On Sun, 31 Mar 2013, Clemens Ladisch wrote: > Eldad Zack wrote: > > On Sun, 31 Mar 2013, Clemens Ladisch wrote: > >> cpu_to_le32() returns not u32 but __le32. > >> Please use that to make sparse happy. > > > > BTW, sparse (0.4.3) didn't complain on my box. > > I'm just calling make C=1. > > make C=1 CF=-D__CHECK_ENDIAN__ Heh, I felt quite silly about this question now that I read Documentation/sparse.txt to the end :) I see why you care so much about it too. I just found 3 places with endian bugs thanks to sparse, where usb_control_msg() is handed a LE instead of CPU native. snd_nativeinstruments_control_get()/put() and snd_usb_nativeinstruments_boot_quirk do this (I'll post a patch to fix this soon). Cheers, Eldad _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
