On Thu, Nov 15, 2012 at 9:09 PM, Anthony Liguori <aligu...@us.ibm.com> wrote:
> Stefan Hajnoczi <stefa...@redhat.com> writes:
>
>> The virtio-blk-data-plane cannot access memory using the usual QEMU
>> functions since it executes outside the global mutex and the memory APIs
>> are this time are not thread-safe.
>>
>> This patch introduces a virtqueue module based on the kernel's vhost
>> vring code. The trick is that we map guest memory ahead of time and
>> access it cheaply outside the global mutex.
>>
>> Once the hardware emulation code can execute outside the global mutex it
>> will be possible to drop this code.
>>
>> Signed-off-by: Stefan Hajnoczi <stefa...@redhat.com>
>> ---
>> hw/Makefile.objs | 2 +-
>> hw/dataplane/Makefile.objs | 3 +
>> hw/dataplane/vring.c | 321
>> +++++++++++++++++++++++++++++++++++++++++++++
>> hw/dataplane/vring.h | 54 ++++++++
>> trace-events | 3 +
>> 5 files changed, 382 insertions(+), 1 deletion(-)
>> create mode 100644 hw/dataplane/Makefile.objs
>> create mode 100644 hw/dataplane/vring.c
>> create mode 100644 hw/dataplane/vring.h
>>
>> diff --git a/hw/Makefile.objs b/hw/Makefile.objs
>> index af4ab0c..da8ef0c 100644
>> --- a/hw/Makefile.objs
>> +++ b/hw/Makefile.objs
>> @@ -1,4 +1,4 @@
>> -common-obj-y = usb/ ide/
>> +common-obj-y = usb/ ide/ dataplane/
>> common-obj-y += loader.o
>> common-obj-$(CONFIG_VIRTIO) += virtio-console.o
>> common-obj-$(CONFIG_VIRTIO_PCI) += virtio-pci.o
>> diff --git a/hw/dataplane/Makefile.objs b/hw/dataplane/Makefile.objs
>> new file mode 100644
>> index 0000000..b58544f
>> --- /dev/null
>> +++ b/hw/dataplane/Makefile.objs
>> @@ -0,0 +1,3 @@
>> +ifeq ($(CONFIG_VIRTIO), y)
>> +common-obj-$(CONFIG_VIRTIO_BLK_DATA_PLANE) += vring.o
>> +endif
>> diff --git a/hw/dataplane/vring.c b/hw/dataplane/vring.c
>> new file mode 100644
>> index 0000000..6aacce8
>> --- /dev/null
>> +++ b/hw/dataplane/vring.c
>> @@ -0,0 +1,321 @@
>> +/* Copyright 2012 Red Hat, Inc.
>> + * Copyright IBM, Corp. 2012
>> + *
>> + * Based on Linux vhost code:
>> + * Copyright (C) 2009 Red Hat, Inc.
>> + * Copyright (C) 2006 Rusty Russell IBM Corporation
>> + *
>> + * Author: Michael S. Tsirkin <m...@redhat.com>
>> + * Stefan Hajnoczi <stefa...@redhat.com>
>> + *
>> + * Inspiration, some code, and most witty comments come from
>> + * Documentation/virtual/lguest/lguest.c, by Rusty Russell
>> + *
>> + * This work is licensed under the terms of the GNU GPL, version 2.
>> + */
>> +
>> +#include "trace.h"
>> +#include "hw/dataplane/vring.h"
>> +
>> +/* Map target physical address to host address
>> + */
>> +static inline void *phys_to_host(Vring *vring, hwaddr phys)
>> +{
>> + /* Adjust for 3.6-4 GB PCI memory range */
>> + if (phys >= 0x100000000) {
>> + phys -= 0x100000000 - 0xe0000000;
>> + } else if (phys >= 0xe0000000) {
>> + fprintf(stderr, "phys_to_host bad physical address in "
>> + "PCI range %#lx\n", phys);
>> + exit(1);
>> + }
>> + return vring->phys_mem_zero_host_ptr + phys;
>> +}
>
> This is too dirty to merge. This code should use a slot mechanism
> similar to what vhost uses. Then it has a plausible chance of working
> on other architectures.
[...]
> You should be able to use a MemoryListener to keep the memory slot table
> up to date. You'll need to use a mutex to protect it but again since
> it's mostly uncontended, that shouldn't be a problem.
Anthony, Paolo,
You're right, we should use MemoryListener. Will fix for v2.
Stefan
