- Oct 24, 2016
-
-
Luker authoredThis was originally meant only to eradicate the interleaving code, so that the raw API could be slightly faster. Turns out a lot had to be done for that, and the API changed to reflect those changes. The RAW_Encoder template now has enabled/disabled methods, so that you can only use the right one, depending on whether you choose interleaving or not. While doing all of this, I realized a nice way to let the user choose the correct symbol number for the block. +1 for typesafety. And since the commit became really big, I threw in a couple of useless style changes, bringing back everything to 80 columns. Really, I should be doing smaller commits. Signed-off-by:Luca Fulchir <luker@fenrirproject.org>
-
- Oct 18, 2016
-
-
Luker authored
Ticket #3 not yet finished, blocked by #7 Signed-off-by:
-
- Oct 09, 2016
-
-
Luker authored
Start portability for VC++14. Still unfinished. Also simplify a bit locking in RFC calls. Signed-off-by:
-
- Sep 24, 2016
-
-
Luker authored
Freebsd/Mac support (completed?), better checks for linkers/profiling, disable CLI tools if requested. Signed-off-by: -
Luker authored
Also disable LTO/profiling for macOSX by default. Buy me a mac, and I might be able to support it better :P Signed-off-by: -
Luker authored
Signed-off-by:
-
- Sep 21, 2016
-
-
Luker authored
See #9 . Uses new RAW API. Completely untested. At least it compiles. Also, did I forget to push Bitmask.hpp? Awkward... Signed-off-by:
-
- Sep 08, 2016
-
-
Luker authored
Signed-off-by:
-
- Sep 06, 2016
-
-
Luker authored
Also, cleanup Cmake. Signed-off-by:
-
- Sep 02, 2016
-
-
Luker authored
caches makes sense for both RaptorQ and RFC6330 namespaces Signed-off-by:
-
- Sep 01, 2016
-
-
Luker authored
LZ4 dependency is now optional, even with caching. Signed-off-by:
-
- Jul 17, 2016
-
-
Luker authored
Note: somehow there is still some interleaving. Which is really bad. There's probably something wrong in both the interleaver and de-interleaver. Signed-off-by:
-
- Jul 14, 2016
-
-
Luker authored
Signed-off-by: -
Luker authored
Signed-off-by: -
Luker authored
Signed-off-by:
-
- Jul 11, 2016
-
-
Luker authored
Signed-off-by:
-
- Jul 06, 2016
-
-
Luker authored
API and decoder still wip Signed-off-by:
-
- Jul 02, 2016
-
-
Luker authored
Signed-off-by:
-
- Jul 01, 2016
-
-
Luker authored
Groundwork started. Started moving things to RFC6330. Next steps will be cleanup & RaptorQ namespaces. Meanwhile we need better file naming. Supporting multiple versions of the same API should be easier now. Signed-off-by:
-
- Jun 24, 2016
-
-
Luker authored
Still, decoding does not work. Signed-off-by: -
Luker authored
Signed-off-by: -
Luker authored
tell people to get the LZ4 library correctly Signed-off-by:
-
- Jun 02, 2016
-
-
Luker authored
Build, make deterministic and statically link the lz4 library if necessary. Signed-off-by: -
Luker authored
It is nice to have syntax highlighting on big files... Signed-off-by: -
Luker authored
port cmake differences from tag v0.1.6 Signed-off-by:
-
- May 30, 2016
-
-
Luker authored
Precomputation not yet working. Need to fix either matrix generation or computation. Signed-off-by: -
Luker authored
Signed-off-by: -
Luker authored
Quite a messy commit, hopefully new ones should be shorter Signed-off-by: -
Luker authored
How come nobody told me I broke everything? Signed-off-by: -
Luker authored
Support for offline precomputation, without overhead if not needed. Not tested yet. Signed-off-by:
-
- Mar 12, 2016
-
-
Luker authored
Sorry to ubuntu12/centos guys, but you don't even have a c++11 compiler, so you can also upgrade the cmake version. Signed-off-by:
-
- Mar 11, 2016
-
-
Luker authored
Ubuntu guys, either fix CMake or update it -.-' Signed-off-by:
-
- Mar 07, 2016
-
-
Luker authored
Rewrote cmake, now a bit cleaner, checks for compiler flags. Added Eigen3, v3.2.8 in case it's not found on the system. Signed-off-by:
-
- Mar 03, 2016
-
-
Luker authored
test_c did not correctly calculate the memory size, although it was still fine due to the data types. the De_Interleaver was reading not-initialized data in the "skip > 0" case. Signed-off-by:
-
- Oct 27, 2015
-
-
Luker authored
Signed-off-by:
-
- Jun 12, 2015
-
-
Luker authored
Strange but true. Let things compile for them, too. Also: threads are required now, not optional Signed-off-by:
-
- Jun 11, 2015
-
-
Luker authored
worked on ArchLinux (cmake 3), plain wrong on Ubuntu (cmake 2.8) things should be ok now. Signed-off-by:
-
- May 12, 2015
-
-
Luker authored
used cppcheck, valgrind on the examples. fix the C interface free(), the example, the build type select. update the cppcheck files, add the clang scanbuild script Signed-off-by: -
Luker authored
Use link time optimization with gcc and clang. Also: add LTO to documentation, and warn that clang and profiling break deterministic builds Signed-off-by:
-
- May 05, 2015
-
-
Luker authored
Enable Profiling for both gcc and clang. This should give us a small boost in performance, specially on few symbols per block. Also: mark CMake features as advanced and use a forward declaration instead of a struct in the C interface. Signed-off-by:
-