Hi there!
I just fresh-installed Xen on new server, Xeon E5-2620, and noticed
following thing:
node1 ~ # xl info -n
host : node1
release : 3.8.8
version : #1 SMP Tue Apr 23 20:54:17 MSK 2013
machine : x86_64
nr_cpus : 6
max_cpu_id : 5
nr_nodes : 1
cores_per_socket : 6
threads_per_core : 1
cpu_mhz : 2000
hw_caps :
bfebfbff:2c100800:00000000:00003f40:17bee3ff:00000000:00000001:00000000
virt_caps : hvm
total_memory : 49127
free_memory : 0
sharing_freed_memory : 0
sharing_used_memory : 0
free_cpus : 0
cpu_topology :
cpu: core socket node
0: 0 0 0
1: 1 0 0
2: 2 0 0
3: 3 0 0
4: 4 0 0
5: 5 0 0
numa_info :
node: memsize memfree distances
0: 51200 0 10
xen_major : 4
xen_minor : 2
xen_extra : .1
xen_caps : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32
hvm-3.0-x86_32p hvm-3.0-x86_64
xen_scheduler : credit
xen_pagesize : 4096
platform_params : virt_start=0xffff800000000000
xen_changeset : unavailable
xen_commandline : mce=1 loglvl=all console=com2,vga
com2=115200,8n1 guest_loglvl=all
cc_compiler : x86_64-pc-linux-gnu-gcc (Gentoo Hardened 4.6.3
p1.11, pie-0.5.2
cc_compile_by :
cc_compile_domain : (none)
cc_compile_date : Mon Apr 22 16:11:02 MSK 2013
xend_config_format : 4
node1~ # dmidecode L | grep -A2 "Core Count"
Core Count: 6
Core Enabled: 6
Thread Count: 12
So, HyperThreading is actually enabled, but Xen shows (and operates)
only phisycal cores.
As far, as I remember, when I used Xen on hexacore i7 some time ago,
there was 12 cores to operate.
So, is it normal behaviour for now, and cant it be fixed somehow?
And, also, maybe I need to paste something else?
_______________________________________________
Xen-users mailing list
Xen-users@lists.xen.org
http://lists.xen.org/xen-users
Sorry, I mean 4.2.1, not 4.1.2 _______________________________________________ Xen-users mailing list Xen-users@lists.xen.org http://lists.xen.org/xen-users
> node1~ # dmidecode L | grep -A2 "Core Count" > Core Count: 6 > Core Enabled: 6 > Thread Count: 12 > > So, HyperThreading is actually enabled, but Xen shows (and operates) > only phisycal cores.Are you sure HT is enabled in the BIOS? I wouldn''t necessarily trust the DMI tables to reflect this...> And, also, maybe I need to paste something else?The Xen bootlogs (xm/xl dmesg) would probably be the most useful thing. Ian.
24.04.2013 16:29, Ian Campbell пишет:> Are you sure HT is enabled in the BIOS? I wouldn''t necessarily trust the > DMI tables to reflect this...AFAIR, it is enabled, and, moreover, it was 12 cores in baremetal linux before reboot. Unfortunately, I can''t reboot until tonight (since there are some production domU)> The Xen bootlogs (xm/xl dmesg) would probably be the most useful thing.Here it is (but it also shows 6 cores): __ __ _ _ ____ _ \ \/ /___ _ __ | || | |___ \ / | \ // _ \ ''_ \ | || |_ __) | | | / \ __/ | | | |__ _| / __/ _| | /_/\_\___|_| |_| |_|(_)_____(_)_| (XEN) Xen version 4.2.1 (@(none)) (x86_64-pc-linux-gnu-gcc (Gentoo Hardened 4.6.3 p1.11, pie-0.5.2) 4.6.3) Mon Apr 22 16:11:02 MSK 2013 (XEN) Latest ChangeSet: unavailable (XEN) Bootloader: GRUB 2.00 (XEN) Command line: mce=1 loglvl=all console=com2,vga com2=115200,8n1 guest_loglvl=all (XEN) Video information: (XEN) VGA is text mode 80x25, font 8x16 (XEN) VBE/DDC methods: none; EDID transfer time: 0 seconds (XEN) EDID info not retrieved because no DDC retrieval method detected (XEN) Disc information: (XEN) Found 1 MBR signatures (XEN) Found 1 EDD information structures (XEN) Multiboot-e820 RAM map: (XEN) 0000000000000000 - 00000000000a0000 (usable) (XEN) 0000000000100000 - 000000007e380000 (usable) (XEN) 000000007e380000 - 000000007e48c000 (ACPI NVS) (XEN) 000000007e48c000 - 000000007f1f7000 (reserved) (XEN) 000000007f1f7000 - 000000007f273000 (ACPI NVS) (XEN) 000000007f273000 - 000000007f2e2000 (reserved) (XEN) 000000007f2e2000 - 000000007f2e5000 (usable) (XEN) 000000007f2e5000 - 000000007f333000 (reserved) (XEN) 000000007f333000 - 000000007f334000 (usable) (XEN) 000000007f334000 - 000000007f33a000 (ACPI NVS) (XEN) 000000007f33a000 - 000000007f347000 (reserved) (XEN) 000000007f347000 - 000000007f34f000 (ACPI NVS) (XEN) 000000007f34f000 - 000000007f37a000 (reserved) (XEN) 000000007f37a000 - 000000007f3fd000 (ACPI NVS) (XEN) 000000007f3fd000 - 000000007f800000 (usable) (XEN) 0000000080000000 - 0000000090000000 (reserved) (XEN) 00000000fed1c000 - 00000000fed40000 (reserved) (XEN) 00000000ff000000 - 0000000100000000 (reserved) (XEN) 0000000100000000 - 0000000c80000000 (usable) (XEN) ACPI Error (tbxfroot-0218): A valid RSDP was not found [20070126] (XEN) System RAM: 49127MB (50306204kB) (XEN) No NUMA configuration found (XEN) Faking a node at 0000000000000000-0000000c80000000 (XEN) Domain heap initialised (XEN) found SMP MP-table at 000fce30 (XEN) DMI 2.7 present. (XEN) Using APIC driver default (XEN) Table is not found! (XEN) Intel MultiProcessor Specification v1.4 (XEN) Virtual Wire compatibility mode. (XEN) OEM ID: A M I Product ID: ALASKA APIC at: 0xFEE00000 (XEN) Processor #0 6:13 APIC version 21 (XEN) Processor #2 6:13 APIC version 21 (XEN) Processor #4 6:13 APIC version 21 (XEN) Processor #6 6:13 APIC version 21 (XEN) Processor #8 6:13 APIC version 21 (XEN) Processor #10 6:13 APIC version 21 (XEN) I/O APIC #0 Version 32 at 0xFEC00000. (XEN) I/O APIC #2 Version 32 at 0xFEC01000. (XEN) Enabling APIC mode: Flat. Using 2 I/O APICs (XEN) Processors: 6 (XEN) SMP: Allowing 6 CPUs (0 hotplug CPUs) (XEN) IRQ limits: 48 GSI, 1120 MSI/MSI-X (XEN) Not enabling x2APIC: depends on iommu_supports_eim. (XEN) XSM Framework v1.0.0 initialized (XEN) Flask: Initializing. (XEN) AVC INITIALIZED (XEN) dummy.c:727: Had to override the machine_address_size security operation with the dummy one. (XEN) dummy.c:732: Had to override the hvm_inject_msi security operation with the dummy one. (XEN) Flask: Starting in permissive mode. (XEN) Using scheduler: SMP Credit Scheduler (credit) (XEN) Detected 2000.071 MHz processor. (XEN) Initing memory sharing. (XEN) xstate_init: using cntxt_size: 0x340 and states: 0x7 (XEN) mce_intel.c:1239: MCA Capability: BCAST 1 SER 1 CMCI 1 firstbank 0 extended MCE MSR 0 (XEN) Intel machine check reporting enabled (XEN) I/O virtualisation disabled (XEN) Enabled directed EOI with ioapic_ack_old on! (XEN) ENABLING IO-APIC IRQs (XEN) -> Using old ACK method (XEN) ..TIMER: vector=0xF0 apic1=0 pin1=2 apic2=-1 pin2=-1 (XEN) TSC deadline timer enabled (XEN) Platform timer is 1.193MHz PIT (XEN) Allocated console ring of 64 KiB. (XEN) VMX: Supported advanced features: (XEN) - APIC MMIO access virtualisation (XEN) - APIC TPR shadow (XEN) - Extended Page Tables (EPT) (XEN) - Virtual-Processor Identifiers (VPID) (XEN) - Virtual NMI (XEN) - MSR direct-access bitmap (XEN) - Unrestricted Guest (XEN) HVM: ASIDs enabled. (XEN) HVM: VMX enabled (XEN) HVM: Hardware Assisted Paging (HAP) detected (XEN) HVM: HAP page sizes: 4kB, 2MB, 1GB (XEN) Brought up 6 CPUs (XEN) ACPI sleep modes: S3 (XEN) mcheck_poll: Machine check polling timer started. (XEN) *** LOADING DOMAIN 0 *** (XEN) Xen kernel: 64-bit, lsb, compat32 (XEN) Dom0 kernel: 64-bit, PAE, lsb, paddr 0x1000000 -> 0x2400000 (XEN) PHYSICAL MEMORY ARRANGEMENT: (XEN) Dom0 alloc.: 0000000c40000000->0000000c50000000 (12342122 pages to be allocated) (XEN) VIRTUAL MEMORY ARRANGEMENT: (XEN) Loaded kernel: ffffffff81000000->ffffffff82400000 (XEN) Init. ramdisk: ffffffff82400000->ffffffff82400000 (XEN) Phys-Mach map: ffffffff82400000->ffffffff882a9b50 (XEN) Start info: ffffffff882aa000->ffffffff882aa4b4 (XEN) Page tables: ffffffff882ab000->ffffffff882f0000 (XEN) Boot stack: ffffffff882f0000->ffffffff882f1000 (XEN) TOTAL: ffffffff80000000->ffffffff88400000 (XEN) ENTRY ADDRESS: ffffffff819831c0 (XEN) Scrubbing Free RAM: .done. (XEN) Initial low memory virq threshold set at 0x4000 pages. (XEN) Std. Loglevel: All (XEN) Guest Loglevel: All (XEN) Xen is relinquishing VGA console. (XEN) *** Serial input -> DOM0 (type ''CTRL-a'' three times to switch input to Xen) (XEN) Freed 260kB init memory. (XEN) PCI add device 0000:00:00.0 (XEN) PCI add device 0000:00:01.0 (XEN) PCI add device 0000:00:01.1 (XEN) PCI add device 0000:00:02.0 (XEN) PCI add device 0000:00:03.0 (XEN) PCI add device 0000:00:03.2 (XEN) PCI add device 0000:00:04.0 (XEN) PCI add device 0000:00:04.1 (XEN) PCI add device 0000:00:04.2 (XEN) PCI add device 0000:00:04.3 (XEN) PCI add device 0000:00:04.4 (XEN) PCI add device 0000:00:04.5 (XEN) PCI add device 0000:00:04.6 (XEN) PCI add device 0000:00:04.7 (XEN) PCI add device 0000:00:05.0 (XEN) PCI add device 0000:00:05.2 (XEN) PCI add device 0000:00:05.4 (XEN) PCI add device 0000:00:11.0 (XEN) PCI add device 0000:00:16.0 (XEN) PCI add device 0000:00:16.1 (XEN) PCI add device 0000:00:1a.0 (XEN) PCI add device 0000:00:1d.0 (XEN) PCI add device 0000:00:1e.0 (XEN) PCI add device 0000:00:1f.0 (XEN) PCI add device 0000:00:1f.2 (XEN) PCI add device 0000:00:1f.3 (XEN) PCI add device 0000:02:00.0 (XEN) PCI add device 0000:02:00.1 (XEN) PCI add device 0000:04:00.0 (XEN) PCI add device 0000:07:00.0 (XEN) PCI add device 0000:07:00.3 (XEN) PCI add device 0000:08:03.0 (XEN) Cannot bind IRQ2 to dom0. In use by ''cascade''. (XEN) Cannot bind IRQ2 to dom0. In use by ''cascade''. (XEN) Cannot bind IRQ2 to dom0. In use by ''cascade''. (XEN) Cannot bind IRQ2 to dom0. In use by ''cascade''. (XEN) grant_table.c:278:d0 Increased maptrack size to 2 frames (XEN) grant_table.c:278:d0 Increased maptrack size to 3 frames (XEN) grant_table.c:278:d0 Increased maptrack size to 4 frames (XEN) grant_table.c:278:d0 Increased maptrack size to 5 frames (XEN) grant_table.c:278:d0 Increased maptrack size to 6 frames (XEN) grant_table.c:278:d0 Increased maptrack size to 7 frames (XEN) grant_table.c:278:d0 Increased maptrack size to 8 frames (XEN) grant_table.c:278:d0 Increased maptrack size to 9 frames _______________________________________________ Xen-users mailing list Xen-users@lists.xen.org http://lists.xen.org/xen-users
On Wed, 2013-04-24 at 11:02 +0100, Vadim A. Misbakh-Soloviov wrote:> 24.04.2013 16:29, Ian Campbell пишет:> (XEN) ACPI Error (tbxfroot-0218): A valid RSDP was not found [20070126]Not having working ACPI is a bit of a problem and could well relate to the lack of HT. Updating to the latest BIOS for your system would seem like a good first step. Ian. _______________________________________________ Xen-users mailing list Xen-users@lists.xen.org http://lists.xen.org/xen-users
> Updating to the latest BIOS for your system would seem like a good first > step.Unfortunately, it is already updated to the latest version (even before Xen installation). _______________________________________________ Xen-users mailing list Xen-users@lists.xen.org http://lists.xen.org/xen-users
I''ve also just talked with guys from datacenter (where my server placed) and discovered, that they using same motherboard on their new VDS server. They also using Xen-4.1. And they not experience that problem (all HT threads shows as they should). The only difference, is they''re using another CPUs (and using both CPU slots on the motherboard, while my server - uses only one). _______________________________________________ Xen-users mailing list Xen-users@lists.xen.org http://lists.xen.org/xen-users
By the way, Ian. JFYI: I just rebooted machine and booted up Xen directly via EFI (previously it was booted via GRUB2, which booted via EFI). Now Xen see all 12 threads-cores. So, as far, as I googled, it seems, it is kind of common bug, that Xen can''t see some processor cores, when boots under GRUB2. //btw, I''ve another issue with dom0, which I can''t fix up about a week:> node1 ~ % hwclock -r > hwclock: select() to /dev/rtc to wait for clock tick timed out: Success > node1 ~ % hwclock -r --directisa > hwclock: select() to /dev/rtc to wait for clock tick timed out: SuccessAnd even making thing like such:> node1 ~ % hwclock --utc --noadjfile --set --date="2013-04-25 05:56:00"don''t fixes anything :( Most strange that it happens only when kernel booted up as dom0, but do not, when baremetal. Hasn''t Xen some ways to fix RTC emulation in hypervisor? :) _______________________________________________ Xen-users mailing list Xen-users@lists.xen.org http://lists.xen.org/xen-users
On Thu, 2013-04-25 at 02:59 +0100, Vadim A. Misbakh-Soloviov wrote:> By the way, Ian. JFYI: > I just rebooted machine and booted up Xen directly via EFI (previously > it was booted via GRUB2, which booted via EFI). > Now Xen see all 12 threads-cores. > > So, as far, as I googled, it seems, it is kind of common bug, that Xen > can''t see some processor cores, when boots under GRUB2.Thanks, I wasn''t aware of this issue but now I know.> //btw, I''ve another issue with dom0, which I can''t fix up about a week: > > > node1 ~ % hwclock -r > > hwclock: select() to /dev/rtc to wait for clock tick timed out: Success > > node1 ~ % hwclock -r --directisa > > hwclock: select() to /dev/rtc to wait for clock tick timed out: Success > > And even making thing like such: > > node1 ~ % hwclock --utc --noadjfile --set --date="2013-04-25 05:56:00" > don''t fixes anything :( > > Most strange that it happens only when kernel booted up as dom0, but do > not, when baremetal. > Hasn''t Xen some ways to fix RTC emulation in hypervisor? :)dom0 only sees PV wallclock time and not an RTC (emulated or real) since the real RTC is owned by the hypervisor. There is supposed to be some way to ask the hypervisor to set the RTC time, but I''m afraid I don''t recall how it fits together on a pvops kernel. I know that on a classic Xen kernel if you ran ntpd (or ntpdate?) it would get pushed down, but I don''t think pvops kernels do that. Sorry, that''s all I can remember. Ian.
> There is supposed to be some way to ask the hypervisor to set the RTC > time, but I''m afraid I don''t recall how it fits together on a pvops > kernel. I know that on a classic Xen kernel if you ran ntpd (or > ntpdate?) it would get pushed down, but I don''t think pvops kernels do > that. > > Sorry, that''s all I can remember.Fortunatelly, all issues was fixed by booting Xen directly from EFI and passing acpi_rsdp (with Xen-reported value) to the dom0 kernel ;) So, the only things I still discouraged is: 1) "xl shutdown -a" segfaults with SELinux enabled (even in permissive mode). While stopping each domU separately is okay. 2) efibootmgr doesn''t work under Xen (which is common bug [1] too, but it seems, nobody wants to fix this thing in Xen. So, it is impossible to change path to Xen on EFI partition without booting to baremetal kernel (<3.8, btw, since =3.8 has issue preventing write to EFI table [2]) [1] http://lists.xen.org/archives/html/xen-devel/2012-05/msg02060.html [2] http://thread.gmane.org/gmane.linux.kernel.efi/1081 _______________________________________________ Xen-users mailing list Xen-users@lists.xen.org http://lists.xen.org/xen-users
On Thu, 2013-04-25 at 09:16 +0100, Vadim A. Misbakh-Soloviov wrote:> > There is supposed to be some way to ask the hypervisor to set the RTC > > time, but I''m afraid I don''t recall how it fits together on a pvops > > kernel. I know that on a classic Xen kernel if you ran ntpd (or > > ntpdate?) it would get pushed down, but I don''t think pvops kernels do > > that. > > > > Sorry, that''s all I can remember. > > Fortunatelly, all issues was fixed by booting Xen directly from EFI and > passing > acpi_rsdp (with Xen-reported value) to the dom0 kernel ;) > > So, the only things I still discouraged is: > 1) "xl shutdown -a" segfaults with SELinux enabled (even in permissive > mode). While stopping each domU separately is okay.Only with selinux enabled? How odd. Please can you report (in a separate thread) with logs and a gdb backtrace. Since it is a segfault I think you may as well go straight to xen-devel with the report.> 2) efibootmgr doesn''t work under Xen (which is common bug [1] too, but > it seems, nobody wants to fix this thing in Xen.Nobody "wants to" is a bit strong, although it is true that nobody *has*. Patches which improve Xen under EFI (either directly booting or via a boot manager) would be greatly appreciated.> So, it is impossible to change path to Xen on EFI partition without > booting to baremetal kernel (<3.8, btw, since =3.8 has issue preventing > write to EFI table [2]) > > [1] http://lists.xen.org/archives/html/xen-devel/2012-05/msg02060.html > [2] http://thread.gmane.org/gmane.linux.kernel.efi/1081 >
That's a problem I've been stuck on twice ;-|. With EFI, xen needs to be booted with xen.efi; any other boot loader detects ACPI tables before starting xen kernel and probably masks or blocks direct access to the raw data so the latter can't find the root pointer for tables anymore. Thus the message about not finding RSDP (Root System Description Pointer). After that anything strange can (and probably will) happen: Xen detects wrong amount of RAM or cores, wrong capabilities on CPUs etc. Probably, most of the Xen admins have already discovered that fact in the process already, yet I still think it wouldn't hurt to document that somewhere in Xen wiki: On an EFI enabled server Xen has to be booted via xen.efi loader. Some OS installs do set the appropriate boot up automatically (SLES for example). Some, OSes, however, do not (Opensuse, for example) and you have to manage the boot entry for Xen manually. Regards, Janno 24.04.2013 13:14, Ian Campbell wrote:> On Wed, 2013-04-24 at 11:02 +0100, Vadim A. Misbakh-Soloviov wrote: >> 24.04.2013 16:29, Ian Campbell пишет: >> (XEN) ACPI Error (tbxfroot-0218): A valid RSDP was not found [20070126] > Not having working ACPI is a bit of a problem and could well relate to > the lack of HT. > > Updating to the latest BIOS for your system would seem like a good first > step. > > Ian. >_______________________________________________ Xen-users mailing list Xen-users@lists.xen.org http://lists.xen.org/xen-users
Possibly Parallel Threads
- xenwatch: page allocation failure: order:4, mode:0x10c0d0 xen_netback:xenvif_alloc: Could not allocate netdev for vif16.0
- [GIT PULL] (xen) stable/for-jens-3.10
- Possibility to dinamically change cHVC size
- Possibility to dinamically change cHVC size
- RFC: [0/2] Remove netloop by lazy copying in netback