From mexas at bristol.ac.uk Sat Jul 5 16:07:07 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Sat Jul 5 16:07:13 2008 Subject: graphviz port build fails on alpha FBSD 6.3 Message-ID: <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> (also posted to ports/graphviz maintainer) I get the following error when building this port on FreeBSD 6.3-STABLE #0: alpha (ds10l). My option set is default (given below as well). Any advice? Anybody else using graphviz? many thanks anton #make [skip] Making all in common gmake[3]: Entering directory `/usr/ports/graphics/graphviz/work/graphviz-2.20.2/lib/common' /bin/sh /usr/local/bin/libtool --tag=CC --mode=compile cc -DHAVE_CONFIG_H -I. -I../.. -I../.. -I../../lib/gvc -I../../lib/pack -I../../lib/fdpgen -I../../lib/pathplan -I../../lib/graph -I../../lib/cdt -I/usr/local/include -I/usr/local/include -I/usr/local/include/tk8.4 -I/usr/local/include/tcl8.4 -I/usr/local/include -O -pipe -mcpu=ev6 -mieee -Wstrict-prototypes -Wpointer-arith -Wall -ffast-math -MT utils.lo -MD -MP -MF .deps/utils.Tpo -c -o utils.lo utils.c cc -DHAVE_CONFIG_H -I. -I../.. -I../.. -I../../lib/gvc -I../../lib/pack -I../../lib/fdpgen -I../../lib/pathplan -I../../lib/graph -I../../lib/cdt -I/usr/local/include -I/usr/local/include -I/usr/local/include/tk8.4 -I/usr/local/include/tcl8.4 -I/usr/local/include -O -pipe -mcpu=ev6 -mieee -Wstrict-prototypes -Wpointer-arith -Wall -ffast-math -MT utils.lo -MD -MP -MF .deps/utils.Tpo -c utils.c -fPIC -DPIC -o .libs/utils.o utils.c:1733: error: conflicting types for 'strncasecmp' utils.h:25: error: previous declaration of 'strncasecmp' was here utils.c:1733: error: conflicting types for 'strncasecmp' utils.h:25: error: previous declaration of 'strncasecmp' was here gmake[3]: *** [utils.lo] Error 1 gmake[3]: Leaving directory `/usr/ports/graphics/graphviz/work/graphviz-2.20.2/lib/common' gmake[2]: *** [all-recursive] Error 1 gmake[2]: Leaving directory `/usr/ports/graphics/graphviz/work/graphviz-2.20.2/lib' gmake[1]: *** [all-recursive] Error 1 gmake[1]: Leaving directory `/usr/ports/graphics/graphviz/work/graphviz-2.20.2' gmake: *** [all] Error 2 *** Error code 2 Stop in /usr/ports/graphics/graphviz. *** Error code 1 Stop in /usr/ports/graphics/graphviz. # I used default option set: # cat /var/db/ports/graphviz/options # This file is auto-generated by 'make config'. # No user-servicable parts inside! # Options for graphviz-2.20.2 _OPTIONS_READ=graphviz-2.20.2 WITH_ICONV=true WITH_XPM=true WITH_DIGCOLA=true WITH_IPSEPCOLA=true WITHOUT_SMYRNA=true WITH_NLS=true WITHOUT_TK=true WITHOUT_PANGOCAIRO=true WITHOUT_RSVG=true WITHOUT_GTK=true WITHOUT_GDK_PIXBUF=true WITHOUT_GNOMEUI=true WITHOUT_MING=true WITHOUT_DEVIL=true WITHOUT_PERL=true WITHOUT_PHP=true WITHOUT_PYTHON=true WITHOUT_RUBY=true WITHOUT_LUA=true WITHOUT_TCL=true WITHOUT_GUILE=true # -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mexas at bristol.ac.uk Sat Jul 5 20:12:55 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Sat Jul 5 20:13:02 2008 Subject: graphviz port build fails on alpha FBSD 6.3 In-Reply-To: <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> References: <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> Message-ID: <20080705201242.GA20124@mech-cluster238.men.bris.ac.uk> On Sat, Jul 05, 2008 at 04:50:36PM +0100, Anton Shterenlikht wrote: > (also posted to ports/graphviz maintainer) > > I get the following error when building this port on > FreeBSD 6.3-STABLE #0: alpha (ds10l). [skip] > cc -DHAVE_CONFIG_H -I. -I../.. -I../.. -I../../lib/gvc -I../../lib/pack -I../../lib/fdpgen -I../../lib/pathplan -I../../lib/graph -I../../lib/cdt -I/usr/local/include -I/usr/local/include -I/usr/local/include/tk8.4 -I/usr/local/include/tcl8.4 -I/usr/local/include -O -pipe -mcpu=ev6 -mieee -Wstrict-prototypes -Wpointer-arith -Wall -ffast-math -MT utils.lo -MD -MP -MF .deps/utils.Tpo -c utils.c -fPIC -DPIC -o .libs/utils.o > utils.c:1733: error: conflicting types for 'strncasecmp' > utils.h:25: error: previous declaration of 'strncasecmp' was here It seems the error is due to configure script concluding there is no strncasecmp in the OS: # cd /usr/ports/graphics/graphviz # make configure > configure.out # grep strncasecmp configure.out checking for strncasecmp... no # But this function seems to be available: # man strncasecmp| head STRCASECMP(3) FreeBSD Library Functions Manual STRCASECMP(3) NAME strcasecmp, strncasecmp -- compare strings, ignoring case LIBRARY Standard C Library (libc, -lc) SYNOPSIS #include # Where in the configure script is the check for strncasecmp done? On i386 FBSD 7.0-stable all is fine: checking for strncasecmp... yes many thanks anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From bugmaster at FreeBSD.org Mon Jul 7 11:06:54 2008 From: bugmaster at FreeBSD.org (FreeBSD bugmaster) Date: Mon Jul 7 11:07:24 2008 Subject: Current problem reports assigned to freebsd-alpha@FreeBSD.org Message-ID: <200807071106.m67B6rEC061885@freefall.freebsd.org> Current FreeBSD problem reports Critical problems Serious problems S Tracker Resp. Description -------------------------------------------------------------------------------- o alpha/61940 alpha [sysinstall] Can't disklabel new disk from FreeBSD/alp o alpha/61973 alpha Machine Check on boot-up of AlphaServer 2100A RM s alpha/67626 alpha X crashes an alpha machine, resulting reboot o alpha/85346 alpha PREEMPTION causes unstability in Alpha4000 SMP kernel o alpha/105134 alpha 'panic: lockmgr: thread ... not exclusive lock owner' 5 problems total. Non-critical problems S Tracker Resp. Description -------------------------------------------------------------------------------- o alpha/25284 alpha PC164 won't reboot with graphics console o alpha/38031 alpha osf1.ko not loaded during boot-time of linux-emu enabl o alpha/48676 alpha Changing the baud rate of serial consoles for Alpha sy o alpha/67903 alpha hw.chipset.memory: 1099511627776 - thats way to much : 4 problems total. From mexas at bristol.ac.uk Mon Jul 7 13:40:10 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Mon Jul 7 13:40:17 2008 Subject: graphviz port build fails on alpha FBSD 6.3 In-Reply-To: References: <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> <20080705201242.GA20124@mech-cluster238.men.bris.ac.uk> Message-ID: <20080707134004.GA74030@mech-cluster238.men.bris.ac.uk> On Mon, Jul 07, 2008 at 03:06:41PM +0200, Dirk Meyer wrote: > Hallo Anton Shterenlikht, > > > > I get the following error when building this port on > > > FreeBSD 6.3-STABLE #0: alpha (ds10l). > > Thanks fo testing Alpha. > > > # make configure > configure.out > > # grep strncasecmp configure.out > > checking for strncasecmp... no > > # > > Can you look into the config.log. > There should be a section with "strncasecmp". > somwhow compile/link of it failed. > > > Where in the configure script is the check for strncasecmp done? > > On i386 FBSD 7.0-stable all is fine: checking for strncasecmp... yes > > search for strncasecmp > its a start of a for block in configure. Dirk I think the following fragment of config.log is relevant: %vi config.log 6929 configure:9333: result: no 6930 configure:9253: checking for strncasecmp 6931 configure:9309: cc -o conftest -O -pipe -mcpu=ev6 -mieee -Wstrict-prototypes -Wpointer-arith -Wall -ffast-math -I/usr/local/include -I/usr/local/include -L/usr/local/lib -L/usr/local/lib conftest.c >&5 6932 conftest.c:86: warning: function declaration isn't a prototype 6933 conftest.c:96: warning: function declaration isn't a prototype 6934 /usr/lib/crtfastmath.o(.text+0x1c): In function `set_fast_math': 6935 : undefined reference to `__ieee_set_fp_control' 6936 /usr/lib/crtfastmath.o(.text+0x20): In function `set_fast_math': 6937 : undefined reference to `__ieee_set_fp_control' 6938 configure:9315: $? = 1 which indicates that lines starting with 9253 in configure are of interest: %vi configure 9250 for ac_func in strcasecmp strncasecmp 9251 do 9252 as_ac_var=`echo "ac_cv_func_$ac_func" | $as_tr_sh` 9253 { echo "$as_me:$LINENO: checking for $ac_func" >&5 9254 echo $ECHO_N "checking for $ac_func... $ECHO_C" >&6; } 9255 if { as_var=$as_ac_var; eval "test \"\${$as_var+set}\" = set"; }; then 9256 echo $ECHO_N "(cached) $ECHO_C" >&6 9257 else 9258 cat >conftest.