[Bug rtl-optimization/23837] [4.0/4.1/4.2 regression] Wrong code with REG_NO_CONFLICT notes (caused by combine)
--- Comment #28 from amylaar at gcc dot gnu dot org 2005-12-16 14:37 --- (In reply to comment #25) > Smarter folks than me (iant ;-) suggest that "a multi-word rotate will > normally > need all the input bits when setting any of the output bits", so the entire > no-conflict thing doesn't make sense here. This is not really relevant, since the the insns that are using other words than the result word are computing their (intermediate) results into pseudos, and thus get emitted in front of (i.e. outside) the REG_NO_CONFLICT block. combine is really to blame for combining insns into a REG_NO_CONFLICT block in a way to invalidate the REG_NO_CONFLICT note without removing the note. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23837 --- You are receiving this mail because: --- You reported the bug, or are watching the reporter. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#333536: Upstream
Filed upstream as gnu pr 25043. Confirmed, fixed in 4.1. -- Stephen R. Marenka If life's not fun, you're not doing it right! <[EMAIL PROTECTED]> signature.asc Description: Digital signature
In order to compile gnat-4.1 under powerpc architecture
Hi everybody, Since two days I have found how to compile gnat-4.1 under ppc architecture, but I struggle with the debian package building procedure. I don't find which files to modify in order to compile a few files without optimisation option. In fact a-except.adb compilation crash the stage1/xgcc if -O1 or -O2 is used... a few others files do so. I have modified a few Makefile.in rules2 and so on but there is two much of it to know where to do all modification. Do I need to recompile gnat-4.1 by removing everywhere O2 and O1 they appear ? Any help welcome, thanks, xavier -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: In order to compile gnat-4.1 under powerpc architecture
Xavier Grave <[EMAIL PROTECTED]> writes: > Hi everybody, > > Since two days I have found how to compile gnat-4.1 under ppc > architecture, but I struggle with the debian package building procedure. > I don't find which files to modify in order to compile a few files > without optimisation option. In fact a-except.adb compilation crash the > stage1/xgcc if -O1 or -O2 is used... > a few others files do so. I have modified a few Makefile.in rules2 and > so on but there is two much of it to know where to do all modification. > > Do I need to recompile gnat-4.1 by removing everywhere O2 and O1 they > appear ? > > Any help welcome, thanks, xavier I'll look into this over the next few days, and I'll try to propose a patch. -- Ludovic Brenta. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[Bug rtl-optimization/21041] [4.0/4.1/4.2 Regression] ICE: output_operand: Cannot decompose address
--- Comment #11 from uweigand at gcc dot gnu dot org 2005-12-16 18:05 --- Patch posted. -- uweigand at gcc dot gnu dot org changed: What|Removed |Added URL||http://gcc.gnu.org/ml/gcc- ||patches/2005- ||12/msg01283.html Component|target |rtl-optimization Keywords||patch http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21041 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#343640: gcc-4.0: FTBFS on hurd-i386: conflicting patches getting applied
Package: gcc-4.0 Version: 4.0.2-5 Severity: important Hi, since libgcj has been split off, both libffi-without-libgcj and disable-libgcj are getting applied on hurd-i386, which leads to a rejected patch in the latter: DEB_VERSION='4.0.2-5'; export DEB_VERSION; \ debian/patches/disable-libgcj.dpatch -patch -d /build/buildd/gcc-4.0-4.0.2/src patching file configure.in Hunk #1 FAILED at 171. 1 out of 1 hunk FAILED -- saving rejects to file configure.in.rej patching file configure Hunk #1 FAILED at 937. 1 out of 1 hunk FAILED -- saving rejects to file configure.rej make: *** [stamps/02-patch-stamp-disable-libgcj] Error 1 The full build log can be found here: http://experimental.ftbfs.de/fetch.php?&pkg=gcc-4.0&ver=4.0.2-5&arch=hurd-i386&stamp=1134709009&file=log&as=raw Before the split, only libffi-without-libgcj got applied. Hurd-i386 is in java_no_systems for the time being, so maybe add another check for with_java to the disable-libgcj patch and only apply it when java would be built anyway? I am not sure how the two source packages are tied together, so I am not sending a patch at this point, but I can try to come up with one if you prefer. cheers, Michael -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#341882: Updated patch
Attached is an updated patch against the -5 package that recently came
out. It also leaves out the dead patch that accidentally made it's
way into the previous submission.
I know I answered questions somewhat piecemeal, but I think I answered
them all. Were there any questions I neglected to answer?
