![]() ![]() > Fedora/RHEL kernel tree for Fedora's upstream tree.Īpologies, it was a bit testy sounding, but no, there is *no* different policy. That would be a gross misuse of having a blended > different for other storage subsystems just because there happens to be > I absolutely do not appreciate the jab here. One last thing to note, is that the current latest btrfs documentation still seems to suggest that no discard should still be the default: `discard=async` is shown on each btrfs mount line when `mount` is ran as of 6.2. Issue occurred on a previous boot, use the journalctl ``-b`` flag. You can get the complete kernel logįor a boot with ``journalctl -no-hostname -k > dmesg.txt``. Are you running any modules that not shipped with directly Fedora's kernel?:ħ. ``sudo dnf update -enablerepo=rawhide kernel``:Ħ. Rawhide kernel, run ``sudo dnf install fedora-repos-rawhide`` followed by Does this problem occur with the latest Rawhide kernel? To install the I have reproduced this issue on two laptops, one running Fedora 37 and one running Fedora 38, both with nvme drives.ĥ. Can you reproduce this issue? If so, please provide the steps to reproduce It first appeared as of the 6.2 kernel release.Ĥ. ![]() *first* appear? Old kernels are available for download at Did it work previously in Fedora? If so, what kernel version did the issue What is the Version-Release number of the kernel:įedora 36 very likely impacted, but I did not test: kernel-6.2.8-100.fc36ģ. The result by this redundant trim (over our current fstrim.timer method) is additional IO overhead and possibly a negative impact to SSD longevity. This impacts all Fedora 37 and 38 users who upgraded to kernel-6.2. As of kernel-6.2, the defaults for btrfs have apparently changed such that the "discard=async" option is applied to btrfs unless "nodiscard" option is explicitly given at mount/fstab. ![]()
0 Comments
Leave a Reply. |