Hi,
As you probably saw, 4.18 was released on Sunday. The 4.18 kernel was built for rawhide yesterday. We will be following roughly the same rebase schedule as in the past. F28 will get the rebase first, typically around 4.18.2 or 4.18.3 followed by F27 shortly thereafter. My best guess for when this would happen would be middle of September. As always, we'll keep you updated on any new developments.
Thanks, Laura
P.S. As a general reminder, if you like F27 please remember to give karma to kernels.
On Tue, Aug 14, 2018 at 11:14 AM, Laura Abbott labbott@redhat.com wrote:
Hi,
As you probably saw, 4.18 was released on Sunday. The 4.18 kernel was built for rawhide yesterday. We will be following roughly the same rebase schedule as in the past. F28 will get the rebase first, typically around 4.18.2 or 4.18.3 followed by F27 shortly thereafter. My best guess for when this would happen would be middle of September. As always, we'll keep you updated on any new developments.
Also of note, the 4.18 test day will be next Tuesday the 21st. Anyone
willing to test would be a great help towards making the rebase go as smoothly as possible. More details can be found https://fedoraproject.org/wiki/Test_Day:2018-08-21_Kernel_4.18_Test_Day with images posted the day before (Monday).
Thanks, Justin
Am 14.08.2018 um 19:16 schrieb Justin Forbes:
On Tue, Aug 14, 2018 at 11:14 AM, Laura Abbott labbott@redhat.com wrote:
As you probably saw, 4.18 was released on Sunday. The 4.18 kernel was built for rawhide yesterday. We will be following roughly the same rebase schedule as in the past. F28 will get the rebase first, typically around 4.18.2 or 4.18.3 followed by F27 shortly thereafter. My best guess for when this would happen would be middle of September. As always, we'll keep you updated on any new developments.
Also of note, the 4.18 test day will be next Tuesday the 21st. Anyone
willing to test would be a great help towards making the rebase go as smoothly as possible. More details can be found https://fedoraproject.org/wiki/Test_Day:2018-08-21_Kernel_4.18_Test_Day with images posted the day before (Monday)
[root@testserver:~]$ uname -a Linux testserver.rhsoft.net 4.18.0-1.fc29.x86_64 #1 SMP Mon Aug 13 16:10:16 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
[root@testserver:~]$ cat /etc/redhat-release Generic release 28 (Generic)
runs just fine with dbmail, dovecot, httpd, php, mariadb, spamassassin, clamav....
Hi,
On 14-08-18 18:14, Laura Abbott wrote:
Hi,
As you probably saw, 4.18 was released on Sunday. The 4.18 kernel was built for rawhide yesterday. We will be following roughly the same rebase schedule as in the past. F28 will get the rebase first, typically around 4.18.2 or 4.18.3 followed by F27 shortly thereafter. My best guess for when this would happen would be middle of September. As always, we'll keep you updated on any new developments.
FYI, there seems to be a problem with 4.18 as a virtualbox guest, sometimes when the user resizes the vm-window the contents of the vm Windows stop updating (when running a GUI) until the next resize.
This is probably an issue in the vboxvideo driver, triggered by changes in 4.18 elsewhere. I've looking into this on my to do list.
Regards,
Hans
On 08/16/2018 05:27 AM, Hans de Goede wrote:
Hi,
On 14-08-18 18:14, Laura Abbott wrote:
Hi,
As you probably saw, 4.18 was released on Sunday. The 4.18 kernel was built for rawhide yesterday. We will be following roughly the same rebase schedule as in the past. F28 will get the rebase first, typically around 4.18.2 or 4.18.3 followed by F27 shortly thereafter. My best guess for when this would happen would be middle of September. As always, we'll keep you updated on any new developments.
FYI, there seems to be a problem with 4.18 as a virtualbox guest, sometimes when the user resizes the vm-window the contents of the vm Windows stop updating (when running a GUI) until the next resize.
This is probably an issue in the vboxvideo driver, triggered by changes in 4.18 elsewhere. I've looking into this on my to do list.
Thanks for the report. I'll check back before doing the push if I don't hear anything before that.
Hi,
On 16-08-18 14:27, Hans de Goede wrote:
Hi,
On 14-08-18 18:14, Laura Abbott wrote:
Hi,
As you probably saw, 4.18 was released on Sunday. The 4.18 kernel was built for rawhide yesterday. We will be following roughly the same rebase schedule as in the past. F28 will get the rebase first, typically around 4.18.2 or 4.18.3 followed by F27 shortly thereafter. My best guess for when this would happen would be middle of September. As always, we'll keep you updated on any new developments.
FYI, there seems to be a problem with 4.18 as a virtualbox guest, sometimes when the user resizes the vm-window the contents of the vm Windows stop updating (when running a GUI) until the next resize.
