similar to: Permissive devices in Xen

Hello, I''m trying to get working infiniband device in Xen. I have Xen v3 with 2.6.18 kernel. I export PCI device from Dom0 into DomU. I''ve enabled permisive device in Dom0. However, inserting infiniband module results in the following messages (with little more verbosity which I have added): pcifront pci-0: Installing PCI frontend pcifront pci-0: Creating PCI Frontend Bus

Hello again, when the iozone writes slow. This is how slabtop looks like: 62476752 62476728 0% 0.10K 1601968 39 6407872K buffer_head 1000678 999168 0% 0.56K 142954 7 571816K radix_tree_node 132184 125911 0% 0.03K 1066 124 4264K kmalloc-32 118496 118224 0% 0.12K 3703 32 14812K kmalloc-node 73206 56467 0% 0.19K 3486 21

Hello, I have cluster with AMD EPYC 7351 cpu. Two CPUs per node. I have performance 8-NUMA configuration: This is from hypervizor: [root@hde10 ~]# lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 64 On-line CPU(s) list: 0-63 Thread(s) per core: 2 Core(s) per socket: 16 Socket(s): 2 NUMA

On 09/14/2018 03:36 PM, Lukas Hejtmanek wrote: > Hello, > > ok, I found that cpu pinning was wrong, so I corrected it to be 1:1. The issue > with iozone remains the same. > > The spec is running, however, it runs slower than 1-NUMA case. > > The corrected XML looks like follows: [Reformated XML for better reading] <cpu mode="host-passthrough">

It seems the problems running mthca in a Xen domU have uncovered a bug in mthca: mthca uses dma_sync_single in mthca_arbel_write_mtt_seg() and mthca_arbel_map_phys_fmr() to sync the MTTs that get written. However, Documentation/DMA-API.txt says: void dma_sync_single(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction) synchronise a single

I''m still playing around with xen-3.0.2-3 and the pciback.hide option. Currently I''m trying to give the guest a pcmcia-to-pci card with a prism54 wlan card in the slot. When starting the guest the following message apears in the bootlog: ... pciback 0000:02:09.0: can''t export pci devices that don''t have a normal (0) or bridge (1) header type! ... The lspci of

On 09/17/2018 04:59 PM, Lukas Hejtmanek wrote: > Hello, > > so the current domain configuration: > <cpu mode='host-passthrough'><topology sockets='8' cores='4' threads='1'/><numa><cell cpus='0-3' memory='62000000' /><cell cpus='4-7' memory='62000000' /><cell cpus='8-11'

Patch allowing drivers in DomU to restore address bars if needed. Signed-off-by: xhejtman@ics.muni.cz -- Lukáš Hejtmánek _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

Hello, I''m trying to get to work InfiniBand card in DomU via exporting PCI device. How does the export works internally? I guess that pci frontend forwards stuff like pci_read_config_dword and pci_write_config_dword to pci backend that actually does it. Am I right? So if the driver of InfiniBand writes some config values to PCI config space and these values are not applied actually,

RPM: http://bits.gluster.org/pub/gluster/glusterfs/3.4.1qa2/ SRC: http://bits.gluster.org/pub/gluster/glusterfs/src/glusterfs-3.4.1qa2.tar.gz This release is made off jenkins-release-42 -- Gluster Build System

Hello, ok, I found that cpu pinning was wrong, so I corrected it to be 1:1. The issue with iozone remains the same. The spec is running, however, it runs slower than 1-NUMA case. The corrected XML looks like follows: <cpu mode='host-passthrough'><topology sockets='8' cores='4' threads='1'/><numa><cell cpus='0-3'

Hello, I use xen for a few months now and I upgraded to opensuse 10.3 with xen 3.1 (64bit versions) recently. Everything works fine, but there is a problem when I change xend-config.sxp to use my own network script (it''s slightly modified version of network-virtual scripts from http://en.opensuse.org/Xen3_yet_another_Virtual_Network_Concept ) instead of default network-bridge script.

Hello, so the current domain configuration: <cpu mode='host-passthrough'><topology sockets='8' cores='4' threads='1'/><numa><cell cpus='0-3' memory='62000000' /><cell cpus='4-7' memory='62000000' /><cell cpus='8-11' memory='62000000' /><cell cpus='12-15'

Hello, I'm using sama 3.0.24 on linux machine that serves a share for a windows client. When editing a file on samba, using MS Word I can see there are severals (probably) backup files. ls looks like this: Dotaznik1.doc ~$taznik1.doc ~WRD0000.tmp ~WRD0001.tmp ~WRD0002.tmp ~WRD0003.tmp ~WRD0004.tmp These files do not disappear after I quit the Word. This does not happen if I edit

Hello, I'm facing a bug with winbind. If I request groups for particular domain user, winbind cache gets corrupted. My winbind version is 4.1.12-23.el7_1 (Centos 7.1, x86_64). 1) this is correct UID to SID mapping: wbinfo -U 100152 S-1-5-21-3451901064-902568176-4053310204-180212 2) I request groups info for a user: id -G -n 209459 [reply with 33 groups] 3) I request UID to SID mapping

On Tue, Jun 16, 2015 at 09:12:24AM -0700, Jeremy Allison wrote: > Not that I know of. Can you run winbindd under valgrind ? > Should help debug this. this is the only waring regardless read/write that valgrind reports: ==16475== Syscall param writev(vector[...]) points to uninitialised byte(s) ==16475== at 0xBB32337: writev (in /usr/lib64/libc-2.17.so) ==16475== by 0xA4AC8E4:

Hello, I'm using the Samba server 3.0.33 that exports volume from a GPFS. The GPFS strongly dislikes unlinking files that are locked (resulting in permission denied) using fcntl F_SETLEASE. It seems that the Samba *sometimes* tries to unlink a file that is oplocked. Why? Is this a bug? Why it does not happen always but only sometimes? I have strace logs showing: Wrong case: 8711

Hi everybody, I needed change field error rendering for some forms created with Rail''s form builder. I found solution by changing ActionView::Base::field_error_ proc to own before form and changing it back after form rendered. I don''t like this solution :) It''s not thread-safe (but it''s not so important, nobody really wants it :), but I don''t like

I have a RedHat 6.0 x86 server which is serving a number of minor things, which I wish to add shell access to. I'm currently running sshd and am quite happy with it, the exceptiong being that I am unable to make sshd perform a chroot for shell account users. I have been reading man pages and howto's, many of which discuss sshd or chroot, but never the two together. Is this not an

Is there anyone actually using VT-d for GPU cards? From what I understand NVidia is impossible, but AMD does work for some cards. I would appreciate any hints from people that are actively using this technology. -- Simon Toth