$ac_ext <<_ACEOF I'd be very happy to provide any other build output. many thanks for supporting this port on alpha! anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mexas at bristol.ac.uk Tue Jul 8 09:19:30 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Tue Jul 8 09:19:36 2008 Subject: lang/gcc42 on alpha Message-ID: <20080708091924.GA9171@mech-cluster238.men.bris.ac.uk> Hi I've been using lang/gcc42 on alpha for more than a year now. My calculation shows that the port was updated 25 times during last year. Each time I upgraded the port with no problem, except for as failure, which happens on my i386 just the same: +pid 3154 (as), uid 0: exited on signal 11 (core dumped) +pid 3160 (as), uid 0: exited on signal 11 (core dumped) +pid 33039 (as), uid 0: exited on signal 11 (core dumped) +pid 33045 (as), uid 0: exited on signal 11 (core dumped) Accordingly I wonder why lang/gcc42 is still marked NOT_FOR_ARCHS= alpha Are there other alpha users who have trouble with this port? If you don't think removing NOT_FOR_ARCHS is justified, perhaps you could change it to .if !defined(NOT_FOR_ARCHS) NOT_FOR_ARCHS= alpha .endif so that I can put NOT_FOR_ARCHS= '' in my make.conf and avoid having to edit the Makefile each upgrade. many thanks anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mexas at bristol.ac.uk Tue Jul 8 12:07:47 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Tue Jul 8 12:07:53 2008 Subject: firefox3 build fails on alpha Message-ID: <20080708120738.GA74833@mech-cluster238.men.bris.ac.uk> Hi firefox3 build fails on my alpha. Details below. Please advise anton # uname -mrs FreeBSD 6.3-STABLE alpha # sysctl -n kern.osreldate 603100 # # cd /usr/ports/www/firefox3 # make [skip] gmake[8]: Entering directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect /xptcall/src/md/unix' gmake[8]: *** No rule to make target `xptcinvoke_freebsd_alpha.o', needed by `li bxptcmd.a'. Stop. gmake[8]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ xptcall/src/md/unix' gmake[7]: *** [libs] Error 2 gmake[7]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ xptcall/src/md' gmake[6]: *** [libs] Error 2 gmake[6]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ xptcall/src' gmake[5]: *** [libs] Error 2 gmake[5]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ xptcall' gmake[4]: *** [libs] Error 2 gmake[4]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect' gmake[3]: *** [libs] Error 2 gmake[3]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom' gmake[2]: *** [libs_tier_xpcom] Error 2 gmake[2]: Leaving directory `/usr/ports/www/firefox3/work/mozilla' gmake[1]: *** [tier_xpcom] Error 2 gmake[1]: Leaving directory `/usr/ports/www/firefox3/work/mozilla' gmake: *** [default] Error 2 *** Error code 2 Stop in /usr/ports/www/firefox3. *** Error code 1 Stop in /usr/ports/www/firefox3. # I used the default configuration: # make showconfig ===> The following configuration options are available for firefox-3.0_2,1: DBUS=on "Enable D-BUS support" NEWTAB=on "Open external links in a new tab" SMB=off "Enable smb:// URI support using gnomevfs" DEBUG=off "Build a debugging image" LOGGING=off "Enable additional log messages" OPTIMIZED_CFLAGS=off "Enable some additional optimizations" ===> Use 'make config' to modify these settings # Please advise. many thanks anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mexas at bristol.ac.uk Tue Jul 8 12:23:59 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Tue Jul 8 12:24:11 2008 Subject: firefox3 build fails on alpha In-Reply-To: <20080708120738.GA74833@mech-cluster238.men.bris.ac.uk> References: <20080708120738.GA74833@mech-cluster238.men.bris.ac.uk> Message-ID: <20080708122352.GA76546@mech-cluster238.men.bris.ac.uk> On Tue, Jul 08, 2008 at 01:07:38PM +0100, Anton Shterenlikht wrote: > Hi > > firefox3 build fails on my alpha. Details below. > > Please advise > anton > > # uname -mrs > FreeBSD 6.3-STABLE alpha > # sysctl -n kern.osreldate > 603100 > # > > # cd /usr/ports/www/firefox3 > # make > > [skip] > > gmake[8]: Entering directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect > /xptcall/src/md/unix' > gmake[8]: *** No rule to make target `xptcinvoke_freebsd_alpha.o', needed by `li > bxptcmd.a'. Stop. > gmake[8]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ > xptcall/src/md/unix' > gmake[7]: *** [libs] Error 2 > gmake[7]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ > xptcall/src/md' > gmake[6]: *** [libs] Error 2 > gmake[6]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ > xptcall/src' > gmake[5]: *** [libs] Error 2 > gmake[5]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ > xptcall' > gmake[4]: *** [libs] Error 2 > gmake[4]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect' > > gmake[3]: *** [libs] Error 2 > gmake[3]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom' > gmake[2]: *** [libs_tier_xpcom] Error 2 > gmake[2]: Leaving directory `/usr/ports/www/firefox3/work/mozilla' > gmake[1]: *** [tier_xpcom] Error 2 > gmake[1]: Leaving directory `/usr/ports/www/firefox3/work/mozilla' > gmake: *** [default] Error 2 > *** Error code 2 > > Stop in /usr/ports/www/firefox3. > *** Error code 1 > > Stop in /usr/ports/www/firefox3. > # > > > I used the default configuration: > > > # make showconfig > ===> The following configuration options are available for firefox-3.0_2,1: > DBUS=on "Enable D-BUS support" > NEWTAB=on "Open external links in a new tab" > SMB=off "Enable smb:// URI support using gnomevfs" > DEBUG=off "Build a debugging image" > LOGGING=off "Enable additional log messages" > OPTIMIZED_CFLAGS=off "Enable some additional optimizations" > ===> Use 'make config' to modify these settings > # > A followup. It seems xptcinvoke_freebsd_alpha.cpp is missing. It is used in Makefile: # pwd /usr/ports/www/firefox3/work/mozilla/xpcom/reflect/xptcall/src/md/unix # ls xptcinvoke*alpha*cpp xptcinvoke_alpha_openbsd.cpp xptcinvoke_openvms_alpha.cpp xptcinvoke_linux_alpha.cpp xptcinvoke_osf1_alpha.cpp # vi Makefile [skip first 144 lines] 145 ###################################################################### 146 # Alpha 147 ###################################################################### 148 # 149 # Tru64/Alpha 150 # 151 ifeq ($(OS_ARCH)$(OS_TEST),OSF1alpha) 152 CPPSRCS := xptcinvoke_osf1_alpha.cpp xptcstubs_osf1_alpha.cpp 