On 12.09.2017 16:44, Paolo Bonzini wrote:
> The purpose of this series is to separate the "check" sources from
> the tests. After these patches, common.config is reduced to simple
> shell initialization, and common.rc is only included by the tests.
>
> Along the way, a lot of dead code is removed
On 09/12/2017 04:26 PM, Paolo Bonzini wrote:
>
>
>>> +then
>>> +if [ -x "$build_iotests/qemu" ]; then
>>> +export QEMU_PROG="$build_iotests/qemu"
>>> +elif [ -x "$build_root/$arch-softmmu/qemu-system-$arch" ]; then
>>> +export QEMU_PROG="$build_root/$arch-softmmu/qemu-syst
On 09/12/2017 09:44 AM, Paolo Bonzini wrote:
> "check" is full of qemu-iotests--specific details. Separating it
> from "common" does not make much sense anymore.
>
> Signed-off-by: Paolo Bonzini
> ---
> tests/qemu-iotests/check | 533 +++-
> tests/qemu-i
On 09/12/2017 09:44 AM, Paolo Bonzini wrote:
> The variable is almost unused, and one of the two uses is actually
> uninitialized.
>
> Signed-off-by: Paolo Bonzini
> ---
> tests/qemu-iotests/check | 5 +
> tests/qemu-iotests/common.rc | 2 +-
> 2 files changed, 2 insertions(+), 5 deletio
On 09/12/2017 09:44 AM, Paolo Bonzini wrote:
> The variable is used in "common" but defined only after the file
> is sourced.
>
> Signed-off-by: Paolo Bonzini
> ---
> tests/qemu-iotests/check | 2 --
> tests/qemu-iotests/common | 2 ++
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
Revi
> > +then
> > +if [ -x "$build_iotests/qemu" ]; then
> > +export QEMU_PROG="$build_iotests/qemu"
> > +elif [ -x "$build_root/$arch-softmmu/qemu-system-$arch" ]; then
> > +export QEMU_PROG="$build_root/$arch-softmmu/qemu-system-$arch"
> > +else
> > +pushd "$buil
On 09/12/2017 09:44 AM, Paolo Bonzini wrote:
> Split "check" parts from tests part.
>
> For the directory setup, the actual computation of directories goes
> in "check", while the sanity checks go in the tests.
>
> Signed-off-by: Paolo Bonzini
> ---
> tests/qemu-iotests/common| 24 +
On 09/12/2017 09:44 AM, Paolo Bonzini wrote:
> It only provides functions used by the test programs.
>
> Signed-off-by: Paolo Bonzini
> ---
> tests/qemu-iotests/check | 6 --
> tests/qemu-iotests/common.rc | 13 +
> 2 files changed, 5 insertions(+), 14 deletions(-)
>
Revie
On 09/12/2017 09:44 AM, Paolo Bonzini wrote:
> These are never used by "check", with one exception that does not need
> $QEMU_OPTIONS. Keep them in common.rc, which will be soon included only
> by the tests.
>
> Signed-off-by: Paolo Bonzini
> ---
> tests/qemu-iotests/039.out | 10 +++---
>
On 09/12/2017 09:44 AM, Paolo Bonzini wrote:
> Instead of ./check failing when a binary is missing, we try each test
> case now and each one fails with tons of test case diffs. Also, all the
> variables were initialized by "check" prior to "common" being sourced,
> and then (uselessly) checked for
On 09/12/2017 09:44 AM, Paolo Bonzini wrote:
> Some functions in common.rc are never used by the tests. Move
> them out of that file and into common, which is already included
> only by "check".
>
> Code that actually *is* common to "check" and tests can be placed in
> common.config.
>
> Signed-
On 09/12/2017 09:44 AM, Paolo Bonzini wrote:
> Signed-off-by: Paolo Bonzini
> ---
> tests/qemu-iotests/check | 4 ++--
> tests/qemu-iotests/common| 2 +-
> tests/qemu-iotests/common.config | 3 ---
> 3 files changed, 3 insertions(+), 6 deletions(-)
Reviewed-by: Eric Blake
--
E
Both callers already had bytes available, but were scaling to
sectors. Move the scaling to internal code. In the case of
bdrv_aligned_pwritev(), we are now passing the exact offset
rather than a rounded sector-aligned value, but that's okay
as long as dirty bitmap widens start/bytes to granularit
Now that we have adjusted the majority of the calls this function
makes to be byte-based, it is easier to read the code if it makes
passes over the image using bytes rather than sectors.
