Merge lp:~ams-codesourcery/gcc-linaro/thumb-ifcvt2 into lp:gcc-linaro/4.7

Proposed by Andrew Stubbs
Status: Merged
Approved by: Andrew Stubbs
Approved revision: not available
Merge reported by: Andrew Stubbs
Merged at revision: not available
Proposed branch: lp:~ams-codesourcery/gcc-linaro/thumb-ifcvt2
Merge into: lp:gcc-linaro/4.7
To merge this branch: bzr merge lp:~ams-codesourcery/gcc-linaro/thumb-ifcvt2
Reviewer Review Type Date Requested Status
Andrew Stubbs (community) Approve
Michael Hope Pending
Ulrich Weigand Pending
Review via email: mp+94530@code.launchpad.net

This proposal supersedes a proposal from 2012-02-20.

Description of the change

Remove peephole optimizations that impede if-conversion.

UPDATE 20120217: Fixed regressions.

Posted upstream: http://gcc.gnu.org/ml/gcc-patches/2012-02/msg00734.html

UPDATE 20120220: Fixed ICE.

UPDATE 20120224: Fixed another ICE, and added some missed cases.

To post a comment you must log in.
Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal

cbuild has taken a snapshot of this branch at r114966 and queued it for build.

The diff against the ancestor r114965 is available at:
 http://builds.linaro.org/toolchain/snapshots/gcc-linaro-4.7+bzr114966~ams-codesourcery~thumb-ifcvt2.diff

and will be built on the following builders:
 a9-builder armv5-builder i686 x86_64

You can track the build queue at:
 http://ex.seabright.co.nz/helpers/scheduler

cbuild-snapshot: gcc-linaro-4.7+bzr114966~ams-codesourcery~thumb-ifcvt2
cbuild-ancestor: lp:gcc-linaro/4.7+bzr114965
cbuild-state: check

Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal
Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal
Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal
Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal

cbuild successfully built this on armv7l-natty-cbuild245-tcpanda06-cortexa9r1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114966~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild245-tcpanda06-cortexa9r1

-PASS: gcc.target/arm/ivopts-2.c object-size text <= 26
+FAIL: gcc.target/arm/ivopts-2.c object-size text <= 26
-PASS: gcc.target/arm/ivopts-3.c object-size text <= 30
+FAIL: gcc.target/arm/ivopts-3.c object-size text <= 30
-PASS: gcc.target/arm/ivopts-4.c object-size text <= 36
+FAIL: gcc.target/arm/ivopts-4.c object-size text <= 36
-PASS: gcc.target/arm/ivopts-5.c object-size text <= 30
+FAIL: gcc.target/arm/ivopts-5.c object-size text <= 30
-PASS: gcc.target/arm/ivopts.c object-size text <= 20
+FAIL: gcc.target/arm/ivopts.c object-size text <= 20
-PASS: gcc.target/arm/pr43597.c object-size text <= 30
+FAIL: gcc.target/arm/pr43597.c object-size text <= 30
-PASS: gcc.target/arm/pr43920-2.c object-size text <= 54
+FAIL: gcc.target/arm/pr43920-2.c object-size text <= 54
+PASS: gcc.target/arm/thumb-ifcvt.c (test for excess errors)
+PASS: gcc.target/arm/thumb-ifcvt.c scan-assembler asrne
+PASS: gcc.target/arm/thumb-ifcvt.c scan-assembler lslne
-PASS: gcc.target/arm/thumb2-mul-space.c scan-assembler muls
+FAIL: gcc.target/arm/thumb2-mul-space.c scan-assembler muls

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114966~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild245-tcpanda06-cortexa9r1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114966~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild245-tcpanda06-cortexa9r1/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild245-tcpanda06-cortexa9r1

Revision history for this message
Ulrich Weigand (uweigand) wrote : Posted in a previous version of this proposal

Those regressions appear to be real at first glance --- can you have a look?

review: Needs Fixing
Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal

cbuild has taken a snapshot of this branch at r114968 and queued it for build.

