summaryrefslogtreecommitdiff
path: root/volk/gen
AgeCommit message (Collapse)Author
2012-04-19volk: updated build system for avx checking supportJosh Blum
updated copy of cpuid.h with the latest from gcc 4.6
2012-04-19volk: make orc a normal arch with overruleJosh Blum
2012-04-19volk: added compile utils and cleanup cmakelistsJosh Blum
2012-04-19volk: working build w/ cmakelistsJosh Blum
2012-04-19volk: created other templates for runtime + machinesJosh Blum
2012-04-19volk: added kernel defs and typedefsJosh Blum
2012-04-19volk: work on template stuffJosh Blum
2012-04-19Merge branch 'maint'Johnathan Corgan
2012-04-18volk: gcc version check without __GNUC_PREREQJosh Blum
2012-04-18volk: added xgetbv stuff from volk_work to maintJosh Blum
This ensures that the compiler has support for xgetbv. This also fixes MSVC by checking for _xgetbv. Also, restored copy of cpuid.h, this should not be modified.
2012-04-16Merge branch 'maint'Johnathan Corgan
Conflicts: volk/gen/make_cpuid_c.py
2012-04-16Volk: also check to make sure OSXSAVE is enabled so you don't check XGETBV ↵Nick Foster
when OS has it disabled.
2012-04-16Volk: add support for checking AVX enable state of OS.Nick Foster
Some systems (notably Xen hypervisor) appear to use XSETBV to disable AVX. This causes SIGILL when running AVX instructions. This commit makes Volk check XCR0 on the AVX arch before proceeding.
2012-04-13Removed whitespace and added dtools/bin/remove-whitespace as a tool to do ↵Tom Rondeau
this in the future. The sed script was provided by Moritz Fischer.
2012-03-22Merge branch 'master' into nextJohnathan Corgan
2012-03-21volk: add include for config.h to cc fileJosh Blum
volk_32f_s32f_convert_16i_a was also a problem, previously volk_32f_s32f_convert_16i_u was fixed to have config.h. I heard putting config h stuff into public headers was bad practice. Including config.h into the generated cc file should fix this issue from now on.
2011-12-08Removed autotools, gr-waveform, some cleanupJohnathan Corgan
Nick Foster owes Nick Corgan a six-pack of beer!
2011-11-30volk: fix #466 deprecated string use, from PinkavaJosh Blum
2011-11-30volk: avx support for MSVCJosh Blum
2011-11-07volk: building on msvc again with sse2 archJosh Blum
2011-11-07volk: work to improve compiler and multi-lib supportJosh Blum
2011-10-20volk: give priority to generated includes (helps with dirty tree)Josh Blum
2011-10-20volk: fixed signed/unsigned comparison warnings.Tom Rondeau
2011-09-17volk: use glob rather than parsing makefile for kernelsJosh Blum
2011-07-22Volk: add in mutex capability to archs.xml. Turns out some compilers can do ↵Nick Foster
both 32-bit and 64-bit compilations, and this breaks Volk. Only works in CMake right now.
2011-07-19volk: fix whitespace in Python scriptMoritz Fischer
2011-05-30volk: make fptrs in volk.h visibleNick Foster
2011-05-20Volk: runtime NEON detection based on /proc/self/auxvNick Foster
2011-05-20Volk: Use "unsafe" math optimizations (read: NEON) in GCC on NEON archNick Foster
2011-05-20Volk: modify NEON test flagNick Foster
2011-05-19Volk: first stab at NEON support. Using compile-time detection.Nick Foster
2011-05-18Volk: added alignment prop to each machine. call volk_get_alignment() to get ↵Nick Foster
your machine's reqd alignment for malloc.
2011-05-18Volk: renamed everything else to _a instead of _a16, makefiles includedNick Foster
2011-05-13volk: do not install library-only headersJosh Blum
2011-05-12Volk: Forgot to put sse4.1 support in sse4_2 and avx machinesNick Foster
2011-05-12Volk: profiling works. loads prefs on init. volk_rank_archs looks in prefs ↵Nick Foster
first.
2011-05-11Volk: Profiler is in apps/ now. Added name to function info. Going to ↵Nick Foster
C++-ify the whole thing.
2011-05-10volk: cmake support for volk (gcc + msvc)Josh Blum
2011-05-10volk: generate two machine structs which are conditional on LV_HAVE_ORCJosh Blum
2011-05-06volk: various backports from MSVC buildingJosh Blum
1) Added support for __cpuid intrinsic under MSVC 2) Fixed disambiguation for std::abs overload in qa code 3) Fixed bit128 union, the ifdefs were completely wrong
2011-05-06volk: top-level common header cleanupJosh Blum
Since we already have a volk_common.h, moved the attributes, API declaration, and c-linkage macros into volk_common.h This change removes volk_attributes.h, in favor of one common include header. The implementation headers that require attributes now include volk_common.h This change removes the emit_omnilog.py script, in favor of pre-processor macros in volk_common.h In addition, extern C is only defined when in C++ and in GCC because non-GCC does not have complex.h
2011-05-06volk: removed volk_registry.h, it was superseded by the machinesJosh Blum
2011-05-06volk: make volk_machine_xxx.cc c-safeJosh Blum
Initialize each machine struct using C constants, (rather than relying on C++ to copy objects). Each machine file is far simpler, because we know what archs are in a machine, we can generate exactly the right code. In addition, rename the file extensions to .c so we know its C only.
2011-05-05volk: make volk_machines.cc c-safeJosh Blum
Initialize the list of machine structs w/ pointers, this does not rely on C++ initialization and is smaller. In addition, rename the file extension to .c so we know its C only. Interesting note, this also fixes a bug when compiling under MSVC, the machines list initialization seemed to have problems prior to this change.
2011-04-26volk: reorganization of generation sources and generated filesJosh Blum
All generation sources have been moved to the gen/ subdirectory. Bootstrap and volk_register.py generate the files into to gen/ subdirectory in an effort to cleanly separate the static/generated parts of the build tree. Define top_gendir in Makefile.common, all generated sources listed in Makefile.ams are prefixed with $(top_gendir) to differentiate them from static in-tree sources.
generated by cgit (git 2.34.1) at 2024-12-21 04:40:21 +0000