Hi,
I just did a git clone and tried to make SR deb packages, but I'm getting the following error:
../../lib/kcore/../../atomic/atomic_unknown.h: At top level: ../../lib/kcore/../../atomic/atomic_unknown.h:60: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘*’ token ../../lib/kcore/../../atomic/atomic_unknown.h:136: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘*’ token ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_inc_int’: ../../lib/kcore/../../atomic/atomic_unknown.h:206: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h:206: error: (Each undeclared identifier is reported only once ../../lib/kcore/../../atomic/atomic_unknown.h:206: error: for each function it appears in.) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_dec_int’: ../../lib/kcore/../../atomic/atomic_unknown.h:207: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_and_int’: ../../lib/kcore/../../atomic/atomic_unknown.h:208: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_or_int’: ../../lib/kcore/../../atomic/atomic_unknown.h:209: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_inc_and_test_int’: ../../lib/kcore/../../atomic/atomic_unknown.h:210: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_dec_and_test_int’: ../../lib/kcore/../../atomic/atomic_unknown.h:211: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_get_and_set_int’: ../../lib/kcore/../../atomic/atomic_unknown.h:212: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_cmpxchg_int’: ../../lib/kcore/../../atomic/atomic_unknown.h:213: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_add_int’: ../../lib/kcore/../../atomic/atomic_unknown.h:216: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_inc_long’: ../../lib/kcore/../../atomic/atomic_unknown.h:218: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_dec_long’: ../../lib/kcore/../../atomic/atomic_unknown.h:219: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_and_long’: ../../lib/kcore/../../atomic/atomic_unknown.h:220: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_or_long’: ../../lib/kcore/../../atomic/atomic_unknown.h:221: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_inc_and_test_long’: ../../lib/kcore/../../atomic/atomic_unknown.h:222: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_dec_and_test_long’: ../../lib/kcore/../../atomic/atomic_unknown.h:223: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_get_and_set_long’: ../../lib/kcore/../../atomic/atomic_unknown.h:224: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_cmpxchg_long’: ../../lib/kcore/../../atomic/atomic_unknown.h:225: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘atomic_add_long’: ../../lib/kcore/../../atomic/atomic_unknown.h:228: error: ‘_atomic_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘mb_atomic_get_int’: ../../lib/kcore/../../atomic/atomic_unknown.h:257: error: ‘__membar_lock’ undeclared (first use in this function) ../../lib/kcore/../../atomic/atomic_unknown.h: In function ‘mb_atomic_get_long’: ../../lib/kcore/../../atomic/atomic_unknown.h:270: error: ‘__membar_lock’ undeclared (first use in this function) enum_mod.c: At top level: enum_mod.c:40: error: ‘NAME’ undeclared here (not in a function) enum_mod.c:40: error: expected ‘,’ or ‘;’ before string constant make[3]: *** [enum_mod.o] Error 1 make[3]: se sale del directorio `/usr/src/sip-router/modules/enum' make[2]: *** [modules] Error 1 make[2]: se sale del directorio `/usr/src/sip-router' make[1]: *** [build-stamp] Error 2 make[1]: se sale del directorio `/usr/src/sip-router' dpkg-buildpackage: failure: debian/rules build gave error exit status 2 make: [deb] Error 2 (no tiene efecto)
All dependencies seem to de installed, and I'm using Debian Lenny. Is there anything I can test to try to get it compile?
Thanks in advance!
On Jun 18, 2009 at 01:24, Sa?l Ibarra saghul@gmail.com wrote:
Hi,
I just did a git clone and tried to make SR deb packages, but I'm getting the following error:
../../lib/kcore/../../atomic/atomic_unknown.h: At top level: ../../lib/kcore/../../atomic/atomic_unknown.h:60: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?*? token ../../lib/kcore/../../atomic/atomic_unknown.h:136: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?*? token ../../lib/kcore/../../atomic/atomic_unknown.h: In function ?atomic_inc_int?:
It looks very strange, first of all atomic_unkown.h is included only for unknown architecures (I suppose you use an x86, sun, ppc, arm or mips machine) and even then you shouldn't get any errors.
