public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
* [edk2-devel] Patch OvmfPkg: Use heap memory for virtio-scsi request causing regression
@ 2024-08-01 11:28 Aithal, Srikanth via groups.io
  2024-08-01 15:53 ` Sami Mujawar
  0 siblings, 1 reply; 4+ messages in thread
From: Aithal, Srikanth via groups.io @ 2024-08-01 11:28 UTC (permalink / raw)
  To: devel, sami.mujawar, Ard Biesheuvel, Jiewen Yao, Gerd Hoffmann

[-- Attachment #1: Type: text/plain, Size: 1921 bytes --]

Hello all,

After the patches [1fc55a39, b342070c] OvmfPkg: Use heap memory for 
virtio-blk request/OvmfPkg: Use heap memory for virtio-scsi request got 
merged, this regression is spotted.

AMD SVM/SEV/SEVES guest boot fails with OVMF packages [Ia3264/X64] built 
from today's upstream edk2 repository[Head: 85fad99].

Guest boot with OVMF.fd or only with OVMF_CODE.fd gets stuck at below point:

/Divisor   = 1
BaudRate/Actual (115200/115200) = 100%
PciSioSerial: Create SIO child serial device - Device Error
PlatformBootManagerAfterConsole
Found Mass Storage device: PciRoot(0x0)/Pci(0x3,0x0)
InstallProtocolInterface: FA920010-6785-4941-B6EC-498C579F160A 3DEB82A0
InstallProtocolInterface: 143B7632-B81B-4CB7-ABD3-B625A5B9BFFE 3DE10AA0
InstallProtocolInterface: 0167CCC4-D0F7-4F21-A3EF-9E64B7CDCE8B 3DE5D4A0
qemu-system-x86_64: wrong size for virtio-scsi headers <========== gets 
hung/

If both OVMF code and vars file are provided the issue is not hit.

Qemu command line to recreate the issue:

/qemu-system-x86_64 \
-machine q35,kernel_irqchip=split \
-name guest=vm,debug-threads=on \
-drive if=pflash,format=raw,unit=0,file=OVMF_CODE.fd,readonly \
-cpu host \
-m 1024 \
-smp 1,cores=1,threads=1,dies=1,sockets=1 \
-kernel bzImage \
-append "root=/dev/sda rw console=ttyS0 net.ifnames=0 biosdevname=0 
movable_node swiotlb=65536 " \
-drive id=disk0,file=22.04-server.qcow2,format=qcow2,if=none \
-device virtio-scsi-pci,id=scsi0,disable-legacy=on,iommu_platform=true \
-device scsi-hd,drive=disk0 \
--enable-kvm \
--nographic
/

Thanks,

Srikanth Aithal


-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#120171): https://edk2.groups.io/g/devel/message/120171
Mute This Topic: https://groups.io/mt/107663187/7686176
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [rebecca@openfw.io]
-=-=-=-=-=-=-=-=-=-=-=-



[-- Attachment #2: Type: text/html, Size: 2949 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [edk2-devel] Patch OvmfPkg: Use heap memory for virtio-scsi request causing regression
  2024-08-01 11:28 [edk2-devel] Patch OvmfPkg: Use heap memory for virtio-scsi request causing regression Aithal, Srikanth via groups.io
@ 2024-08-01 15:53 ` Sami Mujawar
  2024-08-02  3:56   ` Sami Mujawar
  0 siblings, 1 reply; 4+ messages in thread
From: Sami Mujawar @ 2024-08-01 15:53 UTC (permalink / raw)
  To: devel@edk2.groups.io, srikanth.aithal@amd.com, Ard Biesheuvel,
	Jiewen Yao, Gerd Hoffmann

[-- Attachment #1: Type: text/plain, Size: 3124 bytes --]

Hi Srikanth,

Thanks for reporting this issue. It is strange that the virtio-blk driver is working with a similar change.
I am looking into why this may be failing.

Regards,

Sami Mujawar
From: <devel@edk2.groups.io> on behalf of "Aithal, Srikanth via groups.io" <srikanth.aithal=amd.com@groups.io>
Reply to: "devel@edk2.groups.io" <devel@edk2.groups.io>, "srikanth.aithal@amd.com" <srikanth.aithal@amd.com>
Date: Thursday 1 August 2024 at 12:28
To: "devel@edk2.groups.io" <devel@edk2.groups.io>, Sami Mujawar <Sami.Mujawar@arm.com>, Ard Biesheuvel <ardb+tianocore@kernel.org>, Jiewen Yao <jiewen.yao@intel.com>, Gerd Hoffmann <kraxel@redhat.com>
Subject: [edk2-devel] Patch OvmfPkg: Use heap memory for virtio-scsi request causing regression
Resent from: <srikanth.aithal@amd.com>
Resent date: Thursday 1 August 2024 at 12:28


Hello all,

After the patches [1fc55a39, b342070c] OvmfPkg: Use heap memory for virtio-blk request/OvmfPkg: Use heap memory for virtio-scsi request got merged, this regression is spotted.

AMD SVM/SEV/SEVES guest boot fails with OVMF packages [Ia3264/X64] built from today's upstream edk2 repository[Head: 85fad99].

Guest boot with OVMF.fd or only with OVMF_CODE.fd gets stuck at below point:

Divisor   = 1
BaudRate/Actual (115200/115200) = 100%
PciSioSerial: Create SIO child serial device - Device Error
PlatformBootManagerAfterConsole
Found Mass Storage device: PciRoot(0x0)/Pci(0x3,0x0)
InstallProtocolInterface: FA920010-6785-4941-B6EC-498C579F160A 3DEB82A0
InstallProtocolInterface: 143B7632-B81B-4CB7-ABD3-B625A5B9BFFE 3DE10AA0
InstallProtocolInterface: 0167CCC4-D0F7-4F21-A3EF-9E64B7CDCE8B 3DE5D4A0
qemu-system-x86_64: wrong size for virtio-scsi headers   <========== gets hung

If both OVMF code and vars file are provided the issue is not hit.
Qemu command line to recreate the issue:

qemu-system-x86_64 \
-machine q35,kernel_irqchip=split \
-name guest=vm,debug-threads=on \
-drive if=pflash,format=raw,unit=0,file=OVMF_CODE.fd,readonly \
-cpu host \
-m 1024 \
-smp 1,cores=1,threads=1,dies=1,sockets=1 \
-kernel bzImage \
-append "root=/dev/sda rw console=ttyS0 net.ifnames=0 biosdevname=0 movable_node swiotlb=65536 " \
-drive id=disk0,file=22.04-server.qcow2,format=qcow2,if=none \
-device virtio-scsi-pci,id=scsi0,disable-legacy=on,iommu_platform=true \
-device scsi-hd,drive=disk0 \
--enable-kvm \
--nographic

Thanks,

Srikanth Aithal

IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.


-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#120191): https://edk2.groups.io/g/devel/message/120191
Mute This Topic: https://groups.io/mt/107663187/7686176
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [rebecca@openfw.io]
-=-=-=-=-=-=-=-=-=-=-=-



[-- Attachment #2: Type: text/html, Size: 7350 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [edk2-devel] Patch OvmfPkg: Use heap memory for virtio-scsi request causing regression
  2024-08-01 15:53 ` Sami Mujawar
@ 2024-08-02  3:56   ` Sami Mujawar
  2024-08-02  5:22     ` Aithal, Srikanth via groups.io
  0 siblings, 1 reply; 4+ messages in thread
From: Sami Mujawar @ 2024-08-02  3:56 UTC (permalink / raw)
  To: devel@edk2.groups.io, srikanth.aithal@amd.com, Ard Biesheuvel,
	Jiewen Yao, Gerd Hoffmann

[-- Attachment #1: Type: text/plain, Size: 3860 bytes --]

Hi Shrikanth,

It appears I had missed out passing the correct virtio-scsi request size when enqueuing the request.
I have now fixed that in the pull request at https://github.com/tianocore/edk2/pull/6038.
Can you try the patch in the pull request at and let me know if it fixes the issue, please?

Regards,

Sami Mujawar

From: Sami Mujawar <Sami.Mujawar@arm.com>
Date: Thursday 1 August 2024 at 16:53
To: "devel@edk2.groups.io" <devel@edk2.groups.io>, "srikanth.aithal@amd.com" <srikanth.aithal@amd.com>, Ard Biesheuvel <ardb+tianocore@kernel.org>, Jiewen Yao <jiewen.yao@intel.com>, Gerd Hoffmann <kraxel@redhat.com>
Subject: Re: [edk2-devel] Patch OvmfPkg: Use heap memory for virtio-scsi request causing regression

Hi Srikanth,

Thanks for reporting this issue. It is strange that the virtio-blk driver is working with a similar change.
I am looking into why this may be failing.

