similar to: 4316952 ufs_fault_v should not attempt to queue more then 1 handler for a failure event

Hi, I looked at the pid provider''s trap handler code (usr/src/uts/sun4u/ml/trap_table.s and usr/src/uts/sun4/os/dtrace_subr.c), and try to understand how it works. Can someone point me some directions about trap handlers? I am interesed in writing a new one (like dtrace does it in pid provider with the 0x38 trap id). At first, I would be pleased if I could write a trap handler which just

On Thu, Sep 11, 2014 at 09:08:03PM +0930, Rusty Russell wrote: > Amos Kong <akong at redhat.com> writes: > > When I check hwrng attributes in sysfs, cat process always gets > > stuck if guest has only 1 vcpu and uses a slow rng backend. > > > > Currently we check if there is any tasks waiting to be run on > > current cpu in rng_dev_read() by need_resched().

On Mon, Mar 16, 2015 at 02:16:13PM +0100, Peter Zijlstra wrote: > Hi Waiman, > > As promised; here is the paravirt stuff I did during the trip to BOS last week. > > All the !paravirt patches are more or less the same as before (the only real > change is the copyright lines in the first patch). > > The paravirt stuff is 'simple' and KVM only -- the Xen code was a

On Thu, Mar 26, 2015 at 09:21:53PM +0100, Peter Zijlstra wrote: > On Wed, Mar 25, 2015 at 03:47:39PM -0400, Konrad Rzeszutek Wilk wrote: > > Ah nice. That could be spun out as a seperate patch to optimize the existing > > ticket locks I presume. > > Yes I suppose we can do something similar for the ticket and patch in > the right increment. We'd need to restructure the

On Mon, Mar 16, 2015 at 02:16:13PM +0100, Peter Zijlstra wrote: > Hi Waiman, > > As promised; here is the paravirt stuff I did during the trip to BOS last week. > > All the !paravirt patches are more or less the same as before (the only real > change is the copyright lines in the first patch). > > The paravirt stuff is 'simple' and KVM only -- the Xen code was a

On Thu, Mar 26, 2015 at 09:21:53PM +0100, Peter Zijlstra wrote: > On Wed, Mar 25, 2015 at 03:47:39PM -0400, Konrad Rzeszutek Wilk wrote: > > Ah nice. That could be spun out as a seperate patch to optimize the existing > > ticket locks I presume. > > Yes I suppose we can do something similar for the ticket and patch in > the right increment. We'd need to restructure the

On Wed, Mar 25, 2015 at 03:47:39PM -0400, Konrad Rzeszutek Wilk wrote: > Ah nice. That could be spun out as a seperate patch to optimize the existing > ticket locks I presume. Yes I suppose we can do something similar for the ticket and patch in the right increment. We'd need to restructure the code a bit, but its not fundamentally impossible. We could equally apply the head hashing to

On Wed, Mar 25, 2015 at 03:47:39PM -0400, Konrad Rzeszutek Wilk wrote: > Ah nice. That could be spun out as a seperate patch to optimize the existing > ticket locks I presume. Yes I suppose we can do something similar for the ticket and patch in the right increment. We'd need to restructure the code a bit, but its not fundamentally impossible. We could equally apply the head hashing to

On Sun, Sep 14, 2014 at 01:12:58AM +0800, Amos Kong wrote: > On Thu, Sep 11, 2014 at 09:08:03PM +0930, Rusty Russell wrote: > > Amos Kong <akong at redhat.com> writes: > > > When I check hwrng attributes in sysfs, cat process always gets > > > stuck if guest has only 1 vcpu and uses a slow rng backend. > > > > > > Currently we check if there is

On Sun, Sep 14, 2014 at 01:12:58AM +0800, Amos Kong wrote: > On Thu, Sep 11, 2014 at 09:08:03PM +0930, Rusty Russell wrote: > > Amos Kong <akong at redhat.com> writes: > > > When I check hwrng attributes in sysfs, cat process always gets > > > stuck if guest has only 1 vcpu and uses a slow rng backend. > > > > > > Currently we check if there is

On Sun, Jun 15, 2014 at 02:47:07PM +0200, Peter Zijlstra wrote: > Add minimal paravirt support. > > The code aims for minimal impact on the native case. Woot! > > On the lock side we add one jump label (asm_goto) and 4 paravirt > callee saved calls that default to NOPs. The only effects are the > extra NOPs and some pointless MOVs to accomodate the calling > convention.

On Sun, Jun 15, 2014 at 02:47:07PM +0200, Peter Zijlstra wrote: > Add minimal paravirt support. > > The code aims for minimal impact on the native case. Woot! > > On the lock side we add one jump label (asm_goto) and 4 paravirt > callee saved calls that default to NOPs. The only effects are the > extra NOPs and some pointless MOVs to accomodate the calling > convention.

Author: rab Repository: /hg/zfs-crypto/gate Revision: fc8d85bf341da4f066bb72e4b29334a5099f633b Log message: 6260557 sparc post_syscall handler clears t_sysnum earlier than x86 Files: update: usr/src/uts/sparc/os/syscall.c

On 03/25/2015 03:47 PM, Konrad Rzeszutek Wilk wrote: > On Mon, Mar 16, 2015 at 02:16:13PM +0100, Peter Zijlstra wrote: >> Hi Waiman, >> >> As promised; here is the paravirt stuff I did during the trip to BOS last week. >> >> All the !paravirt patches are more or less the same as before (the only real >> change is the copyright lines in the first patch).

On 03/25/2015 03:47 PM, Konrad Rzeszutek Wilk wrote: > On Mon, Mar 16, 2015 at 02:16:13PM +0100, Peter Zijlstra wrote: >> Hi Waiman, >> >> As promised; here is the paravirt stuff I did during the trip to BOS last week. >> >> All the !paravirt patches are more or less the same as before (the only real >> change is the copyright lines in the first patch).

Hi Is there a way i can trigger a script within a libvirt guest immediately after resume? E.g. i don't see any log message that indicates to a guest that it had been paused and was just woken up. The problem that i am trying to solve is that when i reboot the host (pausing all guests), the guests' system clocks are a few minutes late after resume. The guests RTC (current_clocksource =

So I get a phone call about my companie's controler not being able to log into samba. About two weeks ago we migrated from Win2k Server to Samba running on LDAP. What would be the default value that sambaPwdMustChange would be incremented? This is NT Time Right (1 unit for every 100 ms from 1600 right?) I just had to bump everybody what I'm guessing is three weeks, but I need to know

I'd like to get disk I/O down to a minimum for my new Centos 5.5 installation. The machine will not be used as a web server anymore, as that's now hosted on a cloud platform. So there are no HTTP requests coming down the line. If I move the SWAP partition and /var/log/ to a small spare drive, and install Centos on the new larger drive, is there anything else that would cause disk

Amos Kong <akong at redhat.com> writes: > When I check hwrng attributes in sysfs, cat process always gets > stuck if guest has only 1 vcpu and uses a slow rng backend. > > Currently we check if there is any tasks waiting to be run on > current cpu in rng_dev_read() by need_resched(). But need_resched() > doesn't work because rng_dev_read() is executing in user context.

Amos Kong <akong at redhat.com> writes: > When I check hwrng attributes in sysfs, cat process always gets > stuck if guest has only 1 vcpu and uses a slow rng backend. > > Currently we check if there is any tasks waiting to be run on > current cpu in rng_dev_read() by need_resched(). But need_resched() > doesn't work because rng_dev_read() is executing in user context.