ebasov
/
qemu-kvm
forked from rpms/qemu-kvm
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

import qemu-kvm-9.1.0-5.el10

Browse Source
cs10 imports/cs10/qemu-kvm-9.1.0-5.el10
MSVSphere Packaging Team 4 weeks ago
parent 0bb7a025d0
commit 932a9d1ab7
Signed by untrusted user who does not match committer: sys_gitsync
GPG Key ID: B2B0B9F29E528FE8
9 changed files with 1012 additions and 1 deletions
Show Stats Download Patch File Download Diff File

50
SOURCES/kvm-KVM-Define-KVM_MEMSLOTS_NUM_MAX_DEFAULT.patch
Unescape View File

@ -0,0 +1,50 @@
From 111d70a5bdc3ee0dde0a6def9e0c75ed20b4f093 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Tue, 17 Sep 2024 12:38:33 -0400
Subject: [PATCH 6/9] KVM: Define KVM_MEMSLOTS_NUM_MAX_DEFAULT
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 285: KVM: Dynamic sized kvm memslots array
RH-Jira: RHEL-57685
RH-Acked-by: Juraj Marcin <None>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [5/7] e4c2a2c2f3a809c8efb709521c7a94ba0627c69b (peterx/qemu-kvm)
Make the default max nr_slots a macro, it's only used when KVM reports
nothing.
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Link: https://lore.kernel.org/r/20240917163835.194664-3-peterx@redhat.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit b34a908c8f24eedb0a8e5ff486b059b58fd793f4)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
accel/kvm/kvm-all.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index 38393bc86b..87db0f9494 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -71,6 +71,8 @@
/* Default num of memslots to be allocated when VM starts */
#define KVM_MEMSLOTS_NR_ALLOC_DEFAULT 16
+/* Default max allowed memslots if kernel reported nothing */
+#define KVM_MEMSLOTS_NR_MAX_DEFAULT 32
struct KVMParkedVcpu {
unsigned long vcpu_id;
@@ -2617,7 +2619,7 @@ static int kvm_init(MachineState *ms)
/* If unspecified, use the default value */
if (!s->nr_slots) {
- s->nr_slots = 32;
+ s->nr_slots_max = KVM_MEMSLOTS_NR_MAX_DEFAULT;
}
s->nr_as = kvm_check_extension(s, KVM_CAP_MULTI_ADDRESS_SPACE);
--
2.39.3

