[llvm-dev] Work on DAG Isel for TableGen and compiler

[Krzysztof Parzyszek via llvm-dev]

I can answer that.

The way HwModes are handled is that they are translated into macher’s predicates.  Same selection pattern may infer different value types for different modes, and all these variants are then predicated on the condition defining the mode which was applied to produce the variant.  There is not much there to common out, because the created variants are really different patterns at this moment.

With the hw modes you can write one pattern in the .td file and get it automatically expanded into variants instead of having to write the variants yourself, so in that sense you still get the large matcher table one way or the other.  Now, for the DefaultMode that your patch removes---yes, that’s both a shortcoming of the implementation (that it’s not specified in some other way), and that of the Hexagon .td files (that it duplicated another mode), which inspired the RISCV’s usage.


--
Krzysztof Parzyszek  kparzysz at quicinc.com<mailto:kparzysz at quicinc.com>   AI tools development

From: llvm-dev <llvm-dev-bounces at lists.llvm.org> On Behalf Of Craig Topper via llvm-dev
Sent: Thursday, November 19, 2020 12:31 AM
To: Paul C. Anagnostopoulos <paul at windfall.com>
Cc: llvm-dev <llvm-dev at lists.llvm.org>
Subject: [EXT] Re: [llvm-dev] Work on DAG Isel for TableGen and compiler

I would like to know why this patch https://reviews.llvm.org/D90973 had such a drastic size effect on the global isel tablegened matcher table for riscv.  It only changed the DAG ISel table by about 15K. But the global isel table shrinks by over 200K.

~Craig


On Wed, Nov 18, 2020 at 1:37 PM Paul C. Anagnostopoulos via llvm-dev <llvm-dev at lists.llvm.org<mailto:llvm-dev at lists.llvm.org>> wrote:
Given that I'm only somewhat up-to-speed on the DAG ISel scheme and not much at all on the Global ISel scheme, I'm tempted to work on the former and then the latter. So I'll look at the CodeGenDAGPatterns messages first. Then I will take a look at Global ISel.

Matt: Can you suggest one or two things about Global ISel that could use some work? I won't get to it quickly, but it will give me something to focus on.


At 11/18/2020 04:17 PM, Thomas Lively wrote:
>Yes, the CodeGenDAGPatterns is exactly right. Try applying the patch below and rebuilding and you'll see what I mean about the error messages ;) That being said, I'm sympathetic to Matt's point about shifting effort to GlobalISel. Maybe it has similar problems you could work on? A nicer development experience would certainly be a good carrot to get me excited to switch over sooner.

_______________________________________________
LLVM Developers mailing list
llvm-dev at lists.llvm.org<mailto:llvm-dev at lists.llvm.org>
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20201119/6aa601d0/attachment.html>