Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -702,7 +702,7 @@
"The host code below shows the construction of the TV layout. By composing\n",
"a thread layout of ``(4,64):(64,1)`` (64 threads read contiguous elements on the row dimension,\n",
"then 64-thread-groups(2 warps) read different rows) with a value layout of ``(16,8):(8,1)`` (each thread reads\n",
"8 contiguous 16b elements on the row dimension across 4 contiguous rows).\n",
"8 contiguous 16b elements on the row dimension across 16 contiguous rows).\n",
"\n",
"In order to generalize, we started with byte-layout to describe layout for elements in bytes. This is\n",
"to ensure use of 128bit vectorized load store. Then we leverage ``recast_layout`` to convert into\n",
Expand Down