251
SOURCES/kvm-KVM-Dynamic-sized-kvm-memslots-array.patch
Unescape View File

@ -0,0 +1,251 @@
From c77a30265b8d0db43174b040ea82103f8fdb9911 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Tue, 17 Sep 2024 12:38:32 -0400
Subject: [PATCH 5/9] KVM: Dynamic sized kvm memslots array
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 285: KVM: Dynamic sized kvm memslots array
RH-Jira: RHEL-57685
RH-Acked-by: Juraj Marcin <None>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [4/7] 46d4abec352a92112e593ea61b7cbf5ce5f94cdc (peterx/qemu-kvm)
Zhiyi reported an infinite loop issue in VFIO use case. The cause of that
was a separate discussion, however during that I found a regression of
dirty sync slowness when profiling.
Each KVMMemoryListerner maintains an array of kvm memslots. Currently it's
statically allocated to be the max supported by the kernel. However after
Linux commit 4fc096a99e ("KVM: Raise the maximum number of user memslots"),
the max supported memslots reported now grows to some number large enough
so that it may not be wise to always statically allocate with the max
reported.
What's worse, QEMU kvm code still walks all the allocated memslots entries
to do any form of lookups. It can drastically slow down all memslot
operations because each of such loop can run over 32K times on the new
kernels.
Fix this issue by making the memslots to be allocated dynamically.
Here the initial size was set to 16 because it should cover the basic VM
usages, so that the hope is the majority VM use case may not even need to
grow at all (e.g. if one starts a VM with ./qemu-system-x86_64 by default
it'll consume 9 memslots), however not too large to waste memory.
There can also be even better way to address this, but so far this is the
simplest and should be already better even than before we grow the max
supported memslots. For example, in the case of above issue when VFIO was
attached on a 32GB system, there are only ~10 memslots used. So it could
be good enough as of now.
In the above VFIO context, measurement shows that the precopy dirty sync
shrinked from ~86ms to ~3ms after this patch applied. It should also apply
to any KVM enabled VM even without VFIO.
NOTE: we don't have a FIXES tag for this patch because there's no real
commit that regressed this in QEMU. Such behavior existed for a long time,
but only start to be a problem when the kernel reports very large
nr_slots_max value. However that's pretty common now (the kernel change
was merged in 2021) so we attached cc:stable because we'll want this change
to be backported to stable branches.
Cc: qemu-stable <qemu-stable@nongnu.org>
Reported-by: Zhiyi Guo <zhguo@redhat.com>
Tested-by: Zhiyi Guo <zhguo@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Link: https://lore.kernel.org/r/20240917163835.194664-2-peterx@redhat.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 5504a8126115d173687b37e657312a8ffe29fc0c)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
accel/kvm/kvm-all.c | 87 +++++++++++++++++++++++++++++++++-------
accel/kvm/trace-events | 1 +
include/sysemu/kvm_int.h | 1 +
3 files changed, 74 insertions(+), 15 deletions(-)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index 8187ad3964..38393bc86b 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -69,6 +69,9 @@
#define KVM_GUESTDBG_BLOCKIRQ 0
#endif
+/* Default num of memslots to be allocated when VM starts */
+#define KVM_MEMSLOTS_NR_ALLOC_DEFAULT 16
+
struct KVMParkedVcpu {
unsigned long vcpu_id;
int kvm_fd;
@@ -165,6 +168,57 @@ void kvm_resample_fd_notify(int gsi)
}
}
+/**
+ * kvm_slots_grow(): Grow the slots[] array in the KVMMemoryListener
+ *
+ * @kml: The KVMMemoryListener* to grow the slots[] array
+ * @nr_slots_new: The new size of slots[] array
+ *
+ * Returns: True if the array grows larger, false otherwise.
+ */
+static bool kvm_slots_grow(KVMMemoryListener *kml, unsigned int nr_slots_new)
+{
+ unsigned int i, cur = kml->nr_slots_allocated;
+ KVMSlot *slots;
+
+ if (nr_slots_new > kvm_state->nr_slots) {
+ nr_slots_new = kvm_state->nr_slots;
+ }
+
+ if (cur >= nr_slots_new) {
+ /* Big enough, no need to grow, or we reached max */
+ return false;
+ }
+
+ if (cur == 0) {
+ slots = g_new0(KVMSlot, nr_slots_new);
+ } else {
+ assert(kml->slots);
+ slots = g_renew(KVMSlot, kml->slots, nr_slots_new);
+ /*
+ * g_renew() doesn't initialize extended buffers, however kvm
+ * memslots require fields to be zero-initialized. E.g. pointers,
+ * memory_size field, etc.
+ */
+ memset(&slots[cur], 0x0, sizeof(slots[0]) * (nr_slots_new - cur));
+ }
+
+ for (i = cur; i < nr_slots_new; i++) {
+ slots[i].slot = i;
+ }
+
+ kml->slots = slots;
+ kml->nr_slots_allocated = nr_slots_new;
+ trace_kvm_slots_grow(cur, nr_slots_new);
+
+ return true;
+}
+
+static bool kvm_slots_double(KVMMemoryListener *kml)
+{
+ return kvm_slots_grow(kml, kml->nr_slots_allocated * 2);
+}
+
unsigned int kvm_get_max_memslots(void)
{
KVMState *s = KVM_STATE(current_accel());
@@ -193,15 +247,26 @@ unsigned int kvm_get_free_memslots(void)
/* Called with KVMMemoryListener.slots_lock held */
static KVMSlot *kvm_get_free_slot(KVMMemoryListener *kml)
{
- KVMState *s = kvm_state;
+ unsigned int n;
int i;
- for (i = 0; i < s->nr_slots; i++) {
+ for (i = 0; i < kml->nr_slots_allocated; i++) {
if (kml->slots[i].memory_size == 0) {
return &kml->slots[i];
}
}
+ /*
+ * If no free slots, try to grow first by doubling. Cache the old size
+ * here to avoid another round of search: if the grow succeeded, it
+ * means slots[] now must have the existing "n" slots occupied,
+ * followed by one or more free slots starting from slots[n].
+ */
+ n = kml->nr_slots_allocated;
+ if (kvm_slots_double(kml)) {
+ return &kml->slots[n];
+ }
+
return NULL;
}
@@ -222,10 +287,9 @@ static KVMSlot *kvm_lookup_matching_slot(KVMMemoryListener *kml,
hwaddr start_addr,
hwaddr size)
{
- KVMState *s = kvm_state;
int i;
- for (i = 0; i < s->nr_slots; i++) {
+ for (i = 0; i < kml->nr_slots_allocated; i++) {
KVMSlot *mem = &kml->slots[i];
if (start_addr == mem->start_addr && size == mem->memory_size) {
@@ -267,7 +331,7 @@ int kvm_physical_memory_addr_from_host(KVMState *s, void *ram,
int i, ret = 0;
kvm_slots_lock();
- for (i = 0; i < s->nr_slots; i++) {
+ for (i = 0; i < kml->nr_slots_allocated; i++) {
KVMSlot *mem = &kml->slots[i];
if (ram >= mem->ram && ram < mem->ram + mem->memory_size) {
@@ -1071,7 +1135,7 @@ static int kvm_physical_log_clear(KVMMemoryListener *kml,
kvm_slots_lock();
- for (i = 0; i < s->nr_slots; i++) {
+ for (i = 0; i < kml->nr_slots_allocated; i++) {
mem = &kml->slots[i];
/* Discard slots that are empty or do not overlap the section */
if (!mem->memory_size ||
@@ -1719,12 +1783,8 @@ static void kvm_log_sync_global(MemoryListener *l, bool last_stage)
/* Flush all kernel dirty addresses into KVMSlot dirty bitmap */
kvm_dirty_ring_flush();
- /*
- * TODO: make this faster when nr_slots is big while there are
- * only a few used slots (small VMs).
- */
kvm_slots_lock();
- for (i = 0; i < s->nr_slots; i++) {
+ for (i = 0; i < kml->nr_slots_allocated; i++) {
mem = &kml->slots[i];
if (mem->memory_size && mem->flags & KVM_MEM_LOG_DIRTY_PAGES) {
kvm_slot_sync_dirty_pages(mem);
@@ -1839,12 +1899,9 @@ void kvm_memory_listener_register(KVMState *s, KVMMemoryListener *kml,
{
int i;
- kml->slots = g_new0(KVMSlot, s->nr_slots);
kml->as_id = as_id;
- for (i = 0; i < s->nr_slots; i++) {
- kml->slots[i].slot = i;
- }
+ kvm_slots_grow(kml, KVM_MEMSLOTS_NR_ALLOC_DEFAULT);
QSIMPLEQ_INIT(&kml->transaction_add);
QSIMPLEQ_INIT(&kml->transaction_del);
diff --git a/accel/kvm/trace-events b/accel/kvm/trace-events
index 37626c1ac5..ad2ae6fca5 100644
--- a/accel/kvm/trace-events
+++ b/accel/kvm/trace-events
@@ -36,3 +36,4 @@ kvm_io_window_exit(void) ""
kvm_run_exit_system_event(int cpu_index, uint32_t event_type) "cpu_index %d, system_even_type %"PRIu32
kvm_convert_memory(uint64_t start, uint64_t size, const char *msg) "start 0x%" PRIx64 " size 0x%" PRIx64 " %s"
kvm_memory_fault(uint64_t start, uint64_t size, uint64_t flags) "start 0x%" PRIx64 " size 0x%" PRIx64 " flags 0x%" PRIx64
+kvm_slots_grow(unsigned int old, unsigned int new) "%u -> %u"
diff --git a/include/sysemu/kvm_int.h b/include/sysemu/kvm_int.h
index 1d8fb1473b..48e496b3d4 100644
--- a/include/sysemu/kvm_int.h
+++ b/include/sysemu/kvm_int.h
@@ -46,6 +46,7 @@ typedef struct KVMMemoryListener {
MemoryListener listener;
KVMSlot *slots;
unsigned int nr_used_slots;
+ unsigned int nr_slots_allocated;
int as_id;
QSIMPLEQ_HEAD(, KVMMemoryUpdate) transaction_add;
QSIMPLEQ_HEAD(, KVMMemoryUpdate) transaction_del;
--
2.39.3

