-
Notifications
You must be signed in to change notification settings - Fork 39
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Shared object support #15
Commits on Feb 15, 2023
-
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 66642d8 - Browse repository at this point
Copy the full SHA 66642d8View commit details -
fix various errors in the Makefiles that lead to building on a non-cl…
…ean directory not always working Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 403ec5f - Browse repository at this point
Copy the full SHA 403ec5fView commit details -
add test program for cuda code loaded using libdl
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 0e13cbf - Browse repository at this point
Copy the full SHA 0e13cbfView commit details -
when the client dlopens libraries containing cuda kernels, also open …
…them at the server. Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for cb391b3 - Browse repository at this point
Copy the full SHA cb391b3View commit details
Commits on Feb 16, 2023
-
add decoding of fatbinary data
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 905fefe - Browse repository at this point
Copy the full SHA 905fefeView commit details
Commits on Feb 17, 2023
-
add decoding of embedded fatbinaries
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 0997d44 - Browse repository at this point
Copy the full SHA 0997d44View commit details -
add temporary test code that launches a kernel on the server from an …
…elf retrieved via RPC. Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 4ff4b5c - Browse repository at this point
Copy the full SHA 4ff4b5cView commit details
Commits on Feb 18, 2023
-
add registry for tranferred cubins and kernel functions so Cricket is…
… able to identify them when launching kernels Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e72c11c - Browse repository at this point
Copy the full SHA e72c11cView commit details
Commits on Feb 21, 2023
-
fix segfault on cleanup because CUDA accesses nonexisting fatcubinHandle
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 15eb759 - Browse repository at this point
Copy the full SHA 15eb759View commit details -
code cleanup. fix wrong passing of dimensions
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for ff493f6 - Browse repository at this point
Copy the full SHA ff493f6View commit details -
use an infinite timeout for kernel calls
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 98832e2 - Browse repository at this point
Copy the full SHA 98832e2View commit details -
remove timeout for cudaDeviceSynchronize
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 6eeef6c - Browse repository at this point
Copy the full SHA 6eeef6cView commit details
Commits on Mar 9, 2023
-
make cpu_utils_contains_kernel return the right value
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 74179ef - Browse repository at this point
Copy the full SHA 74179efView commit details -
add cudaRegisterVar client function
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 36ead03 - Browse repository at this point
Copy the full SHA 36ead03View commit details -
add gdb commands file for debugging client apps
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for babb70c - Browse repository at this point
Copy the full SHA babb70cView commit details
Commits on Mar 10, 2023
-
fix cpu_utils_contains_kernel and cpu_utils_parameter_info returning …
…the wrong value Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for d1f6173 - Browse repository at this point
Copy the full SHA d1f6173View commit details -
make cpu_utils_launch_child also redirect stderr of child processes t…
…o pipe Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 9cb6aaf - Browse repository at this point
Copy the full SHA 9cb6aafView commit details -
reduce debugging output verbosity and add some NULL checks
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for ac36e85 - Browse repository at this point
Copy the full SHA ac36e85View commit details -
make dlopen return a handle to the main program if it is called with …
…a NULL filename Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 07ed931 - Browse repository at this point
Copy the full SHA 07ed931View commit details -
fix ci error by making tests/cpu/cubin/main.cpp compile
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e9b2c1c - Browse repository at this point
Copy the full SHA e9b2c1cView commit details
Commits on Mar 24, 2023
-
parse kernel parameter infos from in-memory elf using libbfd
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for dec25d1 - Browse repository at this point
Copy the full SHA dec25d1View commit details -
fix cpu-server not using the new name of elf_symbol_address
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 09b34f6 - Browse repository at this point
Copy the full SHA 09b34f6View commit details -
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 701d4bd - Browse repository at this point
Copy the full SHA 701d4bdView commit details
Commits on Mar 27, 2023
-
make higher log levels configurable from makefile
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 89f78e6 - Browse repository at this point
Copy the full SHA 89f78e6View commit details
Commits on Mar 30, 2023
-
add comments and additional error handling
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 4d7dc55 - Browse repository at this point
Copy the full SHA 4d7dc55View commit details -
add elf_init function to avoid multiple initializations of libbfd
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for d9870e0 - Browse repository at this point
Copy the full SHA d9870e0View commit details
Commits on Apr 11, 2023
-
use libelf instead of libbfd for elf manuipulation because of better …
…stability. New code is located in cpu-elf.c Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 45e7e18 - Browse repository at this point
