summaryrefslogtreecommitdiffstats
path: root/Modules/FindCUDA
Commit message (Collapse)AuthorAgeFilesLines
* FindCUDA: Filter unrelated content in compute capabilities outputSoumith Chintala2018-09-191-0/+3
| | | | | | | Working around CUDA-level nvrm_gpu log statements to stdout on some embedded platforms (ex. Drive PX2). See-also: https://github.com/pytorch/pytorch/issues/11518#issue-359113249
* FindCUDA/select_compute_arch: Restore two-component CUDA_VERSIONBrad King2018-08-011-2/+3
| | | | | | | | | | Since commit v3.12.0-rc1~332^2 (FindCUDA/select_compute_arch: Add support for CUDA as a language, 2018-03-15) this module sets `CUDA_VERSION` based on `CMAKE_CUDA_COMPILER_VERSION` when the language is enabled. Limit it to two components for consistency with the normal `FindCUDA` documentation and behavior. Fixes: #18231
* CUDA/select_compute_arch: Fix detection on old CUDA with new card/driverHenry Schreiner2018-04-201-13/+45
| | | | | | | | Fix version comparisons to handle patch components. List and check known archs for each version of CUDA so mismatching versions are not suggested. Fixes: #17921
* FindCUDA/select_compute_arch: Add support for CUDA as a languageHenry Fredrick Schreiner2018-03-201-5/+20
| | | | | | | Even though this is an internal module, we can still prepare it to be used in another public-facing module outside of `FindCUDA`. Issue: #16586
* Merge topic 'msvc2017-findcuda'Brad King2018-01-091-1/+1
|\ | | | | | | | | | | | | | | fab1b432 FindCUDA: Update to properly find MSVC 2017 compiler tools Acked-by: Kitware Robot <kwrobot@kitware.com> Acked-by: Robert Maynard <robert.maynard@kitware.com> Merge-request: !1631
| * FindCUDA: Update to properly find MSVC 2017 compiler toolsAaron Bray2018-01-081-1/+1
| | | | | | | | | | An implementation that handles the different installation locations of visual studio compiler tools
* | Various typo fixesLuz Paz2018-01-032-2/+2
|/ | | | Some are user-facing. Others are source comments.
* Fix trivial typos in textluzpaz2017-11-031-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | Some are user facing. Found using codespell -q 3 --skip="./Utilities" -I .cmake-whitelist.txt` whereby the whitelist contained: ans dum helpfull emmited emmitted buil iff isnt nto ot pathes substract te todays upto whitespaces
* FindCUDA: Fix CMP0007 warning in run_nvcc.cmakeUmar Arshad2017-10-041-5/+4
| | | | | | | | Set `CMP0007` to `NEW` for the entire script. The script comes with CMake and is aware of the policy's behavior, so set the policy to ensure that its warnings do not show up during the build phase. Fixes: #16579
* FindCUDA: Use long brackets to handle spaces in CUDA_NVCC_COMPILE_DEFINITIONSRobert Maynard2017-08-151-1/+1
| | | | Fixes #16510. Quoting of compile definitions breaks string value definitions
* FindCUDA: Add CUDA9/Volta recognitionBoris Fomitchev2017-08-101-2/+10
|
* FindCUDA: Fix PTX selection with multiple architecturesBjoern Thiel2017-02-221-0/+1
|
* FindCUDA: Fix typo in CUDA_DETECT_INSTALLED_GPUS doc stringPierre Moreau2017-02-021-1/+1
|
* FindCUDA: Enable CUDA_SELECT_NVCC_ARCH_FLAGS Auto for MSVCBjoern Thiel2017-01-251-9/+11
| | | | Fixes: #16525
* FindCUDA: Support `2.1(2.0)` architecture notationBoris Fomitchev2016-08-231-4/+4
| | | | Also tweak some messages and comments.
* FindCUDA: Take NVCC include directories from target propertiesPeter Boettcher2016-08-161-1/+16
| | | | | | | | Fixes issue where include directories specified on the target are not passed on to NVCC. This includes both target_include_directories() as well as include directories added by dependency chaining. Closes: #14201
* Use string(APPEND) in ModulesDaniel Pfeifer2016-07-272-2/+2
| | | | | | | Automate with: find Modules -type f -print0 | xargs -0 perl -i -0pe \ 's/set\(([a-zA-Z0-9_]+)(\s+)"\$\{\1\}([^"])/string(APPEND \1\2"\3/g'
* FindCUDA: Refactor Android(Tegra) supportBoris Fomitchev2016-06-091-0/+195
| | | | Reviewed-by: James Bigler <jamesbigler@gmail.com>
* FindCUDA: Only warn about non-existent dependency files in verbose modeJames Bigler2016-02-082-1/+14
|
* FindCUDA: Added some additional comments about non-existent dependency files.James Bigler2016-02-051-0/+2
|
* FindCUDA: Fix for when a non-existent dependency file is found.James Bigler2016-02-051-1/+1
| | | | | | | | | | Previously if a non-existent dependency file is found we set the file to "" and then do if(NOT IS_DIRECTORY "${file}"). Later we call get_filename_component on the empty file which returns basically the current build directory. Having a dependency on the current build directory is really annoying, because anything that compiles into that directory will change the file stamp and cause your files to rebuild every time you call make. :(
* FindCUDA: Add specific cuda_language_flag instead of using nvcc.James Bigler2015-04-091-0/+2
| | | | | | I was previously appending to nvcc_flags inside the file loop. This caused the flag to be appended multiple times which freaks out nvcc. Now the flag is specifically handled per file.
