Hi, "PÁSZTOR György" <pasz...@linux.gyakg.u-szeged.hu> írta 2011-03-28 19:32-kor: > Ötleteket keresek, hogy mit nézzek még meg, mert már kezdenek a hajszálaim > kihullani ettől a problémától. ... > Amit semmi nem magyaráz: Miért lassul rettenetesen le, amikor a xen-es > kernelt XEN hypervisorral együtt bootolom a node05-re:
Közben figyelmes lettem néhány üzenetre a xen dmesgjéből, hátha ez valakinek mond valamit: (XEN) ACPI: RSDP 000FDFD0, 0024 (r2 IBM ) (XEN) ACPI: XSDT 7F7FE120, 0094 (r1 IBM BLADE 0 1000013) (XEN) ACPI: FACP 7F7FB000, 00F4 (r4 IBM BLADE 0 IBM 1000013) (XEN) ACPI: DSDT 7F7F8000, 224E (r1 IBM BLADE 3 IBM 1000013) (XEN) ACPI: FACS 7F72C000, 0040 (XEN) ACPI: TCPA 7F7FD000, 0064 (r0 0 0) ... (XEN) HVM: Hardware Assisted Paging detected. (XEN) Intel VT-d Snoop Control supported. (XEN) Intel VT-d DMA Passthrough not supported. (XEN) Intel VT-d Queued Invalidation supported. (XEN) Intel VT-d Interrupt Remapping not supported. - Nem gáz az, h. dma passthrough not supported? A többi VT-d -s featúrát nem is értem mi. dom0 dmesg-jében: [ 0.000000] ACPI: RSDP 00000000000fdfd0 00024 (v02 IBM ) [ 0.000000] ACPI: XSDT 000000007f7fe120 00094 (v01 IBM BLADE 00000000 01000013) [ 0.000000] ACPI: FACP 000000007f7fb000 000F4 (v04 IBM BLADE 00000000 IBM 01000013) [ 0.000000] ACPI Warning: Invalid length for Pm1aControlBlock: 32, using default 16 (20090903/tbfadt-607) [ 0.000000] ACPI: DSDT 000000007f7f8000 0224E (v01 IBM BLADE 00000003 IBM 01000013) [ 0.000000] ACPI: FACS 000000007f72c000 00040 [ 0.000000] ACPI: TCPA 000000007f7fd000 00064 (v00 00000000 00000000) ... - Mi az, hogy a FACP meg a DSDT közt vmi. "invalid length" lesz? A xen érti, ha nincs hypervisor, akkor a "normál körülmények" közt futó kernel érti. Ez a retek mit, és miért nem ért? ... [ 0.000000] ACPI: Local APIC address 0xfee00000 [ 0.000000] No NUMA configuration found [ 0.000000] Faking a node at 0000000000000000-00000003e9c46000 ... - Miért nem talált numa konfigot? Nem kellett volna neki? normál kernelnél így néz ki:[ 0.000000] NUMA: Using 31 for the hash shift. (Meg ott lát is egy 0-s és egy 1-es node-t, és nem "faking"-el) Ez a numa fel nem ismerés milyen következményekkel járhat? Folytatva a hypervisoros környezet dom0-jának dmesgjét: [ 0.000000] IOAPIC[0]: apic_id 8, version 0, address 0xfec00000, GSI 0-0 [ 0.000000] ACPI: IOAPIC (id[0x09] address[0xfec80000] gsi_base[24]) [ 0.000000] IOAPIC[1]: apic_id 9, version 0, address 0xfec80000, GSI 24-24 [ 0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl) [ 0.000000] ERROR: Unable to locate IOAPIC for GSI 2 [ 0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level) [ 0.000000] ERROR: Unable to locate IOAPIC for GSI 9 [ 0.000000] Using ACPI (MADT) for SMP configuration information ... [ 3.839310] bio: create slab <bio-0> at 0 [ 3.840406] ERROR: Unable to locate IOAPIC for GSI 9 [ 3.840494] ACPI: EC: Look up EC in DSDT [ 3.847800] ACPI: Interpreter enabled [ 3.847804] ACPI: (supports S0 S1 S5) [ 3.847828] ACPI: Using IOAPIC for interrupt routing ... [ 4.296983] Broadcom NetXtreme II 5771x 10Gigabit Ethernet Driver bnx2x 1.52.1 (2009/08/12) [ 4.297114] xen: registering gsi 24 triggering 0 polarity 1 [ 4.297119] xen_allocate_pirq: returning irq 24 for gsi 24 [ 4.297121] xen: --> irq=24 [ 4.297127] Already setup the GSI :24 [ 4.297131] bnx2x 0000:15:00.0: PCI INT A -> GSI 24 (level, low) -> IRQ 24 [ 4.297144] bnx2x 0000:15:00.0: setting latency timer to 64 [ 4.299312] bnx2x: part number 394D4342-31373735-31314131-473036 [ 4.299333] bnx2x: Loading bnx2x-e1h-5.0.21.0.fw [ 4.299338] bnx2x 0000:15:00.0: firmware: requesting bnx2x-e1h-5.0.21.0.fw ... [ 4.352898] eth0: Broadcom NetXtreme II BCM57711 XGb (A0) PCI-E x8 5GHz (Gen2) found at mem 96000000, IRQ 24, node addr 00:10:18:5e:e6:b0 [ 4.352935] xen: registering gsi 34 triggering 0 polarity 1 [ 4.352945] alloc irq_desc for 34 on node -1 [ 4.352947] alloc kstat_irqs on node -1 [ 4.352952] xen: --> irq=34 [ 4.352960] bnx2x 0000:15:00.1: PCI INT B -> GSI 34 (level, low) -> IRQ 34 [ 4.352970] bnx2x 0000:15:00.1: setting latency timer to 64 [ 4.354851] bnx2x: part number 394D4342-31373735-31314131-473036 [ 4.354866] bnx2x: Loading bnx2x-e1h-5.0.21.0.fw [ 4.354869] bnx2x 0000:15:00.1: firmware: requesting bnx2x-e1h-5.0.21.0.fw ... [ 4.361848] eth1: Broadcom NetXtreme II BCM57711 XGb (A0) PCI-E x8 5GHz (Gen2) found at mem 96800000, IRQ 34, node addr 00:10:18:5e:e6:b2 ... [ 23.314723] bnx2x: eth0: using MSI-X IRQs: sp 2243 fp[0] 2242 ... fp[15] 2227 ... [ 23.756889] bnx2x: eth1: using MSI-X IRQs: sp 2226 fp[0] 2225 ... fp[15] 2210 ... [ 24.321285] bnx2x: eth0: using MSI-X IRQs: sp 2243 fp[0] 2242 ... fp[15] 2227 [ 24.856620] bnx2x: eth1: using MSI-X IRQs: sp 2226 fp[0] 2225 ... fp[15] 2210 [ 25.126880] bnx2x: eth0 NIC Link is Up, 10000 Mbps full duplex ... [ 27.260729] XENBUS: Unable to read cpu state [ 27.260814] device-mapper: ioctl: error adding target to table [ 27.260849] XENBUS: Unable to read cpu state [ 27.261189] XENBUS: Unable to read cpu state [ 27.261531] XENBUS: Unable to read cpu state [ 27.261753] XENBUS: Unable to read cpu state [ 27.261934] XENBUS: Unable to read cpu state [ 27.262108] XENBUS: Unable to read cpu state [ 27.262286] XENBUS: Unable to read cpu state Please help, aki tud! Üdv:Gyur! _________________________________________________ linux lista - linux@mlf.linux.rulez.org http://mlf2.linux.rulez.org/mailman/listinfo/linux
