Il 27/08/2013 07:53, Bharata B Rao ha scritto:
> Currently if gluster AIO callback thread fails to notify the QEMU thread about
> AIO completion, we try graceful recovery by marking the disk drive as
> inaccessible. This error recovery code is race-prone as found by Asias and
> Stefan. However as f
On 08/27/2013 01:29 AM, David Gibson wrote:
> On Mon, Aug 26, 2013 at 02:24:49PM +0200, Alexander Graf wrote:
>>
>> On 23.08.2013, at 13:30, Alexey Kardashevskiy wrote:
>>
>>> PAPR+ requires two RTAS calls to be supported by the hypervisor in
>>> order to allow hotplugging VCPUs from the guest. The
its in the first sentence. let me rephrase: the (windows) guest can select some
screen resolution which SDL cannot
provide. what happens is that qemu quits without giving the quest a chance to
shut down. thats like having a monitor
that crashes windows when it doesnt support the video mode.
--
On Mon, Aug 26, 2013 at 09:12:41AM +0200, Markus Armbruster wrote:
> "Michael S. Tsirkin" writes:
>
> > On Thu, Aug 22, 2013 at 01:24:50PM +0200, Markus Armbruster wrote:
> >> "Michael S. Tsirkin" writes:
> >>
> >> > On Thu, Aug 22, 2013 at 10:39:34AM +0200, Markus Armbruster wrote:
> >> >> "Mi
Currently if gluster AIO callback thread fails to notify the QEMU thread about
AIO completion, we try graceful recovery by marking the disk drive as
inaccessible. This error recovery code is race-prone as found by Asias and
Stefan. However as found out by Paolo, this kind of error is impossible and
Alexander Graf writes:
> On 26.08.2013, at 13:46, Nikunj A Dadhania wrote:
>
>> Alexander Graf writes:
>>
>>> On 26.08.2013, at 12:58, Nikunj A Dadhania wrote:
>>>
This implements capabilities exchange between host and client.
As at the moment no capability is supported, put zero fla
On Tue, Aug 27, 2013 at 03:01:03PM +1000, Alexey Kardashevskiy wrote:
> On 08/12/2013 07:02 PM, Michael S. Tsirkin wrote:
> > On Mon, Aug 12, 2013 at 12:36:57AM +1000, Alexey Kardashevskiy wrote:
> >> On 08/11/2013 11:58 PM, Michael S. Tsirkin wrote:
> >>> On Sat, Aug 10, 2013 at 01:09:08AM +1000,
Alexander Graf writes:
+rc = spapr_vio_dma_read(&s->vdev, buffer, &cap, len);
+if (rc) {
+fprintf(stderr, "vscsi_send_capabilities: DMA read failure !\n");
>>>
>>> At this point cap contains random host data, no?
>>
>> Yes, and we zero it out in this case.
>
> The
Paolo Bonzini writes:
> Il 26/08/2013 11:06, Nikunj A Dadhania ha scritto:
+fprintf(stderr, "vscsi_send_capabilities: size out of bound !\n");
+goto error_out;
+}
>>>
>>> I am not 100% familiar with the protocol, could it be that we should
>>> just read sizeof(c
On 08/12/2013 07:02 PM, Michael S. Tsirkin wrote:
> On Mon, Aug 12, 2013 at 12:36:57AM +1000, Alexey Kardashevskiy wrote:
>> On 08/11/2013 11:58 PM, Michael S. Tsirkin wrote:
>>> On Sat, Aug 10, 2013 at 01:09:08AM +1000, Alexey Kardashevskiy wrote:
A PCI device's DMA address space (possibly an
On Tue, 2013-08-27 at 03:48 +0200, Andreas Färber wrote:
> Also, QEMU is definitely not the only project that has higher
> acceptance
> criteria than patch-works-for-the-patch-author. :)
There's a difference between high acceptance criteria and systematic
bike shed painting including in some case
After disabling the QemuClock, we should make sure that no QemuTimers
are still in flight. To implement that with light overhead, we resort
to QemuEvent. The caller of disabling will wait on QemuEvent of each
timerlist.
