maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

Mark mlir code snippets as being written in mlir 

Forgot to add these in previous change :/

PiperOrigin-RevId: 221444322
This commit is contained in:
Jacques Pienaar 2018-11-14 07:58:42 -08:00 committed by jpienaar
parent b1f7e03add
commit fb4b74ccb3
2 changed files with 12 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
mlir/g3doc/MLIRForGraphAlgorithms.md
View File

@ -209,7 +209,7 @@ test the dependence analysis infra in the code generator, Andy Davis wrote a
simple pass that checks dependencies and emits them as "notes", allowing him to simple pass that checks dependencies and emits them as "notes", allowing him to
write tests like this: write tests like this:
``` {.mlir} ```mlir {.mlir}
// RUN: mlir-opt %s -memref-dependence-check -verify // RUN: mlir-opt %s -memref-dependence-check -verify
mlfunc @different_memrefs() { mlfunc @different_memrefs() {
%m.a = alloc() : memref<100xf32> %m.a = alloc() : memref<100xf32>
@ -243,7 +243,7 @@ with ShapeRefiner.
The [MLIR Tensor Type](g3doc/LangRef.md#tensor-type) directly captures shape The [MLIR Tensor Type](g3doc/LangRef.md#tensor-type) directly captures shape
information, so you can have things like: information, so you can have things like:
``` {.mlir} ```mlir {.mlir}
%x = tf.Add %x, %y : tensor<128 x 8 x ? x f32> %x = tf.Add %x, %y : tensor<128 x 8 x ? x f32>
``` ```

20
mlir/g3doc/Rationale.md
View File

@ -140,7 +140,7 @@ unknown dimension can be queried using the "dim" builtin as shown below.
Example: Example:
``` {.mlir} ```mlir {.mlir}
mlfunc foo(...) { mlfunc foo(...) {
%A = alloc <8x?xf32, #lmap> (%N) %A = alloc <8x?xf32, #lmap> (%N)
... ...
@ -334,7 +334,7 @@ functions and CFG functions work together.
### Non-affine control flow {#non-affine-control-flow} ### Non-affine control flow {#non-affine-control-flow}
``` {.mlir} ```mlir {.mlir}
// A simple linear search in every row of a matrix // A simple linear search in every row of a matrix
for (i=0; i<N; i++) { for (i=0; i<N; i++) {
for (j=0; j<N; j++) { for (j=0; j<N; j++) {
@ -351,7 +351,7 @@ The presence of dynamic control flow leads to a CFG function nested in an ML
function: an ML function captures the outer loop while the inner loop is function: an ML function captures the outer loop while the inner loop is
represented in the CFG function. represented in the CFG function.
``` {.mlir} ```mlir {.mlir}
mlfunc @search(memref<?x?xi32 %A, <?xi32> %S, i32 %key) { mlfunc @search(memref<?x?xi32 %A, <?xi32> %S, i32 %key) {
%ni = dim %A, 0 : memref<?x?xi32> %ni = dim %A, 0 : memref<?x?xi32>
// This loop can be parallelized // This loop can be parallelized
@ -410,7 +410,7 @@ for (i=0; i <N; i++)
       }        }
``` ```
``` {.mlir} ```mlir {.mlir}
mlfunc @outer_nest(%n) : (i32) { mlfunc @outer_nest(%n) : (i32) {
for %i = 0 to %n { for %i = 0 to %n {
for %j = 0 to %n { for %j = 0 to %n {
@ -442,7 +442,7 @@ The following example illustrates a reference implementation of a 2D
convolution, which uses an integer set `@@domain` to represent valid input data convolution, which uses an integer set `@@domain` to represent valid input data
in a dilated convolution. in a dilated convolution.
``` {.mlir} ```mlir {.mlir}
// Dilation factors S0 and S1 can be constant folded if constant at compile time. // Dilation factors S0 and S1 can be constant folded if constant at compile time.
@@domain = (d0, d1)[S0,S1,S2,S3]: (d0 % S0 == 0, d1 % S1 == 0, d0 >= 0, d1 >= 0, @@domain = (d0, d1)[S0,S1,S2,S3]: (d0 % S0 == 0, d1 % S1 == 0, d0 >= 0, d1 >= 0,
S3 - d0 - 1 >= 0, S4 - d1 - 1 >= 0) S3 - d0 - 1 >= 0, S4 - d1 - 1 >= 0)
@ -556,7 +556,7 @@ which is called a schedule tree. Each non-leaf node of the tree is an abstract
polyhedral dimension corresponding to an abstract fused loop for each ML polyhedral dimension corresponding to an abstract fused loop for each ML
instruction that appears in that branch. Each leaf node is an ML Instruction. instruction that appears in that branch. Each leaf node is an ML Instruction.
``` {.mlir} ```mlir {.mlir}
// A tiled matmul code (128x128x128) represented in schedule tree form // A tiled matmul code (128x128x128) represented in schedule tree form
// #map0 = (d0, d1, d2, d3, d4, d5) -> (128*d0 + d3, 128*d1 + d4, 128*d2 + d5) // #map0 = (d0, d1, d2, d3, d4, d5) -> (128*d0 + d3, 128*d1 + d4, 128*d2 + d5)
@ -641,7 +641,7 @@ dimensions.
Example: Example:
``` {.mlir} ```mlir {.mlir}
// read relation: two elements ( d0 <= r0 <= d0+1 ) // read relation: two elements ( d0 <= r0 <= d0+1 )
##aff_rel9 = (d0) -> (r0) : r0 - d0 >= 0, d0 - r0 + 1 >= 0 ##aff_rel9 = (d0) -> (r0) : r0 - d0 >= 0, d0 - r0 + 1 >= 0
@ -673,7 +673,7 @@ TODO: figure out the right syntax.
Example: Example:
``` {.mlir} ```mlir {.mlir}
##rel9 ( ) [s0] -> (r0, r1) : 0 <= r0 <= 1023, 0 <= r1 <= s0 - 1 ##rel9 ( ) [s0] -> (r0, r1) : 0 <= r0 <= 1023, 0 <= r1 <= s0 - 1
extfunc @cblas_reduce_ffi(memref<1024 x ? x f32, #layout_map0, hbm> %M) -> f32 [ extfunc @cblas_reduce_ffi(memref<1024 x ? x f32, #layout_map0, hbm> %M) -> f32 [
@ -744,7 +744,7 @@ a list of statements that may also include a yield statement.
Example: Example:
``` {.mlir} ```mlir {.mlir}
// Return sum of elements in 1-dimensional mref A // Return sum of elements in 1-dimensional mref A
mlfunc int32 @sum(%A : memref<?xi32>, %N : i32) -> (i32) { mlfunc int32 @sum(%A : memref<?xi32>, %N : i32) -> (i32) {
%init = 0 %init = 0
@ -774,7 +774,7 @@ situation.
Example: Example:
``` {.mlir} ```mlir {.mlir}
// Compute sum of half of the array // Compute sum of half of the array
mlfunc int32 @sum_half(%A, %N) { mlfunc int32 @sum_half(%A, %N) {
%s0 = 0 %s0 = 0