https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
Mark Linimon changed:
What|Removed |Added
Keywords|needs-qa|
Assignee|virtualization@
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #34 from commit-h...@freebsd.org ---
A commit in branch main references this bug:
URL:
https://cgit.FreeBSD.org/src/commit/?id=927358dd98cb902160093e0dc0bac002d6b43858
commit 927358dd98cb902160093e0dc0bac002d6b43858
Author:
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #33 from commit-h...@freebsd.org ---
A commit in branch main references this bug:
URL:
https://cgit.FreeBSD.org/src/commit/?id=8ea7fa16d9fe5acb7d42a223efbfa23627aa5e0c
commit 8ea7fa16d9fe5acb7d42a223efbfa23627aa5e0c
Author:
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #32 from Wei Hu ---
(In reply to Brad Davis from comment #31)
This problem only happens on Gen2 VMs in Azure. On-prem Hyper-V don't have this
hang issue. Unfortunately I don't have too much experience debugging the
FreeBSD seria
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
Brad Davis changed:
What|Removed |Added
CC||b...@freebsd.org
--- Comment #31 from
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #30 from Michael Baptist ---
Wei Hu, thanks for the update.
Commenting out the printfs does "fix" the issue for now.
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #29 from Wei Hu ---
We have narrowed down the problem area to serial console output in FreeBSD and
Azure COM emulator. Seems these two don't work very well at the early boot
time. Maybe some synchronizing event was missing which
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #28 from Michael Baptist ---
Tested 12.3 and HEAD and saw similar result.
Skus tested:
Standard D8 V3 - booted once failed once
Standard D8 V4 - failed to boot
Standard D8 V5 - booted up fine
Boot issue presents in the same p
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #27 from Michael Baptist ---
(In reply to Wei Hu from comment #26)
Yes I will try that.
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #26 from Wei Hu ---
Michael, can you try 12.3 Gen2 VM on the same Azure VM size? I would like to
know if it was something new in 13 which causes this problem.
--
You are receiving this mail because:
You are the assignee for th
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #25 from Michael Baptist ---
Made it a bit further
Also I am at: 8020c05683f266c5513cfe8ad3e58e7e06f85ffa
Does't seem terribly far behind.
$git diff --stat HEAD origin/main -- stand/
stand/common/isapnp.c| 2 +-
sta
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #24 from Michael Baptist ---
Is the EFI console the same console after the tramplone passes execution to the
kernel?
For instance the code in hammer_time/mi_startup is a different console correct?
Not sure its EFI console rela
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #23 from Wei Hu ---
Do you have evidence that it crashed? Or just the output stalled at different
points during boot?
The reason I am asking is the EFI console seemed to have changed on 13.x. I
have suspicion maybe the new con
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #22 from Michael Baptist ---
OK set module_verbose=3
OK memmap
Type Physical Virtual #Pages Attr
ConventionalMemory 00a0 UC WC WT WB
BootServicesData 00
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #21 from Michael Baptist ---
Also forgot an important step Azure doesn't have vidconsole for user input so
you'll need to mount the memory device root parition from the img file and
update loader.conf
boot_multicons="YES"
boot_
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #20 from Michael Baptist ---
(In reply to Mark Johnston from comment #16)
Yep I'll try a boot verbose as well.
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #19 from Michael Baptist ---
Also I build on FreeBSD then scp the vhd to a linux host for upload, since
FreeBSD support for Azure CLI is behind and some of the patches are out of date
for the copy.
--
You are receiving this ma
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #18 from Michael Baptist ---
(In reply to Mark Johnston from comment #17)
Not sure the instructions that show the steps for Gen 1 that you were
referencing, but the main difference when using the Azure CLI is specifying
--hyper
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #17 from Mark Johnston ---
Is there some written set of steps available for creating a 13.1 Gen2 instance
on Azure? The only option I see is Gen1.
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #16 from Mark Johnston ---
(In reply to Michael Baptist from comment #15)
Could you be able to try a verbose boot as well? i.e., try booting from the
loader with "boot -v". It might be an option under "loader options" as well.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #15 from Michael Baptist ---
At this point output just stops.
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #14 from Michael Baptist ---
/ __ _ _
| | | _ \ / | __ \
| |___ _ __ ___ ___ | |_) | (___ | | | |
| ___| '__/ _ \/ _ \| _ < \___ \| | | |
| | | | | __/
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #13 from Michael Baptist ---
(In reply to Mark Johnston from comment #12)
I've seen output like the trademark, and copyright in init_main.c get output,
however its "random" (to me) as to how much makes it to the console before
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #12 from Mark Johnston ---
(In reply to Michael Baptist from comment #4)
Is there any output following the BOOT tag? The first comment suggests so, but
how far are we getting, exactly?
--
You are receiving this mail because:
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #11 from Wei Hu ---
(In reply to Michael Baptist from comment #10)
This is from an on-prem Windows Server 2019 Build 10.0.17763. I created a Gen2
VM from a 12.3 iso, and then built and booted a kernel from 13 stable branch.
So,
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #10 from Michael Baptist ---
> By the way, I was able to boot FreeBSD 13.1 Gen2 VM on Intel Xeon 8260
> (Cascade Lake) host in my lab.
Is this a hyper V or Azure Stack HCI host or Azure Cloud host?
I also have been able to bo
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #9 from Michael Baptist ---
I've tried a few different sizes.
For example E32sV4/V5, D32s.
The LA57 page table code is not enabled I believe that is correct, however
booting in place from EFI memmap is part of 13.1 release. Th
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #8 from Wei Hu ---
Another thing I can think of is the LA57 support on FreeBSD is not working on
Hyper-V. But it is disabled on 13.1 by default as far as I know.
Maybe some new CPU features are supported well on FreeBSD. That's
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
Wei Hu changed:
What|Removed |Added
CC||w...@microsoft.com
--- Comment #7 from We
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #6 from Michael Baptist ---
OK memmap
Type Physical Virtual #Pages Attr
ConventionalMemory 00a0 UC WC WT WB
BootServicesData 0010
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #5 from Michael Baptist ---
I'll dump memmap from the EFI loader as well when I get a change. I'll be away
for a few days.
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #4 from Michael Baptist ---
/ __ _ _
| | | _ \ / | __ \
| |___ _ __ ___ ___ | |_) | (___ | | | |
| ___| '__/ _ \/ _ \| _ < \___ \| | | |
| | | | | __/
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
--- Comment #3 from Li-Wen Hsu ---
I'm also debugging this issue, in my tests that -CURRENT doesn't with this
issue, but in 13.1 and stable/13 this issue happens randomly. (I have an
13.1-R image which boots fine in some VM sizes during my
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=264267
Kubilay Kocak changed:
What|Removed |Added
Keywords||crash, needs-qa
Assignee
34 matches
Mail list logo
