On Fri, Jan 30, 2015 at 08:37:47AM +0100, Jan Kiszka wrote:
On 2015-01-30 00:06, Paolo Bonzini wrote:
On 29/01/2015 20:37, Marc MarĂ wrote:
Is this an expected behaviour? I can't see why.
I'd like to know if there is a certain reason why it doesn't work. Or
if it should work and
On 02/02/15 23:40, Peter Lieven wrote:
Am 02.02.2015 um 21:09 schrieb Denis V. Lunev:
qemu_gluster_co_discard calculates size to discard as follows
size_t size = nb_sectors * BDRV_SECTOR_SIZE;
ret = glfs_discard_async(s-fd, offset, size, gluster_finish_aiocb, acb);
glfs_discard_async
On 2015-02-02 at 14:00, Kevin Wolf wrote:
Am 26.01.2015 um 16:00 hat Max Reitz geschrieben:
Signed-off-by: Max Reitz mre...@redhat.com
---
qemu-img.c | 57 -
1 file changed, 24 insertions(+), 33 deletions(-)
diff --git a/qemu-img.c
Am 26.01.2015 um 16:00 hat Max Reitz geschrieben:
Now that request clamping is done in the BlockBackend, the growable
field can be removed from the BlockDriverState. All BDSs are now treated
as being growable (that is, they are allowed to grow; they are not
necessarily actually able to).
Am 02.02.2015 um 19:29 schrieb Denis V. Lunev:
nbd_co_discard calls nbd_client_session_co_discard which uses uint32_t
as the length in bytes of the data to discard due to the following
definition:
struct nbd_request {
uint32_t magic;
uint32_t type;
uint64_t handle;
qemu_gluster_co_discard calculates size to discard as follows
size_t size = nb_sectors * BDRV_SECTOR_SIZE;
ret = glfs_discard_async(s-fd, offset, size, gluster_finish_aiocb, acb);
glfs_discard_async is declared as follows:
int glfs_discard_async (glfs_fd_t *fd, off_t length, size_t
On Wed, 2015-01-28 at 16:37 +0800, Chen Fan wrote:
when vfio device support FLR, then when device reset,
we call VFIO_DEVICE_RESET ioctl to reset the device first,
at kernel side, we also can see the order of reset:
3330 rc = pcie_flr(dev, probe);
3331 if (rc != -ENOTTY)
3332
Right now, bdrv_swap() on NBD BDSs results in a segmentation fault
pretty much all of the time. This series fixes this.
Note that this is not a common case, as bdrv_swap() is generally only
performed on root BDSs (there are exceptions, though) and NBD BDSs
normally have a format BDS above them.
Before this patch, the opaque pointer in an NBD BDS points to a
BDRVNBDState, which contains an NbdClientSession object, which in turn
contains a pointer to the BDS. This pointer may become invalid due to
bdrv_swap(), so drop it, and instead pass the BDS directly to the
nbd-client.c functions
On 2015-02-02 at 14:36, Kevin Wolf wrote:
Am 27.01.2015 um 18:11 hat Max Reitz geschrieben:
On 2015-01-27 at 12:10, Eric Blake wrote:
On 01/27/2015 10:04 AM, Max Reitz wrote:
On 2015-01-27 at 11:59, Eric Blake wrote:
On 01/26/2015 08:00 AM, Max Reitz wrote:
Remove growable option from the
On Wed, 2015-01-28 at 16:37 +0800, Chen Fan wrote:
if we detect the aer capability in vfio device, then
we should initialize the vfio device aer rigister bits.
so guest OS can set this bits as needed.
