Re: [PSEUDOPATCH] rename is_compat_task

[Ingo Molnar <mingo@kernel.org>]

* Al Viro <viro@ZenIV.linux.org.uk> wrote:

> On Tue, Dec 08, 2015 at 05:36:49AM +0100, Ingo Molnar wrote:
> 
> > So are there any deep objections to doing this rename in a single, quick, 
> > pain-minimized fashion right at the end of the next merge window, when the 
> > amount of pending patches in various maintainer trees is at a cyclical 
> > minimum? We can also keep an is_compat_task() migratory define for one more 
> > cycle just in case.
> 
> Again, what about sparc?  There we have both 64bit and 32bit syscalls possible 
> to issue from the same process *and* no indication which trap had been used; how 
> do you implement is_compat_syscall() there?  There's a TIF_32BIT, which is used 
> by mmap() and friends, signal delivery, etc., but that's not a matter of which 
> syscall flavour had been issued.  Said that, arch/sparc doesn't use 
> is_compat_task(); it's open-coded everywhere...

Hm, so if Sparc has no notion of compat-ness of the system call then how does it 
implement runtime compat checks, such as AUDIT_ARCH et al?

Thanks,

	Ingo