search for: cpumasks

As the subject says, building syslinux fails when built against 2.6.8.1 kernel headers. Anybody out there have a patch? The compile errors are below. set -e ; for i in memdisk dos win32 mtools unix extlinux sample com32 ; do make DATE=0x41e470ae HEXDATE=0x41e470ae -C $i all ; done make[1]: Entering directory `/usr/src/syslinux-3.07-pre1/memdisk' make[1]: Nothing to be done for `all'.

To improve TLB shootdown performance, flush the remote and local TLBs concurrently. Introduce flush_tlb_multi() that does so. Introduce paravirtual versions of flush_tlb_multi() for KVM, Xen and hyper-v (Xen and hyper-v are only compile-tested). While the updated smp infrastructure is capable of running a function on a single local core, it is not optimized for this case. The multiple function

To improve TLB shootdown performance, flush the remote and local TLBs concurrently. Introduce flush_tlb_multi() that does so. The current flush_tlb_others() interface is kept, since paravirtual interfaces need to be adapted first before it can be removed. This is left for future work. In such PV environments, TLB flushes are not performed, at this time, concurrently. Add a static key to tell

To improve TLB shootdown performance, flush the remote and local TLBs concurrently. Introduce flush_tlb_multi() that does so. The current flush_tlb_others() interface is kept, since paravirtual interfaces need to be adapted first before it can be removed. This is left for future work. In such PV environments, TLB flushes are not performed, at this time, concurrently. Add a static key to tell

To improve TLB shootdown performance, flush the remote and local TLBs concurrently. Introduce flush_tlb_multi() that does so. Introduce paravirtual versions of flush_tlb_multi() for KVM, Xen and hyper-v (Xen and hyper-v are only compile-tested). While the updated smp infrastructure is capable of running a function on a single local core, it is not optimized for this case. The multiple function

An embedded and charset-unspecified text was scrubbed... Name: cpumask:convert-drivers.patch Url: http://lists.linux-foundation.org/pipermail/virtualization/attachments/20090107/f459eb35/attachment.txt

An embedded and charset-unspecified text was scrubbed... Name: cpumask:convert-drivers.patch Url: http://lists.linux-foundation.org/pipermail/virtualization/attachments/20090107/f459eb35/attachment.txt

To improve TLB shootdown performance, flush the remote and local TLBs concurrently. Introduce flush_tlb_multi() that does so. The current flush_tlb_others() interface is kept, since paravirtual interfaces need to be adapted first before it can be removed. This is left for future work. In such PV environments, TLB flushes are not performed, at this time, concurrently. Add a static key to tell

To improve TLB shootdown performance, flush the remote and local TLBs concurrently. Introduce flush_tlb_multi() that does so. The current flush_tlb_others() interface is kept, since paravirtual interfaces need to be adapted first before it can be removed. This is left for future work. In such PV environments, TLB flushes are not performed, at this time, concurrently. Add a static key to tell

Currently, local and remote TLB flushes are not performed concurrently, which introduces unnecessary overhead - each INVLPG can take 100s of cycles. This patch-set allows TLB flushes to be run concurrently: first request the remote CPUs to initiate the flush, then run it locally, and finally wait for the remote CPUs to finish their work. In addition, there are various small optimizations to avoid

...n that there are no nested TLB > + * flushes, an assumption that is already made in > + * flush_tlb_mm_range(). > + */ > + struct cpumask *cond_cpumask = this_cpu_ptr(&flush_tlb_mask); This is logically a stack-local variable, right? But, since we've got preempt off and cpumasks can be huge, we don't want to allocate it on the stack. That might be worth a comment somewhere. > + int cpu; > + > + cpumask_clear(cond_cpumask); > + > + for_each_cpu(cpu, cpumask) { > + if (tlb_is_not_lazy(cpu)) > + __cpumask_set_cpu(cpu, cond_cpumask); > +...

To improve TLB shootdown performance, flush the remote and local TLBs concurrently. Introduce flush_tlb_multi() that does so. The current flush_tlb_others() interface is kept, since paravirtual interfaces need to be adapted first before it can be removed. This is left for future work. In such PV environments, TLB flushes are not performed, at this time, concurrently. Add a static key to tell

To improve TLB shootdown performance, flush the remote and local TLBs concurrently. Introduce flush_tlb_multi() that does so. The current flush_tlb_others() interface is kept, since paravirtual interfaces need to be adapted first before it can be removed. This is left for future work. In such PV environments, TLB flushes are not performed, at this time, concurrently. Add a static key to tell

[ Cover-letter is identical to v2, including benchmark results, excluding the change log. ] Currently, local and remote TLB flushes are not performed concurrently, which introduces unnecessary overhead - each INVLPG can take 100s of cycles. This patch-set allows TLB flushes to be run concurrently: first request the remote CPUs to initiate the flush, then run it locally, and finally wait for

...>> + * flushes, an assumption that is already made in >> + * flush_tlb_mm_range(). >> + */ >> + struct cpumask *cond_cpumask = this_cpu_ptr(&flush_tlb_mask); > > This is logically a stack-local variable, right? But, since we've got > preempt off and cpumasks can be huge, we don't want to allocate it on > the stack. That might be worth a comment somewhere. I will add a comment here. > >> + int cpu; >> + >> + cpumask_clear(cond_cpumask); >> + >> + for_each_cpu(cpu, cpumask) { >> + if (tlb_is_not_lazy...

...>> + * flushes, an assumption that is already made in >> + * flush_tlb_mm_range(). >> + */ >> + struct cpumask *cond_cpumask = this_cpu_ptr(&flush_tlb_mask); > > This is logically a stack-local variable, right? But, since we've got > preempt off and cpumasks can be huge, we don't want to allocate it on > the stack. That might be worth a comment somewhere. I will add a comment here. > >> + int cpu; >> + >> + cpumask_clear(cond_cpumask); >> + >> + for_each_cpu(cpu, cpumask) { >> + if (tlb_is_not_lazy...

Hi Ingo, Please pull a few more cpus4096 changes. Most are fairly lightweight with the biggest change being to use cpumask_var_t's for the domain and pending_mask cpumask's in irq_desc. It also deals with this cpu_evtchn_mask bss space bump: 65536 +2031616 2097152 +3100% cpu_evtchn_mask(.bss) Everything merges into tip/master cleanly. Thanks! Mike The following changes since

We used to use a percpu structure vq_index to record the cpu to queue mapping, this is suboptimal since it duplicates the work of XPS and loses all other XPS functionality such as allowing use to configure their own transmission steering strategy. So this patch switches to use XPS and suggest a default mapping when the number of cpus is equal to the number of queues. With XPS support, there's

We used to use a percpu structure vq_index to record the cpu to queue mapping, this is suboptimal since it duplicates the work of XPS and loses all other XPS functionality such as allowing use to configure their own transmission steering strategy. So this patch switches to use XPS and suggest a default mapping when the number of cpus is equal to the number of queues. With XPS support, there's

Hi Ingo, Please pull a few more cpus4096 changes. Most are fairly lightweight with the biggest change being to use cpumask_var_t's for the domain and pending_mask cpumask's in irq_desc. It also deals with this cpu_evtchn_mask bss space bump: 65536 +2031616 2097152 +3100% cpu_evtchn_mask(.bss) Everything merges into tip/master cleanly. Thanks! Mike The following changes since