Commit Graph

51 Commits

Author SHA1 Message Date
Dr-Noob
ed35cb872b [v0.25] Leave cuda/intel backend to decide how to report PCI vendor failure 2023-03-31 16:16:46 +02:00
Dr-Noob
68619aa03e [v0.25] Avoid segfault when the pci vendor is not found 2023-03-31 15:50:37 +02:00
Dr-Noob
06dc50b6a5 [v0.25] Updated cuda_helper to support latest GPUs 2022-12-03 16:39:18 +00:00
Dr-Noob
a6f0c18fcb [v0.25] Add missing Ampere GPU chips and new uarchs: ada and hopper 2022-10-25 20:13:29 +02:00
Dr-Noob
94490b3f38 [v0.24] Fix typo in error message (thanks #22 and #28) 2022-10-25 19:41:46 +02:00
Dr-Noob
5faac7a756 [v0.24] Update PCI ids to pciutils/pciids@06c4c9a 2022-10-25 19:30:24 +02:00
Dr-Noob
4d948eb80a [v0.24] Remove CUDA driver initialization message before printing any other message 2022-05-21 23:19:03 +02:00
Dr-Noob
af52d2850c [v0.24] Remove cuda-samples dependency 2022-04-17 13:55:05 +02:00
Dr-Noob
bd1158c139 [v0.23] Sort PCI devices; this makes the devices list to match CUDA driver ordering, which fixes a bug when there was more than one NVIDIA GPU 2022-01-22 13:25:22 +01:00
Dr-Noob
23586a18e9 [v0.22] Fix for previous commit (dont show tensor cores in TU116) 2022-01-20 22:57:19 +01:00
Dr-Noob
d3aaf7cfe5 [v0.22] Do not show tensor cores in TU116 2022-01-12 19:34:11 +01:00
Dr-Noob
98a70d5c9e [v0.21] Print only one error message when the GPU chip is not found in the LUT 2021-12-28 16:21:04 +01:00
Dr-Noob
8fbf97c47a [v0.21] Add verbose option. Fix CUDA driver initialization message when verbose output is used 2021-12-27 22:37:51 +01:00
Dr-Noob
4e0e6b5ab5 [v0.20] Rename CUDA topology struct to improve consistency 2021-12-19 11:34:05 +01:00
Dr-Noob
e7477610e1 [v0.20] Always use C++ includes when possible for consistency 2021-12-19 11:30:10 +01:00
Dr-Noob
3e730468d8 [v0.20] Fixes from previous commit 2021-12-19 10:18:23 +01:00
Dr-Noob
981bfabdc8 [v0.20] Merge Intel iGPU branch for preeliminary Intel GPU support 2021-12-19 10:11:23 +01:00
Dr-Noob
a397eb398e [v0.11] Handle the case where the GPU is not found in the pci LUT 2021-12-18 20:12:41 +01:00
Dr-Noob
6d4d8b621b [v0.11] Fix compilation error and ambiguity with CUDA and Intel backend when enabled at the same time due to functions with the same name 2021-12-18 10:14:14 +01:00
Dr-Noob
844377f17a [v0.11] Add support for printing EUs (currently only in Gen9/Gen9.5) 2021-12-08 11:15:59 +01:00
Dr-Noob
38b8949e1c [v0.11] Fix tensor cores calculation for Ampere. Add a brief explanation 2021-11-30 16:03:36 +01:00
Dr-Noob
2034bac006 [v0.11] Displaying Graphics Tier in Intel iGPUs 2021-11-27 14:02:02 +01:00
Dr-Noob
b00050e739 [v0.11] Print available more information for iGPU 2021-11-27 11:22:16 +01:00
Dr-Noob
8740337145 [v0.11] Adding uarch backend for intel iGPUs 2021-11-26 12:52:45 +01:00
Dr-Noob
310486a6a2 [v0.11] Fixes to recover CUDA functionality, ready for implementing Intel iGPU code 2021-11-26 09:33:57 +01:00
Dr-Noob
461e0d2ede [v0.11] Working in master GPU handler for supporting diverse GPU vendors 2021-11-26 08:22:30 +01:00
Dr-Noob
149e5ad62c [v0.11] Working for future support of Intel iGPUs 2021-11-25 19:03:52 +01:00
Dr-Noob
32b2c59b50 [v0.11] Add peak performance with tensor cores to the output 2021-11-23 18:49:34 +01:00
Dr-Noob
8bf0276aae [v0.10] Simple refactoring 2021-11-23 18:17:12 +01:00
Dr-Noob
821b6e760e [v0.10] Add support for displaying the number of tensor cores 2021-11-23 18:09:13 +01:00
Dr-Noob
039e7c350d [v0.10] Replace nvml by pciutils to get pci ids. Needs work to integrate it properly. NVML is enough in the case of NVIDIA GPUs, but because more GPUs will be added in the future, a solution like pciutils is needed 2021-09-04 12:19:42 +02:00
Dr-Noob
4b4d1bc030 [v0.10] Add --list-gpus option 2021-08-23 22:39:31 +02:00
Dr-Noob
a7036218a0 [v0.06] Add check in memory guess to allow guessing only valid memory types for the uarch 2021-08-17 15:09:13 +02:00
Dr-Noob
ea11da2368 [v0.05] Discard specific chips endings when printing the chip to stdout, which looks cleaner 2021-08-16 18:36:45 +02:00
Dr-Noob
dea5211791 [v0.04] Fix bug in which NVML was using the wrong GPU id 2021-08-16 15:18:55 +02:00
Dr-Noob
fd15008ab2 [v0.04] Add error checking for CUDA calls 2021-08-16 13:52:57 +02:00
Dr-Noob
c7a08e7fd0 [v0.04] Add option to select GPU 2021-08-16 13:41:39 +02:00
Dr-Noob
eecc040e98 [v0.04] Small fixes to presentation 2021-08-16 13:11:54 +02:00
Dr-Noob
7ad484b938 [v0.03] Add printer backend from cpufetch (still printing CPU logo) 2021-08-15 23:23:36 +02:00
Dr-Noob
8f31748d1e [v0.02] Fix typos. Divide memory detection in two phases. Show real memory frequency, instead of the returned by CUDA 2021-08-15 22:17:46 +02:00
Dr-Noob
8386052b10 [v0.02] Add warning message. Hide the waiting message when CUDA initialization is done. Fix compiler warnings 2021-08-15 21:56:51 +02:00
Dr-Noob
2687fa5016 [v0.02] Experimental algorithm to detect memory type finished. Information gathering finished 2021-08-15 21:22:41 +02:00
Dr-Noob
7d505dcc07 [v0.01] Experimental feature: guess data rate from gpu and memory clocks 2021-08-14 09:47:16 +02:00
Dr-Noob
10295f2955 [v0.01] Retrieve memory info. All fields done except memory type (how to do it?) 2021-08-13 22:26:55 +02:00
Dr-Noob
981daef728 [v0.01] Retrieve more info; frequency, topology and peak performance 2021-08-13 16:36:10 +02:00
Dr-Noob
59f723ff12 [v0.01] Deduce manufacturing process, chip string and uarch from gpu chip 2021-08-13 16:04:54 +02:00
Dr-Noob
a08c06f67a [v0.01] First approach to map pci ids to gpu chip/uarch 2021-08-13 13:50:25 +02:00
Dr-Noob
bdc4fd7c45 [v0.01] Reading PCI vendor/device values, needed to deduce the GPU chip and thus manufacturing process 2021-08-13 11:31:51 +02:00
Dr-Noob
8bc37d9d71 [v0.01] Detecting uarch. Looks like getting manufacturing process is much harder since we need GPU chip name 2021-08-12 15:51:12 +02:00
Dr-Noob
a15f20a2cc [v0.01] Defining a first approach of the application general structure 2021-08-11 11:34:47 +02:00