The diff against the ancestor r114965 is available at:
 http://builds.linaro.org/toolchain/snapshots/gcc-linaro-4.7+bzr114968~ams-codesourcery~thumb-ifcvt2.diff

and will be built on the following builders:
 a9-builder armv5-builder i686 x86_64

You can track the build queue at:
 http://ex.seabright.co.nz/helpers/scheduler

cbuild-snapshot: gcc-linaro-4.7+bzr114968~ams-codesourcery~thumb-ifcvt2
cbuild-ancestor: lp:gcc-linaro/4.7+bzr114965
cbuild-state: check

Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal

cbuild successfully built this on armv7l-natty-cbuild250-tcpanda05-armv5r2.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114968~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild250-tcpanda05-armv5r2

+UNSUPPORTED: gcc.target/arm/thumb-16bit-ops.c
+UNSUPPORTED: gcc.target/arm/thumb-ifcvt.c

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114968~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild250-tcpanda05-armv5r2/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114968~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild250-tcpanda05-armv5r2/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild250-tcpanda05-armv5r2

Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal

cbuild had trouble building this on armv7l-natty-cbuild250-tcpanda06-cortexa9r1.

See the *failed.txt logs under the build results at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114968~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild250-tcpanda06-cortexa9r1

The test suite results were not checked.

cbuild-checked: armv7l-natty-cbuild250-tcpanda06-cortexa9r1

review: Needs Fixing
Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal
Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal
Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal

cbuild has taken a snapshot of this branch at r114969 and queued it for build.

The diff against the ancestor r114965 is available at:
 http://builds.linaro.org/toolchain/snapshots/gcc-linaro-4.7+bzr114969~ams-codesourcery~thumb-ifcvt2.diff

and will be built on the following builders:
 a9-builder armv5-builder i686 x86_64

You can track the build queue at:
 http://ex.seabright.co.nz/helpers/scheduler

cbuild-snapshot: gcc-linaro-4.7+bzr114969~ams-codesourcery~thumb-ifcvt2
cbuild-ancestor: lp:gcc-linaro/4.7+bzr114965
cbuild-state: check

Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal
Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal
Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal

cbuild successfully built this on armv7l-natty-cbuild250-tcpanda05-armv5r2.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114969~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild250-tcpanda05-armv5r2

+UNSUPPORTED: gcc.target/arm/thumb-16bit-ops.c
+UNSUPPORTED: gcc.target/arm/thumb-ifcvt.c

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114969~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild250-tcpanda05-armv5r2/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114969~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild250-tcpanda05-armv5r2/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild250-tcpanda05-armv5r2

Revision history for this message
Michael Hope (michaelh1) wrote : Posted in a previous version of this proposal
Download full text (3.3 KiB)

cbuild successfully built this on armv7l-natty-cbuild250-tcpanda06-cortexa9r1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114969~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild250-tcpanda06-cortexa9r1

-PASS: gcc.c-torture/execute/20040629-1.c compilation, -Os
+FAIL: gcc.c-torture/execute/20040629-1.c compilation, -Os (internal compiler error)
-PASS: gcc.c-torture/execute/20040629-1.c execution, -Os
+UNRESOLVED: gcc.c-torture/execute/20040629-1.c execution, -Os
-PASS: gcc.c-torture/execute/20040705-1.c compilation, -Os
+FAIL: gcc.c-torture/execute/20040705-1.c compilation, -Os (internal compiler error)
-PASS: gcc.c-torture/execute/20040705-1.c execution, -Os
+UNRESOLVED: gcc.c-torture/execute/20040705-1.c execution, -Os
-PASS: gcc.c-torture/execute/20040705-2.c compilation, -Os
+FAIL: gcc.c-torture/execute/20040705-2.c compilation, -Os (internal compiler error)
-PASS: gcc.c-torture/execute/20040705-2.c execution, -Os
+UNRESOLVED: gcc.c-torture/execute/20040705-2.c execution, -Os
-FAIL: gcc.target/arm/combine-movs.c scan-assembler movs\tr[0-9]
+PASS: gcc.target/arm/combine-movs.c scan-assembler movs\tr[0-9]
+PASS: gcc.target/arm/thumb-16bit-ops.c (test for excess errors)
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler add\tr0, r0, #256
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler add\tr0, r1, #8
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler adds\tr0, r0, #255
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler adds\tr0, r0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler adds\tr0, r1, #7
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler asr\tr0, r1, r2
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler asrs\tr0, r0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler asrs\tr0, r1, #15
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsl\tr0, r1, r2
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsls\tr0, r0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsls\tr0, r1, #15
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsr\tr0, r1, r2
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsrs\tr0, r0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsrs\tr0, r1, #15
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler mov\tr0, #256
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler mov\tr0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler movs\tr0, #255
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler muls\tr0, r1, r0
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler mvns\tr0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler negs\tr0, r1
+PASS: gcc.target/arm/thumb-ifcvt.c (test for excess errors)
+PASS: gcc.target/arm/thumb-ifcvt.c scan-assembler asrne
+PASS: gcc.target/arm/thumb-ifcvt.c scan-assembler lslne

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114969~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild250-tcpanda06-cortexa9r1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114969~ams-codesourcery~thumb-if...