s/rigister/register/
Signed-off-by: Chen Fan chen.fan.f...@cn.fujitsu.com
---
On Wed, 2015-01-28 at 16:37 +0800, Chen Fan wrote:
The why should go here
Signed-off-by: Chen Fan chen.fan.f...@cn.fujitsu.com
---
hw/vfio/pci.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index 54eb6b4..65247ee 100644
---
On 02/02/2015 05:27 AM, Paolo Bonzini wrote:
On 30/01/2015 19:41, John Snow wrote:
+/* Set the #cx'th command of port #px. */
+void ahci_set_command_header(AHCIQState *ahci, uint8_t px,
+ uint8_t cx, AHCICommandHeader *cmd)
+{
+uint64_t ba =
Am 26.01.2015 um 16:00 hat Max Reitz geschrieben:
Signed-off-by: Max Reitz mre...@redhat.com
---
qemu-io.c | 31 ---
1 file changed, 12 insertions(+), 19 deletions(-)
diff --git a/qemu-io.c b/qemu-io.c
index 91a445a..81f8f64 100644
--- a/qemu-io.c
+++
From: Dr. David Alan Gilbert dgilb...@redhat.com
Turn all the D/DD/DDDPRINTFs into trace events
Turn most of the fprintf(stderr, into error_report
Signed-off-by: Dr. David Alan Gilbert dgilb...@redhat.com
---
migration/rdma.c | 333 ---
On 02/02/15 22:58, Peter Lieven wrote:
Am 02.02.2015 um 19:29 schrieb Denis V. Lunev:
qemu_gluster_co_discard calculates size to discard as follows
size_t size = nb_sectors * BDRV_SECTOR_SIZE;
ret = glfs_discard_async(s-fd, offset, size, gluster_finish_aiocb, acb);
glfs_discard_async
Changes from v2:
- dropped conditional in patch gluster code
Changes from v1:
- switched to UINT32_MAX for NBD
- limited max_discard for gluster to SIZE_MAX on i686 only
Signed-off-by: Denis V. Lunev d...@openvz.org
CC: Kevin Wolf kw...@redhat.com
CC: Peter Lieven p...@kamp.de
On 02/02/2015 05:10 AM, Markus Armbruster wrote:
Kevin Wolf kw...@redhat.com writes:
Am 30.01.2015 um 18:04 hat John Snow geschrieben:
On 01/30/2015 09:32 AM, Kevin Wolf wrote:
Am 21.01.2015 um 10:34 hat Markus Armbruster geschrieben:
I'm afraid I forgot much of the discussion we had
On 2015-02-02 at 13:27, Kevin Wolf wrote:
Am 26.01.2015 um 16:00 hat Max Reitz geschrieben:
Signed-off-by: Max Reitz mre...@redhat.com
---
hw/block/xen_disk.c | 28
1 file changed, 12 insertions(+), 16 deletions(-)
diff --git a/hw/block/xen_disk.c
On 2015-02-02 at 13:35, Kevin Wolf wrote:
Am 26.01.2015 um 16:00 hat Max Reitz geschrieben:
Signed-off-by: Max Reitz mre...@redhat.com
---
qemu-img.c | 20 ++--
1 file changed, 6 insertions(+), 14 deletions(-)
diff --git a/qemu-img.c b/qemu-img.c
index 4e9a7f5..be1953d
Am 26.01.2015 um 16:00 hat Max Reitz geschrieben:
This series removes the growable field from the BlockDriverState
object. Its use was to clamp guest requests against the limits of the
BDS; however, this can now be done more easily by moving those checks
into the BlockBackend functions.
In
On 2015-02-02 at 14:34, Kevin Wolf wrote:
Am 26.01.2015 um 16:00 hat Max Reitz geschrieben:
Signed-off-by: Max Reitz mre...@redhat.com
---
qemu-io.c | 31 ---
1 file changed, 12 insertions(+), 19 deletions(-)
diff --git a/qemu-io.c b/qemu-io.c
index
Am 02.02.2015 um 19:29 schrieb Denis V. Lunev:
qemu_gluster_co_discard calculates size to discard as follows
size_t size = nb_sectors * BDRV_SECTOR_SIZE;
ret = glfs_discard_async(s-fd, offset, size, gluster_finish_aiocb, acb);
glfs_discard_async is declared as follows:
int
Expand out and remove the STATUS macro.
