search for: smp_affin

Hi All, We have a xen server and using 8 core processor. I can see that there is 99% iowait on only core 0. 02:28:49 AM CPU %user %nice %sys %iowait %irq %soft %steal %idle intr/s 02:28:54 AM all 0.00 0.00 0.00 12.65 0.00 0.02 2.24 85.08 1359.88 02:28:54 AM 0 0.00 0.00 0.00 96.21 0.00 0.20 3.19 0.40 847.11 02:28:54 AM

...ere is heavy network activity, so if the IRQs are not balanced the server will be CPU bound by the lone processor assigned to handle them. Initially, I solved this problem by adding the following line to "/etc/rc.local" (82 is the IRQ of the ethernet device): echo 0f > /proc/irq/82/smp_affinity This worked like a charm. The IRQs from the ethernet device were balanced across the four processors and each of their idle times was comparable. Unfortunately, when I duplicated this on our backup machine (adjusting the IRQ accordingly), I discovered that something was overwriting the value...

...ult_idle+0x29/0x50 [<ffffffff8004923a>] cpu_idle+0x95/0xb8 [<ffffffff8007796f>] start_secondary+0x498/0x4a7 handlers: [<ffffffff801f74cf>] (usb_hcd_irq+0x0/0x55) Disabling IRQ #58 Looking at /proc/irq/58, it apepars to be USB related: /proc/irq/58/ehci_hcd:usb2 /proc/irq/58/smp_affinity cat /proc/irq/58/smp_affinity cat /proc/irq/58/smp_affinity 00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000002 This is an Asus M4N75TD main board with 4GB non-ECC RAM, and AMD Phenom(tm) II X4 925 Processor. It has an ancient ATI Rage PCI graphics board, booting into ini...

Has anyone had any compatibility issues with a TE110P card installed on a Dell Poweredge 1950? I noted the following error on the LCD display of the Dell Poweredge 1950: E1711 PCI PErr Slot 1 E171F PCIE Fatal Error B0 D4 F0. The Dell hardware owners manual states that it means the system BIOS has reported a PCI parity error on a component that resides in PCI configuration space at bus 0,

...pci-assignable-add 0000:1f:0.0 2.xm cr -c vm.cfg [root@rac10box2 ~]# cat /proc/interrupts |grep mpt2sas0 48: 340449 0 0 0 0 0 0 0 0 0 0 0 PCI-MSI-edge mpt2sas0-msix0 [root@rac10box2 ~]# cat /proc/irq/48/smp_affinity 0fff [root@rac10box2 ~]# echo 2>/proc/irq/48/smp_affinity [root@rac10box2 ~]# cat /proc/irq/48/smp_affinity 0002 [root@rac10box2 ~]# cat /proc/interrupts |grep mpt 48: 342051 0 0 0 0 0 0 0 0 0 0 0 PCI-MSI-e...

Hello: I create a centos7.6 virtual machine which use 3.10.0-957 kernel?using pci passthrough to support nvme device? and I found interrupt can not distribute to target CPU which in smp_affinity set?the problem solved when I update kernel to 5.0.20,is anyone can tell me how to solve this problem without updating kernel?? thanks advance!! -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.centos.org/pipermail/centos-virt/attachments/20190604...

Could anyone either recommend a website or howto on optimizing Linux to run asterisk. Such examples of what I mean are.. Renice of asterisk pid's Forcing irq smp_affinity (For interupt hogging T1 cards) .. That kind of stuff, I looked on the wiki and nothing directly mentions server optimization. Or, is this something that *should* be totally irrelevent when dealing with Asterisk. P.S. I don't mean obvious things such as limiting I/O (ie: turn off debug lo...

...y. I see the packets coming in on eth0, the bridge, and sometimes on the vif device but the domU itself doesn''t see these packets. Has anyone seen this type of behaviour before? I''ve tried all the following with little success: - increasing txqueuelen - changing the smp_affinity settings for our network devces - dedicating the first 2 cores to dom0 and letting the domUs use the remaining 14. If anyone has any ideas or additional things I can try, I would be most grateful :) Thanks, Jeff

...ere we can hear ourselves speaking. The other party doesn't hear this however and its perfectly clean. After solving some of the IRQ/processing issues in which we seemed to get HDLC errors pop up after a week and take down the connection completely (only solvable with a reboot); we changed the smp_affinity settings for all devices to basically run on their own core. This seemed to have solved us getting extended IRQ misses and increase in delays. However it hadn't fixed the echo issue. After some digging on this it seems that there could be a problem with DAHDI and I want to get to the bottom...

