| Message ID | alpine.LFD.2.02.1201041752270.17663@bbs.intern |
|---|---|
| State | New |
| Headers | show |
Am 04.01.2012 17:54, schrieb Gerhard Wiesinger: > Bugfix after reboot when vmmouse was enabled and another OS which uses > e.g. PS/2 > mouse. Could you please be a bit more verbose on what symptoms that fixes? Thanks, Andreas > > Signed-off-by: Gerhard Wiesinger <lists@wiesinger.com> > --- > hw/vmmouse.c | 2 ++ > 1 files changed, 2 insertions(+), 0 deletions(-) > > diff --git a/hw/vmmouse.c b/hw/vmmouse.c > index 1113f33..f9f5b53 100644 > --- a/hw/vmmouse.c > +++ b/hw/vmmouse.c > @@ -254,6 +254,8 @@ static void vmmouse_reset(DeviceState *d) > > s->status = 0xffff; > s->queue_size = VMMOUSE_QUEUE_SIZE; > + > + vmmouse_disable(s); > } > > static int vmmouse_initfn(ISADevice *dev)
Testscenario: 1.) Boot e.g. OS with VMMouse support (e.g. Windows) 2.) reboot 3.) Boot e.g. OS without VMMouse support (e.g. DOS) => PS/2 mouse doesn't work any more. Fixes that issue. Reason is that VMMouse has priority and disables all other mouse entities and therefore must be disabled on reset. Ciao, Gerhard -- http://www.wiesinger.com/ On Wed, 4 Jan 2012, Andreas Färber wrote: > Am 04.01.2012 17:54, schrieb Gerhard Wiesinger: >> Bugfix after reboot when vmmouse was enabled and another OS which uses >> e.g. PS/2 >> mouse. > > Could you please be a bit more verbose on what symptoms that fixes? > > Thanks, > Andreas > >> >> Signed-off-by: Gerhard Wiesinger <lists@wiesinger.com> >> --- >> hw/vmmouse.c | 2 ++ >> 1 files changed, 2 insertions(+), 0 deletions(-) >> >> diff --git a/hw/vmmouse.c b/hw/vmmouse.c >> index 1113f33..f9f5b53 100644 >> --- a/hw/vmmouse.c >> +++ b/hw/vmmouse.c >> @@ -254,6 +254,8 @@ static void vmmouse_reset(DeviceState *d) >> >> s->status = 0xffff; >> s->queue_size = VMMOUSE_QUEUE_SIZE; >> + >> + vmmouse_disable(s); >> } >> >> static int vmmouse_initfn(ISADevice *dev) > > -- > SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany > GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg > > >
On Wed, 4 Jan 2012, Gerhard Wiesinger wrote: > Testscenario: > 1.) Boot e.g. OS with VMMouse support (e.g. Windows) > 2.) reboot > 3.) Boot e.g. OS without VMMouse support (e.g. DOS) => PS/2 mouse doesn't > work any more. Fixes that issue. > > Reason is that VMMouse has priority and disables all other mouse entities and > therefore must be disabled on reset. Any further comments before integrating the patch? Ciao, Gerhard -- http://www.wiesinger.com/
Am 08.01.2012 13:01, schrieb Gerhard Wiesinger: > On Wed, 4 Jan 2012, Gerhard Wiesinger wrote: > >> Testscenario: >> 1.) Boot e.g. OS with VMMouse support (e.g. Windows) >> 2.) reboot >> 3.) Boot e.g. OS without VMMouse support (e.g. DOS) => PS/2 mouse >> doesn't work any more. Fixes that issue. >> >> Reason is that VMMouse has priority and disables all other mouse >> entities and therefore must be disabled on reset. > > Any further comments before integrating the patch? I implicitly asked you for a more verbose commit message. For example, based on the info you provided as a reply above: --8<-- hw/vmmouse.c: Disable vmmouse on reboot When a guest activated the vmmouse, following a reboot the vmmouse stayed enabled and the PS/2 mouse was unusable. That way, when rebooting to guests without vmmouse support the mouse would not work. Fix this by disabling the vmmouse in its reset handler. --8<-- I have it queued for testing tomorrow - I am hoping this will fix a 0.15.1 mouse issue after kdump as well. As a bugfix, it would definitely be a candidate for stable-1.0. Andreas
Am 08.01.2012 17:31, schrieb Andreas Färber: > Am 08.01.2012 13:01, schrieb Gerhard Wiesinger: >> On Wed, 4 Jan 2012, Gerhard Wiesinger wrote: >> >>> Testscenario: >>> 1.) Boot e.g. OS with VMMouse support (e.g. Windows) >>> 2.) reboot >>> 3.) Boot e.g. OS without VMMouse support (e.g. DOS) => PS/2 mouse >>> doesn't work any more. Fixes that issue. >>> >>> Reason is that VMMouse has priority and disables all other mouse >>> entities and therefore must be disabled on reset. >> >> Any further comments before integrating the patch? > > I implicitly asked you for a more verbose commit message. > > For example, based on the info you provided as a reply above: > > --8<-- > hw/vmmouse.c: Disable vmmouse on reboot > > When a guest activated the vmmouse, following a reboot the vmmouse > stayed enabled and the PS/2 mouse was unusable. That way, when rebooting > to guests without vmmouse support the mouse would not work. > > Fix this by disabling the vmmouse in its reset handler. > --8<-- I have been unable to reproduce this issue with our qemu-kvm 0.15.1. SLES 11 uses vmmouse, whereas SLES installation does not use vmmouse. After doing a kdump or reboot in SLES, the mouse worked just fine in either of them. The patch doesn't break anything for me though. Andreas
On 2012-01-12 19:37, Andreas Färber wrote: > Am 08.01.2012 17:31, schrieb Andreas Färber: >> Am 08.01.2012 13:01, schrieb Gerhard Wiesinger: >>> On Wed, 4 Jan 2012, Gerhard Wiesinger wrote: >>> >>>> Testscenario: >>>> 1.) Boot e.g. OS with VMMouse support (e.g. Windows) >>>> 2.) reboot >>>> 3.) Boot e.g. OS without VMMouse support (e.g. DOS) => PS/2 mouse >>>> doesn't work any more. Fixes that issue. >>>> >>>> Reason is that VMMouse has priority and disables all other mouse >>>> entities and therefore must be disabled on reset. >>> >>> Any further comments before integrating the patch? >> >> I implicitly asked you for a more verbose commit message. >> >> For example, based on the info you provided as a reply above: >> >> --8<-- >> hw/vmmouse.c: Disable vmmouse on reboot >> >> When a guest activated the vmmouse, following a reboot the vmmouse >> stayed enabled and the PS/2 mouse was unusable. That way, when rebooting >> to guests without vmmouse support the mouse would not work. >> >> Fix this by disabling the vmmouse in its reset handler. >> --8<-- > > I have been unable to reproduce this issue with our qemu-kvm 0.15.1. > SLES 11 uses vmmouse, whereas SLES installation does not use vmmouse. > After doing a kdump or reboot in SLES, the mouse worked just fine in > either of them. I can confirm the that this patch fixes a real issue. Setup: qemu.git, opensuse 11.4 guest, SDL graphic, system_reset while guest is using the vmmouse. Without the patch, the vmmouse become unusable after the reboot. Also, the mouse stays in absolute mode even before X starts again. So please merge. Jan
On Tue, 24 Jan 2012, Jan Kiszka wrote: > I can confirm the that this patch fixes a real issue. Setup: qemu.git, > opensuse 11.4 guest, SDL graphic, system_reset while guest is using the > vmmouse. Without the patch, the vmmouse become unusable after the > reboot. Also, the mouse stays in absolute mode even before X starts again. > > So please merge. My bug reports are always real issues :-) I submitted patch again with more detailed commit information, please merge into master and 1.0 bugfix. No code change. Thnx. Ciao, Gerhard -- http://www.wiesinger.com/
diff --git a/hw/vmmouse.c b/hw/vmmouse.c index 1113f33..f9f5b53 100644 --- a/hw/vmmouse.c +++ b/hw/vmmouse.c @@ -254,6 +254,8 @@ static void vmmouse_reset(DeviceState *d) s->status = 0xffff; s->queue_size = VMMOUSE_QUEUE_SIZE; + + vmmouse_disable(s); } static int vmmouse_initfn(ISADevice *dev)
Bugfix after reboot when vmmouse was enabled and another OS which uses e.g. PS/2 mouse. Signed-off-by: Gerhard Wiesinger <lists@wiesinger.com> --- hw/vmmouse.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)