From: Jeff B. <jb...@ra...> - 2006-09-11 11:23:06
|
On Sun, Sep 10, 2006 at 11:47:23PM -0400, Mike Frysinger wrote: > On Sunday 10 September 2006 21:49, Yi CDL Yang wrote: > > In my perspective, a glibc syscall is just a simple wrapper for a > > kernel syscall > but not all are ... some contain extra logic that can cause troubles > (ive had LTP catch such things before by running against uClibc) This is the type of thing I'm thinking of as a problem. We also theoretically now have newlib, klibc, dietlibc and who knows how many others to worry about if we test the actual end user interfaces for things rather than using kernel services directly where reasonable. Most of the C libraries include relevant testsuites, whereas the kernel has only ltp. Tks, Jeff Bailey -- I do not agree with a word you say, but I will defend to the death your right to say it. - Voltaire |