Note, qemu_clock_enable(foo,false) can _not_ be called from timer's cb.
And th
The vm_clock may be read outside BQL. This will make timers_state
--the foundation of vm_clock exposed to race condition.
Using private lock to protect it.
Note in tcg mode, vm_clock still read inside BQL, so icount is
left without private lock's protection. As for cpu_ticks_* in
timers_state, it
From: Paolo Bonzini
This emulates Win32 manual-reset events using futexes or conditional
variables. Typical ways to use them are with multi-producer,
single-consumer data structures, to test for a complex condition whose
elements come from different threads:
for (;;) {
qemu_event_re
From: Paolo Bonzini
This lets the read-side access run outside the BQL.
Signed-off-by: Paolo Bonzini
---
include/qemu/seqlock.h | 72 ++
1 file changed, 72 insertions(+)
create mode 100644 include/qemu/seqlock.h
diff --git a/include/qemu/seqloc
Saw the Alex's patches has been merged, rebase mine onto his.
v3:
1. rename seqlock_read_check as seqlock_read_retry
2. Document timerlist were protected by BQL, and discard private lock around
"qemu_event_wait(tl->ev)".
v2:
1. fix comment in commit and code
2. fix race issue for qemu_cl
On 08/24/2013 04:42 AM, Stefan Weil wrote:
Am 23.08.2013 21:47, schrieb Richard Henderson:
I've been thinking for a while about how to reliably test TCG backends, and
maybe how to do regression testing on them. Having to begin the test from a
guest binary, especially considering the vast cross-
在 2013-08-26一的 19:23 -0300，Eduardo Habkost写道：
> This is an attempt to make the CPUID cache topology code clearer, by
> replacing the magic numbers in the code with #defines, and moving all
> the cache information to the same place in the file.
>
> I took care of comparing the assembly output of co
In the block-migration period, each chunk of data is read by bdrv_aio_readv(),
then bdrv_co_aio_rw_vector() , then bdrv_co_do_rw().
Inside the bdrv_co_do_rw() function, the bdrv_co_do_readv() function is called
twice. and exit without finish the execution of the bdrv_co_do_readv()
function. Al
This case will test whether the monitor can receive fd at runtime.
To verify better, additional monitor is created to see if qemu
can handler two monitor instance correctly.
Signed-off-by: Wenchao Xia
---
tests/qemu-iotests/045 | 37 -
tests/qemu-iotests
This patch make use of the compiled scm helper program to transfer
fd via unix socket at runtime.
Signed-off-by: Wenchao Xia
---
QMP/qmp.py|6 ++
tests/qemu-iotests/check |1 +
tests/qemu-iotests/iotests.py | 26 ++
3 files changed,
This series add test case for fd passing with unix socket at runtime. Since
getfd and closefd interface will interact with monitor's data, so it will
help to do regression test for monitor patches. Since python2 do not support
sendmsg(), so a C helper program is added to do the job.
v2:
1: add m
This program can do a sendmsg call to transfer fd with unix
socket, which is not supported in python2.
The built binary will not be deleted in clean, but it is a
existing issue in ./tests, which should be solved in another
patch.
Signed-off-by: Wenchao Xia
---
configure
于 2013-8-26 11:13, Wenchao Xia 写道:
> This series add test case for fd passing with unix socket at runtime. Since
> getfd and closefd interface will interact with monitor's data, so it will
> help to do regression test for monitor patches. Since python2 do not support
> sendmsg(), so a C helper prog
Sorry, a little confusion what's the problem you want to solve?
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1216368
Title:
unsupported screen resolution crashes sdl-qemu
Status in QEMU:
New
B
Am 27.08.2013 02:43, schrieb Benjamin Herrenschmidt:
> Am I dreaming ? Those patches were written about a year ago and you guys
> are still nitpicking on names ? They should have been merged a LONG time
> ago ...
[snip]
Ben, this patch was first posted to qemu-devel June 4th, this year.
If patche
于 2013-8-26 23:55, Luiz Capitulino 写道:
On Fri, 23 Aug 2013 16:17:52 +0800
Wenchao Xia wrote:
This series make auto completion and help functions works normal for sub
command, by using reentrant functions. In order to do that, global variables
are not directly used in those functions any more.
On Mon, Aug 26, 2013 at 02:24:49PM +0200, Alexander Graf wrote:
>
> On 23.08.2013, at 13:30, Alexey Kardashevskiy wrote:
>
> > PAPR+ requires two RTAS calls to be supported by the hypervisor in
> > order to allow hotplugging VCPUs from the guest. The "start-cpu" RTAS
> > call was already there bu
On Mon, 2013-08-26 at 15:43 +0200, Paolo Bonzini wrote:
> Il 23/08/2013 11:23, Alexey Kardashevskiy ha scritto:
> > The existing driver just dropped unsupported requests. This adds error
> > responses to those unhandled requests.
> >
> > Signed-off-by: Alexey Kardashevskiy
> > ---
> > hw/scsi/sp
On Mon, 2013-08-26 at 15:37 +0200, Paolo Bonzini wrote:
> There are certainly cases where time-of-check-to-time-of-use
> vulnerability could make QEMU access uninitialized memory (or worse,
> out-of-bounds arrays). For example, you could try racing the host on
> the length of a scatter/gather list
On Mon, 2013-08-26 at 15:11 +0200, Alexander Graf wrote:
> On 19.08.2013, at 07:55, Alexey Kardashevskiy wrote:
>
> > From: David Gibson
> >
> > Recent PowerKVM allows the kernel to intercept some RTAS calls from
> the
> > guest directly. This is used to implement the more efficient
> in-kernel
On 26 August 2013 22:00, Richard Henderson wrote:
> Allow the code that tcg generates to be less obtuse, passing in
> the return address directly instead of computing it in the helper.
> +uint8_t helper_ret_ldb_mmu(CPUArchState *env, target_ulong addr,
> + int mmu_idx, u
On 26 August 2013 23:33, Paolo Bonzini wrote:
> Il 26/08/2013 19:30, Richard Henderson ha scritto:
>> This isn't the kernel, where non-pagable code size is a concern, so I don't
>> see
>> how full configuration of machine emulations and devices is helpful. I'd be
>> more inclined to go the other
On 08/26/2013 03:33 PM, Paolo Bonzini wrote:
> Il 26/08/2013 19:30, Richard Henderson ha scritto:
>> This isn't the kernel, where non-pagable code size is a concern, so I don't
>> see
>> how full configuration of machine emulations and devices is helpful. I'd be
>> more inclined to go the other w
Il 26/08/2013 21:47, Peter Maydell ha scritto:
> In general I'm not convinced it's wise to give end-users and distros
> even more config options to shoot themselves in the foot with
Note that the Kconfig project is about giving end users _less_ config
options. default-configs/ does not capture de
Il 26/08/2013 19:15, Andreas Färber ha scritto:
>> > PCI devices are generally configurable, so you need to add prompts to them.
> IndustryPack is really misplaced in hw/char/ and I believe I posted
> patches to remedy that and let one actually find it in our source tree.
> There were no objections
On 08/26/2013 03:26 PM, Paolo Bonzini wrote:
> Something that can be done on top of this patch: what about moving the
> "-1" to helper_ret_*? It is common to pretty much all the targets
> (except ARM has -2), and it would allow some simplifications.
I suppose so, yes.
> li rN, retaddr
>
Il 26/08/2013 19:30, Richard Henderson ha scritto:
> This isn't the kernel, where non-pagable code size is a concern, so I don't
> see
> how full configuration of machine emulations and devices is helpful. I'd be
> more inclined to go the other way, where all qemu-system-cpu images always
> build
Il 26/08/2013 23:00, Richard Henderson ha scritto:
> Allow the code that tcg generates to be less obtuse, passing in
> the return address directly instead of computing it in the helper.
>
> Maintain the old entrance point unchanged as an alternate entry point.
>
> Delete the helper_st*_cmmu proto
Il 26/08/2013 23:48, Mike Day ha scritto:
>
>
> Mathieu - Thanks for the review! And thanks for the code, I'm working
> with it right now. I like the idea of using a flag to provide a form of
> atomicity for the doubly-linked list elements. I'm also planning on
> running some timing tests to see
This is an attempt to make the CPUID cache topology code clearer, by
replacing the magic numbers in the code with #defines, and moving all
the cache information to the same place in the file.
I took care of comparing the assembly output of compiling
target-i386/cpu.c before and after applying this
On Sun, Aug 25, 2013 at 3:18 PM, Mathieu Desnoyers <
mathieu.desnoy...@efficios.com> wrote:
>
> I'm not very comfortable with your DQ implementation not providing any
> kind of guarantee to a forward traversal followed by backward traversal,
> nor for backward followed by forward traversal. If a li
On 24/08/13 18:46, Andreas Färber wrote:
Am 21.08.2013 20:52, schrieb Mark Cave-Ayland:
On 21/08/13 18:54, Andreas Färber wrote:
Shouldn't this blob come in the same patch as an update to some
git module, so that we keep track of the sources used to build
the blob?
I concur. Independent of
Discontinue the jump-around-jump-to-jump scheme, trading it for a single
immediate move instruction. The two extra jumps always consume 7 bytes,
whereas the immediate move is either 5 or 7 bytes depending on where the
code_gen_buffer gets located.
Signed-off-by: Richard Henderson
---
include/ex
Use existing stack space for arguments; don't push/pop.
Use less ifdefs and more C ifs.
Signed-off-by: Richard Henderson
---
tcg/i386/tcg-target.c | 165 ++
1 file changed, 74 insertions(+), 91 deletions(-)
diff --git a/tcg/i386/tcg-target.c b/tcg
Avoid a loop in the tlb_fill path; the fill will either succeed or
generate an exception.
Inline the slow_ld/st function; it was a complete copy of the main
helper except for the actual cross-page unaligned code, and the
compiler was inlining it anyway.
Add unlikely markers optimizing for the mos
Use a 7 byte lea before the ultimate 10 byte movq.
Signed-off-by: Richard Henderson
---
tcg/i386/tcg-target.c | 25 -
1 file changed, 20 insertions(+), 5 deletions(-)
diff --git a/tcg/i386/tcg-target.c b/tcg/i386/tcg-target.c
index 841bd75..8226171 100644
--- a/tcg/i386/
The following is a rebase of the i386 portion of the v2 patch set.
That incorporated the minor comments from Aurelien from v1. This
pull request omits the ARM portion of the patch set, as that has
yet to receive any review.
I'd like to get this patch set pulled, because I have three other
patch
Allow the code that tcg generates to be less obtuse, passing in
the return address directly instead of computing it in the helper.
Maintain the old entrance point unchanged as an alternate entry point.
Delete the helper_st*_cmmu prototypes; the implementations did not exist.
Signed-off-by: Richa
No point in splitting the write into 32-bit pieces.
Reviewed-by: Aurelien Jarno
Signed-off-by: Richard Henderson
---
tcg/i386/tcg-target.c | 3 +--
tcg/tcg.c | 7 +++
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/tcg/i386/tcg-target.c b/tcg/i386/tcg-target.c
ind
Aliasing was forcing s->code_ptr to be re-read after the store.
Keep the pointer in a local variable to help the compiler.
Signed-off-by: Richard Henderson
---
tcg/tcg.c | 10 ++
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/tcg/tcg.c b/tcg/tcg.c
index dac8224..42c95af 10
On Mon, Aug 26, 2013 at 06:45:27PM +0200, Benoît Canet wrote:
> > Talk to Eduardo since this is related to cpuid configuration and he is
> > an expert.
>
> I found the following in target-i386/cpu.c
>
> case 0x8006:
> /* cache info (L2 cache) */
> *eax = 0;
> *ebx
Ping.
r~
On 08/17/2013 03:38 PM, Richard Henderson wrote:
> The getauxval routine was added to glibc for 2.16. In order to
> better support qemu on systems prior to 2.16, add a qemu_getauxval
> entry point, and add the relevant defines to our own elf.h.
>
>
> r~
>
>
> Richard Henderson (6):
Ping.
r~
On 08/16/2013 11:24 PM, Richard Henderson wrote:
> Ping.
>
> r~
>
>
> On 07/24/2013 12:50 PM, Richard Henderson wrote:
>> It has been pointed out on LKML that the alpha umount syscall numbers
>> are named wrong, and a patch to rectify that has been posted for 3.11.
>>
>> Glibc works a
On 26 August 2013 19:44, Lluís Vilanova wrote:
> Richard Henderson writes:
>> This isn't the kernel, where non-pagable code size is a concern, so I don't
>> see
>> how full configuration of machine emulations and devices is helpful. I'd be
>> more inclined to go the other way, where all qemu-sys
Richard Henderson writes:
> On 08/26/2013 09:59 AM, Paolo Bonzini wrote:
>>> I don't understand the point of this. There is one alpha emulation;
>>> if you disable it you're left with a program that's not useful. So
>>> why the configuration for it at all?
>>
>> I think this was just for consis
On 08/23/2013 08:09 AM, Yongbok Kim wrote:
> +/* alias for pre IO hole access */
> +memory_region_init_alias(ram_low_preio, NULL, "mips_malta_low_preio.ram",
> + ram_high, 0, MIN(ram_size, (256 << 20)));
> +memory_region_add_subregion(system_memory, 0, ram_lo
On 08/24/2013 12:42 PM, Max Filippov wrote:
> Signed-off-by: Max Filippov
> ---
> target-xtensa/translate.c | 8
> 1 file changed, 8 insertions(+)
Reviewed-by: Richard Henderson
Excellent, thanks. That was the only target that did not call in for IN_ASM
dumping.
r~
On 08/25/2013 03:20 AM, Tobias Markus wrote:
> When in Long Mode, cpu_x86_seg_cache() logs "DS16" because the Default
> operation size bit (D/B bit) is not set for Long Mode Data Segments since
> there are only Data Segments in Long Mode and no explicit 16/32/64-bit
> Descriptors.
> This patch fixe
On 08/26/2013 09:59 AM, Paolo Bonzini wrote:
>> I don't understand the point of this. There is one alpha emulation;
>> if you disable it you're left with a program that's not useful. So
>> why the configuration for it at all?
>
> I think this was just for consistency with other targets. Either
Am 26.08.2013 12:43, schrieb Paolo Bonzini:
> Il 26/08/2013 00:58, Ákos Kovács ha scritto:
>> Signed-off-by: Ákos Kovács
>> ---
>> hw/char/Kconfig | 24
>> 1 file changed, 24 insertions(+)
>> create mode 100644 hw/char/Kconfig
>>
>> diff --git a/hw/char/Kconfig b/hw/ch
Am 26.08.2013 00:58, schrieb Ákos Kovács:
> Signed-off-by: Ákos Kovács
> ---
> hw/cpu/Kconfig | 11 +++
> 1 file changed, 11 insertions(+)
> create mode 100644 hw/cpu/Kconfig
>
> diff --git a/hw/cpu/Kconfig b/hw/cpu/Kconfig
> new file mode 100644
> index 000..d90cbe5
> --- /dev/nu
Il 26/08/2013 16:41, Richard Henderson ha scritto:
> On 08/25/2013 03:58 PM, Ákos Kovács wrote:
>> Set CONFIG_CLIPPER as default for default-configs/alpha-softmmu.mak
>>
>> Signed-off-by: Ákos Kovács
>> ---
>> default-configs/alpha-softmmu.mak |2 ++
>> hw/alpha/Makefile.objs|
From: Eduardo Habkost
It just needs to set has_pvpanic=false after calling it. This way, it
won't be a special case anymore.
Signed-off-by: Eduardo Habkost
Signed-off-by: Michael S. Tsirkin
---
hw/i386/pc_piix.c | 4 ++--
hw/i386/pc_q35.c | 4 ++--
2 files changed, 4 insertions(+), 4 deletio
> Talk to Eduardo since this is related to cpuid configuration and he is
> an expert.
I found the following in target-i386/cpu.c
case 0x8006:
/* cache info (L2 cache) */
*eax = 0;
*ebx = 0x42004200;
*ecx = 0x02008140;
*edx = 0;
break;
>From
From: yinyin
virtqueue_get_avail_bytes: when found a indirect desc, we need loop over it.
/* loop over the indirect descriptor table */
indirect = 1;
max = vring_desc_len(desc_pa, i) / sizeof(VRingDesc);
num_bufs = i = 0;
desc_pa = vring_desc
From: Eduardo Habkost
The pc_init_pci_1_2()/pc_init_pci_1_0() split was made on commit
6fd028f64f662c801fd5a54d0e3a1d2baeee93ea, in preparation for commit
9953f8822cc316eec9962f0a2858c3439a80adec. The latter was reverted, so there's
no reason to keep two separate functions that do exactly the sam
From: Markus Armbruster
Pass on the generic arguments unadulterated, and the machine-specific
ones as separate argument.
Signed-off-by: Markus Armbruster
Acked-by: Alexander Graf
Reviewed-by: Laszlo Ersek
Signed-off-by: Michael S. Tsirkin
---
hw/ppc/e500.c | 35 ++--
From: Eduardo Habkost
All callers always use the same values (get_system_memory(),
get_system_io()), so the parameters are pointless.
If one day we decide to eliminate get_system_memory() and
get_system_io(), we will be able to do that more easily by adding the
values to struct QEMUMachineInitAr
From: Eduardo Habkost
Making the older compat functions call the newer compat functions at the
beginning allows the older functions undo what's done by newer compat
functions. e.g.: pc_compat_1_4() will be able to call pc_compat_1_5()
and then set has_pvpanic=false.
Signed-off-by: Eduardo Habkos
From: Markus Armbruster
Don't explode QEMUMachineInitArgs before passing it to
sun4m_hw_init(), sun4uv_init().
Signed-off-by: Markus Armbruster
Reviewed-by: Laszlo Ersek
Signed-off-by: Michael S. Tsirkin
---
hw/sparc/sun4m.c | 113 -
hw/s
From: Markus Armbruster
Don't explode when the variable is used just a few times, and never
changed.
Signed-off-by: Markus Armbruster
Reviewed-by: Laszlo Ersek
Reviewed-by: Eduardo Habkost
Signed-off-by: Michael S. Tsirkin
---
hw/i386/pc_q35.c | 22 +-
1 file changed, 9
From: Markus Armbruster
Don't explode QEMUMachineInitArgs before passing it to pc_init1().
Signed-off-by: Markus Armbruster
Reviewed-by: Laszlo Ersek
Reviewed-by: Eduardo Habkost
Signed-off-by: Michael S. Tsirkin
---
hw/i386/pc_piix.c | 65 ++-
From: Alex Williamson
Signed-off-by: Alex Williamson
Signed-off-by: Michael S. Tsirkin
---
docs/q35-chipset.cfg | 23 +++
1 file changed, 23 insertions(+)
diff --git a/docs/q35-chipset.cfg b/docs/q35-chipset.cfg
index 1b6efc0..e4ddb7d 100644
--- a/docs/q35-chipset.cfg
+++
From: Markus Armbruster
Don't explode when the variable is used just once, and never changed.
Signed-off-by: Markus Armbruster
Acked-by: Alexander Graf
Reviewed-by: Laszlo Ersek
Signed-off-by: Michael S. Tsirkin
---
hw/ppc/e500plat.c | 18 ++
hw/ppc/mpc8544ds.c | 18 ++-
Migration code assumes that each MR is a multiple of TARGET_PAGE_SIZE:
MR size is divided by TARGET_PAGE_SIZE, so if it isn't migration
never completes.
But this isn't really required for regions set up with
memory_region_init_ram, since that calls qemu_ram_alloc
which aligns size up using TARGET_P
Make 1.4 compat code call the 1.6 one, reducing
code duplication. Add comment explaining why we can't
make 1.4 call 1.5 as usual.
Signed-off-by: Michael S. Tsirkin
Reviewed-by: Andreas Färber
Reviewed-by: Eduardo Habkost
---
hw/i386/pc_piix.c | 4 ++--
hw/i386/pc_q35.c | 4 ++--
2 files chang
ROM files that are put in FW CFG are copied to guest ram, by BIOS, but
they are not backed by RAM so they don't get migrated.
Each time we change two bytes in such a ROM this breaks cross-version
migration: since we can migrate after BIOS has read the first byte but
before it has read the second o
The following changes since commit ecfe10c9a6f9bc77d0e4b7eb5d0f5d61e8fbaed8:
Merge remote-tracking branch 'pmaydell/tags/pull-target-arm-20130820' into
staging (2013-08-20 11:23:52 -0500)
are available in the git repository at:
git://git.kernel.org/pub/scm/virt/kvm/mst/qemu.git tags/for_an
On Mon, Aug 26, 2013 at 05:17:07PM +0200, Andreas Färber wrote:
> Am 26.08.2013 17:01, schrieb Paolo Bonzini:
> > Il 26/08/2013 16:34, Andreas Färber ha scritto:
> >> Am 26.08.2013 14:21, schrieb Paolo Bonzini:
> >>> Il 13/08/2013 00:02, Paolo Bonzini ha scritto:
> vapic requires the VAPIC ROM
On Fri, 23 Aug 2013 16:17:52 +0800
Wenchao Xia wrote:
> This series make auto completion and help functions works normal for sub
> command, by using reentrant functions. In order to do that, global variables
> are not directly used in those functions any more. With this series, cmd_table
> is a m
On Thu, 22 Aug 2013 21:30:09 +0200
Stefan Weil wrote:
> Function expr_error gets a format string and variable arguments like printf.
> It also never returns. Add the necessary attributes.
>
> Signed-off-by: Stefan Weil
Applied to the qmp branch, thanks.
> ---
> monitor.c |3 ++-
> 1 file
Am 26.08.2013 17:01, schrieb Paolo Bonzini:
> Il 26/08/2013 16:34, Andreas Färber ha scritto:
>> Am 26.08.2013 14:21, schrieb Paolo Bonzini:
>>> Il 13/08/2013 00:02, Paolo Bonzini ha scritto:
vapic requires the VAPIC ROM to be mapped into RAM. This is not
possible without PAM hardware.
I'm a newer in QEMU, and I have the following questions:
1. Is it true that all the requests to disk images need to go through function
bdrv_co_do_readv() or bdrv_co_do_writev()?
2. In block-migration thread, the bdrv_co_do_readv is also called to read
blocks from disk images, in order to finis
Il 26/08/2013 16:34, Andreas Färber ha scritto:
> Am 26.08.2013 14:21, schrieb Paolo Bonzini:
>> Il 13/08/2013 00:02, Paolo Bonzini ha scritto:
>>> vapic requires the VAPIC ROM to be mapped into RAM. This is not
>>> possible without PAM hardware. This fixes a segmentation fault
>>> running with -
Il 16/08/2013 16:46, Anthony PERARD ha scritto:
In some cases (Xen), it will not be initialized before to be used. This
leads to segv.
Signed-off-by: Anthony PERARD
Tested-by: Fabio Fantoni
This patch have solved the critical regression of all hvm domUs that
prevented domUs starting with qe
This feature can be used in case where users are avoiding the iops limit by
doing jumbo I/Os hammering the storage backend.
Signed-off-by: Benoit Canet
---
block/qapi.c |3 +++
blockdev.c | 21 ++---
hmp.c|8 ++--
qapi-schema.json | 10 ++
The max parameter of the leaky bucket throttling algorithm can be used to
allow the guest to do bursts.
The max value is a pool of I/O that the guest can use without being throttled
at all. Throttling is triggered once this pool is empty.
Signed-off-by: Benoit Canet
---
block/qapi.c | 26 +
Signed-off-by: Benoit Canet
---
block.c | 363 +++--
block/qapi.c | 21 ++-
blockdev.c| 100 +++--
include/block/block.h |1 -
include/block/block_int.h | 32 +---
5 files changed, 187 ins
Implement the continuous leaky bucket algorithm devised on IRC as a separate
module.
Signed-off-by: Benoit Canet
---
include/qemu/throttle.h | 103 +
util/Makefile.objs |1 +
util/throttle.c | 390 +++
3 files changed, 49
On 08/25/2013 03:58 PM, Ákos Kovács wrote:
> Set CONFIG_CLIPPER as default for default-configs/alpha-softmmu.mak
>
> Signed-off-by: Ákos Kovács
> ---
> default-configs/alpha-softmmu.mak |2 ++
> hw/alpha/Makefile.objs|2 +-
> 2 files changed, 3 insertions(+), 1 deletion(-)
>
This patchset implement continous leaky bucket throttling.
It use two requests queue to enable to do silly unbalanced throttling like
block_set_io_throttle 0 0 0 0 6000 1
It use two timer to get the timer callbacks and the throttle.c code simple
in this version:
The throttling core is pretty so
Signed-off-by: Benoit Canet
---
tests/Makefile|2 +
tests/test-throttle.c | 481 +
2 files changed, 483 insertions(+)
create mode 100644 tests/test-throttle.c
diff --git a/tests/Makefile b/tests/Makefile
index b0200fd..6d2f3ad 100644
Am 26.08.2013 14:21, schrieb Paolo Bonzini:
> Il 13/08/2013 00:02, Paolo Bonzini ha scritto:
>> vapic requires the VAPIC ROM to be mapped into RAM. This is not
>> possible without PAM hardware. This fixes a segmentation fault
>> running with -M isapc.
>>
>> Cc: qemu-sta...@nongnu.org
>> Signed-of
On Mon, Aug 26, 2013 at 02:49:41PM +0200, Benoît Canet wrote:
>
> Hi,
>
> Thanks for the answer.
>
> > On Tue, Aug 13, 2013 at 08:17:13PM +0200, Benoît Canet wrote:
> > >
> > > Hi,
> > >
> > > I noticed that the l3 cache size of a guest /proc/cpuinfo is not the same
> > > as
> > > the l3 cach
Am 26.08.2013 15:04, schrieb Alexander Graf:
>
> On 19.08.2013, at 06:06, Alexey Kardashevskiy wrote:
>
>> IBM POWERPC processors encode PVR as a CPU family in higher 16 bits and
>> a CPU version in lower 16 bits. Since there is no significant change
>> in behavior between versions, there is no p
On 26.08.2013, at 16:20, Andreas Färber wrote:
> Am 26.08.2013 15:36, schrieb Alexander Graf:
>>
>> On 19.08.2013, at 07:55, Alexey Kardashevskiy wrote:
>>
>>> Before XICS-KVM comes, it makes sense to clean up the emulated XICS a bit.
>>>
>>> This does:
>>> 1. add assert in ics_realize()
>>> 2
Hi,
Okay, so this did arrive after all (albeit 4 h late) – sorry for the
duplicate (see the series from 15:18).
Kind regards,
Max
Am 26.08.2013 15:36, schrieb Alexander Graf:
>
> On 19.08.2013, at 07:55, Alexey Kardashevskiy wrote:
>
>> Before XICS-KVM comes, it makes sense to clean up the emulated XICS a bit.
>>
>> This does:
>> 1. add assert in ics_realize()
>> 2. change variable names from "k" to more informative ones
>>
On 08/25/2013 07:59 AM, Peter Maydell wrote:
> Peter Maydell (9):
> bswap.h: Remove cpu_to_le16wu()
> bswap.h: Remove cpu_to_le32wu()
> bswap.h: Remove le16_to_cpupu()
> bswap.h: Remove le32_to_cpupu()
> bswap.h: Remove be32_to_cpupu()
> bswap.h: Remove cpu_to_be16wu()
> bswap.h: Remo
1 - 100 of 216 matches
Mail list logo
