llvm.org GIT mirror llvm / stable docs / AMDGPU / gfx8_data_buf_d16_128.rst
stable

Tree @stable (Download .tar.gz)

gfx8_data_buf_d16_128.rst @stableview markup · raw · history · blame

vdata

16-bit data to store by a buffer instruction.

Size: depends on GFX8 GPU revision:

  • 4 dwords for GFX8.0. This H/W supports no packing.
  • 2 dwords for GFX8.1+. This H/W supports data packing.

Operands: :ref:`v<amdgpu_synid_v>`