Lines Matching refs:executing
727 wavefronts executing on the same device will access the same memory for any
729 executing on different devices will access different memory. It is higher
1812 executing in a 32-bit process
1817 executing in wavefront 32 mode.
1821 executing in a 64-bit process
1826 executing in wavefront 64 mode.
1837 when executing in wavefront 32
1841 executing in wavefront 32 mode.*
1843 when executing in wavefront 64
1847 executing in wavefront 64 mode.*
1853 when executing in wavefront 32
1857 executing in wavefront 32 mode.*
1859 when executing in wavefront 32
1863 executing in wavefront 32 mode.*
1865 when executing in wavefront 64
1869 executing in wavefront 64 mode.*
1871 when executing in wavefront 64
1875 executing in wavefront 64 mode.*
1883 register corresponding to the lane that is executing the current thread of
1889 mode register definitions are used. Some AMDGPU targets support executing in
1985 local address space, then it corresponds to the wavefront that is executing the
1987 private address space, then it corresponds to the lane that is executing the
2000 specify the local address space corresponding to the wavefront that is executing
2004 to specify the private address space corresponding to the lane that is executing
2010 that is executing the focused thread of execution. The wavefront view of private
2384 restore the ``EXEC`` mask from the saved value. While executing the whole
3698 CPU host program, or from an HSA kernel executing on a GPU.
3731 8. CP ensures that when the a wavefront starts executing the kernel machine
3739 before continuing executing the machine code that corresponds to the kernel.
4120 Start executing wavefront
4180 Start executing wavefront
4200 Start executing wavefront
4508 …3:0 4 bits SHARED_VGPR_COUNT Number of shared VGPR blocks when executing in sub…
4637 actually executing.
4870 being managed by SPI for the queue executing the kernel dispatch. This is
4918 memory being managed by SPI for the queue executing the kernel dispatch.
4937 backing memory being managed by SPI for the queue executing the kernel
5126 executing on it.
5147 wavefronts executing in different work-groups as they may be executing on
5158 scalar memory operations performed by wavefronts executing in different
5159 work-groups (which may be executing on different CUs) of an agent can be
5162 ensures a previous vector memory operation has completed before executing a
6219 before executing
6316 before executing
6388 executing on it. The exception is when in tgsplit execution mode when no LDS
6419 executing in different work-groups as they may be executing on different
6432 scalar memory operations performed by wavefronts executing in different
6433 work-groups (which may be executing on different CUs), or the same
6434 work-group if executing in tgsplit mode, of an agent can be reordered
6437 ensures a previous vector memory operation has completed before executing a
8601 before executing
8704 before executing
8776 executing on it. The exception is when in tgsplit execution mode when no LDS
8807 between wavefronts executing in different work-groups as they may be
8808 executing on different CUs.
8829 executing with different L1 caches and the same L2 cache can be reordered
8833 memory operation has completed before executing a subsequent vector memory
11020 before executing
11123 before executing
11198 executing on it.
11218 wavefronts executing in the same work-group as they may be executing on SIMDs
11220 required for coherence between wavefronts executing in different work-groups
11221 as they may be executing on different WGPs.
11229 required for coherence between wavefronts executing in different work-groups
11230 as they may be executing on different SAs that access different L1s.
11235 executing in the same or different work-groups (which may be executing on
11239 ensures a previous vector memory operation has completed before executing a
11247 and scalar memory operations performed by wavefronts executing in different
11248 work-groups (which may be executing on different SAs) of an agent can be
11252 before executing a subsequent vector memory and so can be used to meet the
13012 before executing
13105 before executing
13213 before executing
13300 the error state, the waves executing
13342 the error state, the waves executing
13383 … the error state, the waves executing
14063 … executing this pipeline. Typically, the ``reg offsets`` are the