Hi,
So an update on a potential demo platform for the AGL demo.
Platform:
MACCHIATObin 1.2 board + manually applied PCIe reset fix + removed JTAG & hacked PCIe + NVIDIA GK208B (originally from SynQaucer) + EFI firmware on sdcard from Steve McIntyre
I have a fairly stable box running Debian Buster + backported kernel + testing Grub (for Xen) with managed graphics (via nouveau) and working sound (via a USB audio dongle). With the following caveats:
I use the DT system info instead of ACPI.
I keep hitting ESC on serial at boot prompt and selecting "Reset" until the graphics card wakes up and displays EFI splash screen on the monitor.
and
Only kernel 5.6.0-0.bpo.2-arm64 #1 SMP Debian 5.6.14-2~bpo10+1 (2020-06-09)
While I can get a newer kernel to build and boot up the graphics either comes up corrupted or keeps crashing the X server with a kernel related nouvaue bug. I've tried to narrow it down but both vanilla 5.6.0 and 5.6.14 have the same problem so I suspect there is a patch in the Debian series which fixes it which might need up-streaming.
(aside: anyone know how to make a traditional bisectable git tree from a series of Debian patches?)
KVM works and I'm able to boot up my test kernel and get to a userspace prompt on the serial console. I still have to boot up with virtio-gpu and prove that works.
I've been unable to get Xen up and running on it so far. I've tested with vanilla and also the current testing tree that I finally got working on QEMU TCG. I have gotten some Xen diagnostics which I've dumped into a separate thread so as not to confuse things too much.
Subject: Current status of Xen on MachiatoBin Date: Fri, 02 Oct 2020 17:00:43 +0100 Message-ID: 87lfgomxck.fsf@linaro.org
Hi Alex, could you share tfa and uboot versions ?
Le ven. 2 oct. 2020 à 18:05, Alex Bennée alex.bennee@linaro.org a écrit :
Hi,
So an update on a potential demo platform for the AGL demo.
Platform:
MACCHIATObin 1.2 board
+ manually applied PCIe reset fix + removed JTAG & hacked PCIe + NVIDIA GK208B (originally from SynQaucer) + EFI firmware on sdcard from Steve McIntyreI have a fairly stable box running Debian Buster + backported kernel +
testing Grub (for Xen) with managed graphics (via nouveau) and working
sound (via a USB audio dongle). With the following caveats:
I use the DT system info instead of ACPI.
I keep hitting ESC on serial at boot prompt and selecting "Reset"
until the graphics card wakes up and displays EFI splash screen on the
monitor.
and
Only kernel 5.6.0-0.bpo.2-arm64 #1 SMP Debian 5.6.14-2~bpo10+1 (2020-06-09)
While I can get a newer kernel to build and boot up the graphics either
comes up corrupted or keeps crashing the X server with a kernel related
nouvaue bug. I've tried to narrow it down but both vanilla 5.6.0 and
5.6.14 have the same problem so I suspect there is a patch in the Debian
series which fixes it which might need up-streaming.
(aside: anyone know how to make a traditional bisectable git tree from
a series of Debian patches?)
KVM works and I'm able to boot up my test kernel and get to a userspace
prompt on the serial console. I still have to boot up with virtio-gpu
and prove that works.
I've been unable to get Xen up and running on it so far. I've tested
with vanilla and also the current testing tree that I finally got
working on QEMU TCG. I have gotten some Xen diagnostics which I've
dumped into a separate thread so as not to confuse things too much.
Subject: Current status of Xen on MachiatoBin
Date: Fri, 02 Oct 2020 17:00:43 +0100
Message-ID: 87lfgomxck.fsf@linaro.org
--
Alex Bennée
--
François-Frédéric Ozog | *Director Linaro Edge & Fog Computing Group* T: +33.67221.6485 francois.ozog@linaro.org | Skype: ffozog
François Ozog francois.ozog@linaro.org writes:
Hi Alex, could you share tfa and uboot versions ?
