* [PATCH 1/2] xen: enter/exit lazy_mmu_mode around m2p_override calls
@ 2012-03-27 13:52 Stefano Stabellini
2012-03-27 13:52 ` [PATCH 2/2] m2p_find_override: use list_for_each_entry_safe Stefano Stabellini
2012-03-27 14:31 ` [PATCH 1/2] xen: enter/exit lazy_mmu_mode around m2p_override calls Konrad Rzeszutek Wilk
0 siblings, 2 replies; 8+ messages in thread
From: Stefano Stabellini @ 2012-03-27 13:52 UTC (permalink / raw)
To: konrad.wilk
Cc: Stefano.Stabellini, linux-kernel, xen-devel, Stefano Stabellini
This patch is a significant performance improvement for the
m2p_override.
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
drivers/block/xen-blkback/blkback.c | 4 ++++
drivers/xen/grant-table.c | 8 ++++++++
2 files changed, 12 insertions(+), 0 deletions(-)
diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c
index 0088bf6..60abff6 100644
--- a/drivers/block/xen-blkback/blkback.c
+++ b/drivers/block/xen-blkback/blkback.c
@@ -342,6 +342,7 @@ static void xen_blkbk_unmap(struct pending_req *req)
* Note, we use invcount, so nr->pages, so we can't index
* using vaddr(req, i).
*/
+ arch_enter_lazy_mmu_mode();
for (i = 0; i < invcount; i++) {
ret = m2p_remove_override(
virt_to_page(unmap[i].host_addr), false);
@@ -351,6 +352,7 @@ static void xen_blkbk_unmap(struct pending_req *req)
continue;
}
}
+ arch_leave_lazy_mmu_mode();
}
static int xen_blkbk_map(struct blkif_request *req,
@@ -386,6 +388,7 @@ static int xen_blkbk_map(struct blkif_request *req,
* so that when we access vaddr(pending_req,i) it has the contents of
* the page from the other domain.
*/
+ arch_enter_lazy_mmu_mode();
for (i = 0; i < nseg; i++) {
if (unlikely(map[i].status != 0)) {
pr_debug(DRV_PFX "invalid buffer -- could not remap it\n");
@@ -410,6 +413,7 @@ static int xen_blkbk_map(struct blkif_request *req,
seg[i].buf = map[i].dev_bus_addr |
(req->u.rw.seg[i].first_sect << 9);
}
+ arch_leave_lazy_mmu_mode();
return ret;
}
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index b4d4eac..c7dc2d6 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -751,6 +751,8 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
if (xen_feature(XENFEAT_auto_translated_physmap))
return ret;
+ arch_enter_lazy_mmu_mode();
+
for (i = 0; i < count; i++) {
/* Do not add to override if the map failed. */
if (map_ops[i].status)
@@ -769,6 +771,8 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
return ret;
}
+ arch_leave_lazy_mmu_mode();
+
return ret;
}
EXPORT_SYMBOL_GPL(gnttab_map_refs);
@@ -785,12 +789,16 @@ int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
if (xen_feature(XENFEAT_auto_translated_physmap))
return ret;
+ arch_enter_lazy_mmu_mode();
+
for (i = 0; i < count; i++) {
ret = m2p_remove_override(pages[i], clear_pte);
if (ret)
return ret;
}
+ arch_leave_lazy_mmu_mode();
+
return ret;
}
EXPORT_SYMBOL_GPL(gnttab_unmap_refs);
--
1.7.2.5
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH 2/2] m2p_find_override: use list_for_each_entry_safe
2012-03-27 13:52 [PATCH 1/2] xen: enter/exit lazy_mmu_mode around m2p_override calls Stefano Stabellini
@ 2012-03-27 13:52 ` Stefano Stabellini
2012-03-27 14:26 ` Konrad Rzeszutek Wilk
2012-03-27 14:31 ` [PATCH 1/2] xen: enter/exit lazy_mmu_mode around m2p_override calls Konrad Rzeszutek Wilk
1 sibling, 1 reply; 8+ messages in thread
From: Stefano Stabellini @ 2012-03-27 13:52 UTC (permalink / raw)
To: konrad.wilk
Cc: Stefano.Stabellini, linux-kernel, xen-devel, Stefano Stabellini
Use list_for_each_entry_safe and remove the spin_lock acquisition in
m2p_find_override.
