rpms/qemu-kvm
7
0
Fork 0
Code Issues Pull Requests Releases Activity

* Wed Sep 04 2019 Danilo Cesar Lemes de Paula <ddepaula@redhat.com> - 4.1.0-6.el8

Browse Source 
- kvm-memory-Refactor-memory_region_clear_coalescing.patch [bz#1743142]
- kvm-memory-Split-zones-when-do-coalesced_io_del.patch [bz#1743142]
- kvm-memory-Remove-has_coalesced_range-counter.patch [bz#1743142]
- kvm-memory-Fix-up-memory_region_-add-del-_coalescing.patch [bz#1743142]
- kvm-enable-virgl-for-real-this-time.patch [bz#1559740]
- Resolves: bz#1559740
  ([RFE] Enable virgl as TechPreview (qemu))
- Resolves: bz#1743142
  (Boot guest with multiple e1000 devices, qemu will crash after several guest reboots: kvm_mem_ioeventfd_add: error adding ioeventfd: No space left on device (28))
This commit is contained in:
Danilo C. L. de Paula 2019-09-04 17:07:50 +01:00
parent 37b050692f
commit 549f70b1e3
5 changed files with 425 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

118
kvm-memory-Fix-up-memory_region_-add-del-_coalescing.patch Normal file
View File

@ -0,0 +1,118 @@
From b9b48ed46d2b0a3dd6e8406946eb0516ec75a004 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Fri, 23 Aug 2019 06:14:31 +0100
Subject: [PATCH 4/5] memory: Fix up memory_region_{add|del}_coalescing
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
Message-id: <20190823061431.31759-5-peterx@redhat.com>
Patchwork-id: 90136
O-Subject: [RHEL-AV-8.1 qemu-kvm PATCH 4/4] memory: Fix up memory_region_{add|del}_coalescing
Bugzilla: 1743142
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: Auger Eric <eric.auger@redhat.com>
RH-Acked-by: Philippe Mathieu-Daudé <philmd@redhat.com>
The old memory_region_{add|clear}_coalescing() has some defects
because they both changed mr->coalesced before updating the regions
using memory_region_update_coalesced_range_as(). Then when the
regions were updated in memory_region_update_coalesced_range_as() the
mr->coalesced will always be either one more or one less. So:
- For memory_region_add_coalescing: it'll always trying to remove the
newly added coalesced region while it shouldn't, and,
- For memory_region_clear_coalescing: when it calls the update there
will be no coalesced ranges on mr->coalesced because they were all
removed before hand so the update will probably do nothing for real.
Let's fix this. Now we've got flat_range_coalesced_io_notify() to
notify a single CoalescedMemoryRange instance change, so use it in the
existing memory_region_update_coalesced_range() logic by only notify
either an addition or deletion. Then we hammer both the
memory_region_{add|clear}_coalescing() to use it.
Fixes: 3ac7d43a6fbb5d4a3
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20190820141328.10009-5-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit b960fc1796fb078c21121abf01499603b66b3f57)
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Danilo C. L. de Paula <ddepaula@redhat.com>
---
memory.c | 36 +++++++++++++++++-------------------
1 file changed, 17 insertions(+), 19 deletions(-)
diff --git a/memory.c b/memory.c
index c7cd43f..2f15180 100644
--- a/memory.c
+++ b/memory.c
@@ -2238,27 +2238,26 @@ void memory_region_ram_resize(MemoryRegion *mr, ram_addr_t newsize, Error **errp
qemu_ram_resize(mr->ram_block, newsize, errp);
}
-static void memory_region_update_coalesced_range_as(MemoryRegion *mr, AddressSpace *as)
+/*
+ * Call proper memory listeners about the change on the newly
+ * added/removed CoalescedMemoryRange.
+ */
+static void memory_region_update_coalesced_range(MemoryRegion *mr,
+ CoalescedMemoryRange *cmr,
+ bool add)
{
+ AddressSpace *as;
FlatView *view;
FlatRange *fr;
- view = address_space_get_flatview(as);
- FOR_EACH_FLAT_RANGE(fr, view) {
- if (fr->mr == mr) {
- flat_range_coalesced_io_del(fr, as);
- flat_range_coalesced_io_add(fr, as);
- }
- }
- flatview_unref(view);
-}
-
-static void memory_region_update_coalesced_range(MemoryRegion *mr)
-{
- AddressSpace *as;
-
QTAILQ_FOREACH(as, &address_spaces, address_spaces_link) {
- memory_region_update_coalesced_range_as(mr, as);
+ view = address_space_get_flatview(as);
+ FOR_EACH_FLAT_RANGE(fr, view) {
+ if (fr->mr == mr) {
+ flat_range_coalesced_io_notify(fr, as, cmr, add);
+ }
+ }
+ flatview_unref(view);
}
}
@@ -2276,7 +2275,7 @@ void memory_region_add_coalescing(MemoryRegion *mr,
cmr->addr = addrrange_make(int128_make64(offset), int128_make64(size));
QTAILQ_INSERT_TAIL(&mr->coalesced, cmr, link);
- memory_region_update_coalesced_range(mr);
+ memory_region_update_coalesced_range(mr, cmr, true);
memory_region_set_flush_coalesced(mr);
}
@@ -2294,10 +2293,9 @@ void memory_region_clear_coalescing(MemoryRegion *mr)
while (!QTAILQ_EMPTY(&mr->coalesced)) {
cmr = QTAILQ_FIRST(&mr->coalesced);
QTAILQ_REMOVE(&mr->coalesced, cmr, link);
+ memory_region_update_coalesced_range(mr, cmr, false);
g_free(cmr);
}
-
- memory_region_update_coalesced_range(mr);
}
void memory_region_set_flush_coalesced(MemoryRegion *mr)
--
1.8.3.1

