Bruno Haible wrote: > Hi Jim, > >> > Upon inspection, it appears that the fu_cv_sys_stat_statfs2_bsize >> > test in m4/fsusage.m4 (labelled "AIX") would succeed on >> > glibc, MacOS X, FreeBSD, NetBSD, OpenBSD, AIX, HP-UX, Solaris, Cygwin >> > ... >> > without a clear overview of which platform ends up using which code, >> > I can't assert this. >> ... >> If necessary, we add additional guards later. > > Sorry, slight misunderstanding: I did not want to object against Miklos' > patch, but rather ask for better comments. Currently, > - the code labelled "POSIX" is not used for glibc/Linux, > - the code labelled "AIX" is not used on AIX 4 and 5, > - the code laballed "4.4BSD and NetBSD" is not used on NetBSD 3. > > Here's a proposed patch that updates the comments to match currently available > platforms, in a way that indicates why the order of the tests in fsusage.m4 > matters. > > Also it changes the message > "checking for two-argument statfs with statfs.bsize" > to > "checking for two-argument statfs with statfs.f_bsize" > > and > "checking for two-argument statfs with statfs.fsize" > to > "checking for two-argument statfs with statfs.f_fsize" > > since that's what the tests actually check for. > > OK to commit? > > > 2010-07-04 Bruno Haible <br...@clisp.org> > > fsusage: Clarify which code applies to which platforms. > * m4/fsusage.m4 (gl_FSUSAGE): Clarify which test succeeds on which > platform. > * lib/fsusage.c (get_fs_usage): Likewise.
Yes, that looks fine. Thank you.