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
arch/x86/xen/p2m.c | 8 ++------
1 files changed, 2 insertions(+), 6 deletions(-)
diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c
index 1b267e7..7ed8cc3 100644
--- a/arch/x86/xen/p2m.c
+++ b/arch/x86/xen/p2m.c
@@ -809,21 +809,17 @@ struct page *m2p_find_override(unsigned long mfn)
{
unsigned long flags;
struct list_head *bucket = &m2p_overrides[mfn_hash(mfn)];
- struct page *p, *ret;
+ struct page *p, *t, *ret;
ret = NULL;
- spin_lock_irqsave(&m2p_override_lock, flags);
-
- list_for_each_entry(p, bucket, lru) {
+ list_for_each_entry_safe(p, t, bucket, lru) {
if (page_private(p) == mfn) {
ret = p;
break;
}
}
- spin_unlock_irqrestore(&m2p_override_lock, flags);
-
return ret;
}
--
1.7.2.5
^ permalink raw reply related [flat|nested] 8+ messages in thread
* Re: [PATCH 2/2] m2p_find_override: use list_for_each_entry_safe
2012-03-27 13:52 ` [PATCH 2/2] m2p_find_override: use list_for_each_entry_safe Stefano Stabellini
@ 2012-03-27 14:26 ` Konrad Rzeszutek Wilk
2012-03-27 16:37 ` Stefano Stabellini
0 siblings, 1 reply; 8+ messages in thread
From: Konrad Rzeszutek Wilk @ 2012-03-27 14:26 UTC (permalink / raw)
To: Stefano Stabellini; +Cc: linux-kernel, xen-devel
On Tue, Mar 27, 2012 at 02:52:44PM +0100, Stefano Stabellini wrote:
> Use list_for_each_entry_safe and remove the spin_lock acquisition in
> m2p_find_override.
So this would allow us to get stale entries. Is that OK?
>
> Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> ---
> arch/x86/xen/p2m.c | 8 ++------
> 1 files changed, 2 insertions(+), 6 deletions(-)
>
> diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c
> index 1b267e7..7ed8cc3 100644
> --- a/arch/x86/xen/p2m.c
> +++ b/arch/x86/xen/p2m.c
> @@ -809,21 +809,17 @@ struct page *m2p_find_override(unsigned long mfn)
> {
> unsigned long flags;
> struct list_head *bucket = &m2p_overrides[mfn_hash(mfn)];
> - struct page *p, *ret;
> + struct page *p, *t, *ret;
>
> ret = NULL;
>
> - spin_lock_irqsave(&m2p_override_lock, flags);
> -
> - list_for_each_entry(p, bucket, lru) {
> + list_for_each_entry_safe(p, t, bucket, lru) {
> if (page_private(p) == mfn) {
> ret = p;
> break;
> }
> }
>
> - spin_unlock_irqrestore(&m2p_override_lock, flags);
> -
> return ret;
> }
>
> --
> 1.7.2.5
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 1/2] xen: enter/exit lazy_mmu_mode around m2p_override calls
2012-03-27 13:52 [PATCH 1/2] xen: enter/exit lazy_mmu_mode around m2p_override calls Stefano Stabellini
2012-03-27 13:52 ` [PATCH 2/2] m2p_find_override: use list_for_each_entry_safe Stefano Stabellini
@ 2012-03-27 14:31 ` Konrad Rzeszutek Wilk
2012-03-27 16:56 ` Stefano Stabellini
1 sibling, 1 reply; 8+ messages in thread
From: Konrad Rzeszutek Wilk @ 2012-03-27 14:31 UTC (permalink / raw)
To: Stefano Stabellini; +Cc: linux-kernel, xen-devel
On Tue, Mar 27, 2012 at 02:52:43PM +0100, Stefano Stabellini wrote:
> This patch is a significant performance improvement for the
How much? 10%? 90%?
> m2p_override.
Can you explain what those calls do in the git commit description please?
>
> Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> ---
> drivers/block/xen-blkback/blkback.c | 4 ++++
> drivers/xen/grant-table.c | 8 ++++++++
> 2 files changed, 12 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c
> index 0088bf6..60abff6 100644
> --- a/drivers/block/xen-blkback/blkback.c
> +++ b/drivers/block/xen-blkback/blkback.c
> @@ -342,6 +342,7 @@ static void xen_blkbk_unmap(struct pending_req *req)
> * Note, we use invcount, so nr->pages, so we can't index
> * using vaddr(req, i).
> */
> + arch_enter_lazy_mmu_mode();
> for (i = 0; i < invcount; i++) {
> ret = m2p_remove_override(
> virt_to_page(unmap[i].host_addr), false);
> @@ -351,6 +352,7 @@ static void xen_blkbk_unmap(struct pending_req *req)
> continue;
> }
> }
> + arch_leave_lazy_mmu_mode();
> }
>
> static int xen_blkbk_map(struct blkif_request *req,
> @@ -386,6 +388,7 @@ static int xen_blkbk_map(struct blkif_request *req,
> * so that when we access vaddr(pending_req,i) it has the contents of
> * the page from the other domain.
> */
> + arch_enter_lazy_mmu_mode();
> for (i = 0; i < nseg; i++) {
> if (unlikely(map[i].status != 0)) {
> pr_debug(DRV_PFX "invalid buffer -- could not remap it\n");
> @@ -410,6 +413,7 @@ static int xen_blkbk_map(struct blkif_request *req,
> seg[i].buf = map[i].dev_bus_addr |
> (req->u.rw.seg[i].first_sect << 9);
> }
> + arch_leave_lazy_mmu_mode();
> return ret;
> }
>
> diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
> index b4d4eac..c7dc2d6 100644
> --- a/drivers/xen/grant-table.c
> +++ b/drivers/xen/grant-table.c
> @@ -751,6 +751,8 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
> if (xen_feature(XENFEAT_auto_translated_physmap))
> return ret;
>
> + arch_enter_lazy_mmu_mode();
> +
> for (i = 0; i < count; i++) {
> /* Do not add to override if the map failed. */
> if (map_ops[i].status)
> @@ -769,6 +771,8 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
> return ret;
> }
>
> + arch_leave_lazy_mmu_mode();
> +
> return ret;
> }
> EXPORT_SYMBOL_GPL(gnttab_map_refs);
> @@ -785,12 +789,16 @@ int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
> if (xen_feature(XENFEAT_auto_translated_physmap))
> return ret;
>
> + arch_enter_lazy_mmu_mode();
> +
> for (i = 0; i < count; i++) {
> ret = m2p_remove_override(pages[i], clear_pte);
> if (ret)
> return ret;
> }
>
> + arch_leave_lazy_mmu_mode();
> +
> return ret;
> }
> EXPORT_SYMBOL_GPL(gnttab_unmap_refs);
> --
> 1.7.2.5
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 2/2] m2p_find_override: use list_for_each_entry_safe
2012-03-27 14:26 ` Konrad Rzeszutek Wilk
@ 2012-03-27 16:37 ` Stefano Stabellini
2012-03-27 17:59 ` [Xen-devel] " Konrad Rzeszutek Wilk
0 siblings, 1 reply; 8+ messages in thread
From: Stefano Stabellini @ 2012-03-27 16:37 UTC (permalink / raw)
To: Konrad Rzeszutek Wilk; +Cc: Stefano Stabellini, linux-kernel, xen-devel
On Tue, 27 Mar 2012, Konrad Rzeszutek Wilk wrote:
> On Tue, Mar 27, 2012 at 02:52:44PM +0100, Stefano Stabellini wrote:
> > Use list_for_each_entry_safe and remove the spin_lock acquisition in
> > m2p_find_override.
>
> So this would allow us to get stale entries. Is that OK?
I think it is reasonable.
In practice we should never get an m2p_find_override call looking for an
entry that we are about to add/delete.
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 1/2] xen: enter/exit lazy_mmu_mode around m2p_override calls
2012-03-27 14:31 ` [PATCH 1/2] xen: enter/exit lazy_mmu_mode around m2p_override calls Konrad Rzeszutek Wilk
@ 2012-03-27 16:56 ` Stefano Stabellini
0 siblings, 0 replies; 8+ messages in thread
From: Stefano Stabellini @ 2012-03-27 16:56 UTC (permalink / raw)
To: Konrad Rzeszutek Wilk; +Cc: Stefano Stabellini, linux-kernel, xen-devel
On Tue, 27 Mar 2012, Konrad Rzeszutek Wilk wrote:
> On Tue, Mar 27, 2012 at 02:52:43PM +0100, Stefano Stabellini wrote:
> > This patch is a significant performance improvement for the
>
> How much? 10%? 90%?
I think you are overestimating the meaning of "significant" :)
It is about 6% using grant_table and 2% using blkback.
