Merge lp:~uweigand/gcc-linaro/mov-vld1.64-4.7 into lp:gcc-linaro/4.7

Proposed by Ulrich Weigand
Status: Rejected
Rejected by: Ulrich Weigand
Proposed branch: lp:~uweigand/gcc-linaro/mov-vld1.64-4.7
Merge into: lp:gcc-linaro/4.7
To merge this branch: bzr merge lp:~uweigand/gcc-linaro/mov-vld1.64-4.7
Reviewer Review Type Date Requested Status
Linaro Toolchain Developers Pending
Review via email: mp+103860@code.launchpad.net

This proposal supersedes a proposal from 2012-03-31.

Description of the change

Use vld1.64/vst1.64 instead of vldm/vstm for vector moves where possible.

Updated version: Fix invalid alignment hints.
Update 2012-04-27: Fix EEMBC ICE (support double-word modes in neon_vector_mem_operand).

Not yet upstream; merge request for test and benchmark purposes.

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 r114975 and queued it for build.

The diff against the ancestor r114974 is available at:
 http://builds.linaro.org/toolchain/snapshots/gcc-linaro-4.7+bzr114975~uweigand~mov-vld1.64-4.7.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+bzr114975~uweigand~mov-vld1.64-4.7
cbuild-ancestor: lp:gcc-linaro/4.7+bzr114974
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-cbuild280-tcpanda04-cortexa9r1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114975~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild280-tcpanda04-cortexa9r1

-PASS: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
-PASS: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
-PASS: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
+FAIL: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
-PASS: gcc.dg/pr48616.c (test for excess errors)
-PASS: gcc.dg/pr48616.c execution test
+FAIL: gcc.dg/pr48616.c (test for excess errors)
+UNRESOLVED: gcc.dg/pr48616.c compilation failed to produce executable
-PASS: gcc.dg/torture/vector-2.c -O0 execution test
+FAIL: gcc.dg/torture/vector-2.c -O0 execution test
-PASS: gcc.dg/vect/vect-shift-4.c (test for excess errors)
-PASS: gcc.dg/vect/vect-shift-4.c -flto (test for excess errors)
-PASS: gcc.dg/vect/vect-shift-4.c -flto execution test
+FAIL: gcc.dg/vect/vect-shift-4.c (test for excess errors)
+FAIL: gcc.dg/vect/vect-shift-4.c -flto (test for excess errors)
+UNRESOLVED: gcc.dg/vect/vect-shift-4.c -flto compilation failed to produce executable
-PASS: gcc.dg/vect/vect-shift-4.c execution test
+UNRESOLVED: gcc.dg/vect/vect-shift-4.c compilation failed to produce executable
-PASS: 27_io/basic_istream/seekg/char/exceptions_badbit_throw.cc execution test
+FAIL: 27_io/basic_istream/seekg/char/exceptions_badbit_throw.cc execution test
-PASS: 27_io/basic_ostream/seekp/char/exceptions_badbit_throw.cc execution test
+FAIL: 27_io/basic_ostream/seekp/char/exceptions_badbit_throw.cc execution test

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114975~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild280-tcpanda04-cortexa9r1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114975~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild280-tcpanda04-cortexa9r1/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild280-tcpanda04-cortexa9r1

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

cbuild successfully built this on armv7l-natty-cbuild280-tcpanda06-armv5r2.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114975~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild280-tcpanda06-armv5r2

-PASS: gcc.dg/vect/vect-shift-4.c (test for excess errors)
-PASS: gcc.dg/vect/vect-shift-4.c -flto (test for excess errors)
-PASS: gcc.dg/vect/vect-shift-4.c -flto execution test
+FAIL: gcc.dg/vect/vect-shift-4.c (test for excess errors)
+FAIL: gcc.dg/vect/vect-shift-4.c -flto (test for excess errors)
+UNRESOLVED: gcc.dg/vect/vect-shift-4.c -flto compilation failed to produce executable
-PASS: gcc.dg/vect/vect-shift-4.c execution test
+UNRESOLVED: gcc.dg/vect/vect-shift-4.c compilation failed to produce executable

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114975~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild280-tcpanda06-armv5r2/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114975~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild280-tcpanda06-armv5r2/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild280-tcpanda06-armv5r2

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 r114976 and queued it for build.

The diff against the ancestor r114974 is available at:
 http://builds.linaro.org/toolchain/snapshots/gcc-linaro-4.6+bzr114976~uweigand~mov-vld1.64-4.7.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-4.6+bzr114976~uweigand~mov-vld1.64-4.7
cbuild-ancestor: lp:gcc-linaro+bzr114974
cbuild-state: check

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

Re-spawning due to a bug in the taker. The devfocus was still 4.6 so the build completed and tested but has '4.6' everywhere across it.

cbuild-state: queue

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 r114976 and queued it for build.

