If we repeatedly fail to (fake) offline memory, we won't be sending
any unplug requests to the device. However, we only check if the config
changed when sending such (un)plug requests.
So we could end up trying for a long time to offline memory, even though
the config changed already and we're not supposed to unplug memory
anymore.
Let's optimize for that case, identified while testing the
offline_and_remove() memory timeout and simulating it repeatedly running
into the timeout.
Signed-off-by: David Hildenbrand <david@xxxxxxxxxx>
---
drivers/virtio/virtio_mem.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/drivers/virtio/virtio_mem.c b/drivers/virtio/virtio_mem.c
index 7468b4a907e3..247fb3e0ce61 100644
--- a/drivers/virtio/virtio_mem.c
+++ b/drivers/virtio/virtio_mem.c
@@ -1922,6 +1922,10 @@ static int virtio_mem_sbm_unplug_sb_online(struct virtio_mem *vm,
unsigned long start_pfn;
int rc;
+ /* Stop fake offlining attempts if the config changed. */
+ if (atomic_read(&vm->config_changed))
+ return -EAGAIN;
+
start_pfn = PFN_DOWN(virtio_mem_mb_id_to_phys(mb_id) +
sb_id * vm->sbm.sb_size);
@@ -2233,6 +2237,10 @@ static int virtio_mem_bbm_unplug_request(struct virtio_mem *vm, uint64_t diff)
virtio_mem_bbm_for_each_bb_rev(vm, bb_id, VIRTIO_MEM_BBM_BB_ADDED) {
cond_resched();
+ /* Stop (fake) offlining attempts if the config changed. */
+ if (atomic_read(&vm->config_changed))
+ return -EAGAIN;
+
/*
* As we're holding no locks, these checks are racy,
* but we don't care.
--
2.40.1
_______________________________________________
Virtualization mailing list
Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
