Merge lp:~ramana/gcc-linaro/47-nobble-promote-mode into lp:gcc-linaro/4.7

Proposed by Ramana Radhakrishnan
Status: Rejected
Rejected by: Michael Hope
Proposed branch: lp:~ramana/gcc-linaro/47-nobble-promote-mode
Merge into: lp:gcc-linaro/4.7
To merge this branch: bzr merge lp:~ramana/gcc-linaro/47-nobble-promote-mode
Reviewer Review Type Date Requested Status
Linaro Toolchain Developers Pending
Review via email: mp+107593@code.launchpad.net
To post a comment you must log in.
Revision history for this message
Ramana Radhakrishnan (ramana) wrote :

This is a benchmark run for nobbling promote_mode. This changes the code generated for this particular testcase by removing the extra uxtb in brackets below.

void f(unsigned char * p,unsigned short *s, int c)
{
 unsigned short or = 0;
  while (c)
    {
      or = or | s[c];
      c --;
    }
  *p = (unsigned char)or;
}

f:
 @ args = 0, pretend = 0, frame = 0
 @ frame_needed = 0, uses_anonymous_args = 0
 @ link register save eliminated.
 cmp r2, #0
 str r4, [sp, #-4]!
 beq .L2
 rsb r4, r2, r2, asl #31
 mov r3, #0
 add r1, r1, r2, asl #1
 mov r2, r3 @ movhi
 mov r4, r4, asl #1
.L4:
 ldrh ip, [r1, r3]
 sub r3, r3, #2
 cmp r3, r4
 orr r2, ip, r2
 bne .L4
        (uxtb r2, r2)
.L2:
 strb r2, [r0, #0]
 ldmfd sp!, {r4}
 bx lr

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

cbuild has taken a snapshot of this branch at r114992 and queued it for build.

The diff against the ancestor r114991 is available at:
 http://builds.linaro.org/toolchain/snapshots/gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode.diff

and will be built on the following builders:
 a9-builder a9hf-builder armv5-builder i686 i686-precise x86_64 x86_64-precise

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

cbuild-snapshot: gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode
cbuild-ancestor: lp:gcc-linaro+bzr114991
cbuild-state: check

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

cbuild successfully built this on armv7l-precise-cbuild311-ursa3-cortexa9hfr1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode/logs/armv7l-precise-cbuild311-ursa3-cortexa9hfr1

-PASS: gcc.target/arm/unsigned-extend-2.c scan-assembler ands
+FAIL: gcc.target/arm/unsigned-extend-2.c scan-assembler ands

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode/logs/armv7l-precise-cbuild311-ursa3-cortexa9hfr1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode/logs/armv7l-precise-cbuild311-ursa3-cortexa9hfr1/gcc-testsuite.txt

cbuild-checked: armv7l-precise-cbuild311-ursa3-cortexa9hfr1

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

cbuild successfully built this on armv7l-natty-cbuild311-tcpanda05-cortexa9r1.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode/logs/armv7l-natty-cbuild311-tcpanda05-cortexa9r1

-FAIL: g++.dg/cdce3.C -std=gnu++11 execution test
+PASS: g++.dg/cdce3.C -std=gnu++11 execution test
-FAIL: g++.dg/cdce3.C -std=gnu++98 execution test
+PASS: g++.dg/cdce3.C -std=gnu++98 execution test
-PASS: gcc.target/arm/unsigned-extend-2.c scan-assembler ands
+FAIL: gcc.target/arm/unsigned-extend-2.c scan-assembler ands
-PASS: 27_io/basic_filebuf/close/char/9964.cc execution test
+FAIL: 27_io/basic_filebuf/close/char/9964.cc execution test

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode/logs/armv7l-natty-cbuild311-tcpanda05-cortexa9r1/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode/logs/armv7l-natty-cbuild311-tcpanda05-cortexa9r1/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild311-tcpanda05-cortexa9r1

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

cbuild successfully built this on armv7l-natty-cbuild311-tcpanda02-armv5r2.

The build results are available at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode/logs/armv7l-natty-cbuild311-tcpanda02-armv5r2

-PASS: gcc.target/arm/pr40900.c scan-assembler-not asr
+FAIL: gcc.target/arm/pr40900.c scan-assembler-not asr

The full diff is at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode/logs/armv7l-natty-cbuild311-tcpanda02-armv5r2/testsuite-diff.txt

The full testsuite results are at:
 http://ex.seabright.co.nz/build/gcc-linaro-4.7+bzr114992~ramana~47-nobble-promote-mode/logs/armv7l-natty-cbuild311-tcpanda02-armv5r2/gcc-testsuite.txt

cbuild-checked: armv7l-natty-cbuild311-tcpanda02-armv5r2

Revision history for this message
Linaro Toolchain Builder (cbuild) wrote :
Revision history for this message
Linaro Toolchain Builder (cbuild) wrote :
Revision history for this message
Michael Hope (michaelh1) wrote :

Benchmarks are running. Results will be at:

 http://ex.seabright.co.nz/benchmarks/gcc-linaro-4.7%2bbzr114992~ramana~47-nobble-promote-mode/logs/armv7l-natty-cbuild321-tcpanda06-cortexa9r1/

 http://ex.seabright.co.nz/benchmarks/gcc-linaro-4.7%2bbzr114991/logs/armv7l-natty-cbuild321-tcpanda04-cortexa9r1/

coremark regresses slightly, but that will probably clear up if you fix +FAIL: gcc.target/arm/unsigned-extend-2.c scan-assembler ands.

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

This one didn't do as well as expected as it ended up causing more zero extends than necessary if my memory serves me right. This could be a course for a low priority future blueprint.

Subscribers

People subscribed via source and target branches