This patchset overhauls the stellaris_enet TX and RX handling code,
and converts it to use vmstate.
The principal motivation is to fix the buffer overrun noted
in the first patch, and to reimplement things using simpler
state fields which are easier to migrate and to validate in
post_load. I also
From: Paul Moore pmo...@redhat.com
Additional testing reveals that PulseAudio requires shmctl() and the
mlock()/munlock() syscalls on some systems/configurations. As before,
on systems that do require these syscalls, the problem can be seen with
the following command line:
# qemu -monitor
On Mon, Apr 28, 2014 at 10:25 PM, Andreas Färber afaer...@suse.de wrote:
Hi Marc,
Am 28.04.2014 10:26, schrieb Marc Marí:
From: Marc Marí 5.markm...@gmail.com
Modify debug macros as explained in
https://lists.gnu.org/archive/html/qemu-devel/2014-04/msg03642.html
Signed-off-by: Marc Marí
From: Felix Geyer de...@fobos.de
libusb calls timerfd_create() and timerfd_settime() when it's built with
timerfd support.
Command to reproduce:
-device usb-host,hostbus=1,hostaddr=3,id=hostdev0
Log messages:
audit(1390730418.924:135): auid=4294967295 uid=121 gid=103 ses=4294967295
The following changes since commit 750036a848ea913ba6343718ffa70da98f7eef6b:
Merge remote-tracking branch 'remotes/afaerber/tags/prep-for-upstream' into
staging (2014-03-12 17:53:37 +)
are available in the git repository at:
git://github.com/otubo/qemu.git seccomp
for you to fetch
On Mon, 2014-04-28 at 13:32 +0200, Paolo Bonzini wrote:
Il 28/04/2014 11:38, Marcel Apfelbaum ha scritto:
Doesn't Jun's patch fix a dangling pointer? If so, that would come
first anyway.
It removes pointers from a dead list, and it only deals with
some symptom.
Yeah, but dangling
Amos Kong ak...@redhat.com writes:
Not a serious issue, but it's helpful if we can fix it.
V2: split change of scripts/qapi-visit.py to a split patch,
eat space by using a special char as Markus suggested
V3: update commitlog, update special string, fix of adding
const replace
On 25 April 2014 19:29, Luiz Capitulino lcapitul...@redhat.com wrote:
The following changes since commit 7931b05987564b07ada5a4467d8e78a786a3e7d4:
Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging
(2014-04-25 12:22:37 +0100)
are available in the git repository
[fixing Bharata's address]
Am 28.04.2014 13:29, schrieb Greg Kurz:
POWER7, POWER7+ and POWER8 families use the ILE bit of the LPCR
special purpose register to decide the endianness to use when
entering interrupt handlers. When running a linux guest, this
provides a hint on the endianness used
When source directory can be arrived at by two paths,
configure might misdetect an out of tree build.
The simplest way to trigger the problem is running
configure using a full path. E.g. (firstpath refers to qemu source
tree):
ln -s firstpath secondpath
cd firstpath
On Sunday, April 27, 2014 11:10:50 AM Paolo Bonzini wrote:
Il 14/04/2014 16:47, Paul Moore ha scritto:
Yes. Also the commits don't have your signed-off-by:
so I can't apply it.
Eduardo?
It is absurd that we have had two fixes held up this long for such silly
things.
It's not
Il 28/04/2014 14:43, Marcel Apfelbaum ha scritto:
Yeah, but dangling pointers are bad anyway. I think we should include
Jun's patch.
I have nothing against it, but maybe you/someone answer this question
which I honestly don't have the answer for:
After this patch you can replace a boot-able
28.04.2014 15:48, Michael Tokarev wrote:
[]
I've added a tiny (but hackish and fun) wrapper header for all
this, and pushed whole thing into a branch on my site -- see
http://git.corpit.ru/?p=qemu.git;a=shortlog;h=refs/heads/libcacard-standalone
The wrapper is here:
On Mon, 2014-04-28 at 14:48 +0200, Paolo Bonzini wrote:
Il 28/04/2014 14:43, Marcel Apfelbaum ha scritto:
Yeah, but dangling pointers are bad anyway. I think we should include
Jun's patch.
