| .. |
|
Makefile.am
|
9da0482007
r17358@pc-10-8-1-079: nickm | 2008-07-25 16:41:03 +0200
|
17 years ago |
|
OpenBSD_malloc_Linux.c
|
1a564901e4
Fix the rest of the GCC warnings on OpenBSD_malloc_linux.c
|
17 years ago |
|
address.c
|
122170c1d3
Downlgrade tweak, and answer lots of XXX021s. No actual code fixes in this patch.
|
17 years ago |
|
address.h
|
f6879caa04
Try once again to make BSD compilation happy.
|
17 years ago |
|
aes.c
|
11e464c331
r18753@catbus: nickm | 2008-03-11 14:56:39 -0400
|
17 years ago |
|
aes.h
|
eecc44dab8
r17963@catbus: nickm | 2008-02-07 10:14:25 -0500
|
17 years ago |
|
ciphers.inc
|
d395135e2f
fix a few typos, and give the bootstrap phase stuff a changelog entry.
|
17 years ago |
|
compat.c
|
98066d62bc
Lower sprintf buffer max to ~SSIZE_T_MAX from SIZE_T_CEILING, since we need to compare it to a signed int.
|
17 years ago |
|
compat.h
|
4277b0e926
Remove some cargo-cult gcc hacks around tor_assert and predict_unlikely; instead, use the standard convert-to-boolean hack of "svn st"
|
17 years ago |
|
container.c
|
191197eff7
Revert my older supposed gcc-4.4 warning workaround. GCC was not being needlessly prissy; it was hinting at the wrongly pure smartlist_bsearch_idx.
|
17 years ago |
|
container.h
|
651a0a2fb5
Stop marking bsearch_idx as pure; it is not.
|
17 years ago |
|
crypto.c
|
6c6b0283cb
Ben confirms that the MUST in rfc2631 is only for compatibility with X9.42, and isn't actually a security thing.
|
17 years ago |
|
crypto.h
|
cebdf93949
Fix bug 889: share deep-copied keys between threads to avoid races in reference counts. Bugfix on 0.1.0.1-rc.
|
17 years ago |
|
ht.h
|
69300eb606
r14374@tombo: nickm | 2008-02-21 16:57:39 -0500
|
17 years ago |
|
log.c
|
b6f89a647a
One log.c XXX021 was a misunderstanding. Also, clip log messages passed to syslog to their maximum length when there is a maximum.
|
17 years ago |
|
log.h
|
bd6b3072f9
Change logging code to use fds instead of stdio. Fixes bug 861, and probably makes logging slightly faster. Not a backport candidate: bug 861 is too obscure and harmless.
|
17 years ago |
|
memarea.c
|
6693f32530
Resolve many DOCDOCs.
|
17 years ago |
|
memarea.h
|
3ebd1ebeca
The chunk_size field in memarea_t was never actually set. Remove the whole thing.
|
17 years ago |
|
mempool.c
|
2905291af2
r17659@tombo: nickm | 2008-08-06 12:22:11 -0400
|
17 years ago |
|
mempool.h
|
f3eaeb99a3
r18051@catbus: nickm | 2008-02-12 15:20:43 -0500
|
17 years ago |
|
strlcat.c
|
29fa9eb216
r8860@totoro: nickm | 2006-10-03 14:48:33 -0400
|
19 years ago |
|
strlcpy.c
|
29fa9eb216
r8860@totoro: nickm | 2006-10-03 14:48:33 -0400
|
19 years ago |
|
test.h
|
3be88b2c70
Change test_memeq macro to not leak memory. Addresses coverity CID 47.
|
17 years ago |
|
torgzip.c
|
69300eb606
r14374@tombo: nickm | 2008-02-21 16:57:39 -0500
|
17 years ago |
|
torgzip.h
|
842a33ff20
Update some copyright notices: it is now 2008.
|
17 years ago |
|
torint.h
|
35bef7fefd
make read_all and write_all return ssize_t.
|
17 years ago |
|
tortls.c
|
9c3d17ebb5
Fix a small memory leak of around 32 bytes per TLS connection opened. Bugfix on 0.2.1.1-alpha.
|
17 years ago |
|
tortls.h
|
122170c1d3
Downlgrade tweak, and answer lots of XXX021s. No actual code fixes in this patch.
|
17 years ago |
|
util.c
|
bf80e2df3f
Replace calls to time(NULL) that occur on the order of once per read, one per write, or once per cell with calls to a function that looks at a cached value of time. This is tricksy to benchmark, since it will only help on systems where time() is a syscall and syscalls are relatively slow.
|
17 years ago |
|
util.h
|
bf80e2df3f
Replace calls to time(NULL) that occur on the order of once per read, one per write, or once per cell with calls to a function that looks at a cached value of time. This is tricksy to benchmark, since it will only help on systems where time() is a syscall and syscalls are relatively slow.
|
17 years ago |
Nick Mathewson
bf80e2df3f
Replace calls to time(NULL) that occur on the order of once per read, one per write, or once per cell with calls to a function that looks at a cached value of time. This is tricksy to benchmark, since it will only help on systems where time() is a syscall and syscalls are relatively slow.