Eric Blake writes:
> On 04/20/2018 11:14 AM, Junpeng.Li wrote:
>>
>> Hi, everyone. I found a question or maybe a bug during use qemu-io to read
>> a range file and take the params -vq to dump buffer and do not show
>> I/O statistics. I use the params -vq what i want is just dump buffer
>> but do
On 04/20/2018 11:52 AM, Peter Maydell wrote:
> Following commit 12051d82f004024, UART devices should handle
> being passed a NULL pointer chardev, so we don't need to
> create "null" backends in board code. Remove the code that
> does this and updates serial_hds[].
>
> Signed-off-by: Peter Maydel
On Fri, Apr 20, 2018 at 8:12 PM, Tomáš Golembiovský wrote:
> The documentation for kernel-version and kernel-release on Windows was
> swapped.
>
> Signed-off-by: Tomáš Golembiovský
Reviewed-by: Marc-André Lureau
> ---
> qga/qapi-schema.json | 4 ++--
> 1 file changed, 2 insertions(+), 2 dele
Hi Marc-André,
On 04/20/18 17:25, Marc-André Lureau wrote:
> Iterate over the PCI bridges to lookup the PCI device associated with
> the block device.
>
> Fixes:
> https://bugzilla.redhat.com/show_bug.cgi?id=1567041
>
> Signed-off-by: Marc-André Lureau
> ---
> qga/commands-posix.c | 21 +++
On 04/20/18 18:37, Markus Armbruster wrote:
> Laszlo Ersek writes:
>
>> On 04/20/18 14:53, Markus Armbruster wrote:
>>> Laszlo Ersek writes:
>>
>> [snip]
>>
The targets with softmmu are: aarch64, alpha, arm, cris, hppa, i386,
lm32, m68k, microblaze, microblazeel, mips, mips64, mips64el
We'll soon need an enumeration type that lists all the softmmu targets
that QEMU (the project) supports. Introduce @SysEmuTarget to
"common.json".
Cc: "Daniel P. Berrange"
Cc: David Gibson
Cc: Eric Blake
Cc: Gerd Hoffmann
Cc: Kashyap Chamarthy
Cc: Markus Armbruster
Cc: Paolo Bonzini
Cc: Tho
Now that we have @SysEmuTarget, it makes sense to restict
@TargetInfo.@arch to valid sysemu targets at the schema level.
Cc: "Daniel P. Berrange"
Cc: David Gibson
Cc: Eric Blake
Cc: Gerd Hoffmann
Cc: Kashyap Chamarthy
Cc: Markus Armbruster
Cc: Paolo Bonzini
Cc: Thomas Huth
Signed-off-by: L
Add a schema that describes the different uses and properties of virtual
machine firmware.
Each firmware executable installed on a host system should come with at
least one JSON file that conforms to this schema. Each file informs the
management applications about
- the firmware's properties and o
This version seeks to address the RFCv2 feedback. Changes are noted per
patch.
Cc: "Daniel P. Berrange"
Cc: David Gibson
Cc: Eric Blake
Cc: Gerd Hoffmann
Cc: Kashyap Chamarthy
Cc: Markus Armbruster
Cc: Paolo Bonzini
Cc: Thomas Huth
Thanks,
Laszlo
Laszlo Ersek (3):
qapi: add SysEmuTarge
Some block drivers (iscsi and file-posix when dealing with device files)
do not actually support truncation, even though they provide a
.bdrv_truncate() method and will happily return success when providing a
new size that does not exceed the current size. This is because these
drivers expect the
When creating a file, we should take the WRITE and RESIZE permissions.
We do not need either for the creation itself, but we do need them for
clearing and resizing it. So we can take the proper permissions by
replacing O_TRUNC with an explicit truncation to 0, and by taking the
appropriate file lo
Currently we do not take permissions on a file while it is being
created. That is a bit sad. The simplest way to test this is the
following:
$ qemu-img create -f qcow2 foo.qcow2 64M
Formatting 'foo.qcow2', fmt=qcow2 size=67108864 cluster_size=65536
lazy_refcounts=off refcount_bits=16
This patch adds a test case to 153 which tries to overwrite an image
(using qemu-img create) while it is in use. Without the original user
explicitly sharing the necessary permissions (writing and truncation),
this should not be allowed.
