From: Dongli Zhang <dongli.zhang@oracle.com>
To: xen-devel@lists.xen.org
Cc: sstabellini@kernel.org, wei.liu2@citrix.com,
george.dunlap@eu.citrix.com, ian.jackson@eu.citrix.com,
dario.faggioli@citrix.com, tim@xen.org, david.vrabel@citrix.com,
jbeulich@suse.com, andrew.cooper3@citrix.com
Subject: [PATCH v3 1/1] xen: move TLB-flush filtering out into populate_physmap during vm creation
Date: Thu, 8 Sep 2016 13:30:03 +0800 [thread overview]
Message-ID: <1473312603-28581-1-git-send-email-dongli.zhang@oracle.com> (raw)
This patch implemented parts of TODO left in commit id
a902c12ee45fc9389eb8fe54eeddaf267a555c58. It moved TLB-flush filtering out
into populate_physmap. Because of TLB-flush in alloc_heap_pages, it's very
slow to create a guest with memory size of more than 100GB on host with
100+ cpus.
This patch introduced a "MEMF_no_tlbflush" bit to memflags to indicate
whether TLB-flush should be done in alloc_heap_pages or its caller
populate_physmap. Once this bit is set in memflags, alloc_heap_pages will
ignore TLB-flush. To use this bit after vm is created might lead to
security issue, that is, this would make pages accessible to the guest B,
when guest A may still have a cached mapping to them.
Therefore, this patch also introduced a "already_scheduled" field to struct
domain to indicate whether this domain has ever got scheduled by
hypervisor. MEMF_no_tlbflush can be set only during vm creation phase when
already_scheduled is still 0 before this domain gets scheduled for the
first time.
TODO: ballooning very huge amount of memory cannot benefit from this patch
and might still be slow.
Signed-off-by: Dongli Zhang <dongli.zhang@oracle.com>
---
Changed since v2:
* Limit this optimization to domain creation time.
---
xen/common/domain.c | 2 ++
xen/common/memory.c | 33 +++++++++++++++++++++++++++++++++
xen/common/page_alloc.c | 3 ++-
xen/common/schedule.c | 5 +++++
xen/include/xen/mm.h | 2 ++
xen/include/xen/sched.h | 3 +++
6 files changed, 47 insertions(+), 1 deletion(-)
diff --git a/xen/common/domain.c b/xen/common/domain.c
index a8804e4..611a471 100644
--- a/xen/common/domain.c
+++ b/xen/common/domain.c
@@ -303,6 +303,8 @@ struct domain *domain_create(domid_t domid, unsigned int domcr_flags,
if ( !zalloc_cpumask_var(&d->domain_dirty_cpumask) )
goto fail;
+ d->already_scheduled = 0;
+
if ( domcr_flags & DOMCRF_hvm )
d->guest_type = guest_type_hvm;
else if ( domcr_flags & DOMCRF_pvh )
diff --git a/xen/common/memory.c b/xen/common/memory.c
index f34dd56..3641469 100644
--- a/xen/common/memory.c
+++ b/xen/common/memory.c
@@ -141,6 +141,8 @@ static void populate_physmap(struct memop_args *a)
unsigned int i, j;
xen_pfn_t gpfn, mfn;
struct domain *d = a->domain, *curr_d = current->domain;
+ bool_t need_tlbflush = 0;
+ uint32_t tlbflush_timestamp = 0;
if ( !guest_handle_subrange_okay(a->extent_list, a->nr_done,
a->nr_extents-1) )
@@ -150,6 +152,12 @@ static void populate_physmap(struct memop_args *a)
max_order(curr_d)) )
return;
+ /* MEMF_no_tlbflush can be set only during vm creation phase when
+ * already_scheduled is still 0 before this domain gets scheduled for
+ * the first time. */
+ if ( d->already_scheduled == 0 )
+ a->memflags |= MEMF_no_tlbflush;
+
for ( i = a->nr_done; i < a->nr_extents; i++ )
{
if ( i != a->nr_done && hypercall_preempt_check() )
@@ -214,6 +222,21 @@ static void populate_physmap(struct memop_args *a)
goto out;
}
+ if ( d->already_scheduled == 0 )
+ {
+ for ( j = 0; j < (1U << a->extent_order); j++ )
+ {
+ if ( page[j].u.free.need_tlbflush &&
+ (page[j].tlbflush_timestamp <= tlbflush_current_time()) &&
+ (!need_tlbflush ||
+ (page[j].tlbflush_timestamp > tlbflush_timestamp)) )
+ {
