Re: [Qemu-devel] [PATCH] block: add the support to drain throttled requests

[Stefan Hajnoczi <stefanha@gmail.com>]

On Tue, Mar 27, 2012 at 12:29:09PM +0800, Zhi Yong Wu wrote:
> On Mon, Mar 26, 2012 at 10:21 PM, Stefan Hajnoczi <stefanha@gmail.com> wrote:
> > On Tue, Mar 20, 2012 at 11:44 AM, Stefan Hajnoczi
> > <stefanha@linux.vnet.ibm.com> wrote:
> >> On Tue, Mar 20, 2012 at 10:58:10AM +0100, Kevin Wolf wrote:
> >>> Am 20.03.2012 10:47, schrieb Paolo Bonzini:
> >>> > Il 20/03/2012 10:40, Zhi Yong Wu ha scritto:
> >>> >> HI, Kevin,
> >>> >>
> >>> >> We hope that I/O throttling can be shipped without known issue in QEMU
> >>> >> 1.1, so if you are available, can you give this patch some love?
> >>> >
> >>> > I'm sorry to say this, but I think I/O throttling is impossible to save.
> >>> >  As it is implemented now, it just cannot work in the presence of
> >>> > synchronous I/O, except at the cost of busy waiting with the global
> >>> > mutex taken.  See the message from Stefan yesterday.
> >>>
> >>> qemu_aio_flush() is busy waiting with the global mutex taken anyway, so
> >>> it doesn't change that much.
> >>
> >> Yesterday I only posted an analysis of the bug but here are some
> >> thoughts on how to move forward.  Throttling itself is not the problem.
> >> We've known that synchronous operations in the vcpu thread are a problem
> >> long before throttling.  This is just another reason to convert device
> >> emulation to use asynchronous interfaces.
> >>
> >> Here is the list of device models that perform synchronous block I/O:
> >> hw/fdc.c
> >> hw/ide/atapi.c
> >> hw/ide/core.c
> >> hw/nand.c
> >> hw/onenand.c
> >> hw/pflash_cfi01.c
> >> hw/pflash_cfi02.c
> >> hw/sd.c
> >>
> >> Zhi Hui Li is working on hw/fdc.c and recently sent a patch.
> >>
> >> I think it's too close to QEMU 1.1 to convert all the remaining devices
> >> and test them properly before the soft-freeze.  But it's probably
> >> possible to convert IDE before the soft-freeze.
> >>
> >> In the meantime we could add this to bdrv_rw_co():
> >>
> >> if (bs->io_limits_enabled) {
> >>    fprintf(stderr, "Disabling I/O throttling on '%s' due "
> >>            "to synchronous I/O\n", bdrv_get_device_name(bs));
> >>    bdrv_io_limits_disable(bs);
> >> }
> >>
> >> It's not pretty but tells the user there is an issue and avoids
> >> deadlocking.
> >
> > No one has commented on this suggestion.  I think leaving a known hang
> > in QEMU 1.1 is undesirable.  Better to have this warning and disable
> > throttling in the case we cannot support right now.
> IDE is using both sync API and async API when guest boot.

Yes, a warning patch must handle the synchronous I/O case during startup
(guessing disk geometry), which I think happens even for virtio-blk.

Stefan