The diff against the ancestor r114974 is available at:
 http://builds.linaro.org/toolchain/snapshots/gcc-linaro-4.7+bzr114976~uweigand~mov-vld1.64-4.7.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+bzr114976~uweigand~mov-vld1.64-4.7
cbuild-ancestor: lp:gcc-linaro+bzr114974
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

cbuild successfully built this on armv7l-natty-cbuild280-tcpanda04-cortexa9r1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114976~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild280-tcpanda04-cortexa9r1

-PASS: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
-PASS: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
-PASS: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
+FAIL: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
-PASS: gcc.dg/torture/vector-2.c -O0 execution test
+FAIL: gcc.dg/torture/vector-2.c -O0 execution test
-PASS: 27_io/basic_istream/seekg/char/exceptions_badbit_throw.cc execution test
+FAIL: 27_io/basic_istream/seekg/char/exceptions_badbit_throw.cc execution test
-PASS: 27_io/basic_ostream/seekp/char/exceptions_badbit_throw.cc execution test
+FAIL: 27_io/basic_ostream/seekp/char/exceptions_badbit_throw.cc execution test

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114976~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild280-tcpanda04-cortexa9r1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114976~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild280-tcpanda04-cortexa9r1/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild280-tcpanda04-cortexa9r1

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

cbuild successfully built this on armv7l-natty-cbuild283-tcpanda02-cortexa9r1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114976~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild283-tcpanda02-cortexa9r1

-PASS: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
-PASS: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
-PASS: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
+FAIL: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
-PASS: gcc.dg/torture/vector-2.c -O0 execution test
+FAIL: gcc.dg/torture/vector-2.c -O0 execution test
-PASS: 27_io/basic_istream/seekg/char/exceptions_badbit_throw.cc execution test
+FAIL: 27_io/basic_istream/seekg/char/exceptions_badbit_throw.cc execution test
-PASS: 27_io/basic_ostream/seekp/char/exceptions_badbit_throw.cc execution test
+FAIL: 27_io/basic_ostream/seekp/char/exceptions_badbit_throw.cc execution test

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114976~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild283-tcpanda02-cortexa9r1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114976~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild283-tcpanda02-cortexa9r1/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild283-tcpanda02-cortexa9r1

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

ICEs when building EEMBC:

idctrn01/bmark.c:862:1: error: insn does not satisfy its constraints:
(insn 15020 5306 5374 164 (set (reg:V2SI 109 d23)
        (mem/c:V2SI (plus:SI (reg/f:SI 3 r3 [4876])
                (const_int 1016 [0x3f8])) [4 MEM[(long int[8][8] *)&G_2 + 88B]+0 S8 A64])) idctrn01/bmark.c:635 764 {*neon_movv2si}
     (nil))
idctrn01/bmark.c:862:1: internal compiler error: in reload_cse_simplify_operands, at postreload.c:403

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

cbuild successfully built this on armv7l-natty-cbuild288-ursa1-cortexa9r1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114976~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild288-ursa1-cortexa9r1

-PASS: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
-PASS: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
-PASS: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
+FAIL: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
-PASS: gcc.dg/torture/vector-2.c -O0 execution test
+FAIL: gcc.dg/torture/vector-2.c -O0 execution test

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114976~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild288-ursa1-cortexa9r1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114976~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild288-ursa1-cortexa9r1/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild288-ursa1-cortexa9r1

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

cbuild has taken a snapshot of this branch at r114977 and queued it for build.

The diff against the ancestor r114974 is available at:
 http://builds.linaro.org/toolchain/snapshots/gcc-linaro-4.7+bzr114977~uweigand~mov-vld1.64-4.7.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+bzr114977~uweigand~mov-vld1.64-4.7
cbuild-ancestor: lp:gcc-linaro+bzr114974
cbuild-state: check

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

cbuild successfully built this on armv7l-natty-cbuild288-ursa3-cortexa9r1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114977~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild288-ursa3-cortexa9r1

-PASS: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
-PASS: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
-PASS: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
+FAIL: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
-PASS: gcc.dg/torture/vector-2.c -O0 execution test
+FAIL: gcc.dg/torture/vector-2.c -O0 execution test

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114977~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild288-ursa3-cortexa9r1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114977~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild288-ursa3-cortexa9r1/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild288-ursa3-cortexa9r1

Revision history for this message
Michael Hope (michaelh1) wrote :
Revision history for this message
Andrew Stubbs (ams-codesourcery) wrote :

Presumably the test failures are a Bad Thing? I only ask because they're were already there in the previous test results, so maybe you were expecting them?

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

EEMBC now builds and runs.

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

> Presumably the test failures are a Bad Thing? I only ask because they're were
> already there in the previous test results, so maybe you were expecting them?

Yes, those failures are known; they are due to incorrect alignment assumptions
for vector types (we discussed those on Tuesday). The hope is that with Richard's
patch to align vector types to 8 bytes, those failures will go away.

(I'm not planning on merging this patch before this issue is resolved.)

Revision history for this message
Linaro Toolchain Builder (cbuild) wrote :

cbuild successfully built this on armv7l-natty-cbuild292-tcpanda03-cortexa9r1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114977~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild292-tcpanda03-cortexa9r1

-PASS: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-1.c execution, -O0
-PASS: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
+FAIL: gcc.c-torture/execute/vector-subscript-2.c execution, -O0
-PASS: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
+FAIL: gcc.dg/compat/vector-1 c_compat_x_tst.o-c_compat_y_tst.o execute
-PASS: gcc.dg/torture/vector-2.c -O0 execution test
+FAIL: gcc.dg/torture/vector-2.c -O0 execution test
-PASS: 27_io/basic_istream/seekg/char/exceptions_badbit_throw.cc execution test
+FAIL: 27_io/basic_istream/seekg/char/exceptions_badbit_throw.cc execution test
-PASS: 27_io/basic_ostream/seekp/char/exceptions_badbit_throw.cc execution test
+FAIL: 27_io/basic_ostream/seekp/char/exceptions_badbit_throw.cc execution test

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114977~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild292-tcpanda03-cortexa9r1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114977~uweigand~mov-vld1.64-4.7/logs/armv7l-natty-cbuild292-tcpanda03-cortexa9r1/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild292-tcpanda03-cortexa9r1

Subscribers

People subscribed via source and target branches