virtio-net working on PowerPC KVM

virtio-net working on PowerPC KVM

[Hollis Blanchard]

[-- Attachment #1: Type: text/plain, Size: 618 bytes --]

I'm pleased to report that we now have working network support in the
guest, via the virtio-net driver. In fact, we can use NFS for the
guest's root filesystem. :) Boot log attached.

The bad news is that it's very slow, but the good news is that it's nice
to be improving performance rather than debugging mysterious
crashes... ;)

With this milestone reached, in the near future I intend to start
sending patches to Avi and linuxppc-dev for review, hopefully for
inclusion in 2.6.26. However, I do want to see if we can improve the
performance a little bit first...

-- 
Hollis Blanchard
IBM Linux Technology Center

[-- Attachment #2: kvmppc_440_vioboot.txt --]
[-- Type: text/plain, Size: 6683 bytes --]

bash-3.00# ./qemu-system-ppcemb -M bamboo -nographic -kernel ../../uImage.bamboo -L ../pc-bios/ -append "ip=dhcp nfsroot=9.XX.XX.XX:/netroot/ppc_4xx root=/dev/nfs rw debug" -net nic,model=virtio -net tap 
bamboo_init: START
Ram size passed is: 144 MB
Calling function ppc440_init
setup mmio
setup universal controller
trying to setup sdram controller
sdram_unmap_bcr: Unmap RAM area 0000000000000000 00400000
sdram_unmap_bcr: Unmap RAM area 0000000000000000 00400000
sdram_set_bcr: Map RAM area 0000000000000000 08000000
sdram_set_bcr: Map RAM area 0000000000000000 01000000
Initializing first serial port
ppc405_serial_init: offset 0000000000000300
Done calling ppc440_init
bamboo_init: load kernel
kernel is at guest address: 0x0
bamboo_init: load device tree file
device tree address is at guest address: 0x2b2100
bamboo_init: loading kvm registers
bamboo_init: DONE
Using Bamboo machine description
Linux version 2.6.25-rc3-hg1858cec8eb87-dirty (hollisb@basalt) (gcc version 3.4.2) #150 Wed Mar 19 12:51:20 CDT 2008
Found legacy serial port 0 for /plb/opb/serial@ef600300
  mem=ef600300, taddr=ef600300, irq=0, clk=11059200, speed=115200
Found legacy serial port 1 for /plb/opb/serial@ef600400
  mem=ef600400, taddr=ef600400, irq=0, clk=11059200, speed=0
console [udbg0] enabled
Entering add_active_range(0, 0, 36864) 0 entries of 256 used
setup_arch: bootmem
arch: exit
Top of RAM: 0x9000000, Total RAM: 0x9000000
Memory hole size: 0MB
Zone PFN ranges:
  DMA             0 ->    36864
  Normal      36864 ->    36864
Movable zone start PFN for each node
early_node_map[1] active PFN ranges
    0:        0 ->    36864
On node 0 totalpages: 36864
  DMA zone: 288 pages used for memmap
  DMA zone: 0 pages reserved
  DMA zone: 36576 pages, LIFO batch:7
  Normal zone: 0 pages used for memmap
  Movable zone: 0 pages used for memmap
Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 36576
Kernel command line: ip=dhcp nfsroot=9.XX.XX.XX:/netroot/ppc_4xx root=/dev/nfs rw debug
irq: Allocated host of type 2 @0xc03f3900
UIC0 (32 IRQ sources) at DCR 0xc0
irq: Default host set to @0xc03f3900
PID hash table entries: 1024 (order: 10, 4096 bytes)
time_init: decrementer frequency = 666.666660 MHz
time_init: processor frequency   = 666.666660 MHz
clocksource: timebase mult[600000] shift[22] registered
clockevent: decrementer mult[aaaa] shift[16] cpu[0]
Dentry cache hash table entries: 32768 (order: 5, 131072 bytes)
Inode-cache hash table entries: 16384 (order: 4, 65536 bytes)
Memory: 143060k/147456k available (2632k kernel code, 4252k reserved, 100k data, 125k bss, 132k init)
SLUB: Genslabs=10, HWalign=32, Order=0-1, MinObjects=4, CPUs=1, Nodes=1
Calibrating delay loop... 2490.36 BogoMIPS (lpj=4980736)
Mount-cache hash table entries: 512
net_namespace: 156 bytes
NET: Registered protocol family 16
             
PCI host bridge /plb/pci@ec000000 (primary) ranges:
 MEM 0x00000000a0000000..0x00000000bfffffff -> 0x00000000a0000000 
  IO 0x00000000e8000000..0x00000000e800ffff -> 0x0000000000000000
4xx PCI DMA offset set to 0x00000000
PCI: Probing PCI hardware
PCI: Hiding 4xx host bridge resources 0000:00:00.0
irq: irq_create_mapping(0xc03f3900, 0x1c)
irq: -> using host @c03f3900
irq: -> obtained virq 28
Time: timebase clocksource has been installed.
NET: Registered protocol family 2
IP route cache hash table entries: 2048 (order: 1, 8192 bytes)
TCP established hash table entries: 8192 (order: 4, 65536 bytes)
TCP bind hash table entries: 8192 (order: 3, 32768 bytes)
TCP: Hash tables configured (established 8192 bind 8192)
TCP reno registered
irq: irq_create_mapping(0xc03f3900, 0x0)
irq: -> using host @c03f3900
irq: -> obtained virq 16
irq: irq_create_mapping(0xc03f3900, 0x1)
irq: -> using host @c03f3900
irq: -> obtained virq 17
io scheduler noop registered
io scheduler anticipatory registered (default)
io scheduler deadline registered
io scheduler cfq registered
Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing enabled
serial8250.0: ttyS0 at MMIO 0xef600300 (irq = 16) is a 16450
console handover: boot [udbg0] -> real [ttyS0]
irq: irq_create_mapping(0xc03f3900, 0x0)
irq: -> using host @c03f3900
irq: -> existing mapping on virq 16
ef600300.serial: ttyS0 at MMIO 0xef600300 (irq = 16) is a 16450
irq: irq_create_mapping(0xc03f3900, 0x1)
irq: -> using host @c03f3900
irq: -> existing mapping on virq 17
brd: module loaded
Intel(R) PRO/1000 Network Driver - version 7.3.20-k2
Copyright (c) 1999-2006 Intel Corporation.
pcnet32.c:v1.34 14.Aug.2007 tsbogend@alpha.franken.de
tun: Universal TUN/TAP device driver, 1.6
tun: (C) 1999-2004 Max Krasnyansky <maxk@qualcomm.com>
PCI: Enabling device 0000:00:01.0 (0000 -> 0001)
TCP cubic registered
NET: Registered protocol family 1
NET: Registered protocol family 17
RPC: Registered udp transport module.
RPC: Registered tcp transport module.
Sending DHCP requests .., OK
IP-Config: Got DHCP answer from 255.255.255.255, my address is 9.XX.XX.XX
IP-Config: Complete:
     device=eth0, addr=9.XX.XX.XX, mask=255.255.255.0, gw=9.XX.XX.XX,
     host=9.XX.XX.XX, domain=austin.ibm.com, nis-domain=(none),
     bootserver=255.255.255.255, rootserver=9.XX.XX.XX, rootpath=
Looking up port of RPC 100003/2 on 9.XX.XX.XX
Looking up port of RPC 100005/1 on 9.XX.XX.XX
VFS: Mounted root (nfs filesystem).
Freeing unused kernel memory: 132k init
modprobe: FATAL: Could not load /lib/modules/2.6.25-rc3-hg1858cec8eb87-dirty/modules.dep: No such file or directory

modprobe: FATAL: Could not load /lib/modules/2.6.25-rc3-hg1858cec8eb87-dirty/modules.dep: No such file or directory

INIT: version 2.85 booting
		Welcome to DENX Embedded Linux Environment
		Press 'I' to enter interactive startup.
Building the cache [  OK  ]
 storage network audio done[  OK  ]
modprobe: FATAL: Could not load /lib/modules/2.6.25-rc3-hg1858cec8eb87-dirty/modules.dep: No such file or directory

Cannot access the Hardware Clock via any known method.
Use the --debug option to see the details of our search for an access method.
Setting clock : Thu Jan  1 00:04:09 UTC 1970 [  OK  ]
Setting hostname 9.53.41.171:  [  OK  ]
Mounting local filesystems:  [  OK  ]
Enabling swap space:  [  OK  ]
INIT: Entering runlevel: 3
Entering non-interactive startup
Bringing up loopback interface:  [  OK  ]
Starting system logger: [  OK  ]
Starting kernel logger: [  OK  ]
Starting portmap: [  OK  ]
Mounting NFS filesystems:  [  OK  ]
Mounting other filesystems:  [  OK  ]
Starting xinetd: [  OK  ]
add bridge failed: Package not installed
interface eth1 does not exist!
SIOCGIFFLAGS: No such device
SIOCGIFFLAGS: No such device

