All of lore.kernel.org
 help / color / mirror / Atom feed
* [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
@ 2020-03-16  2:58 Zou Wei
  2020-03-16  4:31 ` Yang Xu
  2020-03-17  7:14 ` [LTP] " Xiao Yang
  0 siblings, 2 replies; 10+ messages in thread
From: Zou Wei @ 2020-03-16  2:58 UTC (permalink / raw)
  To: ltp

pthread_cond_init/s-c.c testcase lose a #ifndef line, fix it

Reported-by: Zou Wei <zou_wei@huawei.com>
Signed-off-by: Zou Wei <zou_wei@huawei.com>
---
 testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
index 6a5c70a..8d18484 100644
--- a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
+++ b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
@@ -80,6 +80,7 @@
 /***********************************    Test case   *****************************************/
 /********************************************************************************************/
+#ifndef WITHOUT_XOPEN
 typedef struct _teststruct {
 	pthread_cond_t cnd[10 * SCALABILITY_FACTOR];
 	pthread_condattr_t ca[4];
--
2.6.2


^ permalink raw reply related	[flat|nested] 10+ messages in thread

* [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
  2020-03-16  2:58 [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line Zou Wei
@ 2020-03-16  4:31 ` Yang Xu
  2020-03-16  7:53   ` Yang Xu
  2020-03-17  2:08   ` Yang Xu
  2020-03-17  7:14 ` [LTP] " Xiao Yang
  1 sibling, 2 replies; 10+ messages in thread
From: Yang Xu @ 2020-03-16  4:31 UTC (permalink / raw)
  To: ltp

Hi Zou

It was introduced by commit 310c9eb6e6("Remove explicit #define's for 
`_POSIX_C_SOURCE` and `_XOPEN_SOURCE`").

IMO, we can remove this WITHOUT_XOPEN macro directly because we have 
setted -D_POSIX_C_SOURCE=200809L -D_XOPEN_SOURCE=700 in 
ltp/testcases/open_posix_testsuite/CFLAGS.

Best Regards
Yang Xu
> pthread_cond_init/s-c.c testcase lose a #ifndef line, fix it
> 
> Reported-by: Zou Wei <zou_wei@huawei.com>
> Signed-off-by: Zou Wei <zou_wei@huawei.com>
> ---
>   testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c | 1 +
>   1 file changed, 1 insertion(+)
> 
> diff --git a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
> index 6a5c70a..8d18484 100644
> --- a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
> +++ b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
> @@ -80,6 +80,7 @@
>   /***********************************    Test case   *****************************************/
>   /********************************************************************************************/
> +#ifndef WITHOUT_XOPEN
>   typedef struct _teststruct {
>   	pthread_cond_t cnd[10 * SCALABILITY_FACTOR];
>   	pthread_condattr_t ca[4];
> --
> 2.6.2
> 
> 



^ permalink raw reply	[flat|nested] 10+ messages in thread

* [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
  2020-03-16  4:31 ` Yang Xu
@ 2020-03-16  7:53   ` Yang Xu
  2020-03-17  1:13     ` Enji Cooper
  2020-03-17  2:08   ` Yang Xu
  1 sibling, 1 reply; 10+ messages in thread
From: Yang Xu @ 2020-03-16  7:53 UTC (permalink / raw)
  To: ltp

Hi Cyril

On my environment, when I test openposix, this case missed run and 
complie because locate-test has filter rules called by 
open_posix_testsuite/scripts/generate-makefiles.sh.  filter rules as below:

runnable)
         find "$WHERE/conformance" "$WHERE/stress" -type f -name 
'*[0-9].c' -o -name '[0-9]*-[0-9]*.sh' | grep -v buildonly | grep -v 
'^./tools'
         find "$WHERE/functional" -type f -name '*.c'

I have a question why we leave it here if they are no meaningful. If
it was meaningful, why we filter this case? I see the history, but found 
no thing.

@Enji, you often change openposix code, can you give me some advise?