This is probably an issue in the vboxvideo driver, triggered by changes in 4.18 elsewhere. I've looking into this on my to do list.
Never mind it turns out that this is a pre-existing problem which got triggered due to other changes (it requires the user-session to be GNOME on Wayland to trigger and I was mostly testing with GNOME on Xorg).
TL;DR: There still is a bug, but it is not a regression in 4.18.
Regards,
Hans
On 08/14/2018 09:14 AM, Laura Abbott wrote:
Hi,
As you probably saw, 4.18 was released on Sunday. The 4.18 kernel was built for rawhide yesterday. We will be following roughly the same rebase schedule as in the past. F28 will get the rebase first, typically around 4.18.2 or 4.18.3 followed by F27 shortly thereafter. My best guess for when this would happen would be middle of September. As always, we'll keep you updated on any new developments.
Thanks, Laura
P.S. As a general reminder, if you like F27 please remember to give karma to kernels.
4.18.2 is nominally expected to be released this weekend but stable updates have been accelerated due to needing to get L1TF fixes out. At the very least we need to wait until the merge window closes and the test day happens before pushing the rebase. F29 will be up to date for anyone who wants to test that kernel.
On 08/17/2018 06:57 AM, Laura Abbott wrote:
On 08/14/2018 09:14 AM, Laura Abbott wrote:
Hi,
As you probably saw, 4.18 was released on Sunday. The 4.18 kernel was built for rawhide yesterday. We will be following roughly the same rebase schedule as in the past. F28 will get the rebase first, typically around 4.18.2 or 4.18.3 followed by F27 shortly thereafter. My best guess for when this would happen would be middle of September. As always, we'll keep you updated on any new developments.
Thanks, Laura
P.S. As a general reminder, if you like F27 please remember to give karma to kernels.
4.18.2 is nominally expected to be released this weekend but stable updates have been accelerated due to needing to get L1TF fixes out. At the very least we need to wait until the merge window closes and the test day happens before pushing the rebase. F29 will be up to date for anyone who wants to test that kernel.
My current plan is to push the F28 rebase early next week. If you get a chance, please test the F29 kernel or the package in the kernel-stabilization COPR.
Laura
"LA" == Laura Abbott labbott@redhat.com writes:
LA> My current plan is to push the F28 rebase early next week. If you LA> get a chance, please test the F29 kernel or the package in the LA> kernel-stabilization COPR.
I thought we weren't doing kernel-stabilization this time around because the F29 kernels were sufficient, but looking now I see it's there (though the 4.18.5 build failed).
I've been running F29 kernels (on my F28 desktop) since 4.18.0 showed up there and everything has been well. Currently running kernel-4.18.5-300.fc29 from https://koji.fedoraproject.org/koji/buildinfo?buildID=1139115 with no issues at all. I just had to do this:
mkdir kernel; cd kernel koji download-build -a x86_64 kernel-4.18.5-300.fc29 dnf update *rpm
and reboot into the new kernel.
- J<
On 08/29/2018 01:05 PM, Jason L Tibbitts III wrote:
"LA" == Laura Abbott labbott@redhat.com writes:
LA> My current plan is to push the F28 rebase early next week. If you LA> get a chance, please test the F29 kernel or the package in the LA> kernel-stabilization COPR.
I thought we weren't doing kernel-stabilization this time around because the F29 kernels were sufficient, but looking now I see it's there (though the 4.18.5 build failed).
I didn't end up pushing anything to the kernel-stabilization branch but I did use the stabilization repo to test the actual rebased versions since the work looks slightly different between a straight F29 branch and a rebase (dropping some patches and config options)
I've been running F29 kernels (on my F28 desktop) since 4.18.0 showed up there and everything has been well. Currently running kernel-4.18.5-300.fc29 from https://koji.fedoraproject.org/koji/buildinfo?buildID=1139115 with no issues at all. I just had to do this:
Oh I missed that. It looks like it was an infra problem. I'll keep an eye out when I do the final rebase.
mkdir kernel; cd kernel koji download-build -a x86_64 kernel-4.18.5-300.fc29 dnf update *rpm
and reboot into the new kernel.
Excellent, thanks for your report and support as always
On 08/14/2018 09:14 AM, Laura Abbott wrote:
Hi,
As you probably saw, 4.18 was released on Sunday. The 4.18 kernel was built for rawhide yesterday. We will be following roughly the same rebase schedule as in the past. F28 will get the rebase first, typically around 4.18.2 or 4.18.3 followed by F27 shortly thereafter. My best guess for when this would happen would be middle of September. As always, we'll keep you updated on any new developments.
Thanks, Laura
P.S. As a general reminder, if you like F27 please remember to give karma to kernels.