+ need_tlbflush = 1;
+ tlbflush_timestamp = page[j].tlbflush_timestamp;
+ }
+ }
+ }
+
mfn = page_to_mfn(page);
}
@@ -232,6 +255,16 @@ static void populate_physmap(struct memop_args *a)
}
out:
+ if ( need_tlbflush )
+ {
+ cpumask_t mask = cpu_online_map;
+ tlbflush_filter(mask, tlbflush_timestamp);
+ if ( !cpumask_empty(&mask) )
+ {
+ perfc_incr(need_flush_tlb_flush);
+ flush_tlb_mask(&mask);
+ }
+ }
a->nr_done = i;
}
diff --git a/xen/common/page_alloc.c b/xen/common/page_alloc.c
index 18ff6cf..e0283fc 100644
--- a/xen/common/page_alloc.c
+++ b/xen/common/page_alloc.c
@@ -827,7 +827,8 @@ static struct page_info *alloc_heap_pages(
BUG_ON(pg[i].count_info != PGC_state_free);
pg[i].count_info = PGC_state_inuse;
- if ( pg[i].u.free.need_tlbflush &&
+ if ( !(memflags & MEMF_no_tlbflush) &&
+ pg[i].u.free.need_tlbflush &&
(pg[i].tlbflush_timestamp <= tlbflush_current_time()) &&
(!need_tlbflush ||
(pg[i].tlbflush_timestamp > tlbflush_timestamp)) )
diff --git a/xen/common/schedule.c b/xen/common/schedule.c
index 32a300f..593541a 100644
--- a/xen/common/schedule.c
+++ b/xen/common/schedule.c
@@ -1376,6 +1376,11 @@ static void schedule(void)
next = next_slice.task;
+ /* Set already_scheduled to 1 when this domain gets scheduled for the
+ * first time */
+ if ( next->domain->already_scheduled == 0 )
+ next->domain->already_scheduled = 1;
+
sd->curr = next;
if ( next_slice.time >= 0 ) /* -ve means no limit */
diff --git a/xen/include/xen/mm.h b/xen/include/xen/mm.h
index 58bc0b8..880ca88 100644
--- a/xen/include/xen/mm.h
+++ b/xen/include/xen/mm.h
@@ -221,6 +221,8 @@ struct npfec {
#define MEMF_exact_node (1U<<_MEMF_exact_node)
#define _MEMF_no_owner 5
#define MEMF_no_owner (1U<<_MEMF_no_owner)
+#define _MEMF_no_tlbflush 6
+#define MEMF_no_tlbflush (1U<<_MEMF_no_tlbflush)
#define _MEMF_node 8
#define MEMF_node_mask ((1U << (8 * sizeof(nodeid_t))) - 1)
#define MEMF_node(n) ((((n) + 1) & MEMF_node_mask) << _MEMF_node)
diff --git a/xen/include/xen/sched.h b/xen/include/xen/sched.h
index 2f9c15f..cbd8329 100644
--- a/xen/include/xen/sched.h
+++ b/xen/include/xen/sched.h
@@ -474,6 +474,9 @@ struct domain
unsigned int guest_request_enabled : 1;
unsigned int guest_request_sync : 1;
} monitor;
+
+ /* set to 1 the first time this domain gets scheduled. */
+ bool_t already_scheduled;
};
/* Protect updates/reads (resp.) of domain_list and domain_hash. */
--
1.9.1
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
next reply other threads:[~2016-09-08 5:30 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-09-08 5:30 Dongli Zhang [this message]
2016-09-08 7:19 ` [PATCH v3 1/1] xen: move TLB-flush filtering out into populate_physmap during vm creation Dario Faggioli
2016-09-08 10:50 ` Wei Liu
2016-09-08 11:01 ` Dario Faggioli
2016-09-08 11:11 ` Wei Liu
2016-09-08 11:19 ` Juergen Gross
2016-09-08 14:49 ` Dario Faggioli
2016-09-08 14:52 ` Andrew Cooper
2016-09-08 14:28 ` Ian Jackson
2016-09-08 15:36 ` Jan Beulich
2016-09-08 15:53 ` Dario Faggioli
-- strict thread matches above, loose matches on Subject: below --
2016-09-09 5:37 Dongli Zhang
2016-09-09 7:56 ` Jan Beulich
2016-09-09 9:34 ` Andrew Cooper
2016-09-09 9:46 ` Jan Beulich
2016-09-09 9:52 ` Andrew Cooper
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1473312603-28581-1-git-send-email-dongli.zhang@oracle.com \
--to=dongli.zhang@oracle.com \
--cc=andrew.cooper3@citrix.com \
--cc=dario.faggioli@citrix.com \
--cc=david.vrabel@citrix.com \
--cc=george.dunlap@eu.citrix.com \
--cc=ian.jackson@eu.citrix.com \
--cc=jbeulich@suse.com \
--cc=sstabellini@kernel.org \
--cc=tim@xen.org \
--cc=wei.liu2@citrix.com \
--cc=xen-devel@lists.xen.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).