length of the transported string
add documentation
|
9 years ago |
|
leitner
|
fb1f19042f
|
support gcc 5 builtins for multiplication overflow in safemult.h
fix copypasted comment in fmt_asn1dertag
add man pages for add_of, sub_of and assign
|
9 years ago |
|
leitner
|
6cfc16fc40
|
add gcc 5 integer overflow intrinsics support to rangecheck.h
in the process, identify a gnarly problem that was not only not found by
the test suite, there was a test testing for the wrong behavior!
|
9 years ago |
|
leitner
|
7bdc1b29bb
|
a few missing headers and minor fixes
fix corking behavior for the case where a buffer follows a buffer in
iob_send
|
9 years ago |
|
leitner
|
cdecd16ab3
|
add documentation
|
10 years ago |
|
leitner
|
e5865bcc52
|
fix timezone handling
|
10 years ago |
|
leitner
|
ac6979e4af
|
add scan_iso8601+test
|
10 years ago |
|
leitner
|
ad5ce55281
|
add fmt_iso8601
|
10 years ago |
|
leitner
|
695843c79e
|
small buffer optimization
fix botched noipv6 case in buffer_tcp6
|
10 years ago |
|
leitner
|
214ad3a92e
|
use gcc magic and a #define to convert buffer_puts to buffer_put if the
string argument length is known at compile time (same for
buffer_putsflush)
|
10 years ago |
|
leitner
|
fce165fdcb
|
remove dead store
|
10 years ago |
|
leitner
|
c523dcbc48
|
also install critbit.h
documentation clarification
|
10 years ago |
|
leitner
|
1b3e55485b
|
make critbit0_contains as pure
|
10 years ago |
|
leitner
|
3dff6a2630
|
bump version
|
10 years ago |
|
leitner
|
6d4c3850d6
|
update dependency
|
10 years ago |
|
leitner
|
8a6186ba03
|
add critbit
|
10 years ago |
|
leitner
|
42eb4c9830
|
use xz instead of bz2 in make tar
|
10 years ago |
|
leitner
|
12caae78b2
|
cleaner make clean
|
10 years ago |
|
leitner
|
10affedd80
|
update Makefile
|
10 years ago |
|
leitner
|
79f534ffdf
|
SECURITY: fix botched integer overflow handling logic in stralloc_ready (Giorgio)
|
10 years ago |
|
leitner
|
d0e735e2c2
|
support noipv6 being set externally even when there is perfectly good
IPv6 support.
|
10 years ago |
|
leitner
|
1856c8a2d7
|
damn it, fucked up the return value
|
10 years ago |
|
leitner
|
6e022f98c9
|
... but if errno==EINVAL still fall back to socket+fcntl (Robert Henney)
|
10 years ago |
|
leitner
|
35f5a9692a
|
if SOCK_NONBLOCK is defined, use it instead of socket+fcntl
|
10 years ago |
|
leitner
|
a255cebc57
|
mmap_read/mmap_shared on zero length files no longer fail but return a
zero length buffer
|
10 years ago |