Could you send me please the output of make (a pasted gcc ... line will be enough), the config.mak file and the output of: uname -a git rev-parse HEAD git log -1 --pretty=oneline atomic/atomic_unknown.h md5sum atomic/atomic_unknown.h
Andrei
Hi Andrei,
First, sorry for the late reply, but I've been unable to touch a computer since :-/
When you mentioned the arch stuff I realised what the issue could be: it's a Xen virtual domain, not a phisycal server.
Uname -a show correct info anyway:
playground:/usr/src/sip-router# uname -a Linux playground 2.6.26-1-xen-686 #1 SMP Sat Jan 10 22:52:47 UTC 2009 i686 GNU/Linux
Do you still need all other information?
Thanks in advance!!
On Fri, Jun 19, 2009 at 4:26 PM, Andrei Pelinescu-Onciulandrei@iptel.org wrote:
On Jun 18, 2009 at 01:24, Sa?l Ibarra saghul@gmail.com wrote:
Hi,
I just did a git clone and tried to make SR deb packages, but I'm getting the following error:
../../lib/kcore/../../atomic/atomic_unknown.h: At top level: ../../lib/kcore/../../atomic/atomic_unknown.h:60: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?*? token ../../lib/kcore/../../atomic/atomic_unknown.h:136: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?*? token ../../lib/kcore/../../atomic/atomic_unknown.h: In function ?atomic_inc_int?:
It looks very strange, first of all atomic_unkown.h is included only for unknown architecures (I suppose you use an x86, sun, ppc, arm or mips machine) and even then you shouldn't get any errors.
Could you send me please the output of make (a pasted gcc ... line will be enough), the config.mak file and the output of: uname -a git rev-parse HEAD git log -1 --pretty=oneline atomic/atomic_unknown.h md5sum atomic/atomic_unknown.h
Andrei
On Jun 30, 2009 at 14:33, Sa?l Ibarra saghul@gmail.com wrote:
Hi Andrei,
First, sorry for the late reply, but I've been unable to touch a computer since :-/
When you mentioned the arch stuff I realised what the issue could be: it's a Xen virtual domain, not a phisycal server.
Uname -a show correct info anyway:
playground:/usr/src/sip-router# uname -a Linux playground 2.6.26-1-xen-686 #1 SMP Sat Jan 10 22:52:47 UTC 2009 i686 GNU/Linux
There shouldn't be any problem.
What compiler do you use? Is it a normal gcc?
Do you still need all other information?
Yes, I need at least config.mak (if you don't have it run make cfg) and the make output (make >/tmp/log and then send me /tmp/log). md5sum atomic/atomic_unknown.h might also help.
Andrei
On Fri, Jun 19, 2009 at 4:26 PM, Andrei Pelinescu-Onciulandrei@iptel.org wrote:
On Jun 18, 2009 at 01:24, Sa?l Ibarra saghul@gmail.com wrote:
Hi,
I just did a git clone and tried to make SR deb packages, but I'm getting the following error:
../../lib/kcore/../../atomic/atomic_unknown.h: At top level: ../../lib/kcore/../../atomic/atomic_unknown.h:60: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?*? token ../../lib/kcore/../../atomic/atomic_unknown.h:136: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?*? token ../../lib/kcore/../../atomic/atomic_unknown.h: In function ?atomic_inc_int?:
It looks very strange, first of all atomic_unkown.h is included only for unknown architecures (I suppose you use an x86, sun, ppc, arm or mips machine) and even then you shouldn't get any errors.
Could you send me please the output of make ?(a pasted gcc ... line will be enough), the config.mak file and the output of: uname -a git rev-parse HEAD git log -1 --pretty=oneline atomic/atomic_unknown.h md5sum atomic/atomic_unknown.h
Andrei
-- Sa?l -- "Nunca subestimes el ancho de banda de un cami?n lleno de disketes."
Hi Andrei,
I had a very small config.mak file, just this 2 lines:
C_DEFS:=$(filter-out $(DEFS_RM) $(extra_defs),$(C_DEFS)) $(extra_defs) CFLAGS:=$(filter-out $(CFLAGS_RM) $(CC_EXTRA_OPTS), $(CFLAGS)) $(CC_EXTRA_OPTS)
But after doing make cfg it got a lot more content and now I don't get that error. I think the error was produced because when doing make deb, make cfg is not executed and thus ARCH variable is undefined.
