Re: [Qemu-devel] [PULL v2 00/44] Block patches

[Stefan Hajnoczi]

On Mon, Jan 12, 2015 at 09:43:28PM +0000, Peter Maydell wrote:
> On 12 January 2015 at 16:39, Stefan Hajnoczi <address@hidden> wrote:
> > The following changes since commit 64ea8038ffbf703dcd438a108d2d5499c8ff95d9:
> >
> >   Merge remote-tracking branch 
> > 'remotes/awilliam/tags/vfio-update-20150109.0' into staging (2015-01-10 
> > 22:29:09 +0000)
> >
> > are available in the git repository at:
> >
> >   git://github.com/stefanha/qemu.git tags/block-pull-request
> >
> > for you to fetch changes up to bed6f0227ef9da1505758a63bd2b6cd2a45874c6:
> >
> >   NVMe: Set correct VS Value for 1.1 Compliant Controllers (2015-01-12 
> > 16:30:28 +0000)
> 
> I'm afraid this fails 'make check' on at least one config and
> generates a lot of ugly warnings on several.
> 
> (1) Ugly warning (seems to happen everywhere):
> 
> TEST: tests/virtio-blk-test... (pid=28666)
>   /arm/virtio/blk/mmio/basic:
> WARNING: Image format was not specified for '/tmp/qtest.q0ywpl' and
> probing guessed raw.
>          Automatically detecting the format is dangerous for raw
> images, write operations on block 0 will be restricted.
>          Specify the 'raw' format explicitly to remove the restrictions.
> OK

I am dropping the series because of the timeout you discovered below.

The fix for the warning is easy and I will let Marc know.

> (2) Apparent errors on OSX when trying to use "/dev/null" as an
> image file. Weirdly these don't actually cause "make check" to fail,
> which suggests something's not propagating an error correctly.
> 
> GTESTER check-qtest-i386
> blkdebug: Suspended request 'A'
> blkdebug: Resuming request 'A'
> qemu-system-i386: -drive if=none,id=drive0,file=/dev/null,format=raw:
> could not open disk image /dev/null: Could not refresh total sector
> count: Operation not supported by device

I have dropped "block/raw-posix.c: Fixes raw_getlength() on Mac OS X so
that it reports the correct length of a real CD".  It breaks non-CD-ROM
character devices like /dev/null.

> (3) Actual failure, x86_64 host:
> 
> TEST: tests/virtio-blk-test... (pid=16564)
>   /arm/virtio/blk/mmio/basic:
> WARNING: Image format was not specified for '/tmp/qtest.4s4Naf' and
> probing guessed raw.
>          Automatically detecting the format is dangerous for raw
> images, write operations on block 0 will be restricted.
>          Specify the 'raw' format explicitly to remove the restrictions.
> **
> ERROR:/home/petmay01/linaro/qemu-for-merges/tests/libqos/virtio.c:91:qvirtio_wait_queue_isr:
> assertion failed: (g_get_monotonic_time() - start_time <= timeout_us)
> FAIL
> GTester: last random seed: R02S2b2c6e34351d711919b868b92fde3543
> (pid=16570)
> FAIL: tests/virtio-blk-test
> 
> This might be an intermittent failure; I think my x86-64 gcc
> build passed the first time and failed on a rerun. The clang
> build also failed this.

This is strange.  The timeout is 30 seconds so something must be very
wrong.

I haven't been able to reproduce it yet but will ask Marc to
investigate.

Stefan

pgpL9WdbHr68N.pgp
Description: PGP signature