Best Regards
Yang Xu

> Hi Zou
> 
> It was introduced by commit 310c9eb6e6("Remove explicit #define's for 
> `_POSIX_C_SOURCE` and `_XOPEN_SOURCE`").
> 
> IMO, we can remove this WITHOUT_XOPEN macro directly because we have 
> setted -D_POSIX_C_SOURCE=200809L -D_XOPEN_SOURCE=700 in 
> ltp/testcases/open_posix_testsuite/CFLAGS.
> 
> Best Regards
> Yang Xu
>> pthread_cond_init/s-c.c testcase lose a #ifndef line, fix it
>>
>> Reported-by: Zou Wei <zou_wei@huawei.com>
>> Signed-off-by: Zou Wei <zou_wei@huawei.com>
>> ---
>>   
>> testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c 
>> | 1 +
>> ? 1 file changed, 1 insertion(+)
>>
>> diff --git 
>> a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c 
>>
>> index 6a5c70a..8d18484 100644
>> --- 
>> a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>> +++ 
>> b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>> @@ -80,6 +80,7 @@
>> ? /***********************************??? Test case   
>> *****************************************/
>>   
>> /********************************************************************************************/ 
>>
>> +#ifndef WITHOUT_XOPEN
>> ? typedef struct _teststruct {
>> ????? pthread_cond_t cnd[10 * SCALABILITY_FACTOR];
>> ????? pthread_condattr_t ca[4];
>> -- 
>> 2.6.2
>>
>>
> 
> 
> 



^ permalink raw reply	[flat|nested] 10+ messages in thread

* [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
  2020-03-16  7:53   ` Yang Xu
@ 2020-03-17  1:13     ` Enji Cooper
  2020-03-17  3:38       ` Yang Xu
  0 siblings, 1 reply; 10+ messages in thread
From: Enji Cooper @ 2020-03-17  1:13 UTC (permalink / raw)
  To: ltp

Hi Yang,
	I?ll have to look back, but this might have been the original set of items which never compiled/built cleanly.
	I think it?s time to go back to a more sane gmake-based build framework. My goal in making things work with POSIX make was noble for the time but it seems incredibly unnecessary now and I think it?s overcomplicating things.
Thanks!
-Enji

> On Mar 16, 2020, at 12:53 AM, Yang Xu <xuyang2018.jy@cn.fujitsu.com> wrote:
> 
> Hi Cyril
> 
> On my environment, when I test openposix, this case missed run and complie because locate-test has filter rules called by open_posix_testsuite/scripts/generate-makefiles.sh.  filter rules as below:
> 
> runnable)
>        find "$WHERE/conformance" "$WHERE/stress" -type f -name '*[0-9].c' -o -name '[0-9]*-[0-9]*.sh' | grep -v buildonly | grep -v '^./tools'
>        find "$WHERE/functional" -type f -name '*.c'
> 
> I have a question why we leave it here if they are no meaningful. If
> it was meaningful, why we filter this case? I see the history, but found no thing.
> 
> @Enji, you often change openposix code, can you give me some advise?
> 
> Best Regards
> Yang Xu
> 
>> Hi Zou
>> It was introduced by commit 310c9eb6e6("Remove explicit #define's for `_POSIX_C_SOURCE` and `_XOPEN_SOURCE`").
>> IMO, we can remove this WITHOUT_XOPEN macro directly because we have setted -D_POSIX_C_SOURCE=200809L -D_XOPEN_SOURCE=700 in ltp/testcases/open_posix_testsuite/CFLAGS.
>> Best Regards
>> Yang Xu
>>> pthread_cond_init/s-c.c testcase lose a #ifndef line, fix it
>>> 
>>> Reported-by: Zou Wei <zou_wei@huawei.com>
>>> Signed-off-by: Zou Wei <zou_wei@huawei.com>
>>> ---
>>>  testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c | 1 +
>>>   1 file changed, 1 insertion(+)
>>> 
>>> diff --git a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c 
>>> index 6a5c70a..8d18484 100644
>>> --- a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>>> +++ b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>>> @@ -80,6 +80,7 @@
>>>   /***********************************    Test case   *****************************************/
>>>  /********************************************************************************************/ 
>>> +#ifndef WITHOUT_XOPEN
>>>   typedef struct _teststruct {
>>>       pthread_cond_t cnd[10 * SCALABILITY_FACTOR];
>>>       pthread_condattr_t ca[4];
>>> -- 
>>> 2.6.2
>>> 
>>> 
> 
> 


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
  2020-03-16  4:31 ` Yang Xu
  2020-03-16  7:53   ` Yang Xu
@ 2020-03-17  2:08   ` Yang Xu
  2020-03-17  3:06     ` [LTP] 答复: " Zouwei
  1 sibling, 1 reply; 10+ messages in thread
From: Yang Xu @ 2020-03-17  2:08 UTC (permalink / raw)
  To: ltp

Hi Zou

Even this case does't compile or run on ltp default envrionment, I think 
it is a framework problem about ltp openposix. This fix is obviously 
correct and simple.

