aboutsummaryrefslogtreecommitdiff
path: root/src/or/buffers.c
Commit message (Collapse)AuthorAge
...
* r17947@catbus: nickm | 2008-02-06 11:57:53 -0500Nick Mathewson2008-02-06
| | | | | | | Fix a bunch of DOCDOC items; document the --quiet flag; refactor a couple of XXXX020 items. svn:r13405
* r17936@catbus: nickm | 2008-02-06 00:31:11 -0500Nick Mathewson2008-02-06
| | | | | | | Fix/downgrade some more XXX020s. svn:r13397
* r15890@tombo: nickm | 2008-01-12 17:19:51 -0500Nick Mathewson2008-01-13
| | | | | | | Add a --disable-buffer-freelists configure argument to build without buffer RAM chunk freelists, so we can benchmark the impact of this. svn:r13121
* r17467@catbus: nickm | 2008-01-05 19:54:20 -0500Nick Mathewson2008-01-06
| | | | | | | fix whitespace svn:r13035
* r15750@tombo: nickm | 2007-12-29 19:12:51 -0500Nick Mathewson2007-12-30
| | | | | | | Make HTTP-processing code in buffers.c faster: do not pull up 50K of text into the first chunk if the headers end earlier than that. svn:r13006
* r17433@catbus: nickm | 2007-12-29 12:35:57 -0500Nick Mathewson2007-12-29
| | | | | | | Incomplete code to avoid doing needless pull-ups on HTTP. Also, use memstr instead of strstr to find the content-length header. svn:r13005
* r15745@tombo: nickm | 2007-12-29 00:15:17 -0500Nick Mathewson2007-12-29
| | | | | | | Try even harder not to allocate chunks bigger than MAX_CHUNK_ALLOC. svn:r13003
* r17426@catbus: nickm | 2007-12-28 21:12:29 -0500Nick Mathewson2007-12-29
| | | | | | | Remove need for buf_pullup in fetch_line_from_buf(). svn:r13002
* r17410@catbus: nickm | 2007-12-27 09:20:27 -0500Nick Mathewson2007-12-27
| | | | | | | Exciting new entries in buf_dump_freelist_sizes() to make sure our freelist sizes are reasonable. svn:r12996
* r17407@catbus: nickm | 2007-12-27 00:18:31 -0500Nick Mathewson2007-12-27
| | | | | | | Fix windows compile error svn:r12995
* r17405@catbus: nickm | 2007-12-26 22:27:37 -0500Nick Mathewson2007-12-27
| | | | | | | And clean up an assertion failure. svn:r12993
* r17403@catbus: nickm | 2007-12-26 22:23:50 -0500Nick Mathewson2007-12-27
| | | | | | | Apparently loops work better when you actually increment/decrement the loop variable. Who knew? svn:r12992
* r15723@tombo: nickm | 2007-12-26 19:25:51 -0500Nick Mathewson2007-12-27
| | | | | | | Actually trim freelists. svn:r12991
* r15721@tombo: nickm | 2007-12-26 17:07:10 -0500Nick Mathewson2007-12-26
| | | | | | | Fix a nasty infinite loop in flush_buf[_tls]. svn:r12990
* r15719@tombo: nickm | 2007-12-26 14:02:12 -0500Nick Mathewson2007-12-26
| | | | | | | More tweaks on 0-byte TLS writes. svn:r12989
* r15717@tombo: nickm | 2007-12-26 13:55:53 -0500Nick Mathewson2007-12-26
| | | | | | | Oops. flush_buf_tls can request more than the requested number of bytes. When that happens, do not let the size_t sz wrap around. svn:r12988
* r15715@tombo: nickm | 2007-12-26 13:09:27 -0500Nick Mathewson2007-12-26
| | | | | | | Peacetime managed to run overnight in paranoia mode. Disabling paranoia, so that buffers.c can be a little speedy again. svn:r12987
* r15710@tombo: nickm | 2007-12-25 19:36:03 -0500Nick Mathewson2007-12-26
| | | | | | | Fix in flush_buf_tls: it is okay to flush an empty buffer, since we may have a partial TLS record pending. svn:r12984
* r15693@tombo: nickm | 2007-12-25 19:11:29 -0500Nick Mathewson2007-12-26
| | | | | | | Here, have some terribly clever new buffer code. It uses a mbuf-like strategy rather than a ring buffer strategy, so it should require far far less extra memory to hold any given amount of data. Also, it avoids access patterns like x=malloc(1024);x=realloc(x,1048576);x=realloc(x,1024);append_to_freelist(x) that might have been contributing to memory fragmentation. I've tested it out a little on peacetime, and it seems to work so far. If you want to benchmark it for speed, make sure to remove the #define PARANOIA; #define NOINLINE macros at the head of the module. svn:r12983
* clean up copyrights, and assign 2007 copyrights to The Tor Project, IncRoger Dingledine2007-12-12
| | | | svn:r12786
* patch from karsten to clean up documentation and to integrateRoger Dingledine2007-12-07
| | | | | | | more fixes into rend-spec.txt. svn:r12715
* r16997@catbus: nickm | 2007-12-06 18:56:33 -0500Nick Mathewson2007-12-06
| | | | | | | Make http://torproject urls into https://torproject urls. Patch from steve. svn:r12708
* r16995@catbus: nickm | 2007-12-06 18:42:53 -0500Nick Mathewson2007-12-06
| | | | | | | patch from steve: replace more tor.eff.org instances with torproject.org svn:r12707
* easy tweaks on r12607Roger Dingledine2007-11-29
| | | | svn:r12608
* karsten's second refactoring patchRoger Dingledine2007-11-29
| | | | svn:r12607
* r14727@31-33-67: nickm | 2007-11-05 18:34:35 -0500Nick Mathewson2007-11-05
| | | | | | | Make VERSIONS variable-length. svn:r12391
* r16438@catbus: nickm | 2007-11-05 16:45:45 -0500Nick Mathewson2007-11-05
| | | | | | | Initial code for variable-length cells. CERT and VERSIONS need to use them. svn:r12390
* Separate "SOCKS_COMMAND_CONNECT_DIR" into two flags inRoger Dingledine2007-10-28
| | | | | | | | | | | | edge_connection_t: want_onehop if it must attach to a circuit with only one hop (e.g. for the current tunnelled connections that use begin_dir), and use_begindir if we mean to use a BEGIN_DIR relay command to establish the stream rather than the normal BEGIN. Now we can make anonymized begin_dir connections for (e.g.) more secure hidden service posting and fetching. svn:r12244
* Say "using socksX _to_ port y" instead of "socksX _on_ port y". it confuses ↵Peter Palfrader2007-10-10
| | | | | | weasel less svn:r11834
* r14830@catbus: nickm | 2007-08-29 13:50:10 -0400Nick Mathewson2007-08-29
| | | | | | | Make controllers accept LF as well as CRLF. Update spec to reflect this. Remove now-dead code. Make controller warning about v0 protocol more accurate. svn:r11299
* r14034@catbus: nickm | 2007-07-31 05:22:56 -0400Nick Mathewson2007-07-31
| | | | | | | possible fix for bug 473: always set buf->cur on buf_remove_from_front. svn:r11004
* Disable SENTINELS in buffers: it may be responsible for making platform ↵Nick Mathewson2007-07-30
| | | | | | mallocs() be inefficient, and we havent needed it in a long long time. svn:r10994
* Yet another buffer RAM patch: tNever ever ever keep a buffer memory chunk ↵Nick Mathewson2007-07-30
| | | | | | around for an empty buffer that could go on the freelist. This wants profiling to make sure that performance doesnt suffer. svn:r10993
* r13952@catbus: nickm | 2007-07-27 19:18:46 -0400Nick Mathewson2007-07-27
| | | | | | | Weasel noticed that many buffers spend their time with empty 4k, 8k, and 16k memory chunks. Thus, be more aggressive about putting empty chunks on the freelist, regardless of their high water marks. (Also, run buffer_shrink_freelist on the 8k-chunk freelist.) svn:r10953
* r13944@catbus: nickm | 2007-07-27 15:52:35 -0400Nick Mathewson2007-07-27
| | | | | | | | | Fix warnings on platforms where rlim values can be signed. Add an 8k buffer freelist. svn:r10948
* r13939@catbus: nickm | 2007-07-27 14:33:22 -0400Nick Mathewson2007-07-27
| | | | | | | When dumping memory usage, list bytes used in buffer memory free-lists. svn:r10947
* free bridge list on exit; try harder to free buffer freelists on exit.Roger Dingledine2007-07-17
| | | | svn:r10854
* free another string, and the buffer freelists, on exit.Roger Dingledine2007-07-17
| | | | svn:r10851
* r13477@catbus: nickm | 2007-06-17 14:22:03 -0400Nick Mathewson2007-06-17
| | | | | | | Sun CC likes to give warnings for the do { } while(0) construction for making statement-like macros. Define STMT_BEGIN/STMT_END macros that do the right thing, and use them everywhere. svn:r10645
* r13274@catbus: nickm | 2007-06-05 19:06:30 -0400Nick Mathewson2007-06-06
| | | | | | | Remove some deadcode that seems unlikely to return svn:r10509
* r12934@catbus: nickm | 2007-05-24 14:11:18 -0400Nick Mathewson2007-05-24
| | | | | | | When freeing slack buffers, free the buffers at the end of the freelist first. svn:r10313
* r12764@catbus: nickm | 2007-05-15 17:17:39 -0400Nick Mathewson2007-05-15
| | | | | | | Enable (and cope with) more GCC 4.2 warnings. svn:r10196
* r12935@Kushana: nickm | 2007-05-10 11:25:09 -0400Nick Mathewson2007-05-10
| | | | | | | Make buffer unit tests pass again svn:r10158
* fix a printf size_t/int mismatch from r10150Roger Dingledine2007-05-10
| | | | svn:r10152
* r12706@catbus: nickm | 2007-05-09 18:39:46 -0400Nick Mathewson2007-05-09
| | | | | | | Keep two freelists for buffer ram chunks: one of 4k chunks, and one of 16k chunks. Also, document the whole business. svn:r10150
* r12704@catbus: nickm | 2007-05-09 17:43:34 -0400Nick Mathewson2007-05-09
| | | | | | | Periodically clean the freelist of buffer memory chunks. svn:r10149
* simplify connection_watch_events()Roger Dingledine2007-04-25
| | | | | | | hope this doesn't break it svn:r10025
* r12784@Kushana: nickm | 2007-04-23 19:24:09 -0400Nick Mathewson2007-04-23
| | | | | | | Avoid another assert failure in the new buffer memory code. (This time, it hit when freeing a 4k buffer with some data on it.) svn:r10011
* r12782@Kushana: nickm | 2007-04-23 18:20:19 -0400Nick Mathewson2007-04-23
| | | | | | | Avoid a segfault when freeing a buffer with size 0. svn:r10010
* r12499@catbus: nickm | 2007-04-23 10:42:23 -0400Nick Mathewson2007-04-23
| | | | | | | Keep a freelist of unused 4k buffer chunks, rather than wasting 8k for every inactive connection_t. svn:r10006
generated by cgit v1.2.3 (git 2.25.4) at 2024-09-24 12:30:26 +0000