Signed-off-by: Peter Maydell peter.mayd...@linaro.org
---
fpu/softfloat-specialize.h | 24 ++---
fpu/softfloat.c| 239 ++---
include/fpu/softfloat.h| 30 +++---
3 files changed, 167 insertions(+),
Am 02.02.2015 um 21:09 schrieb Denis V. Lunev:
qemu_gluster_co_discard calculates size to discard as follows
size_t size = nb_sectors * BDRV_SECTOR_SIZE;
ret = glfs_discard_async(s-fd, offset, size, gluster_finish_aiocb, acb);
glfs_discard_async is declared as follows:
int
On 02/02/2015 05:35 AM, Paolo Bonzini wrote:
On 30/01/2015 19:42, John Snow wrote:
+void ahci_command_set_sizes(AHCICommand *cmd, uint64_t xbytes,
+unsigned prd_size)
+{
+/* Each PRD can describe up to 4MiB, and must not be odd. */
+
On 02/02/2015 12:31 PM, Peter Maydell wrote:
Peter Maydell (3):
softfloat: Expand out the STATUS_PARAM macro
softfloat: expand out STATUS_VAR
softfloat: expand out STATUS macro
fpu/softfloat-specialize.h | 124 ++--
fpu/softfloat.c| 1609
On 02/02/2015 12:31 PM, Peter Maydell wrote:
-void float_raise( int8 flags STATUS_PARAM )
+void float_raise(int8 flags , float_status *status)
Extra space before comma.
r~
On 13/01/2015 09:53, Fam Zheng wrote:
We have module build support around for a while, but also had it bitrot
several times. It probably makes sense to enable it by default so that
people can notice and use it.
Counterpart to --enable-modules, which is turned as default,
--disable-modules
On 02/02/2015 05:37 AM, Paolo Bonzini wrote:
On 30/01/2015 19:42, John Snow wrote:
+/**
+ * Boot and fully enable the HBA device.
+ * @see ahci_boot, ahci_pci_enable and ahci_hba_enable.
+ */
+static AHCIQState *ahci_macro_bootup(void)
Ugly name... I would just leave out this patch.
On 02/02/2015 05:25 AM, Paolo Bonzini wrote:
On 30/01/2015 19:41, John Snow wrote:
+/* Construct our Command Header (set_command_header handles endianness.) */
+memset(cmd, 0x00, sizeof(cmd));
+cmd.flags = 5; /* reg_h2d_fis is 5 double-words long */
+cmd.flags = 0x400;
The qemu process does not always need to be killed, just waiting for it
can be fine, too. This introduces a way to do so.
Signed-off-by: Max Reitz mre...@redhat.com
---
tests/qemu-iotests/common.qemu | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git
When the drive-mirror block job is completed, it will call bdrv_swap()
on the source and the target BDS; this should obviously not result in a
segmentation fault.
Signed-off-by: Max Reitz mre...@redhat.com
---
tests/qemu-iotests/094 | 81 ++
Am 27.01.2015 um 18:11 hat Max Reitz geschrieben:
On 2015-01-27 at 12:10, Eric Blake wrote:
On 01/27/2015 10:04 AM, Max Reitz wrote:
On 2015-01-27 at 11:59, Eric Blake wrote:
On 01/26/2015 08:00 AM, Max Reitz wrote:
Remove growable option from the open command and from the qemu-io
command
nbd_co_discard calls nbd_client_session_co_discard which uses uint32_t
as the length in bytes of the data to discard due to the following
definition:
struct nbd_request {
uint32_t magic;
uint32_t type;
uint64_t handle;
uint64_t from;
uint32_t len; -- the length of data to be
On Wed, 2015-01-28 at 16:37 +0800, Chen Fan wrote:
for old machine types, we should disable aercap feature.
Signed-off-by: Chen Fan chen.fan.f...@cn.fujitsu.com
---
hw/vfio/pci.c | 13 ++---
include/hw/compat.h | 4
2 files changed, 14 insertions(+), 3 deletions(-)
On Wed, 2015-01-28 at 16:37 +0800, Chen Fan wrote:
when the vfio device encounters an uncorrectable error in host,
the vfio_pci driver will signal the eventfd registered by this
vfio device, the results in the qemu eventfd handler getting
invoked.
this patch is to pass the error to guest
This patchset expands out all the uses of the softfloat STATUS_VAR,
STATUS_PARAM and STATUS macros and removes their definitions.
