Re: [LTP] Build error of "regress.c"

Re: [LTP] Build error of "regress.c"

[Caspar Zhang]

[-- Attachment #1: Type: text/plain, Size: 3663 bytes --]



----- "Mitani" <mitani@ryobi.co.jp> wrote:

> Hi,
> 
> I ran "make" with "ltp-2009-12-16" and "ltp-2009-12-18" cvs, 
> but it failed:


I met same problem. try this patch.

Thanks,
Caspar

> ------------
> [...]
> gcc -g -O2 -g -O2 -fno-strict-aliasing -pipe -Wall 
> -I../../../include
> -I../../../include
>   -L../../../lib
> -L/home/mitani/LTP/ltp-2009-12-18/testcases/kernel/syscalls/libevent/lib
>  libevent/test/regress.c  -levent -o libevent/test/regress
> libevent/test/regress.c:44:19: error: event.h: No such file or
> directory
> libevent/test/regress.c: In function 'simple_read_cb':
> libevent/test/regress.c:69: warning: implicit declaration of function
> 'event_add'
> libevent/test/regress.c: In function 'multiple_write_cb':
> libevent/test/regress.c:100: warning: implicit declaration of
> function
> 'event_del'
> libevent/test/regress.c: In function 'signal_cb':
> libevent/test/regress.c:159: warning: implicit declaration of
> function
> 'signal_del'
> libevent/test/regress.c: At top level:
> libevent/test/regress.c:164: error: field 'ev' has incomplete type
> [...]
> libevent/test/regress.c:455: warning: implicit declaration of
> function
> 'bufferevent_write'
> libevent/test/regress.c:459: warning: implicit declaration of
> function
> 'bufferevent_free'
> libevent/test/regress.c: In function 'main':
> libevent/test/regress.c:473: warning: implicit declaration of
> function
> 'event_init'
> make[3]: *** [libevent/test/regress] Error 1
> make[3]: Leaving directory
> `/home/mitani/LTP/ltp-2009-12-18/testcases/kernel/syscalls'
> make[2]: *** [all] Error 2
> [...]
> ------------
> 
> "regress.c" was not compiled before "ltp-2009-12-15" cvs.
> It seems to have been compiled after "ltp-2009-12-16" cvs 
> for the first time.
> 
> 
> The results of "make" are various by environment as follows:
> 
>   (1) RHEL5.4(2.6.18-164.el5) x86    : succeeded
>   (2) RHEL5.4(2.6.18-164.el5) x86_64 : failed
>   (3) RHEL5.4(2.6.18-164.el5) ia64   : failed
>   (4) RHEL4.8(2.6.9-89.ELsmp) x86    : succeeded
> 
> There is "/usr/local/include/event.h" file in the successful system.
> There is not "/usr/local/include/event.h" file in the failed system.
> 
> "/usr/local/include/event.h" file completely accords with 
> "${LTPROOT}/testcases/kernel/syscalls/libevent/event.h" file.
> I think that "${LTPROOT}/testcases/kernel/syscalls/libevent/event.h"
> copied to "/usr/local/include/event.h" when installed.
> 
> However, depending on system, there seems to be the case that 
> this file isn't copied.
> 
> 
> ---
> 
> Instead of "/usr/local/include/event.h", I installed "libevent-devel"
> (libevent-devel-1.1a-3.2.1.x86_64.rpm) to above (2) system.
> Then "/usr/include/event.h" file was installed.
> This "/usr/include/event.h" file does not completely accord with
> "/usr/local/include/event.h".
> But after that, "make" succeeded.
> 
> I don't think that this measure is right. This is the wrong way, I
> think.
> Because, I judge that LTP is going to copy a
> "/usr/local/include/event.h"
> file by oneself.
> 
> 
> 
> Thank you--
> 
> -Tomonori Mitani
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> This SF.Net email is sponsored by the Verizon Developer Community
> Take advantage of Verizon's best-in-class app development support
> A streamlined, 14 day to market process makes app distribution fast
> and easy
> Join now and get one step closer to millions of Verizon customers
> http://p.sf.net/sfu/verizon-dev2dev 
> _______________________________________________
> Ltp-list mailing list
> Ltp-list@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ltp-list


[-- Attachment #2: syscalls-libevent01.patch --]
[-- Type: text/x-patch, Size: 2820 bytes --]

diff --git a/runtest/syscalls b/runtest/syscalls
index 2acb906..755306a 100644
--- a/runtest/syscalls
+++ b/runtest/syscalls
@@ -465,7 +465,7 @@ lchown01_16 lchown01_16
 lchown02 prep_create_link; lchown02
 lchown02_16 prep_create_link; lchown02_16
 
-libevent01 export LTPROOT; run_libevent.sh
+libevent01 export LTPROOT; $LTPROOT/testcases/bin/libevent/run_libevent.sh
 
 link01 symlink01 -T link01
 link02 link02
diff --git a/testcases/kernel/syscalls/Makefile b/testcases/kernel/syscalls/Makefile
index 34ba9dc..695a0c3 100644
--- a/testcases/kernel/syscalls/Makefile
+++ b/testcases/kernel/syscalls/Makefile
@@ -29,36 +29,37 @@ FILTER_OUT_DIRS		:= libevent
 # pcllib is busted as all heck because of libtool
 FILTER_OUT_DIRS		+= pcllib
 
-# libevent doesn't compile right now... I have no bloody clue what breaks it
-# when doing make clean...
-ifdef DO_NO_COMPILE_ME
 ifeq ($(wildcard $(abs_srcdir)/libevent/test/Makefile),)
 $(warning libevent/test/Makefile does not exist; skipping to avoid compile errors)
 $(warning Please run "make autotools" and "configure" at the top-level LTP)
 $(warning directory if you want to run this test)
 else
 
-LIBEVENT		:= libevent/lib/libevent.a
-
-LIBEVENT_LINK		:= libevent/libevent.a
+LIBEVENT		:= libevent/libevent.a
 
 CLEAN_DEPS		:= libevent-clean
 
-CLEAN_TARGETS		:= $(LIBEVENT) $(LIBEVENT_LINK)
+CLEAN_TARGETS		:= $(LIBEVENT)
 
 INSTALL_DEPS		:= libevent-install
 
-INSTALL_TARGETS		:= libevent/run-libevent.sh libevent/test/test-libevent.sh
+INSTALL_TARGETS		:= libevent/run_libevent.sh libevent/test/test-libevent.sh
 
 LIBEVENT_APPS		:= regress test-eof test-init test-time test-weof
 
+TEST_APPS		:= $(addprefix libevent/test/,$(LIBEVENT_APPS))
+
 MAKE_DEPS		:= libevent-all
 
-MAKE_TARGETS		:= $(addprefix libevent/test/,$(LIBEVENT_APPS))
+MAKE_TARGETS		+= $(TEST_APPS)
 
 libevent libevent/test: %:
 	mkdir -p "$@"
 
+$(TEST_APPS): LDFLAGS += -L$(abs_builddir)/libevent
+$(TEST_APPS): LDLIBS += -levent
+$(TEST_APPS): CPPFLAGS += -I$(abs_builddir)/libevent			   
+
 $(LIBEVENT): $(abs_srcdir)/libevent/Makefile libevent
 	$(MAKE) -C libevent -f "$(firstword $^)" libevent.a
 
@@ -69,7 +70,6 @@ libevent-clean libevent-install: libevent-%: $(abs_srcdir)/libevent/test/Makefil
 	$(MAKE) -C libevent/test -f "$(firstword $^)" $*
 
 endif
-endif
 
 ifeq ($(UCLINUX),1)
 FILTER_OUT_DIRS	+= capget capset chmod chown clone fork getcontext llseek \
--- a/testcases/kernel/syscalls/libevent/run_libevent.sh	2009-11-20 03:08:36.000000000 +0800
+++ b/testcases/kernel/syscalls/libevent/run_libevent.sh	2009-12-15 16:55:12.750261436 +0800
@@ -31,7 +31,7 @@
 	tst_resm TCONF "You need to be root to run these tests"
 	TST_EXIT=0
 else
-	"$LTPROOT/testcases/bin/tests/test-libevent.sh"
+	"$LTPROOT/testcases/bin/libevent/test/test-libevent.sh"
 	TST_EXIT=$?
 fi
 tst_cleanup

[-- Attachment #3: Type: text/plain, Size: 390 bytes --]

------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev 

[-- Attachment #4: Type: text/plain, Size: 155 bytes --]