The kernel is now in testing for F28. Please test and give karma as I would like to push very soon.
"LA" == Laura Abbott labbott@redhat.com writes:
LA> The kernel is now in testing for F28. Please test and give karma as LA> I would like to push very soon.
For the record, that's https://bodhi.fedoraproject.org/updates/FEDORA-2018-a8b6cb002c
which can be fetched with dnf update --enablerepo=updates-testing --advisory=FEDORA-2018-a8b6cb002c
- J<
Am 07.09.18 um 02:41 schrieb Laura Abbott:
On 08/14/2018 09:14 AM, Laura Abbott wrote:
Hi,
As you probably saw, 4.18 was released on Sunday. The 4.18 kernel was built for rawhide yesterday. We will be following roughly the same rebase schedule as in the past. F28 will get the rebase first, typically around 4.18.2 or 4.18.3 followed by F27 shortly thereafter. My best guess for when this would happen would be middle of September. As always, we'll keep you updated on any new developments.
Thanks, Laura
P.S. As a general reminder, if you like F27 please remember to give karma to kernels.
The kernel is now in testing for F28. Please test and give karma as I would like to push very soon.
on virtual machines i use the f29 for some time
for the phyiscal hosts i would be thankful if someone starts a 4.18.6 build for F28 to avoid another reboot cycle the next day given the amount of bugfixes and that even Phoronix wrote an article with "Linux 4.18.6 Rolls Out With L1TF Foreshadow Fixes"
some items in the upstream changelog may even fix the "/sys/devices/system/cpu/vulnerabilities/l1tf:Vulnerable" on my machine with 32 GB RAM while all others have "/sys/devices/system/cpu/vulnerabilities/l1tf:Mitigation: PTE Inversion" even vm's on top of the 32 GB machine which seems to be still vulnerable
https://www.phoronix.com/scan.php?page=news_item&px=Linux-4.18.6-Stable-...
https://koji.fedoraproject.org/koji/packageinfo?packageID=8
kernel-4.18.6-300.fc29 labbott 2018-09-06 01:05:35 kernel-4.18.5-200.fc28 labbott 2018-09-04 21:12:52
Am 07.09.18 um 06:11 schrieb Reindl Harald:
https://www.phoronix.com/scan.php?page=news_item&px=Linux-4.18.6-Stable-...
https://koji.fedoraproject.org/koji/packageinfo?packageID=8
kernel-4.18.6-300.fc29 labbott 2018-09-06 01:05:35 kernel-4.18.5-200.fc28 labbott 2018-09-04 21:12:52
FTR: kernel-4.18.5 shoukd have been replaced by kernel-4.18.6 and i *really* don't get why a build for F29 got fired up but not for F28
there are machines which need it for l1tf, see changelog below
"extremely unlikely system" is nonsense, it's an ordinary "HP Compaq Elite 8300 CMT" simply running with 32 GB
----------------
[root@srv-rhsoft:~]$ grep . /sys/devices/system/cpu/vulnerabilities/* /sys/devices/system/cpu/vulnerabilities/l1tf:Mitigation: PTE Inversion /sys/devices/system/cpu/vulnerabilities/meltdown:Mitigation: PTI /sys/devices/system/cpu/vulnerabilities/spec_store_bypass:Mitigation: Speculative Store Bypass disabled via prctl and seccomp /sys/devices/system/cpu/vulnerabilities/spectre_v1:Mitigation: __user pointer sanitization /sys/devices/system/cpu/vulnerabilities/spectre_v2:Mitigation: Full generic retpoline, IBPB, IBRS_FW
[root@srv-rhsoft:~]$ uname -r 4.18.6-300.fc29.x86_64 #1 SMP Wed Sep 5 20:11:09 UTC 2018
[root@srv-rhsoft:~]$ cat /etc/redhat-release Generic release 28 (Generic) ----------------
here you go: https://cdn.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.18.6
x86/speculation/l1tf: Fix off-by-one error when warning that system has too much RAM
commit b0a182f875689647b014bc01d36b340217792852 upstream.
Two users have reported [1] that they have an "extremely unlikely" system with more than MAX_PA/2 memory and L1TF mitigation is not effective. In fact it's a CPU with 36bits phys limit (64GB) and 32GB memory, but due to holes in the e820 map, the main region is almost 500MB over the 32GB limit:
[ 0.000000] BIOS-e820: [mem 0x0000000100000000-0x000000081effffff] usable
Suggestions to use 'mem=32G' to enable the L1TF mitigation while losing the 500MB revealed, that there's an off-by-one error in the check in l1tf_select_mitigation().
kernel@lists.fedoraproject.org
-
Hans de Goede -
Jason L Tibbitts III -
Justin Forbes -
Laura Abbott -
Reindl Harald