DENX ELDK version 4.1 build 2007-01-19
Linux 2.6.25-rc3-hg1858cec8eb87-dirty on a ppc

9 login: 

[-- Attachment #3: Type: text/plain, Size: 278 bytes --]

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

[-- Attachment #4: Type: text/plain, Size: 158 bytes --]

_______________________________________________
kvm-devel mailing list
kvm-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/kvm-devel

Re: [kvm-ppc-devel] virtio-net working on PowerPC KVM

[Zhang Wei]

Hi, Hollis,

Hollis Blanchard wrote:
> I'm pleased to report that we now have working network support in the
> guest, via the virtio-net driver. In fact, we can use NFS for the
> guest's root filesystem. :) Boot log attached.
>

Congratulations! That's a milestone!
 
> The bad news is that it's very slow, but the good news is that it's nice
> to be improving performance rather than debugging mysterious
> crashes... ;)
> 
> With this milestone reached, in the near future I intend to start
> sending patches to Avi and linuxppc-dev for review, hopefully for
> inclusion in 2.6.26. However, I do want to see if we can improve the
> performance a little bit first...
> 
> 
> 
> ------------------------------------------------------------------------
> 
> bash-3.00# ./qemu-system-ppcemb -M bamboo -nographic -kernel ../../uImage.bamboo -L ../pc-bios/ -append "ip=dhcp nfsroot=9.XX.XX.XX:/netroot/ppc_4xx root=/dev/nfs rw debug" -net nic,model=virtio -net tap 
> bamboo_init: START
> Ram size passed is: 144 MB
> Calling function ppc440_init
> setup mmio
> setup universal controller
> trying to setup sdram controller
> sdram_unmap_bcr: Unmap RAM area 0000000000000000 00400000
> sdram_unmap_bcr: Unmap RAM area 0000000000000000 00400000
> sdram_set_bcr: Map RAM area 0000000000000000 08000000
> sdram_set_bcr: Map RAM area 0000000000000000 01000000
> Initializing first serial port
> ppc405_serial_init: offset 0000000000000300
> Done calling ppc440_init
> bamboo_init: load kernel
> kernel is at guest address: 0x0
> bamboo_init: load device tree file
> device tree address is at guest address: 0x2b2100
> bamboo_init: loading kvm registers
> bamboo_init: DONE
> Using Bamboo machine description
> Linux version 2.6.25-rc3-hg1858cec8eb87-dirty (hollisb@basalt) (gcc version 3.4.2) #150 Wed Mar 19 12:51:20 CDT 2008
> Found legacy serial port 0 for /plb/opb/serial@ef600300
>   mem=ef600300, taddr=ef600300, irq=0, clk=11059200, speed=115200
> Found legacy serial port 1 for /plb/opb/serial@ef600400
>   mem=ef600400, taddr=ef600400, irq=0, clk=11059200, speed=0
> console [udbg0] enabled
> Entering add_active_range(0, 0, 36864) 0 entries of 256 used
> setup_arch: bootmem
> arch: exit
> Top of RAM: 0x9000000, Total RAM: 0x9000000
> Memory hole size: 0MB
> Zone PFN ranges:
>   DMA             0 ->    36864
>   Normal      36864 ->    36864
> Movable zone start PFN for each node
> early_node_map[1] active PFN ranges
>     0:        0 ->    36864
> On node 0 totalpages: 36864
>   DMA zone: 288 pages used for memmap
>   DMA zone: 0 pages reserved
>   DMA zone: 36576 pages, LIFO batch:7
>   Normal zone: 0 pages used for memmap
>   Movable zone: 0 pages used for memmap
> Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 36576
> Kernel command line: ip=dhcp nfsroot=9.XX.XX.XX:/netroot/ppc_4xx root=/dev/nfs rw debug
> irq: Allocated host of type 2 @0xc03f3900
> UIC0 (32 IRQ sources) at DCR 0xc0
> irq: Default host set to @0xc03f3900
> PID hash table entries: 1024 (order: 10, 4096 bytes)
> time_init: decrementer frequency = 666.666660 MHz
> time_init: processor frequency   = 666.666660 MHz
> clocksource: timebase mult[600000] shift[22] registered
> clockevent: decrementer mult[aaaa] shift[16] cpu[0]
> Dentry cache hash table entries: 32768 (order: 5, 131072 bytes)
> Inode-cache hash table entries: 16384 (order: 4, 65536 bytes)
> Memory: 143060k/147456k available (2632k kernel code, 4252k reserved, 100k data, 125k bss, 132k init)
> SLUB: Genslabs=10, HWalign=32, Order=0-1, MinObjects=4, CPUs=1, Nodes=1
> Calibrating delay loop... 2490.36 BogoMIPS (lpj=4980736)
> Mount-cache hash table entries: 512
> net_namespace: 156 bytes
> NET: Registered protocol family 16
>              
> PCI host bridge /plb/pci@ec000000 (primary) ranges:
>  MEM 0x00000000a0000000..0x00000000bfffffff -> 0x00000000a0000000 
>   IO 0x00000000e8000000..0x00000000e800ffff -> 0x0000000000000000
> 4xx PCI DMA offset set to 0x00000000
> PCI: Probing PCI hardware
> PCI: Hiding 4xx host bridge resources 0000:00:00.0
> irq: irq_create_mapping(0xc03f3900, 0x1c)
> irq: -> using host @c03f3900
> irq: -> obtained virq 28
> Time: timebase clocksource has been installed.
> NET: Registered protocol family 2
> IP route cache hash table entries: 2048 (order: 1, 8192 bytes)
> TCP established hash table entries: 8192 (order: 4, 65536 bytes)
> TCP bind hash table entries: 8192 (order: 3, 32768 bytes)
> TCP: Hash tables configured (established 8192 bind 8192)
> TCP reno registered
> irq: irq_create_mapping(0xc03f3900, 0x0)
> irq: -> using host @c03f3900
> irq: -> obtained virq 16
> irq: irq_create_mapping(0xc03f3900, 0x1)
> irq: -> using host @c03f3900
> irq: -> obtained virq 17
> io scheduler noop registered
> io scheduler anticipatory registered (default)
> io scheduler deadline registered
> io scheduler cfq registered
> Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing enabled
> serial8250.0: ttyS0 at MMIO 0xef600300 (irq = 16) is a 16450
> console handover: boot [udbg0] -> real [ttyS0]
> irq: irq_create_mapping(0xc03f3900, 0x0)
> irq: -> using host @c03f3900
> irq: -> existing mapping on virq 16
> ef600300.serial: ttyS0 at MMIO 0xef600300 (irq = 16) is a 16450
> irq: irq_create_mapping(0xc03f3900, 0x1)
> irq: -> using host @c03f3900
> irq: -> existing mapping on virq 17
> brd: module loaded
> Intel(R) PRO/1000 Network Driver - version 7.3.20-k2
> Copyright (c) 1999-2006 Intel Corporation.
> pcnet32.c:v1.34 14.Aug.2007 tsbogend@alpha.franken.de
> tun: Universal TUN/TAP device driver, 1.6
> tun: (C) 1999-2004 Max Krasnyansky <maxk@qualcomm.com>
> PCI: Enabling device 0000:00:01.0 (0000 -> 0001)
> TCP cubic registered
> NET: Registered protocol family 1
> NET: Registered protocol family 17
> RPC: Registered udp transport module.
> RPC: Registered tcp transport module.
> Sending DHCP requests .., OK
> IP-Config: Got DHCP answer from 255.255.255.255, my address is 9.XX.XX.XX
> IP-Config: Complete:
>      device=eth0, addr=9.XX.XX.XX, mask=255.255.255.0, gw=9.XX.XX.XX,
>      host=9.XX.XX.XX, domain=austin.ibm.com, nis-domain=(none),
>      bootserver=255.255.255.255, rootserver=9.XX.XX.XX, rootpath=
> Looking up port of RPC 100003/2 on 9.XX.XX.XX
> Looking up port of RPC 100005/1 on 9.XX.XX.XX
> VFS: Mounted root (nfs filesystem).
> Freeing unused kernel memory: 132k init
> modprobe: FATAL: Could not load /lib/modules/2.6.25-rc3-hg1858cec8eb87-dirty/modules.dep: No such file or directory
> 
> modprobe: FATAL: Could not load /lib/modules/2.6.25-rc3-hg1858cec8eb87-dirty/modules.dep: No such file or directory
> 
> INIT: version 2.85 booting
> 		Welcome to DENX Embedded Linux Environment
> 		Press 'I' to enter interactive startup.
> Building the cache [  OK  ]
>  storage network audio done[  OK  ]
> modprobe: FATAL: Could not load /lib/modules/2.6.25-rc3-hg1858cec8eb87-dirty/modules.dep: No such file or directory
> 
> Cannot access the Hardware Clock via any known method.
> Use the --debug option to see the details of our search for an access method.
> Setting clock : Thu Jan  1 00:04:09 UTC 1970 [  OK  ]
> Setting hostname 9.53.41.171:  [  OK  ]
> Mounting local filesystems:  [  OK  ]
> Enabling swap space:  [  OK  ]
> INIT: Entering runlevel: 3
> Entering non-interactive startup
> Bringing up loopback interface:  [  OK  ]
> Starting system logger: [  OK  ]
> Starting kernel logger: [  OK  ]
> Starting portmap: [  OK  ]
> Mounting NFS filesystems:  [  OK  ]
> Mounting other filesystems:  [  OK  ]
> Starting xinetd: [  OK  ]
> add bridge failed: Package not installed
> interface eth1 does not exist!
> SIOCGIFFLAGS: No such device
> SIOCGIFFLAGS: No such device
> 
> DENX ELDK version 4.1 build 2007-01-19
> Linux 2.6.25-rc3-hg1858cec8eb87-dirty on a ppc
> 
> 9 login: 
> 
> 
> ------------------------------------------------------------------------
> 
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> kvm-ppc-devel mailing list
> kvm-ppc-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/kvm-ppc-devel

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

Re: virtio-net working on PowerPC KVM

[Avi Kivity]

Hollis Blanchard wrote:
> I'm pleased to report that we now have working network support in the
> guest, via the virtio-net driver. In fact, we can use NFS for the
> guest's root filesystem. :) Boot log attached.
>
>   

Congrats!

> The bad news is that it's very slow, but the good news is that it's nice
> to be improving performance rather than debugging mysterious
> crashes... ;)
>
> With this milestone reached, in the near future I intend to start
> sending patches to Avi and linuxppc-dev for review, hopefully for
> inclusion in 2.6.26. However, I do want to see if we can improve the
> performance a little bit first...
>   

Low virtio net performance may be due to the virtio net host timer.  
What's your guest/host ping latency?

Even if you have a good hrtimer implementation, I think you'll see 
0.25ms latency, and that may be enough to slow down nfs.  Unfortunately 
virtio is tuned for throughput at this time (it should be easy to 
disable the timer when we detect the queue is usually empty).

-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

Re: [kvm-ppc-devel] virtio-net working on PowerPC KVM

[Christian Ehrhardt]

Avi Kivity wrote:
> Hollis Blanchard wrote:
>> I'm pleased to report that we now have working network support in the
>> guest, via the virtio-net driver. In fact, we can use NFS for the
>> guest's root filesystem. :) Boot log attached.
>>
>>   
> 
> Congrats!
> 
>> The bad news is that it's very slow, but the good news is that it's nice
>> to be improving performance rather than debugging mysterious
>> crashes... ;)
>>
>> With this milestone reached, in the near future I intend to start
>> sending patches to Avi and linuxppc-dev for review, hopefully for
>> inclusion in 2.6.26. However, I do want to see if we can improve the
>> performance a little bit first...
>>   
> 
> Low virtio net performance may be due to the virtio net host timer.  
> What's your guest/host ping latency?

I would be happy about 0.25ms atm :-). The current ping latency to the Host or other PC's is around 7-8ms (native sys is ~0.15ms).
We are investigating performance improvements in general and also some changes in the setup e.g. booting from virtio-block as alternative for some speedup.

> Even if you have a good hrtimer implementation, I think you'll see 
> 0.25ms latency, and that may be enough to slow down nfs.  Unfortunately 
> virtio is tuned for throughput at this time (it should be easy to 
> disable the timer when we detect the queue is usually empty).
> 


-- 

Grüsse / regards, 
Christian Ehrhardt
IBM Linux Technology Center, Open Virtualization

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

Re: [kvm-ppc-devel] virtio-net working on PowerPC KVM

[Avi Kivity]

Alexander Graf wrote:
>
> I am experiencing 7-8ms ping latencies (native 0.1ms) on x86_64 as 
> well, when pinging the virtual machine. Maybe it's not related to 
> PowerPC? Is it supposed to be that slow?

If you have a really old host kernel, or a misconfigured one, it can 
happen.  What's your host kernel?  are you using hrtimers?

-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.