153 ASFILES := xptcinvoke_asm_osf1_alpha.s xptcstubs_asm_osf1_alpha.s 154 endif 155 # 156 # FreeBSD/Alpha 157 # -> 158 ifeq ($(OS_ARCH)$(OS_TEST),FreeBSDalpha) -> 159 CPPSRCS := xptcinvoke_freebsd_alpha.cpp xptcstubs_freebsd_alpha.cpp 160 endif 161 # 162 # Linux/Alpha 163 # -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mexas at bristol.ac.uk Thu Jul 10 11:06:26 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Thu Jul 10 11:06:33 2008 Subject: graphviz port build fails on alpha FBSD 6.3 In-Reply-To: <20080707134004.GA74030@mech-cluster238.men.bris.ac.uk> References: <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> <20080705201242.GA20124@mech-cluster238.men.bris.ac.uk> <20080707134004.GA74030@mech-cluster238.men.bris.ac.uk> Message-ID: <20080710110617.GA81447@mech-cluster238.men.bris.ac.uk> On Mon, Jul 07, 2008 at 02:40:04PM +0100, Anton Shterenlikht wrote: > On Mon, Jul 07, 2008 at 03:06:41PM +0200, Dirk Meyer wrote: > > Hallo Anton Shterenlikht, > > > > > > I get the following error when building this port on > > > > FreeBSD 6.3-STABLE #0: alpha (ds10l). > > > > Thanks fo testing Alpha. > > > > > # make configure > configure.out > > > # grep strncasecmp configure.out > > > checking for strncasecmp... no > > > # > > > > Can you look into the config.log. > > There should be a section with "strncasecmp". > > somwhow compile/link of it failed. > > > > > Where in the configure script is the check for strncasecmp done? > > > On i386 FBSD 7.0-stable all is fine: checking for strncasecmp... yes > > > > search for strncasecmp > > its a start of a for block in configure. > > Dirk > > I think the following fragment of config.log is relevant: > > %vi config.log > > 6929 configure:9333: result: no > 6930 configure:9253: checking for strncasecmp > 6931 configure:9309: cc -o conftest -O -pipe -mcpu=ev6 -mieee > -Wstrict-prototypes -Wpointer-arith -Wall -ffast-math > -I/usr/local/include -I/usr/local/include > -L/usr/local/lib -L/usr/local/lib conftest.c >&5 > 6932 conftest.c:86: warning: function declaration isn't a prototype > 6933 conftest.c:96: warning: function declaration isn't a prototype > 6934 /usr/lib/crtfastmath.o(.text+0x1c): In function `set_fast_math': > 6935 : undefined reference to `__ieee_set_fp_control' > 6936 /usr/lib/crtfastmath.o(.text+0x20): In function `set_fast_math': > 6937 : undefined reference to `__ieee_set_fp_control' > 6938 configure:9315: $? = 1 > > > which indicates that lines starting with 9253 in configure are of interest: > > %vi configure > > 9250 for ac_func in strcasecmp strncasecmp > 9251 do > 9252 as_ac_var=`echo "ac_cv_func_$ac_func" | $as_tr_sh` > 9253 { echo "$as_me:$LINENO: checking for $ac_func" >&5 > 9254 echo $ECHO_N "checking for $ac_func... $ECHO_C" >&6; } > 9255 if { as_var=$as_ac_var; eval "test \"\${$as_var+set}\" = set"; }; then > 9256 echo $ECHO_N "(cached) $ECHO_C" >&6 > 9257 else > 9258 cat >conftest.$ac_ext <<_ACEOF Dirk I tried to look at this problem myself but got stuck. Anyway I think I need to show you the full logs for strncasecmp. I don't really understand what goes on with this bit of code. Is the problem in the actual c code at the end? In this case perhaps I could just test my strncasecmp on some simple c code? Or is the problem in the configure script itself? many thanks anton configure:9253: checking for strncasecmp configure:9309: cc -o conftest -O -pipe -mcpu=ev6 -mieee -Wstrict-prototypes -Wp ointer-arith -Wall -ffast-math -I/usr/local/include -I/usr/local/include -L/usr /local/lib -L/usr/local/lib conftest.c >&5 conftest.c:86: warning: function declaration isn't a prototype conftest.c:96: warning: function declaration isn't a prototype /usr/lib/crtfastmath.o(.text+0x1c): In function `set_fast_math': : undefined reference to `__ieee_set_fp_control' /usr/lib/crtfastmath.o(.text+0x20): In function `set_fast_math': : undefined reference to `__ieee_set_fp_control' configure:9315: $? = 1 configure: failed program was: | /* confdefs.h. */ | #define PACKAGE_NAME "graphviz" | #define PACKAGE_TARNAME "graphviz" | #define PACKAGE_VERSION "2.20.2" | #define PACKAGE_STRING "graphviz 2.20.2" | #define PACKAGE_BUGREPORT "http://www.graphviz.org/" | #define GVPLUGIN_VERSION 5 | #define DEFAULT_FONTPATH "/usr/share/fonts/default/Type1:/usr/X11R6/lib/X11/fo nts/TrueType:/usr/X11R6/lib/X11/fonts/truetype:/usr/X11R6/lib/X11/fonts/TTF:/usr /share/fonts/TrueType:/usr/share/fonts/truetype:/usr/openwin/lib/X11/fonts/TrueT ype:/usr/X11R6/lib/X11/fonts/Type1:/usr/common/share/fonts/ttf" | #define PATHSEPARATOR ":" | #define NO_POSTSCRIPT_ALIAS 0 | #define DEFAULT_DPI 96 | #define PACKAGE "graphviz" | #define VERSION "2.20.2" | #define ENABLE_SHARED 1 | #define YYTEXT_POINTER 1 | #define NO_FPERR 1 | #define HAVE_SYS_TYPES_H 1 | #define HAVE_SYS_STAT_H 1 | #define HAVE_STDLIB_H 1 | #define HAVE_STRING_H 1 | #define HAVE_MEMORY_H 1 | #define HAVE_STRINGS_H 1 | #define HAVE_INTTYPES_H 1 | #define HAVE_STDINT_H 1 | #define HAVE_UNISTD_H 1 | #define HAVE_STDARG_H 1 | #define HAVE_STDDEF_H 1 | #define HAVE_STDDEF_H 1 | #define HAVE_STDLIB_H 1 | #define HAVE_STDINT_H 1 | #define HAVE_SEARCH_H 1 | #define HAVE_GETOPT_H 1 | #define HAVE_PTHREAD_H 1 | #define HAVE_FLOAT_H 1 | #define HAVE_LIMITS_H 1 | #define HAVE_TERMIOS_H 1 | #define HAVE_ERRNO_H 1 | #define HAVE_TIME_H 1 | #define HAVE_UNISTD_H 1 | #define HAVE_FENV_H 1 | #define HAVE_STRING_H 1 | #define HAVE_STRINGS_H 1 | #define HAVE_INTTYPES_H 1 | #define HAVE_SETJMP_H 1 | #define HAVE_SYS_TIME_H 1 | #define HAVE_SYS_TIMES_H 1 | #define HAVE_SYS_TYPES_H 1 | #define HAVE_SYS_SELECT_H 1 | #define HAVE_SYS_SOCKET_H 1 | #define HAVE_SYS_STAT_H 1 | #define HAVE_SYS_MMAN_H 1 | #define HAVE_SYS_IOCTL_H 1 | #define HAVE_LANGINFO_H 1 | #define HAVE_LIBINTL_H 1 | #define TIME_WITH_SYS_TIME 1 | #define HAVE_DIRENT_H 1 | #define HAVE__BOOL 1 | #define HAVE_STDBOOL_H 1 | #define HAVE_BOOL 1 | #define C_ALLOCA 1 | #define STACK_DIRECTION -1 | /* end confdefs.h. */ | /* Define strncasecmp to an innocuous variant, in case declares str ncasecmp. | For example, HP-UX 11i declares gettimeofday. */ | #define strncasecmp innocuous_strncasecmp | | /* System header to define __stub macros and hopefully few prototypes, | which can conflict with char strncasecmp (); below. | Prefer to if __STDC__ is defined, since | exists even on freestanding compilers. */ | | #ifdef __STDC__ | # include | #else | # include | #endif | | #undef strncasecmp | | /* Override any GCC internal prototype to avoid an error. | Use char because int might match the return type of a GCC | builtin and then its argument prototype would still apply. */ | #ifdef __cplusplus | extern "C" | #endif | char strncasecmp (); | /* The GNU C library defines this for functions which it implements | to always fail with ENOSYS. Some functions are actually named | something starting with __ and the normal name is an alias. */ | #if defined __stub_strncasecmp || defined __stub___strncasecmp | choke me | #endif | | int | main () | { | return strncasecmp (); | ; | return 0; | } configure:9333: result: no -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From naddy at mips.inka.de Thu Jul 10 14:28:36 2008 From: naddy at mips.inka.de (Christian Weisgerber) Date: Thu Jul 10 14:28:43 2008 Subject: graphviz port build fails on alpha FBSD 6.3 References: <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> <20080705201242.GA20124@mech-cluster238.men.bris.ac.uk> <20080707134004.GA74030@mech-cluster238.men.bris.ac.uk> Message-ID: Anton Shterenlikht wrote: > 6931 configure:9309: cc -o conftest -O -pipe -mcpu=ev6 -mieee > -Wstrict-prototypes -Wpointer-arith -Wall -ffast-math > -I/usr/local/include -I/usr/local/include > -L/usr/local/lib -L/usr/local/lib conftest.c >&5 > 6932 conftest.c:86: warning: function declaration isn't a prototype > 6933 conftest.c:96: warning: function declaration isn't a prototype > 6934 /usr/lib/crtfastmath.o(.text+0x1c): In function `set_fast_math': > 6935 : undefined reference to `__ieee_set_fp_control' -ffast-math isn't supported on alpha. Don't use it. Ports that unconditionally use it need to be fixed. -- Christian "naddy" Weisgerber naddy@mips.inka.de From mexas at bristol.ac.uk Thu Jul 10 14:38:01 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Thu Jul 10 14:38:21 2008 Subject: graphviz port build fails on alpha FBSD 6.3 In-Reply-To: References: <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> <20080705201242.GA20124@mech-cluster238.men.bris.ac.uk> <20080707134004.GA74030@mech-cluster238.men.bris.ac.uk> Message-ID: <20080710143753.GA42554@mech-cluster238.men.bris.ac.uk> On Thu, Jul 10, 2008 at 01:55:48PM +0000, Christian Weisgerber wrote: > Anton Shterenlikht wrote: > > > 6931 configure:9309: cc -o conftest -O -pipe -mcpu=ev6 -mieee > > -Wstrict-prototypes -Wpointer-arith -Wall -ffast-math > > -I/usr/local/include -I/usr/local/include > > -L/usr/local/lib -L/usr/local/lib conftest.c >&5 > > 6932 conftest.c:86: warning: function declaration isn't a prototype > > 6933 conftest.c:96: warning: function declaration isn't a prototype > > 6934 /usr/lib/crtfastmath.o(.text+0x1c): In function `set_fast_math': > > 6935 : undefined reference to `__ieee_set_fp_control' > > -ffast-math isn't supported on alpha. Don't use it. Ports that > unconditionally use it need to be fixed. Naddy thanks, just in time! I was just discussing this with the port maintainer. Actually if I put CFLAGS= -O -pipe -ffast-math in /etc/make.conf I get stopped at the config stage: configure:3832: error: C compiler cannot create executables I cannot find much info on -ffast-math. Could you please expand on your answer. many thanks anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mexas at bristol.ac.uk Thu Jul 10 14:56:17 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Thu Jul 10 14:56:38 2008 Subject: graphviz port build fails on alpha FBSD 6.3 In-Reply-To: <20080710143753.GA42554@mech-cluster238.men.bris.ac.uk> References: <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> <20080705201242.GA20124@mech-cluster238.men.bris.ac.uk> <20080707134004.GA74030@mech-cluster238.men.bris.ac.uk> <20080710143753.GA42554@mech-cluster238.men.bris.ac.uk> Message-ID: <20080710145610.GA55854@mech-cluster238.men.bris.ac.uk> On Thu, Jul 10, 2008 at 03:37:53PM +0100, Anton Shterenlikht wrote: > On Thu, Jul 10, 2008 at 01:55:48PM +0000, Christian Weisgerber wrote: > > Anton Shterenlikht wrote: > > > > > 6931 configure:9309: cc -o conftest -O -pipe -mcpu=ev6 -mieee > > > -Wstrict-prototypes -Wpointer-arith -Wall -ffast-math > > > -I/usr/local/include -I/usr/local/include > > > -L/usr/local/lib -L/usr/local/lib conftest.c >&5 > > > 6932 conftest.c:86: warning: function declaration isn't a prototype > > > 6933 conftest.c:96: warning: function declaration isn't a prototype > > > 6934 /usr/lib/crtfastmath.o(.text+0x1c): In function `set_fast_math': > > > 6935 : undefined reference to `__ieee_set_fp_control' > > > > -ffast-math isn't supported on alpha. Don't use it. Ports that > > unconditionally use it need to be fixed. > > Naddy thanks, just in time! > I was just discussing this with the port maintainer. > Actually if I put > > CFLAGS= -O -pipe -ffast-math > > in /etc/make.conf I get stopped at the config stage: > > configure:3832: error: C compiler cannot create executables > > I cannot find much info on -ffast-math. Could you please expand on your > answer. Found it, should've looked better, sorry. >From gcc man pages: -ffast-math Sets -fno-math-errno, -funsafe-math-optimizations, -fno-trap- ping-math, -ffinite-math-only, -fno-rounding-math and -fno-signal- ing-nans. This option causes the preprocessor macro "__FAST_MATH__" to be defined. This option should never be turned on by any -O option since it can result in incorrect output for programs which depend on an exact implementation of IEEE or ISO rules/specifications for math func- tions. plus PR ports/113599: graphics/dri: fix build on alpha http://www.FreeBSD.org/cgi/query-pr.cgi?pr=113599 FreeBSD/alpha can't handle -ffast-math (fails to link due to some unimplemented functions), so don't try to build graphics/dri with that option there. I'll try your solution and add .if ${ARCH} == alpha FAST_MATH= .else FAST_MATH= -ffast-math .endif to the Makefile Will report on results. many thanks anton > > many thanks > anton > > -- > Anton Shterenlikht > Room 2.6, Queen's Building > Mech Eng Dept > Bristol University > University Walk, Bristol BS8 1TR, UK > Tel: +44 (0)117 928 8233 > Fax: +44 (0)117 929 4423 > _______________________________________________ > freebsd-alpha@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-alpha > To unsubscribe, send any mail to "freebsd-alpha-unsubscribe@freebsd.org" -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mexas at bristol.ac.uk Thu Jul 10 15:11:35 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Thu Jul 10 15:11:43 2008 Subject: graphviz port build fails on alpha FBSD 6.3 In-Reply-To: <20080710145610.GA55854@mech-cluster238.men.bris.ac.uk> References: <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> <20080705201242.GA20124@mech-cluster238.men.bris.ac.uk> <20080707134004.GA74030@mech-cluster238.men.bris.ac.uk> <20080710143753.GA42554@mech-cluster238.men.bris.ac.uk> <20080710145610.GA55854@mech-cluster238.men.bris.ac.uk> Message-ID: <20080710151121.GA92045@mech-cluster238.men.bris.ac.uk> On Thu, Jul 10, 2008 at 03:56:10PM +0100, Anton Shterenlikht wrote: > On Thu, Jul 10, 2008 at 03:37:53PM +0100, Anton Shterenlikht wrote: > > On Thu, Jul 10, 2008 at 01:55:48PM +0000, Christian Weisgerber wrote: > > > Anton Shterenlikht wrote: > > > > > > > 6931 configure:9309: cc -o conftest -O -pipe -mcpu=ev6 -mieee > > > > -Wstrict-prototypes -Wpointer-arith -Wall -ffast-math > > > > -I/usr/local/include -I/usr/local/include > > > > -L/usr/local/lib -L/usr/local/lib conftest.c >&5 > > > > 6932 conftest.c:86: warning: function declaration isn't a prototype > > > > 6933 conftest.c:96: warning: function declaration isn't a prototype > > > > 6934 /usr/lib/crtfastmath.o(.text+0x1c): In function `set_fast_math': > > > > 6935 : undefined reference to `__ieee_set_fp_control' > > > > > > -ffast-math isn't supported on alpha. Don't use it. Ports that > > > unconditionally use it need to be fixed. > > > > Naddy thanks, just in time! > > I was just discussing this with the port maintainer. > > Actually if I put > > > > CFLAGS= -O -pipe -ffast-math > > > > in /etc/make.conf I get stopped at the config stage: > > > > configure:3832: error: C compiler cannot create executables > > > > I cannot find much info on -ffast-math. Could