Signed-off-by: Max Reitz
---
tests/qemu-iotests/153 |
raw_apply_lock_bytes() and raw_check_lock_bytes() currently take a
BDRVRawState *, but they only use the lock_fd field. During image
creation, we do not have a BDRVRawState, but we do have an FD; so if we
want to reuse the functions there, we should modify them to receive only
the FD.
Signed-off-
On 04/19/2018 08:42 PM, Alex Bennée wrote:
>> It seems I can't even build fp-test for x86_64.
>>
>> CROSS-BUILD x86_64 guest-tests with cc
>> In file included from /home/rth/work/qemu/qemu/tests/fp/fp-test.c:14:0:
>> /home/rth/work/qemu/qemu/include/qemu/osdep.h:30:10: fatal error:
>> config-host
On 04/20/2018 05:50 AM, Alex Bennée wrote:
> +tcg_target_ulong __attribute__((flatten)) helper_le_ldq_mmu(CPUArchState
> *env,
> +target_ulong
> addr,
> +TCGMemOpIdx oi,
> +
On 04/19/2018 10:18 PM, Alex Bennée wrote:
>
> Richard Henderson writes:
>
>> On 04/19/2018 03:58 AM, Alex Bennée wrote:
>>> +test-arm-iwmmxt: CFLAGS+=-marm -march=iwmmxt -mabi=aapcs
>>> +test-arm-iwmmxt: test-arm-iwmmxt.S
>>
>> This appears to be insufficient.
>> If I begin with armv7l-linux-gn
Commit d7d218ef02d87c637d20d64da8f575d434ff6f78 attempted to change
dwProtocols to only advertise support for T=0 and not T=1. The change
was incorrect as it changed 0x0003 to 0x0001.
lsusb -v in a linux guest shows:
"dwProtocols 65536 (Invalid values detected)", though the
smart
On Fri, Apr 20, 2018 at 03:19:47PM -0300, Eduardo Habkost wrote:
> From: Amador Pahim
>
> Otherwise we will see the debug messages in our test logs.
Why don't we want to see the QMP debug messages when debugging is
enabled?
>
> Signed-off-by: Amador Pahim
> Signed-off-by: Eduardo Habkost
> -
On Fri, Apr 20, 2018 at 03:19:28PM -0300, Eduardo Habkost wrote:
> From: Amador Pahim
>
> This patch adds the QEMUMachine._create_console() method, which
> returns a list with the chardev console device arguments to be
> used in the qemu command line.
>
> Signed-off-by: Amador Pahim
> [ehabkost
On Fri, Apr 20, 2018 at 16:11:35 -0300, Philippe Mathieu-Daudé wrote:
> On 04/20/2018 03:35 PM, Emilio G. Cota wrote:
> > I also changed the patches to make the series simpler: one patch
> > to add softmmu.c (without compiling it yet), one patch to
> > actually use softmmu.c, and a final patch to r
On Fri, Apr 20, 2018 at 03:19:29PM -0300, Eduardo Habkost wrote:
> From: Amador Pahim
>
> Avocado Testing Framework can help with functional tests in
> qemu development process. This patch creates the basic infrastructure
> to use with Avocado tests. It contains:
>
> - A README file with the ini
On 04/20/2018 12:57 PM, Dr. David Alan Gilbert (git) wrote:
> From: "Dr. David Alan Gilbert"
>
> Update the migration docs:
>
> Among other changes:
> * Added a general list of advice for device authors
> * Reordered the section on conditional state (subsections etc)
> into the order we
On 04/20/2018 03:55 PM, Emilio G. Cota wrote:
> While at it, set is_jmp to DISAS_NORETURN when generating
> an exception.
>
> Reviewed-by: Richard Henderson
Reviewed-by: Philippe Mathieu-Daudé
> Cc: Stafford Horne
> Signed-off-by: Emilio G. Cota
> ---
> target/openrisc/translate.c | 93
> +
On 2018-04-06 17:17, Max Reitz wrote:
> My non-Fedora testing system does not have a kernel with userfaultfd
> support which causes 181 and 201 to fail. That is annoying. This
> series makes those tests recognize the issue and convert it into a
> _notrun.
>
>
> Max Reitz (2):
> iotests: Add f
Emilio G. Cota writes:
> On Fri, Apr 20, 2018 at 16:50:39 +0100, Alex Bennée wrote:
>> Any questions/thoughts/comments?
>
> Very nice!
>
> I modified the style a little bit, namely to pass checkpatch and to
> improve grepability by having
> foo __attribute__((flatten))
> function(arg1, arg2,
The only non-trivial modification is the use of DISAS_TOO_MANY
in the same way is used by the generic translation loop.
Acked-by: Cornelia Huck
Reviewed-by: David Hildenbrand
Reviewed-by: Richard Henderson
Cc: David Hildenbrand
Cc: Cornelia Huck
Cc: Alexander Graf
Cc: qemu-s3...@nongnu.org
S
Reviewed-by: Richard Henderson
Reviewed-by: Philippe Mathieu-Daudé
Cc: Aurelien Jarno
Cc: Yongbok Kim
Signed-off-by: Emilio G. Cota
---
target/mips/translate.c | 346
1 file changed, 175 insertions(+), 171 deletions(-)
diff --git a/target/mips
Notes:
- BS_EXCP in generate_exception_err becomes DISAS_NORETURN,
because we do not return after raising an exception.
- Some uses of BS_EXCP are misleading in that they're used
only as a "not BS_STOP" exit condition, i.e. they have nothing
to do with an actual exception. For those cases,
Notes:
- Did not convert {num,max}_insns and is_jmp, since the corresponding
code will go away in the next patch.
- Avoided a checkpatch error in use_exit_tb.
- As suggested by David, (1) Drop ctx.pc and use
ctx.base.pc_next instead, and (2) Rename ctx.next_pc to
ctx.pc_tmp and add a comme
On 04/20/2018 11:14 AM, Junpeng.Li wrote:
>
> Hi, everyone. I found a question or maybe a bug during use qemu-io to read
> a range file and take the params -vq to dump buffer and do not show
> I/O statistics. I use the params -vq what i want is just dump buffer
> but do not show I/O statistics, bu
On 04/20/2018 03:35 PM, Emilio G. Cota wrote:
> On Fri, Apr 20, 2018 at 16:50:39 +0100, Alex Bennée wrote:
>> Any questions/thoughts/comments?
>
> Very nice!
Yep :)
>
> I modified the style a little bit, namely to pass checkpatch and to
> improve grepability by having
> foo __attribute__((fla
Notes:
- Did not convert {num,max}_insns, since the corresponding code
will go away in the next patch.
- ctx->pc becomes ctx->base.pc_next, and ctx->next_pc becomes
ctx->pc_succ_insn.
While at it, convert the remaining tb->cflags readers to tb_cflags().
Reviewed-by: Richard Henderson
Cc: M
On Fri, Apr 20, 2018 at 07:42:14 +0100, Alex Bennée wrote:
>
> Richard Henderson writes:
(snip)
> >> Finally the end of the series has me adding Emilio's fp-test to the
> >> per-target builds. Unfortunately although some osdep.h and softfloat.c
> >> fiddling allows us to build in most cases I sti
Reviewed-by: Bastian Koppelmann
Reviewed-by: Richard Henderson
Cc: Michael Clark
Cc: Palmer Dabbelt
Cc: Sagar Karandikar
Cc: Bastian Koppelmann
Signed-off-by: Emilio G. Cota
---
target/riscv/translate.c | 72 +++-
1 file changed, 28 insertions(+),
No changes to the logic here; this is just to make the diff
that follows easier to read.
While at it, remove the unnecessary 'struct' in
'struct TranslationBlock'.
Note that checkpatch complains with a false positive:
ERROR: space prohibited after that '&' (ctx:WxW)
#75: FILE: target/mips/tra
* David Gibson (da...@gibson.dropbear.id.au) wrote:
> So. AFAICT the estimate of page dirty rate is based on the assumption
> that page dirties are independent of each other - one page is as
> likely to be dirtied as any other. If we don't make that assumption,
> I don't see how we can really
On Fri, Apr 20, 2018 at 12:31:42 +1200, Michael Clark wrote:
(snip)
> In any case it seems we need some pretty major changes to translate.c
> before we can make misa writable in qemu-riscv. Almost every gen routine
> with the exception of RVI will need predication based on extensions. It
> makes on
Notes:
- Changed the num_insns test in insn_start to check for
dc->base.num_insns > 1, since when tb_start is first
called in a TB, base.num_insns is already set to 1.
- Removed DISAS_NEXT from the switch in tb_stop; use
DISAS_TOO_MANY instead.
- Added an assert_not_reached on tb_stop for
Notes:
- DISAS_TOO_MANY replaces the former "break" in the translation loop.
However, care must be taken not to overwrite a previous condition
in is_jmp; that's why in translate_insn we first check is_jmp and
return if it's != DISAS_NEXT.
- Added an assert in translate_insn, before exiting
Reviewed-by: Richard Henderson
Cc: Michael Clark
Cc: Palmer Dabbelt
Cc: Sagar Karandikar
Cc: Bastian Koppelmann
Signed-off-by: Emilio G. Cota
---
target/riscv/translate.c | 158 ---
1 file changed, 80 insertions(+), 78 deletions(-)
diff --git a/ta
Note: I looked into dropping dc->do_debug. However, I don't see
an easy way to do it given that TOO_MANY is also valid
when we just translate more than max_insns. Thus, the check
for do_debug in "case DISAS_PC_CC_UPDATED" would still need
additional state to know whether or not we came from
breakpo
While at it, use int for both num_insns and max_insns to make
sure we have same-type comparisons.
Reviewed-by: Richard Henderson
Signed-off-by: Emilio G. Cota
---
accel/tcg/translator.c | 21 ++---
include/exec/translator.h | 8
target/alpha/translate.c | 6 ++-
While at it, set is_jmp to DISAS_NORETURN when generating
an exception.
Reviewed-by: Richard Henderson
Cc: Stafford Horne
Signed-off-by: Emilio G. Cota
---
target/openrisc/translate.c | 93 ++---
1 file changed, 46 insertions(+), 47 deletions(-)
diff --
The TB after BS_STOP is not fixed (e.g. helper_mtc0_hwrena
changes hflags, which ends up changing the TB flags via
cpu_get_tb_cpu_state). This requires a full lookup (i.e.
with flags) via lookup_and_goto_ptr instead of gen_goto_tb,
since the latter only looks at the PC for in-page goto's. Fix it.
This was fairly straightforward since it had already been converted
to DisasContextBase; just had to add TARGET_TOO_MANY to the switch
in tb_stop.
Reviewed-by: Richard Henderson
Cc: Aurelien Jarno
Signed-off-by: Emilio G. Cota
---
target/sh4/translate.c | 171 +-
Reviewed-by: Richard Henderson
Cc: Mark Cave-Ayland
Cc: Artyom Tarasenko
Signed-off-by: Emilio G. Cota
---
target/sparc/translate.c | 27 +++
1 file changed, 15 insertions(+), 12 deletions(-)
diff --git a/target/sparc/translate.c b/target/sparc/translate.c
index 5aa367
Notes:
- pc and npc are left unmodified, since they can point to out-of-TB
jump targets.
- Got rid of last_pc in gen_intermediate_code(), using base.pc_next
instead. Only update pc_next (1) on a breakpoint (so that tb->size
includes the insn), and (2) after reading the current instruction
Notes:
- Moved the cross-page check from the end of translate_insn to
init_disas_context.
Reviewed-by: Richard Henderson
Tested-by: Mark Cave-Ayland
Cc: Mark Cave-Ayland
Cc: Artyom Tarasenko
Signed-off-by: Emilio G. Cota
---
target/sparc/translate.c | 174 +++--
On 04/20/2018 12:55 PM, Peter Maydell wrote:
> In commit 210f47840dd62, we changed the bcm2836 SoC object to
> always create a CPU of the correct type for that SoC model. This
> makes the default_cpu_type settings in the MachineClass structs
> for the raspi2 and raspi3 boards redundant. We didn't c
v2: https://lists.nongnu.org/archive/html/qemu-devel/2018-04/msg00837.html
Changes from v2:
- Rebase onto v2.12.0-rc4
- Add R-b's
- mips:
+ Fix comment typo s/intetrupt/interrupt/
+ Fixes suggested by Richard:
- use lookup_and_goto_ptr for BS_STOP (add a separate patch for this)
- u
Awesome, thanks!
On Fri, 20 Apr 2018, 19:35 Emilio G. Cota, wrote:
> On Fri, Apr 20, 2018 at 16:50:39 +0100, Alex Bennée wrote:
> > Any questions/thoughts/comments?
>
> Very nice!
>
> I modified the style a little bit, namely to pass checkpatch and to
> improve grepability by having
> foo __at
On Fri, Apr 20, 2018 at 16:50:39 +0100, Alex Bennée wrote:
> Any questions/thoughts/comments?
Very nice!
I modified the style a little bit, namely to pass checkpatch and to
improve grepability by having
foo __attribute__((flatten))
function(arg1, arg2, ...)
as opposed to
foo __attribute__((
From: Amador Pahim
Not all distros will support cloudinit. Let's use Fedora, which is
proven to work.
Signed-off-by: Amador Pahim
Signed-off-by: Eduardo Habkost
---
tests/avocado/test_nec-usb-xhci.py | 2 +-
tests/avocado/test_numa_hotplug.py | 2 +-
2 files changed, 2 insertions(+), 2 deleti
From: Amador Pahim
The docstrings are enough. Let's remove the 'pass' instructions.
Signed-off-by: Amador Pahim
Signed-off-by: Eduardo Habkost
---
tests/avocado/avocado_qemu/test.py | 6 --
1 file changed, 6 deletions(-)
diff --git a/tests/avocado/avocado_qemu/test.py
b/tests/avocado/av
From: Cleber Rosa
The PortTracker class was introduced on Avocado's own utility
libraries, starting with version 54.0, and it actually received fixes
after that.
Let's avoid this duplicate implementation and rely on the standard
one.
Signed-off-by: Cleber Rosa
Signed-off-by: Eduardo Habkost
-
From: Amador Pahim
Uses can not add an image to the virtual machine with the option to
configure the user/password using cloudinit.
Signed-off-by: Amador Pahim
Signed-off-by: Eduardo Habkost
---
tests/avocado/README.rst | 4 +-
tests/avocado/avocado_qemu/test.py | 136
From: Amador Pahim
Fixed in commit 243afe858b95765b98d16a1f0dd50dca262858ad.
According to the RHBZ1436616, the issue is: usb-storage device under
pci-bridge is unusable after migration.
Test consists in create a VM with an usb-storage device and check
whether it's still available after a live m
From: Amador Pahim
- Using acell=KVM when we have a guest.
- Better resilience for the cases where live-migration fails.
Signed-off-by: Amador Pahim
Signed-off-by: Eduardo Habkost
---
tests/avocado/test_nec-usb-xhci.py | 6 +-
tests/avocado/test_numa_hotplug.py | 1 +
2 files changed, 6 i
From: Lukáš Doktor
Some of the tests require (usually bootable) image to be executed, let's
add a helper function which uses the default params to define the image
or provides useful default and explains what's going on in case the
image is not available.
Signed-off-by: Lukáš Doktor
Signed-off-
From: Amador Pahim
This issue was fixed in commit d81d857.
According to the RHBZ1431939, the issue is 'host nodes' returning '128'.
It should return empty value when using default policy or 0 when using
bind policy.
Test consists in inspect the result of the 'info memdev' QMP command
after hot-
From: Amador Pahim
Otherwise we will see the debug messages in our test logs.
Signed-off-by: Amador Pahim
Signed-off-by: Eduardo Habkost
---
tests/avocado/avocado_qemu/test.py | 1 +
1 file changed, 1 insertion(+)
diff --git a/tests/avocado/avocado_qemu/test.py
b/tests/avocado/avocado_qemu/
From: Cleber Rosa
These exercise the basics of firing up QEMU with and without
a VNC device, and attempting to set the password when a VNC
device and password is given (or not).
Signed-off-by: Cleber Rosa
Signed-off-by: Eduardo Habkost
---
tests/avocado/test_vnc.py | 58 ++
From: Lukáš Doktor
For conveniency let's set default user to root and password to 123456 to
avoid the need to always input those. This password is inspired by the
default password of JeOS available from avocado-vt.
Signed-off-by: Lukáš Doktor
Signed-off-by: Eduardo Habkost
---
tests/avocado/a
From: Amador Pahim
There's a exception name mismatch when callin `QEMUMigrationError`.
Fixing.
Signed-off-by: Amador Pahim
Signed-off-by: Eduardo Habkost
---
tests/avocado/avocado_qemu/test.py | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/tests/avocado/avocado_qemu
From: Cleber Rosa
Let's just give a single command line that can be reliably executed to
install avocado and the dependencies we're aware for the existing
tests.
The installation approach chosen is the most universal one (pip) and
one that impacts the least the developer's system, being limited
From: Amador Pahim
Fixed in commit e85c0d1.
According to the RHBZ1447027, the issue is: guest cannot boot with 240
or above vcpus when using ovmf.
Test consists in set the VM with 240 vcpus and the OVMF device to then
check whether the VM can be started.
A parameters.yaml file is provided in o
From: Lukáš Doktor
The "arch" is quite important info regarding the current VM, let's store
it so tests can query for it.
Signed-off-by: Lukáš Doktor
Signed-off-by: Eduardo Habkost
---
tests/avocado/avocado_qemu/test.py | 1 +
1 file changed, 1 insertion(+)
diff --git a/tests/avocado/avocado
From: Lukáš Doktor
Adds regresion test for RHBZ1473203 which runs VM with 16 numa nodes and
then verifies memory allocation is accurate before and after hotplugging
memory into default and 13th node.
Signed-off-by: Lukáš Doktor
Signed-off-by: Eduardo Habkost
---
tests/avocado/test_numa_hotplu
From: Lukáš Doktor
The get_console (and _handle_prompt) uses the last non-empty line to
check what is going on, but when debug is enabled, kernel produces lots
of lines spoiling the output. Let's also ignore the messages that looks
like kernel debugs ([ $float] $msg).
This significantly improve
From: Lukáš Doktor
The prompt regexp is quite basic one (# or $), let's require at least
one additional space afterwards, which is the case in most modern
systems and decreases the probability of false-positives.
Signed-off-by: Lukáš Doktor
Signed-off-by: Eduardo Habkost
---
tests/avocado/avo
From: Lukáš Doktor
When the VM is booted before calling "vm.get_console" this command fails
to login as there is no new output in the console. Let's just press
enter (which should be relatively harmless as login asks for the user
again and when already logged in we might "only" execute what's alr
From: Lukáš Doktor
The 10s login is quite short, let's use 60s and see whether it fits.
Signed-off-by: Lukáš Doktor
Signed-off-by: Eduardo Habkost
---
tests/avocado/avocado_qemu/test.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/avocado/avocado_qemu/test.py
b/t
From: Amador Pahim
Expose the QMP result on migration error.
Signed-off-by: Amador Pahim
Signed-off-by: Eduardo Habkost
---
tests/avocado/avocado_qemu/test.py | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tests/avocado/avocado_qemu/test.py
b/tests/avocado/avocado_qemu
From: Amador Pahim
This patch adds the QEMUMachine._create_console() method, which
returns a list with the chardev console device arguments to be
used in the qemu command line.
Signed-off-by: Amador Pahim
[ehabkost: reword commit message]
Signed-off-by: Eduardo Habkost
---
scripts/qemu.py | 4
On 20/04/2018 19:55, Peter Maydell wrote:
> There seems to be a race between tb_gen_code() and qemu_ram_free(),
> which results in an abort() in Edgar's test case that exercises the
> xilinx-spips mmio-exec functionality.
>
> Here's what happens:
> (1) memory_region_invalidate_mmio_ptr() is calle
Background
==
A few months ago, Cleber started a thread[1] about the system for
Avocado-based tests. He invited people to take a look at it, we
got some interesting feedback.
However, I saw no feedback on the actual code itself, and I don't
know what needs to be done to get this merged.
The documentation for kernel-version and kernel-release on Windows was
swapped.
Signed-off-by: Tomáš Golembiovský
---
qga/qapi-schema.json | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/qga/qapi-schema.json b/qga/qapi-schema.json
index 17884c7c70..d3a686ba12 100644
--- a
From: "Dr. David Alan Gilbert"
Update the migration docs:
Among other changes:
* Added a general list of advice for device authors
* Reordered the section on conditional state (subsections etc)
into the order we prefer.
* Add a note about firmware
Signed-off-by: Dr. David Alan Gilbert
There seems to be a race between tb_gen_code() and qemu_ram_free(),
which results in an abort() in Edgar's test case that exercises the
xilinx-spips mmio-exec functionality.
Here's what happens:
(1) memory_region_invalidate_mmio_ptr() is called, and it deletes
the temporary ram MemoryRegion. Thi
On 04/20/2018 11:52 AM, Peter Maydell wrote:
> The superio device has a limit on the number of serial
> ports it supports which is really only there because
> it has a fixed-size array serial[]. This limit isn't
> related particularly to the global MAX_SERIAL_PORTS limit,
> so use a different #defi
On 20/04/2018 19:06, Peter Maydell wrote:
> On 20 April 2018 at 17:58, Paolo Bonzini wrote:
>> On 20/04/2018 16:52, Peter Maydell wrote:
>>> Instead of having a fixed sized global serial_hds[] array,
>>> use a local dynamically reallocated one, so we don't have
>>> a compile time limit on how many
I'm seeing this issue in version 2.12.0-rc4. I wasn't seeing this as
much in earlier 2.11, but its a major pain in the current version.
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1520730
Title:
On 04/20/2018 11:52 AM, Peter Maydell wrote:
> Create a new function max_serial_hds() which returns the number of
> serial ports defined by the user. This is needed only by spapr.
> This allows us to remove the MAX_SERIAL_PORTS define.
>
> Signed-off-by: Peter Maydell
> ---
> include/sysemu/syse
These commands did not get their tests in the original commits:
- guest-get-host-name
- guest-get-timezone
- guest-get-users
Trivial tests that mostly only call the commands were added.
Signed-off-by: Tomáš Golembiovský
---
tests/test-qga.c | 54
On 04/20/2018 07:13 AM, Bob Chen wrote:
> 2.11.1 could work, qemu is no longer occupying 100% CPU. That's
> interesting...
>
Does 2.12 use 100% even at the firmware menu? Maybe we're not giving
this VM long enough to hit the spot that causes it to use 100%.
> Now I can see the starting screen
On 19 April 2018 at 08:27, Sai Pavan Boddu
wrote:
> SNOOP_NONE state handle is moved above in the if ladder, as it's same
> as SNOOP_STRIPPING during data cycles.
>
> Signed-off-by: Sai Pavan Boddu
>
Reviewed-by: Francisco Iglesias
> ---
> Changes for V2:
> Fixed commit message
>
On 04/19/2018 10:05 PM, Stefan Hajnoczi wrote:
>>> This patch series implements .bdrv_co_invalidate_cache() for
>>> block/file-posix.c
>>> on Linux so that shared storage live migration works. I have sent it as an
>>> RFC
>>> because cache consistency is not binary, there are corner cases which
On Fri, Apr 20, 2018 at 12:42:10 +0800, Peter Xu wrote:
> Put all the shared qemu-thread implementations into these files. The
> header should be internal to qemu-thread but not for qemu-thread users.
>
> Introduce some hooks correspondingly for the shared part. Note that in
> qemu_mutex_unlock_
On 20 April 2018 at 17:58, Paolo Bonzini wrote:
> On 20/04/2018 16:52, Peter Maydell wrote:
>> Instead of having a fixed sized global serial_hds[] array,
>> use a local dynamically reallocated one, so we don't have
>> a compile time limit on how many serial ports a system has.
>>
>> Signed-off-by:
On 04/20/2018 11:52 AM, Peter Maydell wrote:
> Instead of having a fixed sized global serial_hds[] array,
> use a local dynamically reallocated one, so we don't have
> a compile time limit on how many serial ports a system has.
>
> Signed-off-by: Peter Maydell
> ---
> include/sysemu/sysemu.h |
On 20/04/2018 16:52, Peter Maydell wrote:
> Instead of having a fixed sized global serial_hds[] array,
> use a local dynamically reallocated one, so we don't have
> a compile time limit on how many serial ports a system has.
>
> Signed-off-by: Peter Maydell
Just one question, would it make sense
On 04/20/2018 11:52 AM, Peter Maydell wrote:
> The ISA serial port handling in serial-isa.c imposes a limit
> of 4 serial ports. This is because we only know of 4 IO port
> and IRQ settings for them, and is unrelated to the generic
> MAX_SERIAL_PORTS limit, though they happen to both be set at
> 4
On Fri, Apr 20, 2018 at 17:17:15 +0100, Alex Bennée wrote:
>
> Emilio G. Cota writes:
>
> > v1: http://lists.gnu.org/archive/html/qemu-devel/2018-02/msg06499.html
> >
> > Changes since v1:
> >
>
>
> I was just catching up on reviews but there are a fair few conflicts in
> the later patches. I
On 04/20/2018 11:52 AM, Peter Maydell wrote:
> Following commit 12051d82f004024, UART devices should handle
> being passed a NULL pointer chardev, so we don't need to
> create "null" backends in board code. Remove the code that
> does this and updates serial_hds[].
>
> (fsl-imx7.c was already writ
Laszlo Ersek writes:
> On 04/20/18 14:53, Markus Armbruster wrote:
>> Laszlo Ersek writes:
>
> [snip]
>
>>> The targets with softmmu are: aarch64, alpha, arm, cris, hppa, i386,
>>> lm32, m68k, microblaze, microblazeel, mips, mips64, mips64el, mipsel,
>>> moxie, nios2, or1k, ppc, ppc64, ppcemb, r
On 04/20/2018 11:52 AM, Peter Maydell wrote:
> Following commit 12051d82f004024, UART devices should handle
> being passed a NULL pointer chardev, so we don't need to
> create "null" backends in board code. Remove the code that
> does this and updates serial_hds[].
>
> Signed-off-by: Peter Maydell
On 04/20/2018 11:52 AM, Peter Maydell wrote:
> Currently the serial.c realize code has an explicit check that it is not
> connected to a disconnected backend (ie one with a NULL chardev).
> This isn't what we want -- you should be able to create a serial device
> even if it isn't attached to anythi
On 20 April 2018 at 16:50, Alex Bennée wrote:
> accel/tcg/Makefile.objs | 1 +
> accel/tcg/cputlb.c | 61 +
> accel/tcg/cputlb.h | 20 ++
> accel/tcg/softmmu.c | 497 +++
> accel/tcg/softmmu_template.h | 435 ---
Emilio G. Cota writes:
> v1: http://lists.gnu.org/archive/html/qemu-devel/2018-02/msg06499.html
>
> Changes since v1:
>
I was just catching up on reviews but there are a fair few conflicts in
the later patches. I also note you have some comments from Richard to
address. Should I wait until v3?
1 - 100 of 272 matches
Mail list logo