-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

Re: [kvm-ppc-devel] virtio-net working on PowerPC KVM

[Alexander Graf]

On Mar 31, 2008, at 6:52 AM, Christian Ehrhardt wrote:

> Avi Kivity wrote:
>> Hollis Blanchard wrote:
>>> I'm pleased to report that we now have working network support in  
>>> the
>>> guest, via the virtio-net driver. In fact, we can use NFS for the
>>> guest's root filesystem. :) Boot log attached.
>>>
>>>
>>
>> Congrats!
>>
>>> The bad news is that it's very slow, but the good news is that  
>>> it's nice
>>> to be improving performance rather than debugging mysterious
>>> crashes... ;)
>>>
>>> With this milestone reached, in the near future I intend to start
>>> sending patches to Avi and linuxppc-dev for review, hopefully for
>>> inclusion in 2.6.26. However, I do want to see if we can improve the
>>> performance a little bit first...
>>>
>>
>> Low virtio net performance may be due to the virtio net host timer.
>> What's your guest/host ping latency?
>
> I would be happy about 0.25ms atm :-). The current ping latency to  
> the Host or other PC's is around 7-8ms (native sys is ~0.15ms).
> We are investigating performance improvements in general and also  
> some changes in the setup e.g. booting from virtio-block as  
> alternative for some speedup.

I am experiencing 7-8ms ping latencies (native 0.1ms) on x86_64 as  
well, when pinging the virtual machine. Maybe it's not related to  
PowerPC? Is it supposed to be that slow?

>
>
>> Even if you have a good hrtimer implementation, I think you'll see
>> 0.25ms latency, and that may be enough to slow down nfs.   
>> Unfortunately
>> virtio is tuned for throughput at this time (it should be easy to
>> disable the timer when we detect the queue is usually empty).
>>

Alex

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

Re: [kvm-ppc-devel] virtio-net working on PowerPC KVM

[Alexander Graf]

On Mar 31, 2008, at 9:43 AM, Avi Kivity wrote:

> Alexander Graf wrote:
>>
>> I am experiencing 7-8ms ping latencies (native 0.1ms) on x86_64 as
>> well, when pinging the virtual machine. Maybe it's not related to
>> PowerPC? Is it supposed to be that slow?
>
> If you have a really old host kernel, or a misconfigured one, it can
> happen.  What's your host kernel?  are you using hrtimers?

This is a 2.6.22 SUSE kernel. As far as I can see  
CONFIG_HIGH_RES_TIMERS is not set.

Guess it's my fault then - nevertheless a warning when starting kvm  
would be nice.

Alex

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

Re: [kvm-ppc-devel] virtio-net working on PowerPC KVM

[Avi Kivity]

Alexander Graf wrote:
>
> On Mar 31, 2008, at 9:43 AM, Avi Kivity wrote:
>
>> Alexander Graf wrote:
>>>
>>> I am experiencing 7-8ms ping latencies (native 0.1ms) on x86_64 as
>>> well, when pinging the virtual machine. Maybe it's not related to
>>> PowerPC? Is it supposed to be that slow?
>>
>> If you have a really old host kernel, or a misconfigured one, it can
>> happen.  What's your host kernel?  are you using hrtimers?
>
> This is a 2.6.22 SUSE kernel. As far as I can see 
> CONFIG_HIGH_RES_TIMERS is not set.
>

An alternative is -clock hpet or -clock rtc.

> Guess it's my fault then - nevertheless a warning when starting kvm 
> would be nice.

Yes, indeed.


-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.


-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

Re: [kvm-ppc-devel] virtio-net working on PowerPC KVM

[Dor Laor]

On Mon, 2008-03-31 at 10:17 +0200, Alexander Graf wrote:
> On Mar 31, 2008, at 9:43 AM, Avi Kivity wrote:
> 
> > Alexander Graf wrote:
> >>
> >> I am experiencing 7-8ms ping latencies (native 0.1ms) on x86_64 as
> >> well, when pinging the virtual machine. Maybe it's not related to
> >> PowerPC? Is it supposed to be that slow?
> >
> > If you have a really old host kernel, or a misconfigured one, it can
> > happen.  What's your host kernel?  are you using hrtimers?
> 
> This is a 2.6.22 SUSE kernel. As far as I can see  
> CONFIG_HIGH_RES_TIMERS is not set.
> 
> Guess it's my fault then - nevertheless a warning when starting kvm  
> would be nice.
> 

You can also try with -clock=hpet or -clock=rtc

> Alex
> 
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
> _______________________________________________
> kvm-devel mailing list
> kvm-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/kvm-devel


-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace