Re: LTP: list of failures on 32bit and compat mode

["Arnd Bergmann" <arnd@arndb.de>]

On Thu, Apr 6, 2023, at 12:56, Petr Vorel wrote:
>> On Thu, Apr 6, 2023, at 11:11, Naresh Kamboju wrote:
>
>> > mq_timedreceive01
>> > mq_timedreceive01.c:197: TPASS: mq_timedreceive() failed expectedly: EINTR (4)
>> > tst_test.c:1581: TBROK: Test killed by SIGSEGV!
>
>> I think this is the same problem as fstatfs02, where ltp passes
>> an invalid pointer and expects EFAULT, but musl touches the data
>> first in order to do the time64 conversion. Needs the same fix.
>
> FYI mq_timedreceive01 is broken on 32bit systems with glibc
> (in current LTP master):
>

> EINTR (4)
> tst_test.c:1618: TBROK: Test killed by SIGSEGV!

Right, I see this has the same time64 logic as musl now.

>> > recvmmsg01.c:92: TPASS: recvmmsg() overflow in nanoseconds in timeout
>> > : EINVAL (22)
>> > tst_test.c:1581: TBROK: Test killed by SIGSEGV!
>
>> Same time64 conversion issue as above.
>
> Besides the same problem with shm permissions on musl I see SIGSEGV also on
> 64bit musl on current LTP master.

Ah, I see. This must be the padding code then, not the time64
conversion:

+int recvmmsg(int fd, struct mmsghdr *msgvec, unsigned int vlen, unsigned int flags, struct timespec *timeout)
+{
+#if LONG_MAX > INT_MAX
+       struct mmsghdr *mh = msgvec;
+       unsigned int i;
+       for (i = vlen; i; i--, mh++)
+               mh->msg_hdr.__pad1 = mh->msg_hdr.__pad2 = 0;
+#endif


       Arnd