Hi all, I'd like to run the april snapshot but when I run the binary I
get the error: /lib/libc.so.6: version `GLIBC_2.3' not found.
Will CMUCL still work with glibc-2.2.5 if built on a machine with
glibc-2.2.5 installed?
I figured I'd try it out and see but I've run into problems, here are
the steps I've taken.
I'm using the november binaries to build the april sources on linux.
1> src/tools/create-target.sh build
Lisp = linux_gencgc
Motif = x86
2> src/tools/build-world.sh build lisp
[snip]
Warning: The C header file #p"target:lisp/internals.h" has changed.
Be sure to re-compile the startup code.
[Building Initial Core File (version 0) in file "target:lisp/kernel.core":
Writing 4096 bytes [1 page] from :READ-ONLY space
Writing 4096 bytes [1 page] from :STATIC space
Writing 13414400 bytes [3275 pages] from :DYNAMIC space
done]
T
3> src/tools/rebuild-world.sh build
[snip]
echo 0 > version
echo '1 + ' `cat version` | bc > ,version
mv ,version version
gcc -rdynamic -Wstrict-prototypes -Wall -g -DGENCGC -DLINKAGE_TABLE -DVERSION=`cat
version` -c ../../src/lisp/version.c
gcc -g -rdynamic -Xlinker --export-dynamic -Xlinker -Map -Xlinker foo -o ,lisp \
lisp.o coreparse.o alloc.o monitor.o print.o interr.o vars.o parse.o
interrupt.o search.o validate.o globals.o dynbind.o breakpoint.o regnames.o
backtrace.o save.o purify.o socket.o runprog.o time.o x86-arch.o x86-assem.o
linux-stubs.o Linux-os.o os-common.o elf.o gencgc.o version.o \
-ldl -lm
mv -f ,lisp lisp
echo 'Map file for lisp version ' `cat version` > ,lisp.nm
../../src/lisp/linux-nm lisp | grep -v " [F] " >> ,lisp.nm
mv ,lisp.nm lisp.nm
make: Leaving directory `/root/lisp/cmucl19/build/lisp'
4> src/tools/build-world.sh build lisp
[snip]
Warning: The C header file #p"target:lisp/internals.h" has changed.
Be sure to re-compile the startup code.
[Building Initial Core File (version 1) in file "target:lisp/kernel.core":
Writing 4096 bytes [1 page] from :READ-ONLY space
Writing 4096 bytes [1 page] from :STATIC space
Writing 13414400 bytes [3275 pages] from :DYNAMIC space
done]
T
This doesn't seem right to me, I don't think this second genesis
should have modified internals.h (??). The actual changes made are to
remove these lines:
#define LinkageSpaceStart 2952790016
#define SpaceStart_TargetReadOnly 268435456
#define SpaceStart_TargetStatic 671088640
#define SpaceStart_TargetDynamic 1207959552
#define SpaceStart_TargetForeignLinkage 2952790016
This predictably causes problems when I try to rebuild the runtime as
follows.
5a> src/tools/rebuild-world.sh build
[snip]
gcc -rdynamic -Wstrict-prototypes -Wall -g -DGENCGC -DLINKAGE_TABLE -I.
-I../../p86/lisp -I../../src/lisp -I- -I/usr/X11R6/include -DGENCGC -DLINKAGE_TABLE
-c -o validate.o ../../src/lisp/validate.c
../../src/lisp/validate.c: In function `validate':
../../src/lisp/validate.c:55: `SpaceStart_TargetReadOnly' undeclared (first use in
this function)
../../src/lisp/validate.c:55: (Each undeclared identifier is reported only once
../../src/lisp/validate.c:55: for each function it appears in.)
../../src/lisp/validate.c:62: `SpaceStart_TargetStatic' undeclared (first use in this
function)
../../src/lisp/validate.c:69: `SpaceStart_TargetDynamic' undeclared (first use in this
function)
../../src/lisp/validate.c:107: `LinkageSpaceStart' undeclared (first use in this
function)
make: *** [validate.o] Error 1
Since this failed I tried starting again from scratch, but this time
after finishing step 4 I ignored the warning that internals.h had
changed and tried moving on to the world-loading stage.
5b> src/tools/load-world.sh build "19a 2004-05-04"
Strange ... dynamic space lossage.
interrupt_handle_now: No handler for signal 11?
LDB monitor
ldb> unknown command: ``(in-package''
ldb> ldb> unknown command: ``(setf''
ldb> unknown command: ``'("build/"''
ldb> ldb> unknown command: ``(load''
ldb> ldb> unknown command: ``(pushnew''
ldb> unknown command: ``(pushnew''
ldb> unknown command: ``(pushnew''
ldb> ldb> unknown command: ``(load''
ldb> unknown command: ``19a''
ldb> ldb>
EOF on something other than a tty.
Obviously this isn't right.
Any help would be much appreciated, thanks.
--
jan
