Bug 1083873

The official netkvm driver support for winxp is virtio-win-prewhql-49 ,Pls retest it w/ virtio-win-prewhql-49 or latest virtio-win package.

Mike

From http://alt.fedoraproject.org/pub/alt/virtio-win/latest/images/ I can only get 0.1-74, which also have this issue.

and one mistake in my post that the driver version is 11/20/2013,51.65.104.7400

(In reply to Mike Cao from comment #2)
> The official netkvm driver support for winxp is virtio-win-prewhql-49 ,Pls
> retest it w/ virtio-win-prewhql-49 or latest virtio-win package.
> 
> Mike

Hi,

Can you upload to somewhere kernel memory dump (please archive it before uploading).

Thanks,
Yan.

Created attachment 882249 [details]
memory dump

minidump of windows xp

(In reply to fanlulin from comment #5)
> Created attachment 882249 [details]
> memory dump
> 
> minidump of windows xp

Looks like the file was not uploaded correctly.

Created attachment 882491 [details]
memory dump2

xp dump again.
------------------
i didn't notice the first one was broken.

The crash is in function ParaNdis_PowerOff(PARANDIS_ADAPTER *pContext) when we try to shutdown the control queue. It looks like the control queue was not initialised to begin with.

Hi,

Can you try this engineering build and see if solves the problem?

https://www.dropbox.com/s/wkfamji1tt6u6wv/XP.zip

Thanks,
Yan.

Hi
I have tried this build, and confirm this issue is fixed..
But I tried 4 times, only once I got a BSOD when  resuming from hibernation, so I am not sure this issue is from the new driver.

And I noticed that is speed of NIC is 1.4Gbps after install the new driver.

0: kd> !analyze -v
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

KERNEL_STACK_INPAGE_ERROR (77)
The requested page of kernel data could not be read in.  Caused by
bad block in paging file or disk controller error.
In the case when the first arguments is 0 or 1, the stack signature
in the kernel stack was not found.  Again, bad hardware.
An I/O status of c000009c (STATUS_DEVICE_DATA_ERROR) or
C000016AL (STATUS_DISK_OPERATION_FAILED)  normally indicates
the data could not be read from the disk due to a bad
block.  Upon reboot autocheck will run and attempt to map out the bad
sector.  If the status is C0000185 (STATUS_IO_DEVICE_ERROR) and the paging
file is on a SCSI disk device, then the cabling and termination should be
checked.  See the knowledge base article on SCSI termination.
Arguments:
Arg1: c000000e, status code
Arg2: c000000e, i/o status code
Arg3: 00000000, page file number
Arg4: 007c2000, offset into page file

Debugging Details:
------------------


ERROR_CODE: (NTSTATUS) 0xc000000e - <Unable to get error code text>

DISK_HARDWARE_ERROR: There was error with disk hardware

BUGCHECK_STR:  0x77_c000000e

DEFAULT_BUCKET_ID:  DRIVER_FAULT

PROCESS_NAME:  System

LAST_CONTROL_TRANSFER:  from 80512f79 to 804faf33

STACK_TEXT:  
bad23cdc 80512f79 00000077 c000000e c000000e nt!KeBugCheckEx+0x1b
bad23d50 80513deb c0588610 000169a1 00000001 nt!MiMakeOutswappedPageResident+0x4f5
bad23d8c 80540d76 00789a90 00000000 89a28b30 nt!MmInPageKernelStack+0x149
bad23da4 80541246 896bbe08 805d0f64 00000000 nt!KiInSwapKernelStacks+0x16
bad23dac 805d0f64 00000000 00000000 00000000 nt!KeSwapProcessOrStack+0x7c
bad23ddc 805470de 805411ca 00000000 00000000 nt!PspSystemThreadStartup+0x34
00000000 00000000 00000000 00000000 00000000 nt!KiThreadStartup+0x16


STACK_COMMAND:  kb

FOLLOWUP_IP: 
nt!MiMakeOutswappedPageResident+4f5
80512f79 cc              int     3

SYMBOL_STACK_INDEX:  1

SYMBOL_NAME:  nt!MiMakeOutswappedPageResident+4f5

FOLLOWUP_NAME:  MachineOwner

MODULE_NAME: nt

DEBUG_FLR_IMAGE_TIMESTAMP:  4802516a

IMAGE_NAME:  memory_corruption

FAILURE_BUCKET_ID:  0x77_c000000e_nt!MiMakeOutswappedPageResident+4f5

BUCKET_ID:  0x77_c000000e_nt!MiMakeOutswappedPageResident+4f5

Followup: MachineOwner
---------



(In reply to Yan Vugenfirer from comment #10)
> Hi,
> 
> Can you try this engineering build and see if solves the problem?
> 
> https://www.dropbox.com/s/wkfamji1tt6u6wv/XP.zip
> 
> Thanks,
> Yan.

Created attachment 883819 [details]
BSOD when resume from hibernate

memory dump of BSOD when resume from hibernate

Related bug: BZ #957507

(In reply to Yan Vugenfirer from comment #13)
> Related bug: BZ #957507

Hi,Yan
Do you plan to add the patch to prewhql build ?

Thanks,
Mike

1. The new crash dump is not related to network driver. It is worth to open a new bug (could be some storage controller issue).


2. You are saying the speed now is 1.4Gbps. What was it before? Are you talking about actual NIC throughput or number you see in the GUI?

Thanks,
Yan.

1. I will do more test to confirm it.

2. only Windows GUI shows the speed is 1.4Gbps, before is 1Gbps... I didn't test the actual NIC throughput.

Thanks
Fan

(In reply to Yan Vugenfirer from comment #15)
> 1. The new crash dump is not related to network driver. It is worth to open
> a new bug (could be some storage controller issue).
> 
> 
> 2. You are saying the speed now is 1.4Gbps. What was it before? Are you
> talking about actual NIC throughput or number you see in the GUI?
> 
> Thanks,
> Yan.

lijin, pls help to verify this bug

Reproduced this issue on virtio-win-prewhl-74
Verified this issue on virtio-win-prewhl-87

kernel-2.6.32-491.el6.x86_64
qemu-kvm: 1.4.2

Steps:
1.boot winxp guest with:
qemu-system-x86_64 -drive file=winxp.qcow2,if=none,cache=unsafe,media=disk,format=qcow2,id=drive-ide0-0-1 -device ide-drive,id=ide1,drive=drive-ide0-0-1,bus=ide.1 -monitor stdio -usb -device usb-tablet -boot menu=on -chardev file,path=/root/console.log,id=serial1 -device isa-serial,chardev=serial1,id=s1 -cpu Nehalem,hv_relaxed -smp 2 -m 2G -enable-kvm -global PIIX4_PM.disable_s3=0 -global PIIX4_PM.disable_s4=0 -netdev tap,script=/etc/qemu-ifup,id=tap1 -device virtio-net-pci,netdev=tap1,id=net1 -cdrom virtio-win-0.1-74.iso -vga cirrus -vnc :2
2.do s4 in guest

Actual Results:
on virtio-win-prewhl-74, guest bsod with "7e" code
on  virtio-win-prewhl-87,guest can hibernate and resume correctly.

Based on above ,this issue has been fixed already .

Need to highlight that we can *not* reproduce this issue on RHEL6 qemu-kvm .
We use upstream v1.4.2 for bug verification

Since RHEL users will not affected ,closing this bug

winxp can s4 correctly on rhel7.1 host.

package info:
qemu-kvm-rhev-2.1.2-5.el7.x86_64
kernel-3.10.0-196.el7.x86_64
virtio-win-1.7.2-1.el7
seabios-1.7.5-5.el7.x86_64