you please expand on your > > answer. > > Found it, should've looked better, sorry. > > >From gcc man pages: > > -ffast-math > Sets -fno-math-errno, -funsafe-math-optimizations, -fno-trap- > ping-math, -ffinite-math-only, -fno-rounding-math and -fno-signal- > ing-nans. > > This option causes the preprocessor macro "__FAST_MATH__" to be > defined. > > This option should never be turned on by any -O option since it can > result in incorrect output for programs which depend on an exact > implementation of IEEE or ISO rules/specifications for math func- > tions. > > plus PR ports/113599: graphics/dri: fix build on alpha > http://www.FreeBSD.org/cgi/query-pr.cgi?pr=113599 > > FreeBSD/alpha can't handle -ffast-math (fails to link due to some > unimplemented functions), so don't try to build graphics/dri with > that option there. > > > I'll try your solution and add > > .if ${ARCH} == alpha > FAST_MATH= > .else > FAST_MATH= -ffast-math > .endif > > to the Makefile > > Will report on results. no, adding this to Makefile doesn't help. It is being ignored, e.g.: cc -DHAVE_CONFIG_H -I. -I../.. -I../../lib/ast -I../../lib/sfio -I/usr/local/include -I/usr/local/include -O -pipe -mcpu=ev6 -mieee -Wstrict-prototypes -Wpointer-arith -Wall ---> -ffast-math -MT vmopen.lo -MD -MP -MF .deps/vmopen.Tpo -c vmopen.c -o vmopen.o >/dev/null 2>&1 it must be somewhere in configure.ac probably. -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mexas at bristol.ac.uk Thu Jul 10 16:01:18 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Thu Jul 10 16:01:27 2008 Subject: SOLVED: Re: graphviz port build fails on alpha FBSD 6.3 In-Reply-To: <20080710151121.GA92045@mech-cluster238.men.bris.ac.uk> References: <20080705155035.GA88847@mech-cluster238.men.bris.ac.uk> <20080705201242.GA20124@mech-cluster238.men.bris.ac.uk> <20080707134004.GA74030@mech-cluster238.men.bris.ac.uk> <20080710143753.GA42554@mech-cluster238.men.bris.ac.uk> <20080710145610.GA55854@mech-cluster238.men.bris.ac.uk> <20080710151121.GA92045@mech-cluster238.men.bris.ac.uk> Message-ID: <20080710160104.GA93414@mech-cluster238.men.bris.ac.uk> On Thu, Jul 10, 2008 at 04:11:22PM +0100, Anton Shterenlikht wrote: > On Thu, Jul 10, 2008 at 03:56:10PM +0100, Anton Shterenlikht wrote: > > On Thu, Jul 10, 2008 at 03:37:53PM +0100, Anton Shterenlikht wrote: > > > On Thu, Jul 10, 2008 at 01:55:48PM +0000, Christian Weisgerber wrote: > > > > Anton Shterenlikht wrote: > > > > > > > > > 6931 configure:9309: cc -o conftest -O -pipe -mcpu=ev6 -mieee > > > > > -Wstrict-prototypes -Wpointer-arith -Wall -ffast-math > > > > > -I/usr/local/include -I/usr/local/include > > > > > -L/usr/local/lib -L/usr/local/lib conftest.c >&5 > > > > > 6932 conftest.c:86: warning: function declaration isn't a prototype > > > > > 6933 conftest.c:96: warning: function declaration isn't a prototype > > > > > 6934 /usr/lib/crtfastmath.o(.text+0x1c): In function `set_fast_math': > > > > > 6935 : undefined reference to `__ieee_set_fp_control' > > > > > > > > -ffast-math isn't supported on alpha. Don't use it. Ports that > > > > unconditionally use it need to be fixed. > > > > > > Naddy thanks, just in time! > > > I was just discussing this with the port maintainer. > > > Actually if I put > > > > > > CFLAGS= -O -pipe -ffast-math > > > > > > in /etc/make.conf I get stopped at the config stage: > > > > > > configure:3832: error: C compiler cannot create executables > > > > > > I cannot find much info on -ffast-math. Could you please expand on your > > > answer. > > > > Found it, should've looked better, sorry. > > > > >From gcc man pages: > > > > -ffast-math > > Sets -fno-math-errno, -funsafe-math-optimizations, -fno-trap- > > ping-math, -ffinite-math-only, -fno-rounding-math and -fno-signal- > > ing-nans. > > > > This option causes the preprocessor macro "__FAST_MATH__" to be > > defined. > > > > This option should never be turned on by any -O option since it can > > result in incorrect output for programs which depend on an exact > > implementation of IEEE or ISO rules/specifications for math func- > > tions. > > > > plus PR ports/113599: graphics/dri: fix build on alpha > > http://www.FreeBSD.org/cgi/query-pr.cgi?pr=113599 > > > > FreeBSD/alpha can't handle -ffast-math (fails to link due to some > > unimplemented functions), so don't try to build graphics/dri with > > that option there. > > > > > > I'll try your solution and add > > > > .if ${ARCH} == alpha > > FAST_MATH= > > .else > > FAST_MATH= -ffast-math > > .endif > > > > to the Makefile > > > > Will report on results. > > no, adding this to Makefile doesn't help. It is being ignored, e.g.: > > cc -DHAVE_CONFIG_H -I. -I../.. -I../../lib/ast -I../../lib/sfio > -I/usr/local/include -I/usr/local/include -O -pipe -mcpu=ev6 -mieee > -Wstrict-prototypes -Wpointer-arith -Wall > ---> -ffast-math > -MT vmopen.lo -MD -MP -MF .deps/vmopen.Tpo -c vmopen.c -o vmopen.o > >/dev/null 2>&1 > > it must be somewhere in configure.ac probably. I noted this line in configure: # icc doesn't like -ffast-math CFLAGS=`echo ${CFLAGS} | sed 's/ -ffast-math//'` accordingly I just added the same line at the end of CFLAGS block: # alpha doesn't like -ffast-math CFLAGS=`echo ${CFLAGS} | sed 's/ -ffast-math//'` now configure detects strncasecmp on OS: configure:9256: checking for strcasecmp configure:9312: cc -o conftest -O -pipe -mcpu=ev6 -mieee -Wstrict-prototypes -Wp ointer-arith -Wall -I/usr/local/include -I/usr/local/include -L/usr/local/lib - L/usr/local/lib conftest.c >&5 conftest.c:107: warning: function declaration isn't a prototype conftest.c:117: warning: function declaration isn't a prototype configure:9318: $? = 0 configure:9336: result: yes configure:9256: checking for strncasecmp configure:9312: cc -o conftest -O -pipe -mcpu=ev6 -mieee -Wstrict-prototypes -Wp ointer-arith -Wall -I/usr/local/include -I/usr/local/include -L/usr/local/lib - L/usr/local/lib conftest.c >&5 conftest.c:108: warning: function declaration isn't a prototype conftest.c:118: warning: function declaration isn't a prototype configure:9318: $? = 0 configure:9336: result: yes and build and intall pass fine: install -C -o root -g wheel -m 444 /usr/ports/graphics/graphviz/work/graphviz- 2.20.2/tclpkg/tclpathplan/demo/pathplan_data/*.dat /usr/local/share/examples/gr aphviz/pathplan_data/ ===> Compressing manual pages for graphviz-2.20.2 ===> Registering installation for graphviz-2.20.2 Many thanks to Naddy for the advise. anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mezz7 at cox.net Sat Jul 12 22:55:22 2008 From: mezz7 at cox.net (Jeremy Messenger) Date: Sat Jul 12 22:55:27 2008 Subject: firefox3 build fails on alpha In-Reply-To: <20080708122352.GA76546@mech-cluster238.men.bris.ac.uk> References: <20080708120738.GA74833@mech-cluster238.men.bris.ac.uk> <20080708122352.GA76546@mech-cluster238.men.bris.ac.uk> Message-ID: On Tue, 08 Jul 2008 07:23:52 -0500, Anton Shterenlikht wrote: > On Tue, Jul 08, 2008 at 01:07:38PM +0100, Anton Shterenlikht wrote: >> Hi >> >> firefox3 build fails on my alpha. Details below. >> >> Please advise >> anton >> >> # uname -mrs >> FreeBSD 6.3-STABLE alpha >> # sysctl -n kern.osreldate >> 603100 >> # >> >> # cd /usr/ports/www/firefox3 >> # make >> >> [skip] >> >> gmake[8]: Entering directory >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect >> /xptcall/src/md/unix' >> gmake[8]: *** No rule to make target `xptcinvoke_freebsd_alpha.o', >> needed by `li >> bxptcmd.a'. Stop. >> gmake[8]: Leaving directory >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ >> xptcall/src/md/unix' >> gmake[7]: *** [libs] Error 2 >> gmake[7]: Leaving directory >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ >> xptcall/src/md' >> gmake[6]: *** [libs] Error 2 >> gmake[6]: Leaving directory >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ >> xptcall/src' >> gmake[5]: *** [libs] Error 2 >> gmake[5]: Leaving directory >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ >> xptcall' >> gmake[4]: *** [libs] Error 2 >> gmake[4]: Leaving directory >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect' >> >> gmake[3]: *** [libs] Error 2 >> gmake[3]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom' >> gmake[2]: *** [libs_tier_xpcom] Error 2 >> gmake[2]: Leaving directory `/usr/ports/www/firefox3/work/mozilla' >> gmake[1]: *** [tier_xpcom] Error 2 >> gmake[1]: Leaving directory `/usr/ports/www/firefox3/work/mozilla' >> gmake: *** [default] Error 2 >> *** Error code 2 >> >> Stop in /usr/ports/www/firefox3. >> *** Error code 1 >> >> Stop in /usr/ports/www/firefox3. >> # >> >> >> I used the default configuration: >> >> >> # make showconfig >> ===> The following configuration options are available for >> firefox-3.0_2,1: >> DBUS=on "Enable D-BUS support" >> NEWTAB=on "Open external links in a new tab" >> SMB=off "Enable smb:// URI support using gnomevfs" >> DEBUG=off "Build a debugging image" >> LOGGING=off "Enable additional log messages" >> OPTIMIZED_CFLAGS=off "Enable some additional optimizations" >> ===> Use 'make config' to modify these settings >> # >> > > A followup. It seems xptcinvoke_freebsd_alpha.cpp is missing. It > is used in Makefile: I think most of us don't have any alpha machine. You might have to create patch for us. I think, the alpha support has been dropped so it's pointless for us to work on alpha support. Cheers, Mezz > # pwd > /usr/ports/www/firefox3/work/mozilla/xpcom/reflect/xptcall/src/md/unix > # ls xptcinvoke*alpha*cpp > xptcinvoke_alpha_openbsd.cpp xptcinvoke_openvms_alpha.cpp > xptcinvoke_linux_alpha.cpp xptcinvoke_osf1_alpha.cpp > # vi Makefile > > [skip first 144 lines] > > 145 > ###################################################################### > 146 # Alpha > 147 > ###################################################################### > 148 # > 149 # Tru64/Alpha > 150 # > 151 ifeq ($(OS_ARCH)$(OS_TEST),OSF1alpha) > 152 CPPSRCS := xptcinvoke_osf1_alpha.cpp xptcstubs_osf1_alpha.cpp > 153 ASFILES := xptcinvoke_asm_osf1_alpha.s > xptcstubs_asm_osf1_alpha.s > 154 endif > 155 # > 156 # FreeBSD/Alpha > 157 # > -> 158 ifeq ($(OS_ARCH)$(OS_TEST),FreeBSDalpha) > -> 159 CPPSRCS := xptcinvoke_freebsd_alpha.cpp > xptcstubs_freebsd_alpha.cpp > 160 endif > 161 # > 162 # Linux/Alpha > 163 # -- mezz7@cox.net - mezz@FreeBSD.org FreeBSD GNOME Team http://www.FreeBSD.org/gnome/ - gnome@FreeBSD.org From wb at freebie.xs4all.nl Sat Jul 12 23:11:47 2008 From: wb at freebie.xs4all.nl (Wilko Bulte) Date: Sat Jul 12 23:12:00 2008 Subject: firefox3 build fails on alpha In-Reply-To: References: <20080708120738.GA74833@mech-cluster238.men.bris.ac.uk> <20080708122352.GA76546@mech-cluster238.men.bris.ac.uk> Message-ID: <20080712225929.GA22401@freebie.xs4all.nl> Quoting Jeremy Messenger, who wrote on Sat, Jul 12, 2008 at 05:44:11PM -0500 .. > On Tue, 08 Jul 2008 07:23:52 -0500, Anton Shterenlikht > wrote: > > > On Tue, Jul 08, 2008 at 01:07:38PM +0100, Anton Shterenlikht wrote: > >> Hi > >> > >> firefox3 build fails on my alpha. Details below. > >> > >> Please advise > >> anton > >> > >> # uname -mrs > >> FreeBSD 6.3-STABLE alpha > >> # sysctl -n kern.osreldate > >> 603100 > >> # > >> > >> # cd /usr/ports/www/firefox3 > >> # make > >> > >> [skip] > >> > >> gmake[8]: Entering directory > >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect > >> /xptcall/src/md/unix' > >> gmake[8]: *** No rule to make target `xptcinvoke_freebsd_alpha.o', > >> needed by `li > >> bxptcmd.a'. Stop. > >> gmake[8]: Leaving directory > >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ > >> xptcall/src/md/unix' > >> gmake[7]: *** [libs] Error 2 > >> gmake[7]: Leaving directory > >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ > >> xptcall/src/md' > >> gmake[6]: *** [libs] Error 2 > >> gmake[6]: Leaving directory > >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ > >> xptcall/src' > >> gmake[5]: *** [libs] Error 2 > >> gmake[5]: Leaving directory > >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect/ > >> xptcall' > >> gmake[4]: *** [libs] Error 2 > >> gmake[4]: Leaving directory > >> `/usr/ports/www/firefox3/work/mozilla/xpcom/reflect' > >> > >> gmake[3]: *** [libs] Error 2 > >> gmake[3]: Leaving directory `/usr/ports/www/firefox3/work/mozilla/xpcom' > >> gmake[2]: *** [libs_tier_xpcom] Error 2 > >> gmake[2]: Leaving directory `/usr/ports/www/firefox3/work/mozilla' > >> gmake[1]: *** [tier_xpcom] Error 2 > >> gmake[1]: Leaving directory `/usr/ports/www/firefox3/work/mozilla' > >> gmake: *** [default] Error 2 > >> *** Error code 2 > >> > >> Stop in /usr/ports/www/firefox3. > >> *** Error code 1 > >> > >> Stop in /usr/ports/www/firefox3. > >> # > >> > >> > >> I used the default configuration: > >> > >> > >> # make showconfig > >> ===> The following configuration options are available for > >> firefox-3.0_2,1: > >> DBUS=on "Enable D-BUS support" > >> NEWTAB=on "Open external links in a new tab" > >> SMB=off "Enable smb:// URI support using gnomevfs" > >> DEBUG=off "Build a debugging image" > >> LOGGING=off "Enable additional log messages" > >> OPTIMIZED_CFLAGS=off "Enable some additional optimizations" > >> ===> Use 'make config' to modify these settings > >> # > >> > > > > A followup. It seems xptcinvoke_freebsd_alpha.cpp is missing. It > > is used in Makefile: > > I think most of us don't have any alpha machine. You might have to create > patch for us. I think, the alpha support has been dropped so it's > pointless for us to work on alpha support. The latest release of FreeBSD supporting the Alpha platform port is RELENG_6. Anything newer no longer supports Alpha. Wilko From bugmaster at FreeBSD.org Mon Jul 14 11:06:54 2008 From: bugmaster at FreeBSD.org (FreeBSD bugmaster) Date: Mon Jul 14 11:07:09 2008 Subject: Current problem reports assigned to freebsd-alpha@FreeBSD.org Message-ID: <200807141106.m6EB6slv014336@freefall.freebsd.org> Current FreeBSD problem reports Critical problems Serious problems S Tracker Resp. Description -------------------------------------------------------------------------------- o alpha/61940 alpha [sysinstall] Can't disklabel new disk from FreeBSD/alp o alpha/61973 alpha Machine Check on boot-up of AlphaServer 2100A RM s alpha/67626 alpha X crashes an alpha machine, resulting reboot o alpha/85346 alpha PREEMPTION causes unstability in Alpha4000 SMP kernel o alpha/105134 alpha 'panic: lockmgr: thread ... not exclusive lock owner' 5 problems total. Non-critical problems S Tracker Resp. Description -------------------------------------------------------------------------------- o alpha/25284 alpha PC164 won't reboot with graphics console o alpha/38031 alpha