In my opinion these macros are basically just obfuscating the
passing around and use of a float_status pointer in the softfloat
code, and serve no useful purpose. I
On 01/29/2015 10:55 AM, Peter Maydell wrote:
Support guest CPUs which need 7 MMU index values.
Add a comment about what would be required to raise the limit
further (trivial for 8, TCG backend rework for 9 or more).
Signed-off-by: Peter Maydell peter.mayd...@linaro.org
Reviewed-by: Greg
On 2015/2/2 20:19, Wei Liu wrote:
On Mon, Feb 02, 2015 at 09:17:23AM +0800, Tiejun Chen wrote:
When we're working to support IGD GFX passthrough with qemu
upstream, instead of -gfx_passthru we'd like to make that
a machine option, -machine xxx,-igd-passthru=on. This need
to bring a change on
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 02/02/2015 05:37 PM, David Gibson wrote:
On Thu, Jan 29, 2015 at 08:27:19PM +1100, Alexey Kardashevskiy wrote:
Every sPAPRTCETable object holds an IOMMU memory region which holds
a referenced to the sPAPRTCETable instance. So if we want to free
On 01/30/15 17:32, Eric Blake wrote:
On 01/30/2015 02:06 PM, Don Slutz wrote:
This adds two new inject commands:
inject-vmport-reboot
inject-vmport-halt
And three guest info commands:
vmport-guestinfo-set
vmport-guestinfo-get
query-vmport-guestinfo
More details in qmp-commands.hx
On 2015/2/2 19:08, Ian Campbell wrote:
On Mon, 2015-02-02 at 09:17 +0800, Tiejun Chen wrote:
When we're working to support IGD GFX passthrough with qemu
upstream, instead of -gfx_passthru we'd like to make that
a machine option, -machine xxx,-igd-passthru=on. This need
to bring a change on
On Mon, 02/02 20:34, Paolo Bonzini wrote:
On 13/01/2015 09:53, Fam Zheng wrote:
We have module build support around for a while, but also had it bitrot
several times. It probably makes sense to enable it by default so that
people can notice and use it.
Counterpart to
On 2015/2/2 17:37, Markus Armbruster wrote:
Gonglei arei.gong...@huawei.com writes:
On 2015/1/30 20:32, Markus Armbruster wrote:
Gonglei arei.gong...@huawei.com writes:
On 2015/1/30 20:01, Markus Armbruster wrote:
Gonglei arei.gong...@huawei.com writes:
On 2015/1/30 15:46, Markus
On 2015/2/2 20:54, Ian Jackson wrote:
Wei Liu writes (Re: [v2][PATCH] libxl: add one machine property to support IGD GFX
passthrough):
On Mon, Feb 02, 2015 at 09:17:23AM +0800, Tiejun Chen wrote:
When we're working to support IGD GFX passthrough with qemu
upstream, instead of -gfx_passthru
On 02/02/2015 13:28, Pavel Dovgaluk wrote:
cpu-exception_index = EXCP_INTERRUPT;
next_tb = 0;
+qemu_notify_event();
Why is this needed?
It is needed to wake up iothread in replay mode.
Otherwise
On 02/02/2015 14:12, Peter Maydell wrote:
In particular, you can do it while you are under the big QEMU lock.
...but this is a signal handler, so we can't guarantee that the
thread holds the big lock.
...and if we can iterate over CPU lists in signal handlers,
the correct approach is
On 02/02/2015 13:42, Pavel Dovgaluk wrote:
From: Paolo Bonzini [mailto:pbonz...@redhat.com]
On 02/02/2015 13:28, Pavel Dovgaluk wrote:
cpu-exception_index = EXCP_INTERRUPT;
next_tb = 0;
+
Am 30.01.2015 um 09:42 hat Denis V. Lunev geschrieben:
fallocate() works fine and could handle properly with arbitrary size
requests. There is no sense to reduce the amount of space to fallocate.
The bigger is the size, the better is the performance as the amount of
journal updates is reduced.
some remarks:
- i introduced rd_merged and wr_merged block accounting stats to
blockstats as a generic interface which can be set from any
driver that will introduce multirequest merging in the future.
