> On Sep 19, 2020, at 3:41 PM, Al Viro <viro@xxxxxxxxxxxxxxxxxx> wrote: > > On Sat, Sep 19, 2020 at 03:23:54PM -0700, Andy Lutomirski wrote: >> >>>> On Sep 19, 2020, at 3:09 PM, Al Viro <viro@xxxxxxxxxxxxxxxxxx> wrote: >>> >>> On Fri, Sep 18, 2020 at 05:16:15PM +0200, Christoph Hellwig wrote: >>>>> On Fri, Sep 18, 2020 at 02:58:22PM +0100, Al Viro wrote: >>>>> Said that, why not provide a variant that would take an explicit >>>>> "is it compat" argument and use it there? And have the normal >>>>> one pass in_compat_syscall() to that... >>>> >>>> That would help to not introduce a regression with this series yes. >>>> But it wouldn't fix existing bugs when io_uring is used to access >>>> read or write methods that use in_compat_syscall(). One example that >>>> I recently ran into is drivers/scsi/sg.c. >>> >>> So screw such read/write methods - don't use them with io_uring. >>> That, BTW, is one of the reasons I'm sceptical about burying the >>> decisions deep into the callchain - we don't _want_ different >>> data layouts on read/write depending upon the 32bit vs. 64bit >>> caller, let alone the pointer-chasing garbage that is /dev/sg. >> >> Well, we could remove in_compat_syscall(), etc and instead have an implicit parameter in DEFINE_SYSCALL. Then everything would have to be explicit. This would probably be a win, although it could be quite a bit of work. > > It would not be a win - most of the syscalls don't give a damn > about 32bit vs. 64bit... Any reasonable implementation would optimize it out for syscalls that don’t care. Or it could be explicit: DEFINE_MULTIARCH_SYSCALL(...)
