[LTP] LTP Intermediate Release for JULY 2009 Build Failures

[LTP] LTP Intermediate Release for JULY 2009 Build Failures

[CAI Qian]

Hi,

I have seen those failures,

# uname -a
Linux foo.bar.redhat.com 2.6.18-158.el5 #1 SMP Mon Jul 13 16:23:20 EDT
2009 s390x s390x s390x GNU/Linux

make[5]: Entering directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls/eventfd2'
gcc -g -O2 -I../../include -g -Wall -I../../../../include -Wall    eventfd2_01.c  -L../../../../lib -lltp -o eventfd2_01
gcc -g -O2 -I../../include -g -Wall -I../../../../include -Wall    eventfd2_02.c  -L../../../../lib -lltp -o eventfd2_02
gcc -g -O2 -I../../include -g -Wall -I../../../../include -Wall    eventfd2_03.c  -L../../../../lib -lltp -o eventfd2_03
eventfd2_03.c: In function ‘xsem_wait’:
eventfd2_03.c:60: warning: format ‘%llu’ expects type ‘long long unsigned int’, but argument 5 has type ‘u_int64_t’
eventfd2_03.c: In function ‘main’:
eventfd2_03.c:129: warning: implicit declaration of function ‘waitpid’
eventfd2_03.c: At top level:
../../include/linux_syscall_numbers.h:19: warning: ‘cleanup’ used but never defined
/tmp/ccODR9Wd.o: In function `eventfd2':
/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls/eventfd2/eventfd2_03.c:49: undefined reference to `cleanup'
collect2: ld returned 1 exit status
make[5]: *** [eventfd2_03] Error 1
make[5]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls/eventfd2'
make[4]: *** [all] Error 2
make[4]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls'
make[3]: *** [all] Error 2
make[3]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases'
make[1]: *** [all] Error 2
make[1]: Leaving directory `/root/20090721/ltp-intermediate-20090721'
make: *** [build] Error 2

----------
# uname -a
Linux foo.bar.redhat.com 2.6.18-158.el5 #1 SMP Mon Jul 13
16:27:25 EDT 2009 ppc64 ppc64 ppc64 GNU/Linux

make[5]: Entering directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls/fstatat'
gcc -g -O2 -I../../include -g -Wall -I../../../../include -Wall -DTST_USE_NEWER64_SYSCALL=1 -D_FILE_OFFSET_BITS=64   -c -o fstatat01_64.o fstatat01.c
fstatat01.c: In function ‘main’:
fstatat01.c:126: error: expected ‘)’ before numeric constant
make[5]: *** [fstatat01_64.o] Error 1
make[5]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls/fstatat'
make[4]: *** [all] Error 2
make[4]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls'
make[3]: *** [all] Error 2
make[3]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases'
make[1]: *** [all] Error 2
make[1]: Leaving directory `/root/20090721/ltp-intermediate-20090721'
make: *** [build] Error 2

CAI Qian
------------------------------------------------------------------------------
Enter the BlackBerry Developer Challenge  
This is your chance to win up to $100,000 in prizes! For a limited time, 
vendors submitting new applications to BlackBerry App World(TM) will have
the opportunity to enter the BlackBerry Developer Challenge. See full prize  
details at: http://p.sf.net/sfu/Challenge
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] LTP Intermediate Release for JULY 2009 Build Failures

[Garrett Cooper]

On Tue, Jul 21, 2009 at 3:25 AM, CAI Qian<caiqian@cclom.cn> wrote:
> Hi,
>
> I have seen those failures,
>
> # uname -a
> Linux foo.bar.redhat.com 2.6.18-158.el5 #1 SMP Mon Jul 13 16:23:20 EDT
> 2009 s390x s390x s390x GNU/Linux
>
> make[5]: Entering directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls/eventfd2'
> gcc -g -O2 -I../../include -g -Wall -I../../../../include -Wall    eventfd2_01.c  -L../../../../lib -lltp -o eventfd2_01
> gcc -g -O2 -I../../include -g -Wall -I../../../../include -Wall    eventfd2_02.c  -L../../../../lib -lltp -o eventfd2_02
> gcc -g -O2 -I../../include -g -Wall -I../../../../include -Wall    eventfd2_03.c  -L../../../../lib -lltp -o eventfd2_03
> eventfd2_03.c: In function ‘xsem_wait’:
> eventfd2_03.c:60: warning: format ‘%llu’ expects type ‘long long unsigned int’, but argument 5 has type ‘u_int64_t’
> eventfd2_03.c: In function ‘main’:
> eventfd2_03.c:129: warning: implicit declaration of function ‘waitpid’
> eventfd2_03.c: At top level:
> ../../include/linux_syscall_numbers.h:19: warning: ‘cleanup’ used but never defined
> /tmp/ccODR9Wd.o: In function `eventfd2':
> /root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls/eventfd2/eventfd2_03.c:49: undefined reference to `cleanup'
> collect2: ld returned 1 exit status
> make[5]: *** [eventfd2_03] Error 1
> make[5]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls/eventfd2'
> make[4]: *** [all] Error 2
> make[4]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls'
> make[3]: *** [all] Error 2
> make[3]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel'
> make[2]: *** [all] Error 2
> make[2]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases'
> make[1]: *** [all] Error 2
> make[1]: Leaving directory `/root/20090721/ltp-intermediate-20090721'
> make: *** [build] Error 2

The patch provided below fixes the majority of the compile
errors/warnings. I'm unable to get it to run though by itself, so I
must be missing some critical piece of info.

I was tempted to fix the format string error, but I don't know what
the potential implications are of switching the data type from
u_int64_t to unsigned long long...

> # uname -a
> Linux foo.bar.redhat.com 2.6.18-158.el5 #1 SMP Mon Jul 13
> 16:27:25 EDT 2009 ppc64 ppc64 ppc64 GNU/Linux
>
> make[5]: Entering directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls/fstatat'
> gcc -g -O2 -I../../include -g -Wall -I../../../../include -Wall -DTST_USE_NEWER64_SYSCALL=1 -D_FILE_OFFSET_BITS=64   -c -o fstatat01_64.o fstatat01.c
> fstatat01.c: In function ‘main’:
> fstatat01.c:126: error: expected ‘)’ before numeric constant
> make[5]: *** [fstatat01_64.o] Error 1
> make[5]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls/fstatat'
> make[4]: *** [all] Error 2
> make[4]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel/syscalls'
> make[3]: *** [all] Error 2
> make[3]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases/kernel'
> make[2]: *** [all] Error 2
> make[2]: Leaving directory `/root/20090721/ltp-intermediate-20090721/testcases'
> make[1]: *** [all] Error 2
> make[1]: Leaving directory `/root/20090721/ltp-intermediate-20090721'
> make: *** [build] Error 2

Missing syscall number. Not sure why it fails for you and passes for
me though... the syscall may need to be added to a .in file.

Thanks,
-Garrett

Fix compiler error noted by CAI Qian w.r.t. the intermediate July release, as...
1. The necessary headers weren't being pulled in for waitpid.
2. gcc was complaining about -Wunused with cleanup.

Signed-off-by: Garrett Cooper <yanegomi@gmail.com>

Index: testcases/kernel/syscalls/eventfd2/eventfd2_03.c
===================================================================
RCS file: /cvsroot/ltp/ltp/testcases/kernel/syscalls/eventfd2/eventfd2_03.c,v
retrieving revision 1.2
diff -u -r1.2 eventfd2_03.c
--- testcases/kernel/syscalls/eventfd2/eventfd2_03.c    6 Jul 2009
14:49:18 -0000       1.2
+++ testcases/kernel/syscalls/eventfd2/eventfd2_03.c    21 Jul 2009
14:15:59 -0000
@@ -26,6 +26,7 @@
 #include <sys/types.h>
 #include <sys/syscall.h>
 #include <sys/stat.h>
+#include <sys/wait.h>
 #include <fcntl.h>
 #include <stdlib.h>
 #include <stdio.h>
@@ -44,6 +45,8 @@
 #define EFD_SEMLIKE (1 << 0)
 #endif

+/* Dummy function reference to avoid compiler warning from
linux_syscall_numbers.h. */
+extern void cleanup() { }

 static int eventfd2(int count, int flags) {
        return syscall(__NR_eventfd2, count, flags);

------------------------------------------------------------------------------
Enter the BlackBerry Developer Challenge  
This is your chance to win up to $100,000 in prizes! For a limited time, 
vendors submitting new applications to BlackBerry App World(TM) will have
the opportunity to enter the BlackBerry Developer Challenge. See full prize  
details at: http://p.sf.net/sfu/Challenge
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] LTP Intermediate Release for JULY 2009 Build Failures

[Mike Frysinger]

[-- Attachment #1.1: Type: text/plain, Size: 617 bytes --]

On Tuesday 21 July 2009 10:21:44 Garrett Cooper wrote:
> I was tempted to fix the format string error, but I don't know what
> the potential implications are of switching the data type from
> u_int64_t to unsigned long long...

dont do that.  just use the proper format string from inttypes.h.
uint64_t f;
printf("f = %"PRIu64"\n", f);

> +/* Dummy function reference to avoid compiler warning from
> linux_syscall_numbers.h. */
> +extern void cleanup() { }

this is incorrect for a few reasons.  it should read:
+/* Dummy function as syscall() from linux_syscall_numbers.h uses cleanup() */
void cleanup() { }
-mike

[-- Attachment #1.2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 79 bytes --]

------------------------------------------------------------------------------

[-- Attachment #3: 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] LTP Intermediate Release for JULY 2009 Build Failures

[Garrett Cooper]

On Tue, Jul 21, 2009 at 2:28 PM, Mike Frysinger<vapier@gentoo.org> wrote:
> On Tuesday 21 July 2009 10:21:44 Garrett Cooper wrote:
>> I was tempted to fix the format string error, but I don't know what
>> the potential implications are of switching the data type from
>> u_int64_t to unsigned long long...
>
> dont do that.  just use the proper format string from inttypes.h.
> uint64_t f;
> printf("f = %"PRIu64"\n", f);
>
>> +/* Dummy function reference to avoid compiler warning from
>> linux_syscall_numbers.h. */
>> +extern void cleanup() { }
>
> this is incorrect for a few reasons.  it should read:
> +/* Dummy function as syscall() from linux_syscall_numbers.h uses cleanup() */
> void cleanup() { }
> -mike

    Ok. I'll fix it again, recompile, and test.
    Thanks for the review Mike :) (even though it's a bit post-mortem
now). I do lack a bit when it comes to C coding style and that's one
area that I want to improve upon.
-Garrett

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

Re: [LTP] LTP Intermediate Release for JULY 2009 Build Failures

[Garrett Cooper]

On Tue, Jul 21, 2009 at 9:38 PM, Garrett Cooper<yanegomi@gmail.com> wrote:
> On Tue, Jul 21, 2009 at 2:28 PM, Mike Frysinger<vapier@gentoo.org> wrote:
>> On Tuesday 21 July 2009 10:21:44 Garrett Cooper wrote:
>>> I was tempted to fix the format string error, but I don't know what
>>> the potential implications are of switching the data type from
>>> u_int64_t to unsigned long long...
>>
>> dont do that.  just use the proper format string from inttypes.h.
>> uint64_t f;
>> printf("f = %"PRIu64"\n", f);
>>
>>> +/* Dummy function reference to avoid compiler warning from
>>> linux_syscall_numbers.h. */
>>> +extern void cleanup() { }
>>
>> this is incorrect for a few reasons.  it should read:
>> +/* Dummy function as syscall() from linux_syscall_numbers.h uses cleanup() */
>> void cleanup() { }
>> -mike
>
>    Ok. I'll fix it again, recompile, and test.
>    Thanks for the review Mike :) (even though it's a bit post-mortem
> now). I do lack a bit when it comes to C coding style and that's one
> area that I want to improve upon.

Ok, new patch given Mike's review suggestion...

Fix compiler error for
testcases/kernel/syscalls/eventfd2/eventfd2_03.c noted by CAI Qian
w.r.t. the intermediate July release, as...
1. The necessary headers weren't being pulled in for waitpid.
2. gcc was complaining about -Wunused with cleanup.
3. The wrong format argument was being passed to printf.

Fix for 3 and suggestions on coding style made by Mike Frysinger.

Signed-off-by: Garrett Cooper <yanegomi@gmail.com>

gcooper@orangebox /scratch/ltp-dev2/ltp $ cdiff
testcases/kernel/syscalls/eventfd2/eventfd2_03.c
Index: testcases/kernel/syscalls/eventfd2/eventfd2_03.c
===================================================================
RCS file: /cvsroot/ltp/ltp/testcases/kernel/syscalls/eventfd2/eventfd2_03.c,v
retrieving revision 1.2
diff -u -r1.2 eventfd2_03.c
--- testcases/kernel/syscalls/eventfd2/eventfd2_03.c    6 Jul 2009
14:49:18 -0000       1.2
+++ testcases/kernel/syscalls/eventfd2/eventfd2_03.c    22 Jul 2009
05:00:37 -0000
@@ -26,6 +26,7 @@
 #include <sys/types.h>
 #include <sys/syscall.h>
 #include <sys/stat.h>
+#include <sys/wait.h>
 #include <fcntl.h>
 #include <stdlib.h>
 #include <stdio.h>
@@ -44,6 +45,8 @@
 #define EFD_SEMLIKE (1 << 0)
 #endif

+/* Dummy function as syscall from linux_syscall_numbers.h uses cleanup(). */
+void cleanup() { }

 static int eventfd2(int count, int flags) {
        return syscall(__NR_eventfd2, count, flags);
@@ -56,7 +59,7 @@
                perror("reading eventfd");
                exit(1);
        }
-       fprintf(stdout, "[%u] wait completed on %d: count=%llu\n",
+       fprintf(stdout, "[%u] wait completed on %d: count=%lu64\n",
                getpid(), fd, cntr);
 }

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

Re: [LTP] LTP Intermediate Release for JULY 2009 Build Failures

[Mike Frysinger]

[-- Attachment #1.1: Type: text/plain, Size: 938 bytes --]

On Wednesday 22 July 2009 01:01:34 Garrett Cooper wrote:
> On Tue, Jul 21, 2009 at 9:38 PM, Garrett Cooper wrote:
> > On Tue, Jul 21, 2009 at 2:28 PM, Mike Frysinger wrote:
> >> On Tuesday 21 July 2009 10:21:44 Garrett Cooper wrote:
> >>> I was tempted to fix the format string error, but I don't know what
> >>> the potential implications are of switching the data type from
> >>> u_int64_t to unsigned long long...
> >>
> >> dont do that.  just use the proper format string from inttypes.h.
> >> uint64_t f;
> >> printf("f = %"PRIu64"\n", f);
> >
>
> -       fprintf(stdout, "[%u] wait completed on %d: count=%llu\n",
> +       fprintf(stdout, "[%u] wait completed on %d: count=%lu64\n",
>                 getpid(), fd, cntr);

umm, you sure that's right ?  seems to me it'll output the number followed by 
the string "64".  i think you need to look at my example again (and take 
notice of the quote marks).
-mike

[-- Attachment #1.2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 79 bytes --]

------------------------------------------------------------------------------

[-- Attachment #3: 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] LTP Intermediate Release for JULY 2009 Build Failures

[Garrett Cooper]

On Jul 21, 2009, at 10:22 PM, Mike Frysinger <vapier@gentoo.org> wrote:

> On Wednesday 22 July 2009 01:01:34 Garrett Cooper wrote:
>> On Tue, Jul 21, 2009 at 9:38 PM, Garrett Cooper wrote:
>>> On Tue, Jul 21, 2009 at 2:28 PM, Mike Frysinger wrote:
>>>> On Tuesday 21 July 2009 10:21:44 Garrett Cooper wrote:
>>>>> I was tempted to fix the format string error, but I don't know  
>>>>> what
>>>>> the potential implications are of switching the data type from
>>>>> u_int64_t to unsigned long long...
>>>>
>>>> dont do that.  just use the proper format string from inttypes.h.
>>>> uint64_t f;
>>>> printf("f = %"PRIu64"\n", f);
>>>
>>
>> -       fprintf(stdout, "[%u] wait completed on %d: count=%llu\n",
>> +       fprintf(stdout, "[%u] wait completed on %d: count=%lu64\n",
>>                getpid(), fd, cntr);
>
> umm, you sure that's right ?  seems to me it'll output the number  
> followed by
> the string "64".  i think you need to look at my example again (and  
> take
> notice of the quote marks).

Your suggestion didn't compile and unfortunately the testcase appears  
to be broken. I'll retest when I get home.
Thanks!

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

Re: [LTP] LTP Intermediate Release for JULY 2009 Build Failures

[Mike Frysinger]

[-- Attachment #1.1: Type: text/plain, Size: 1444 bytes --]

On Wednesday 22 July 2009 02:52:25 Garrett Cooper wrote:
> On Jul 21, 2009, at 10:22 PM, Mike Frysinger wrote:
> > On Wednesday 22 July 2009 01:01:34 Garrett Cooper wrote:
> >> On Tue, Jul 21, 2009 at 9:38 PM, Garrett Cooper wrote:
> >>> On Tue, Jul 21, 2009 at 2:28 PM, Mike Frysinger wrote:
> >>>> On Tuesday 21 July 2009 10:21:44 Garrett Cooper wrote:
> >>>>> I was tempted to fix the format string error, but I don't know
> >>>>> what
> >>>>> the potential implications are of switching the data type from
> >>>>> u_int64_t to unsigned long long...
> >>>>
> >>>> dont do that.  just use the proper format string from inttypes.h.
> >>>> uint64_t f;
> >>>> printf("f = %"PRIu64"\n", f);
> >>
> >> -       fprintf(stdout, "[%u] wait completed on %d: count=%llu\n",
> >> +       fprintf(stdout, "[%u] wait completed on %d: count=%lu64\n",
> >>                getpid(), fd, cntr);
> >
> > umm, you sure that's right ?  seems to me it'll output the number
> > followed by
> > the string "64".  i think you need to look at my example again (and
> > take
> > notice of the quote marks).
>
> Your suggestion didn't compile and unfortunately the testcase appears
> to be broken. I'll retest when I get home.

did you include inttypes.h ?  i'm pretty sure your proposed change would (1) 
fix warnings for one arch size, (2) add warnings for the other arch size, and 
(3) add a spurious "64" string to the output.
-mike

[-- Attachment #1.2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 79 bytes --]

------------------------------------------------------------------------------

[-- Attachment #3: Type: text/plain, Size: 155 bytes --]

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