Description of problem: dm-crypt devices located on SSD storage should have the `no_read_workqueue` and `no_write_workqueue` set by default when the devices are created the first time in the installer. See the following discussions: - https://wiki.archlinux.org/title/Dm-crypt/Specialties#Disable_workqueue_for_increased_solid_state_drive_(SSD)_performance - https://github.com/fedora-silverblue/issue-tracker/issues/388 - https://blog.cloudflare.com/speeding-up-linux-disk-encryption/ - https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/drivers/md/dm-crypt.c?id=39d42fa96ba1b7d2544db3f8ed5da8fb0d5cb877 Version-Release number of selected component (if applicable): Latest Fedora How reproducible: Always Steps to Reproduce: 1. Install Fedora 37/Rawhide with an encrypted disk 2. `cryptsetup luksDump /dev/sdaX | grep Flags` returns `Flags: <empty>` Actual results: Low disk read/write performance with `Flags: <empty>` Expected results: Fast disk read/write performance with `Flags: no-read-workqueue no-write-workqueue`
It looks like something that should be handled in out storage library. Reassigning to blivet.
This bug appears to have been reported against 'rawhide' during the Fedora Linux 38 development cycle. Changing version to 38.
This bug appears to have been reported against 'rawhide' during the Fedora Linux 40 development cycle. Changing version to 40.