From: Robert W. <ro...@us...> - 2003-08-22 06:06:10
|
>ltp-full-20030807/testcases/kernel/pty/hangup01.c > This test unnecessarily includes sys/stropts.h and uses > the STREAMS specific getmsg() in conjunction with poll(). > I think using read(), thereby eliminating stropts.h and the > use of getmsg(), is more appropriate here since you are > checking for events on a pty, not on a STREAM. This change > would also be appropriate since this is not a STREAMS > specific test. And it would also be nice, since uClibc does > not support STREAMS (i.e. "the XSI STREAMS Option Group"). I replaced the usage of getmsg() with read() and updated the CVS version. >ltp-full-20030807/testcases/kernel/pty/ptem01.c > unnecessarily includes sys/stropts.h. Since uClibc does not > support STREAMS (i.e. "the XSI STREAMS Option Group") this > causes this to needlessly not-compile. This header should > be removed. Done! >ltp-full-20030807/testcases/kernel/syscalls/fmtmsg > we do not (yet) support fmtmsg, so this test did not compile. > Why is this under kernel/syscalls? We group in glibc functions/calls under "kernel/syscalls". We should probably seperate the two types to make it a bit clearer. However, until I have enough time to do this....they'll stay in their current location. The call is defined in SuSv3. >ltp-full-20030807/testcases/kernel/syscalls/ioctl/ioctl02.c > unnecessarily includes stropts.h. Since uClibc does not > support STREAMS (i.e. "the XSI STREAMS Option Group") this > causes this to needlessly not-compile. This header should > be removed. Done! >ltp-full-20030807/testcases/kernel/syscalls/mallopt/mallopt01.c > The tests for mallinfo() and mallopt(), which are not specified > by SuSv3 (and are absent from uClibc) should probably be > dropped. Again, probably should be moved into a glibc area. However, I don't think the test should be removed, since they are still valid and run under glibc. >ltp-full-20030807/testcases/kernel/syscalls/nftw > nftw64 is not specified by SuSv3. uClibc currently does > not supply either nftw() or ftw() (which needs to be fixed). > Why is this under kernel/syscalls though? You have an awful > lot of tests under kernel/syscalls that do not appear to be > syscall tests... See above. Thanks for the comments. After all this I think I'm going to create a seperate /testcases/kernel directory for the glibc calls. Hopefully I can get it done in time for the September release....if not, then definitely by October. -Robbie Robert V. Williamson <ro...@us...> Linux Test Project IBM Linux Technology Center Web: http://ltp.sourceforge.net IRC: #ltp on freenode.irc.net ==================== "Only two things are infinite, the universe and human stupidity, and I'm not sure about the former." -Albert Einstein |