64
kvm-memory-Refactor-memory_region_clear_coalescing.patch Normal file
View File

@ -0,0 +1,64 @@
From 134ab69ffdfb7e45a0be385595036d0427928306 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Fri, 23 Aug 2019 06:14:28 +0100
Subject: [PATCH 1/5] memory: Refactor memory_region_clear_coalescing
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
Message-id: <20190823061431.31759-2-peterx@redhat.com>
Patchwork-id: 90134
O-Subject: [RHEL-AV-8.1 qemu-kvm PATCH 1/4] memory: Refactor memory_region_clear_coalescing
Bugzilla: 1743142
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: Auger Eric <eric.auger@redhat.com>
RH-Acked-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Removing the update variable and quit earlier if the memory region has
no coalesced range. This prepares for the next patch.
Fixes: 3ac7d43a6fbb5d4a3
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20190820141328.10009-4-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 9c1aa1c235c770d84462d482460a96e957e95b9c)
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Danilo C. L. de Paula <ddepaula@redhat.com>
---
memory.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/memory.c b/memory.c
index 5d8c9a9..9f40742 100644
--- a/memory.c
+++ b/memory.c
@@ -2276,7 +2276,10 @@ void memory_region_add_coalescing(MemoryRegion *mr,
void memory_region_clear_coalescing(MemoryRegion *mr)
{
CoalescedMemoryRange *cmr;
- bool updated = false;
+
+ if (QTAILQ_EMPTY(&mr->coalesced)) {
+ return;
+ }
qemu_flush_coalesced_mmio_buffer();
mr->flush_coalesced_mmio = false;
@@ -2285,12 +2288,9 @@ void memory_region_clear_coalescing(MemoryRegion *mr)
cmr = QTAILQ_FIRST(&mr->coalesced);
QTAILQ_REMOVE(&mr->coalesced, cmr, link);
g_free(cmr);
- updated = true;
}
- if (updated) {
- memory_region_update_coalesced_range(mr);
- }
+ memory_region_update_coalesced_range(mr);
}
void memory_region_set_flush_coalesced(MemoryRegion *mr)
--
1.8.3.1

