[DOCS] Added example for G_EXTRACT and G_INSERT

[xgupta]

Thanks, @sushmaunnibhavi for the patch!

I think the examples you add are for https://llvm.org/docs/GlobalISel/GenericOpcode.html#vector-specific-operations.

For scalar one you should look at /llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-extract.mir and /llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-insert.mir.

[xgupta]

@sushmaunnibhavi, I can't apply your updated patch. I think you should first rebase your branch with LLVM upstream main.
There is a recent tutorial (https://reviews.llvm.org/D100714) on the contribution that also talks about arc uses.

[xgupta]

Or if you are updating diff from Phabticator GUI, make sure you follow the steps mentions here -https://llvm.org/docs/Phabricator.html#requesting-a-review-via-the-web-interface.
But I think using arc (command line) is a better way.

[sushmaunnibhavi]

In D101227#2714434, @xgupta wrote:

Or if you are updating diff from Phabticator GUI, make sure you follow the steps mentions here -https://llvm.org/docs/Phabricator.html#requesting-a-review-via-the-web-interface.
But I think using arc (command line) is a better way.

Yes I did it now, can you please check?

[xgupta]

LGTM!

[xgupta]

Whenever you make changes to docs, make sure you run make docs-llvm-html before updating the revision.

llvm/docs/GlobalISel/GenericOpcode.rst

190

A line break is needed between these two lines

199

same here and %5 should be shifted two keystrokes right for indentation.

[sushmaunnibhavi]

In D101227#2714474, @xgupta wrote:

Whenever you make changes to docs, make sure you run make docs-llvm-html before updating the revision.

Okay I will do it.

[gargaroff]

llvm/docs/GlobalISel/GenericOpcode.rst

191

I think you should add the type for %2 to make it clear that this changes the type. I.e. %2:_(s64)

200

And same here.

[xgupta]

@gargaroff Is the patch looks good?

[gargaroff]

With the nit fixed, this should be good to go. Thanks!

llvm/docs/GlobalISel/GenericOpcode.rst

200

Add the type for %0 as well to make it clear that this is the same type as the result.

[sushmaunnibhavi]

Added type for %0 in G_INSERT operation

[xgupta]

Congratulations

llvm/docs/GlobalISel/GenericOpcode.rst

 
 G_EXTRACT
 ^^^^^^^^^
 
 Extract a register of the specified size, starting from the block given by
 index. This will almost certainly be mapped to sub-register COPYs after
 register banks have been selected.
 
+.. code-block:: none
+
+  %3:_(s32) = G_EXTRACT %2:_(s64), 32
+
 G_INSERT
 ^^^^^^^^
 
 Insert a smaller register into a larger one at the specified bit-index.
 
+.. code-block:: none
+
+  %2:_(s64) = G_INSERT %0:(_s64), %1:_(s32), 0
+
 G_MERGE_VALUES
 ^^^^^^^^^^^^^^
 
 Concatenate multiple registers of the same size into a wider register.
 The input operands are always ordered from lowest bits to highest:
 
 .. code-block:: none