ed0d9049dfAdded an option to install squish library for external use (also made the same for BC6H and BC7 libs).
Sukender
2016-09-16 10:25:08 +0200
02e3438c34Added an (advanced) CMake option to build the whole squish library (with an adequate warning to tell this is not the 'official' way).
Sukender
2016-09-16 10:23:32 +0200
752f78b56aFixed compilation and namespaces of non-included squish source files, for future optional inclusion.
Sukender
2016-09-16 10:21:54 +0200
4a4c6ba882Fixed missing include for MSVC 2015.3.
Sukender
2016-09-16 10:20:51 +0200
d41c7ac33aMerge pull request #243 from fshp/master
Ignacio
2016-09-14 11:37:41 -0700
Fix CMakeLists after rename license and readme files Additions to 831ebec475
Maxim Moseychuk
2016-09-14 13:14:58 +0300
8ac7fdb3f2Initial support for the ktx container. Currently, it only supports BC formats and has no support for key-value pairs. Updated the compress utility with a ktx option and exposed sRGB setting.
Manvydas Šliamka
2016-08-17 00:40:29 +0300
20a0adb665Suppress warning from cmake about policy changes. I'm uncertain whether the OLD behavior or the NEW behavior was intended. Either should be sufficient to suppress the warning, and OLD appears to work.
cfcohen
2015-12-04 20:44:12 -0500
ea2ac753b0Fix undefined behavior in array overflow. The GCC 5.2 warning that brought this to my attention was:
cfcohen
2015-12-04 20:37:04 -0500
f55b7d438aRenable finding of external libraries. That is what cmake is for, right? If these external libraries were intentionally disabled, at least comment out the messages saying that they dependencies couldn't be found, since prior to this commit the message was that there was a problem finding them.
cfcohen
2015-12-04 20:33:01 -0500
fdc798d02bFix minor problem with extern/poshlib not being c++ at all, so -std=c++11 is not a valid option. A cmake 2.8.0 compatible approach that involves CMAKE_CXX_FLAGS causes problems for the CUDA compressor module that _cannot_ accept -std=c++11. This approach seems the most correct, but I'm new to CMake, so there may be a better solution.
cfcohen
2015-12-04 20:27:18 -0500
4c860bf84bAttempted fix for pointer/integer confusion. The warning from GCC was: "ThreadPool.cpp:123:46: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]." It appears that an integer from 0 to threadCount was passed as to a value ultimately assigned to (void *)((Private)p->arg). This commit moved the integer into an unfreed block of memory and passed a pointer to it, but perhaps the threadNumber doesn't really need to be passed at all?
cfcohen
2015-12-04 20:17:11 -0500
47919914a9Suppress warnings about set but unused variables in CUDA compress routines.
cfcohen
2015-12-04 20:10:52 -0500
2d1bbe5415Suppress security warnings about insecure use of nvDebug().
cfcohen
2015-12-04 20:06:18 -0500
9b6191ab0cFix errors linking nvddsview. Undefined reference to nv::Fit::computePrincipalComponent_EigenSolver caused by library ordering (nvmath needs to follow other libraries). Fixed in multiple other places already, but not this one.
cfcohen
2015-12-04 20:02:19 -0500
Justin Meiners
Ignacio
Dmitry Marakasov
Ken Cooke
leper
Sylvain Rochette
Mrkvak
Deleted user
lunkhound
acidicMercury8
Lectem
Sukender
Maxim Moseychuk
Manvydas Šliamka
Kamil Rytarowski
magestik
Timothy Pearson
sanderd17
mpersano
cfcohen