73
SOURCES/kvm-KVM-Rename-KVMMemoryListener.nr_used_slots-to-nr_slo.patch
Unescape View File

@ -0,0 +1,73 @@
From b1d082cfad79245ac0ffed45f723092388d1cf45 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Tue, 17 Sep 2024 12:38:34 -0400
Subject: [PATCH 7/9] KVM: Rename KVMMemoryListener.nr_used_slots to
nr_slots_used
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 285: KVM: Dynamic sized kvm memslots array
RH-Jira: RHEL-57685
RH-Acked-by: Juraj Marcin <None>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [6/7] ed173123ee23edcf62a6c1940ca74cdfd6b545e9 (peterx/qemu-kvm)
This will make all nr_slots counters to be named in the same manner.
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Link: https://lore.kernel.org/r/20240917163835.194664-4-peterx@redhat.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit dbdc00ba5b136bba80d850f61cc79a9cafaae1cd)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
accel/kvm/kvm-all.c | 6 +++---
include/sysemu/kvm_int.h | 2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index 87db0f9494..e99aaba486 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -239,7 +239,7 @@ unsigned int kvm_get_free_memslots(void)
if (!s->as[i].ml) {
continue;
}
- used_slots = MAX(used_slots, s->as[i].ml->nr_used_slots);
+ used_slots = MAX(used_slots, s->as[i].ml->nr_slots_used);
}
kvm_slots_unlock();
@@ -1516,7 +1516,7 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml,
}
start_addr += slot_size;
size -= slot_size;
- kml->nr_used_slots--;
+ kml->nr_slots_used--;
} while (size);
return;
}
@@ -1555,7 +1555,7 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml,
ram_start_offset += slot_size;
ram += slot_size;
size -= slot_size;
- kml->nr_used_slots++;
+ kml->nr_slots_used++;
} while (size);
}
diff --git a/include/sysemu/kvm_int.h b/include/sysemu/kvm_int.h
index 48e496b3d4..b705dfc9b4 100644
--- a/include/sysemu/kvm_int.h
+++ b/include/sysemu/kvm_int.h
@@ -45,7 +45,7 @@ typedef struct KVMMemoryUpdate {
typedef struct KVMMemoryListener {
MemoryListener listener;
KVMSlot *slots;
- unsigned int nr_used_slots;
+ unsigned int nr_slots_used;
unsigned int nr_slots_allocated;
int as_id;
QSIMPLEQ_HEAD(, KVMMemoryUpdate) transaction_add;
--
2.39.3