Copy the full SHA 45e7e18View commit details -
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 8de247f - Browse repository at this point
Copy the full SHA 8de247fView commit details
Commits on May 4, 2023
-
migrate to new elf handling. add decompression support for cuda fatbi…
…naries Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 6acdf43 - Browse repository at this point
Copy the full SHA 6acdf43View commit details
Commits on May 10, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 975cd31 - Browse repository at this point
Copy the full SHA 975cd31View commit details -
fix elf handling to work with a wider variety of CUDA kernels
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for eeb8e48 - Browse repository at this point
Copy the full SHA eeb8e48View commit details
Commits on May 11, 2023
-
fix memory leaks identified by gcc sanitizer
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 66eb961 - Browse repository at this point
Copy the full SHA 66eb961View commit details
Commits on May 12, 2023
-
clean up of uneeded code paths relating to old LD_PRELOADing of serve…
…r. Cleanup of shm support Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for b230687 - Browse repository at this point
Copy the full SHA b230687View commit details -
fix cudaMemcpy using correct shm index references
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 6e46154 - Browse repository at this point
Copy the full SHA 6e46154View commit details
Commits on May 15, 2023
-
fix resource manager add_sorted function inserting and wrong location…
… in list leading to binary search failing sometimes. Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 4a4bd02 - Browse repository at this point
Copy the full SHA 4a4bd02View commit details
Commits on May 16, 2023
-
fix wrong decoding of compressed kernels
when walking through compressed cubins we added the decompressed size to the current location instead of the compressed size leading to jumping over important parts of the cubin. Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for bde6500 - Browse repository at this point
Copy the full SHA bde6500View commit details -
update dockerfiles so they install cuda profiler api and add new Dock…
…erfile for CUDA 12.1 Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 33e0fe4 - Browse repository at this point
Copy the full SHA 33e0fe4View commit details -
if a binary does not contain any kernel cricket should not show any e…
…rros. I replaced them with warnings for now. We should probably do better sanity checking before assuming not found .nv.info sections only occur in binaries without kernels. Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for dcd9009 - Browse repository at this point
Copy the full SHA dcd9009View commit details -
cricket supports binaries with debug symbols so we should not throw a…
…n error when we find binaries with debug symbols Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for d944cd9 - Browse repository at this point
Copy the full SHA d944cd9View commit details
Commits on May 18, 2023
-
implement cudaRegisterVar API so that we support cudaMemcpyToSymbol
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for a0473ac - Browse repository at this point
Copy the full SHA a0473acView commit details
Commits on Jun 1, 2023
-
add some driver apis, fix shadowing CUDA functions not working when t…
…heir name is redefined using macros Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 0df2fd3 - Browse repository at this point
Copy the full SHA 0df2fd3View commit details
Commits on Jun 2, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 0641ccc - Browse repository at this point
Copy the full SHA 0641cccView commit details -
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for c9b9726 - Browse repository at this point
Copy the full SHA c9b9726View commit details -
add license to pytorch_minimal.py
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 3b541b3 - Browse repository at this point
Copy the full SHA 3b541b3View commit details -
add nvml library to dockerfiles
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 433930b - Browse repository at this point
Copy the full SHA 433930bView commit details -
exclude some nvml definitions when compiling with an old CUDA version…
… to make the CI happy Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for f249f8f - Browse repository at this point
Copy the full SHA f249f8fView commit details
Commits on Jun 6, 2023
-
er file and initialization of nvml part Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 6860540 - Browse repository at this point
Copy the full SHA 6860540View commit details -
change c standard to gnu11, improve logging
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for c849bd7 - Browse repository at this point
Copy the full SHA c849bd7View commit details
Commits on Jun 7, 2023
-
add documentation on how to use pytorch to docs/pytorch.md
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 4c78904 - Browse repository at this point
Copy the full SHA 4c78904View commit details -
fix elf decompression handling padding wrong in some circumstances
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 5c64748 - Browse repository at this point
Copy the full SHA 5c64748View commit details -
fix decompression not working for long uncompressed lz4 segments
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 1c7d39f - Browse repository at this point
Copy the full SHA 1c7d39fView commit details
Commits on Jun 12, 2023
-
fix potential segfault because of missing variadic parameter in logging
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for c9f09b9 - Browse repository at this point
Copy the full SHA c9f09b9View commit details -
use uint64_t for decompressions to fix overflowing of range and lengt…