- the old multiwrite interface is still there and might be removed.
v3-v4:
- added a
Am 02.02.2015 um 15:16 schrieb Kevin Wolf:
Am 02.02.2015 um 15:12 hat Peter Lieven geschrieben:
Am 02.02.2015 um 15:04 schrieb Kevin Wolf:
Am 02.02.2015 um 14:55 hat Peter Lieven geschrieben:
Am 02.02.2015 um 14:23 schrieb Kevin Wolf:
Am 30.01.2015 um 09:42 hat Denis V. Lunev geschrieben:
On 01/31/2015 11:23 AM, Jan Kiszka wrote:
On 2015-01-05 12:37, Jan Kiszka wrote:
On 2015-01-05 12:22, Stefan Hajnoczi wrote:
Commit 49d2e648e8087d154d8bf8b91f27c8e05e79d5a6 (machine: remove
qemu_machine_opts global list) removed option descriptions from the
-machine QemuOptsList to avoid
Add the qmp and hmp commands to tune the parameters used in live
migration.
Signed-off-by: Liang Li liang.z...@intel.com
Signed-off-by: Yang Zhang yang.z.zh...@intel.com
---
hmp-commands.hx | 15 ++
hmp.c | 35 ++
hmp.h
Am 09.01.2015 um 16:17 hat Don Slutz geschrieben:
This is the same way vl.c handles this.
Signed-off-by: Don Slutz dsl...@verizon.com
Thanks, applied to the block branch.
Kevin
Am 02.02.2015 um 14:23 schrieb Kevin Wolf:
Am 30.01.2015 um 09:42 hat Denis V. Lunev geschrieben:
fallocate() works fine and could handle properly with arbitrary size
requests. There is no sense to reduce the amount of space to fallocate.
The bigger is the size, the better is the performance as
Signed-off-by: Peter Lieven p...@kamp.de
Reviewed-by: Eric Blake ebl...@redhat.com
Reviewed-by: Max Reitz mre...@redhat.com
---
block.c| 2 ++
block/accounting.c | 7 +++
block/qapi.c | 2 ++
hmp.c | 6 +-
Am 28.01.2015 um 02:51 hat Fam Zheng geschrieben:
The commit 533ffb17a that removed qed_aiocb_info.cancel said to remove
this but didn't do it.
Signed-off-by: Fam Zheng f...@redhat.com
Thanks, applied to the block branch.
Kevin
this adds a knob to disable request merging for debugging or benchmarks if
dedired.
Signed-off-by: Peter Lieven p...@kamp.de
---
hw/block/virtio-blk.c | 5 -
include/hw/virtio/virtio-blk.h | 1 +
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/hw/block/virtio-blk.c
Am 02.02.2015 um 14:55 hat Peter Lieven geschrieben:
Am 02.02.2015 um 14:23 schrieb Kevin Wolf:
Am 30.01.2015 um 09:42 hat Denis V. Lunev geschrieben:
fallocate() works fine and could handle properly with arbitrary size
requests. There is no sense to reduce the amount of space to fallocate.
From: Paolo Bonzini [mailto:pbonz...@redhat.com]
On 02/02/2015 13:28, Pavel Dovgaluk wrote:
cpu-exception_index = EXCP_INTERRUPT;
next_tb = 0;
+qemu_notify_event();
Why is this needed?
It is
Am 30.01.2015 um 09:42 hat Denis V. Lunev geschrieben:
I have performed several tests with non-aligned fallocate calls and
in all cases (with non-aligned fallocates) Linux performs fine, i.e.
