 |
A RHEL user is reporting the same issue: https://access.redhat.com/discussions/5895461 |
|
|
Because this is a bug in the RHEL kernel code, this needs to be reported upstream at bugzilla.redhat.com . The following patch in the changelog is a suspect: [Hyper-V][RHEL-7.9]video: hyperv_fb: Fix the cache type when mapping the VRAM Edit (BZ#1908896) |
|
|
Bug report filed: https://bugzilla.redhat.com/show_bug.cgi?id=1941841 * Description of problem (copied from https://bugs.centos.org/view.php?id=18117): After updating to kernel version 3.10.0-1160.21.1.el7.x86_64 on a Hyper-V (Windows 10) VM, the OS will not boot. There appear to be error messages output but the display is affected (squashed) and they are illegible. Selecting another kernel from the grub menu allows the OS to boot. * The following patch seems to have caused the issue: [Hyper-V][RHEL-7.9]video: hyperv_fb: Fix the cache type when mapping the VRAM Edit (BZ#1908896) kernel commit: commit 5f1251a48c17b54939d7477305e39679a565382c Author: Dexuan Cui <decui@microsoft.com> Date: Tue Nov 17 16:03:05 2020 -0800 video: hyperv_fb: Fix the cache type when mapping the VRAM * The issue has been fixed by: Stable kernel commit: commit 452f087d2ff6decf298149e0bfd9fa5c212a636d Author: Dexuan Cui <decui@microsoft.com> Date: Sat Jan 9 14:53:58 2021 -0800 video: hyperv_fb: Fix the mmap() regression for v5.4.y and older |
|
|
I've built a plus kernel with the patch that is supposed to fix the problem. It is available from: https://people.centos.org/toracat/kernel/7/bugs/18117/ Please test, if you can. |
|
|
@raydecampo The following question was asked in the referenced RH bug report. Could you find out the host build version of your system? =================== I am trying to reproduce the issue with 3.10.0-1160.21.1.el7.x86_64 in VM on Hyper-V host, but failed to reproduce on Windows server 2019 (17763-10.0-1-0.1757) or Windows server 2016 (14393-10.0-4-0.4169), perhaps it is related to specify host build, could you help to check the host build version on which you meet this problem? To get host build, you can run 'dmesg | grep "Hyper-V Host Build"' on the vm =================== |
 |
@toracat Edition: Windows 10 Business Version: 20H2 OS build: 19042.867 Experience: Windows Feature Experience Pack 120.2212.551.0 There is something more going on, yesterday I spun up a fresh install of CentOS 7 on a new VM in the same host (with the intent of testing the plus kernel) and I was unable to reproduce the issue. It still reliably reproduces on the existing CentOS 7 installs however. Perhaps it is because the existing instances have a number of kernels installed? I will try to test the plus kernel on the system exhibiting the issue, although I am a little uncertain concerning how to do that. Should I install the problematic kernel version and then install the plus kernel RPMs from the link? Or just install the plus kernel RPMs? Or something else? |
|
|
Thank you. I copied your response in the RHBZ. Regarding the plus kernel, the best test should be something that determines whether or not the proposed patch fixes the problem. So, take a system that is exhibiting the issue and then install the plus kernel and boot to it. If the problem is corrected, I would say the patch worked. |
|
|
Unfortunately the plus kernel did not address the issue, no change in behavior. First I tried installing just the plus kernel. When that failed I installed all the RPMs from the linked location. |
|
|
Thanks for testing and reporting back. This could mean that the culprit is something else. Needs more digging. |
|
|
@raydecampo As seen in the upstream BZ, a microsoft developer identified the cause of the bug and proposed a fix: https://bugzilla.redhat.com/show_bug.cgi?id=1941841#c53 I have used the suggested patch: --- drivers/video/hyperv_fb.c.old 2021-03-26 19:09:14.694996600 -0700 +++ drivers/video/hyperv_fb.c 2021-03-26 19:09:37.602996600 -0700 @@ -956,13 +956,14 @@ goto error1; } + hvfb_get_option(info); + ret = hvfb_getmem(hdev, info); if (ret) { pr_err("No memory for framebuffer\n"); goto error2; } - hvfb_get_option(info); pr_info("Screen resolution: %dx%d, Color depth: %d\n", screen_width, screen_height, screen_depth); and rebuilt the plus kernel (kernel-plus-3.10.0-1160.21.1.el7.centos.plus.bug18117.1.x86_64.rpm). You can find it here: https://people.centos.org/toracat/kernel/7/bugs/18117/ Could you try testing this kernel and see if the patch has fixed the issue? |
|
|
I have updated the plus kernel set to kernel-plus-3.10.0-1160.21.1.el7.centos.plus.bug18117.2.x86_64.rpm. This version contains the following patch: commit 67e7cdb4829d ("video: hyperv: hyperv_fb: Obtain screen resolution from Hyper-V host") It is referenced as "Patch A" in https://bugzilla.redhat.com/show_bug.cgi?id=1941841#c58 Please test, if you can. |
|
|
@toracat I tested the new plus kernel you provided and that does resolve the issue. Thank you for your help. |
|
|
@raydecampo Thanks for the feedback. Great to hear the fix was confirmed. According to the latest comment on the RHBZ: "We plan to include the missing patch and push a fix out in an upcoming RHEL 7.9.z batch update. You can follow the current status in this BZ." If the patch does not make it in the next update, the plus kernel will carry it until the distro kernel gets fixed. |
|
|
kernel-4.18.0-240.22.1.el8_3 was released upstream. The hyper-v patch is not in there, so the plus kernel will be built with it. |
|
|
Fixed in kernel-3.10.0-1160.31.1.el7. |
- Anonymous