Regards,

Sami Mujawar
From: <devel@edk2.groups.io> on behalf of "Aithal, Srikanth via groups.io" <srikanth.aithal=amd.com@groups.io>
Reply to: "devel@edk2.groups.io" <devel@edk2.groups.io>, "srikanth.aithal@amd.com" <srikanth.aithal@amd.com>
Date: Thursday 1 August 2024 at 12:28
To: "devel@edk2.groups.io" <devel@edk2.groups.io>, Sami Mujawar <Sami.Mujawar@arm.com>, Ard Biesheuvel <ardb+tianocore@kernel.org>, Jiewen Yao <jiewen.yao@intel.com>, Gerd Hoffmann <kraxel@redhat.com>
Subject: [edk2-devel] Patch OvmfPkg: Use heap memory for virtio-scsi request causing regression
Resent from: <srikanth.aithal@amd.com>
Resent date: Thursday 1 August 2024 at 12:28


Hello all,

After the patches [1fc55a39, b342070c] OvmfPkg: Use heap memory for virtio-blk request/OvmfPkg: Use heap memory for virtio-scsi request got merged, this regression is spotted.

AMD SVM/SEV/SEVES guest boot fails with OVMF packages [Ia3264/X64] built from today's upstream edk2 repository[Head: 85fad99].

Guest boot with OVMF.fd or only with OVMF_CODE.fd gets stuck at below point:

Divisor   = 1
BaudRate/Actual (115200/115200) = 100%
PciSioSerial: Create SIO child serial device - Device Error
PlatformBootManagerAfterConsole
Found Mass Storage device: PciRoot(0x0)/Pci(0x3,0x0)
InstallProtocolInterface: FA920010-6785-4941-B6EC-498C579F160A 3DEB82A0
InstallProtocolInterface: 143B7632-B81B-4CB7-ABD3-B625A5B9BFFE 3DE10AA0
InstallProtocolInterface: 0167CCC4-D0F7-4F21-A3EF-9E64B7CDCE8B 3DE5D4A0
qemu-system-x86_64: wrong size for virtio-scsi headers   <========== gets hung

If both OVMF code and vars file are provided the issue is not hit.
Qemu command line to recreate the issue:

qemu-system-x86_64 \
-machine q35,kernel_irqchip=split \
-name guest=vm,debug-threads=on \
-drive if=pflash,format=raw,unit=0,file=OVMF_CODE.fd,readonly \
-cpu host \
-m 1024 \
-smp 1,cores=1,threads=1,dies=1,sockets=1 \
-kernel bzImage \
-append "root=/dev/sda rw console=ttyS0 net.ifnames=0 biosdevname=0 movable_node swiotlb=65536 " \
-drive id=disk0,file=22.04-server.qcow2,format=qcow2,if=none \
-device virtio-scsi-pci,id=scsi0,disable-legacy=on,iommu_platform=true \
-device scsi-hd,drive=disk0 \
--enable-kvm \
--nographic

Thanks,

Srikanth Aithal

IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.


-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#120202): https://edk2.groups.io/g/devel/message/120202
Mute This Topic: https://groups.io/mt/107663187/7686176
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [rebecca@openfw.io]
-=-=-=-=-=-=-=-=-=-=-=-



[-- Attachment #2: Type: text/html, Size: 10043 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [edk2-devel] Patch OvmfPkg: Use heap memory for virtio-scsi request causing regression
  2024-08-02  3:56   ` Sami Mujawar
@ 2024-08-02  5:22     ` Aithal, Srikanth via groups.io
  0 siblings, 0 replies; 4+ messages in thread
From: Aithal, Srikanth via groups.io @ 2024-08-02  5:22 UTC (permalink / raw)
  To: Sami Mujawar, devel@edk2.groups.io, srikanth.aithal@amd.com,
	Ard Biesheuvel, Jiewen Yao, Gerd Hoffmann

[-- Attachment #1: Type: text/plain, Size: 4213 bytes --]


On 8/2/2024 9:26 AM, Sami Mujawar wrote:
>
> Hi Shrikanth,
>
> It appears I had missed out passing the correct virtio-scsi request 
> size when enqueuing the request.
>
> I have now fixed that in the pull request at 
> https://github.com/tianocore/edk2/pull/6038.
>
> Can you try the patch in the pull request at and let me know if it 
> fixes the issue, please?
>
Yes, this pull request fixes the issue. Thank you.

