| Message ID | 1365698186-27355-6-git-send-email-will.deacon@arm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
diff --git a/tools/kvm/include/kvm/virtio-mmio.h b/tools/kvm/include/kvm/virtio-mmio.h index 983c8fc..4d6a671 100644 --- a/tools/kvm/include/kvm/virtio-mmio.h +++ b/tools/kvm/include/kvm/virtio-mmio.h @@ -4,7 +4,7 @@ #include <linux/types.h> #include <linux/virtio_mmio.h> -#define VIRTIO_MMIO_MAX_VQ 3 +#define VIRTIO_MMIO_MAX_VQ 32 #define VIRTIO_MMIO_MAX_CONFIG 1 #define VIRTIO_MMIO_IO_SIZE 0x200
Commit 4d789d4a2050 ("kvm tools: Increase amount of possible interrupts per PCI device") increased the maximum amount of virtio queues for the PCI transport, but neglected to do the same for MMIO. This patch makes the same change for virtio-mmio. Cc: Sasha Levin <sasha.levin@oracle.com> Reported-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com> --- Sasha -- although this fixes a SEGV when booting a guest which tries to use virtio-net over MMIO, it *drastically* increases the memory footprint of a guest kernel to unacceptable levels (38MB -> 170MB). tools/kvm/include/kvm/virtio-mmio.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)