osf1.ko not loaded during boot-time of linux-emu enabl o alpha/48676 alpha Changing the baud rate of serial consoles for Alpha sy o alpha/67903 alpha hw.chipset.memory: 1099511627776 - thats way to much : 4 problems total. From gerald at pfeifer.com Sun Jul 20 08:44:58 2008 From: gerald at pfeifer.com (Gerald Pfeifer) Date: Sun Jul 20 08:45:04 2008 Subject: lang/gcc42 on alpha In-Reply-To: <20080708091924.GA9171@mech-cluster238.men.bris.ac.uk> References: <20080708091924.GA9171@mech-cluster238.men.bris.ac.uk> Message-ID: Hi Anton, On Tue, 8 Jul 2008, Anton Shterenlikht wrote: > I've been using lang/gcc42 on alpha for more than a year now. > My calculation shows that the port was updated 25 times during > last year. Each time I upgraded the port with no problem, except > for as failure, which happens on my i386 just the same: thanks for the report. This is the first feedback I got during the more than three years that we have had NOT_FOR_ARCHS=alpha for the newer lang/gcc ports. > Accordingly I wonder why lang/gcc42 is still marked > NOT_FOR_ARCHS= alpha > > Are there other alpha users who have trouble with this port? The original reports that triggered this change came from the FreeBSD Ports Build Cluster. I did some archeology, and here is what I had sent to a member of the portmgr team back than in response, which resulted in the suggestion to use NOT_FOR_ARCHS: I just wasted a couple of ours bringing sparc64 back to bootstrap-land for GCC 4.0 and GCC 4.1 because *nobody*, not a single FreeBSD developer or user apparently tried to build either for many moons. And now the same is starting for alpha; isn't that a waste of our time, for the sake of one or two enthusiasts? I'm fine giving it a try and remove NOT_FOR_ARCHS=alpha for lang/gcc42. If there are problems with it, I may ask you for help or put this back in, though. ;-) > [I] can put NOT_FOR_ARCHS= '' in my make.conf and avoid having to edit > the Makefile each upgrade. Shouldn't "cvs update" handle this gracefully and keep the change while merging in the other updates? Gerald -- Gerald (Jerry) Pfeifer gerald@pfeifer.com http://www.pfeifer.com/gerald/ From bugmaster at FreeBSD.org Mon Jul 21 11:06:51 2008 From: bugmaster at FreeBSD.org (FreeBSD bugmaster) Date: Mon Jul 21 11:07:02 2008 Subject: Current problem reports assigned to freebsd-alpha@FreeBSD.org Message-ID: <200807211106.m6LB6pVm031790@freefall.freebsd.org> Current FreeBSD problem reports Critical problems Serious problems S Tracker Resp. Description -------------------------------------------------------------------------------- o alpha/61940 alpha [sysinstall] Can't disklabel new disk from FreeBSD/alp o alpha/61973 alpha Machine Check on boot-up of AlphaServer 2100A RM s alpha/67626 alpha X crashes an alpha machine, resulting reboot o alpha/85346 alpha PREEMPTION causes unstability in Alpha4000 SMP kernel o alpha/105134 alpha 'panic: lockmgr: thread ... not exclusive lock owner' 5 problems total. Non-critical problems S Tracker Resp. Description -------------------------------------------------------------------------------- o alpha/25284 alpha PC164 won't reboot with graphics console o alpha/38031 alpha osf1.ko not loaded during boot-time of linux-emu enabl o alpha/48676 alpha Changing the baud rate of serial consoles for Alpha sy o alpha/67903 alpha hw.chipset.memory: 1099511627776 - thats way to much : 4 problems total. From bugmaster at FreeBSD.org Mon Jul 28 11:06:52 2008 From: bugmaster at FreeBSD.org (FreeBSD bugmaster) Date: Mon Jul 28 11:07:02 2008 Subject: Current problem reports assigned to freebsd-alpha@FreeBSD.org Message-ID: <200807281106.m6SB6pFA078828@freefall.freebsd.org> Current FreeBSD problem reports Critical problems Serious problems S Tracker Resp. Description -------------------------------------------------------------------------------- o alpha/61940 alpha [sysinstall] Can't disklabel new disk from FreeBSD/alp o alpha/61973 alpha Machine Check on boot-up of AlphaServer 2100A RM s alpha/67626 alpha X crashes an alpha machine, resulting reboot o alpha/85346 alpha PREEMPTION causes unstability in Alpha4000 SMP kernel o alpha/105134 alpha 'panic: lockmgr: thread ... not exclusive lock owner' 5 problems total. Non-critical problems S Tracker Resp. Description -------------------------------------------------------------------------------- o alpha/25284 alpha PC164 won't reboot with graphics console o alpha/38031 alpha osf1.ko not loaded during boot-time of linux-emu enabl o alpha/48676 alpha Changing the baud rate of serial consoles for Alpha sy o alpha/67903 alpha hw.chipset.memory: 1099511627776 - thats way to much : 4 problems total. From mexas at bristol.ac.uk Thu Jul 31 21:58:10 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Thu Jul 31 21:58:17 2008 Subject: lang/gcc42 on alpha In-Reply-To: References: <20080708091924.GA9171@mech-cluster238.men.bris.ac.uk> Message-ID: <20080731215803.GA35238@mech-cluster238.men.bris.ac.uk> On Sun, Jul 20, 2008 at 10:19:11AM +0200, Gerald Pfeifer wrote: > Hi Anton, > > On Tue, 8 Jul 2008, Anton Shterenlikht wrote: > > I've been using lang/gcc42 on alpha for more than a year now. > > My calculation shows that the port was updated 25 times during > > last year. Each time I upgraded the port with no problem, except > > for as failure, which happens on my i386 just the same: > > thanks for the report. This is the first feedback I got during the > more than three years that we have had NOT_FOR_ARCHS=alpha for the > newer lang/gcc ports. > > > Accordingly I wonder why lang/gcc42 is still marked > > NOT_FOR_ARCHS= alpha > > > > Are there other alpha users who have trouble with this port? > > The original reports that triggered this change came from the FreeBSD > Ports Build Cluster. I did some archeology, and here is what I had sent > to a member of the portmgr team back than in response, which resulted in > the suggestion to use NOT_FOR_ARCHS: > > I just wasted a couple of ours bringing sparc64 back to bootstrap-land > for GCC 4.0 and GCC 4.1 because *nobody*, not a single FreeBSD developer > or user apparently tried to build either for many moons. And now the > same is starting for alpha; isn't that a waste of our time, for the sake > of one or two enthusiasts? > > I'm fine giving it a try and remove NOT_FOR_ARCHS=alpha for lang/gcc42. > If there are problems with it, I may ask you for help or put this back > in, though. ;-) Gerald, many thanks. Yes, please do this. I'm happy to help with testing. thanks anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mexas at bristol.ac.uk Thu Jul 31 22:10:29 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Thu Jul 31 22:10:35 2008 Subject: future for FBSD on alpha Message-ID: <20080731221015.GA35293@mech-cluster238.men.bris.ac.uk> I wonder how many people run FBSD on alpha. According to bsdstats.org very few - less than 5 boxes. There must be more. Is there a way to find out how many people are on this list? Is it a good indicator? I, for one, am very keen to have some level of support for FBSD on alpha, even at 6-stable branch. However, if the actual usage is indeed very small, it must be expected that ports maintainers will stop supporting alpha soon. I'm trying to find out if there are people on this list, who, like myself, are keen to see FBSD supported on alpha for say 1-3 more years. If yes, we might be able to lobby ports maintainers not to drop support for alpha ports. I'm thinking at the moment about gcc42, firefox3, lapack, lapack95, blas. many thanks anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mexas at bristol.ac.uk Thu Jul 31 22:24:52 2008 From: mexas at bristol.ac.uk (Anton Shterenlikht) Date: Thu Jul 31 22:24:58 2008 Subject: firefox3 build fails on alpha In-Reply-To: <20080712225929.GA22401@freebie.xs4all.nl> References: <20080708120738.GA74833@mech-cluster238.men.bris.ac.uk> <20080708122352.GA76546@mech-cluster238.men.bris.ac.uk> <20080712225929.GA22401@freebie.xs4all.nl> Message-ID: <20080731222442.GA35346@mech-cluster238.men.bris.ac.uk> On Sun, Jul 13, 2008 at 12:59:29AM +0200, Wilko Bulte wrote: > Quoting Jeremy Messenger, who wrote on Sat, Jul 12, 2008 at 05:44:11PM -0500 .. > > On Tue, 08 Jul 2008 07:23:52 -0500, Anton Shterenlikht > > wrote: > > > > > > A followup. It seems xptcinvoke_freebsd_alpha.cpp is missing. It > > > is used in Makefile: > > > > I think most of us don't have any alpha machine. You might have to create > > patch for us. I think, the alpha support has been dropped so it's > > pointless for us to work on alpha support. > > The latest release of FreeBSD supporting the Alpha platform port is > RELENG_6. Anything newer no longer supports Alpha. I'm happy with RELENG_6 on alpha. I'm happy to test firefox3 on alpha, but need to learn about pathes. In this particular case it seems that (at least) 2 files are missing from the distribution: xptcinvoke_freebsd_alpha.cpp xptcstubs_freebsd_alpha.cpp (curiously the corresponding *alpha.cpp files are present for osf1, linux, openvms and openbsd). I think that whoever created the tar file simply forgot to add the freebsd files. I did try to contact mozilla developers directly, but no reply so far. I understand that mozilla developers put out the tar file. Perhaps the port maintainers could alert the mozilla team about the missing files, or maybe even get these 2 files. I'd be very greatful. many thanks anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 From mezz7 at cox.net Thu Jul 31 22:42:11 2008 From: mezz7 at cox.net (Jeremy Messenger) Date: Thu Jul 31 22:42:23 2008 Subject: firefox3 build fails on alpha In-Reply-To: <20080731222442.GA35346@mech-cluster238.men.bris.ac.uk> References: <20080708120738.GA74833@mech-cluster238.men.bris.ac.uk> <20080708122352.GA76546@mech-cluster238.men.bris.ac.uk> <20080712225929.GA22401@freebie.xs4all.nl> <20080731222442.GA35346@mech-cluster238.men.bris.ac.uk> Message-ID: On Thu, 31 Jul 2008 17:24:43 -0500, Anton Shterenlikht wrote: > On Sun, Jul 13, 2008 at 12:59:29AM +0200, Wilko Bulte wrote: >> Quoting Jeremy Messenger, who wrote on Sat, Jul 12, 2008 at 05:44:11PM >> -0500 .. >> > On Tue, 08 Jul 2008 07:23:52 -0500, Anton Shterenlikht >> > wrote: >> > > >> > > A followup. It seems xptcinvoke_freebsd_alpha.cpp is missing. It >> > > is used in Makefile: >> > >> > I think most of us don't have any alpha machine. You might have to >> create >> > patch for us. I think, the alpha support has been dropped so it's >> > pointless for us to work on alpha support. >> >> The latest release of FreeBSD supporting the Alpha platform port is >> RELENG_6. Anything newer no longer supports Alpha. > > I'm happy with RELENG_6 on alpha. > I'm happy to test firefox3 on alpha, but need to learn about pathes. > > In this particular case it seems that (at least) 2 files > are missing from the distribution: > > xptcinvoke_freebsd_alpha.cpp > xptcstubs_freebsd_alpha.cpp > > (curiously the corresponding *alpha.cpp files are present for osf1, > linux, > openvms and openbsd). > > I think that whoever created the tar file simply forgot to add the > freebsd > files. > > I did try to contact mozilla developers directly, but no reply so far. > I understand that mozilla developers put out the tar file. Perhaps the > port maintainers could alert the mozilla team about the missing files, > or maybe even get these 2 files. I'd be very greatful. Have you tried 3.0.1 yet? The Makefile.in has changed by Mozilla team: --------------------------------------------------- ###################################################################### # # Tru64/Alpha # ifeq ($(OS_ARCH)$(OS_TEST),OSF1alpha) CPPSRCS := xptcinvoke_osf1_alpha.cpp xptcstubs_osf1_alpha.cpp ASFILES := xptcinvoke_asm_osf1_alpha.s xptcstubs_asm_osf1_alpha.s endif # # Linux/Alpha # ifneq (,$(filter Linuxalpha FreeBSDalpha NetBSDalpha,$(OS_ARCH)$(OS_TEST))) CPPSRCS := xptcinvoke_linux_alpha.cpp xptcstubs_linux_alpha.cpp endif --------------------------------------------------- But I have no idea if it works. Cheers, Mezz > many thanks > anton -- mezz7@cox.net - mezz@FreeBSD.org FreeBSD GNOME Team http://www.FreeBSD.org/gnome/ - gnome@FreeBSD.org From rjv at weblinkmo.com Thu Jul 31 23:39:28 2008 From: rjv at weblinkmo.com (Richard J. Valenta) Date: Thu Jul 31 23:39:42 2008 Subject: future for FBSD on alpha In-Reply-To: <20080731221015.GA35293@mech-cluster238.men.bris.ac.uk> References: <20080731221015.GA35293@mech-cluster238.men.bris.ac.uk> Message-ID: <8F8C8CFE62002A4EB82EE1CE9F2ED7D40ECACD@depot.AxxisCorp.local> I'm using it for an old DS20 server downstairs... the machine doesn't get the use it used to, but I like having a stable OS on it... and love FreeBSD's server environment. So, if you weren't counting me - add me to the list... ? Richard J. Valenta Note:? The information contained in this e-mail message, including any attachments, is legally privileged and confidential.? If you are not the intended recipient, you are hereby notified that any reading, use or dissemination of this message is strictly prohibited.? If you have received this message in error, please immediately, notify us by telephone and delete this message from your system. ? This message is NOT or NOLONGER ENCRYPTED ? -----BEGIN PGP SIGNATURE----- iQA/AwUBPxB11XHywJHFtakmEQLR7QCePuVbfTACQLrLVaN2aHVqHtJJHOIAoJP3 lSDDPwENBV3deRd/CMhMq2+N =xnNu -----END PGP SIGNATURE----- -----Original Message----- From: owner-freebsd-alpha@freebsd.org [mailto:owner-freebsd-alpha@freebsd.org] On Behalf Of Anton Shterenlikht Sent: Thursday, July 31, 2008 5:10 PM To: freebsd-alpha@freebsd.org Subject: future for FBSD on alpha I wonder how many people run FBSD on alpha. According to bsdstats.org very few - less than 5 boxes. There must be more. Is there a way to find out how many people are on this list? Is it a good indicator? I, for one, am very keen to have some level of support for FBSD on alpha, even at 6-stable branch. However, if the actual usage is indeed very small, it must be expected that ports maintainers will stop supporting alpha soon. I'm trying to find out if there are people on this list, who, like myself, are keen to see FBSD supported on alpha for say 1-3 more years. If yes, we might be able to lobby ports maintainers not to drop support for alpha ports. I'm thinking at the moment about gcc42, firefox3, lapack, lapack95, blas. many thanks anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 _______________________________________________ freebsd-alpha@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-alpha To unsubscribe, send any mail to "freebsd-alpha-unsubscribe@freebsd.org"