Commit Graph

9 Commits

Author SHA1 Message Date
Jakob Stoklund Olesen
72847f3057 Reapply r155136 after fixing PR12599.
Original commit message:

Defer some shl transforms to DAGCombine.

The shl instruction is used to represent multiplication by a constant
power of two as well as bitwise left shifts. Some InstCombine
transformations would turn an shl instruction into a bit mask operation,
making it difficult for later analysis passes to recognize the
constsnt multiplication.

Disable those shl transformations, deferring them to DAGCombine time.
An 'shl X, C' instruction is now treated mostly the same was as 'mul X, C'.

These transformations are deferred:

  (X >>? C) << C   --> X & (-1 << C)  (When X >> C has multiple uses)
  (X >>? C1) << C2 --> X << (C2-C1) & (-1 << C2)   (When C2 > C1)
  (X >>? C1) << C2 --> X >>? (C1-C2) & (-1 << C2)  (When C1 > C2)

The corresponding exact transformations are preserved, just like
div-exact + mul:

  (X >>?,exact C) << C   --> X
  (X >>?,exact C1) << C2 --> X << (C2-C1)
  (X >>?,exact C1) << C2 --> X >>?,exact (C1-C2)

The disabled transformations could also prevent the instruction selector
from recognizing rotate patterns in hash functions and cryptographic
primitives. I have a test case for that, but it is too fragile.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@155362 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-23 17:39:52 +00:00
Jakob Stoklund Olesen
eece9dc81c Revert r155136 "Defer some shl transforms to DAGCombine."
While the patch was perfect and defect free, it exposed a really nasty
bug in X86 SelectionDAG that caused an llc crash when compiling lencod.

I'll put the patch back in after fixing the SelectionDAG problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@155181 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-20 00:38:45 +00:00
Jakob Stoklund Olesen
0d5fcae6cd Defer some shl transforms to DAGCombine.
The shl instruction is used to represent multiplication by a constant
power of two as well as bitwise left shifts. Some InstCombine
transformations would turn an shl instruction into a bit mask operation,
making it difficult for later analysis passes to recognize the
constsnt multiplication.

Disable those shl transformations, deferring them to DAGCombine time.
An 'shl X, C' instruction is now treated mostly the same was as 'mul X, C'.

These transformations are deferred:

  (X >>? C) << C   --> X & (-1 << C)  (When X >> C has multiple uses)
  (X >>? C1) << C2 --> X << (C2-C1) & (-1 << C2)   (When C2 > C1)
  (X >>? C1) << C2 --> X >>? (C1-C2) & (-1 << C2)  (When C1 > C2)

The corresponding exact transformations are preserved, just like
div-exact + mul:

  (X >>?,exact C) << C   --> X
  (X >>?,exact C1) << C2 --> X << (C2-C1)
  (X >>?,exact C1) << C2 --> X >>?,exact (C1-C2)

The disabled transformations could also prevent the instruction selector
from recognizing rotate patterns in hash functions and cryptographic
primitives. I have a test case for that, but it is too fragile.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@155136 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-19 16:46:26 +00:00
Jakob Stoklund Olesen
0d77b9c29c Extract the broken part of XFAILed test into its own file.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@155081 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-19 00:20:38 +00:00
Eli Bendersky
0f0c411079 Replace all instances of dg.exp file with lit.local.cfg, since all tests are run with LIT now and now Dejagnu. dg.exp is no longer needed.
Patch reviewed by Daniel Dunbar. It will be followed by additional cleanup patches.




git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@150664 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-16 06:28:33 +00:00
Eric Christopher
a510a1c931 Add file.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138083 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-19 21:21:20 +00:00
Nick Lewycky
9071ac2797 This directory was missing the dg.exp to cause the tests to run. Some time since
it was added, the test has regressed, so XFAIL it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@132686 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-06 20:23:00 +00:00
Chris Lattner
0a9e3d613b final step needed to resolve PR6627, which allows us to flatten the code down to
a nice and tidy:
  %x1 = load i32* %0, align 4
  %1 = icmp eq i32 %x1, 1179403647
  br i1 %1, label %if.then, label %if.end

instead of doing lots of loads and branches.  May the FreeBSD bootloader
long fit in its allocated space.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@130416 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-28 18:15:47 +00:00
Chris Lattner
734d7c2a7e make a couple of changes to the standard pass pipeline:
1. Only run the early (in the module pass pipe) instcombine/simplifycfg
   if the "unit at a time" passes they are cleaning up after runs.

2. Move the "clean up after the unroller" pass to the very end of the
   function-level pass pipeline.  Loop unroll uses instsimplify now,
   so it doesn't create a ton of trash.  Moving instcombine later allows
   it to clean up after opportunities are exposed by GVN, DSE, etc.

3. Introduce some phase ordering tests for things that are specifically
   intended to be simplified by the full optimizer as a whole.

This resolves PR2338, and is progress towards PR6627, which will be 
generating code that looks similar to test2.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@130241 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-26 20:45:33 +00:00