On Wed, Sep 04, 2013 at 09:10:50PM +0100, Zoltan Kiss wrote:
> On 03/09/13 13:42, Konrad Rzeszutek Wilk wrote:
> >Correct. The double buffering code is being run in lib/swiotlb.c not the
> >xen-swiotlb.c. Hence the question of why not move the tracing in there.
> I've put the trace to both location
On 03/09/13 13:42, Konrad Rzeszutek Wilk wrote:
Correct. The double buffering code is being run in lib/swiotlb.c not the
xen-swiotlb.c. Hence the question of why not move the tracing in there.
I've put the trace to both locations before swiotlb_tbl_map_single is
called, so the same tracer will b
On 22/08/13 23:02, Steven Rostedt wrote:
Please use a more specific name. "bounce" is too generic. I know
tracepoints are grouped by systems, but its easier for tools to just
state a tracepoint name than the system:event pair.
Sure, see my v2 patch coming soon.
Zoli
--
To unsubscribe from this
On Mon, Sep 02, 2013 at 06:54:30PM +0100, Zoltan Kiss wrote:
> On 23/08/13 13:55, Konrad Rzeszutek Wilk wrote:
> >On Thu, Aug 22, 2013 at 10:47:28PM +0100, Zoltan Kiss wrote:
> >>Ftrace is currently not able to detect when SWIOTLB has to do double
> >>buffering
> >>under Xen. You can only see it i
On 23/08/13 13:55, Konrad Rzeszutek Wilk wrote:
On Thu, Aug 22, 2013 at 10:47:28PM +0100, Zoltan Kiss wrote:
Ftrace is currently not able to detect when SWIOTLB has to do double buffering
under Xen. You can only see it indirectly in function_graph, when
xen_swiotlb_map_page() doesn't stop after
On Thu, Aug 22, 2013 at 10:47:28PM +0100, Zoltan Kiss wrote:
> Ftrace is currently not able to detect when SWIOTLB has to do double buffering
> under Xen. You can only see it indirectly in function_graph, when
> xen_swiotlb_map_page() doesn't stop after range_straddles_page_boundary(), but
> calls
On Thu, 22 Aug 2013 22:47:28 +0100
Zoltan Kiss wrote:
> /*
> * Used to do a quick range check in swiotlb_tbl_unmap_single and
> * swiotlb_tbl_sync_single_*, to see if the memory was in fact allocated by
> this
> @@ -358,6 +362,9 @@ dma_addr_t xen_swiotlb_map_page(struct device *dev,
> str
7 matches
Mail list logo