Public bug reported:
Hi,
We are using “Windows Server 2008” with qemu-kvm-2.0.1 (linux kernel:3.10.0) as
a host of a VM.
Using drive_mirror to do live-migration on the same host for different disks
Local disk: /sf/data/local/
Shared disk(iscsi): /sf/other/local/--- the disk is busy, the IO
** Attachment added: "livemig.png"
https://bugs.launchpad.net/qemu/+bug/1434779/+attachment/4351204/+files/livemig.png
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1434779
Title:
qemu hangs on
After load elf64 binary, qemu tilegx can finish executing the first
system call (uname) successfully in _dl_discover_osversion(), and
return to __libc_start_main().
Chen Gang (12):
linux-user: tilegx: Firstly add architecture related features
linux-user: tilegx: Add target features support wi
They are based on Linux kernel tilegx architecture for 64 bit binary,
also based on tilegx ABI reference document.
Signed-off-by: Chen Gang
---
linux-user/tilegx/syscall.h| 80
linux-user/tilegx/syscall_nr.h | 278
linux-user/tilegx/term
They are for target features within qemu which independent from outside.
Signed-off-by: Chen Gang
---
linux-user/tilegx/target_cpu.h | 35 +++
linux-user/tilegx/target_signal.h | 28 ++
linux-user/tilegx/target_structs.h | 48 +
Add tilegx architecture in "syscall_defs.h", all related features (ioctrl,
and stat) are based on Linux kernel tilegx 64-bit implementation.
Signed-off-by: Chen Gang
---
linux-user/syscall_defs.h | 38 ++
1 file changed, 34 insertions(+), 4 deletions(-)
diff
Add main working flow feature, system call processing feature, and elf64
tilegx binary loading feature, based on Linux kernel tilegx 64-bit
implementation.
Signed-off-by: Chen Gang
---
include/elf.h| 2 ++
linux-user/elfload.c | 23 ++
linux-user/main.c| 86 +
For tilegx, several syscall macros are not supported, so switch them to
avoid building break.
Signed-off-by: Chen Gang
---
linux-user/syscall.c | 50 +-
1 file changed, 49 insertions(+), 1 deletion(-)
diff --git a/linux-user/syscall.c b/linux-user
Add related configuration, make files for tilegx.
Signed-off-by: Chen Gang
---
configure | 3 +++
default-configs/tilegx-linux-user.mak | 1 +
target-tilegx/Makefile.objs | 1 +
3 files changed, 5 insertions(+)
create mode 100644 default-configs/tilegx-linu
It implements minimized cpu features for linux-user.
Signed-off-by: Chen Gang
---
target-tilegx/cpu-qom.h | 73
target-tilegx/cpu.c | 149
target-tilegx/cpu.h | 94 ++
3 files changed,
For supporting linux-user system call, tilegx need support exception
helper features for it.
Signed-off-by: Chen Gang
---
target-tilegx/helper.c | 31 +++
target-tilegx/helper.h | 1 +
2 files changed, 32 insertions(+)
create mode 100644 target-tilegx/helper.c
crea
I don't run apparmor.
Ubuntu 14.10 host
Have this same issue with USB Host Attachment on Windows and Linux guests.
USB Redirection using SPICE is unaffected and works as expected.
Libvirt logs show...
libusbx: error [_get_usbfs_fd] libusbx couldn't open USB device
/dev/bus/usb/003/006: Permissi
It is from Tilera Corporation, and copied from Linux kernel "arch/tile/
include/uapi/arch/opcode_tilegx.h".
Signed-off-by: Chen Gang
---
target-tilegx/opcode_tilegx.h | 1406 +
1 file changed, 1406 insertions(+)
create mode 100644 target-tilegx/opcode_til
Finish processing tilegx bundle, and reach to related pipes. At present,
the qemu tilegx can pass building.
Signed-off-by: Chen Gang
---
target-tilegx/translate.c | 515 ++
1 file changed, 515 insertions(+)
create mode 100644 target-tilegx/translate.c
For the instructions which need tcg generation, the decoding functions
return directly, or they will direct to the exception.
Signed-off-by: Chen Gang
---
target-tilegx/translate.c | 704 ++
1 file changed, 704 insertions(+)
diff --git a/target-tilegx
Generate related tcg instructions, and qemu tilegx can run to 1st system
call (uname) successfully in _dl_discover_osversion(), and return to
__libc_start_main().
Signed-off-by: Chen Gang
---
target-tilegx/translate.c | 543 ++
1 file changed, 543 ins
On Mar 20, 2015, at 6:31 PM, John Snow wrote:
>
>
> On 03/20/2015 06:25 PM, Programmingkid wrote:
>>
>> On Mar 20, 2015, at 4:35 PM, John Snow wrote:
>>>
>>> If you feel like the CD rom boot failure is "new", I'd be very interested
>>> to know how recent that failure is.
>>>
>>> The macio R
With a mask value of 0x0040, the result will never be 1.
This fixes a Coverity warning.
Signed-off-by: Stefan Weil
---
target-tricore/op_helper.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/target-tricore/op_helper.c b/target-tricore/op_helper.c
index 0e7b4e8..b7d18b2
On 21/03/2015 07:25, Emilio G. Cota wrote:
> It always evaluates to true.
>
> Signed-off-by: Emilio G. Cota
> ---
> translate-all.c | 4
> 1 file changed, 4 deletions(-)
>
> diff --git a/translate-all.c b/translate-all.c
> index 9f47ce7..11763c6 100644
> --- a/translate-all.c
> +++ b/tra
32-bit PPC cannot do atomic operations on long long. Inside the loops,
we are already using local counters that are summed at the end of
the run---with one exception in rcu_read_stress_test: fix it to use
the same technique. Then, use a mutex to protect the global counts.
Performance does not mat
On 19/03/2015 18:47, Andreas Färber wrote:
> Use qtest_add_func() instead of g_test_add_func() to reflect
> the architecture tested, changing GTester paths as follows:
>
> /fw_cfg/foo -> /x86_64/fw_cfg/foo etc.
>
> Signed-off-by: Andreas Färber
> ---
> tests/fw_cfg-test.c | 26 +
On 19/03/2015 19:20, Gabriel L. Somlo wrote:
> Signed-off-by: Gabriel Somlo
> ---
>
> And, once again, this time after testing and making sure semicolons
> were't inadvertently left out :)
>
> Sorry for all the noise,
> Gabriel
Queued for 2.3.0-rc.
Paolo
>> On Thu, Mar 19, 2015 at 06:25:1
On 21 March 2015 at 15:34, Paolo Bonzini wrote:
> 32-bit PPC cannot do atomic operations on long long. Inside the loops,
> we are already using local counters that are summed at the end of
> the run---with one exception in rcu_read_stress_test: fix it to use
> the same technique. Then, use a mut
Add a drive property that can be used with sdhci-pci (instead of the odd
"-drive if=sd,... -device sdhci-pci" that has no equivalent in the rest
of QEMU). Then adjust the zynq platform to set it based on the DriveInfo
that it gets.
Note that in this case the BlockBackend is attached to the SDHCI
Am 21.03.2015 um 16:34 schrieb Paolo Bonzini:
> 32-bit PPC cannot do atomic operations on long long. Inside the loops,
> we are already using local counters that are summed at the end of
> the run---with one exception in rcu_read_stress_test: fix it to use
> the same technique. Then, use a mutex
Suggested-by: Paolo Bonzini
Signed-off-by: Emilio G. Cota
---
target-i386/cpu.h | 2 --
translate-all.c | 4
2 files changed, 6 deletions(-)
diff --git a/target-i386/cpu.h b/target-i386/cpu.h
index 15db6d7..4ee12ca 100644
--- a/target-i386/cpu.h
+++ b/target-i386/cpu.h
@@ -31,8 +31,6 @@
On 20 March 2015 at 23:57, Chen Gang wrote:
> For 'opcode_tilegx.h', it comes from Linux kernel which is already more
> than 1000 lines. For me, I still suggest to let it within one patch.
Right, for that sort of special case you should put it in
a single patch which is just "Import copy of whate
32-bit PPC cannot do atomic operations on long long. Inside the loops,
we are already using local counters that are summed at the end of
the run---with some exceptions (rcu_stress_count for rcutorture,
n_nodes for test-rcu-list): fix them to use the same technique.
For test-rcu-list, remove the mo
On Fri, Mar 20, 2015 at 08:39:24AM +0100, Cornelia Huck wrote:
> On Wed, 18 Mar 2015 14:08:56 +0100
> "Michael S. Tsirkin" wrote:
>
> > On Wed, Mar 18, 2015 at 05:34:56PM +0800, Jason Wang wrote:
> > > There's no need to use vector 0 for invalid virtqueue. So this patch
> > > changes to use VIRTI
On 21/03/2015 18:29, Emilio G. Cota wrote:
> Suggested-by: Paolo Bonzini
> Signed-off-by: Emilio G. Cota
> ---
> target-i386/cpu.h | 2 --
> translate-all.c | 4
> 2 files changed, 6 deletions(-)
>
> diff --git a/target-i386/cpu.h b/target-i386/cpu.h
> index 15db6d7..4ee12ca 100644
> -
On Thu, Mar 19, 2015 at 08:52:35PM +0200, Marcel Apfelbaum wrote:
> Notes:
> - I realize that this qualifies as a late submission, but the code was
> arround
>a lot of time and got lots of reviews and it also tested (by Gerd), so
> should be mature enough.
> - This series is not that big, p
Am 21.03.2015 um 19:13 schrieb Paolo Bonzini:
> 32-bit PPC cannot do atomic operations on long long. Inside the loops,
> we are already using local counters that are summed at the end of
> the run---with some exceptions (rcu_stress_count for rcutorture,
> n_nodes for test-rcu-list): fix them to us
This document covers the guest-side hardware interface, as
well as the host-side programming API of QEMU's firmware
configuration (fw_cfg) device.
Signed-off-by: Jordan Justen
Signed-off-by: Gabriel Somlo
---
docs/specs/fw_cfg.txt | 205 ++
1 file
Exit with an error (instead of simply logging a trace event)
whenever the same fw_cfg file name is added multiple times via
one of the fw_cfg_add_file[_callback]() host-side API calls.
Signed-off-by: Gabriel Somlo
Reviewed-by: Laszlo Ersek
---
hw/nvram/fw_cfg.c | 11 ++-
trace-events
Enforce a single assignment of data for each distinct selector key.
Signed-off-by: Gabriel Somlo
Reviewed-by: Laszlo Ersek
---
hw/nvram/fw_cfg.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/hw/nvram/fw_cfg.c b/hw/nvram/fw_cfg.c
index ed70798..227beaf 100644
--- a/hw/nvram/fw_cfg.c
+++ b/
Document and clean up fw_cfg; additionally, allow user-provided blobs to
be inserted into fw_cfg via the qemu command line.
Changes since v2:
- entire series depends on (applies on top of) another fw_cfg patch:
(http://lists.nongnu.org/archive/html/qemu-devel/2015-03/msg04278.html)
>From this point forward, any guest-side writes to the fw_cfg
data register will be treated as no-ops. This patch also removes
the unused host-side API function fw_cfg_add_callback(), which
allowed the registration of a callback to be executed each time
the guest completed a full overwrite of a giv
Allow user supplied files to be inserted into the fw_cfg
device before starting the guest. Since fw_cfg_add_file()
already disallows duplicate fw_cfg file names, qemu will
exit with an error message if the user supplies multiple
blobs with the same fw_cfg file name, or if a blob name
collides with
On 3/22/15 01:33, Peter Maydell wrote:
> On 20 March 2015 at 23:57, Chen Gang wrote:
>> For 'opcode_tilegx.h', it comes from Linux kernel which is already more
>> than 1000 lines. For me, I still suggest to let it within one patch.
>
> Right, for that sort of special case you should put it in
> a
39 matches
Mail list logo