I have nothing against it, but maybe you/someone answer this question
which I honestly don't have
The 2108 chip supports MSI and MSI-X, so update the emulation
to support both chips.
Signed-off-by: Hannes Reinecke h...@suse.de
---
hw/scsi/megasas.c| 138 ++-
hw/scsi/mfi.h| 7 +++
include/hw/pci/pci_ids.h | 1 +
3 files
To ease debugging we should be decoding
the register names.
Signed-off-by: Hannes Reinecke h...@suse.de
---
hw/scsi/megasas.c | 22 +++---
trace-events | 4 ++--
2 files changed, 21 insertions(+), 5 deletions(-)
diff --git a/hw/scsi/megasas.c b/hw/scsi/megasas.c
index
Hi all,
here is now the revisited version of the megaraid sas update.
The original implementation is being unchanged, and the new
emulation is using a different vmstate field (with an appropriate name).
So there shouldn't be any migration issues left.
Patchset is relative to Paolos scsi-next
The trace events already contain the function name, so the actual
message doesn't need to contain any of these informations.
Signed-off-by: Hannes Reinecke h...@suse.de
---
trace-events | 32
1 file changed, 16 insertions(+), 16 deletions(-)
diff --git
25.04.2014, 21:09, Richard Henderson r...@twiddle.net:
On 04/25/2014 01:13 AM, Dmitry Poletaev wrote:
There is a set of test, that checks QEMU CPU for similar behavior with
real hardware (http://roberto.greyhats.it/projects/pills.html). Test
reg/pill2579.c can detect, that program is
OK now that most of the patches are being merged by Juan, here's
an incremental patchset addressing the remaining issues.
changes from v5:
fix warning in openpic code on 32 bit hosts
fix integer overflow in virtio-net
more validation in ssi-sd
address a new issue in virtio
From: Michael Roth mdr...@linux.vnet.ibm.com
CVE-2013-4534
opp-nb_cpus is read from the wire and used to determine how many
IRQDest elements to read into opp-dst[]. If the value exceeds the
length of opp-dst[], MAX_CPU, opp-dst[] can be overrun with arbitrary
data from the wire.
Fix this by
CVE-2013-4149 QEMU 1.3.0 out-of-bounds buffer write in
virtio_net_load()@hw/net/virtio-net.c
} else if (n-mac_table.in_use) {
uint8_t *buf = g_malloc0(n-mac_table.in_use);
We are allocating buffer of size n-mac_table.in_use
qemu_get_buffer(f, buf,
Malformed input can have config_len in migration stream
exceed the array size allocated on destination, the
result will be heap overflow.
To fix, that config_len matches on both sides.
CVE-2014-0182
Reported-by: Dr. David Alan Gilbert dgilb...@redhat.com
Signed-off-by: Michael S. Tsirkin
CVE-2013-4537
s-arglen is taken from wire and used as idx
in ssi_sd_transfer().
Validate it before access.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
hw/sd/ssi-sd.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/hw/sd/ssi-sd.c b/hw/sd/ssi-sd.c
index 3273c8a..b012e57
The PADEN bit in the transmit control register enables padding of short
data packets out to the required minimum length. However a typo here
meant we were adjusting tx_fifo_len rather than tx_frame_len, so the
padding didn't actually happen. Fix this bug.
Signed-off-by: Peter Maydell
Convert this device to use vmstate for its save/load, including
providing a post_load function that sanitizes inbound data to
avoid possible buffer overflows if it is malicious.