From the serial port I get:
BootROM - 2.03 Starting CP-0 IOROM 1.07 Booting from SD 0 (0x29) SD - wait_for_sd_interrupt: Error interrupt - 00008000 SD - wait_for_sd_interrupt: Error interrupt status 00000003 SD - sd_get_cmd_response: Get command response failed.
and on the splash:
Armada 8040 MacchiatoBin Cortex-A72 1.30 GHz MARVELL UEFI 18.09.2 16364 MB RAM
Is there anyway to interrogate the versions from userspace?
Hello,
I would like to update my status of Xen on DeveloperBox.
Platform: DeveloperBox (96boards EE) + 8GB DDR + Nvidia GeForce GT 710B + Samsung NVMe 512GB
As you may know the DeveloperBox boots with the EDK2 with ACPI or DT (selectable).
Xen Boot EDK2+ACPI (SBBR): Booted EDK2+DT (EBBR): Booted with the upstream GRUB (debian 10 package has a bug)
Dom0 Boot EDK2+ACPI: Booted (EFI detected), but netsec ethernet driver doesn't work. Graphics (EFI FB) doesn't work too. EDK2+DT: Booted (no EFI), netsec works. Graphics (EFI FB) doesn't work.
I think EDK2+DT will be enough for the Demo and Socionext needs because it is EBBR compliant.
Since a known PCIe bus errata of SynQuacer, we can not use the nouveau driver without modification. I applied an old patch but it didn't work. So the options are - Fix nouveau (but anyway that should be out-of-tree) - Enable EFI FB support on Dom0 with DT. - Investigate firmware mitigation and do similar things in Xen (making a shadow page table for PCIe MMIO region?) - Search for a USB-HDMI(or VGA) dongle which can work with Linux.
Thank you,
2020年10月3日(土) 1:05 Alex Bennée alex.bennee@linaro.org:
Hi,
So an update on a potential demo platform for the AGL demo.
Platform:
MACCHIATObin 1.2 board + manually applied PCIe reset fix + removed JTAG & hacked PCIe + NVIDIA GK208B (originally from SynQaucer) + EFI firmware on sdcard from Steve McIntyre
I have a fairly stable box running Debian Buster + backported kernel + testing Grub (for Xen) with managed graphics (via nouveau) and working sound (via a USB audio dongle). With the following caveats:
I use the DT system info instead of ACPI.
I keep hitting ESC on serial at boot prompt and selecting "Reset" until the graphics card wakes up and displays EFI splash screen on the monitor.
and
Only kernel 5.6.0-0.bpo.2-arm64 #1 SMP Debian 5.6.14-2~bpo10+1 (2020-06-09)
While I can get a newer kernel to build and boot up the graphics either comes up corrupted or keeps crashing the X server with a kernel related nouvaue bug. I've tried to narrow it down but both vanilla 5.6.0 and 5.6.14 have the same problem so I suspect there is a patch in the Debian series which fixes it which might need up-streaming.
(aside: anyone know how to make a traditional bisectable git tree from a series of Debian patches?)
KVM works and I'm able to boot up my test kernel and get to a userspace prompt on the serial console. I still have to boot up with virtio-gpu and prove that works.
I've been unable to get Xen up and running on it so far. I've tested with vanilla and also the current testing tree that I finally got working on QEMU TCG. I have gotten some Xen diagnostics which I've dumped into a separate thread so as not to confuse things too much.
Subject: Current status of Xen on MachiatoBin Date: Fri, 02 Oct 2020 17:00:43 +0100 Message-ID: 87lfgomxck.fsf@linaro.org
-- Alex Bennée
Masami Hiramatsu masami.hiramatsu@linaro.org writes:
Hello,
I would like to update my status of Xen on DeveloperBox.
Platform: DeveloperBox (96boards EE)
- 8GB DDR
- Nvidia GeForce GT 710B
- Samsung NVMe 512GB
As you may know the DeveloperBox boots with the EDK2 with ACPI or DT (selectable).
Xen Boot EDK2+ACPI (SBBR): Booted EDK2+DT (EBBR): Booted with the upstream GRUB (debian 10 package has a bug)
Dom0 Boot EDK2+ACPI: Booted (EFI detected), but netsec ethernet driver doesn't work. Graphics (EFI FB) doesn't work too. EDK2+DT: Booted (no EFI), netsec works. Graphics (EFI FB) doesn't work.
I think EDK2+DT will be enough for the Demo and Socionext needs because it is EBBR compliant.
Since a known PCIe bus errata of SynQuacer, we can not use the nouveau driver without modification. I applied an old patch but it didn't work. So the options are
- Fix nouveau (but anyway that should be out-of-tree)
I messed around a lot with the nouveau driver when I first got my Developer Box but it was ultimately frustrating because the GPU drivers are quite big and complex beasts for those not familiar with them. In the end I went headless and was happier with it.
- Enable EFI FB support on Dom0 with DT.
I think this sounds like the easiest to get going.
- Investigate firmware mitigation and do similar things in Xen
(making a shadow page table for PCIe MMIO region?)
AIUI the previous mitigation relied on using EL2 to trap wide accesses and work around the bus limitations. I wasn't able to use it because I needed to boot Linux in EL2 to debug KVM.
- Search for a USB-HDMI(or VGA) dongle which can work with Linux.
I was pleasantly surprised that the random USB dongle/expander with audio worked out of the box on the MB however I don't know of any equivalent for video. Certainly things like the DisplayLink expanders are totally proprietary.
Is the USB a full USB3? Could we use an external PCI GPU enclosure or does that require Thunderbolt as well? As far I understand it docking stations still need the internal GPU to render stuff before pushing the data out over HDMI.
Thank you,
2020年10月3日(土) 1:05 Alex Bennée alex.bennee@linaro.org:
Hi,
So an update on a potential demo platform for the AGL demo.
Platform:
MACCHIATObin 1.2 board + manually applied PCIe reset fix + removed JTAG & hacked PCIe + NVIDIA GK208B (originally from SynQaucer) + EFI firmware on sdcard from Steve McIntyre
I have a fairly stable box running Debian Buster + backported kernel + testing Grub (for Xen) with managed graphics (via nouveau) and working sound (via a USB audio dongle). With the following caveats:
I use the DT system info instead of ACPI.
I keep hitting ESC on serial at boot prompt and selecting "Reset" until the graphics card wakes up and displays EFI splash screen on the monitor.
and
Only kernel 5.6.0-0.bpo.2-arm64 #1 SMP Debian 5.6.14-2~bpo10+1 (2020-06-09)
While I can get a newer kernel to build and boot up the graphics either comes up corrupted or keeps crashing the X server with a kernel related nouvaue bug. I've tried to narrow it down but both vanilla 5.6.0 and 5.6.14 have the same problem so I suspect there is a patch in the Debian series which fixes it which might need up-streaming.
(aside: anyone know how to make a traditional bisectable git tree from a series of Debian patches?)
KVM works and I'm able to boot up my test kernel and get to a userspace prompt on the serial console. I still have to boot up with virtio-gpu and prove that works.
I've been unable to get Xen up and running on it so far. I've tested with vanilla and also the current testing tree that I finally got working on QEMU TCG. I have gotten some Xen diagnostics which I've dumped into a separate thread so as not to confuse things too much.
Subject: Current status of Xen on MachiatoBin Date: Fri, 02 Oct 2020 17:00:43 +0100 Message-ID: 87lfgomxck.fsf@linaro.org
-- Alex Bennée
stratos-dev@op-lists.linaro.org
-
Alex Bennée -
François Ozog -
Masami Hiramatsu