Signed-off-by: Eric Blake
Reviewed-by: John Snow
Reviewed-by: Vladimir Sementsov-Ogievskiy
---
v7: rebase t
On 09/12/2017 09:44 AM, Paolo Bonzini wrote:
> This includes shell function, shell variables and command line options
> (randomize.awk does not exist).
>
> Signed-off-by: Paolo Bonzini
> ---
> tests/qemu-iotests/check | 28 -
> tests/qemu-iotests/common| 23 --
Now that we have adjusted the majority of the calls this function
makes to be byte-based, it is easier to read the code if it makes
passes over the image using bytes rather than sectors.
Signed-off-by: Eric Blake
Reviewed-by: John Snow
Reviewed-by: Vladimir Sementsov-Ogievskiy
---
v5: no chang
Now that we have adjusted the majority of the calls this function
makes to be byte-based, it is easier to read the code if it makes
passes over the image using bytes rather than sectors.
Signed-off-by: Eric Blake
Reviewed-by: John Snow
---
v6: no change
v5: rebase to earlier changes
v2-v4: no c
This is new code, but it is easier to read if it makes passes over
the image using bytes rather than sectors (and will get easier in
the future when bdrv_get_block_status is converted to byte-based).
Signed-off-by: Eric Blake
Reviewed-by: John Snow
---
v7: tweak constant given to MIN (no semant
Some of the callers were already scaling bytes to sectors; others
can be easily converted to pass byte offsets, all in our shift
towards a consistent byte interface everywhere. Making the change
will also make it easier to write the hold-out callers to use byte
rather than sectors for their iterat
Now that all callers are using byte-based interfaces, there's no
reason for our internal hbitmap to remain with sector-based
granularity. It also simplifies our internal scaling, since we
already know that hbitmap widens requests out to granularity
boundaries.
Signed-off-by: Eric Blake
Reviewed-
Half the callers were already scaling bytes to sectors; the other
half can eventually be simplified to use byte iteration. Both
callers were already using the result as a bool, so make that
explicit. Making the change also makes it easier for a future
dirty-bitmap patch to offload scaling over to
All callers to bdrv_dirty_iter_new() passed 0 for their initial
starting point, drop that parameter.
Most callers to bdrv_set_dirty_iter() were scaling a byte offset to
a sector number; the exception qcow2-bitmap will be converted later
to use byte rather than sector iteration. Move the scaling t
Thanks to recent cleanups, most callers were scaling a return value
of sectors into bytes (the exception, in qcow2-bitmap, will be
converted to byte-based iteration later). Update the interface to
do the scaling internally instead.
In qcow2-bitmap, the code was specifically checking for an error
We are still using an internal hbitmap that tracks a size in sectors,
with the granularity scaled down accordingly, because it lets us
use a shortcut for our iterators which are currently sector-based.
But there's no reason we can't track the dirty bitmap size in bytes,
since it is (mostly) an inte
We are gradually converting to byte-based interfaces, as they are
easier to reason about than sector-based. Change the qcow2 bitmap
helper function sectors_covered_by_bitmap_cluster(), renaming it
to bytes_covered_by_bitmap_cluster() in the process.
Signed-off-by: Eric Blake
Reviewed-by: John Sn
Thanks to recent cleanups, all callers were scaling a return value
of sectors into bytes; do the scaling internally instead.
Signed-off-by: Eric Blake
---
v7: fix one more trace caller [Kevin]
v4-v6: no change
v3: no change, add R-b
v2: no change
---
block/dirty-bitmap.c | 4 ++--
block/mirror
We've previously fixed several places where we failed to account
for possible errors from bdrv_nb_sectors(). Fix another one by
making bdrv_dirty_bitmap_truncate() report the error rather then
silently resizing bitmaps to -1. Then adjust the sole caller
bdrv_truncate() to both reduce the likeliho
We had several functions that no one is currently using, and which
use sector-based interfaces. I'm trying to convert towards byte-based
interfaces, so it's easier to just drop the unused functions:
bdrv_dirty_bitmap_get_meta
bdrv_dirty_bitmap_get_meta_locked
bdrv_dirty_bitmap_reset_meta
bdrv_dir
Right now, the dirty-bitmap code exposes the fact that we use
a scale of sector granularity in the underlying hbitmap to anything
that wants to serialize a dirty bitmap. It's nicer to uniformly
expose bytes as our dirty-bitmap interface, matching the previous
change to bitmap size. The only calle
When subdividing a bitmap serialization, the code in hbitmap.c
enforces that start/count parameters are aligned (except that
count can end early at end-of-bitmap). We exposed this required
alignment through bdrv_dirty_bitmap_serialization_align(), but
forgot to actually check that we comply with i
We're already reporting bytes for bdrv_dirty_bitmap_granularity();
mixing bytes and sectors in our return values is a recipe for
confusion. A later cleanup will convert dirty bitmap internals
to be entirely byte-based, but in the meantime, we should report
the bitmap size in bytes.
