Mark McLoughlin wrote:
> On Wed, 2009-05-27 at 13:40 -0400, Gregory Haskins wrote:
>
>> Mark McLoughlin wrote:
>>
>>> On Wed, 2009-05-27 at 15:11 +0300, Avi Kivity wrote:
>>>
>>>
>>>
>>>> Multiple cookies on the same address are required by virtio. You can't
>>>> mux since the data doesn't go anywhere.
>>>>
>>>> Virtio can survive by checking all rings on a notify, and we can later
>>>> add a mechanism that has a distinct address for each ring, but let's see
>>>> if we can cope with multiple cookies. Mark?
>>>>
>>>>
>>> Trying to catch up, but you're talking about replacing virtio-pci
>>> QUEUE_NOTIFY handling with iosignalfd ?
>>>
>>> For a perfect replacement, what you really need is to be able to
>>> register multiple cookies per address range, but only have them trigger
>>> if the written data matches a provided value.
>>>
>>>
>> Hmm..thats an interesting idea. To date, the "cookie" has really been
>> for identifying the proper range selected for deassignment. I never
>> thought of using it as an actual trigger value at run-time.
>>
>>
>>> If the data is lost, virtio has no way of knowing which queue is being
>>> notified, so we either end up with per-device, rather than per-queue,
>>> notifications (probably not too bad for net, at least) or a different
>>> notify address per queue (limiting the number of queues per device).
>>>
>>>
>> The addr-per-queue is how I was envisioning it, but the trigger value
>> concept hadn't occurred to me. I could make this an option during
>> assignment (e.g. "COOKIE" flag means only trigger on writes of the
>> provided cookie, otherwise trigger on any write). Sound good?
>>
>
> Ah, I'd been thinking of the trigger data being provided separately to
> the cookie.
>
> The virtio ABI is fixed, so we couldn't e.g. have the guest use a cookie
> to identify a queue - it's just going to continue using a per-device
> queue number. So, if the cookie was also the trigger, we'd need an
> eventfd per device.
>
> And if this was a device where the guest writes similar values to
> multiple addresses, you'd need an eventfd per address.
>
>
Hi Mark,
So with the v5 release of iosignalfd, we now have the notion of a
"trigger", the API of which is as follows:
-----------------------
/*!
* \brief Assign an eventfd to an IO port (PIO or MMIO)
*
* Assigns an eventfd based file-descriptor to a specific PIO or MMIO
* address range. Any guest writes to the specified range will generate
* an eventfd signal.
*
* A data-match pointer can be optionally provided in "trigger" and only
* writes which match this value exactly will generate an event. The length
* of the trigger is established by the length of the overall IO range, and
* therefore must be in a natural byte-width for the IO routines of your
* particular architecture (e.g. 1, 2, 4, or 8 bytes on x86_64).
*
* \param kvm Pointer to the current kvm_context
* \param addr The IO address
* \param len The length of the IO region at the address
* \param fd The eventfd file-descriptor
* \param trigger A optional pointer providing data-match token
* \param flags FLAG_PIO: PIO, else MMIO
*/
int kvm_assign_iosignalfd(kvm_context_t kvm, unsigned long addr, size_t len,
int fd, void *trigger, int flags);
-----------------
in the kvm-eventfd test harness, I create three unique eventfd handles,
and do the following:
-------------------
unsigned char matchA = 0xa5, matchB = 0x42;
kvm_assign_iosignalfd(kvm_context, addr, 1, fd[0], NULL, 0);
kvm_assign_iosignalfd(kvm_context, addr, 1, fd[1], &matchA, 0);
kvm_assign_iosignalfd(kvm_context, addr, 1, fd[2], &matchB, 0);
-------------------
In otherwords, I register a "NULL" trigger (wildcarded) on the first
fd. The second has a data-match trigger of 0xa5, and the third has
0x42. All three of these eventfd's map to the same mmio address with a
width of 1 byte.
I also fork a task which selects all three fds, and will print out the
eventfd "count" value when tripped.
Then, in the guest, I do:
----------------------
iowrite8(0, iosignalfd_mmio);
iowrite8(0xa5, iosignalfd_mmio);
iowrite8(0x42, iosignalfd_mmio);
-------------------
The result of which is:
IOSIGNALFD 0: event triggered with val 3
IOSIGNALFD 1: event triggered with val 1
IOSIGNALFD 2: event triggered with val 1
on the host, which is my expected outcome. Let me know if you do not
think this is sufficient to implement a solution to your virtio-pci design.
-Greg
Hi Greg,
On Wed, 2009-06-03 at 18:04 -0400, Gregory Haskins wrote:
> Hi Mark,
> So with the v5 release of iosignalfd, we now have the notion of a
> "trigger", the API of which is as follows:
>
> -----------------------
> /*!
> * \brief Assign an eventfd to an IO port (PIO or MMIO)
> *
> * Assigns an eventfd based file-descriptor to a specific PIO or MMIO
> * address range. Any guest writes to the specified range will generate
> * an eventfd signal.
> *
> * A data-match pointer can be optionally provided in "trigger" and only
> * writes which match this value exactly will generate an event. The length
> * of the trigger is established by the length of the overall IO range, and
> * therefore must be in a natural byte-width for the IO routines of your
> * particular architecture (e.g. 1, 2, 4, or 8 bytes on x86_64).
This looks like it'll work fine for virtio-pci.
Thanks,
Mark.