_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] Build error of "regress.c"

[Garrett Cooper]

On Thu, Dec 17, 2009 at 11:54 PM, Caspar Zhang <czhang@redhat.com> wrote:
>
>
> ----- "Mitani" <mitani@ryobi.co.jp> wrote:
>
>> Hi,
>>
>> I ran "make" with "ltp-2009-12-16" and "ltp-2009-12-18" cvs,
>> but it failed:
>
>
> I met same problem. try this patch.
>
> Thanks,
> Caspar
>
>> ------------
>> [...]
>> gcc -g -O2 -g -O2 -fno-strict-aliasing -pipe -Wall
>> -I../../../include
>> -I../../../include
>>   -L../../../lib
>> -L/home/mitani/LTP/ltp-2009-12-18/testcases/kernel/syscalls/libevent/lib
>>  libevent/test/regress.c  -levent -o libevent/test/regress
>> libevent/test/regress.c:44:19: error: event.h: No such file or
>> directory
>> libevent/test/regress.c: In function 'simple_read_cb':
>> libevent/test/regress.c:69: warning: implicit declaration of function
>> 'event_add'
>> libevent/test/regress.c: In function 'multiple_write_cb':
>> libevent/test/regress.c:100: warning: implicit declaration of
>> function
>> 'event_del'
>> libevent/test/regress.c: In function 'signal_cb':
>> libevent/test/regress.c:159: warning: implicit declaration of
>> function
>> 'signal_del'
>> libevent/test/regress.c: At top level:
>> libevent/test/regress.c:164: error: field 'ev' has incomplete type
>> [...]
>> libevent/test/regress.c:455: warning: implicit declaration of
>> function
>> 'bufferevent_write'
>> libevent/test/regress.c:459: warning: implicit declaration of
>> function
>> 'bufferevent_free'
>> libevent/test/regress.c: In function 'main':
>> libevent/test/regress.c:473: warning: implicit declaration of
>> function
>> 'event_init'
>> make[3]: *** [libevent/test/regress] Error 1
>> make[3]: Leaving directory
>> `/home/mitani/LTP/ltp-2009-12-18/testcases/kernel/syscalls'
>> make[2]: *** [all] Error 2
>> [...]
>> ------------
>>
>> "regress.c" was not compiled before "ltp-2009-12-15" cvs.
>> It seems to have been compiled after "ltp-2009-12-16" cvs
>> for the first time.
>>
>>
>> The results of "make" are various by environment as follows:
>>
>>   (1) RHEL5.4(2.6.18-164.el5) x86    : succeeded
>>   (2) RHEL5.4(2.6.18-164.el5) x86_64 : failed
>>   (3) RHEL5.4(2.6.18-164.el5) ia64   : failed
>>   (4) RHEL4.8(2.6.9-89.ELsmp) x86    : succeeded
>>
>> There is "/usr/local/include/event.h" file in the successful system.
>> There is not "/usr/local/include/event.h" file in the failed system.
>>
>> "/usr/local/include/event.h" file completely accords with
>> "${LTPROOT}/testcases/kernel/syscalls/libevent/event.h" file.
>> I think that "${LTPROOT}/testcases/kernel/syscalls/libevent/event.h"
>> copied to "/usr/local/include/event.h" when installed.
>>
>> However, depending on system, there seems to be the case that
>> this file isn't copied.
>>
>>
>> ---
>>
>> Instead of "/usr/local/include/event.h", I installed "libevent-devel"
>> (libevent-devel-1.1a-3.2.1.x86_64.rpm) to above (2) system.
>> Then "/usr/include/event.h" file was installed.
>> This "/usr/include/event.h" file does not completely accord with
>> "/usr/local/include/event.h".
>> But after that, "make" succeeded.
>>
>> I don't think that this measure is right. This is the wrong way, I
>> think.
>> Because, I judge that LTP is going to copy a
>> "/usr/local/include/event.h"
>> file by oneself.
>>
>>
>>
>> Thank you--
>>
>> -Tomonori Mitani

    Please try again off cvs Mitani-san.
Thanks,
-Garrett

------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev 
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] Build error of "regress.c"

[Mitani]

Caspar,
Garrett,

Thank you for useful information.

I arranged the patch to match the latest file, and applied it.

============
--- ./Makefile.org      2009-12-16 04:27:42.000000000 +0900
+++ ./Makefile  2009-12-18 17:47:42.000000000 +0900
@@ -35,7 +35,7 @@
 $(warning directory if you want to run this test)
 else

-LIBEVENT               := libevent/lib/libevent.a
+LIBEVENT               := libevent/libevent.a

 CLEAN_DEPS             := libevent-clean

@@ -56,8 +56,9 @@
 libevent libevent/test: %:
        mkdir -p "$@"

-$(TEST_APPS): LDFLAGS += -L$(abs_builddir)/libevent/lib
+$(TEST_APPS): LDFLAGS += -L$(abs_builddir)/libevent
 $(TEST_APPS): LDLIBS += -levent
+$(TEST_APPS): CPPFLAGS += -I$(abs_builddir)/libevent

 $(LIBEVENT): $(abs_srcdir)/libevent/Makefile libevent
        $(MAKE) -C libevent -f "$(firstword $^)" libevent.a
============

After that, "make" succeeded!


Thank you--

-Tomonori Mitani


-----Original Message-----
From: Garrett Cooper [mailto:yanegomi@gmail.com] 
Sent: Friday, December 18, 2009 5:50 PM
To: Caspar Zhang
Cc: Mitani; ltp-list@lists.sourceforge.net
Subject: Re: [LTP] Build error of "regress.c"

On Thu, Dec 17, 2009 at 11:54 PM, Caspar Zhang <czhang@redhat.com> wrote:
>
>
> ----- "Mitani" <mitani@ryobi.co.jp> wrote:
>
>> Hi,
>>
>> I ran "make" with "ltp-2009-12-16" and "ltp-2009-12-18" cvs,
>> but it failed:
>
>
> I met same problem. try this patch.
>
> Thanks,
> Caspar
>
>> ------------
>> [...]
>> gcc -g -O2 -g -O2 -fno-strict-aliasing -pipe -Wall
>> -I../../../include
>> -I../../../include
>>   -L../../../lib
>> -L/home/mitani/LTP/ltp-2009-12-18/testcases/kernel/syscalls/libevent/lib
>>  libevent/test/regress.c  -levent -o libevent/test/regress
>> libevent/test/regress.c:44:19: error: event.h: No such file or
>> directory
>> libevent/test/regress.c: In function 'simple_read_cb':
>> libevent/test/regress.c:69: warning: implicit declaration of function
>> 'event_add'
>> libevent/test/regress.c: In function 'multiple_write_cb':
>> libevent/test/regress.c:100: warning: implicit declaration of
>> function
>> 'event_del'
>> libevent/test/regress.c: In function 'signal_cb':
>> libevent/test/regress.c:159: warning: implicit declaration of
>> function
>> 'signal_del'
>> libevent/test/regress.c: At top level:
>> libevent/test/regress.c:164: error: field 'ev' has incomplete type
>> [...]
>> libevent/test/regress.c:455: warning: implicit declaration of
>> function
>> 'bufferevent_write'
>> libevent/test/regress.c:459: warning: implicit declaration of
>> function
>> 'bufferevent_free'
>> libevent/test/regress.c: In function 'main':
>> libevent/test/regress.c:473: warning: implicit declaration of
>> function
>> 'event_init'
>> make[3]: *** [libevent/test/regress] Error 1
>> make[3]: Leaving directory
>> `/home/mitani/LTP/ltp-2009-12-18/testcases/kernel/syscalls'
>> make[2]: *** [all] Error 2
>> [...]
>> ------------
>>
>> "regress.c" was not compiled before "ltp-2009-12-15" cvs.
>> It seems to have been compiled after "ltp-2009-12-16" cvs
>> for the first time.
>>
>>
>> The results of "make" are various by environment as follows:
>>
>>   (1) RHEL5.4(2.6.18-164.el5) x86    : succeeded
>>   (2) RHEL5.4(2.6.18-164.el5) x86_64 : failed
>>   (3) RHEL5.4(2.6.18-164.el5) ia64   : failed
>>   (4) RHEL4.8(2.6.9-89.ELsmp) x86    : succeeded
>>
>> There is "/usr/local/include/event.h" file in the successful system.
>> There is not "/usr/local/include/event.h" file in the failed system.
>>
>> "/usr/local/include/event.h" file completely accords with
>> "${LTPROOT}/testcases/kernel/syscalls/libevent/event.h" file.
>> I think that "${LTPROOT}/testcases/kernel/syscalls/libevent/event.h"
>> copied to "/usr/local/include/event.h" when installed.
>>
>> However, depending on system, there seems to be the case that
>> this file isn't copied.
>>
>>
>> ---
>>
>> Instead of "/usr/local/include/event.h", I installed "libevent-devel"
>> (libevent-devel-1.1a-3.2.1.x86_64.rpm) to above (2) system.
>> Then "/usr/include/event.h" file was installed.
>> This "/usr/include/event.h" file does not completely accord with
>> "/usr/local/include/event.h".
>> But after that, "make" succeeded.
>>
>> I don't think that this measure is right. This is the wrong way, I
>> think.
>> Because, I judge that LTP is going to copy a
>> "/usr/local/include/event.h"
>> file by oneself.
>>
>>
>>
>> Thank you--
>>
>> -Tomonori Mitani