90
SOURCES/kvm-KVM-Rename-KVMState-nr_slots-to-nr_slots_max.patch
Unescape View File

@ -0,0 +1,90 @@
From 891fb13363d168760cd21d0c57368e1a413cad27 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Tue, 17 Sep 2024 12:38:35 -0400
Subject: [PATCH 8/9] KVM: Rename KVMState->nr_slots to nr_slots_max
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 285: KVM: Dynamic sized kvm memslots array
RH-Jira: RHEL-57685
RH-Acked-by: Juraj Marcin <None>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [7/7] 7a1b28f04ee6a2c80b07db241fc88cb40f54e376 (peterx/qemu-kvm)
This value used to reflect the maximum supported memslots from KVM kernel.
Rename it to be clearer.
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Link: https://lore.kernel.org/r/20240917163835.194664-5-peterx@redhat.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 943c742868c739c0b14fd996bad3adf744156fec)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
accel/kvm/kvm-all.c | 12 ++++++------
include/sysemu/kvm_int.h | 4 ++--
2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index e99aaba486..dc6253895d 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -183,8 +183,8 @@ static bool kvm_slots_grow(KVMMemoryListener *kml, unsigned int nr_slots_new)
unsigned int i, cur = kml->nr_slots_allocated;
KVMSlot *slots;
- if (nr_slots_new > kvm_state->nr_slots) {
- nr_slots_new = kvm_state->nr_slots;
+ if (nr_slots_new > kvm_state->nr_slots_max) {
+ nr_slots_new = kvm_state->nr_slots_max;
}
if (cur >= nr_slots_new) {
@@ -225,7 +225,7 @@ unsigned int kvm_get_max_memslots(void)
{
KVMState *s = KVM_STATE(current_accel());
- return s->nr_slots;
+ return s->nr_slots_max;
}
unsigned int kvm_get_free_memslots(void)
@@ -243,7 +243,7 @@ unsigned int kvm_get_free_memslots(void)
}
kvm_slots_unlock();
- return s->nr_slots - used_slots;
+ return s->nr_slots_max - used_slots;
}
/* Called with KVMMemoryListener.slots_lock held */
@@ -2615,10 +2615,10 @@ static int kvm_init(MachineState *ms)
(kvm_supported_memory_attributes & KVM_MEMORY_ATTRIBUTE_PRIVATE);
kvm_immediate_exit = kvm_check_extension(s, KVM_CAP_IMMEDIATE_EXIT);
- s->nr_slots = kvm_check_extension(s, KVM_CAP_NR_MEMSLOTS);
+ s->nr_slots_max = kvm_check_extension(s, KVM_CAP_NR_MEMSLOTS);
/* If unspecified, use the default value */
- if (!s->nr_slots) {
+ if (!s->nr_slots_max) {
s->nr_slots_max = KVM_MEMSLOTS_NR_MAX_DEFAULT;
}
diff --git a/include/sysemu/kvm_int.h b/include/sysemu/kvm_int.h
index b705dfc9b4..2c57194b6b 100644
--- a/include/sysemu/kvm_int.h
+++ b/include/sysemu/kvm_int.h
@@ -103,8 +103,8 @@ struct KVMDirtyRingReaper {
struct KVMState
{
AccelState parent_obj;
-
- int nr_slots;
+ /* Max number of KVM slots supported */
+ int nr_slots_max;
int fd;
int vmfd;
int coalesced_mmio;
--
2.39.3

144
SOURCES/kvm-accel-kvm-refactor-dirty-ring-setup.patch
Unescape View File

@ -0,0 +1,144 @@
From 00a2dbf483a077bb31b1c9f70cced36319d22628 Mon Sep 17 00:00:00 2001
From: Ani Sinha <anisinha@redhat.com>
Date: Thu, 12 Sep 2024 11:48:38 +0530
Subject: [PATCH 4/9] accel/kvm: refactor dirty ring setup
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 285: KVM: Dynamic sized kvm memslots array
RH-Jira: RHEL-57685
RH-Acked-by: Juraj Marcin <None>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [3/7] 94f345d1e7ad6437dd2ce67ca7cad224c67aa48f (peterx/qemu-kvm)
Refactor setting up of dirty ring code in kvm_init() so that is can be
reused in the future patchsets.
Signed-off-by: Ani Sinha <anisinha@redhat.com>
Link: https://lore.kernel.org/r/20240912061838.4501-1-anisinha@redhat.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 28ed7f9761eb273e7dedcfdc0507d158106d0451)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
accel/kvm/kvm-all.c | 88 +++++++++++++++++++++++++--------------------
1 file changed, 50 insertions(+), 38 deletions(-)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index d86d1b515a..8187ad3964 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -2439,6 +2439,55 @@ static int find_kvm_machine_type(MachineState *ms)
return type;
}
+static int kvm_setup_dirty_ring(KVMState *s)
+{
+ uint64_t dirty_log_manual_caps;
+ int ret;
+
+ /*
+ * Enable KVM dirty ring if supported, otherwise fall back to
+ * dirty logging mode
+ */
+ ret = kvm_dirty_ring_init(s);
+ if (ret < 0) {
+ return ret;
+ }
+
+ /*
+ * KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is not needed when dirty ring is
+ * enabled. More importantly, KVM_DIRTY_LOG_INITIALLY_SET will assume no
+ * page is wr-protected initially, which is against how kvm dirty ring is
+ * usage - kvm dirty ring requires all pages are wr-protected at the very
+ * beginning. Enabling this feature for dirty ring causes data corruption.
+ *
+ * TODO: Without KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 and kvm clear dirty log,
+ * we may expect a higher stall time when starting the migration. In the
+ * future we can enable KVM_CLEAR_DIRTY_LOG to work with dirty ring too:
+ * instead of clearing dirty bit, it can be a way to explicitly wr-protect
+ * guest pages.
+ */
+ if (!s->kvm_dirty_ring_size) {
+ dirty_log_manual_caps =
+ kvm_check_extension(s, KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2);
+ dirty_log_manual_caps &= (KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE |
+ KVM_DIRTY_LOG_INITIALLY_SET);
+ s->manual_dirty_log_protect = dirty_log_manual_caps;
+ if (dirty_log_manual_caps) {
+ ret = kvm_vm_enable_cap(s, KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2, 0,
+ dirty_log_manual_caps);
+ if (ret) {
+ warn_report("Trying to enable capability %"PRIu64" of "
+ "KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 but failed. "
+ "Falling back to the legacy mode. ",
+ dirty_log_manual_caps);
+ s->manual_dirty_log_protect = 0;
+ }
+ }
+ }
+
+ return 0;
+}
+
static int kvm_init(MachineState *ms)
{
MachineClass *mc = MACHINE_GET_CLASS(ms);
@@ -2458,7 +2507,6 @@ static int kvm_init(MachineState *ms)
const KVMCapabilityInfo *missing_cap;
int ret;
int type;
- uint64_t dirty_log_manual_caps;
qemu_mutex_init(&kml_slots_lock);
@@ -2570,47 +2618,11 @@ static int kvm_init(MachineState *ms)
s->coalesced_pio = s->coalesced_mmio &&
kvm_check_extension(s, KVM_CAP_COALESCED_PIO);
- /*
- * Enable KVM dirty ring if supported, otherwise fall back to
- * dirty logging mode
- */
- ret = kvm_dirty_ring_init(s);
+ ret = kvm_setup_dirty_ring(s);
if (ret < 0) {
goto err;
}
- /*
- * KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is not needed when dirty ring is
- * enabled. More importantly, KVM_DIRTY_LOG_INITIALLY_SET will assume no
- * page is wr-protected initially, which is against how kvm dirty ring is
- * usage - kvm dirty ring requires all pages are wr-protected at the very
- * beginning. Enabling this feature for dirty ring causes data corruption.
- *
- * TODO: Without KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 and kvm clear dirty log,
- * we may expect a higher stall time when starting the migration. In the
- * future we can enable KVM_CLEAR_DIRTY_LOG to work with dirty ring too:
- * instead of clearing dirty bit, it can be a way to explicitly wr-protect
- * guest pages.
- */
- if (!s->kvm_dirty_ring_size) {
- dirty_log_manual_caps =
- kvm_check_extension(s, KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2);
- dirty_log_manual_caps &= (KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE |
- KVM_DIRTY_LOG_INITIALLY_SET);
- s->manual_dirty_log_protect = dirty_log_manual_caps;
- if (dirty_log_manual_caps) {
- ret = kvm_vm_enable_cap(s, KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2, 0,
- dirty_log_manual_caps);
- if (ret) {
- warn_report("Trying to enable capability %"PRIu64" of "
- "KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 but failed. "
- "Falling back to the legacy mode. ",
- dirty_log_manual_caps);
- s->manual_dirty_log_protect = 0;
- }
- }
- }
-
#ifdef KVM_CAP_VCPU_EVENTS
s->vcpu_events = kvm_check_extension(s, KVM_CAP_VCPU_EVENTS);
#endif
--
2.39.3