The sanitizing fixes CVE-2013-4532 (though nobody should be
relying on the security properties of most of the
Packet transmission for the stellaris ethernet controller can be triggered
in one of two ways:
* by setting a threshold value in the THR register; when the FIFO
fill level reaches the threshold, the h/w starts transmitting.
Software has to finish filling the FIFO before the transmit
Fix various debug format strings which were incorrect for the
data type, so that building with debug enabled is possible.
Signed-off-by: Peter Maydell peter.mayd...@linaro.org
Reviewed-by: Dr. David Alan Gilbert dgilb...@redhat.com
---
hw/net/stellaris_enet.c | 6 +++---
1 file changed, 3
Incoming migration with stellaris_enet is unsafe.
It's being reworked, but for now, simply block it
since noone is using it anyway.
Block outgoing migration for good measure.
CVE-2013-4532
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
hw/net/stellaris_enet.c | 11 ++-
1 file
The datasheet is clear that the frame length written to the DATA
register is actually stored in the TX FIFO; this means we don't
need to keep both tx_frame_len and tx_fifo_len state separately.
Signed-off-by: Peter Maydell peter.mayd...@linaro.org
Reviewed-by: Dr. David Alan Gilbert
On 28 April 2014 14:08, Michael S. Tsirkin m...@redhat.com wrote:
Incoming migration with stellaris_enet is unsafe.
It's being reworked, but for now, simply block it
since noone is using it anyway.
Block outgoing migration for good measure.
CVE-2013-4532
Signed-off-by: Michael S. Tsirkin
The current tx_fifo code has a corner case where the guest can overrun
the fifo buffer: if automatic CRCs are disabled we allow the guest to write
the CRC word even if there isn't actually space for it in the FIFO.
The datasheet is unclear about exactly how the hardware deals with this
situation;
Il 28/04/2014 13:23, Peter Lieven ha scritto:
if the iscsi driver receives a write zeroes request with
the BDRV_REQ_MAY_UNMAP flag set it fails with -ENOTSUP
if the iscsi target does not support WRITE SAME with
UNMAP. However, the BDRV_REQ_MAY_UNMAP is only a hint
and writing zeroes with WRITE
On Mon, Apr 28, 2014 at 04:08:26PM +0300, Michael S. Tsirkin wrote:
Incoming migration with stellaris_enet is unsafe.
It's being reworked, but for now, simply block it
since noone is using it anyway.
Block outgoing migration for good measure.
CVE-2013-4532
Signed-off-by: Michael S.
Am 28.04.2014 14:41, schrieb Peter Crosthwaite:
On Mon, Apr 28, 2014 at 10:25 PM, Andreas Färber afaer...@suse.de wrote:
Hi Marc,
Am 28.04.2014 10:26, schrieb Marc Marí:
From: Marc Marí 5.markm...@gmail.com
Modify debug macros as explained in
2014-04-28 14:25 GMT+02:00 Andreas Färber afaer...@suse.de:
This is exactly how I told you not to do it in response to Peter C.'s
proposal. I had done so in my v1 [1] and it was rejected.
In your response to the proposal, you sent me the link to your dprintf
branch, which uses functions, no
On 28 April 2014 06:07, Michael Tokarev m...@tls.msk.ru wrote:
Here's another trivial-patches pull request. I removed all
questionable or still-being-discussed patchses from there
which accumulated recently.
What's left are either code cleanups and rearrangements,
spelling fixes or --help
On 04/28/2014 12:20 AM, Chun Yan Liu wrote:
Isn't your conversion pair-wise per driver, in that you always pair
bdrv_create2 with options, and bdrv_create with opts? That is, won't
cco-opts always be false if cco-drv-bdrv_create2 is non-NULL, since
we already guaranteed that there is at
Il 28/04/2014 13:11, Peter Lieven ha scritto:
diff --git a/block/iscsi.c b/block/iscsi.c
index b490e98..9f5b4a0 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -30,6 +30,8 @@
#include qemu-common.h
#include qemu/config-file.h
#include qemu/error-report.h
+#include qemu/bitops.h
+#include
On Mon, Apr 28, 2014 at 02:09:50PM +0100, Peter Maydell wrote:
On 28 April 2014 14:08, Michael S. Tsirkin m...@redhat.com wrote:
Incoming migration with stellaris_enet is unsafe.
It's being reworked, but for now, simply block it
since noone is using it anyway.
Block outgoing migration for
On Mon, 2014-04-28 at 13:33 +0200, Markus Armbruster wrote:
Peter Maydell peter.mayd...@linaro.org writes:
QEMU tends to have rather a lot of half-finished API or coding style
transitions, as an inevitable consequence of our having such a large
code base with wildly varying levels of
On 17 April 2014 19:54, Michael S. Tsirkin m...@redhat.com wrote:
On Thu, Apr 17, 2014 at 09:10:12AM -0700, Anthony Liguori wrote:
On Thu, Apr 17, 2014 at 6:54 AM, Michael S. Tsirkin m...@redhat.com wrote:
People sometimes detect security issues in upstream
QEMU and don't know where to
For some weird reason, Git added this changes, which I didn't do (I should
have checked before sending). I'll check if I missed something in my dev
environment.
Marc
Am 28.04.2014 12:56, schrieb Peter Maydell:
QEMU tends to have rather a lot of half-finished API or coding style
transitions, as an inevitable consequence of our having such a large
code base with wildly varying levels of maintainedness. I thought it
would be useful to at least document them
Am 28.04.2014 13:21, schrieb Peter Crosthwaite:
Hi Marc,
On such a long series, it's usual to include a cover letter
summarising the entire series. Its subject is PATCH 00/NN and can be
generated by adding the --cover-letter switch to git send-email. Hand
edit the file them send along with
2014-04-28 15:16 GMT+02:00 Andreas Färber afaer...@suse.de:
The real question to ask is, does the code have any #ifdef DEBUG_FOO, or
does the respective maintainer intend to use it that way? If not, then
your if (DEBUG_FOO) {...} is perfectly valid and makes more sense than
having ..._ENABLED
On Mo, 2014-04-28 at 12:11 +, Gonglei (Arei) wrote:
Hi, Gerd.
What's your opinion about this issue? Thanks!
qemu 2.0 should not need this.
background info:
https://www.kraxel.org/blog/2014/03/qemu-and-usb-tablet-cpu-consumtion/
cheers,
Gerd
On Mon, Apr 28, 2014 at 11:16 PM, Andreas Färber afaer...@suse.de wrote:
Am 28.04.2014 14:41, schrieb Peter Crosthwaite:
On Mon, Apr 28, 2014 at 10:25 PM, Andreas Färber afaer...@suse.de wrote:
Hi Marc,
Am 28.04.2014 10:26, schrieb Marc Marí:
From: Marc Marí 5.markm...@gmail.com
Modify
On 28 April 2014 13:40, Eduardo Otubo ot...@linux.vnet.ibm.com wrote:
The following changes since commit 750036a848ea913ba6343718ffa70da98f7eef6b:
Merge remote-tracking branch 'remotes/afaerber/tags/prep-for-upstream' into
staging (2014-03-12 17:53:37 +)
are available in the git
On Mon, Apr 28, 2014 at 02:24:45PM +0100, Peter Maydell wrote:
On 17 April 2014 19:54, Michael S. Tsirkin m...@redhat.com wrote:
On Thu, Apr 17, 2014 at 09:10:12AM -0700, Anthony Liguori wrote:
On Thu, Apr 17, 2014 at 6:54 AM, Michael S. Tsirkin m...@redhat.com
wrote:
People sometimes
27.04.2014, 20:59, Peter Maydell peter.mayd...@linaro.org:
On 27 April 2014 17:46, Michael Tokarev m...@tls.msk.ru wrote:
25.04.2014 21:24, Peter Maydell wrote:
It is always going to be possible to determine that you're
running on an emulator rather than real hardware, so changing
On 28 April 2014 14:28, Andreas Färber afaer...@suse.de wrote:
Am 28.04.2014 12:56, schrieb Peter Maydell:
QEMU tends to have rather a lot of half-finished API or coding style
transitions, as an inevitable consequence of our having such a large
code base with wildly varying levels of
Am 28.04.2014 15:35, schrieb Peter Crosthwaite:
On Mon, Apr 28, 2014 at 11:16 PM, Andreas Färber afaer...@suse.de wrote:
Am 28.04.2014 14:41, schrieb Peter Crosthwaite:
On Mon, Apr 28, 2014 at 10:25 PM, Andreas Färber afaer...@suse.de wrote:
Hi Marc,
Am 28.04.2014 10:26, schrieb Marc Marí:
On 28 April 2014 14:41, Dmitry Poletaev poletaev-q...@yandex.ru wrote:
Let's imagine we analyse a program(may be a malware) and so
run it in emulator. Malware can execute that test and understand
that it run in an emulator. After that malware can make decision,
that someone analyse it and
Am 28.04.2014 15:22, schrieb Paolo Bonzini:
Il 28/04/2014 13:11, Peter Lieven ha scritto:
diff --git a/block/iscsi.c b/block/iscsi.c
index b490e98..9f5b4a0 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -30,6 +30,8 @@
#include qemu-common.h
#include qemu/config-file.h
#include
Il 09/04/2014 15:52, Michael S. Tsirkin ha scritto:
On Wed, Apr 09, 2014 at 10:56:57AM +, Gonglei (Arei) wrote:
Hi,
QEMU only mmap MSIX_PAGE_SIZE memory for all pci devices in
assigned_dev_register_msix_mmio(), meanwhile the set the one
page memmory to zero, so the rest memory will be
Il 28/04/2014 14:04, Gonglei (Arei) ha scritto:
Hi,
Subject: Re: [Qemu-devel] [Xen-devel] Hvmloader: Modify ACPI to only supply
_EJ0 methods for PCIslots that support hotplug by runtime patching
On Mon, 2014-04-28 at 12:14 +0200, Fabio Fantoni wrote:
Il 22/01/2014 15:32, Fabio Fantoni ha
KVM only supports MSIX table size up to 256 vectors,
but some assigned devices support more vectors,
at the moment attempts to assign them fail with EINVAL.
Tweak the MSIX capability exposed to guest to limit table size
to a supported value.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
On Mon, 2014-04-28 at 17:02 +0300, Michael S. Tsirkin wrote:
KVM only supports MSIX table size up to 256 vectors,
but some assigned devices support more vectors,
at the moment attempts to assign them fail with EINVAL.
Tweak the MSIX capability exposed to guest to limit table size
to a
On Mon, Apr 28, 2014 at 02:24:45PM +0100, Peter Maydell wrote:
On 17 April 2014 19:54, Michael S. Tsirkin m...@redhat.com wrote:
On Thu, Apr 17, 2014 at 09:10:12AM -0700, Anthony Liguori wrote:
On Thu, Apr 17, 2014 at 6:54 AM, Michael S. Tsirkin m...@redhat.com
wrote:
People sometimes
On 04/28/2014 09:47 AM, Paul Moore wrote:
On Sunday, April 27, 2014 11:10:50 AM Paolo Bonzini wrote:
Il 14/04/2014 16:47, Paul Moore ha scritto:
Yes. Also the commits don't have your signed-off-by:
so I can't apply it.
Eduardo?
It is absurd that we have had two fixes held up this long for
I'm understand your position.
But why in TCG undefined flags obviously change to zero in some cases?
For example:
af = 0; /* undefined */
It is not a part of Intel specification, what reason was apply that convention?
28.04.2014, 17:49, Peter Maydell peter.mayd...@linaro.org:
On 28 April
I'll play around once I get the password.
From what I've seen so far,
I'm not sure it's the right server to use for security :(
The list now appears here
https://lists.nongnu.org/mailman/listinfo
under the heading Below is a listing of all the public mailing lists on
lists.nongnu.org.
The list
Il 28/04/2014 15:43, Peter Lieven ha scritto:
b) the allocationmap should be allocated even for out-of-range opt_unmap_gran,
using a granularity of 64 sectors in that case.
Would the increase of the resolution bring any benefit? If we increase the
resolution I think all sectors falling into
On 04/28/2014 06:05 AM, Dmitry Poletaev wrote:
-env-cc_src = (eflags ~(CC_C | CC_O)) |
-(lshift(src ^ t0, 11 - (DATA_BITS - 1)) CC_O) |
+env-cc_src = (eflags ~CC_C) |
((src (DATA_BITS - count)) CC_C);
+if (count == 1) {
+
Il 28/04/2014 15:01, Hannes Reinecke ha scritto:
The 2108 chip supports MSI and MSI-X, so update the emulation
to support both chips.
Should the 2108 PCIDeviceClass set the is_express member to true?
Otherwise you get no QEMU_PCI_CAP_EXPRESS capability.
Paolo
Signed-off-by: Hannes
On 04/28/2014 04:44 PM, Paolo Bonzini wrote:
Il 28/04/2014 15:01, Hannes Reinecke ha scritto:
The 2108 chip supports MSI and MSI-X, so update the emulation
to support both chips.
Should the 2108 PCIDeviceClass set the is_express member to true?
Otherwise you get no QEMU_PCI_CAP_EXPRESS
Am 28.04.2014 16:39, schrieb Paolo Bonzini:
Il 28/04/2014 15:43, Peter Lieven ha scritto:
b) the allocationmap should be allocated even for out-of-range
opt_unmap_gran, using a granularity of 64 sectors in that case.
Would the increase of the resolution bring any benefit? If we increase the
On 04/28/2014 06:21 AM, Michael S. Tsirkin wrote:
When source directory can be arrived at by two paths,
configure might misdetect an out of tree build.
The simplest way to trigger the problem is running
configure using a full path. E.g. (firstpath refers to qemu source
tree):
ln -s
Paolo Bonzini pbonz...@redhat.com writes:
Remove the scsi-block: prefix for error messages as suggested
by Markus.
Improve the previous patch by making the message the same for both
scsi-block and scsi-generic, including the strerror() output in both
and making an explicit reference to
Il 28/04/2014 16:48, Hannes Reinecke ha scritto:
On 04/28/2014 04:44 PM, Paolo Bonzini wrote:
Il 28/04/2014 15:01, Hannes Reinecke ha scritto:
The 2108 chip supports MSI and MSI-X, so update the emulation
to support both chips.
Should the 2108 PCIDeviceClass set the is_express member to
On 28 April 2014 01:45, Peter Crosthwaite peter.crosthwa...@xilinx.com wrote:
Implement named GPIOs on the Device layer. Listifies the existing GPIOs
stuff using string keys. Legacy un-named GPIOs are preserved by using
a NULL name string - they are just a single matchable element in the
name
Il 28/04/2014 16:02, Michael S. Tsirkin ha scritto:
KVM only supports MSIX table size up to 256 vectors,
but some assigned devices support more vectors,
at the moment attempts to assign them fail with EINVAL.
Tweak the MSIX capability exposed to guest to limit table size
to a supported value.
Il 28/04/2014 16:41, Peter Lieven ha scritto:
What if opt_unmap_gran is 32K or lower? In this case you're not using an
allocationmap.
As written I am fine with lowering this to 4K.
Follow-Up or v3?
Follow up is okay.
Paolo
When the deprecated -nographic option is used with the -mon option in
readline mode, qemu will create a second character device for stdio
and place it over the stdio chardev put into place by the -mon
option. This causes the terminal to stop echoeing characters upon exit
from Qemu.
Fix by
From: Michael Tokarev m...@tls.msk.ru
Cherry-pick QEMU commit 0dbcf95a1, because it is still needed
for libvixl 1.3:
disas/libvixl/ contains functions which uses 64bit constants
without using appropriate suffixes, which fails on 32bits.
Fix this by using ULL suffix.
Signed-off-by: Michael
Temporarily disable building the A64 disassembler, to avoid
bisection failures across the commits which update libvixl and
reapply our local build fixes to it.
Signed-off-by: Peter Maydell peter.mayd...@linaro.org
---
configure | 16
1 file changed, 8 insertions(+), 8
On Fri, Apr 25, 2014 at 01:08:18AM -0600, Chun Yan Liu wrote:
Hi,
Hello,
I'm looking at xen hvm guest direct kernel boot and interested to do
it. I found there were some discussions about it and an early work
around by Daniel (based on xen qemu-dm).
[meta-comment]
On 04/27/2014 11:07 PM, TeLeMan wrote:
--
SUN OF A BEACH
On Tue, Oct 15, 2013 at 2:33 PM, Michael S. Tsirkin m...@redhat.com wrote:
This adds C code for generating ACPI tables at runtime,
snip ~ 50k
+if (misc.has_hpet) {
+build_hpet(tables-table_data,
Upstream libvixl 1.3 fixes a number of the places which needed
ULL suffixes, but not all of them; reapply those ones from
commit 37fd5b53b which are still relevant.
Signed-off-by: Peter Maydell peter.mayd...@linaro.org
---
disas/libvixl/a64/instructions-a64.h | 24
Signed-off-by: Peter Lieven p...@kamp.de
---
block/iscsi.c | 12 ++--
1 file changed, 10 insertions(+), 2 deletions(-)
diff --git a/block/iscsi.c b/block/iscsi.c
index 9f5b4a0..eca8a22 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -95,7 +95,15 @@ typedef struct IscsiAIOCB {
depending on the target the opt_unmap_gran might be as low
as 4K. As we know use this also as a knob to activate the allocationmap
feature lower the barrier. The limit 4K (and not 512) is choosen
to avoid a potentially too big allocationmap.
Signed-off-by: Peter Lieven p...@kamp.de
---
This patchset updates our copy of libvixl to the upstream
1.3 release. I don't think there's anything particularly
earthshattering in 1.3 compared to what we had before.
Changes v1-v2:
* add top-and-tail patches which disable building libvixl
for the duration of the commits which update and
Am 28.04.2014 15:01, schrieb Hannes Reinecke:
The 2108 chip supports MSI and MSI-X, so update the emulation
to support both chips.
Signed-off-by: Hannes Reinecke h...@suse.de
---
hw/scsi/megasas.c| 138
++-
hw/scsi/mfi.h|
When source directory can be arrived at by two paths,
configure might misdetect an out of tree build.
The simplest way to trigger the problem is running
configure using a full path. E.g. (firstpath refers to qemu source
tree):
ln -s firstpath secondpath
cd firstpath
This reverts the previous commit disabling the A64 disassembler,
since libvixl will now build on all platforms again.
Signed-off-by: Peter Maydell peter.mayd...@linaro.org
---
configure | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/configure b/configure
Hi,
This is v2 patch for qemu-ga to add argument to specify which filesystems
to be frozen by guest-fsfreeze-freeze command.
Changes to v1:
added documentation of the new field
(v1: http://lists.gnu.org/archive/html/qemu-devel/2014-04/msg04085.html)
---
Tomoki Sekiyama (1):
qga: Add
When an array of mount point paths is specified as 'mountpoints' argument
of guest-fsfreeze-freeze, qemu-ga with this patch will only freeze the file
systems mounted on specified paths in Linux.
This would be useful when the host wants to create partial disk snapshots.
Signed-off-by: Tomoki
Peter Maydell peter.mayd...@linaro.org writes:
Upstream libvixl 1.3 fixes a number of the places which needed
ULL suffixes, but not all of them; reapply those ones from
commit 37fd5b53b which are still relevant.
Any idea why they didn't apply them all?
--
Alex Bennée
On 28 April 2014 16:27, Alex Bennée alex.ben...@linaro.org wrote:
Peter Maydell peter.mayd...@linaro.org writes:
Upstream libvixl 1.3 fixes a number of the places which needed
ULL suffixes, but not all of them; reapply those ones from
commit 37fd5b53b which are still relevant.
Any idea why
On Fri, 2014-04-25 at 01:08 -0600, Chun Yan Liu wrote:
Hi,
I'm looking at xen hvm guest direct kernel boot and interested to do
it. I found there were some discussions about it and an early work
around by Daniel (based on xen qemu-dm).
On Mon, 2014-04-28 at 16:36 +0100, Ian Campbell wrote:
On Fri, 2014-04-25 at 01:08 -0600, Chun Yan Liu wrote:
Hi,
I'm looking at xen hvm guest direct kernel boot and interested to do
it. I found there were some discussions about it and an early work
around by Daniel (based on xen
Eric Blake ebl...@redhat.com writes:
On 04/28/2014 06:21 AM, Michael S. Tsirkin wrote:
When source directory can be arrived at by two paths,
configure might misdetect an out of tree build.
The simplest way to trigger the problem is running
configure using a full path. E.g. (firstpath refers
Thanks to everybody who helped with getting QEMU 2.0 released.
The traditional reward for a job well done is another job, which
means we should probably work out what the release schedule for
2.1 is going to be.
We started 2.1's development phase on 17th April, which means that
for a standard 3
On Fri, Apr 25, 2014 at 11:08:00PM +0200, Paolo Bonzini wrote:
Il 25/04/2014 01:18, Eduardo Habkost ha scritto:
On Fri, Apr 25, 2014 at 12:57:48AM +0200, Paolo Bonzini wrote:
Il 24/04/2014 22:57, Eduardo Habkost ha scritto:
If that didn't break other use cases, I would agree.
But -cpu host
Il 28/04/2014 16:23, Michael S. Tsirkin ha scritto:
When source directory can be arrived at by two paths,
configure might misdetect an out of tree build.
The simplest way to trigger the problem is running
configure using a full path. E.g. (firstpath refers to qemu source
tree):
ln -s
Il 28/04/2014 09:02, Amos Kong ha scritto:
def mcgen(code, **kwds):
-return cgen('\n'.join(code.split('\n')[1:-1]), **kwds)
+raw = cgen('\n'.join(code.split('\n')[1:-1]), **kwds)
+return re.sub(re.escape(eatspace) + ' +', '', raw)
I'd feel a little more confident with ' *' instead
On 04/28/2014 08:02 AM, Peter Maydell wrote:
Michael Tokarev (1):
libvixl: fix 64bit constants usage
Peter Maydell (4):
configure: Disable building A64 disassembler for libvixl update
disas/libvixl: Update to libvixl 1.3
disas/libvixl: Add missing ULL suffixes
Revert configure:
Juan Quintela quint...@redhat.com writes:
Hi
Please, send any topic that you are interested in covering.
[...]
I'd like to have these things settled sooner than five minutes before
the scheduled hour, so here goes: call or no call? Agenda?
On 28 April 2014 15:23, Michael S. Tsirkin m...@redhat.com wrote:
@@ -5179,7 +5187,7 @@ do
done
mkdir -p $DIRS
for f in $FILES ; do
-if [ -e $source_path/$f ] [ $source_path != `pwd` ]; then
+if [ -e $source_path/$f ] [ $pwd_is_source_path != y ]; then
symlink
101 - 200 of 298 matches
Mail list logo
