pool/xen
SHA256
4
0
Fork 1
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

- fate#319989 - Update to Xen 4.7 RC2

Browse Source 
xen-4.7.0-testing-src.tar.bz2

- bsc#961600 - L3: poor performance when Xen HVM domU configured
  with max memory > current memory
  x86-PoD-only-reclaim-if-needed.patch

OBS-URL: https://build.opensuse.org/package/show/Virtualization/xen?expand=0&rev=430
This commit is contained in:
Charles Arnold 2016-05-10 22:03:52 +00:00 committed by Git OBS Bridge
parent 66b4cba252
commit f0be2374a0
5 changed files with 50 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
libxl.pvscsi.patch
View File

@ -105,7 +105,7 @@ Index: xen-4.7.0-testing/docs/man/xl.pod.1
===================================================================
--- xen-4.7.0-testing.orig/docs/man/xl.pod.1
+++ xen-4.7.0-testing/docs/man/xl.pod.1
@@ -1416,6 +1416,24 @@ List virtual trusted platform modules fo
@@ -1417,6 +1417,24 @@ List virtual trusted platform modules fo
=back

32
x86-PoD-only-reclaim-if-needed.patch Normal file
View File

@ -0,0 +1,32 @@
References: bsc#961600
Reclaiming pages is pointless when the cache can already satisfy all
outstanding PoD entries, and doing reclaims in that case can be very
harmful to performance when that memory gets used by the guest, but
only to store zeroes there.
Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
Index: xen-4.7.0-testing/xen/arch/x86/mm/p2m-pod.c
===================================================================
--- xen-4.7.0-testing.orig/xen/arch/x86/mm/p2m-pod.c
+++ xen-4.7.0-testing/xen/arch/x86/mm/p2m-pod.c
@@ -1031,7 +1031,6 @@ static void pod_eager_record(struct p2m_
{
struct pod_mrp_list *mrp = &p2m->pod.mrp;
- ASSERT(mrp->list[mrp->idx] == INVALID_GFN);
ASSERT(gfn != INVALID_GFN);
mrp->list[mrp->idx++] =
@@ -1079,7 +1078,9 @@ p2m_pod_demand_populate(struct p2m_domai
return 0;
}
- pod_eager_reclaim(p2m);
+ /* Only reclaim if we're in actual need of more cache. */
+ if ( p2m->pod.entry_count > p2m->pod.count )
+ pod_eager_reclaim(p2m);
/* Only sweep if we're actually out of memory. Doing anything else
* causes unnecessary time and fragmentation of superpages in the p2m. */

4
xen-4.7.0-testing-src.tar.bz2
View File

@ -1,3 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:524bc5df998a153fe4fc84a878057cf14e93c97ca0f45a38f0c30647a359f4da
size 4457990
oid sha256:2b89d13516c53564c901ceb6599543812d7ad18e95fac06c062987984eb43a22
size 4463595

13
xen.changes
View File

@ -1,3 +1,16 @@
-------------------------------------------------------------------
Tue May 10 14:39:00 MDT 2016 - carnold@suse.com
- fate#319989 - Update to Xen 4.7 RC2
xen-4.7.0-testing-src.tar.bz2
-------------------------------------------------------------------
Tue May 10 11:06:56 MDT 2016 - carnold@suse.com
- bsc#961600 - L3: poor performance when Xen HVM domU configured
with max memory > current memory
x86-PoD-only-reclaim-if-needed.patch
-------------------------------------------------------------------
Fri May 6 11:47:24 UTC 2016 - ohering@suse.de

2
xen.spec
View File

@ -281,6 +281,7 @@ Patch463: libxl.set-migration-constraints-from-cmdline.patch
# Hypervisor and PV driver Patches
Patch501: x86-ioapic-ack-default.patch
Patch502: x86-cpufreq-report.patch
Patch503: x86-PoD-only-reclaim-if-needed.patch
Patch520: xen_pvonhvm.xen_emul_unplug.patch
Patch521: supported_module.patch
Patch601: xen.build-compare.doc_html.patch
@ -599,6 +600,7 @@ Authors:
# Hypervisor and PV driver Patches
%patch501 -p1
%patch502 -p1
%patch503 -p1
%patch520 -p1
%patch521 -p1
%patch601 -p1