The only exter
All callers of bdrv_img_create() pass in a size, or -1 to read the
size from the backing file. We then set that size as the QemuOpt
default, which means we will reuse that default rather than the
final parameter to qemu_opt_get_size() several lines later. But
it is rather confusing to read subseq
The only client of hbitmap_serialization_granularity() is dirty-bitmap's
bdrv_dirty_bitmap_serialization_align(). Keeping the two names consistent
is worthwhile, and the shorter name is more representative of what the
function returns (the required alignment to be used for start/count of
other ser
There are patches floating around to add NBD_CMD_BLOCK_STATUS,
but NBD wants to report status on byte granularity (even if the
reporting will probably be naturally aligned to sectors or even
much higher levels). I've therefore started the task of
converting our block status code to report at a byt
On 09/08/2017 09:04 AM, Eric Blake wrote:
>>> void bdrv_dirty_bitmap_truncate(BlockDriverState *bs)
>>> {
>>> BdrvDirtyBitmap *bitmap;
>>> -uint64_t size = bdrv_nb_sectors(bs);
>>> +int64_t size = bdrv_getlength(bs);
>>>
>>> +assert(size >= 0);
>>
>> How can you assert that ther
On 12/09/2017 18:30, Philippe Mathieu-Daudé wrote:
> Hi Paolo,
>
> Hmm did you just resend your v1? Except the cover the patches don't have
> "v2" and I remember reviewing this one (also Eric Blake replied with his
> R-b).
I didn't include the R-bs, sorry (most of the later patches had
non-trivia
Hi Paolo,
Hmm did you just resend your v1? Except the cover the patches don't have
"v2" and I remember reviewing this one (also Eric Blake replied with his
R-b).
On 09/12/2017 11:44 AM, Paolo Bonzini wrote:
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/check | 4 ++--
tests
Am 12.09.2017 um 18:05 hat Cornelia Huck geschrieben:
> On Fri, 8 Sep 2017 17:55:29 +0200
> Thomas Huth wrote:
>
> > On 08.09.2017 13:54, Kevin Wolf wrote:
> > > Am 08.09.2017 um 13:24 hat Cornelia Huck geschrieben:
> > >> On Fri, 8 Sep 2017 13:04:25 +0200
> > >> Kevin Wolf wrote:
> > >>
> >
On Fri, 8 Sep 2017 17:55:29 +0200
Thomas Huth wrote:
> On 08.09.2017 13:54, Kevin Wolf wrote:
> > Am 08.09.2017 um 13:24 hat Cornelia Huck geschrieben:
> >> On Fri, 8 Sep 2017 13:04:25 +0200
> >> Kevin Wolf wrote:
> >>
> >>> Am 05.09.2017 um 17:16 hat Cornelia Huck geschrieben:
> The
These are never used by "check", with one exception that does not need
$QEMU_OPTIONS. Keep them in common.rc, which will be soon included only
by the tests.
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/039.out | 10 +++---
tests/qemu-iotests/061.out | 4 +--
tests/qemu-iotes
The variable is almost unused, and one of the two uses is actually
uninitialized.
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/check | 5 +
tests/qemu-iotests/common.rc | 2 +-
2 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/tests/qemu-iotests/check b/tests/qemu-iot
"check" is full of qemu-iotests--specific details. Separating it
from "common" does not make much sense anymore.
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/check | 533 +++-
tests/qemu-iotests/common | 552 ---
It only provides functions used by the test programs.
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/check | 6 --
tests/qemu-iotests/common.rc | 13 +
2 files changed, 5 insertions(+), 14 deletions(-)
diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check
in
The variable is used in "common" but defined only after the file
is sourced.
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/check | 2 --
tests/qemu-iotests/common | 2 ++
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check
ind
Split "check" parts from tests part.
For the directory setup, the actual computation of directories goes
in "check", while the sanity checks go in the tests.
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/common| 24
tests/qemu-iotests/common.config | 49 --
This includes shell function, shell variables and command line options
(randomize.awk does not exist).
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/check | 28 -
tests/qemu-iotests/common| 23 --
tests/qemu-iotests/common.config | 26 ---
Some functions in common.rc are never used by the tests. Move
them out of that file and into common, which is already included
only by "check".
Code that actually *is* common to "check" and tests can be placed in
common.config.
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/common
Instead of ./check failing when a binary is missing, we try each test
case now and each one fails with tons of test case diffs. Also, all the
variables were initialized by "check" prior to "common" being sourced,
and then (uselessly) checked for emptiness again in "check".
Centralize the search f
The purpose of this series is to separate the "check" sources from
the tests. After these patches, common.config is reduced to simple
shell initialization, and common.rc is only included by the tests.
Along the way, a lot of dead code is removed too.
In v2, the following patches:
qemu-iotests
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/check | 4 ++--
tests/qemu-iotests/common| 2 +-
tests/qemu-iotests/common.config | 3 ---
3 files changed, 3 insertions(+), 6 deletions(-)
diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check
index 4a6ed67b42..5c0d0
On 09/12/2017 05:45 AM, Thomas Huth wrote:
> On 11.09.2017 19:20, Eric Blake wrote:
>> Maintaining two layers of libqtest APIs, one that takes an explicit
>> QTestState object, and the other that uses the implicit global_qtest,
>> is annoying. In the interest of getting rid of global implicit
>> s
On 09/12/2017 05:14 AM, Thomas Huth wrote:
> On 11.09.2017 19:20, Eric Blake wrote:
>> We have several callers that were formatting the argument strings
>> themselves; consolidate this effort by adding new convenience
>> functions directly in libqtest, and update all call-sites that
>> can benefit
On 09/12/2017 02:29 AM, Thomas Huth wrote:
> On 11.09.2017 19:19, Eric Blake wrote:
>> Commit 2f8b2767 originally added qpci_plug_device_test() and
>> qpci_unplug_acpi_device_test() as a pair, both in pci-pc.c.
>> Later, commit cf716b31 moved one half of the pair to pci.c
>> when adding PPC64 suppo
On 09/12/2017 02:21 AM, Thomas Huth wrote:
> On 11.09.2017 19:19, Eric Blake wrote:
>> When initializing a QVirtioDevice (which always has an associated
>> QVirtioBus), we want to track which QTestState to use for all
>> I/O processed through that bus and device. Copy the paradigm
>> used for QPCI
On 12/09/2017 14:57, Kevin Wolf wrote:
> Am 12.09.2017 um 14:31 hat Paolo Bonzini geschrieben:
>> On 12/09/2017 12:40, Kevin Wolf wrote:
>>> Am 09.08.2017 um 23:55 hat Paolo Bonzini geschrieben:
Split "check" parts from tests part.
For the directory setup, the actual computation of d
Am 12.09.2017 um 14:31 hat Paolo Bonzini geschrieben:
> On 12/09/2017 12:40, Kevin Wolf wrote:
> > Am 09.08.2017 um 23:55 hat Paolo Bonzini geschrieben:
> >> Split "check" parts from tests part.
> >>
> >> For the directory setup, the actual computation of directories goes
> >> in "check", while the
On 12/09/2017 14:51, Kevin Wolf wrote:
> Am 12.09.2017 um 14:28 hat Paolo Bonzini geschrieben:
>> On 12/09/2017 12:31, Kevin Wolf wrote:
>>> Hm, does this mean that instead of ./check failing when a binary is
>>> missing, we try each test case now and each one fails with the same
>>> error message?
Am 12.09.2017 um 14:28 hat Paolo Bonzini geschrieben:
> On 12/09/2017 12:31, Kevin Wolf wrote:
> > Hm, does this mean that instead of ./check failing when a binary is
> > missing, we try each test case now and each one fails with the same
> > error message?
