[Bug target/61055] [avr] wrong test instruction after increment with -O1

2014-05-09 Thread gjl at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=61055 --- Comment #2 from Georg-Johann Lay gjl at gcc dot gnu.org --- Author: gjl Date: Fri May 9 11:20:43 2014 New Revision: 210267 URL: http://gcc.gnu.org/viewcvs?rev=210267root=gccview=rev Log: gcc/config/avr PR target/61055 *

[Bug target/61055] [avr] wrong test instruction after increment with -O1

2014-05-09 Thread gjl at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=61055 --- Comment #3 from Georg-Johann Lay gjl at gcc dot gnu.org --- Author: gjl Date: Fri May 9 11:25:11 2014 New Revision: 210268 URL: http://gcc.gnu.org/viewcvs?rev=210268root=gccview=rev Log: gcc/config/avr Backport from 2014-05-09 trunk

[Bug target/61055] [avr] wrong test instruction after increment with -O1

2014-05-09 Thread gjl at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=61055 --- Comment #4 from Georg-Johann Lay gjl at gcc dot gnu.org --- Author: gjl Date: Fri May 9 11:29:58 2014 New Revision: 210269 URL: http://gcc.gnu.org/viewcvs?rev=210269root=gccview=rev Log: gcc/ Backport from 2014-05-09 trunk r210267 PR

[Bug target/61055] [avr] wrong test instruction after increment with -O1

2014-05-09 Thread gjl at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=61055 --- Comment #5 from Georg-Johann Lay gjl at gcc dot gnu.org --- Author: gjl Date: Fri May 9 11:34:46 2014 New Revision: 210270 URL: http://gcc.gnu.org/viewcvs?rev=210270root=gccview=rev Log: gcc/ Backport from 2014-05-09 trunk r210267 PR

[Bug target/61055] [avr] wrong test instruction after increment with -O1

2014-05-09 Thread gjl at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=61055 Georg-Johann Lay gjl at gcc dot gnu.org changed: What|Removed |Added Keywords||wrong-code

[Bug target/61055] [avr] wrong test instruction after increment with -O1

2014-05-05 Thread gjl at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=61055 --- Comment #1 from Georg-Johann Lay gjl at gcc dot gnu.org --- I see similar code with -fno-peephole2 What about the following that is also turned into a signed test and generates an ADIW / BRGE sequence. This is also wrong then when we start