On 2012-08-14 00:41, Chris Wilson wrote: > On Mon, 13 Aug 2012 22:41:10 -0700, Ben Widawsky <ben at bwidawsk.net> > wrote: >> This is possible with the new force paramter in do_switch. As stated >> in >> that patch, the goal is to get a real context stored at the time of >> initialization. >> >> References: https://bugs.freedesktop.org/show_bug.cgi?id=52429 >> Tested-by: Guang A Yang <guang.a.yang at intel.com> >> Signed-off-by: Ben Widawsky <ben at bwidawsk.net> > > I'm missing the rationalisation for this pair of patches... For > instance, I can't see how this closes the hole we have upon resume > where > ring->context_obj == DEFAULT_CONTEXT but CCID is 0. > -Chris Yeah this doesn't fix that problem. The problem this is trying to solve is suspend/resume before any context switch actually occurs. Basically jam the default context obj in, and this allows us to force restore it on resume. However, as you point out, I guess that force restore is missing. Let me think a bit more/chat on IRC. -- Ben Widawsky, Intel Open Source Technology Center