144
SOURCES/kvm-kvm-refactor-core-virtual-machine-creation-into-its-.patch
Unescape View File

@ -0,0 +1,144 @@
From 67180363bdc1898462f90e16c1909db7331cc5e2 Mon Sep 17 00:00:00 2001
From: Ani Sinha <anisinha@redhat.com>
Date: Thu, 8 Aug 2024 17:08:38 +0530
Subject: [PATCH 3/9] kvm: refactor core virtual machine creation into its own
function
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 285: KVM: Dynamic sized kvm memslots array
RH-Jira: RHEL-57685
RH-Acked-by: Juraj Marcin <None>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [2/7] a783111d9a2ef6590103543f1bd103bf90052872 (peterx/qemu-kvm)
Refactoring the core logic around KVM_CREATE_VM into its own separate function
so that it can be called from other functions in subsequent patches. There is
no functional change in this patch.
CC: pbonzini@redhat.com
CC: zhao1.liu@intel.com
Signed-off-by: Ani Sinha <anisinha@redhat.com>
Link: https://lore.kernel.org/r/20240808113838.1697366-1-anisinha@redhat.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 67388078da1cf6dac89e5a7c748cca3444d49690)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
accel/kvm/kvm-all.c | 89 ++++++++++++++++++++++++++++-----------------
1 file changed, 56 insertions(+), 33 deletions(-)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index 7432a54f39..d86d1b515a 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -2385,6 +2385,60 @@ uint32_t kvm_dirty_ring_size(void)
return kvm_state->kvm_dirty_ring_size;
}
+static int do_kvm_create_vm(MachineState *ms, int type)
+{
+ KVMState *s;
+ int ret;
+
+ s = KVM_STATE(ms->accelerator);
+
+ do {
+ ret = kvm_ioctl(s, KVM_CREATE_VM, type);
+ } while (ret == -EINTR);
+
+ if (ret < 0) {
+ error_report("ioctl(KVM_CREATE_VM) failed: %s", strerror(-ret));
+
+#ifdef TARGET_S390X
+ if (ret == -EINVAL) {
+ error_printf("Host kernel setup problem detected."
+ " Please verify:\n");
+ error_printf("- for kernels supporting the"
+ " switch_amode or user_mode parameters, whether");
+ error_printf(" user space is running in primary address space\n");
+ error_printf("- for kernels supporting the vm.allocate_pgste"
+ " sysctl, whether it is enabled\n");
+ }
+#elif defined(TARGET_PPC)
+ if (ret == -EINVAL) {
+ error_printf("PPC KVM module is not loaded. Try modprobe kvm_%s.\n",
+ (type == 2) ? "pr" : "hv");
+ }
+#endif
+ }
+
+ return ret;
+}
+
+static int find_kvm_machine_type(MachineState *ms)
+{
+ MachineClass *mc = MACHINE_GET_CLASS(ms);
+ int type;
+
+ if (object_property_find(OBJECT(current_machine), "kvm-type")) {
+ g_autofree char *kvm_type;
+ kvm_type = object_property_get_str(OBJECT(current_machine),
+ "kvm-type",
+ &error_abort);
+ type = mc->kvm_type(ms, kvm_type);
+ } else if (mc->kvm_type) {
+ type = mc->kvm_type(ms, NULL);
+ } else {
+ type = kvm_arch_get_default_type(ms);
+ }
+ return type;
+}
+
static int kvm_init(MachineState *ms)
{
MachineClass *mc = MACHINE_GET_CLASS(ms);
@@ -2467,45 +2521,14 @@ static int kvm_init(MachineState *ms)
}
s->as = g_new0(struct KVMAs, s->nr_as);
- if (object_property_find(OBJECT(current_machine), "kvm-type")) {
- g_autofree char *kvm_type = object_property_get_str(OBJECT(current_machine),
- "kvm-type",
- &error_abort);
- type = mc->kvm_type(ms, kvm_type);
- } else if (mc->kvm_type) {
- type = mc->kvm_type(ms, NULL);
- } else {
- type = kvm_arch_get_default_type(ms);
- }
-
+ type = find_kvm_machine_type(ms);
if (type < 0) {
ret = -EINVAL;
goto err;
}
- do {
- ret = kvm_ioctl(s, KVM_CREATE_VM, type);
- } while (ret == -EINTR);
-
+ ret = do_kvm_create_vm(ms, type);
if (ret < 0) {
- error_report("ioctl(KVM_CREATE_VM) failed: %s", strerror(-ret));
-
-#ifdef TARGET_S390X
- if (ret == -EINVAL) {
- error_printf("Host kernel setup problem detected."
- " Please verify:\n");
- error_printf("- for kernels supporting the"
- " switch_amode or user_mode parameters, whether");
- error_printf(" user space is running in primary address space\n");
- error_printf("- for kernels supporting the vm.allocate_pgste"
- " sysctl, whether it is enabled\n");
- }
-#elif defined(TARGET_PPC)
- if (ret == -EINVAL) {
- error_printf("PPC KVM module is not loaded. Try modprobe kvm_%s.\n",
- (type == 2) ? "pr" : "hv");
- }
-#endif
goto err;
}
--
2.39.3

132
SOURCES/kvm-kvm-replace-fprintf-with-error_report-printf-in-kvm_.patch
Unescape View File

@ -0,0 +1,132 @@
From 522e19dd84eb5c4d88b3b70193ee104f67a5b89d Mon Sep 17 00:00:00 2001
From: Ani Sinha <anisinha@redhat.com>
Date: Wed, 28 Aug 2024 18:15:39 +0530
Subject: [PATCH 2/9] kvm: replace fprintf with error_report()/printf() in
kvm_init()
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 285: KVM: Dynamic sized kvm memslots array
RH-Jira: RHEL-57685
RH-Acked-by: Juraj Marcin <None>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [1/7] 6c1230a6d5033d928817df9458938a675058e995 (peterx/qemu-kvm)
error_report() is more appropriate for error situations. Replace fprintf with
error_report() and error_printf() as appropriate. Some improvement in error
reporting also happens as a part of this change. For example:
From:
$ ./qemu-system-x86_64 --accel kvm
Could not access KVM kernel module: No such file or directory
To:
$ ./qemu-system-x86_64 --accel kvm
qemu-system-x86_64: --accel kvm: Could not access KVM kernel module: No such file or directory
CC: qemu-trivial@nongnu.org
CC: zhao1.liu@intel.com
CC: armbru@redhat.com
Reviewed-by: Zhao Liu <zhao1.liu@intel.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Ani Sinha <anisinha@redhat.com>
Link: https://lore.kernel.org/r/20240828124539.62672-1-anisinha@redhat.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 804dfbe3ef5e950328b162ae85741be2e228544f)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
accel/kvm/kvm-all.c | 40 ++++++++++++++++++----------------------
1 file changed, 18 insertions(+), 22 deletions(-)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index c7f1cc64b6..7432a54f39 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -2427,7 +2427,7 @@ static int kvm_init(MachineState *ms)
QLIST_INIT(&s->kvm_parked_vcpus);
s->fd = qemu_open_old(s->device ?: "/dev/kvm", O_RDWR);
if (s->fd == -1) {
- fprintf(stderr, "Could not access KVM kernel module: %m\n");
+ error_report("Could not access KVM kernel module: %m");
ret = -errno;
goto err;
}
@@ -2437,13 +2437,13 @@ static int kvm_init(MachineState *ms)
if (ret >= 0) {
ret = -EINVAL;
}
- fprintf(stderr, "kvm version too old\n");
+ error_report("kvm version too old");
goto err;
}
if (ret > KVM_API_VERSION) {
ret = -EINVAL;
- fprintf(stderr, "kvm version not supported\n");
+ error_report("kvm version not supported");
goto err;
}
@@ -2488,26 +2488,22 @@ static int kvm_init(MachineState *ms)
} while (ret == -EINTR);
if (ret < 0) {
- fprintf(stderr, "ioctl(KVM_CREATE_VM) failed: %d %s\n", -ret,
- strerror(-ret));
+ error_report("ioctl(KVM_CREATE_VM) failed: %s", strerror(-ret));
#ifdef TARGET_S390X
if (ret == -EINVAL) {
- fprintf(stderr,
- "Host kernel setup problem detected. Please verify:\n");
- fprintf(stderr, "- for kernels supporting the switch_amode or"
- " user_mode parameters, whether\n");
- fprintf(stderr,
- " user space is running in primary address space\n");
- fprintf(stderr,
- "- for kernels supporting the vm.allocate_pgste sysctl, "
- "whether it is enabled\n");
+ error_printf("Host kernel setup problem detected."
+ " Please verify:\n");
+ error_printf("- for kernels supporting the"
+ " switch_amode or user_mode parameters, whether");
+ error_printf(" user space is running in primary address space\n");
+ error_printf("- for kernels supporting the vm.allocate_pgste"
+ " sysctl, whether it is enabled\n");
}
#elif defined(TARGET_PPC)
if (ret == -EINVAL) {
- fprintf(stderr,
- "PPC KVM module is not loaded. Try modprobe kvm_%s.\n",
- (type == 2) ? "pr" : "hv");
+ error_printf("PPC KVM module is not loaded. Try modprobe kvm_%s.\n",
+ (type == 2) ? "pr" : "hv");
}
#endif
goto err;
@@ -2526,9 +2522,9 @@ static int kvm_init(MachineState *ms)
nc->name, nc->num, soft_vcpus_limit);
if (nc->num > hard_vcpus_limit) {
- fprintf(stderr, "Number of %s cpus requested (%d) exceeds "
- "the maximum cpus supported by KVM (%d)\n",
- nc->name, nc->num, hard_vcpus_limit);
+ error_report("Number of %s cpus requested (%d) exceeds "
+ "the maximum cpus supported by KVM (%d)",
+ nc->name, nc->num, hard_vcpus_limit);
exit(1);
}
}
@@ -2542,8 +2538,8 @@ static int kvm_init(MachineState *ms)
}
if (missing_cap) {
ret = -EINVAL;
- fprintf(stderr, "kvm does not support %s\n%s",
- missing_cap->name, upgrade_note);
+ error_report("kvm does not support %s", missing_cap->name);
+ error_printf("%s", upgrade_note);
goto err;
}
--
2.39.3

92
SOURCES/kvm-migration-Ensure-vmstate_save-sets-errp.patch
Unescape View File

@ -0,0 +1,92 @@
From 6be2f51c147df1ab1dd7c68c6b554512dfc05e6f Mon Sep 17 00:00:00 2001
From: Hanna Czenczek <hreitz@redhat.com>
Date: Tue, 15 Oct 2024 19:04:37 +0200
Subject: [PATCH 1/9] migration: Ensure vmstate_save() sets errp
RH-Author: Hanna Czenczek <hreitz@redhat.com>
RH-MergeRequest: 288: migration: Ensure vmstate_save() sets errp
RH-Jira: RHEL-63051
RH-Acked-by: Stefano Garzarella <sgarzare@redhat.com>
RH-Acked-by: German Maglione <None>
RH-Commit: [1/1] 4d5a65c294ae83a29db885e42fb3f2ca913c36f0 (hreitz/qemu-kvm-c-9-s)
migration/savevm.c contains some calls to vmstate_save() that are
followed by migrate_set_error() if the integer return value indicates an
error. migrate_set_error() requires that the `Error *` object passed to
it is set. Therefore, vmstate_save() is assumed to always set *errp on
error.
Right now, that assumption is not met: vmstate_save_state_v() (called
internally by vmstate_save()) will not set *errp if
vmstate_subsection_save() or vmsd->post_save() fail. Fix that by adding
an *errp parameter to vmstate_subsection_save(), and by generating a
generic error in case post_save() fails (as is already done for
pre_save()).
Without this patch, qemu will crash after vmstate_subsection_save() or
post_save() have failed inside of a vmstate_save() call (unless
migrate_set_error() then happen to discard the new error because
s->error is already set). This happens e.g. when receiving the state
from a virtio-fs back-end (virtiofsd) fails.
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Link: https://lore.kernel.org/r/20241015170437.310358-1-hreitz@redhat.com
Signed-off-by: Peter Xu <peterx@redhat.com>
(cherry picked from commit 37dfcba1a04989830c706f9cbc00450e5d3a7447)
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
---
migration/vmstate.c | 13 ++++++++-----
1 file changed, 8 insertions(+), 5 deletions(-)
diff --git a/migration/vmstate.c b/migration/vmstate.c
index ff5d589a6d..fa002b24e8 100644
--- a/migration/vmstate.c
+++ b/migration/vmstate.c
@@ -22,7 +22,8 @@
#include "trace.h"
static int vmstate_subsection_save(QEMUFile *f, const VMStateDescription *vmsd,
- void *opaque, JSONWriter *vmdesc);
+ void *opaque, JSONWriter *vmdesc,
+ Error **errp);
static int vmstate_subsection_load(QEMUFile *f, const VMStateDescription *vmsd,
void *opaque);
@@ -441,12 +442,13 @@ int vmstate_save_state_v(QEMUFile *f, const VMStateDescription *vmsd,
json_writer_end_array(vmdesc);
}
- ret = vmstate_subsection_save(f, vmsd, opaque, vmdesc);
+ ret = vmstate_subsection_save(f, vmsd, opaque, vmdesc, errp);
if (vmsd->post_save) {
int ps_ret = vmsd->post_save(opaque);
- if (!ret) {
+ if (!ret && ps_ret) {
ret = ps_ret;
+ error_setg(errp, "post-save failed: %s", vmsd->name);
}
}
return ret;
@@ -518,7 +520,8 @@ static int vmstate_subsection_load(QEMUFile *f, const VMStateDescription *vmsd,
}
static int vmstate_subsection_save(QEMUFile *f, const VMStateDescription *vmsd,
- void *opaque, JSONWriter *vmdesc)
+ void *opaque, JSONWriter *vmdesc,
+ Error **errp)
{
const VMStateDescription * const *sub = vmsd->subsections;
bool vmdesc_has_subsections = false;
@@ -546,7 +549,7 @@ static int vmstate_subsection_save(QEMUFile *f, const VMStateDescription *vmsd,
qemu_put_byte(f, len);
qemu_put_buffer(f, (uint8_t *)vmsdsub->name, len);
qemu_put_be32(f, vmsdsub->version_id);
- ret = vmstate_save_state(f, vmsdsub, opaque, vmdesc);
+ ret = vmstate_save_state_with_err(f, vmsdsub, opaque, vmdesc, errp);
if (ret) {
return ret;
}
--
2.39.3

37
SPECS/qemu-kvm.spec
Unescape View File

@ -143,7 +143,7 @@ Obsoletes: %{name}-block-ssh <= %{epoch}:%{version} \
Summary: QEMU is a machine emulator and virtualizer
Name: qemu-kvm
Version: 9.1.0
Release: 4%{?rcrel}%{?dist}%{?cc_suffix}
Release: 5%{?rcrel}%{?dist}%{?cc_suffix}
# Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
# Epoch 15 used for RHEL 8
# Epoch 17 used for RHEL 9 (due to release versioning offset in RHEL 8.5)
@ -320,6 +320,22 @@ Patch72: kvm-target-i386-Introduce-GraniteRapids-v2-model.patch
# For RHEL-30315 - [Intel 10.0 FEAT] [GNR] Virt-QEMU: Add AVX10.1 instruction support
# For RHEL-45110 - [Intel 10.0 FEAT] [CWF][DMR] Virt-QEMU: Advertise new instructions SHA2-512NI, SM3, and SM4
Patch73: kvm-target-i386-add-sha512-sm3-sm4-feature-bits.patch
# For RHEL-63051 - qemu crashed after killed virtiofsd during migration
Patch74: kvm-migration-Ensure-vmstate_save-sets-errp.patch
# For RHEL-57685 - Bad migration performance when performing vGPU VM live migration
Patch75: kvm-kvm-replace-fprintf-with-error_report-printf-in-kvm_.patch
# For RHEL-57685 - Bad migration performance when performing vGPU VM live migration
Patch76: kvm-kvm-refactor-core-virtual-machine-creation-into-its-.patch
# For RHEL-57685 - Bad migration performance when performing vGPU VM live migration
Patch77: kvm-accel-kvm-refactor-dirty-ring-setup.patch
# For RHEL-57685 - Bad migration performance when performing vGPU VM live migration
Patch78: kvm-KVM-Dynamic-sized-kvm-memslots-array.patch
# For RHEL-57685 - Bad migration performance when performing vGPU VM live migration
Patch79: kvm-KVM-Define-KVM_MEMSLOTS_NUM_MAX_DEFAULT.patch
# For RHEL-57685 - Bad migration performance when performing vGPU VM live migration
Patch80: kvm-KVM-Rename-KVMMemoryListener.nr_used_slots-to-nr_slo.patch
# For RHEL-57685 - Bad migration performance when performing vGPU VM live migration
Patch81: kvm-KVM-Rename-KVMState-nr_slots-to-nr_slots_max.patch
%if %{have_clang}
BuildRequires: clang
@ -357,6 +373,8 @@ BuildRequires: librbd-devel
# We need both because the 'stap' binary is probed for by configure
BuildRequires: systemtap
BuildRequires: systemtap-sdt-devel
# Required as we use dtrace for trace backend
BuildRequires: /usr/bin/dtrace
# For VNC PNG support
BuildRequires: libpng-devel
# For virtiofs
@ -1384,6 +1402,23 @@ useradd -r -u 107 -g qemu -G kvm -d / -s /sbin/nologin \
%endif
%changelog
* Tue Nov 19 2024 Miroslav Rezanina <mrezanin@redhat.com> - 9.1.0-5
- kvm-migration-Ensure-vmstate_save-sets-errp.patch [RHEL-63051]
- kvm-kvm-replace-fprintf-with-error_report-printf-in-kvm_.patch [RHEL-57685]
- kvm-kvm-refactor-core-virtual-machine-creation-into-its-.patch [RHEL-57685]
- kvm-accel-kvm-refactor-dirty-ring-setup.patch [RHEL-57685]
- kvm-KVM-Dynamic-sized-kvm-memslots-array.patch [RHEL-57685]
- kvm-KVM-Define-KVM_MEMSLOTS_NUM_MAX_DEFAULT.patch [RHEL-57685]
- kvm-KVM-Rename-KVMMemoryListener.nr_used_slots-to-nr_slo.patch [RHEL-57685]
- kvm-KVM-Rename-KVMState-nr_slots-to-nr_slots_max.patch [RHEL-57685]
- kvm-Require-new-dtrace-package.patch [RHEL-67899]
- Resolves: RHEL-63051
(qemu crashed after killed virtiofsd during migration)
- Resolves: RHEL-57685
(Bad migration performance when performing vGPU VM live migration )
- Resolves: RHEL-67899
(Failed to build qemu-kvm due to missing dtrace [rhel-10.0])
* Tue Nov 12 2024 Miroslav Rezanina <mrezanin@redhat.com> - 9.1.0-4.el10
- kvm-accel-kvm-check-for-KVM_CAP_READONLY_MEM-on-VM.patch [RHEL-58928]
- kvm-hw-s390x-ipl-Provide-more-memory-to-the-s390-ccw.img.patch [RHEL-58153]

Write Preview
Loading…
Cancel
Save