Hi, I am after a feel of the throughput capabilities for TC and Iptables in comparison to dedicated hardware. I have heard talk about 1Gb+ throughput with minimal performance impact using 50ish TC rules and 100+ Iptables rules. Is there anyone here running large throughput / large configurations, and if so, what sort of figures? Regards Dan

I am trying to get to the bottom of audio clicks, pops, dropouts with my Asterisk server. These occur even when the system is under minimal load (e.g. 1 Zap device in a queue being played music on hold) and occurs with both Zap and Sip devices so isn't network related. The audio problems occur at the same time on all channels and seems to be when Asterisk "gets busy" and uses

...l vqs, and the above throughput >> data has been very close to same fio test in host side with single >> job. So more improvement should be observed once more IOthreads are >> used for handling requests from multi vqs. >> >> TODO: >> - adjust vq's irq smp_affinity according to blk-mq hw queue's cpumask >> >> V3: >> - fix use-after-free on vq->name reported by Michael >> >> V2: (suggestions from Michael and Dave Chinner) >> - allocate virtqueues' pointers dynamically >> - make sur...

...l vqs, and the above throughput >> data has been very close to same fio test in host side with single >> job. So more improvement should be observed once more IOthreads are >> used for handling requests from multi vqs. >> >> TODO: >> - adjust vq's irq smp_affinity according to blk-mq hw queue's cpumask >> >> V3: >> - fix use-after-free on vq->name reported by Michael >> >> V2: (suggestions from Michael and Dave Chinner) >> - allocate virtqueues' pointers dynamically >> - make sur...

Hi, In the course of my experiments with rt kernels it so happens that the gui versions of tuna built from upstream (RHEL, F11) SRPMs show wrong affinity settings for IRQ threads. Where would be a proper place to report this and get some help on other rt-related issues? Thanks, Sasha

...dmesg|grep "IRQ 21" [ 1.141040] ACPI: PCI Interrupt Link [GSIF] enabled at IRQ 21 $ ls -lA /proc/irq/21 total 0 -r--r--r-- 1 root root 0 Aug 10 19:42 affinity_hint -r--r--r-- 1 root root 0 Aug 10 19:42 node dr-xr-xr-x 2 root root 0 Aug 10 19:42 qxl -rw-r--r-- 1 root root 0 Aug 10 19:42 smp_affinity -rw-r--r-- 1 root root 0 Aug 10 19:42 smp_affinity_list -r--r--r-- 1 root root 0 Aug 10 19:42 spurious dr-xr-xr-x 2 root root 0 Aug 10 19:42 virtio2 $ ls -lA /proc/irq/21/qxl total 0 $ ls -lA /proc/irq/21/virtio2 total 0 ===== {{{ on host, everything else is on guest }}} virsh # managedsave <...

...handles requests from all vqs, and the above throughput > data has been very close to same fio test in host side with single > job. So more improvement should be observed once more IOthreads are > used for handling requests from multi vqs. > > TODO: > - adjust vq's irq smp_affinity according to blk-mq hw queue's cpumask > > V3: > - fix use-after-free on vq->name reported by Michael > > V2: (suggestions from Michael and Dave Chinner) > - allocate virtqueues' pointers dynamically > - make sure the per-queue spinlock isn...

Hello, Here is the situation. There is a machine with 3 intel gigabit card, 2 of them on PCI-X and in bridge, the 3rd is used only for management access. The machine is a dual Xeon 2.8GHz with HT. With 2.6.8 kernel from debian (testing) and htb with u32 on, i usually get about 30-40% software interrupts on CPU0 and CPU2, and without htb and u32, 10% less. Now, if I boot with 2.6.17.9 kernel,

...) Could anyody please help me with a solution so that iam able to balance equally the interrupts on both cores? b.) Is there anyway to actually assign more interrupts to this IRQ (25) or is that something that the kernel discretly takes care of? Thanks Alex [root at node2 ~]# cat /proc/irq/25/smp_affinity 3 Btw I also tried setting this value to "2" and "1" to get to choose which core it handles, and that did work but again the problem is I can only use one core at a time and not balance! [root at node2 ~]# cat /proc/interrupts CPU0 CPU1 0: 229...

...ughput >>> data has been very close to same fio test in host side with single >>> job. So more improvement should be observed once more IOthreads are >>> used for handling requests from multi vqs. >>> >>> TODO: >>> - adjust vq's irq smp_affinity according to blk-mq hw queue's cpumask >>> >>> V3: >>> - fix use-after-free on vq->name reported by Michael >>> >>> V2: (suggestions from Michael and Dave Chinner) >>> - allocate virtqueues' pointers dynamically &...

...io-blk-mq device, only one IOthread handles requests from all vqs, and the above throughput data has been very close to same fio test in host side with single job. So more improvement should be observed once more IOthreads are used for handling requests from multi vqs. TODO: - adjust vq's irq smp_affinity according to blk-mq hw queue's cpumask V3: - fix use-after-free on vq->name reported by Michael V2: (suggestions from Michael and Dave Chinner) - allocate virtqueues' pointers dynamically - make sure the per-queue spinlock isn't kept in same cache line - make each queue'...