Stuart
Stuart R. Anderson [EMAIL PROTECTED]
Network & Software Engineering http://www.netsweng.com/
1024D/37A79149: 0791 D3B8 9A4C 2CDC A31F
BD03 0A62 E534 37A7 9149Index: debian/rules.patch
===
--- debian/rules.patch (.../vendor/gcc-4.0/current)(revision 150)
+++ debian/rules.patch (.../src/gcc-4.0) (revision 150)
@@ -109,6 +109,7 @@
endif
ifneq (,$(findstring /$(DEB_TARGET_ARCH)/,/mips/mipsel/))
debian_patches += libffi-mips libmudflap-entry-point libmudflap-mips
+ debian_patches += mips-biarch
endif
ifeq ($(DEB_TARGET_ARCH_OS),kfreebsd)
Index: debian/patches/mips-biarch.dpatch
===
--- debian/patches/mips-biarch.dpatch (.../vendor/gcc-4.0/current)
(revision 0)
+++ debian/patches/mips-biarch.dpatch (.../src/gcc-4.0) (revision 150)
@@ -0,0 +1,85 @@
+#! /bin/sh -e
+
+# DP: Patch author: Stuart Anderson <[EMAIL PROTECTED]>
+# DP: Upstream status: Not submitted
+# DP: Description: add full tri-arch support. Include linux64.h also fix up
+# DP: Description: the directory names so that o32 is the default and follow
+# DP: Description: the glibc convention for 32 & 64 bit names
+
+dir=
+if [ $# -eq 3 -a "$2" = '-d' ]; then
+pdir="-d $3"
+dir="$3/"
+elif [ $# -ne 1 ]; then
+echo >&2 "`basename $0`: script expects -patch|-unpatch as argument"
+exit 1
+fi
+case "$1" in
+-patch)
+patch $pdir -c -f --no-backup-if-mismatch -p1 < $0
+;;
+-unpatch)
+patch $pdir -c -f --no-backup-if-mismatch -R -p1 < $0
+;;
+*)
+echo >&2 "`basename $0`: script expects -patch|-unpatch as argument"
+exit 1
+esac
+exit 0
+
+*** src/gcc/config.gcc.bak Mon Oct 3 16:12:53 2005
+--- src/gcc/config.gcc Mon Oct 3 16:19:45 2005
+***
+*** 1424,1430
+ gas=yes
+ ;;
+ mips*-*-linux*) # Linux MIPS, either endian.
+! tm_file="dbxelf.h elfos.h svr4.h linux.h ${tm_file} mips/linux.h"
+ case ${target} in
+ mipsisa32*-*)
+ target_cpu_default="MASK_SOFT_FLOAT"
+--- 1424,1431
+ gas=yes
+ ;;
+ mips*-*-linux*) # Linux MIPS, either endian.
+! tm_file="dbxelf.h elfos.h svr4.h linux.h ${tm_file} mips/linux.h
mips/linux64.h"
+! tmake_file="${tmake_file} mips/t-linux64"
+ case ${target} in
+ mipsisa32*-*)
+ target_cpu_default="MASK_SOFT_FLOAT"
+
+*** src/gcc/config/mips/linux64.h.bak Mon Oct 3 11:28:01 2005
+--- src/gcc/config/mips/linux64.h Mon Oct 3 11:28:29 2005
+***
+*** 23,29
+ in order to make the other specs easier to write. */
+ #define DRIVER_SELF_SPECS \
+ "%{!EB:%{!EL:%(endian_spec)}}", \
+! "%{!mabi=*: -mabi=n32}"
+
+ #undef SUBTARGET_ASM_SPEC
+ #define SUBTARGET_ASM_SPEC "\
+--- 23,29
+ in order to make the other specs easier to write. */
+ #define DRIVER_SELF_SPECS \
+ "%{!EB:%{!EL:%(endian_spec)}}", \
+! "%{!mabi=*: -mabi=32}"
+
+ #undef SUBTARGET_ASM_SPEC
+ #define SUBTARGET_ASM_SPEC "\
+
+*** src/gcc/config/mips/t-linux64.orig Wed Oct 19 06:19:32 2005
+--- src/gcc/config/mips/t-linux64 Tue Oct 18 18:08:02 2005
+***
+*** 1,5
+ MULTILIB_OPTIONS = mabi=n32/mabi=32/mabi=64
+! MULTILIB_DIRNAMES = n32 32 64
+ MULTILIB_OSDIRNAMES = ../lib32 ../lib ../lib64
+
+ EXTRA_MULTILIB_PARTS=crtbegin.o crtend.o crtbeginS.o crtendS.o crtbeginT.o
+--- 1,5
+ MULTILIB_OPTIONS = mabi=n32/mabi=32/mabi=64
+! MULTILIB_DIRNAMES = 32 . 64
+ MULTILIB_OSDIRNAMES = ../lib32 ../lib ../lib64
+
+ EXTRA_MULTILIB_PARTS=crtbegin.o crtend.o crtbeginS.o crtendS.o crtbeginT.o
Property changes on: debian/patches/mips-biarch.dpatch
___
Name: svn:executable
+ *
Index: debian/rules.defs
===
--- debian/rules.defs (.../vendor/gcc-4.0/current)(revision 150)
+++ debian/rules.defs (.../src/gcc-4.0) (revision 150)
@@ -611,7 +611,7 @@
biarch_cpu := $(patsubst $(DEB_TARGET_GNU_CPU)=%, %, \
$(filter $(DEB_TARGET_GNU_CPU)=%, $(biarch_map)))
-biarch_archs := /i386/powerpc/sparc/s390/
+biarch_archs := /i386/powerpc/sparc/s390/mips/mipsel/
ifeq (biarch, $(findstring biarch,$(WITHOUT_LANG)))
biarch_archs :=
endif
@@ -647,7 +647,7 @@
export TARGET64_MA
Bug#343640: gcc-4.0: FTBFS on hurd-i386: conflicting patches getting applied
Michael Banck a écrit : Package: gcc-4.0 Version: 4.0.2-5 Severity: important Hi, since libgcj has been split off, both libffi-without-libgcj and disable-libgcj are getting applied on hurd-i386, which leads to a rejected patch in the latter: There is the same problem for kfreebsd-i386, I have sent a patch to Matthias Klose. It should be fixed in the SVN, at least it is the case for kfreebsd-i386. -- .''`. Aurelien Jarno | GPG: 1024D/F1BCDB73 : :' : Debian developer | Electrical Engineer `. `' [EMAIL PROTECTED] | [EMAIL PROTECTED] `-people.debian.org/~aurel32 | www.aurel32.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Processed: Re: Bug#333536: Upstream
Processing commands for [EMAIL PROTECTED]: > # Stephen R Marenka <[EMAIL PROTECTED]> writes: > # > Filed upstream as gnu pr 25043. Confirmed, fixed in 4.1. > forwarded 333536 http://gcc.gnu.org/PR25043 Bug#333536: [m68k] ICE in instantiate_virtual_regs_lossage Noted your statement that Bug has been forwarded to http://gcc.gnu.org/PR25043. > tags 333536 + upstream fixed-upstream Bug#333536: [m68k] ICE in instantiate_virtual_regs_lossage There were no tags set. Tags added: upstream, fixed-upstream > retitle 333536 [fixed in 4.1] [m68k] ICE in instantiate_virtual_regs_lossage Bug#333536: [m68k] ICE in instantiate_virtual_regs_lossage Changed Bug title. > End of message, stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[Bug rtl-optimization/23837] [4.0/4.1/4.2 regression] Wrong code with REG_NO_CONFLICT notes (caused by combine)
--- Comment #29 from steven at gcc dot gnu dot org 2005-12-16 22:19 --- Subject: Bug 23837 Author: steven Date: Fri Dec 16 22:19:09 2005 New Revision: 108690 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=108690 Log: PR rtl-optimization/23837 * optabs.c (expand_binop): For a multi-word rotate, never emit a REG_NO_CONFLICT block. Modified: trunk/gcc/ChangeLog trunk/gcc/optabs.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23837 --- You are receiving this mail because: --- You reported the bug, or are watching the reporter. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[Bug rtl-optimization/23837] [4.0/4.1 regression] Wrong code with REG_NO_CONFLICT notes (caused by combine)
--- Comment #30 from steven at gcc dot gnu dot org 2005-12-16 22:25 --- Should be fixed on the trunk. I guess the patch could be backported to all active branches, the bug is latent but present in all GCCs releases since 2.early. -- steven at gcc dot gnu dot org changed: What|Removed |Added Summary|[4.0/4.1/4.2 regression]|[4.0/4.1 regression] Wrong |Wrong code with |code with REG_NO_CONFLICT |REG_NO_CONFLICT notes |notes (caused by combine) |(caused by combine) | http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23837 --- You are receiving this mail because: --- You reported the bug, or are watching the reporter. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[Bug libobjc/13946] ObjC configured --with-objc-gc needs external Boehm gc
-- pinskia at gcc dot gnu dot org changed: What|Removed |Added Last reconfirmed|2005-09-10 16:35:43 |2005-12-17 00:32:51 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13946 --- You are receiving this mail because: --- You reported the bug, or are watching the reporter. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