Reviewed-by: Yang Xu <xuyang2018.jy@cn.fujitsu.com>

Someone who merging this patch can add fixes tag.

  Fixes: 310c9eb6e ("Remove explicit #define's for `_POSIX_C_SOURCE` and 
`_XOPEN_SOURCE`")

Best Regards
Yang Xu

> Hi Zou
> 
> It was introduced by commit 310c9eb6e6("Remove explicit #define's for 
> `_POSIX_C_SOURCE` and `_XOPEN_SOURCE`").
> 
> IMO, we can remove this WITHOUT_XOPEN macro directly because we have 
> setted -D_POSIX_C_SOURCE=200809L -D_XOPEN_SOURCE=700 in 
> ltp/testcases/open_posix_testsuite/CFLAGS.
> 
> Best Regards
> Yang Xu
>> pthread_cond_init/s-c.c testcase lose a #ifndef line, fix it
>>
>> Reported-by: Zou Wei <zou_wei@huawei.com>
>> Signed-off-by: Zou Wei <zou_wei@huawei.com>
>> ---
>>   
>> testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c 
>> | 1 +
>> ? 1 file changed, 1 insertion(+)
>>
>> diff --git 
>> a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c 
>>
>> index 6a5c70a..8d18484 100644
>> --- 
>> a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>> +++ 
>> b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>> @@ -80,6 +80,7 @@
>> ? /***********************************??? Test case   
>> *****************************************/
>>   
>> /********************************************************************************************/ 
>>
>> +#ifndef WITHOUT_XOPEN
>> ? typedef struct _teststruct {
>> ????? pthread_cond_t cnd[10 * SCALABILITY_FACTOR];
>> ????? pthread_condattr_t ca[4];
>> -- 
>> 2.6.2
>>
>>
> 
> 
> 



^ permalink raw reply	[flat|nested] 10+ messages in thread

* [LTP] 答复:  [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
  2020-03-17  2:08   ` Yang Xu
@ 2020-03-17  3:06     ` Zouwei
  2020-03-17  5:52       ` Xiao Yang
  0 siblings, 1 reply; 10+ messages in thread
From: Zouwei @ 2020-03-17  3:06 UTC (permalink / raw)
  To: ltp

Hi Xu

Thanks for your answers and reviews. 

When I used the open-posix testsuite, I modified multiple files before compiling successfully. 

It is hoped that in future LTP release versions, open-posix can be automatically enabled and can be successfully compiled without modification.

BTW, The files I modified are as follows:
ltp/testcases/Makefile
@@ -30,9 +30,9 @@ include $(top_srcdir)/include/mk/env_pre.mk
 #    run).
 FILTER_OUT_DIRS                := kdump

-ifneq ($(WITH_OPEN_POSIX_TESTSUITE),yes)
-FILTER_OUT_DIRS                += open_posix_testsuite
-endif

 ifneq ($(WITH_REALTIME_TESTSUITE),yes)
 FILTER_OUT_DIRS                += realtime

ltp/testcases/open_posix_testsuite/LDFLAGS
@@ -0,0 +1 @@
+-lpthread -lrt -lm

ltp/testcases/open_posix_testsuite/functional/Makefile
@@ -5,7 +5,7 @@
 #
 
 all clean install test:
-	@for dir in `ls -d */Makefile 2>/dev/null | sed -e 's,/Makefile$$,,g'`; do \
+	@for dir in `find . -name *Makefile | egrep -v '^./Makefile' | sed -e 's,/Makefile$$,,g'`; do \
 		$(MAKE) -C $$dir $@;						   \
 	done
 
