On 08/25/2011 11:34 AM, Caspar Zhang wrote:
> On 08/25/2011 11:25 AM, Wanlong Gao wrote:
>> As the patch of Caspar Zhang<[email protected]> said:
>> In new kernels,
>> this wouldn't happen, despite some ENOMEM/EAGAIN might occur during
>> testing. So the correct PASS/FAIL criteria is: the test passes when the
>> program finishes running and exits normally; the test fails when the
>> program exits abnormally, e.g. a SIGABRT received.
>>
>> Signed-off-by: Wanlong Gao<[email protected]>
>
> Hi Wanlong, thanks for your patch. One question, have you tested it on a
> buggy kernel on IA64 to see if the reproducer could hit the bug
> described in https://lkml.org/lkml/2009/10/2/85?
>
> Thanks,
> Caspar
No yet, I just made the patch from yours.
This bug seems no effect for this case now, right?
Thanks
-Wanlong Gao
>
>> ---
>> testcases/kernel/syscalls/mmap/mmap11.c | 6 ++----
>> 1 files changed, 2 insertions(+), 4 deletions(-)
>>
>> diff --git a/testcases/kernel/syscalls/mmap/mmap11.c
>> b/testcases/kernel/syscalls/mmap/mmap11.c
>> index 786724b..4523f47 100644
>> --- a/testcases/kernel/syscalls/mmap/mmap11.c
>> +++ b/testcases/kernel/syscalls/mmap/mmap11.c
>> @@ -112,13 +112,13 @@ void check(void)
>> for (lc = 0; TEST_LOOPING(lc); lc++) {
>> Tst_count = 0;
>> ret = pthread_create(&th,&attr, wait_thread, NULL);
>> - if (ret) {
>> + if (ret&& errno != ENOMEM&& errno != EAGAIN) {
>> tst_resm(TINFO, "[%d] ", count);
>> tst_brkm(TBROK|TERRNO, cleanup, "pthread_create");
>> }
>> count++;
>> ret = pthread_create(&thread[lc],&attr, wait_thread2, NULL);
>> - if (ret) {
>> + if (ret&& errno != ENOMEM&& errno != EAGAIN) {
>> tst_resm(TINFO, "[%d] ", count);
>> tst_brkm(TBROK|TERRNO, cleanup, "pthread_create");
>> }
>> @@ -133,8 +133,6 @@ void *wait_thread(void *args)
>> void *addr;
>>
>> addr = malloc(MAL_SIZE);
>> - if (addr)
>> - memset(addr, 1, MAL_SIZE);
>> sleep(1);
>> return NULL;
>> }
>
>
------------------------------------------------------------------------------
EMC VNX: the world's simplest storage, starting under $10K
The only unified storage solution that offers unified management
Up to 160% more powerful than alternatives and 25% more efficient.
Guaranteed. http://p.sf.net/sfu/emc-vnx-dev2dev
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
