From: Al Viro <viro@xxxxxxxxxxxxxxxxxx>
Date: Sat, 6 Oct 2012 07:06:09 +0100
> - set_fs((mm_segment_t) { get_thread_current_ds() });
> + __asm__ __volatile__ ("wr %%g0, %0, %%asi" : :
> + "r" (current_thread_info()->current_ds ));
Please use set_fs() or create a __set_fs() if you're trying to avoid
the weird cast. This is also indented improperly, should be:
__asm__ __volatile__ ("wr %%g0, %0, %%asi" : :
"r" (current_thread_info()->current_ds ));
Also, this whole bit about rearranging thead flags is an enormous,
unnecessary, and hard to audit distraction from fixing the syscall
tracing bug.
I've decided that I don't want to keep reviewing this risky
rearrangement change, it's bitten us once already. We can do
this in sparc-next if you want next cycle.
Could you please, instead, just rearrange the order of operations in
ret_sys_call, as needed, to purely fix the strace bug?
Thanks.
--
To unsubscribe from this list: send the line "unsubscribe sparclinux" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
