y.
Bonus: info qtree now shows the geometry.
Signed-off-by: Markus Armbruster
---
hw/ide/core.c | 19 ++-
hw/ide/internal.h |4 +++-
hw/ide/qdev.c | 22 +-
3 files changed, 38 insertions(+), 7 deletions(-)
diff --git a/hw/ide/core.c b/hw/ide/core.
Signed-off-by: Markus Armbruster
---
hw/hd-geometry.c | 32
1 files changed, 20 insertions(+), 12 deletions(-)
diff --git a/hw/hd-geometry.c b/hw/hd-geometry.c
index db47846..1a58894 100644
--- a/hw/hd-geometry.c
+++ b/hw/hd-geometry.c
@@ -97,14 +97,31
ranslation, set with -drive trans=...
Bonus: info qtree now shows the translation. Except when it shows
"auto": that's resolved by pc_cmos_init_late(). To be addressed
shortly.
Signed-off-by: Markus Armbruster
---
hw/ide/internal.h |1 +
hw/ide/qdev.c | 10 +-
nders the hints unused. Purge them from the block layer.
Signed-off-by: Markus Armbruster
---
block.c | 32
block.h | 12
block_int.h |1 -
blockdev.c | 14 ++
hw/block-common.h|
Currently, it is split between hd_geometry_guess() and
pc_cmos_init_late(). Confusing. info qtree shows the result of the
former. Also confusing.
Fold the part done in pc_cmos_init_late() into hd_geometry_guess().
Signed-off-by: Markus Armbruster
---
hw/block-common.h |1 +
hw/hd
0,1,1 1023,14,55 631032057
:12: 0,1,163,14,55 63 64377
The actual FAT file system nevertheless assumes the partition has
1032129 or 64449 sectors. Oops.
Signed-off-by: Markus Armbruster
---
block/vvfat.c |7 ---
1 files changed, 4
Leave it to ide_init_drive().
Signed-off-by: Markus Armbruster
---
blockdev.c |6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/blockdev.c b/blockdev.c
index 06c997e..5f8677e 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -330,15 +330,15 @@ DriveInfo *drive_init
now obeyed, and the
default hard disk geometry with :floppy: now matches the one without
format=vvfat.
Signed-off-by: Markus Armbruster
---
block/vvfat.c | 52
1 files changed, 28 insertions(+), 24 deletions(-)
diff --git a/block/vvfat.c b/blo
Signed-off-by: Markus Armbruster
---
hw/hd-geometry.c | 84 +++---
1 files changed, 42 insertions(+), 42 deletions(-)
diff --git a/hw/hd-geometry.c b/hw/hd-geometry.c
index f0dd021..db47846 100644
--- a/hw/hd-geometry.c
+++ b/hw/hd-geometry.c
So far covers only IDE and tests only CMOS contents.
Signed-off-by: Markus Armbruster
---
tests/Makefile |2 +
tests/hd-geo-test.c | 403 +++
2 files changed, 405 insertions(+), 0 deletions(-)
create mode 100644 tests/hd-geo-test.c
Signed-off-by: Markus Armbruster
---
block.h |8 +++-
1 files changed, 7 insertions(+), 1 deletions(-)
diff --git a/block.h b/block.h
index 993894e..1cd8a01 100644
--- a/block.h
+++ b/block.h
@@ -426,6 +426,8 @@ typedef struct BlockConf {
uint32_t opt_io_size;
int32_t
In preparation of purging it from the block layer, which will happen
later in this series.
Signed-off-by: Markus Armbruster
---
blockdev.c |4
blockdev.h |1 +
2 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/blockdev.c b/blockdev.c
index a85a429..161985b 100644
--- a
Signed-off-by: Markus Armbruster
---
hw/hd-geometry.c |7 +++
trace-events |4
2 files changed, 7 insertions(+), 4 deletions(-)
diff --git a/hw/hd-geometry.c b/hw/hd-geometry.c
index c45eafd..f0dd021 100644
--- a/hw/hd-geometry.c
+++ b/hw/hd-geometry.c
@@ -32,6 +32,7
New limits straight from ATA4 6.2 Register delivered data transfer
command sector addressing.
I figure the old sector limit 63 was blindly copied from the BIOS
int 13 limit. Doesn't apply to the hardware. No idea where the old
cylinder limit comes from.
Signed-off-by: Markus Armbr
Signed-off-by: Markus Armbruster
---
hw/hd-geometry.c | 22 --
1 files changed, 8 insertions(+), 14 deletions(-)
diff --git a/hw/hd-geometry.c b/hw/hd-geometry.c
index 1a58894..fb849a3 100644
--- a/hw/hd-geometry.c
+++ b/hw/hd-geometry.c
@@ -119,8 +119,7 @@ static void
"Daniel P. Berrange" writes:
> On Mon, Jul 09, 2012 at 02:17:04PM -0500, miny...@acm.org wrote:
>> diff --git a/qemu-options.hx b/qemu-options.hx
>> index 125a4da..823f6bc 100644
>> --- a/qemu-options.hx
>> +++ b/qemu-options.hx
>> @@ -2204,6 +2204,41 @@ Three button serial mouse. Configure the g
Scenario:
1. Start a guest with a blank disk (need not be the only disk) and
default disk geometry.
2. Examine the disk's physical geometry
Details depend on the device model. scsi-hd exposes it in mode pages
4 and 5. ide-hd in command IDENTIFY, and in its translation from CHS
to L
Luiz Capitulino writes:
> It allows for specifying an alias for each option name, see next commits
> examples.
>
> Signed-off-by: Luiz Capitulino
> ---
> qemu-option.c | 9 -
> qemu-option.h | 1 +
> 2 files changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/qemu-option.c b/qemu-o
Luiz Capitulino writes:
> Today, machine options use underscores to separate words (eg. kernel_irqchip),
> however upcoming QOM conversion wants to use dashes instead.
Command line users will appreciate consistent use of '-', too.
> This series converts all machine type options to use dashes. C
Blue Swirl writes:
> On Tue, Jul 10, 2012 at 6:34 PM, Markus Armbruster wrote:
>> Scenario:
>>
>> 1. Start a guest with a blank disk (need not be the only disk) and
>>default disk geometry.
>>
>> 2. Examine the disk's physical geometry
>>
A quick git-grep for more instances of '_' in option parameter names
finds only vcard_emul_options(). Yet another ad hoc option argument
parser. I'm not asking you to do anything about it.
x27;t limit serial number length, also uses
DriveInfo for -usbdevice. But that doesn't provide access to
DriveInfo serial.
Signed-off-by: Markus Armbruster
---
blockdev.c|4 +---
blockdev.h|4 +---
hw/block-common.c |2 +-
hw/ide/core.c |6 +++---
4 files
Applies on top of "[PATCH v3 00/29] Disk geometry cleanup".
Markus Armbruster (4):
hw/block-common: Move BlockConf & friends from block.h
hw/block-common: Factor out fall back to legacy -drive serial=...
blockdev: Don't limit DriveInfo serial to 20 characters
hw/block
This stuff doesn't belong to block layer, and was put there only
because a better home didn't exist then. Now it does.
Signed-off-by: Markus Armbruster
---
block.h | 45 -
hw/block-commo
Signed-off-by: Markus Armbruster
---
hw/Makefile.objs |2 +-
hw/block-common.c| 24
hw/block-common.h|3 +++
hw/ide/qdev.c| 12 ++--
hw/scsi-disk.c |8 +---
hw/usb/dev-storage.c | 10 ++
hw/virtio-blk.c
Kevin Wolf writes:
> Am 05.07.2012 18:39, schrieb Markus Armbruster:
>> Markus Armbruster writes:
>>
>>> Kevin Wolf writes:
>>>
>>>> Am 29.06.2012 17:34, schrieb Markus Armbruster:
>>>>> New limits straight from ATA4 6.2 Regi
Signed-off-by: Markus Armbruster
---
hw/block-common.c | 40
hw/block-common.h |2 ++
hw/ide/core.c | 24
hw/ide/qdev.c | 19 ++-
hw/scsi-disk.c| 31 +++
hw
Kevin Wolf writes:
> Am 10.07.2012 11:12, schrieb Markus Armbruster:
>> Commit f3d54fc4 factored it out of hw/ide.c for reuse. Sensible,
>> except it was put into block.c. Device-specific functionality should
>> be kept in device code, not the block layer. Move it to
>
[Alex's illegibly long lines wrapped]
Alexander Graf writes:
> On 09.07.2012, at 10:50, Markus Armbruster wrote:
>
>> Alexander Graf writes:
>>
>>> We've had support for creating AHCI devices using -device for a while now,
>>> but it's cumb
Gerd Hoffmann writes:
> Hi,
>
>> What I'd like to avoid is casual users setting up new guests with our
>> shiny new q35 board getting their IDE drives connected to some slow, old
>> controller just because they haven't discovered that if=ide doesn't cut
>> it anymore, and they need to use if=ah
Alexander Graf writes:
> On 12.07.2012, at 10:17, Markus Armbruster wrote:
>
>> [Alex's illegibly long lines wrapped]
>>
>> Alexander Graf writes:
>>
>>> On 09.07.2012, at 10:50, Markus Armbruster wrote:
>>>
>>>> Alexander Graf
Alexander Graf writes:
> On 12.07.2012, at 13:09, Markus Armbruster wrote:
>
>> Alexander Graf writes:
>>
>>> On 12.07.2012, at 10:17, Markus Armbruster wrote:
>>>
>>>> [Alex's illegibly long lines wrapped]
>>>>
>>
le of 4096:
$ qemu-system-x86_64 -S -vnc :0 -bios nonexistant
qemu: PC system firmware (pflash) must be a multiple of 0x1000
[Exit 1 ]
Fix by handling the qemu_find_file() failure.
Signed-off-by: Markus Armbruster
---
v2: Don't report errno, because that can produce misleading
Blue Swirl writes:
> On Wed, Dec 5, 2012 at 7:41 PM, Hans de Goede wrote:
>> Hi,
>>
>>
>> On 12/05/2012 08:28 PM, Blue Swirl wrote:
>>>
>>> On Tue, Dec 4, 2012 at 10:00 PM, Anthony Liguori
>>> wrote:
Peter Maydell writes:
> On 4 December 2012 18:38, Blue Swirl wrote:
>>
Gerhard Wiesinger writes:
> On 03.12.2012 21:51, Anthony Liguori wrote:
>> Hi,
>>
>> Major features include:
>> - After nearly 6 years of work, all remaining differences between the
>> qemu-kvm.git and qemu.git have been merged into qemu.git
>>
>
> How is qemu-kvm enabled?
> --enable-kvm ?
Gerhard Wiesinger writes:
> On 07.12.2012 08:37, Markus Armbruster wrote:
>> Gerhard Wiesinger writes:
>>
>>> On 03.12.2012 21:51, Anthony Liguori wrote:
>>>> Hi,
>>>>
>>>> Major features include:
>>>>- After near
Stefan Hajnoczi writes:
> On Fri, Dec 07, 2012 at 11:49:49AM +0800, liguang wrote:
>> if value to be translated is larger than INT64_MAX,
>> this function will not be convenient for caller to
>> be aware of it, so change a little for this.
>>
>> Signed-off-by: liguang
>> ---
>> cutils.c |5
liguang writes:
> qemu-img will complain when qcow or qcow2
> size overflow for 64 bits, report the right
> message in this condition.
Recommend to show the reproducer with output before and after the patch
in the commit message.
Please-separate-words-with-spaces-in-your-subject-line-thank-you-:)
Igor Mammedov writes:
> On Fri, 7 Dec 2012 11:49:49 +0800
> liguang wrote:
>
>> if value to be translated is larger than INT64_MAX,
>> this function will not be convenient for caller to
>> be aware of it, so change a little for this.
>>
>> Signed-off-by: liguang
>> ---
>> cutils.c |5 +++-
Igor Mammedov writes:
> On Thu, 13 Dec 2012 09:03:50 +0100
> Markus Armbruster wrote:
>
>> Igor Mammedov writes:
>>
>> > On Fri, 7 Dec 2012 11:49:49 +0800
>> > liguang wrote:
>> >
>> >> if value to be translated is larger than INT6
liguang writes:
> if value to be translated is larger than INT64_MAX,
> this function will not be convenient for caller to
> be aware of it, so change a little for this.
>
> Signed-off-by: liguang
> ---
> cutils.c |3 ++-
> 1 files changed, 2 insertions(+), 1 deletions(-)
>
> diff --git a/c
liguang writes:
> qemu-img will complain when qcow or qcow2
> size overflow for 64 bits, report the right
> message in this condition.
>
> before change:
> qemu-img: Invalid image size specified! You may use k, M, G or T suffixes for
> qemu-img: kilobytes, megabytes, gigabytes and terabytes.
>
>
Gerd Hoffmann writes:
> https://bugzilla.redhat.com/show_bug.cgi?id=885644
>
> Reported-by: Markus Armbruster
> Signed-off-by: Gerd Hoffmann
Spotted by Coverity.
Impact of the bug unknown; would be nice to have in the commit message.
Reviewed-by: Markus Armbruster
"Michael S. Tsirkin" writes:
> On Fri, Dec 14, 2012 at 09:37:17PM +, Blue Swirl wrote:
>> Perhaps the use of '-' vs. '_' in file names could be unified while
>> renaming. I think most new files use dash.
Yes, please!
> I can do this in a follow up patch but first let's put this rule in
> co
"Michael S. Tsirkin" writes:
> On Mon, Dec 17, 2012 at 11:59:59AM +0100, Markus Armbruster wrote:
>> "Michael S. Tsirkin" writes:
>>
>> > On Fri, Dec 14, 2012 at 09:37:17PM +, Blue Swirl wrote:
>> >> Perhaps the use of '-&#x
code several times in the past
> (cpkt->id not set before send_control_event() passed it on to
> send_control_msg()). This will finally eliminate the false-positive.
>
> CC: Markus Armbruster
> Signed-off-by: Amit Shah
Patch makes sense to me, and the Coverity defect
Signed-off-by: Markus Armbruster
---
hw/ide/core.c | 2 +-
hw/ide/internal.h | 3 +--
hw/ide/qdev.c | 12 +++-
3 files changed, 5 insertions(+), 12 deletions(-)
diff --git a/hw/ide/core.c b/hw/ide/core.c
index 2c0c978..1e94dc4 100644
--- a/hw/ide/core.c
+++ b/hw/ide/core.c
a non-qdevified controller. The block configuration
stuff I'm working on also profits from it, and is real reason I'm
posting this.
Markus Armbruster (10):
ide: Break all non-qdevified controllers
ide: Move IDEDevice pointer from IDEBus to IDEState
ide: Use IDEState member dev for &q
copy was used
only when the copy of bs was non-null, thus no bug there.
Signed-off-by: Markus Armbruster
---
hw/ide/core.c | 18 --
hw/ide/internal.h | 4 +---
hw/ide/qdev.c | 3 +--
3 files changed, 10 insertions(+), 15 deletions(-)
diff --git a/hw/ide/core.c b/hw/id
the copy was used
only when the copy of bs was non-null, thus no bug there.
Signed-off-by: Markus Armbruster
---
hw/ide/core.c | 14 --
hw/ide/internal.h | 3 +--
hw/ide/qdev.c | 10 +-
3 files changed, 10 insertions(+), 17 deletions(-)
diff --git a/hw/ide/core.c
the copy was used
only when the copy of bs was non-null, thus no bug there.
Signed-off-by: Markus Armbruster
---
hw/ide/core.c | 22 +++---
hw/ide/internal.h | 2 --
hw/ide/qdev.c | 16 +++-
3 files changed, 18 insertions(+), 22 deletions(-)
diff --git a/hw/id
the copy was used
only when the copy of bs was non-null, thus no bug there.
Signed-off-by: Markus Armbruster
---
hw/ide/atapi.c| 2 +-
hw/ide/core.c | 14 --
hw/ide/internal.h | 3 +--
hw/ide/qdev.c | 9 +
4 files changed, 11 insertions(+), 17 deletions(-)
diff
Testing dev is more obvious than testing bs, in my opinion.
Signed-off-by: Markus Armbruster
---
hw/ide/ahci.c | 8
hw/ide/core.c | 56 -
hw/ide/microdrive.c | 5 +++--
hw/ide/qdev.c | 2 +-
4 files changed, 38
). Begs the question whether stale data could
have been used after unplug. As far as I can tell, the copies were
used only when the copy of bs was non-null, thus no bug there.
Signed-off-by: Markus Armbruster
---
hw/ide/core.c | 52
hw/id
It's a copy of dev->conf.bs. The copy was needed for non-qdevified
controllers, which lacked dev.
Note how pci_piix3_xen_ide_unplug() cleared the copy. We'll get back
to that in the next few commits.
Signed-off-by: Markus Armbruster
---
hw/ide/ahci.c | 11 +-
h
All uses are simple array subscripts.
Signed-off-by: Markus Armbruster
---
hw/ide/core.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/hw/ide/core.c b/hw/ide/core.c
index 5876862..66e56c2 100644
--- a/hw/ide/core.c
+++ b/hw/ide/core.c
@@ -2006,8 +2006,7 @@ static void
e writing has been on the wall for a few years.
Signed-off-by: Markus Armbruster
---
hw/ide/core.c | 43 +--
1 file changed, 9 insertions(+), 34 deletions(-)
diff --git a/hw/ide/core.c b/hw/ide/core.c
index c4f93d0..2c0c978 100644
--- a/hw/ide/core.c
+++ b/
Alexander Graf writes:
> On 17.12.2012, at 15:43, Markus Armbruster wrote:
>
>> Alexander Graf writes:
>>
>>> On 17.12.2012, at 15:05, Markus Armbruster wrote:
>>>
>>>> They complicate IDE data structures and keep getting in the way.
>>
Alexander Graf writes:
> On 17.12.2012, at 15:05, Markus Armbruster wrote:
>
>> They complicate IDE data structures and keep getting in the way.
>> Also, TRIM support (commit d353fb72) is broken for them, because
>> ide_identify() accesses IDEDevice member conf, but IDEDe
Alexander Graf writes:
> On 17.12.2012, at 16:15, Markus Armbruster wrote:
>
>> Alexander Graf writes:
>>
>>> On 17.12.2012, at 15:43, Markus Armbruster wrote:
>>>
>>>> Alexander Graf writes:
>>>>
>>>>> On 17.12.2012
Amit Shah writes:
> On (Mon) 17 Dec 2012 [14:14:17], Markus Armbruster wrote:
>> Amit Shah writes:
>>
>> > Stuff the cpkt before calling send_control_msg(). it should not be
>> > concerned about contents, just send across the buffer.
>> >
>> >
Amit Shah writes:
> On (Mon) 17 Dec 2012 [18:23:53], Markus Armbruster wrote:
>> >> However, it now worries find_port_by_id() in remove_port() could return
>> >> a null pointer, which is then dereferenced. No idea why it didn't
>> >
Amit Shah writes:
> remove_port() is called from qdev's unplug callback, and we're certain
> the port will be found in our list of ports. Adding an assert()
> documents this.
>
> This was flagged by Coverity, fix suggested by Markus.
>
> CC: Markus Armbrus
Andreas Färber writes:
> Am 17.12.2012 15:43, schrieb Markus Armbruster:
>> Alexander Graf writes:
>>
>>> On 17.12.2012, at 15:05, Markus Armbruster wrote:
>>>
>>>> They complicate IDE data structures and keep getting in the way.
>>>>
Peter Maydell writes:
> On 17 December 2012 14:05, Markus Armbruster wrote:
>> The writing has been on the wall for a few years.
>
> ...behind a filing cabinet in a disused lavatory with a sign on the door
> saying "beware of the leopard"?
>
> We really ne
Anthony Liguori writes:
> Markus Armbruster writes:
>
>> *** Important ***
>> This *breaks* all non-qdevified controllers, see PATCH 01/10.
>> Maintainers are cc'ed.
>>
>> If you want still more time to qdevify your controller, please speak
>> up n
Alexey Zaytsev writes:
> Hi.
>
> Would it be a good idea to provide some non-empty serial number for
> VIRTIO_BLK_T_GET_ID, when the user does not specify one?
> QEMU assigns a default serial to IDE drives, but not virtio.
SCSI devices don't get one either.
The IDE default serial numbers aren't
Paolo Bonzini writes:
> Two build fixes for configurations not covered by my and Anthony's testing.
Fixes build for me, thanks!
Lei Li writes:
> When controlling a qemu instance from another program, it's
> hard to know which serial port or monitor device is redirected
> to which pty. With more than one device using "pty" a lot of
> guesswork is involved.
>
> $ ./x86_64-softmmu/qemu-system-x86_64 -serial pty -serial pty -
Jason Baron writes:
> On Thu, Dec 20, 2012 at 08:07:02PM +, Blue Swirl wrote:
>> On Thu, Dec 20, 2012 at 5:14 PM, Jason Baron wrote:
>> > From: Jason Baron
[...]
>> > @@ -172,9 +173,11 @@ void qtest_quit(QTestState *s)
>> > unlink(s->pid_file);
>> > unlink(s->socket_path);
>> >
tch
> restores the behavior documented by Intel when cpuid_xlevel2 is 0.
>
> The existing behavior when cpuid_xlevel2 is set (falling back to
> level=cpuid_xlevel) is being kept, as I couldn't find any public
> documentation on the CPUID 0xC000 function range on Centaur CPUs
t;pty:%s", q_ptsname(master_fd));
> qemu_opt_set(opts, "path", q_ptsname(master_fd));
> -fprintf(stderr, "char device redirected to %s\n", q_ptsname(master_fd));
> +
> +label = qemu_opts_id(opts);
> + fprintf(stderr, "char device%s%s redirected to %s\n",
> +label ? " " : "",
> +label ?: "",
> +q_ptsname(master_fd));
>
> s = g_malloc0(sizeof(PtyCharDriver));
> chr->opaque = s;
Reviewed-by: Markus Armbruster
Lluís Vilanova writes:
> This interface decouples event obtention from interaction.
Are you sure "obtention" is a word? :)
> Events can be obtained through three different methods:
>
> * identifier
> * name
> * simple wildcard pattern
Wenchao Xia writes:
> This patch change all info call back function to take
> additional QDict * parameter, which allow those command
> take parameter.
[...]
> diff --git a/monitor.c b/monitor.c
> index c0e32d6..797680f 100644
> --- a/monitor.c
> +++ b/monitor.c
> @@ -123,7 +123,7 @@ typedef st
Wenchao Xia writes:
> This patch add an option to show snapshots on a single block
> device, so some snapshot do not exist on other block device
> could be shown.
>
> Signed-off-by: Wenchao Xia
> ---
> monitor.c |6 +++---
> savevm.c | 55 +++
Luiz Capitulino writes:
> On Wed, 19 Dec 2012 18:17:09 +0800
> Wenchao Xia wrote:
>
>> This patch enable sub info command handler getting meaningful
>> parameter.
>>
>> Signed-off-by: Wenchao Xia
>> ---
>> hmp-commands.hx |2 +-
>> monitor.c | 79
>>
Peter Lieven writes:
> Latest git doesn't compile with this patch:
>
> hw/fw_cfg.c: In function 'fw_cfg_write':
> hw/fw_cfg.c:207:5: error: implicit declaration of function
> trace_fw_cfg_write' [-Werror=implicit-function-declaration]
> hw/fw_cfg.c:207:5: error: nested extern declaration of
> tra
Kevin Wolf writes:
> Add comments to help static analysers detect that these cases are
> intentional, and clean up some whitespace in the environment of these
> comments.
>
> Signed-off-by: Kevin Wolf
Reviewed-by: Markus Armbruster
Kevin Wolf writes:
> Return -errno instead of -1 on errors. While touching the
> code, fix a memory leak.
>
> Signed-off-by: Kevin Wolf
> ---
> block/bochs.c | 22 +++---
> 1 files changed, 15 insertions(+), 7 deletions(-)
>
> diff --git a/block/bochs.c b/block/bochs.c
> index
Peter Lieven writes:
> Am 24.01.2013 um 11:32 schrieb Markus Armbruster :
>
>> Peter Lieven writes:
>>
>>> Latest git doesn't compile with this patch:
>>>
>>> hw/fw_cfg.c: In function 'fw_cfg_write':
>>> hw/fw_cfg.c:
Kevin Wolf writes:
> Return -errno instead of -1 on errors. While touching the
> code, fix a memory leak.
>
> Signed-off-by: Kevin Wolf
> ---
> block/cloop.c | 27 +--
> 1 files changed, 17 insertions(+), 10 deletions(-)
>
> diff --git a/block/cloop.c b/block/cloop.c
>
Kevin Wolf writes:
> Return -errno instead of -1 on errors. While touching the
> code, fix a memory leak.
>
> Signed-off-by: Kevin Wolf
> ---
> block/vpc.c | 36 +---
> 1 files changed, 25 insertions(+), 11 deletions(-)
>
> diff --git a/block/vpc.c b/block/vpc.
Kevin Wolf writes:
> Return -errno instead of -1 on errors. While touching the
> code, fix even some more memory leaks than in the other drivers...
LOL
> Signed-off-by: Kevin Wolf
> ---
> block/dmg.c | 136 --
> 1 files changed, 103 ins
Kevin Wolf writes:
> The buffers are allocated with g_(re)alloc, so use g_free to free them.
>
> Signed-off-by: Kevin Wolf
> ---
> block/dmg.c | 16 +---
> 1 files changed, 9 insertions(+), 7 deletions(-)
>
> diff --git a/block/dmg.c b/block/dmg.c
> index d47d2d8..316d3c0 100644
>
Kevin Wolf writes:
> Return -errno instead of -1 on errors. Hey, no memory leak to fix here
> while we're touching it!
I'm flabbergasted!
> Signed-off-by: Kevin Wolf
> ---
> block/parallels.c | 23 +++
> 1 files changed, 15 insertions(+), 8 deletions(-)
>
> diff --git a/
Dong Xu Wang writes:
> qemu_opts_print has no user now, so can re-write the function safely.
>
> qemu_opts_print will be used while using "qemu-img create", it will
> produce the same output as previous code.
>
> The behavior of this function has changed:
>
> 1. Print every possible option, wheth
Dong Xu Wang writes:
> This patch will create 4 functions, count_opts_list, append_opts_list,
> free_opts_list and print_opts_list, they will used in following commits.
>
> Signed-off-by: Dong Xu Wang
> ---
> v6->v7):
> 1) Fix typo.
>
> v5->v6):
> 1) allocate enough space in append_opts_list fun
Eric Blake writes:
> On 01/15/2013 03:24 AM, Wenchao Xia wrote:
>
+int bdrv_snapshot_find(BlockDriverState *bs, QEMUSnapshotInfo *sn_info,
+ const char *name)
>
+if (!strcmp(sn->id_str, name) || !strcmp(sn->name, name)) {
>>>
>
>>>
>>> This code c
Philipp Hahn writes:
> Hello,
>
> I tried to open a "twoGbMaxExtentSparse" VMDK file, which uses spaces in its
> own and for the referenced file names. This breaks in line 646 of
> block/vmdk.c because "%511s" stops at the first space and thus fname is
> incomplete:
> ret = sscanf(p, "
Laszlo Ersek writes:
> On 01/25/13 09:14, Amos Kong wrote:
>> FD_SET() and FD_CLR() are used to add and remove one descriptor from a
>> set, the 'fd' should be less than FD_SETSIZE. Glibc will give a warning
>> and crash the qemu when we set a fd (1024) to a set.
>>
>> # qemu -device virtio-net
Tim Hardeck writes:
> Hi Markus,
>
> thanks for your input.
>
> On Wed, 2013-01-23 at 18:16 +0100, Markus Armbruster wrote:
>> Use appropriate types, drop superfluous casts, use sizeof, don't
>> exploit that this particular call of gnutls_fingerprint() doe
Use appropriate types, drop superfluous casts, use sizeof, don't
exploit that this particular call of gnutls_fingerprint() doesn't
change its last argument.
Signed-off-by: Markus Armbruster
---
ui/vnc-ws.c | 11 ++-
1 file changed, 6 insertions(+), 5 deletions(-)
diff --gi
Kevin Wolf writes:
> Am 24.01.2013 14:09, schrieb Markus Armbruster:
>> Kevin Wolf writes:
>>
>>> Return -errno instead of -1 on errors. While touching the
>>> code, fix a memory leak.
>>>
>>> Signed-
Michal Privoznik writes:
> Currently, we are using 'tray_open' in QMP and 'tray-open' in
> HMP. However, the QMP documentation was mistakenly using the
> HMP version.
Could you point me to where we're using "tray_open"?
You screwed up your e-mail address: vnet.linux.ibm.com doesn't exist.
My review bounced. Please find it on the list.
More old news: casting pointers considered harmful, and atomics
considered hard to use correctly.
Markus Armbruster (4):
trace: Fix simple trace dropped event record for big endian
trace: Direct access of atomics is verboten, use the API
trace: Clean up the "try to update atomic unt
GLib's
atomic int operations work just fine for unsigned. Drop it.
Signed-off-by: Markus Armbruster
---
trace/simple.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/trace/simple.c b/trace/simple.c
index ce17d64..ccbdb6a 100644
--- a/trace/simple.c
--
always use this API.
Signed-off-by: Markus Armbruster
---
trace/simple.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/trace/simple.c b/trace/simple.c
index ccbdb6a..592ff48 100644
--- a/trace/simple.c
+++ b/trace/simple.c
@@ -166,13 +166,13 @@ static gpointer
Eric Blake writes:
> On 01/25/2013 07:07 AM, Michal Privoznik wrote:
>> On 25.01.2013 14:55, Markus Armbruster wrote:
>>> Michal Privoznik writes:
>>>
>>>> Currently, we are using 'tray_open' in QMP and 'tray-open' in
>>>>
401 - 500 of 28273 matches
Mail list logo