    Please try again off cvs Mitani-san.
Thanks,
-Garrett



------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev 
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

[LTP] Build error of "regress.c"

[Mitani]

Hi,

I ran "make" with "ltp-2009-12-16" and "ltp-2009-12-18" cvs, 
but it failed:
------------
[...]
gcc -g -O2 -g -O2 -fno-strict-aliasing -pipe -Wall  -I../../../include
-I../../../include
  -L../../../lib
-L/home/mitani/LTP/ltp-2009-12-18/testcases/kernel/syscalls/libevent/lib
 libevent/test/regress.c  -levent -o libevent/test/regress
libevent/test/regress.c:44:19: error: event.h: No such file or directory
libevent/test/regress.c: In function 'simple_read_cb':
libevent/test/regress.c:69: warning: implicit declaration of function
'event_add'
libevent/test/regress.c: In function 'multiple_write_cb':
libevent/test/regress.c:100: warning: implicit declaration of function
'event_del'
libevent/test/regress.c: In function 'signal_cb':
libevent/test/regress.c:159: warning: implicit declaration of function
'signal_del'
libevent/test/regress.c: At top level:
libevent/test/regress.c:164: error: field 'ev' has incomplete type
[...]
libevent/test/regress.c:455: warning: implicit declaration of function
'bufferevent_write'
libevent/test/regress.c:459: warning: implicit declaration of function
'bufferevent_free'
libevent/test/regress.c: In function 'main':
libevent/test/regress.c:473: warning: implicit declaration of function
'event_init'
make[3]: *** [libevent/test/regress] Error 1
make[3]: Leaving directory
`/home/mitani/LTP/ltp-2009-12-18/testcases/kernel/syscalls'
make[2]: *** [all] Error 2
[...]
------------

"regress.c" was not compiled before "ltp-2009-12-15" cvs.
It seems to have been compiled after "ltp-2009-12-16" cvs 
for the first time.


The results of "make" are various by environment as follows:

  (1) RHEL5.4(2.6.18-164.el5) x86    : succeeded
  (2) RHEL5.4(2.6.18-164.el5) x86_64 : failed
  (3) RHEL5.4(2.6.18-164.el5) ia64   : failed
  (4) RHEL4.8(2.6.9-89.ELsmp) x86    : succeeded

There is "/usr/local/include/event.h" file in the successful system.
There is not "/usr/local/include/event.h" file in the failed system.

"/usr/local/include/event.h" file completely accords with 
"${LTPROOT}/testcases/kernel/syscalls/libevent/event.h" file.
I think that "${LTPROOT}/testcases/kernel/syscalls/libevent/event.h"
copied to "/usr/local/include/event.h" when installed.

However, depending on system, there seems to be the case that 
this file isn't copied.


---

Instead of "/usr/local/include/event.h", I installed "libevent-devel"
(libevent-devel-1.1a-3.2.1.x86_64.rpm) to above (2) system.
Then "/usr/include/event.h" file was installed.
This "/usr/include/event.h" file does not completely accord with
"/usr/local/include/event.h".
But after that, "make" succeeded.

I don't think that this measure is right. This is the wrong way, I think.
Because, I judge that LTP is going to copy a "/usr/local/include/event.h"
file by oneself.



Thank you--

-Tomonori Mitani




------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev 
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list