areas are zeroed correctly. Checks were made on
Linux 3.16.0-28-generic #38-Ubuntu SMP
This should
At Mon, 2 Feb 2015 12:36:00 +0100,
Kevin Wolf wrote:
Am 02.02.2015 um 10:51 hat Hitoshi Mitake geschrieben:
At Thu, 22 Jan 2015 18:08:11 +0900,
Hitoshi Mitake wrote:
Calling error_get_progname() in the context of qemu-io can cause
segmentation fault because qemu-io doesn't
On Sun, 1 Feb 2015 23:29:20 +0200
Michael S. Tsirkin m...@redhat.com wrote:
On Fri, Jan 30, 2015 at 03:08:08PM +0100, Cornelia Huck wrote:
On Wed, 7 Jan 2015 21:10:07 +0200
Michael S. Tsirkin m...@redhat.com wrote:
On Wed, Jan 07, 2015 at 05:22:32PM +0100, Cornelia Huck wrote:
On
Am 02.02.2015 um 15:04 schrieb Kevin Wolf:
Am 02.02.2015 um 14:55 hat Peter Lieven geschrieben:
Am 02.02.2015 um 14:23 schrieb Kevin Wolf:
Am 30.01.2015 um 09:42 hat Denis V. Lunev geschrieben:
fallocate() works fine and could handle properly with arbitrary size
requests. There is no sense to
On Sun, 01 Feb 2015 14:56:26 +0200
Gal Hammer gham...@redhat.com wrote:
On 22/01/2015 15:52, Igor Mammedov wrote:
On Tue, 16 Dec 2014 17:50:43 +0200
Gal Hammer gham...@redhat.com wrote:
Based on Microsoft's sepecifications (paper can be dowloaded from
Wei Liu writes (Re: [v2][PATCH] libxl: add one machine property to support IGD
GFX passthrough):
On Mon, Feb 02, 2015 at 09:17:23AM +0800, Tiejun Chen wrote:
When we're working to support IGD GFX passthrough with qemu
upstream, instead of -gfx_passthru we'd like to make that
a machine
On 2 February 2015 at 10:03, Paolo Bonzini pbonz...@redhat.com wrote:
You can't do this -- this code is a signal handler so it could
get run at any time including while the list of CPUs is being
updated. (This is why we have the exit_request flag in the
first place rather than just setting the
On 02/02/2015 14:46, Igor Mammedov wrote:
On Sun, 01 Feb 2015 14:56:26 +0200
Gal Hammer gham...@redhat.com wrote:
On 22/01/2015 15:52, Igor Mammedov wrote:
On Tue, 16 Dec 2014 17:50:43 +0200
Gal Hammer gham...@redhat.com wrote:
Based on Microsoft's sepecifications (paper can be dowloaded
From: Paolo Bonzini [mailto:pbonz...@redhat.com]
On 22/01/2015 09:52, Pavel Dovgalyuk wrote:
+if (replay_mode == REPLAY_MODE_RECORD) {
+replay_save_instructions();
+replay_put_event(EVENT_EXCEPTION);
+return true;
Missing mutex lock/unlock.
I think not. We
this patch finally introduces multiread support to virtio-blk. While
multiwrite support was there for a long time, read support was missing.
The complete merge logic is moved into virtio-blk.c which has
been the only user of request merging ever since. This is required
to be able to merge chunks
As it was not obvious (at least for me) where the 32 comes from;
add a constant for it.
Signed-off-by: Peter Lieven p...@kamp.de
Reviewed-by: Eric Blake ebl...@redhat.com
Reviewed-by: Max Reitz mre...@redhat.com
---
hw/block/virtio-blk.c | 2 +-
include/hw/virtio/virtio-blk.h | 4 +++-
On Mon, Feb 02, 2015 at 07:27:10AM +, David Woodhouse wrote:
On Sun, 2015-02-01 at 21:07 -0800, David Miller wrote:
We might as well have not have implemented the IPSEC stack at all,
because as a result of the userland VPN stuff our IPSEC stack is
largely unused except by a very
Am 02.02.2015 um 15:12 hat Peter Lieven geschrieben:
Am 02.02.2015 um 15:04 schrieb Kevin Wolf:
Am 02.02.2015 um 14:55 hat Peter Lieven geschrieben:
Am 02.02.2015 um 14:23 schrieb Kevin Wolf:
Am 30.01.2015 um 09:42 hat Denis V. Lunev geschrieben:
fallocate() works fine and could handle
Signed-off-by: Peter Lieven p...@kamp.de
Reviewed-by: Max Reitz mre...@redhat.com
---
block/block-backend.c | 5 +
include/sysemu/block-backend.h | 1 +
2 files changed, 6 insertions(+)
diff --git a/block/block-backend.c b/block/block-backend.c
index d00c129..c28e240 100644
---
On Mon, 02 Feb 2015 15:13:39 +0200
Gal Hammer gham...@redhat.com wrote:
On 02/02/2015 14:46, Igor Mammedov wrote:
On Sun, 01 Feb 2015 14:56:26 +0200
Gal Hammer gham...@redhat.com wrote:
On 22/01/2015 15:52, Igor Mammedov wrote:
On Tue, 16 Dec 2014 17:50:43 +0200
Gal Hammer
Hi,
are there any plans to release a 2.2 maintainance release anytime soon?