Now I'm getting the following error:
lex.yy.c:5075: warning: ‘input’ defined but not used gcc -g -O9 -funroll-loops -Wcast-align -minline-all-stringops -falign-loops -ftree-vectorize -mtune=athlon64 -Wall -DNAME='"ser"' -DVERSION='"2.99.0-dev01-serk"' -DARCH='"i386"' -DOS='linux_' -DOS_QUOTED='"linux"' -DCOMPILER='"gcc 4.3.2"' -D__CPU_i386 -D__OS_linux -DSER_VER=2099000 -DCFG_DIR='"/usr/local/etc/ser/"' -DPKG_MALLOC -DSHM_MEM -DSHM_MMAP -DDNS_IP_HACK -DUSE_IPV6 -DUSE_MCAST -DUSE_TCP -DDISABLE_NAGLE -DHAVE_RESOLV_RES -DUSE_DNS_CACHE -DUSE_DNS_FAILOVER -DUSE_DST_BLACKLIST -DUSE_NAPTR -DDBG_QM_MALLOC -DUSE_TLS -DTLS_HOOKS -DFAST_LOCK -DADAPTIVE_WAIT -DADAPTIVE_WAIT_LOOPS=1024 -DCC_GCC_LIKE_ASM -DHAVE_GETHOSTBYNAME2 -DHAVE_UNION_SEMUN -DHAVE_SCHED_YIELD -DHAVE_MSG_NOSIGNAL -DHAVE_MSGHDR_MSG_CONTROL -DHAVE_ALLOCA_H -DHAVE_TIMEGM -DHAVE_SCHED_SETSCHEDULER -DHAVE_EPOLL -DHAVE_SIGIO_RT -DSIGINFO64_WORKARROUND -DUSE_FUTEX -DHAVE_SELECT -c cfg.tab.c -o cfg.tab.o gcc -Wl,-O2 -Wl,-E action.o atomic_ops.o basex.o bit_scan.o cfg_core.o cfg_parser.o core_cmd.o crc.o daemonize.o data_lump.o data_lump_rpl.o dns_cache.o dprint.o dset.o dst_blacklist.o endianness.o error.o flags.o forward.o hash_func.o id.o io_wait.o ip_addr.o local_timer.o lock_ops.o lvalue.o main.o md5.o md5utils.o mod_fix.o modparam.o msg_translator.o nonsip_hooks.o pass_fd.o proxy.o pt.o pvapi.o qvalue.o re.o receive.o resolve.o route.o route_struct.o rpc_lookup.o rvalue.o script_cb.o sctp_options.o sctp_server.o select_buf.o select.o select_core.o ser_stun.o signals.o socket_info.o sr_compat.o sr_module.o stats.o switch.o tcp_main.o tcp_options.o tcp_read.o timer.o timer_proc.o tls_hooks.o tsend.o udp_server.o usr_avp.o ut.o y.tab.o mem/dl_malloc.o mem/f_malloc.o mem/ll_malloc.o mem/mem.o mem/memtest.o mem/q_malloc.o mem/sf_malloc.o mem/shm_mem.o mem/vq_malloc.o parser/hf.o parser/msg_parser.o parser/parse_allow.o parser/parse_body.o parser/parse_content.o parser/parse_cseq.o parser/parse_date.o parser/parse_disposition.o parser/parse_diversion.o parser/parse_event.o parser/parse_expires.o parser/parse_fline.o parser/parse_from.o parser/parse_hname2.o parser/parse_hostport.o parser/parse_identity.o parser/parse_identityinfo.o parser/parse_methods.o parser/parse_nameaddr.o parser/parse_param.o parser/parse_refer_to.o parser/parse_retry_after.o parser/parser_f.o parser/parse_rpid.o parser/parse_rr.o parser/parse_sipifmatch.o parser/parse_subscription_state.o parser/parse_to.o parser/parse_uri.o parser/parse_via.o parser/digest/digest.o parser/digest/digest_parser.o parser/digest/param_parser.o parser/contact/contact.o parser/contact/parse_contact.o parser/sdp/sdp.o parser/sdp/sdp_helpr_funcs.o rand/fastrand.o rand/isaac/rand.o cfg/cfg.o cfg/cfg_ctx.o cfg/cfg_script.o cfg/cfg_select.o cfg/cfg_struct.o lex.yy.o cfg.tab.o -ldl -lresolv -o ser cfg.tab.o: In function `yyparse': /usr/src/sip-router/cfg.tab.c:4478: multiple definition of `yyparse' y.tab.o:/usr/src/sip-router/y.tab.c:4478: first defined here collect2: ld returned 1 exit status make[2]: *** [ser] Error 1 make[2]: se sale del directorio `/usr/src/sip-router' make[1]: *** [build-stamp] Error 2 make[1]: se sale del directorio `/usr/src/sip-router' dpkg-buildpackage: failure: debian/rules build gave error exit status 2 make: [deb] Error 2 (no tiene efecto)
Should I open a bug report on the bugtracker regarding the Debian packaging?
