On Thu, Jul 28, 2005 at 11:43:41PM +1000, Stephen Rothwell wrote: > Hi all, > > Someone mentioned the mess in evdev.c that is caused by the fact that the > structures that are passed to/from user mode via read/write require > conversion when this API is used from 32 bit tasks on 64 bit kernels. > Some "discussion" followed during which I suggested an idea originally > from Matthew Wilcox of an arch-specific is_compat_task() function so that > these places could be identified. However it was considered better to > instead implement compat_sys_read/write.
This looks totally horrible, especially as we'd need readv/writev and pread/pwrite aswell. I don't think anyone but Andi actually liked this approach when discussed earlier. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