ltp/testcases/open_posix_testsuite/scripts/locate-test
@@ -92,7 +92,8 @@ runnable)
 	# waltz down the tools directory and try and build t0 (which doesn't
 	# make sense as it's a tool, not a test). Better criterion needs to
 	# be established for this file.
-	find "$WHERE/conformance" "$WHERE/stress" -type f -name '*[0-9].c' -o -name '[0-9]*-[0-9]*.sh' | grep -v buildonly | grep -v '^./tools'
+	find "$WHERE/conformance" "$WHERE/stress" -type f -name '*[0-9a-z].c' -o -name '[0-9]*-[0-9]*.sh' | grep -v buildonly | grep -v '^./tools' | grep -v 'testfrmw.c$'
 	find "$WHERE/functional" -type f -name '*.c'
 	;;
 test-tools)

ltp/testcases/open_posix_testsuite/stress/Makefile
@@ -5,7 +5,7 @@
 #
 
 all clean install test:
-	@for dir in `ls -d */Makefile 2>/dev/null | sed -e 's,/Makefile$$,,g'`; do \
+	@for dir in `find . -name *Makefile | egrep -v '^./Makefile' | sed -e 's,/Makefile$$,,g'`; do \
 		$(MAKE) -C $$dir $@;						  \
 	done

Best Regards
Zou Wei
-----????-----
???: Yang Xu [mailto:xuyang2018.jy@cn.fujitsu.com] 
????: 2020?3?17? 10:08
???: Zouwei (Samuel) <zou_wei@huawei.com>
??: ltp@lists.linux.it
??: Re: [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line

Hi Zou

Even this case does't compile or run on ltp default envrionment, I think it is a framework problem about ltp openposix. This fix is obviously correct and simple.

Reviewed-by: Yang Xu <xuyang2018.jy@cn.fujitsu.com>

Someone who merging this patch can add fixes tag.

  Fixes: 310c9eb6e ("Remove explicit #define's for `_POSIX_C_SOURCE` and
`_XOPEN_SOURCE`")

Best Regards
Yang Xu

> Hi Zou
> 
> It was introduced by commit 310c9eb6e6("Remove explicit #define's for 
> `_POSIX_C_SOURCE` and `_XOPEN_SOURCE`").
> 
> IMO, we can remove this WITHOUT_XOPEN macro directly because we have 
> setted -D_POSIX_C_SOURCE=200809L -D_XOPEN_SOURCE=700 in 
> ltp/testcases/open_posix_testsuite/CFLAGS.
> 
> Best Regards
> Yang Xu
>> pthread_cond_init/s-c.c testcase lose a #ifndef line, fix it
>>
>> Reported-by: Zou Wei <zou_wei@huawei.com>
>> Signed-off-by: Zou Wei <zou_wei@huawei.com>
>> ---
>>   
>> testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>> | 1 +
>> ? 1 file changed, 1 insertion(+)
>>
>> diff --git
>> a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c
>> .c 
>> b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c
>> .c
>>
>> index 6a5c70a..8d18484 100644
>> ---
>> a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c
>> .c
>> +++ 
>> b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c
>> .c
>> @@ -80,6 +80,7 @@
>> ? /***********************************??? Test case   
>> *****************************************/
>>   
>> /********************************************************************
>> ************************/
>>
>> +#ifndef WITHOUT_XOPEN
>> ? typedef struct _teststruct {
>> ????? pthread_cond_t cnd[10 * SCALABILITY_FACTOR];
>> ????? pthread_condattr_t ca[4];
>> --
>> 2.6.2
>>
>>
> 
> 
> 



^ permalink raw reply	[flat|nested] 10+ messages in thread

* [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
  2020-03-17  1:13     ` Enji Cooper
@ 2020-03-17  3:38       ` Yang Xu
  0 siblings, 0 replies; 10+ messages in thread
From: Yang Xu @ 2020-03-17  3:38 UTC (permalink / raw)
  To: ltp

Hi Enji


> Hi Yang,
> 	I?ll have to look back, but this might have been the original set of items which never compiled/built cleanly.
I guess commit bec80d891a ("Fix oversight with `test tools'") changed 
this, as below:

--- a/testcases/open_posix_testsuite/scripts/locate-test
+++ b/testcases/open_posix_testsuite/scripts/locate-test
...
-       find "$WHERE" -type f -name '*.c' -o -name '[0-9]*-[0-9]*.sh' | 
grep -v buildonly | grep -v '^./tools' | grep -v testfrmw | grep -v 
'_scenarii'
+       find "$WHERE" -type f -name '*[0-9].c' -o -name 
'[0-9]*-[0-9]*.sh' | grep -v buildonly | grep -v '^./tools'

> 	I think it?s time to go back to a more sane gmake-based build framework. My goal in making things work with POSIX make was noble for the time but it seems incredibly unnecessary now and I think it?s overcomplicating things.
Yes. We should refactor this with using gmake, at lease we should 
compile them automatically.

Best Regards
Yang Xu
> Thanks!
> -Enji
> 
>> On Mar 16, 2020, at 12:53 AM, Yang Xu <xuyang2018.jy@cn.fujitsu.com> wrote:
>>
>> Hi Cyril
>>
>> On my environment, when I test openposix, this case missed run and complie because locate-test has filter rules called by open_posix_testsuite/scripts/generate-makefiles.sh.  filter rules as below:
>>
>> runnable)
>>         find "$WHERE/conformance" "$WHERE/stress" -type f -name '*[0-9].c' -o -name '[0-9]*-[0-9]*.sh' | grep -v buildonly | grep -v '^./tools'
>>         find "$WHERE/functional" -type f -name '*.c'
>>
>> I have a question why we leave it here if they are no meaningful. If
>> it was meaningful, why we filter this case? I see the history, but found no thing.
>>
>> @Enji, you often change openposix code, can you give me some advise?
>>
>> Best Regards
>> Yang Xu
>>
>>> Hi Zou
>>> It was introduced by commit 310c9eb6e6("Remove explicit #define's for `_POSIX_C_SOURCE` and `_XOPEN_SOURCE`").
>>> IMO, we can remove this WITHOUT_XOPEN macro directly because we have setted -D_POSIX_C_SOURCE=200809L -D_XOPEN_SOURCE=700 in ltp/testcases/open_posix_testsuite/CFLAGS.
>>> Best Regards
>>> Yang Xu
>>>> pthread_cond_init/s-c.c testcase lose a #ifndef line, fix it
>>>>
>>>> Reported-by: Zou Wei <zou_wei@huawei.com>
>>>> Signed-off-by: Zou Wei <zou_wei@huawei.com>
>>>> ---
>>>>   testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c | 1 +
>>>>    1 file changed, 1 insertion(+)
>>>>
>>>> diff --git a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>>>> index 6a5c70a..8d18484 100644
>>>> --- a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>>>> +++ b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>>>> @@ -80,6 +80,7 @@
>>>>    /***********************************    Test case   *****************************************/
>>>>   /********************************************************************************************/
>>>> +#ifndef WITHOUT_XOPEN
>>>>    typedef struct _teststruct {
>>>>        pthread_cond_t cnd[10 * SCALABILITY_FACTOR];
>>>>        pthread_condattr_t ca[4];
>>>> -- 
>>>> 2.6.2
>>>>
>>>>
>>
>>
> 
> 
> 



^ permalink raw reply	[flat|nested] 10+ messages in thread

* [LTP]  答复:  [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
  2020-03-17  3:06     ` [LTP] 答复: " Zouwei
@ 2020-03-17  5:52       ` Xiao Yang
  0 siblings, 0 replies; 10+ messages in thread
From: Xiao Yang @ 2020-03-17  5:52 UTC (permalink / raw)
  To: ltp

On 2020/3/17 11:06, Zouwei (Samuel) wrote:
> Hi Xu
>
> Thanks for your answers and reviews.
>
> When I used the open-posix testsuite, I modified multiple files before compiling successfully.
>
> It is hoped that in future LTP release versions, open-posix can be automatically enabled and can be successfully compiled without modification.
Hi Zou,

You can send some patches to fix them directly.

Thanks,
Xiao Yang
>
> BTW, The files I modified are as follows:
> ltp/testcases/Makefile
> @@ -30,9 +30,9 @@ include $(top_srcdir)/include/mk/env_pre.mk
>   #    run).
>   FILTER_OUT_DIRS                := kdump
>
> -ifneq ($(WITH_OPEN_POSIX_TESTSUITE),yes)
> -FILTER_OUT_DIRS                += open_posix_testsuite
> -endif
>
>   ifneq ($(WITH_REALTIME_TESTSUITE),yes)
>   FILTER_OUT_DIRS                += realtime
>
> ltp/testcases/open_posix_testsuite/LDFLAGS
> @@ -0,0 +1 @@
> +-lpthread -lrt -lm
>
> ltp/testcases/open_posix_testsuite/functional/Makefile
> @@ -5,7 +5,7 @@
>   #
>
>   all clean install test:
> -	@for dir in `ls -d */Makefile 2>/dev/null | sed -e 's,/Makefile$$,,g'`; do \
> +	@for dir in `find . -name *Makefile | egrep -v '^./Makefile' | sed -e 's,/Makefile$$,,g'`; do \
>   		$(MAKE) -C $$dir $@;						   \
>   	done
>
> ltp/testcases/open_posix_testsuite/scripts/locate-test
> @@ -92,7 +92,8 @@ runnable)
>   	# waltz down the tools directory and try and build t0 (which doesn't
>   	# make sense as it's a tool, not a test). Better criterion needs to
>   	# be established for this file.
> -	find "$WHERE/conformance" "$WHERE/stress" -type f -name '*[0-9].c' -o -name '[0-9]*-[0-9]*.sh' | grep -v buildonly | grep -v '^./tools'
> +	find "$WHERE/conformance" "$WHERE/stress" -type f -name '*[0-9a-z].c' -o -name '[0-9]*-[0-9]*.sh' | grep -v buildonly | grep -v '^./tools' | grep -v 'testfrmw.c$'
>   	find "$WHERE/functional" -type f -name '*.c'
>   	;;
>   test-tools)
>
> ltp/testcases/open_posix_testsuite/stress/Makefile
> @@ -5,7 +5,7 @@
>   #
>
>   all clean install test:
> -	@for dir in `ls -d */Makefile 2>/dev/null | sed -e 's,/Makefile$$,,g'`; do \
> +	@for dir in `find . -name *Makefile | egrep -v '^./Makefile' | sed -e 's,/Makefile$$,,g'`; do \
>   		$(MAKE) -C $$dir $@;						  \
>   	done
>
> Best Regards
> Zou Wei
> -----????-----
> ???: Yang Xu [mailto:xuyang2018.jy@cn.fujitsu.com]
> ????: 2020?3?17? 10:08
> ???: Zouwei (Samuel)<zou_wei@huawei.com>
> ??: ltp@lists.linux.it
> ??: Re: [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
>
> Hi Zou
>
> Even this case does't compile or run on ltp default envrionment, I think it is a framework problem about ltp openposix. This fix is obviously correct and simple.
>
> Reviewed-by: Yang Xu<xuyang2018.jy@cn.fujitsu.com>
>
> Someone who merging this patch can add fixes tag.
>
>    Fixes: 310c9eb6e ("Remove explicit #define's for `_POSIX_C_SOURCE` and
> `_XOPEN_SOURCE`")
>
> Best Regards
> Yang Xu
>
>> Hi Zou
>>
>> It was introduced by commit 310c9eb6e6("Remove explicit #define's for
>> `_POSIX_C_SOURCE` and `_XOPEN_SOURCE`").
>>
>> IMO, we can remove this WITHOUT_XOPEN macro directly because we have
>> setted -D_POSIX_C_SOURCE=200809L -D_XOPEN_SOURCE=700 in
>> ltp/testcases/open_posix_testsuite/CFLAGS.
>>
>> Best Regards
>> Yang Xu
>>> pthread_cond_init/s-c.c testcase lose a #ifndef line, fix it
>>>
>>> Reported-by: Zou Wei<zou_wei@huawei.com>
>>> Signed-off-by: Zou Wei<zou_wei@huawei.com>
>>> ---
>>>
>>> testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
>>> | 1 +
>>>    1 file changed, 1 insertion(+)
>>>
>>> diff --git
>>> a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c
>>> .c
>>> b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c
>>> .c
>>>
>>> index 6a5c70a..8d18484 100644
>>> ---
>>> a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c
>>> .c
>>> +++
>>> b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c
>>> .c
>>> @@ -80,6 +80,7 @@
>>>    /***********************************    Test case
>>> *****************************************/
>>>
>>> /********************************************************************
>>> ************************/
>>>
>>> +#ifndef WITHOUT_XOPEN
>>>    typedef struct _teststruct {
>>>        pthread_cond_t cnd[10 * SCALABILITY_FACTOR];
>>>        pthread_condattr_t ca[4];
>>> --
>>> 2.6.2
>>>
>>>
>>
>>
>>
>
>
>




^ permalink raw reply	[flat|nested] 10+ messages in thread

* [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
  2020-03-16  2:58 [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line Zou Wei
  2020-03-16  4:31 ` Yang Xu
@ 2020-03-17  7:14 ` Xiao Yang
  2020-03-17  7:34   ` [LTP] 答复: " Zouwei
  1 sibling, 1 reply; 10+ messages in thread
From: Xiao Yang @ 2020-03-17  7:14 UTC (permalink / raw)
  To: ltp

Hi Zou,

Pushed.

BTW:
1) we also need to update open_posix_testsuite so that it can compile 
these stress tests by default.
2) Your patch seems to be broken so I recreate and apply it.

Thanks,
Xiao Yang
On 2020/3/16 10:58, Zou Wei wrote:
> pthread_cond_init/s-c.c testcase lose a #ifndef line, fix it
>
> Reported-by: Zou Wei<zou_wei@huawei.com>
> Signed-off-by: Zou Wei<zou_wei@huawei.com>
> ---
>   testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c | 1 +
>   1 file changed, 1 insertion(+)
>
> diff --git a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
> index 6a5c70a..8d18484 100644
> --- a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
> +++ b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c
> @@ -80,6 +80,7 @@
>   /***********************************    Test case   *****************************************/
>   /********************************************************************************************/
> +#ifndef WITHOUT_XOPEN
>   typedef struct _teststruct {
>   	pthread_cond_t cnd[10 * SCALABILITY_FACTOR];
>   	pthread_condattr_t ca[4];
> --
> 2.6.2
>
>




^ permalink raw reply	[flat|nested] 10+ messages in thread

* [LTP] 答复:  [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line
  2020-03-17  7:14 ` [LTP] " Xiao Yang
@ 2020-03-17  7:34   ` Zouwei
  0 siblings, 0 replies; 10+ messages in thread
From: Zouwei @ 2020-03-17  7:34 UTC (permalink / raw)
  To: ltp

Hi Xu

Thanks for merge the patch.

I sent another 2 patches, one is to enable the open_posix_testsuite in the makefile, and the other one is to fix the open_posix compilation issue.

Best Regards
Zou Wei
-----????-----
???: Xiao Yang [mailto:yangx.jy@cn.fujitsu.com] 
????: 2020?3?17? 15:15
???: Zouwei (Samuel) <zou_wei@huawei.com>
??: ltp@lists.linux.it
??: Re: [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line

Hi Zou,

Pushed.

BTW:
1) we also need to update open_posix_testsuite so that it can compile these stress tests by default.
2) Your patch seems to be broken so I recreate and apply it.

Thanks,
Xiao Yang
On 2020/3/16 10:58, Zou Wei wrote:
> pthread_cond_init/s-c.c testcase lose a #ifndef line, fix it
>
> Reported-by: Zou Wei<zou_wei@huawei.com>
> Signed-off-by: Zou Wei<zou_wei@huawei.com>
> ---
>   testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.c | 1 +
>   1 file changed, 1 insertion(+)
>
> diff --git 
> a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.
> c 
> b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.
> c
> index 6a5c70a..8d18484 100644
> --- 
> a/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/s-c.
> c
> +++ b/testcases/open_posix_testsuite/stress/threads/pthread_cond_init/
> +++ s-c.c
> @@ -80,6 +80,7 @@
>   /***********************************    Test case   *****************************************/
>   
> /*********************************************************************
> ***********************/
> +#ifndef WITHOUT_XOPEN
>   typedef struct _teststruct {
>   	pthread_cond_t cnd[10 * SCALABILITY_FACTOR];
>   	pthread_condattr_t ca[4];
> --
> 2.6.2
>
>




^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2020-03-17  7:34 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-03-16  2:58 [LTP] [PATCH] pthread_cond_init/s-c.c: Add the lose ifndef line Zou Wei
2020-03-16  4:31 ` Yang Xu
2020-03-16  7:53   ` Yang Xu
2020-03-17  1:13     ` Enji Cooper
2020-03-17  3:38       ` Yang Xu
2020-03-17  2:08   ` Yang Xu
2020-03-17  3:06     ` [LTP] 答复: " Zouwei
2020-03-17  5:52       ` Xiao Yang
2020-03-17  7:14 ` [LTP] " Xiao Yang
2020-03-17  7:34   ` [LTP] 答复: " Zouwei

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.