Re: [Xen-devel] [PATCHv6 3/5] gnttab: split grant table lock into table and maptrack locks

2015-04-29 Thread Jan Beulich
David Vrabel david.vra...@citrix.com 04/29/15 12:54 PM On 23/04/15 16:04, Jan Beulich wrote: --- a/xen/include/xen/grant_table.h +++ b/xen/include/xen/grant_table.h @@ -82,7 +82,12 @@ struct grant_table { struct grant_mapping **maptrack; unsigned int maptrack_head;

Re: [Xen-devel] [PATCHv6 3/5] gnttab: split grant table lock into table and maptrack locks

2015-04-29 Thread David Vrabel
On 23/04/15 16:04, Jan Beulich wrote: --- a/xen/include/xen/grant_table.h +++ b/xen/include/xen/grant_table.h @@ -82,7 +82,12 @@ struct grant_table { struct grant_mapping **maptrack; unsigned int maptrack_head; unsigned int maptrack_limit; -/* Lock

Re: [Xen-devel] [PATCHv6 3/5] gnttab: split grant table lock into table and maptrack locks

2015-04-23 Thread Jan Beulich
On 22.04.15 at 18:00, david.vra...@citrix.com wrote: Subsequent changes make both these locks uncontented. Is this patch really necessary? -- dvrabel Yeah, particularly if this ... @@ -540,6 +550,16 @@ static void mapcount( *wrc = *rdc = 0; +/* + * While taking the

[Xen-devel] [PATCHv6 3/5] gnttab: split grant table lock into table and maptrack locks

2015-04-22 Thread David Vrabel
From: Matt Wilson m...@amazon.com The maptrack lock protects the maptrack state only. Signed-off-by: Matt Wilson m...@amazon.com Signed-off-by: David Vrabel david.vra...@citrix.com --- Subsequent changes make both these locks uncontented. Is this patch really necessary? -- dvrabel ---