Thank you very much for your help and interest,
On Tue, Jun 30, 2009 at 2:43 PM, Andrei Pelinescu-Onciulandrei@iptel.org wrote:
On Jun 30, 2009 at 14:33, Sa?l Ibarra saghul@gmail.com wrote:
Hi Andrei,
First, sorry for the late reply, but I've been unable to touch a computer since :-/
When you mentioned the arch stuff I realised what the issue could be: it's a Xen virtual domain, not a phisycal server.
Uname -a show correct info anyway:
playground:/usr/src/sip-router# uname -a Linux playground 2.6.26-1-xen-686 #1 SMP Sat Jan 10 22:52:47 UTC 2009 i686 GNU/Linux
There shouldn't be any problem.
What compiler do you use? Is it a normal gcc?
Do you still need all other information?
Yes, I need at least config.mak (if you don't have it run make cfg) and the make output (make >/tmp/log and then send me /tmp/log). md5sum atomic/atomic_unknown.h might also help.
Andrei
On Fri, Jun 19, 2009 at 4:26 PM, Andrei Pelinescu-Onciulandrei@iptel.org wrote:
On Jun 18, 2009 at 01:24, Sa?l Ibarra saghul@gmail.com wrote:
Hi,
I just did a git clone and tried to make SR deb packages, but I'm getting the following error:
../../lib/kcore/../../atomic/atomic_unknown.h: At top level: ../../lib/kcore/../../atomic/atomic_unknown.h:60: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?*? token ../../lib/kcore/../../atomic/atomic_unknown.h:136: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?*? token ../../lib/kcore/../../atomic/atomic_unknown.h: In function ?atomic_inc_int?:
It looks very strange, first of all atomic_unkown.h is included only for unknown architecures (I suppose you use an x86, sun, ppc, arm or mips machine) and even then you shouldn't get any errors.
Could you send me please the output of make ?(a pasted gcc ... line will be enough), the config.mak file and the output of: uname -a git rev-parse HEAD git log -1 --pretty=oneline atomic/atomic_unknown.h md5sum atomic/atomic_unknown.h
Andrei
-- Sa?l -- "Nunca subestimes el ancho de banda de un cami?n lleno de disketes."
On Jun 30, 2009 at 18:58, Sa?l Ibarra saghul@gmail.com wrote:
Hi Andrei,
I had a very small config.mak file, just this 2 lines:
C_DEFS:=$(filter-out $(DEFS_RM) $(extra_defs),$(C_DEFS)) $(extra_defs) CFLAGS:=$(filter-out $(CFLAGS_RM) $(CC_EXTRA_OPTS), $(CFLAGS)) $(CC_EXTRA_OPTS)
But after doing make cfg it got a lot more content and now I don't get that error. I think the error was produced because when doing make deb, make cfg is not executed and thus ARCH variable is undefined.
Not running make cfg should not produce any problem. If config.mak is missing it will be created automatically (make cfg is automatically run if config.mak is missing).
I don't understand why you did get such a small config.mak. If you manage to reproduce the problem, please let me know.
Now I'm getting the following error:
[...]
cfg.tab.o: In function `yyparse': /usr/src/sip-router/cfg.tab.c:4478: multiple definition of `yyparse' y.tab.o:/usr/src/sip-router/y.tab.c:4478: first defined here
^^^^^^^ The file y.tab.c should not exist. It is not created by the sip-router compilation process (it creates cfg.tab.c instead). I have this reported before, but in the other case there was some mixup will old source files.
Did you run bison cfg.y by hand? (that would explain it)
The fix is: rm y.tab.c y.tab.h and then retry compiling. If you manage to reproduce it somehow, please let me know. (I can't reproduce it)
[...]
Should I open a bug report on the bugtracker regarding the Debian packaging?
No, because is not yet clear that this is a bug.
Note also that the debian packaging has not yet been updated, is still the original one from ser and it most likely doesn't build a proper sip-router deb (if it builds one at all).
Andrei
Andrei Pelinescu-Onciul wrote:
On Jun 30, 2009 at 18:58, Sa?l Ibarra saghul@gmail.com wrote:
Now I'm getting the following error:
[...]
cfg.tab.o: In function `yyparse': /usr/src/sip-router/cfg.tab.c:4478: multiple definition of `yyparse' y.tab.o:/usr/src/sip-router/y.tab.c:4478: first defined here
^^^^^^^
The file y.tab.c should not exist. It is not created by the sip-router compilation process (it creates cfg.tab.c instead). I have this reported before, but in the other case there was some mixup will old source files.
Did you run bison cfg.y by hand? (that would explain it)
The fix is: rm y.tab.c y.tab.h and then retry compiling. If you manage to reproduce it somehow, please let me know. (I can't reproduce it)
I remember I had the same problem and I also tried a "make deb" some time before.
regards klaus
Hi,
Not running make cfg should not produce any problem. If config.mak is missing it will be created automatically (make cfg is automatically run if config.mak is missing).
I don't understand why you did get such a small config.mak. If you manage to reproduce the problem, please let me know.
I think it's debian packaging related...
The file y.tab.c should not exist. It is not created by the sip-router compilation process (it creates cfg.tab.c instead). I have this reported before, but in the other case there was some mixup will old source files.
Did you run bison cfg.y by hand? (that would explain it)
The fix is: rm y.tab.c y.tab.h and then retry compiling. If you manage to reproduce it somehow, please let me know. (I can't reproduce it)
I dind't run bison by hand, but after removing those two files compilation continues well :)
Note also that the debian packaging has not yet been updated, is still the original one from ser and it most likely doesn't build a proper sip-router deb (if it builds one at all).
That's right, I see debian packaging is not yet quite ready, because now I'm getting some errors regarding directories. I'm going to leave the make deb for now and move on to 'regular' compilation :) Thank you for your help and when someone diggs into debian packaging let me know if I can help with some testing :)
Best regards,
Andrei Pelinescu-Onciul schrieb:
Now I'm getting the following error:
[...]
cfg.tab.o: In function `yyparse': /usr/src/sip-router/cfg.tab.c:4478: multiple definition of `yyparse' y.tab.o:/usr/src/sip-router/y.tab.c:4478: first defined here
^^^^^^^
The file y.tab.c should not exist. It is not created by the sip-router compilation process (it creates cfg.tab.c instead). I have this reported before, but in the other case there was some mixup will old source files.
I found the bug - it happens if you start "make deb" a second time
1. make deb -> this will fail with an error 2. make deb -> this will generate y.tab.c as yacc (bison is yacc modues) instead of bison is used
regards klaus
On Jul 08, 2009 at 13:35, Klaus Darilion klaus.mailinglists@pernau.at wrote:
Andrei Pelinescu-Onciul schrieb:
Now I'm getting the following error:
[...]
cfg.tab.o: In function `yyparse': /usr/src/sip-router/cfg.tab.c:4478: multiple definition of `yyparse' y.tab.o:/usr/src/sip-router/y.tab.c:4478: first defined here
^^^^^^^
The file y.tab.c should not exist. It is not created by the sip-router compilation process (it creates cfg.tab.c instead). I have this reported before, but in the other case there was some mixup will old source files.
I found the bug - it happens if you start "make deb" a second time
- make deb
-> this will fail with an error 2. make deb -> this will generate y.tab.c as yacc (bison is yacc modues) instead of bison is used
Thanks, it should be fixed. make deb works now! (although some work regarding what exactly is packaged is still needed)
Andrei