Reported-by: Aithal Srikanth <sraithal@amd.com>

Tested-by: Aithal Srikanth <sraithal@amd.com>

> Regards,
>
> Sami Mujawar
>
> *From: *Sami Mujawar <Sami.Mujawar@arm.com>
> *Date: *Thursday 1 August 2024 at 16:53
> *To: *"devel@edk2.groups.io" <devel@edk2.groups.io>, 
> "srikanth.aithal@amd.com" <srikanth.aithal@amd.com>, Ard Biesheuvel 
> <ardb+tianocore@kernel.org>, Jiewen Yao <jiewen.yao@intel.com>, Gerd 
> Hoffmann <kraxel@redhat.com>
> *Subject: *Re: [edk2-devel] Patch OvmfPkg: Use heap memory for 
> virtio-scsi request causing regression
>
> Hi Srikanth,
>
> Thanks for reporting this issue. It is strange that the virtio-blk 
> driver is working with a similar change.
>
> I am looking into why this may be failing.
>
> Regards,
>
> Sami Mujawar
>
> *From: *<devel@edk2.groups.io> on behalf of "Aithal, Srikanth via 
> groups.io" <srikanth.aithal=amd.com@groups.io>
> *Reply to: *"devel@edk2.groups.io" <devel@edk2.groups.io>, 
> "srikanth.aithal@amd.com" <srikanth.aithal@amd.com>
> *Date: *Thursday 1 August 2024 at 12:28
> *To: *"devel@edk2.groups.io" <devel@edk2.groups.io>, Sami Mujawar 
> <Sami.Mujawar@arm.com>, Ard Biesheuvel <ardb+tianocore@kernel.org>, 
> Jiewen Yao <jiewen.yao@intel.com>, Gerd Hoffmann <kraxel@redhat.com>
> *Subject: *[edk2-devel] Patch OvmfPkg: Use heap memory for virtio-scsi 
> request causing regression
> *Resent from: *<srikanth.aithal@amd.com>
> *Resent date: *Thursday 1 August 2024 at 12:28
>
> Hello all,
>
> After the patches [1fc55a39, b342070c] OvmfPkg: Use heap memory for 
> virtio-blk request/OvmfPkg: Use heap memory for virtio-scsi request 
> got merged, this regression is spotted.
>
> AMD SVM/SEV/SEVES guest boot fails with OVMF packages [Ia3264/X64] 
> built from today's upstream edk2 repository[Head: 85fad99].
>
> Guest boot with OVMF.fd or only with OVMF_CODE.fd gets stuck at below 
> point:
>
> /Divisor   = 1
> BaudRate/Actual (115200/115200) = 100%
> PciSioSerial: Create SIO child serial device - Device Error
> PlatformBootManagerAfterConsole
> Found Mass Storage device: PciRoot(0x0)/Pci(0x3,0x0)
> InstallProtocolInterface: FA920010-6785-4941-B6EC-498C579F160A 3DEB82A0
> InstallProtocolInterface: 143B7632-B81B-4CB7-ABD3-B625A5B9BFFE 3DE10AA0
> InstallProtocolInterface: 0167CCC4-D0F7-4F21-A3EF-9E64B7CDCE8B 3DE5D4A0
> qemu-system-x86_64: wrong size for virtio-scsi headers <========== 
> gets hung/
>
> If both OVMF code and vars file are provided the issue is not hit.
>
> Qemu command line to recreate the issue:
>
> /qemu-system-x86_64 \
> -machine q35,kernel_irqchip=split \
> -name guest=vm,debug-threads=on \
> -drive if=pflash,format=raw,unit=0,file=OVMF_CODE.fd,readonly \
> -cpu host \
> -m 1024 \
> -smp 1,cores=1,threads=1,dies=1,sockets=1 \
> -kernel bzImage \
> -append "root=/dev/sda rw console=ttyS0 net.ifnames=0 biosdevname=0 
> movable_node swiotlb=65536 " \
> -drive id=disk0,file=22.04-server.qcow2,format=qcow2,if=none \
> -device virtio-scsi-pci,id=scsi0,disable-legacy=on,iommu_platform=true \
> -device scsi-hd,drive=disk0 \
> --enable-kvm \
> --nographic/
>
> Thanks,
>
> Srikanth Aithal
>
> 
>
> IMPORTANT NOTICE: The contents of this email and any attachments are 
> confidential and may also be privileged. If you are not the intended 
> recipient, please notify the sender immediately and do not disclose 
> the contents to any other person, use it for any purpose, or store or 
> copy the information in any medium. Thank you. 


-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#120204): https://edk2.groups.io/g/devel/message/120204
Mute This Topic: https://groups.io/mt/107663187/7686176
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [rebecca@openfw.io]
-=-=-=-=-=-=-=-=-=-=-=-



[-- Attachment #2: Type: text/html, Size: 13194 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2024-08-02  5:23 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-08-01 11:28 [edk2-devel] Patch OvmfPkg: Use heap memory for virtio-scsi request causing regression Aithal, Srikanth via groups.io
2024-08-01 15:53 ` Sami Mujawar
2024-08-02  3:56   ` Sami Mujawar
2024-08-02  5:22     ` Aithal, Srikanth via groups.io

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox