[RFC/PATCH] Make powerpc64 use __thread for per-cpu variables
David S. Miller
davem at davemloft.net
Thu May 11 08:25:02 EST 2006
From: Paul Mackerras <paulus at samba.org>
Date: Thu, 11 May 2006 07:05:24 +1000
> No, Richard has a point, it's not the value that is the concern, it's
> the address, which gcc could assume is still valid after a barrier.
Oh right, and that's currently part of why we obfuscate the
address computation with the RELOC_HIDE() buisness.
Once we expose what's really going on with something like
__thread, gcc can now be "smart" about it.
More information about the Linuxppc-dev