> > m2p_override.
>
> Can you explain what those calls do in the git commit description please?
Right, I have been way too concise.
Each m2p_add/remove_override call issues a MULTI_grant_table_op and a
__flush_tlb_single if kmap_op != NULL.
Batching all the calls together is a great performance benefit because
it means issuing one hypercall total rather than two hypercall per page.
I'll resend with a proper commit message.
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [Xen-devel] [PATCH 2/2] m2p_find_override: use list_for_each_entry_safe
2012-03-27 16:37 ` Stefano Stabellini
@ 2012-03-27 17:59 ` Konrad Rzeszutek Wilk
2012-03-29 14:03 ` Stefano Stabellini
0 siblings, 1 reply; 8+ messages in thread
From: Konrad Rzeszutek Wilk @ 2012-03-27 17:59 UTC (permalink / raw)
To: Stefano Stabellini; +Cc: Konrad Rzeszutek Wilk, xen-devel, linux-kernel
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=unknown-8bit, Size: 862 bytes --]
On Tue, Mar 27, 2012 at 05:37:07PM +0100, Stefano Stabellini wrote:
> On Tue, 27 Mar 2012, Konrad Rzeszutek Wilk wrote:
> > On Tue, Mar 27, 2012 at 02:52:44PM +0100, Stefano Stabellini wrote:
> > > Use list_for_each_entry_safe and remove the spin_lock acquisition in
> > > m2p_find_override.
> >
> > So this would allow us to get stale entries. Is that OK?
>
> I think it is reasonable.
> In practice we should never get an m2p_find_override call looking for an
> entry that we are about to add/delete.
Why don't you add that in the comment and also fix:
/home/konrad/ssd/linux/arch/x86/xen/p2m.c: In function
‘m2p_find_override’:
/home/konrad/ssd/linux/arch/x86/xen/p2m.c:810: warning: unused variable
‘flags’
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [Xen-devel] [PATCH 2/2] m2p_find_override: use list_for_each_entry_safe
2012-03-27 17:59 ` [Xen-devel] " Konrad Rzeszutek Wilk
@ 2012-03-29 14:03 ` Stefano Stabellini
0 siblings, 0 replies; 8+ messages in thread
From: Stefano Stabellini @ 2012-03-29 14:03 UTC (permalink / raw)
To: Konrad Rzeszutek Wilk
Cc: Stefano Stabellini, Konrad Rzeszutek Wilk, xen-devel, linux-kernel
[-- Attachment #1: Type: text/plain, Size: 830 bytes --]
On Tue, 27 Mar 2012, Konrad Rzeszutek Wilk wrote:
> On Tue, Mar 27, 2012 at 05:37:07PM +0100, Stefano Stabellini wrote:
> > On Tue, 27 Mar 2012, Konrad Rzeszutek Wilk wrote:
> > > On Tue, Mar 27, 2012 at 02:52:44PM +0100, Stefano Stabellini wrote:
> > > > Use list_for_each_entry_safe and remove the spin_lock acquisition in
> > > > m2p_find_override.
> > >
> > > So this would allow us to get stale entries. Is that OK?
> >
> > I think it is reasonable.
> > In practice we should never get an m2p_find_override call looking for an
> > entry that we are about to add/delete.
>
> Why don't you add that in the comment and also fix:
>
> /home/konrad/ssd/linux/arch/x86/xen/p2m.c: In function
> ‘m2p_find_override’:
> /home/konrad/ssd/linux/arch/x86/xen/p2m.c:810: warning: unused variable
> ‘flags’
Ooops. I'll do that.
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2012-03-29 14:01 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-03-27 13:52 [PATCH 1/2] xen: enter/exit lazy_mmu_mode around m2p_override calls Stefano Stabellini
2012-03-27 13:52 ` [PATCH 2/2] m2p_find_override: use list_for_each_entry_safe Stefano Stabellini
2012-03-27 14:26 ` Konrad Rzeszutek Wilk
2012-03-27 16:37 ` Stefano Stabellini
2012-03-27 17:59 ` [Xen-devel] " Konrad Rzeszutek Wilk
2012-03-29 14:03 ` Stefano Stabellini
2012-03-27 14:31 ` [PATCH 1/2] xen: enter/exit lazy_mmu_mode around m2p_override calls Konrad Rzeszutek Wilk
2012-03-27 16:56 ` Stefano Stabellini
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.