96
kvm-memory-Remove-has_coalesced_range-counter.patch Normal file
View File

@ -0,0 +1,96 @@
From c1db31bce6d2e5f49e34a2e7282e50bea3f92278 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Fri, 23 Aug 2019 06:14:30 +0100
Subject: [PATCH 3/5] memory: Remove has_coalesced_range counter
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
Message-id: <20190823061431.31759-4-peterx@redhat.com>
Patchwork-id: 90135
O-Subject: [RHEL-AV-8.1 qemu-kvm PATCH 3/4] memory: Remove has_coalesced_range counter
Bugzilla: 1743142
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: Auger Eric <eric.auger@redhat.com>
RH-Acked-by: Philippe Mathieu-Daudé <philmd@redhat.com>
The has_coalesced_range could potentially be problematic in that it
only works for additions of coalesced mmio ranges but not deletions.
The reason is that has_coalesced_range information can be lost when
the FlatView updates the topology again when the updated region is not
covering the coalesced regions. When that happens, due to
flatrange_equal() is not checking against has_coalesced_range, the new
FlatRange will be seen as the same one as the old and the new
instance (whose has_coalesced_range will be zero) will replace the old
instance (whose has_coalesced_range _could_ be non-zero).
The counter was originally used to make sure every FlatRange will only
notify once for coalesced_io_{add|del} memory listeners, because each
FlatRange can be used by multiple address spaces, so logically
speaking it could be called multiple times. However we should not
limit that, because memory listeners should will only be registered
with specific address space rather than multiple address spaces.
So let's fix this up by simply removing the whole has_coalesced_range.
Fixes: 3ac7d43a6fbb5d4a3
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20190820141328.10009-3-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 264ef5a5c52c249ff51a16d141fc03df71714a13)
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Danilo C. L. de Paula <ddepaula@redhat.com>
---
memory.c | 14 --------------
1 file changed, 14 deletions(-)
diff --git a/memory.c b/memory.c
index 7b24cb8..c7cd43f 100644
--- a/memory.c
+++ b/memory.c
@@ -217,7 +217,6 @@ struct FlatRange {
bool romd_mode;
bool readonly;
bool nonvolatile;
- int has_coalesced_range;
};
#define FOR_EACH_FLAT_RANGE(var, view) \
@@ -654,7 +653,6 @@ static void render_memory_region(FlatView *view,
fr.romd_mode = mr->romd_mode;
fr.readonly = readonly;
fr.nonvolatile = nonvolatile;
- fr.has_coalesced_range = 0;
/* Render the region itself into any gaps left by the current view. */
for (i = 0; i < view->nr && int128_nz(remain); ++i) {
@@ -888,14 +886,6 @@ static void flat_range_coalesced_io_del(FlatRange *fr, AddressSpace *as)
{
CoalescedMemoryRange *cmr;
- if (!fr->has_coalesced_range) {
- return;
- }
-
- if (--fr->has_coalesced_range > 0) {
- return;
- }
-
QTAILQ_FOREACH(cmr, &fr->mr->coalesced, link) {
flat_range_coalesced_io_notify(fr, as, cmr, false);
}
@@ -910,10 +900,6 @@ static void flat_range_coalesced_io_add(FlatRange *fr, AddressSpace *as)
return;
}
- if (fr->has_coalesced_range++) {
- return;
- }
-
QTAILQ_FOREACH(cmr, &mr->coalesced, link) {
flat_range_coalesced_io_notify(fr, as, cmr, true);
}
--
1.8.3.1