Thanks,
Peter
On 02/02/2015 14:50, Pavel Dovgaluk wrote:
I think not. We just have to write number of already executed instructions.
This number is not linked to exception event. They could be read in replay
while
processing some other event.
I was referring to replay_put_event(EVENT_EXCEPTION) only.
On 2 February 2015 at 08:23, Maxim Ostapenko
m.ostape...@partner.samsung.com wrote:
Sorry, missed ML.
You missed qemu-devel as well :-)
On 02/02/2015 12:20 PM, Maxim Ostapenko wrote:
Hi,
this patch fixes https://bugs.launchpad.net/qemu/+bug/1416988.
There is a small ABI incompatibility
On 30/01/2015 19:42, John Snow wrote:
memset(cmd, 0x00, sizeof(cmd));
-cmd.flags = 5; /* reg_h2d_fis is 5 double-words long */
-cmd.flags = 0x400; /* clear PxTFD.STS.BSY when done */
-cmd.prdtl = 1; /* One PRD table entry. */
+cmd.flags = 5;/*
On 30/01/2015 19:42, John Snow wrote:
The Invalid Command Sentinel here that caps the property array is an
invalid ATA command, namely 0x01. 0x00 is NOP and 0xFF is reserved for
vendor usage, so I chose the first invalid one instead.
You can use ARRAY_SIZE instead.
Paolo
Define the data structure and variables used to do multiple thread
decompression, and add the code to initialize and free them.
Signed-off-by: Liang Li liang.z...@intel.com
Signed-off-by: Yang Zhang yang.z.zh...@intel.com
---
arch_init.c | 13 -
1 file changed, 12 insertions(+), 1
On 30/01/2015 19:42, John Snow wrote:
+void ahci_command_set_sizes(AHCICommand *cmd, uint64_t xbytes,
+unsigned prd_size)
+{
+/* Each PRD can describe up to 4MiB, and must not be odd. */
+g_assert_cmphex(prd_size, =, 4096 * 1024);
+
On 30/01/2015 19:41, John Snow wrote:
Add a simple DMA r/w test to ahci-test.
Oh, and for the first 18 patches, refactor everything into helpers so
that each ahci_test isn't a thousand lines long.
This patch depends upon the ahci test preliminary refactoring series
upstream, which
The multiple compression threads can be turned on/off through
qmp and hmp interface before doing live migration.
Signed-off-by: Liang Li liang.z...@intel.com
Signed-off-by: Yang Zhang yang.z.zh...@intel.com
Reviewed-by: Dr.David Alan Gilbert dgilb...@redhat.com
---
migration/migration.c | 7
Am 02.02.2015 um 10:51 hat Hitoshi Mitake geschrieben:
At Thu, 22 Jan 2015 18:08:11 +0900,
Hitoshi Mitake wrote:
Calling error_get_progname() in the context of qemu-io can cause
segmentation fault because qemu-io doesn't initialize its progname
with error_set_progname(). This patch adds
From: Paolo Bonzini [mailto:pbonz...@redhat.com]
On 22/01/2015 09:52, Pavel Dovgalyuk wrote:
This patch adds calls to replay functions into the icount setup block.
In record mode number of executed instructions is written to the log.
In replay mode number of istructions to execute is taken
Hi Peter,
Sorry for my late reply.
