On Tue, 2007-02-27 at 04:17 +0530, Debarshi 'Rishi' Ray wrote: > > Linux-specific _IOR() macro. See include/parted/fdasd.h > > Do I have the green signal to remove fdasd support from Hurd for the time > being, as David suggested? I have checked that this particular problem is > solved by doing so?
Conditionalize the DASD support code (fdasd*, vtoc*) so it's Linux only. There will likely never be a need for DASD support on Hurd. > >> The problem in 1.7.1 is that in libparted/arch/gnu.c function gnu_write has > >> PED_SECTOR_SIZE in line number 657. This should be PED_SECTOR_SIZE_DEFAULT, > >> which is #define'd in include/parted/unit.h. There is no #define for > >> PED_SECTOR_SIZE. This is issue persists in the Git snapshot too. > > > Yeah, this should change to PED_SECTOR_SIZE_DEFAULT > > Done. Pushed to 'master' branch. I also fixed a small problem in the signature > of the function gnu_write in the same file. The firest parameter is now a > const PedDevice. > > A new problem has cropped up. In parted/ui.c, the function init_ui uses the > SA_SIGINFO constant to set the flags passed to sigaction. Although the entire > block is within a '#ifdef HAVE_SIGACTION....#endif', we are not checking > whether the SA_SIGINFO is defined in glibc or not. While building in Debian > GNU/Hurd 4.0 I found out that although the sigaction function is mentioned in > /usr/include/bits/sigaction.h, SA_SIGINFO is not defined in the library. This > seems to be a known issue in the Hurd kernel, and is confirmed by #hurd. > According to Guillem Jover (braindmg) we should check for the availability of > SA_SIGINFO at build-time, since it is a POSIX XSI extension. > > What do I do? Do I do as suggested and post the patch for review before > pushing? Yes. -- David Cantrell <[EMAIL PROTECTED]> Red Hat / Westford, MA
signature.asc
Description: This is a digitally signed message part
_______________________________________________ parted-devel mailing list [email protected] http://lists.alioth.debian.org/mailman/listinfo/parted-devel