> >
> > *tries it out*
> >
> > Okay, it
On 12/09/2017 12:40, Kevin Wolf wrote:
> Am 09.08.2017 um 23:55 hat Paolo Bonzini geschrieben:
>> Split "check" parts from tests part.
>>
>> For the directory setup, the actual computation of directories goes
>> in "check", while the sanity checks go in the tests.
>>
>> Signed-off-by: Paolo Bonzini
On 12/09/2017 12:31, Kevin Wolf wrote:
> Hm, does this mean that instead of ./check failing when a binary is
> missing, we try each test case now and each one fails with the same
> error message?
>
> *tries it out*
>
> Okay, it's already broken today because the strings are never empty but
> cont
On Tue, Sep 12, 2017 at 12:28:53PM +0100, Daniel P. Berrange wrote:
> Make the crypto driver implement the bdrv_co_preadv|pwritev
> callbacks, and also use bdrv_co_preadv|pwritev for I/O
> with the protocol driver beneath.
>
> Signed-off-by: Daniel P. Berrange
> ---
> block/crypto.c | 104
> +++
The BDRV_REQ_FUA flag can trivially be allowed in the crypt driver
as a passthrough to the underlying block driver.
Signed-off-by: Daniel P. Berrange
---
block/crypto.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/block/crypto.c b/block/crypto.c
index 1f77095336..d1
Using 16KB bounce buffers creates a significant performance
penalty for I/O to encrypted volumes on storage which high
I/O latency (rotating rust & network drives), because it
triggers lots of fairly small I/O operations.
On tests with rotating rust, and cache=none|directsync,
write speed increase
Instead of sector offset, take the bytes offset when encrypting
or decrypting data.
Signed-off-by: Daniel P. Berrange
---
block/crypto.c | 12
block/qcow.c | 7 +--
block/qcow2-cluster.c | 8 +++-
block/qcow2.c | 4 ++--
crypto/block-luks.c
Make the crypto driver implement the bdrv_co_preadv|pwritev
callbacks, and also use bdrv_co_preadv|pwritev for I/O
with the protocol driver beneath.
Signed-off-by: Daniel P. Berrange
---
block/crypto.c | 104 +++--
1 file changed, 56 insertions
Use the qcrypto_block_get_sector_size() value in the block
crypto driver instead of hardcoding 512 as the sector size.
Signed-off-by: Daniel P. Berrange
---
block/crypto.c | 34 ++
1 file changed, 18 insertions(+), 16 deletions(-)
diff --git a/block/crypto.c b/bl
While current encryption schemes all have a fixed sector size of
512 bytes, this is not guaranteed to be the case in future. Expose
the sector size in the APIs so the block layer can remove assumptions
about fixed 512 byte sectors.
Signed-off-by: Daniel P. Berrange
---
crypto/block-luks.c|
The crypto APIs report the offset of the data payload as an uint64_t
type, but the block driver is casting to size_t or ssize_t which will
potentially truncate.
Most of the block APIs use int64_t for offsets meanwhile, so even if
using uint64_t in the crypto block driver we are still at risk of
tr
This is a followup to
v1: https://lists.gnu.org/archive/html/qemu-devel/2017-08/msg00781.html
v2: https://lists.gnu.org/archive/html/qemu-devel/2017-08/msg06464.html
This collection of patches first improves the performance of the
crypto block driver and then does various cleanups to improve
On 11.09.2017 19:20, Eric Blake wrote:
> Maintaining two layers of libqtest APIs, one that takes an explicit
> QTestState object, and the other that uses the implicit global_qtest,
> is annoying. In the interest of getting rid of global implicit
> state and having less code to maintain, merge:
>
On 11.09.2017 19:20, Eric Blake wrote:
> Maintaining two layers of libqtest APIs, one that takes an explicit
> QTestState object, and the other that uses the implicit global_qtest,
> is annoying. In the interest of getting rid of global implicit
> state and having less code to maintain, merge:
>
On 11.09.2017 19:20, Eric Blake wrote:
> Maintaining two layers of libqtest APIs, one that takes an explicit
> QTestState object, and the other that uses the implicit global_qtest,
> is annoying. In the interest of getting rid of global implicit
> state and having less code to maintain, merge:
>
Am 09.08.2017 um 23:55 hat Paolo Bonzini geschrieben:
> Split "check" parts from tests part.
>
> For the directory setup, the actual computation of directories goes
> in "check", while the sanity checks go in the tests.