123
kvm-memory-Split-zones-when-do-coalesced_io_del.patch Normal file
View File

@ -0,0 +1,123 @@
From bdd5394047f7fbecac82d067b9e67db8a20c49d2 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Fri, 23 Aug 2019 06:14:29 +0100
Subject: [PATCH 2/5] memory: Split zones when do coalesced_io_del()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
Message-id: <20190823061431.31759-3-peterx@redhat.com>
Patchwork-id: 90133
O-Subject: [RHEL-AV-8.1 qemu-kvm PATCH 2/4] memory: Split zones when do coalesced_io_del()
Bugzilla: 1743142
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: Auger Eric <eric.auger@redhat.com>
RH-Acked-by: Philippe Mathieu-Daudé <philmd@redhat.com>
It is a workaround of current KVM's KVM_UNREGISTER_COALESCED_MMIO
interface. The kernel interface only allows to unregister an mmio
device with exactly the zone size when registered, or any smaller zone
that is included in the device mmio zone. It does not support the
userspace to specify a very large zone to remove all the small mmio
devices within the zone covered.
Logically speaking it would be nicer to fix this from KVM side, though
in all cases we still need to coop with old kernels so let's do this.
Fixes: 3ac7d43a6fbb5d4a3
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20190820141328.10009-2-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 23f1174aac4181f86bb7e13ca8bc2d4a0bdf1e5c)
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Danilo C. L. de Paula <ddepaula@redhat.com>
---
memory.c | 49 +++++++++++++++++++++++++++++++++++--------------
1 file changed, 35 insertions(+), 14 deletions(-)
diff --git a/memory.c b/memory.c
index 9f40742..7b24cb8 100644
--- a/memory.c
+++ b/memory.c
@@ -855,8 +855,39 @@ static void address_space_update_ioeventfds(AddressSpace *as)
flatview_unref(view);
}
+/*
+ * Notify the memory listeners about the coalesced IO change events of
+ * range `cmr'. Only the part that has intersection of the specified
+ * FlatRange will be sent.
+ */
+static void flat_range_coalesced_io_notify(FlatRange *fr, AddressSpace *as,
+ CoalescedMemoryRange *cmr, bool add)
+{
+ AddrRange tmp;
+
+ tmp = addrrange_shift(cmr->addr,
+ int128_sub(fr->addr.start,
+ int128_make64(fr->offset_in_region)));
+ if (!addrrange_intersects(tmp, fr->addr)) {
+ return;
+ }
+ tmp = addrrange_intersection(tmp, fr->addr);
+
+ if (add) {
+ MEMORY_LISTENER_UPDATE_REGION(fr, as, Forward, coalesced_io_add,
+ int128_get64(tmp.start),
+ int128_get64(tmp.size));
+ } else {
+ MEMORY_LISTENER_UPDATE_REGION(fr, as, Reverse, coalesced_io_del,
+ int128_get64(tmp.start),
+ int128_get64(tmp.size));
+ }
+}
+
static void flat_range_coalesced_io_del(FlatRange *fr, AddressSpace *as)
{
+ CoalescedMemoryRange *cmr;
+
if (!fr->has_coalesced_range) {
return;
}
@@ -865,16 +896,15 @@ static void flat_range_coalesced_io_del(FlatRange *fr, AddressSpace *as)
return;
}
- MEMORY_LISTENER_UPDATE_REGION(fr, as, Reverse, coalesced_io_del,
- int128_get64(fr->addr.start),
- int128_get64(fr->addr.size));
+ QTAILQ_FOREACH(cmr, &fr->mr->coalesced, link) {
+ flat_range_coalesced_io_notify(fr, as, cmr, false);
+ }
}
static void flat_range_coalesced_io_add(FlatRange *fr, AddressSpace *as)
{
MemoryRegion *mr = fr->mr;
CoalescedMemoryRange *cmr;
- AddrRange tmp;
if (QTAILQ_EMPTY(&mr->coalesced)) {
return;
@@ -885,16 +915,7 @@ static void flat_range_coalesced_io_add(FlatRange *fr, AddressSpace *as)
}
QTAILQ_FOREACH(cmr, &mr->coalesced, link) {
- tmp = addrrange_shift(cmr->addr,
- int128_sub(fr->addr.start,
- int128_make64(fr->offset_in_region)));
- if (!addrrange_intersects(tmp, fr->addr)) {
- continue;
- }
- tmp = addrrange_intersection(tmp, fr->addr);
- MEMORY_LISTENER_UPDATE_REGION(fr, as, Forward, coalesced_io_add,
- int128_get64(tmp.start),
- int128_get64(tmp.size));
+ flat_range_coalesced_io_notify(fr, as, cmr, true);
}
}
--
1.8.3.1

26
qemu-kvm.spec
View File

@ -67,7 +67,7 @@ Obsoletes: %1-rhev
Summary: QEMU is a machine emulator and virtualizer
Name: qemu-kvm
Version: 4.1.0
Release: 5%{?dist}
Release: 6%{?dist}
# Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
Epoch: 15
License: GPLv2 and GPLv2+ and CC-BY
@ -142,6 +142,14 @@ Patch28: kvm-redhat-s390x-Rename-s390-ccw-virtio-rhel8.0.0-to-s39.patch
Patch29: kvm-redhat-s390x-Add-proper-compatibility-options-for-th.patch
# For bz#1744170 - [IBM Power] New 8.1.0 machine type for pseries
Patch31: kvm-redhat-update-pseries-rhel8.1.0-machine-type.patch
# For bz#1743142 - Boot guest with multiple e1000 devices, qemu will crash after several guest reboots: kvm_mem_ioeventfd_add: error adding ioeventfd: No space left on device (28)
Patch32: kvm-memory-Refactor-memory_region_clear_coalescing.patch
# For bz#1743142 - Boot guest with multiple e1000 devices, qemu will crash after several guest reboots: kvm_mem_ioeventfd_add: error adding ioeventfd: No space left on device (28)
Patch33: kvm-memory-Split-zones-when-do-coalesced_io_del.patch
# For bz#1743142 - Boot guest with multiple e1000 devices, qemu will crash after several guest reboots: kvm_mem_ioeventfd_add: error adding ioeventfd: No space left on device (28)
Patch34: kvm-memory-Remove-has_coalesced_range-counter.patch
# For bz#1743142 - Boot guest with multiple e1000 devices, qemu will crash after several guest reboots: kvm_mem_ioeventfd_add: error adding ioeventfd: No space left on device (28)
Patch35: kvm-memory-Fix-up-memory_region_-add-del-_coalescing.patch
BuildRequires: wget
BuildRequires: rpm-build
@ -575,7 +583,6 @@ buildldflags="VL_LDFLAGS=-Wl,--build-id"
--enable-vhost-vsock \
--enable-vnc \
--enable-mpath \
--disable-virglrenderer \
--disable-xen-pci-passthrough \
--enable-tcg \
--with-git=git \
@ -1037,6 +1044,10 @@ useradd -r -u 107 -g qemu -G kvm -d / -s /sbin/nologin \
%if 0%{have_memlock_limits}
%{_sysconfdir}/security/limits.d/95-kvm-memlock.conf
%endif
%if %{have_spice}
%{_libexecdir}/vhost-user-gpu
%{_datadir}/%{name}/vhost-user/50-qemu-gpu.json
%endif
%files -n qemu-img
%defattr(-,root,root)
@ -1080,6 +1091,17 @@ useradd -r -u 107 -g qemu -G kvm -d / -s /sbin/nologin \
%changelog
* Wed Sep 04 2019 Danilo Cesar Lemes de Paula <ddepaula@redhat.com> - 4.1.0-6.el8
- kvm-memory-Refactor-memory_region_clear_coalescing.patch [bz#1743142]
- kvm-memory-Split-zones-when-do-coalesced_io_del.patch [bz#1743142]
- kvm-memory-Remove-has_coalesced_range-counter.patch [bz#1743142]
- kvm-memory-Fix-up-memory_region_-add-del-_coalescing.patch [bz#1743142]
- kvm-enable-virgl-for-real-this-time.patch [bz#1559740]
- Resolves: bz#1559740
([RFE] Enable virgl as TechPreview (qemu))
- Resolves: bz#1743142
(Boot guest with multiple e1000 devices, qemu will crash after several guest reboots: kvm_mem_ioeventfd_add: error adding ioeventfd: No space left on device (28))
* Tue Aug 27 2019 Danilo Cesar Lemes de Paula <ddepaula@redhat.com> - 4.1.0-5.el8
- kvm-redhat-s390x-Rename-s390-ccw-virtio-rhel8.0.0-to-s39.patch [bz#1693772]
- kvm-redhat-s390x-Add-proper-compatibility-options-for-th.patch [bz#1693772]