Read more...

Revision history for this message
Andrew Stubbs (ams-codesourcery) wrote : Posted in a previous version of this proposal

Clearly those ICEs are a problem ..... :(

review: Needs Fixing
Revision history for this message
Michael Hope (michaelh1) wrote :

cbuild has taken a snapshot of this branch at r114970 and queued it for build.

The diff against the ancestor r114965 is available at:
 http://builds.linaro.org/toolchain/snapshots/gcc-linaro-4.7+bzr114970~ams-codesourcery~thumb-ifcvt2.diff

and will be built on the following builders:
 a9-builder armv5-builder i686 x86_64

You can track the build queue at:
 http://ex.seabright.co.nz/helpers/scheduler

cbuild-snapshot: gcc-linaro-4.7+bzr114970~ams-codesourcery~thumb-ifcvt2
cbuild-ancestor: lp:gcc-linaro/4.7+bzr114965
cbuild-state: check

Revision history for this message
Michael Hope (michaelh1) wrote :

cbuild successfully built this on armv7l-natty-cbuild254-tcpanda04-armv5r2.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114970~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild254-tcpanda04-armv5r2

+UNSUPPORTED: gcc.target/arm/thumb-16bit-ops.c
+UNSUPPORTED: gcc.target/arm/thumb-ifcvt.c

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114970~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild254-tcpanda04-armv5r2/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114970~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild254-tcpanda04-armv5r2/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild254-tcpanda04-armv5r2

Revision history for this message
Michael Hope (michaelh1) wrote :

cbuild successfully built this on armv7l-natty-cbuild254-tcpanda05-cortexa9r1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114970~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild254-tcpanda05-cortexa9r1

-FAIL: gcc.target/arm/combine-movs.c scan-assembler movs\tr[0-9]
+PASS: gcc.target/arm/combine-movs.c scan-assembler movs\tr[0-9]
+PASS: gcc.target/arm/thumb-16bit-ops.c (test for excess errors)
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler add\tr0, r0, #256
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler add\tr0, r1, #8
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler adds\tr0, r0, #255
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler adds\tr0, r0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler adds\tr0, r1, #7
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler asr\tr0, r1, r2
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler asrs\tr0, r0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler asrs\tr0, r1, #15
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsl\tr0, r1, r2
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsls\tr0, r0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsls\tr0, r1, #15
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsr\tr0, r1, r2
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsrs\tr0, r0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler lsrs\tr0, r1, #15
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler mov\tr0, #256
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler mov\tr0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler movs\tr0, #255
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler muls\tr0, r1, r0
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler mvns\tr0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler negs\tr0, r1
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler sub\tr0, r0, #256
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler sub\tr0, r1, #8
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler subs\tr0, r0, #255
+PASS: gcc.target/arm/thumb-16bit-ops.c scan-assembler subs\tr0, r1, #7
+PASS: gcc.target/arm/thumb-ifcvt.c (test for excess errors)
+PASS: gcc.target/arm/thumb-ifcvt.c scan-assembler asrne
+PASS: gcc.target/arm/thumb-ifcvt.c scan-assembler lslne

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114970~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild254-tcpanda05-cortexa9r1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114970~ams-codesourcery~thumb-ifcvt2/logs/armv7l-natty-cbuild254-tcpanda05-cortexa9r1/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild254-tcpanda05-cortexa9r1

Revision history for this message
Michael Hope (michaelh1) wrote :
Revision history for this message
Michael Hope (michaelh1) wrote :
Revision history for this message
Ulrich Weigand (uweigand) wrote :

What's the upstream status of this? Has it been approved yet?

Revision history for this message
Andrew Stubbs (ams-codesourcery) wrote :

I've just resubmitted the version of this patch from this branch:
http://gcc.gnu.org/ml/gcc-patches/2012-03/msg00571.html

Awaiting upstream review.

Revision history for this message
Andrew Stubbs (ams-codesourcery) wrote :

Now approved by Richard E.

review: Approve

Subscribers

People subscribed via source and target branches