search for: num_cpus

...ublic License for + * more details. + */ + +#include <linux/kernel.h> +#include <linux/types.h> +#include <acpi/acpi_bus.h> +#include <acpi/acpi_drivers.h> + +#include <asm/xen/hypercall.h> + +static DEFINE_MUTEX(xen_cpu_lock); + +static int xen_acpi_pad_idle_cpus(int *num_cpus) +{ + int ret; + + struct xen_platform_op op = { + .cmd = XENPF_core_parking, + .interface_version = XENPF_INTERFACE_VERSION, + }; + + /* set cpu nums expected to be idled */ + op.u.core_parking.type = XEN_CORE_PARKING_SET; + op.u.core_parking.idle_nums = (uint32_t)*num_cpus; + ret = HYPERVISOR_d...

...info[''total_memory''] = info[''total_memory''] / 1024 diff -r 51045f276c90 tools/xenmon/xenbaked.c --- a/tools/xenmon/xenbaked.c Mon Jul 31 10:48:48 2006 -0500 +++ b/tools/xenmon/xenbaked.c Mon Jul 31 10:49:10 2006 -0500 @@ -462,10 +462,7 @@ unsigned int get_num_cpus(void) xc_interface_close(xc_handle); opts.cpu_freq = (double)op.u.physinfo.cpu_khz/1000.0; - return (op.u.physinfo.threads_per_core * - op.u.physinfo.cores_per_socket * - op.u.physinfo.sockets_per_node * - op.u.physinfo.nr_nodes); + return op.u.phy...

What is the correct way to globally configure R to default to single (or at least, << NUM_CPUS) threaded operation? Using R 3.3.1 (both in debian unstable or using the CRAN repository for xenial) with OpenBLAS (0.2.18) defaults to using one thread per available CPU, which isn't ideal for machines more than a couple of CPUs. Setting the environment (OMP_NUM_THREADS or OPENBLAS_NUM_THRE...

...M ",systems[iter],") ",sep="") if(QUERY_DATE=="Year") { usageAVGQuery<-paste("SELECT STR_TO_DATE(CONCAT(YEAR(FROM_UNIXTIME(dispatch_time)),'-01-01'),'%Y-%m-%d') AS date, AVG(cpu_seconds_user/total_time) AS time, AVG(num_cpus) AS AVG_CPUs, AVG(dispatch_time-end_time) AS AVG_job_length FROM ",systems[iter],usageDateClamp,"GROUP BY date;",sep="") usageQuery<-paste("SELECT STR_TO_DATE(CONCAT(YEAR(FROM_UNIXTIME(dispatch_time)),'-01-01'),'%Y-%m-%d') AS date, cp...

...eues in guest, with something like below? > > That sounds better, as both transports and drivers can opt-in here. > > However, maybe it would be even better to try to come up with a better > strategy of allocating msix vectors in virtio-pci. More vectors in the > num_queues > num_cpus case, even if they still need to be shared? > Individual vectors for n-1 cpus and then a shared one for the remaining > queues? > > It might even be device-specific: Have some low-traffic status queues > share a vector, and provide an individual vector for high-traffic > queues....

...eues in guest, with something like below? > > That sounds better, as both transports and drivers can opt-in here. > > However, maybe it would be even better to try to come up with a better > strategy of allocating msix vectors in virtio-pci. More vectors in the > num_queues > num_cpus case, even if they still need to be shared? > Individual vectors for n-1 cpus and then a shared one for the remaining > queues? > > It might even be device-specific: Have some low-traffic status queues > share a vector, and provide an individual vector for high-traffic > queues....

Testing on an 8-node 128-way NUMA machine has exposed a problem with Xen''s nr_cpus calculation. In this case, since Xen cuts off recognized CPUs at 32, the machine appears to have 16 CPUs on the first and second nodes and none on the remaining nodes. Given this asymmetry, the calculation of sockets_per_node (which is later used to calculate nr_cpus) is incorrect:

Please provide a patch to Open Projects list. On Wed, Mar 4, 2015 at 8:25 PM, Vassil Vassilev <vvasilev at cern.ch> wrote: > On 17/02/15 09:47, Anton Korobeynikov wrote: >> >> John, >> >> Yes, I'm taking care about application as usual. > > I saw the the LLVM mentoring org was accepted. Congrats! > Anton, could you tell me what is the procedure of

...NUM_THREADS = 1 export OPENBLAS_NUM_THREADS OPENBLAS_NUM_THREADS environment variable does not affect the OMP_NUM_THREADS. 2016-08-01 20:06 GMT+09:00 Gordon Ball <gordon at chronitis.net>: > What is the correct way to globally configure R to default to single (or > at least, << NUM_CPUS) threaded operation? > > > Using R 3.3.1 (both in debian unstable or using the CRAN repository for > xenial) with OpenBLAS (0.2.18) defaults to using one thread per > available CPU, which isn't ideal for machines more than a couple of CPUs. > > Setting the environment (OMP_...

+Easily, some of the code snippets end up being copied dozens of +times, which leads to worse maintainability, understandability and logical +design. The project description stresses code maintainability and logical design more than bug finding due to omissions in copy and pasted code. Reading this made me think of a check that would suggest people to replace copy and pasted code with a function

...imit > the max number of queues in guest, with something like below? That sounds better, as both transports and drivers can opt-in here. However, maybe it would be even better to try to come up with a better strategy of allocating msix vectors in virtio-pci. More vectors in the num_queues > num_cpus case, even if they still need to be shared? Individual vectors for n-1 cpus and then a shared one for the remaining queues? It might even be device-specific: Have some low-traffic status queues share a vector, and provide an individual vector for high-traffic queues. Would need some device<-&gt...

...ke below? > > > > That sounds better, as both transports and drivers can opt-in here. > > > > However, maybe it would be even better to try to come up with a better > > strategy of allocating msix vectors in virtio-pci. More vectors in the > > num_queues > num_cpus case, even if they still need to be shared? > > Individual vectors for n-1 cpus and then a shared one for the remaining > > queues? > > > > It might even be device-specific: Have some low-traffic status queues > > share a vector, and provide an individual vector for h...

> On Mar 11, 2015, at 2:14 AM, Vassil Vassilev <vvasilev at cern.ch> wrote: > > On 10/03/15 19:13, Anna Zaks wrote: >> >>> On Mar 10, 2015, at 1:03 AM, Vassil Vassilev <vvasilev at cern.ch <mailto:vvasilev at cern.ch>> wrote: >>> >>> On 09/03/15 21:52, Anna Zaks wrote: >>>> >>>> +Easily, some of the code

On 3/13/19 1:33 AM, Cornelia Huck wrote: > On Tue, 12 Mar 2019 10:22:46 -0700 (PDT) > Dongli Zhang <dongli.zhang at oracle.com> wrote: > >> I observed that there is one msix vector for config and one shared vector >> for all queues in below qemu cmdline, when the num-queues for virtio-blk >> is more than the number of possible cpus: >> >> qemu:

On 3/13/19 1:33 AM, Cornelia Huck wrote: > On Tue, 12 Mar 2019 10:22:46 -0700 (PDT) > Dongli Zhang <dongli.zhang at oracle.com> wrote: > >> I observed that there is one msix vector for config and one shared vector >> for all queues in below qemu cmdline, when the num-queues for virtio-blk >> is more than the number of possible cpus: >> >> qemu:

..., with something like below? >> That sounds better, as both transports and drivers can opt-in here. >> >> However, maybe it would be even better to try to come up with a better >> strategy of allocating msix vectors in virtio-pci. More vectors in the >> num_queues > num_cpus case, even if they still need to be shared? >> Individual vectors for n-1 cpus and then a shared one for the remaining >> queues? >> >> It might even be device-specific: Have some low-traffic status queues >> share a vector, and provide an individual vector for high-tr...

...em to cause any particular trouble (yet). Gordon [1]: docs.rstudio.com/ide/server-pro/r-sessions.html > > 2016-08-01 20:06 GMT+09:00 Gordon Ball <gordon at chronitis.net>: >> What is the correct way to globally configure R to default to single (or >> at least, << NUM_CPUS) threaded operation? >> >> >> Using R 3.3.1 (both in debian unstable or using the CRAN repository for >> xenial) with OpenBLAS (0.2.18) defaults to using one thread per >> available CPU, which isn't ideal for machines more than a couple of CPUs. >> >> S...

When I say busybox it''s actually VMware ESX server which seems to use busybox (which I guess is the case a number of other software appliances) . Reason for wanting to install puppet is to run the cli tools to create nightly vmware snapshots. I''m happy to give it a go (and add the docs to the wiki) but I''m not too sure at this stage how big a task this might be and what,