http://blog.gmane.org/gmane.comp.emulators.qemu hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://comments.gmane.org/gmane.comp.emulators.qemu/152690 <pre> Paolo, list Please find a set of patches for iscsi. 1, The recent change to the eventsystem in iscsi is racy. If the async connect of the socket takes longer than almost instant, the shortcircuit trying to write directly to the socket may try to do so before the socket is established. So we hcange the logic here and ONLY try the shortcircuit avoiding setting up POLLOUT if we have been fully logged in. This avoids the race where we might crash if the socket is not yet connected. 2, Swithc to using READCAPACITY10 for MMC devices. Only SBC devices support READCAPACITY16 so far. 3, Switch to use READ10 for MMC devices instead of READ16. With this patch ISCSI and MMC devices now work correctly. 4, If the device is a SMC device, then force the device to be sg. We dont support emulation as far as i know for SMV devices, so these device types would only work for sg or iscsi passthrough anyway. The reason we need to forcer sg is that since we can not READ from a SMC device, we need to shortcircuit find_image_format() before it tries to do the bdrv_pread() trying to guess what format the image is. With this patch I have verified I can access and manage a media changer via iscsi-passthrough. regards ronnie sahblerg </pre> Ronnie Sahlberg 2012-05-26T04:56:37 http://comments.gmane.org/gmane.comp.emulators.qemu/152687 <pre>I found a GSOC11 project that attempted to port the Android "Goldfish" platform to mainline QEMU. Was this project successful, or is this currently being worked on? </pre> Ira Ray Jenkins 2012-05-25T22:13:25 http://comments.gmane.org/gmane.comp.emulators.qemu/152671 <pre>Converting the screendump command is simple and shouldn't take more than or or two patches, the complicated part is to report all errors correctly. I hope I didn't go too far there, but at least this series does the right thing (or is very near to). console.c | 7 ++-- console.h | 5 +-- cutils.c | 100 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ hmp-commands.hx | 5 ++- hmp.c | 9 +++++ hmp.h | 1 + hw/blizzard.c | 4 +-- hw/g364fb.c | 51 ++++++++++++++++++++-------- hw/omap_lcdc.c | 60 +++++++++++++++++++++++---------- hw/qxl.c | 7 ++-- hw/tcx.c | 96 ++++++++++++++++++++++++++++++++++++++++------------ hw/vga.c | 38 +++++++++++++-------- hw/vga_int.h | 3 +- hw/vmware_vga.c | 7 ++-- monitor.c | 6 ---- qapi-schema.json | 24 +++++++++++++ qemu-common.h | 7 ++++ qerror.c | 28 ++++++++++++++-- qerror.h | 22 ++++++++++-- qmp-commands.hx | 5 +-- 20 files changed, 390 insertions(+), 95 deletions(-) </pre> Luiz Capitulino 2012-05-25T19:41:05 http://comments.gmane.org/gmane.comp.emulators.qemu/152668 <pre>Hello, Here's the 2nd iteration of the virtio hwrng device. This device just accepts data from the chardev and sends it up to the guest. There's no policy here on restricting the number of bytes we send to the guest, as that can be done by management apps. libvirt has a better picture of the host's entropy and multiple guests, so it can better make decisions on when to supply entropy, and how much. Of course, Linux has a low watermark, via /proc/sys/kernel/random/entropy_avail and it will stop giving out entropy once that watermark is reached (currently 64). /dev/urandom doesn't use all the bits at once, so using that device to feed in entropy to the guest is safer and faster (without compromising on the quality of randomness). libvirt can also use daemons like egd to feed in entropy. v2: * Remove hard-wiring to /dev/urandom * Use chardev for input * Add a QMP event for notifying listeners about entropy needed and the bytes asked for by the guest. * Add s390 code Please review, and apply if appropriate. Amit Shah (1): virtio-rng: hardware random number generator device Makefile.objs | 1 + hw/pci.h | 1 + hw/s390-virtio-bus.c | 35 +++++++++ hw/s390-virtio-bus.h | 2 + hw/virtio-pci.c | 51 +++++++++++++ hw/virtio-pci.h | 2 + hw/virtio-rng.c | 199 ++++++++++++++++++++++++++++++++++++++++++++++++++ hw/virtio-rng.h | 24 ++++++ hw/virtio.h | 3 + monitor.c | 3 + monitor.h | 1 + 11 files changed, 322 insertions(+), 0 deletions(-) create mode 100644 hw/virtio-rng.c create mode 100644 hw/virtio-rng.h </pre> Amit Shah 2012-05-25T19:32:48 http://comments.gmane.org/gmane.comp.emulators.qemu/152654 <pre>The following changes since commit aeb29b6459cb9496b38c820f3faff64cf2369d0d: audio: Always call fini on exit (2012-05-24 19:35:27 +0400) are available in the git repository at: git://repo.or.cz/qemu/kevin.git for-anthony Jim Meyering (1): qcow2: don't leak buffer for unexpected qcow_version in header Kevin Wolf (1): qcow2: Check qcow2_alloc_clusters_at() return value MORITA Kazutaka (3): sheepdog: mark image as snapshot when tag is specified sheepdog: return -errno on error sheepdog: use heap instead of stack for BDRVSheepdogState Pavel Hrdina (3): fdc: floppy drive should be visible after start without media fdc: fix media detection fdc-test: introduced qtest no_media_on_start and cmos qtest for floppy Richard W.M. Jones (1): qemu-img: Explain how rebase operation can be used to perform a 'diff' operation. Stefan Hajnoczi (1): qemu-iotests: mark 035 qcow2-only block/qcow2-cluster.c | 23 ++++++---- block/qcow2.c | 3 +- block/sheepdog.c | 113 +++++++++++++++++++++++++++++------------------- hw/fdc.c | 14 +++--- hw/pc.c | 2 +- qemu-img.texi | 18 ++++++++ tests/fdc-test.c | 65 ++++++++++++++++++++------- tests/qemu-iotests/035 | 2 +- 8 files changed, 159 insertions(+), 81 deletions(-) </pre> Kevin Wolf 2012-05-25T17:38:44 http://comments.gmane.org/gmane.comp.emulators.qemu/152586 <pre> Hi, Here comes a patch series with a bunch of updates for the ehci host controller. Two important changes: First ehci supports queuing now, i.e. more than a single packet per endpoint can be in flight at the same time, which improves usb-host performance. Second the wakeup rate is adaptive now, so ehci will have alot less wakeups when the usb bus is idle. Also a bunch of minor fixes and cleanups. Post-freeze material. please review, Gerd Gerd Hoffmann (18): ehci: add EHCIPacket ehci: make ehci_execute work on EHCIPacket instead of EHCIQueue ehci: cache USBDevice in EHCIQueue ehci: move ehci_flush_qh ehci: add queuing support ehci: tweak queue initialization ehci: add async field to EHCIQueue ehci: move async schedule to bottom half ehci: schedule async bh on async packet completion ehci: kick async schedule on wakeup ehci: fix reset ehci: add ehci_*_enabled() helpers ehci: update status bits in ehci_set_state ehci: fix halt status handling ehci: remove unused attach_poll_counter ehci: create ehci_update_frindex ehci: adaptive wakeup rate. ehci: rework frame skipping hw/usb/hcd-ehci.c | 610 +++++++++++++++++++++++++++++++++++------------------ trace-events | 1 + 2 files changed, 408 insertions(+), 203 deletions(-) </pre> Gerd Hoffmann 2012-05-25T12:40:15 http://comments.gmane.org/gmane.comp.emulators.qemu/152573 <pre> Paolo, Thanks for your review! I have implemented your suggested changes and resubmit the patch. I didnt know about -device scsi-generic arguments but using it allowed to get rid of the bs-&gt;iscsi field altogether as you suggested and the patch is now a low cleaner and doesnt touch so many other areas anymore. The commit message contains an example fragment from how force the passthrough mode for iscsi on the commandline. I will update the manpage and documentation to illustrate how to use passthrough with iscsi in a later patch during the weekend or next week. Thanks ronnie sahlberg </pre> Ronnie Sahlberg 2012-05-25T11:59:00 http://comments.gmane.org/gmane.comp.emulators.qemu/152571 <pre>Make qemu_find_file() check for the passed in name as a straight pathname even if it doesn't have any path separator character in it. This means that "-bios foo", "-dtb foo" etc will find a file 'foo' in the current directory. This removes an inconsistency with -kernel and -initrd, which both accept plain filenames as meaning files in the current directory. It's also less confusing for the user than an undocumented restriction that "this option accepts a filename, except for the special case where the filename you pass happens not to have a '/' in it, in which case we'll ignore it." Signed-off-by: Peter Maydell &lt;peter.maydell&lt; at &gt;linaro.org&gt; --- vl.c | 5 ++--- 1 files changed, 2 insertions(+), 3 deletions(-) diff --git a/vl.c b/vl.c index 23ab3a3..4639526 100644 --- a/vl.c +++ b/vl.c &lt; at &gt;&lt; at &gt; -1801,9 +1801,8 &lt; at &gt;&lt; at &gt; char *qemu_find_file(int type, const char *name) const char *subdir; char *buf; - /* If name contains path separators then try it as a straight path. */ - if ((strchr(name, '/') || strchr(name, '\\')) - &amp;&amp; access(name, R_OK) == 0) { + /* Try the name as a straight path first */ + if (access(name, R_OK) == 0) { return g_strdup(name); } switch (type) { </pre> Peter Maydell 2012-05-25T12:07:01 http://comments.gmane.org/gmane.comp.emulators.qemu/152566 <pre>Any code that depends on a particular CPU type can now go through callbacks on the QOM UniCore32CPUClass. Signed-off-by: Andreas Färber &lt;afaerber&lt; at &gt;suse.de&gt; --- target-unicore32/cpu.h | 4 ---- 1 files changed, 0 insertions(+), 4 deletions(-) diff --git a/target-unicore32/cpu.h b/target-unicore32/cpu.h index 81c14ff..50d5695 100644 --- a/target-unicore32/cpu.h +++ b/target-unicore32/cpu.h &lt; at &gt;&lt; at &gt; -120,10 +120,6 &lt; at &gt;&lt; at &gt; void cpu_asr_write(CPUUniCore32State *env1, target_ulong val, target_ulong mask) #define UC32_HWCAP_CMOV 4 /* 1 &lt;&lt; 2 */ #define UC32_HWCAP_UCF64 8 /* 1 &lt;&lt; 3 */ -#define UC32_CPUID(env) (env-&gt;cp0.c0_cpuid) -#define UC32_CPUID_UCV2 0x40010863 -#define UC32_CPUID_ANY 0xffffffff - #define cpu_init uc32_cpu_init #define cpu_exec uc32_cpu_exec #define cpu_signal_handler uc32_cpu_signal_handler </pre> Andreas Färber 2012-05-25T11:51:58 http://comments.gmane.org/gmane.comp.emulators.qemu/152565 <pre>The multifunction ich9 ehci controller with uhci companions uses a different interrupt pin for each function. The three uhci devices get pins A, B and C, whereas ehci uses pin D. This way the guest can assign different IRQ lines to each controller. Signed-off-by: Gerd Hoffmann &lt;kraxel&lt; at &gt;redhat.com&gt; --- hw/usb/hcd-uhci.c | 21 +++++++++++++++++++-- 1 files changed, 19 insertions(+), 2 deletions(-) diff --git a/hw/usb/hcd-uhci.c b/hw/usb/hcd-uhci.c index 3ea388c..9871e24 100644 --- a/hw/usb/hcd-uhci.c +++ b/hw/usb/hcd-uhci.c &lt; at &gt;&lt; at &gt; -138,6 +138,7 &lt; at &gt;&lt; at &gt; struct UHCIState { /* Interrupts that should be raised at the end of the current frame. */ uint32_t pending_int_mask; + int irq_pin; /* Active packets */ QTAILQ_HEAD(, UHCIQueue) queues; &lt; at &gt;&lt; at &gt; -340,7 +341,7 &lt; at &gt;&lt; at &gt; static void uhci_update_irq(UHCIState *s) } else { level = 0; } - qemu_set_irq(s-&gt;dev.irq[3], level); + qemu_set_irq(s-&gt;dev.irq[s-&gt;irq_pin], level); } static void uhci_reset(void *opaque) &lt; at &gt;&lt; at &gt; -1184,15 +1185,31 &lt; at &gt;&lt; at &gt; static USBBusOps uhci_bus_ops = { static int usb_uhci_common_initfn(PCIDevice *dev) { + PCIDeviceClass *pc = PCI_DEVICE_GET_CLASS(dev); UHCIState *s = DO_UPCAST(UHCIState, dev, dev); uint8_t *pci_conf = s-&gt;dev.config; int i; pci_conf[PCI_CLASS_PROG] = 0x00; /* TODO: reset value should be 0. */ - pci_conf[PCI_INTERRUPT_PIN] = 4; /* interrupt pin D */ pci_conf[USB_SBRN] = USB_RELEASE_1; // release number + switch (pc-&gt;device_id) { + case PCI_DEVICE_ID_INTEL_82801I_UHCI1: + s-&gt;irq_pin = 0; /* A */ + break; + case PCI_DEVICE_ID_INTEL_82801I_UHCI2: + s-&gt;irq_pin = 1; /* B */ + break; + case PCI_DEVICE_ID_INTEL_82801I_UHCI3: + s-&gt;irq_pin = 2; /* C */ + break; + default: + s-&gt;irq_pin = 3; /* D */ + break; + } + pci_config_set_interrupt_pin(pci_conf, s-&gt;irq_pin + 1); + if (s-&gt;masterbus) { USBPort *ports[NB_PORTS]; for(i = 0; i &lt; NB_PORTS; i++) { </pre> Gerd Hoffmann 2012-05-25T11:50:57 http://comments.gmane.org/gmane.comp.emulators.qemu/152545 <pre>These patches implement softmmu support on unicore32 architecture. UniCore32 CPU is embedded in PKUnity-3 SoC, so we add hw/pkunity dir to contain all puv3 devices simulation codes together. Only minimal system control modules are simulated, to make linux kernel boot and busybox run in initramfs. Any advice is greatly appreciated. Thanks, Guan Xuetao --- Guan Xuetao (9): unicore32-softmmu: make exceptions generated in unicore32 architecture unicore32-softmmu: implement some softmmu specific functions unicore32-softmmu: add coprocessor 0 and ocd instruction support unicore32-softmmu: make sure that kernel can access user space unicore32-softmmu: initialize ucv2 cpu unicore32-softmmu: add generic cpu state save/load functions unicore32-softmmu: add puv3 soc support unicore32-softmmu: add config and makefile support unicore32-softmmu: add maintainer information MAINTAINERS | 7 + Makefile.target | 5 + arch_init.c | 2 + arch_init.h | 1 + configure | 4 + cpu-exec.c | 1 + default-configs/unicore32-softmmu.mak | 4 + hw/pkunity/puv3.c | 130 ++++++++++++++++ hw/pkunity/puv3.h | 49 ++++++ hw/pkunity/puv3_dma.c | 109 +++++++++++++ hw/pkunity/puv3_gpio.c | 141 +++++++++++++++++ hw/pkunity/puv3_intc.c | 135 +++++++++++++++++ hw/pkunity/puv3_ost.c | 151 +++++++++++++++++++ hw/pkunity/puv3_pm.c | 148 ++++++++++++++++++ linux-user/main.c | 3 +- target-unicore32/cpu.c | 17 ++- target-unicore32/cpu.h | 8 +- target-unicore32/helper.c | 48 +----- target-unicore32/helper.h | 15 +- target-unicore32/machine.c | 99 ++++++++++++ target-unicore32/op_helper.c | 187 +++++++++++++++++++++++- target-unicore32/translate.c | 106 ++++++++++++- target-unicore32/uc32_softmmu.c | 267 +++++++++++++++++++++++++++++++++ 23 files changed, 1568 insertions(+), 69 deletions(-) create mode 100644 default-configs/unicore32-softmmu.mak create mode 100644 hw/pkunity/puv3.c create mode 100644 hw/pkunity/puv3.h create mode 100644 hw/pkunity/puv3_dma.c create mode 100644 hw/pkunity/puv3_gpio.c create mode 100644 hw/pkunity/puv3_intc.c create mode 100644 hw/pkunity/puv3_ost.c create mode 100644 hw/pkunity/puv3_pm.c create mode 100644 target-unicore32/machine.c create mode 100644 target-unicore32/uc32_softmmu.c </pre> Guan Xuetao 2012-05-25T11:28:58 http://comments.gmane.org/gmane.comp.emulators.qemu/152532 <pre> Hi, This patch series brings some bandwidth management tweaks for uhci, and also a little uhci migration fix. It also brings live migration support for usb-storage, including an update for scsi-disk live migration support. Paolo, feel free to cherry-pick the scsi patch, I'll go drop it for the pull request then. Obviously post-freeze material. please review, Gerd The following changes since commit aeb29b6459cb9496b38c820f3faff64cf2369d0d: audio: Always call fini on exit (2012-05-24 19:35:27 +0400) are available in the git repository at: git://git.kraxel.org/qemu usb.51 Gerd Hoffmann (10): uhci: fix bandwidth management uhci: use bottom half uhci: make bandwidth tunable uhci: fix trace format strings uhci: zap uhci_pre_save scsi: prepare migration code for usb-storage support usb-storage: remove MSDState-&gt;residue usb-storage: add usb_msd_packet_complete() usb-storage: add scsi_off, remove scsi_buf usb-storage: migration support hw/scsi-bus.c | 8 +++--- hw/scsi-disk.c | 13 ++++++++- hw/usb/dev-storage.c | 69 ++++++++++++++++++++++++++++++++----------------- hw/usb/hcd-uhci.c | 46 +++++++++++++++++++------------- trace-events | 10 +++--- 5 files changed, 92 insertions(+), 54 deletions(-) </pre> Gerd Hoffmann 2012-05-25T09:44:11 http://comments.gmane.org/gmane.comp.emulators.qemu/152523 <pre> Paolo, Please find an updated patch that implements your suggestion. This patch updates how iscsi sets up the read and write events so that we only call qemu_aio_set_fd_handler() and qemu_notify_event() when we actually need to. regards ronnie sahlberg </pre> Ronnie Sahlberg 2012-05-25T10:25:11 http://comments.gmane.org/gmane.comp.emulators.qemu/152507 <pre> Hi, A bunch of xhci updates. Nothing earth-shaking: A little reset fix from Jan and support for tracepoints. Scheduled for post-freeze merging. cheers, Gerd Gerd Hoffmann (7): xhci: trace: mmio reads+writes xhci: trace: run+stop xhci: trace: irq + events xhci: trace: ring fetch xhci: trace: endpoints xhci: trace: transfers xhci: trace: slots Jan Kiszka (1): xhci: Clean up reset function hw/usb/hcd-xhci.c | 251 +++++++++++++++++++++++++++++------------------------ trace-events | 35 ++++++++ 2 files changed, 173 insertions(+), 113 deletions(-) </pre> Gerd Hoffmann 2012-05-25T09:47:26 http://comments.gmane.org/gmane.comp.emulators.qemu/152485 <pre> Paolo, list Please find two patches. Both are big so maybe best to only add to the post 1.1 branches. The first patch tries to be more intelligent when setting up the fd events and only call these functions if something has changed. The second patch adds support for passthrough of SCSI comamnds. Just like /dev/sg allows the guest to talk to a real scsi device directly without qemu emulating the commands, this patch adds the same capability when using iscsi devices. With this patch the guest can talk directly to the target iscsi devices. regards ronnie sahlberg </pre> Ronnie Sahlberg 2012-05-25T07:43:14 http://comments.gmane.org/gmane.comp.emulators.qemu/152484 <pre>LinkedIn ------------ Jagan Teki requested to add you as a connection on LinkedIn: ------------------------------------------ Jiajun, I'd like to add you to my professional network on LinkedIn. - Jagan Accept invitation from Jagan Teki http://www.linkedin.com/e/-kkb1ec-h2my9bzj-w/qTMmi8QEI_f3FNXUkL1mvZgy00BGYniwg3/blk/I310587048_11/1BpC5vrmRLoRZcjkkZt5YCpnlOt3RApnhMpmdzgmhxrSNBszYNclYUd30Te3kMcjd9bSERjRpHsPBHbPAVczgNcjoPczsLrCBxbOYWrSlI/EML_comm_afe/?hs=false&amp;tok=0frruqu87Rg5g1 View invitation from Jagan Teki http://www.linkedin.com/e/-kkb1ec-h2my9bzj-w/qTMmi8QEI_f3FNXUkL1mvZgy00BGYniwg3/blk/I310587048_11/34NnPwQc3sUdj0NcQALqnpPbOYWrSlI/svi/?hs=false&amp;tok=0wOvalp-XRg5g1 ------------------------------------------ Why might connecting with Jagan Teki be a good idea? Jagan Teki's connections could be useful to you: After accepting Jagan Teki's invitation, check Jagan Teki's connections to see who else you may know and who you might want an introduction to. Building these connections can create opportunities in the future. </pre> Jagan Teki via LinkedIn 2012-05-25T07:49:12 http://comments.gmane.org/gmane.comp.emulators.qemu/152478 <pre>Hello, Am new around here and am working with the QEMU in a project. I would like to ask if there is anyone that can let me know how to compile or build the images to be used to run the QEMU from the source code. I already got the source code from the wiki.qemu.org and I have the version 1.0.1. Now I would like to build it from there to start doing some test with my project. At the beginning I was working with the version 0.14.50 but it was already build with all the executable and everything and am having issues when I try to put the system into an Idle state using the "wait" so I was looking to use the newest version and go from there. I appreciate all the help and guidance you can give me since am kind of newbie with this :) Arturo C.V. </pre> Arturo CV 2012-05-24T22:08:30 http://comments.gmane.org/gmane.comp.emulators.qemu/152473 <pre>I encountered the serious bug on QEMU Guest Agent. environment Guest OS : RHEL 5.8 / 5.7 (i686) Guest Agent Version : qemu-1.1.0rc2 and rc3 I am trying to take snapshot via virsh snapshot-create-as command. And to freeze guest's filesystem and take snapshot is succeed. But after sending the thaw command to Guest, time error occurs on libvirt qemu agent because of not catch Guest's answer. In addition, its situation is worst because the Guest Filesystem is kept as frozen. The problem does not occur on RHEL 6.2 Guest OS and in about qemu-1.0 it does not occur. Regards MATSUDA Daiki </pre> MATSUDA, Daiki 2012-05-25T05:57:10 http://comments.gmane.org/gmane.comp.emulators.qemu/152467 <pre>Signed-off-by: Amos Kong &lt;akong&lt; at &gt;redhat.com&gt; --- qerror.c | 4 ++++ qerror.h | 3 +++ 2 files changed, 7 insertions(+), 0 deletions(-) diff --git a/qerror.c b/qerror.c index 5092fe7..a5b6fd4 100644 --- a/qerror.c +++ b/qerror.c &lt; at &gt;&lt; at &gt; -172,6 +172,10 &lt; at &gt;&lt; at &gt; static const QErrorStringTable qerror_table[] = { .desc = "Parameter '%(name)' expects %(expected)", }, { + .error_fmt = QERR_TOO_MANY_PARAMETERS, + .desc = "Too many parameters", + }, + { .error_fmt = QERR_INVALID_PASSWORD, .desc = "Password incorrect", }, diff --git a/qerror.h b/qerror.h index 4cbba48..5341ea6 100644 --- a/qerror.h +++ b/qerror.h &lt; at &gt;&lt; at &gt; -151,6 +151,9 &lt; at &gt;&lt; at &gt; QError *qobject_to_qerror(const QObject *obj); #define QERR_INVALID_PARAMETER_VALUE \ "{ 'class': 'InvalidParameterValue', 'data': { 'name': %s, 'expected': %s } }" +#define QERR_TOO_MANY_PARAMETERS \ + "{ 'class': 'TooManyParameters', 'data': {} }" + #define QERR_INVALID_PASSWORD \ "{ 'class': 'InvalidPassword', 'data': {} }" </pre> Amos Kong 2012-05-25T03:31:59 http://comments.gmane.org/gmane.comp.emulators.qemu/152434 <pre>The following changes since commit aeb29b6459cb9496b38c820f3faff64cf2369d0d: audio: Always call fini on exit (2012-05-24 19:35:27 +0400) are available in the git repository at: git://github.com/mdroth/qemu.git qga-pull-5-24-12 Luiz Capitulino (2): configure: check if environ is declared qemu-ga: Fix missing environ declaration configure | 19 +++++++++++++++++++ qga/commands-posix.c | 6 +++++- 2 files changed, 24 insertions(+), 1 deletions(-) </pre> Michael Roth 2012-05-24T18:52:39 http://comments.gmane.org/gmane.comp.emulators.qemu/152394 <pre>The 035 parallel aio write test relies on knowledge of qcow2 metadata layout to stress parallel L2 table accesses. This only works for qcow2 unless we add additional calculations for qed or other formats. Mark this test as qcow2-only. Note that the test is strictly speaking non-deterministic although the output produced is reliable with qcow2. This is because the aio_write command returns before the aio write request has completed. Completions can occur at any time afterwards and cause a message to be printed. Therefore the exact output of this test is not deterministic but we seem to get away with it for qcow2 (maybe due to coroutine and main loop scheduling). Signed-off-by: Stefan Hajnoczi &lt;stefanha&lt; at &gt;linux.vnet.ibm.com&gt; --- I was originally going to filter out the "qemu-io&gt; " prompt so that non-determinism is not an issue. But it turns out that aio write completion messages can be printed at any time too. Short of sorting the output before golden master comparison, we don't have a good solution here. Sorting output would be confusing since we normally don't do that in other tests. tests/qemu-iotests/035 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/qemu-iotests/035 b/tests/qemu-iotests/035 index 56616a1..9d2d347 100755 --- a/tests/qemu-iotests/035 +++ b/tests/qemu-iotests/035 &lt; at &gt;&lt; at &gt; -39,7 +39,7 &lt; at &gt;&lt; at &gt; trap "_cleanup; exit \$status" 0 1 2 3 15 . ./common.rc . ./common.filter -_supported_fmt generic +_supported_fmt qcow2 _supported_proto generic _supported_os Linux </pre> Stefan Hajnoczi 2012-05-24T15:26:51 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.comp.emulators.qemu