Chris Lattner 61b8c9bf94 The pattern isel is aggressively codegen'ing all of the loads in these ...

functions together at the start of the basic block, causing massive spillage.
The old isel codegened the loads wherever they happened to land, so it
generated good code for the first case, but bad code for the second.

We really want the pattern isel to generate (the same) good code for both.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19448 91177308-0d34-0410-b5e6-96231b3b80d8

2005-01-11 03:05:03 +00:00

..

Analysis

Make this testcase a bit more challanging

2004-12-17 17:14:00 +00:00

Archive

…

Assembler

Update this test to pass.

2005-01-05 15:23:06 +00:00

BugPoint

…

Bytecode

…

C++Frontend

Add a testcase to check that initializer that CAN be evaluated statically ARE.

2005-01-03 22:27:58 +00:00

CFrontend

Test fails on all platforms, not just linux

2005-01-08 23:44:03 +00:00

CodeGen

The pattern isel is aggressively codegen'ing all of the loads in these

2005-01-11 03:05:03 +00:00

Debugger

This test still fails on Darwin and Sparc/Solaris.

2004-12-06 22:05:02 +00:00

ExecutionEngine

Make this test actually capable of passing.

2004-12-05 07:15:21 +00:00

Feature

Update testcase

2004-12-10 05:39:57 +00:00

lib

…

Linker

Check in an old test that never made it into CVS

2005-01-02 02:46:29 +00:00

Other

…

Scripts

Remove last remnants of Python/QMTest support

2004-12-08 23:07:27 +00:00

TableGen

…

Transforms

New testcases that we should combine.

2005-01-01 16:14:46 +00:00

Verifier

…

Failure.sh

…

Makefile

Make sure the path is properly set in the llvmgcc and llvmgxx variables.

2005-01-03 04:05:32 +00:00

Makefile.tests

Make sure LLVMGCCWITHPATH is used instead of just LLVMGCC so that the

2004-12-24 03:44:24 +00:00

TestRunner.sh

Fix this script to work with hyphenated llvm-gcc and friend.

2005-01-03 22:26:38 +00:00