At Thu, 22 Jan 2015 10:07:27 +,
Peter Maydell wrote:
On 22 January 2015 at 09:08, Hitoshi Mitake
mitake.hito...@lab.ntt.co.jp wrote:
Calling error_get_progname() in the context of qemu-x86_64 can cause
segmentation fault because linux-user/main.c
On 29/01/2015 16:52, Peter Maydell wrote:
+CPU_FOREACH(cpu) {
+cpu-exit_loop_request = 1;
+}
}
You can't do this -- this code is a signal handler so it could
get run at any time including while the list of CPUs is being
updated. (This is why we have the exit_request
Add the code to create and destroy the multiple threads those will be
used to do data decompression. Left some functions empty just to keep
clearness, and the code will be added later.
Signed-off-by: Liang Li liang.z...@intel.com
Signed-off-by: Yang Zhang yang.z.zh...@intel.com
---
arch_init.c
Add the qmp and hmp commands to query the parameters used in live
migration.
Signed-off-by: Liang Li liang.z...@intel.com
Signed-off-by: Yang Zhang yang.z.zh...@intel.com
---
hmp-commands.hx | 2 ++
hmp.c | 21 +
hmp.h | 1 +
qemu_put_compression_data() compress the data and put it to QEMUFile.
qemu_put_qemu_file() put the data in the buffer of source QEMUFile to
destination QEMUFile.
Signed-off-by: Liang Li liang.z...@intel.com
Signed-off-by: Yang Zhang yang.z.zh...@intel.com
---
include/migration/qemu-file.h | 3
Give some details about the multiple thread (de)compression and
how to use it in live migration.
Signed-off-by: Liang Li liang.z...@intel.com
Signed-off-by: Yang Zhang yang.z.zh...@intel.com
Reviewed-by: Dr.David Alan Gilbert dgilb...@redhat.com
---
docs/multi-thread-compression.txt | 149
Gonglei arei.gong...@huawei.com writes:
On 2015/1/30 20:32, Markus Armbruster wrote:
Gonglei arei.gong...@huawei.com writes:
On 2015/1/30 20:01, Markus Armbruster wrote:
Gonglei arei.gong...@huawei.com writes:
On 2015/1/30 15:46, Markus Armbruster wrote:
Gonglei
On 30/01/2015 19:41, John Snow wrote:
+/* Construct our Command Header (set_command_header handles endianness.)
*/
+memset(cmd, 0x00, sizeof(cmd));
+cmd.flags = 5; /* reg_h2d_fis is 5 double-words long */
+cmd.flags = 0x400; /* clear PxTFD.STS.BSY when done */
And here
On 30/01/2015 19:41, John Snow wrote:
@@ -703,8 +703,8 @@ static void ahci_test_identify(AHCIQState *ahci)
/* Copy the existing Command #0 structure from the CLB into local memory,
* and build a new command #0. */
memread(ahci-port[i].clb, cmd, sizeof(cmd));
-cmd.b1 = 5;
On 30/01/2015 19:42, John Snow wrote:
+memwrite(ptr, tx, bufsize);
+
+/* Write this buffer to disk, then read it back to the DMA buffer. */
+ahci_guest_io(ahci, px, CMD_WRITE_DMA, ptr, bufsize);
I would qmemset the buffer here.
Paolo
+ahci_guest_io(ahci, px, CMD_READ_DMA,
On 30/01/2015 19:42, John Snow wrote:
+/**
+ * Boot and fully enable the HBA device.
+ * @see ahci_boot, ahci_pci_enable and ahci_hba_enable.
+ */
+static AHCIQState *ahci_macro_bootup(void)
Ugly name... I would just leave out this patch.
Paolo
+{
+AHCIQState *ahci;
+ahci =
Implement the core logic of multiple thread decompression,
the decompression can work now.
Signed-off-by: Liang Li liang.z...@intel.com
Signed-off-by: Yang Zhang yang.z.zh...@intel.com
---
arch_init.c | 49 +++--
1 file changed, 47 insertions(+), 2
This feature can help to reduce the data transferred about 60%, and the
migration time can also be reduced about 70%.
Summary of changed from v3-v4
-Update the test data in the document
-Fix some typo errors
-Use compressBound instead of MACRO define
-Optimize the
1 - 100 of 187 matches
Mail list logo