search for: last_timer

Hi, I am doing some domain resource monitoring work. I use cat /dev/urandom | md5sum to simulate vCPU load and write a script to calculate vCPU utilization. It seems all good at the beginning. After I abort the ``cat ... md5sum`` command in domain, I see some strange data as below: CPU time: 8410960000000 CPU util: 99.8410609331% CPU time: 8411970000000 CPU util: 100.843672381% CPU time:

I was getting vCPU use time outside of guest with libvirt-python API, and them calculate utilization with (cpuTime2 - cpuTime1) / (t2 - t1). I was not doing this inside the guest os. 2015-07-24 15:09 GMT+08:00 2020human <human2020@qq.com>: > You calculate is vCPU use time not utilization。 > > use_time/total_cpu_time is utilization。 > > total_cpu_time=`cat /proc/stat |sed -n

The following patches are the latest update of the TSC and hrtimer patches I posted on 29/03. Rusty's original TSC patch has been resynced to the latest lguest repo, as has the hrtimer patch, which also incorporates feedback from Jeremy & Rusty: - Change clock event hrtimer to absolute time. 'now' is captured in the host during the hypercall. - Propagate -ETIME back to the

The following patches are the latest update of the TSC and hrtimer patches I posted on 29/03. Rusty's original TSC patch has been resynced to the latest lguest repo, as has the hrtimer patch, which also incorporates feedback from Jeremy & Rusty: - Change clock event hrtimer to absolute time. 'now' is captured in the host during the hypercall. - Propagate -ETIME back to the

Some time ago someone posted a nice script to monitor HTB classes, classmon.pl. A friend of mine ported it to CBQ, and attached is the result. Suggestions are welcome. Rubens > #!/usr/bin/perl > > # Classy CBQ Operations Monitor...in Perl > # Based on classmon.pl by Toby Cantor > # By BLFC > > # The following short command line options are parsed as you might expect. >

...'\0'; + kill_guest(lg, "CRASH: %s", msg); + break; + } + case LHCALL_FLUSH_TLB: + if (regs->edx) + guest_pagetable_clear_all(lg); + else + guest_pagetable_flush_user(lg); + break; + case LHCALL_TIMER_READ: { + u32 now = jiffies; + mb(); + regs->eax = now - lg->last_timer; + lg->last_timer = now; + break; + } + case LHCALL_GET_WALLCLOCK: { + struct timespec ts; + ktime_get_real_ts(&ts); + regs->eax = ts.tv_sec; + break; + } + case LHCALL_BIND_DMA: + regs->eax = bind_dma(lg, regs->edx, regs->ebx, + regs->ecx >> 8, regs->e...

...'\0'; + kill_guest(lg, "CRASH: %s", msg); + break; + } + case LHCALL_FLUSH_TLB: + if (regs->edx) + guest_pagetable_clear_all(lg); + else + guest_pagetable_flush_user(lg); + break; + case LHCALL_TIMER_READ: { + u32 now = jiffies; + mb(); + regs->eax = now - lg->last_timer; + lg->last_timer = now; + break; + } + case LHCALL_GET_WALLCLOCK: { + struct timespec ts; + ktime_get_real_ts(&ts); + regs->eax = ts.tv_sec; + break; + } + case LHCALL_BIND_DMA: + regs->eax = bind_dma(lg, regs->edx, regs->ebx, + regs->ecx >> 8, regs->e...

Bradley, and the rsync development team, I came across the following message on the rsync bug tracking system, while searching for just this feature (access time preservation)... http://rsync.samba.org/cgi-bin/rsync/incoming?id=2509;expression=atime;user=guest ... where it is written... > Date: Mon, 7 Feb 2000 02:59:42 -0500 > From: "Bradley M. Kuhn" <bkuhn@ebb.org> >

..._guest_idt_entry(vcpu, regs->rdx,&g); + break; + } + case LHCALL_SET_STACK: + guest_set_stack(vcpu, regs->rdx, regs->rbx); + break; + case LHCALL_TS: + vcpu->ts = regs->rdx; + break; + case LHCALL_TIMER_READ: { + u32 now = jiffies; + mb(); + regs->rax = now - linfo->last_timer; + linfo->last_timer = now; + break; + } + case LHCALL_TIMER_START: + linfo->timer_on = 1; + if (regs->rdx != HZ) + kill_guest(linfo, "Bad clock speed %lli", regs->rdx); + linfo->last_timer = jiffies; + break; + case LHCALL_HALT: + linfo->halted = 1; + break; +...

..._guest_idt_entry(vcpu, regs->rdx,&g); + break; + } + case LHCALL_SET_STACK: + guest_set_stack(vcpu, regs->rdx, regs->rbx); + break; + case LHCALL_TS: + vcpu->ts = regs->rdx; + break; + case LHCALL_TIMER_READ: { + u32 now = jiffies; + mb(); + regs->rax = now - linfo->last_timer; + linfo->last_timer = now; + break; + } + case LHCALL_TIMER_START: + linfo->timer_on = 1; + if (regs->rdx != HZ) + kill_guest(linfo, "Bad clock speed %lli", regs->rdx); + linfo->last_timer = jiffies; + break; + case LHCALL_HALT: + linfo->halted = 1; + break; +...

Unlike full virtualization in which the virtual machine provides the same platform interface as running natively on the hardware, paravirtualization requires modification to the guest operating system to work with the platform interface provided by the hypervisor. Xen was designed with performance in mind. Calls to the hypervisor are minimized, batched if necessary, and non-critical codepaths

Unlike full virtualization in which the virtual machine provides the same platform interface as running natively on the hardware, paravirtualization requires modification to the guest operating system to work with the platform interface provided by the hypervisor. Xen was designed with performance in mind. Calls to the hypervisor are minimized, batched if necessary, and non-critical codepaths

Is there another way of doing the bindings so that the Windows share doesn't run on TCP/IP but will still work with SAMBA? If so, where in the docs should I be looking, and if not - maybe this should be something discussed by the SAMBA dev people.. Regards, Marc. Return-Path: <twunder@iwmail.com> Delivered-To: samba@lists.samba.org Received: from femail14.sdc1.sfba.home.com