* FindCUDA: Avoid if() auto-dereference in string comparisonsAdam Strzelecki2014-09-241-1/+1
| | | | This silences possible CMP0054 warnings.
* FindCUDA: Fix OSX Clang & no C language enabledAdam Strzelecki2014-06-301-2/+2
| | | | | | | | | | | When setting default CUDA_HOST_COMPILER we must dereference CMAKE_C_COMPILER, i.e. /usr/bin/clang should be used instead /usr/bin/cc which is symlink. Otherwise CUDA thinks it is GCC and issues -dumpspecs which is unknown option to Clang. Also in case neither CMAKE_C_COMPILER is defined (project does not use C language) nor CUDA_HOST_COMPILER is specified manually we should skip -ccbin and let nvcc use its own default C compiler.
* Replace MATCHES ".+" tests with NOT STREQUAL ""Rolf Eike Beer2014-04-142-2/+2
|
* Replace string(REGEX REPLACE) with string(REPLACE) where possibleRolf Eike Beer2014-04-142-5/+3
| | | | The simple replacement is much faster.
* Clean up usage of if(... MATCHES regex) followed string(REGEX REPLACE regex)Rolf Eike Beer2014-04-141-4/+4
| | | | | | The matches have already been calculated and can simply be taken from CMAKE_MATCH_n variables. This avoids multiple compilations of the same or very similar regular expressions.
* Fix spelling and typos (affecting binary data / module messages)Andreas Mohr2013-05-071-1/+1
|
* Merge topic 'topics/FindCUDA/Add-CUDA_HOST_COMPILER'David Cole2012-09-181-2/+10
|\ | | | | | | | | db207e4 FindCUDA: Added CUDA_HOST_COMPILER variable.
| * FindCUDA: Added CUDA_HOST_COMPILER variable.James Bigler2012-08-201-2/+10
| | | | | | | | | | | | | | | | | | Added a new CUDA variable for specifying the CUDA_HOST_COMPILER. This will allow users to be able to specify which host compiler to use for invoking NVCC with. By default it will use the compiler used for host compilation. This is convenient for when you want to specify a different compiler than the default compiler. You end up using the same compiler for both the NVCC compilation and the host compilation instead of using the default compiler in the path for NVCC.
* | Remove CMake-language block-end command argumentsKitware Robot2012-08-133-10/+10
|/ | | | | | | | | | | | | | | | | Ancient versions of CMake required else(), endif(), and similar block termination commands to have arguments matching the command starting the block. This is no longer the preferred style. Run the following shell code: for c in else endif endforeach endfunction endmacro endwhile; do echo 's/\b'"$c"'\(\s*\)(.\+)/'"$c"'\1()/' done >convert.sed && git ls-files -z -- bootstrap '*.cmake' '*.cmake.in' '*CMakeLists.txt' | egrep -z -v '^(Utilities/cm|Source/kwsys/)' | egrep -z -v 'Tests/CMakeTests/While-Endwhile-' | xargs -0 sed -i -f convert.sed && rm convert.sed
* Merge topic 'topics/FindCUDA/Quote-fixes'David Cole2011-12-061-15/+15
|\ | | | | | | | | c3c7a0c Fixes for handling quotes in args and other places (Fix Bug 11726 and 12099).
| * Fixes for handling quotes in args and other places (Fix Bug 11726 and 12099).James Bigler2011-12-061-15/+15
| |
* | Add work around for CUDA in UNC paths.James Bigler2011-12-051-2/+16
|/ | | | | | Nvcc can emit '/path' instead of '//path' which can cause a lot of grief later. We test to see if the file exists, if it doesn't then we see if the file exists with '/' prepended. Files that don't exist won't be added to the list.
* Define __CUDACC__ during dependency scanning, since NVCC doesn't.James Bigler2010-01-281-0/+7
|
* Reordered the arguments, so that user generated arguments go last and thus ↵James Bigler2010-01-081-7/+7
| | | | get precedence.
* Don't allow -G -M on CUDA versions less than 3.0.James Bigler2009-11-301-1/+15
|
* Fix Xcode build.James Bigler2009-10-071-6/+0
| | | | | Move the make_directory command to the main target, so that CMAKE_CFG_INTDIR will get expanded by the build tool.
* Updated copyright notice to conform to NVIDIA guidelines.James Bigler2009-10-053-6/+3
|
* Updated the license on top of each file.James Bigler2009-09-293-48/+93
|
* Convert FindCUDA helper newlines from CRLF to LFBrad King2009-09-282-309/+309
| | | | | These files were committed to the repository with Windows newlines. This converts them to Unix newlines so they will show up natively.
* Added a command to make the output directory. This is to fix the XCode ↵James Bigler2009-09-231-0/+6
| | | | build that uses a different output directory than other systems, and rather than try to match that we'll just make it.
* Initial version of FindCUDA script. Still needs documentation formatting.James Bigler2009-09-153-0/+408
generated by cgit v0.12 at 2024-11-10 20:16:27 (GMT)