github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-10 07:28:15 -05:00
Code Issues Packages Projects Releases Wiki Activity

nv: do not use va_addr for cpu accesses (#12697)

Browse Source 
* nv: do not use va_addr for cpu accesses

* mypy
This commit is contained in:
nimlgen
2025-10-15 22:48:12 +08:00
committed by GitHub
parent 3ab23af829
commit db5ae846aa
1 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
tinygrad/runtime/ops_nv.py
View File

@@ -135,7 +135,7 @@ class NVComputeQueue(NVCommandQueue):
qmd_buf.cpu_view().view(size=prg.qmd.mv.nbytes, fmt='B')[:] = prg.qmd.mv
assert qmd_buf.va_addr < (1 << 40), f"large qmd addr {qmd_buf.va_addr:x}"
qmd = QMD(dev=prg.dev, addr=cast(int, qmd_buf.va_addr)) # Save qmd for later update
qmd = QMD(dev=prg.dev, addr=qmd_buf.cpu_view().addr) # Save qmd for later update
self.bind_sints_to_mem(*global_size, mem=qmd_buf.cpu_view(), fmt='I', offset=qmd.field_offset('cta_raster_width' if qmd.ver<4 else 'grid_width'))
self.bind_sints_to_mem(*(local_size[:2]), mem=qmd_buf.cpu_view(), fmt='H', offset=qmd.field_offset('cta_thread_dimension0'))
@@ -519,7 +519,7 @@ class NVDevice(HCQCompiled[HCQSignal]):
self.cmdq_page:HCQBuffer = self.iface.alloc(0x200000, cpu_access=True)
self.cmdq_allocator = BumpAllocator(size=self.cmdq_page.size, base=cast(int, self.cmdq_page.va_addr), wrap=True)
self.cmdq = MMIOInterface(cast(int, self.cmdq_page.va_addr), 0x200000, fmt='I')
self.cmdq = self.cmdq_page.cpu_view().view(fmt='I')
self.num_gpcs, self.num_tpc_per_gpc, self.num_sm_per_tpc, self.max_warps_per_sm, self.sm_version = self._query_gpu_info('num_gpcs',
'num_tpc_per_gpc', 'num_sm_per_tpc', 'max_warps_per_sm', 'sm_version')
@@ -552,8 +552,8 @@ class NVDevice(HCQCompiled[HCQSignal]):
nv_gpu.NVC36F_CTRL_CMD_GPFIFO_GET_WORK_SUBMIT_TOKEN_PARAMS(workSubmitToken=-1))
self.iface.setup_gpfifo_vm(gpfifo)
return GPFifo(ring=MMIOInterface(gpfifo_area.va_addr + offset, entries*8, fmt='Q'), entries_count=entries, token=ws_token_params.workSubmitToken,
controls=nv_gpu.AmpereAControlGPFifo.from_address(gpfifo_area.va_addr + offset + entries * 8))
return GPFifo(ring=gpfifo_area.cpu_view().view(offset, entries*8, fmt='Q'), entries_count=entries, token=ws_token_params.workSubmitToken,
controls=nv_gpu.AmpereAControlGPFifo.from_address(gpfifo_area.cpu_view().addr + offset + entries * 8))
def _query_gpu_info(self, *reqs):
nvrs = [getattr(nv_gpu,'NV2080_CTRL_GR_INFO_INDEX_'+r.upper(), getattr(nv_gpu,'NV2080_CTRL_GR_INFO_INDEX_LITTER_'+r.upper(), None)) for r in reqs]