…h specifiers for very long compressed segments Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for c709acf - Browse repository at this point
Copy the full SHA c709acfView commit details -
update docs to not deactivate compression as we now support compresse…
…d pytorch kernels Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 56ce060 - Browse repository at this point
Copy the full SHA 56ce060View commit details
Commits on Jun 13, 2023
-
add v2 implementation of cudaGetDeviceProperties
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for da4682e - Browse repository at this point
Copy the full SHA da4682eView commit details -
add libgl dependency to pytorch documentation
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 9f4e797 - Browse repository at this point
Copy the full SHA 9f4e797View commit details -
improve support for cuGetProcAddress
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 8de9fb8 - Browse repository at this point
Copy the full SHA 8de9fb8View commit details
Commits on Jun 14, 2023
-
add cuDNN tests to tests/samples
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for d41d195 - Browse repository at this point
Copy the full SHA d41d195View commit details -
use fixed size rpc array instead of opaque variable length array for …
…cudaDeviceProp Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 523d86e - Browse repository at this point
Copy the full SHA 523d86eView commit details
Commits on Jun 16, 2023
-
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for d786d9c - Browse repository at this point
Copy the full SHA d786d9cView commit details
Commits on Jun 20, 2023
-
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 2de0e2e - Browse repository at this point
Copy the full SHA 2de0e2eView commit details -
use resource managers for cudnn api
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 0a01b07 - Browse repository at this point
Copy the full SHA 0a01b07View commit details -
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 9de7292 - Browse repository at this point
Copy the full SHA 9de7292View commit details -
add cudnn activation and pooling apis
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 32796d7 - Browse repository at this point
Copy the full SHA 32796d7View commit details
Commits on Jun 21, 2023
-
implement cudaMemset Async APIs
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e180369 - Browse repository at this point
Copy the full SHA e180369View commit details -
add cudnn dependency to Dockerfiles
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 14838e6 - Browse repository at this point
Copy the full SHA 14838e6View commit details -
Configuration menu - View commit details
-
Copy full SHA for b392420 - Browse repository at this point
Copy the full SHA b392420View commit details -
add server side cudnn lrn implementations, fix some function names
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 26e19bd - Browse repository at this point
Copy the full SHA 26e19bdView commit details -
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 15fc3a2 - Browse repository at this point
Copy the full SHA 15fc3a2View commit details -
implement cudnn tensor functions
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 762cada - Browse repository at this point
Copy the full SHA 762cadaView commit details
Commits on Jun 22, 2023
-
implement three more cudnn tensor APIs
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 5d381a7 - Browse repository at this point
Copy the full SHA 5d381a7View commit details
Commits on Jun 26, 2023
-
add cublas and cudnn functions to support mnistCUDNN sample
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 6da2f8d - Browse repository at this point
Copy the full SHA 6da2f8dView commit details -
fix faulty if statement when intercepting dlopen calls
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 8a911ca - Browse repository at this point
Copy the full SHA 8a911caView commit details -
improve logging for unloading of modules
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 122b721 - Browse repository at this point
Copy the full SHA 122b721View commit details -
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e8813ea - Browse repository at this point
Copy the full SHA e8813eaView commit details
Commits on Jun 29, 2023
-
improve cublas implementation, add cudnnBackend implementation
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e5dbebf - Browse repository at this point
Copy the full SHA e5dbebfView commit details
Commits on Jul 13, 2023
-
improve debug output for cuModuleLoad
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for ce21d8a - Browse repository at this point
Copy the full SHA ce21d8aView commit details -
add support for cuModuleLoadData
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 481dec9 - Browse repository at this point
Copy the full SHA 481dec9View commit details -
cublas: remove usage of new APIs if we compile for CUDA 10
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for fbf7dad - Browse repository at this point
Copy the full SHA fbf7dadView commit details
Commits on Jul 17, 2023
-
fix using logger function before initialization
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for bf3a15e - Browse repository at this point
Copy the full SHA bf3a15eView commit details -
fix no output on weird shells, e.g. ssh
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for f30d9b0 - Browse repository at this point
Copy the full SHA f30d9b0View commit details
Commits on Jul 18, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 07db2ba - Browse repository at this point
Copy the full SHA 07db2baView commit details -
remove cuda 10 support, add cudnn CI test
Signed-off-by: Niklas Eiling <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 088b6fc - Browse repository at this point
Copy the full SHA 088b6fcView commit details