>
> Signed-off-by: Paolo Bonzini
Same comment as for patch 1, we may want t
On 11.09.2017 19:20, Eric Blake wrote:
> Remove the trivial wrapper qtest_init(), and change qtest_start()
> to no longer implicitly set global_qtest, to make it obvious in the
> rest of the testsuite where we are still relying on global_qtest.
> Everything now uses qtest_start() (and friends) and
Am 09.08.2017 um 23:55 hat Paolo Bonzini geschrieben:
> These are never used by "check", with one exception that does not need
> $QEMU_OPTIONS. Keep them in common.rc, which will be soon included only
> by the tests.
>
> Signed-off-by: Paolo Bonzini
> diff --git a/tests/qemu-iotests/common b/te
On 11.09.2017 19:20, Eric Blake wrote:
> We have several callers that were formatting the argument strings
> themselves; consolidate this effort by adding new convenience
> functions directly in libqtest, and update all call-sites that
> can benefit from it.
[...]
> diff --git a/tests/libqtest.c b/
On Thu, Aug 31, 2017 at 10:08:00AM -0500, Eric Blake wrote:
> On 08/31/2017 06:05 AM, Daniel P. Berrange wrote:
> > Make the crypto driver implement the bdrv_co_preadv|pwritev
> > callbacks, and also use bdrv_co_preadv|pwritev for I/O
> > with the protocol driver beneath.
> >
> > Signed-off-by: D
Am 10.08.2017 um 00:43 hat Paolo Bonzini geschrieben:
>
>
> - Original Message -
> > From: "Eric Blake"
> > To: "Paolo Bonzini" , qemu-de...@nongnu.org
> > Cc: kw...@redhat.com, qemu-block@nongnu.org
> > Sent: Thursday, August 10, 2017 12:18:54 AM
> > Subject: Re: [Qemu-devel] [PATCH 01/
Am 11.09.2017 um 18:51 hat Vladimir Sementsov-Ogievskiy geschrieben:
> Hi Kevin!
>
> I'm confused with relations of permissions and invalidation, can you please
> help?
>
> Now dirty bitmaps are loaded in invalidate_cache. Here is a problem with
> migration:
>
> 1. destination starts (inactive)
On 11.09.2017 19:20, Eric Blake wrote:
> Rather than have two similar shutdown functions, where one requires
> the use of global_qtest in the header, it is better to have a single
> shutdown function that still takes care of cleaning up global_qtest
> if it is set. All callers are updated.
>
> Si
On 11.09.2017 19:20, Eric Blake wrote:
> Drop one more client of global_qtest by teaching all ahci test
> functionality to pass in an explicit QTestState. The state was
> already available, so no callers had to be adjusted.
>
> Signed-off-by: Eric Blake
Reviewed-by: Thomas Huth
On 11.09.2017 19:19, Eric Blake wrote:
> Drop one more client of global_qtest by teaching all fw_cfg test
> functionality (invoked through alloc-pc) to pass in an explicit
> QTestState, adjusting all callers. In particular, fw_cfg-test
> had to reorder things to create the test state prior to crea
On 11.09.2017 19:19, Eric Blake wrote:
> Now that QVirtioDevice and QVirtQueue point back to QVirtioBus,
> we can reuse the explicit QTestState stored there rather than
> relying on implicit global_qtest. We also have to pass QTestState
> through a few functions that can't trace back through
> QVi
On 11.09.2017 19:19, Eric Blake wrote:
> Commit 2f8b2767 originally added qpci_plug_device_test() and
> qpci_unplug_acpi_device_test() as a pair, both in pci-pc.c.
> Later, commit cf716b31 moved one half of the pair to pci.c
> when adding PPC64 support. Keep the implementations of the
> two functi
On 11.09.2017 19:19, Eric Blake wrote:
> When initializing a QVirtioDevice (which always has an associated
> QVirtioBus), we want to track which QTestState to use for all
> I/O processed through that bus and device. Copy the paradigm
> used for QPCIBus, and track the test state at the bus level; t
On 11.09.2017 19:19, Eric Blake wrote:
> When initializing a QPCIBus, track which QTestState the bus is
> associated with (so that a later patch can then explicitly use
> that test state for all communication on the bus, rather than
> blindly relying on global_qtest). Update the initialization
> f
86 matches
Mail list logo
