1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-21 03:53:04 +02:00
Commit Graph

8 Commits

Author SHA1 Message Date
Jakob Stoklund Olesen
434f6a8d64 LowerSubregsInstructionPass::LowerExtract should not extend the live range of registers.
When LowerExtract eliminates an EXTRACT_SUBREG with a kill flag, it moves the
kill flag to the place where the sub-register is killed. This can accidentally
overlap with the use of a sibling sub-register, and we have trouble.

In the test case we have this code:

Live Ins: %R0 %R1 %R2
	%R2L<def> = EXTRACT_SUBREG %R2<kill>, 1
	%R2H<def> = LOAD16fi <fi#-1>, 0, Mem:LD(2,4) [FixedStack-1 + 0]
	%R1L<def> = EXTRACT_SUBREG %R1<kill>, 1
	%R0L<def> = EXTRACT_SUBREG %R0<kill>, 1
	%R0H<def> = ADD16 %R2H<kill>, %R2L<kill>, %AZ<imp-def>, %AN<imp-def>, %AC0<imp-def>, %V<imp-def>, %VS<imp-def>

subreg: CONVERTING: %R2L<def> = EXTRACT_SUBREG %R2<kill>, 1
subreg: eliminated!
subreg: killed here: %R0H<def> = ADD16 %R2H, %R2L, %R2<imp-use,kill>, %AZ<imp-def>, %AN<imp-def>, %AC0<imp-def>, %V<imp-def>, %VS<imp-def>

The kill flag on %R2 is moved to the last instruction, and the live range overlaps with the definition of %R2H:

*** Bad machine code: Redefining a live physical register ***
- function:    f
- basic block:  0x18358c0 (#0)
- instruction: %R2H<def> = LOAD16fi <fi#-1>, 0, Mem:LD(2,4) [FixedStack-1 + 0]
Register R2H was defined but already live.

The fix is to replace EXTRACT_SUBREG with IMPLICIT_DEF instead of eliminating
it completely:

subreg: CONVERTING: %R2L<def> = EXTRACT_SUBREG %R2<kill>, 1
subreg: replace by: %R2L<def> = IMPLICIT_DEF %R2<kill>

Note that these IMPLICIT_DEF instructions survive to the asm output. It is
necessary to fix the stack-color-with-reg test case because of that.

llvm-svn: 78093
2009-08-04 20:01:11 +00:00
Jakob Stoklund Olesen
0f9cf8a32b Remove unneeded intrinsics from Blackfin backend.
__builtin_bfin_ones does the same as ctpop, so it can be implemented in the front-end.

__builtin_bfin_loadbytes loads from an unaligned pointer with the disalignexcpt instruction. It does the same as loading from a pointer with the low bits masked. It is better if the front-end creates a masked load. We can always instruction select the masked to disalignexcpt+load.

We keep csync/ssync/idle. These intrinsics represent instructions that need workarounds for some silicon revisions. We may even want to convert inline assembler to intrinsics to enable the workarounds.

llvm-svn: 77917
2009-08-02 21:49:05 +00:00
Jakob Stoklund Olesen
9e71890531 Fix issue in regscavenger when scavenging a callee-saved register that has not been spilled.
llvm-svn: 77912
2009-08-02 20:29:41 +00:00
Jakob Stoklund Olesen
fe8f36d66b Never add a kill flag to a constrained physical register in a two-addr instruction.
llvm-svn: 77906
2009-08-02 19:13:03 +00:00
Jakob Stoklund Olesen
a726715be6 Scavenger asserts.
Allow imp-def and imp-use of anything in the scavenger asserts, just like the machine code verifier.
Allow redefinition of a sub-register of a live register.

llvm-svn: 77904
2009-08-02 18:28:41 +00:00
Jakob Stoklund Olesen
23ad8d1848 Add some basic blackfin intrinsics.
llvm-svn: 77903
2009-08-02 18:28:11 +00:00
Jakob Stoklund Olesen
5c6e03f669 Inline assembly support for Blackfin.
We use the same constraints as GCC, including those that are slightly insane for inline assembler.

llvm-svn: 77899
2009-08-02 17:39:17 +00:00
Jakob Stoklund Olesen
e7eb74ef39 Analog Devices Blackfin back-end.
Generate code for the Blackfin family of DSPs from Analog Devices:

  http://www.analog.com/en/embedded-processing-dsp/blackfin/processors/index.html
  
We aim to be compatible with the exsisting GNU toolchain found at:

  http://blackfin.uclinux.org/gf/project/toolchain
  
The back-end is experimental.

llvm-svn: 77897
2009-08-02 17:32:10 +00:00