| Message ID | 20230321180610.2620012-1-neal@gompa.dev (mailing list archive) |
|---|---|
| Headers | show |
| Series | Enforce 4k sectorize by default for mkfs | expand |
On 2023/3/22 02:06, Neal Gompa wrote: > The Fedora Asahi SIG[0] is working on bringing up support for > Apple Silicon Macintosh computers through the Asahi Fedora Remix[1]. > > Apple Silicon Macs are unusual in that they currently require 16k > page sizes, which means that the current default for mkfs.btrfs(8) > makes a filesystem that is unreadable on x86 PCs and most other ARM > PCs. > > Soon, this will be even more of a problem within Apple Silicon Macs > as Asahi Lina is working on 4k support to enable x86 emulation[2] > and since Linux does not support dynamically switching page sizes at > runtime, users will likely regularly switch back and forth depending > on their needs. > > Thus, I'd like to see us finally make the switchover to 4k sectorsize > for new filesystems by default, regardless of page size. > > The initial test run by Hector Martin[3] at request of Qu Wenruo > looks promising[4], and I hope we can get this to land upstream soon. Reviewed-by: Qu Wenruo <wqu@suse.com> And I'm already pulling my aarch64 VM to run subpage tests again, and this time with 16K page size, just like the one utilized by Asahi Linux project. The failures reported by Hector is being investigated, and such switch in mkfs would bring more feedback from other distros too. Hopefully we can also remove the experimental warning from kernel soon. Thanks, Qu > > This is an update on the initial RFC patch[5], which addresses the > documentation feedback from Anand Jain. > > [0]: https://fedoraproject.org/wiki/SIGs/Asahi > [1]: https://asahi-fedora-remix.org/ > [2]: https://vt.social/@lina/110060963422545117 > [3]: https://lore.kernel.org/linux-btrfs/fdffeecd-964f-0c69-f869-eb9ceca20263@suse.com/T/#m11d7939de96c43b3a7cdabc7c568d8bcafc7ca83 > [4]: https://lore.kernel.org/linux-btrfs/fdffeecd-964f-0c69-f869-eb9ceca20263@suse.com/T/#mf382b78a8122b0cb82147a536c85b6a9098a2895 > [5]: https://lore.kernel.org/linux-btrfs/fdffeecd-964f-0c69-f869-eb9ceca20263@suse.com/T/#t > > Neal Gompa (1): > btrfs-progs: mkfs: Enforce 4k sectorsize by default > > Documentation/Subpage.rst | 15 ++++++++------- > Documentation/mkfs.btrfs.rst | 13 +++++++++---- > mkfs/main.c | 2 +- > 3 files changed, 18 insertions(+), 12 deletions(-) >
On Tue, Mar 21, 2023 at 02:06:09PM -0400, Neal Gompa wrote: > The Fedora Asahi SIG[0] is working on bringing up support for > Apple Silicon Macintosh computers through the Asahi Fedora Remix[1]. > > Apple Silicon Macs are unusual in that they currently require 16k > page sizes, which means that the current default for mkfs.btrfs(8) > makes a filesystem that is unreadable on x86 PCs and most other ARM > PCs. > > Soon, this will be even more of a problem within Apple Silicon Macs > as Asahi Lina is working on 4k support to enable x86 emulation[2] > and since Linux does not support dynamically switching page sizes at > runtime, users will likely regularly switch back and forth depending > on their needs. > > Thus, I'd like to see us finally make the switchover to 4k sectorsize > for new filesystems by default, regardless of page size. > > The initial test run by Hector Martin[3] at request of Qu Wenruo > looks promising[4], and I hope we can get this to land upstream soon. > > This is an update on the initial RFC patch[5], which addresses the > documentation feedback from Anand Jain. > > [0]: https://fedoraproject.org/wiki/SIGs/Asahi > [1]: https://asahi-fedora-remix.org/ > [2]: https://vt.social/@lina/110060963422545117 > [3]: https://lore.kernel.org/linux-btrfs/fdffeecd-964f-0c69-f869-eb9ceca20263@suse.com/T/#m11d7939de96c43b3a7cdabc7c568d8bcafc7ca83 > [4]: https://lore.kernel.org/linux-btrfs/fdffeecd-964f-0c69-f869-eb9ceca20263@suse.com/T/#mf382b78a8122b0cb82147a536c85b6a9098a2895 > [5]: https://lore.kernel.org/linux-btrfs/fdffeecd-964f-0c69-f869-eb9ceca20263@suse.com/T/#t > This all looks good to me, you can add Reviewed-by: Josef Bacik <josef@toxicpanda.com> I've got a mac studio on the way, I'll add this to our CI testing to make sure we've got good coverage of this code going forward. Thanks, Josef
On Wed, Mar 22, 2023 at 10:33 AM Josef Bacik <josef@toxicpanda.com> wrote: > > On Tue, Mar 21, 2023 at 02:06:09PM -0400, Neal Gompa wrote: > > The Fedora Asahi SIG[0] is working on bringing up support for > > Apple Silicon Macintosh computers through the Asahi Fedora Remix[1]. > > > > Apple Silicon Macs are unusual in that they currently require 16k > > page sizes, which means that the current default for mkfs.btrfs(8) > > makes a filesystem that is unreadable on x86 PCs and most other ARM > > PCs. > > > > Soon, this will be even more of a problem within Apple Silicon Macs > > as Asahi Lina is working on 4k support to enable x86 emulation[2] > > and since Linux does not support dynamically switching page sizes at > > runtime, users will likely regularly switch back and forth depending > > on their needs. > > > > Thus, I'd like to see us finally make the switchover to 4k sectorsize > > for new filesystems by default, regardless of page size. > > > > The initial test run by Hector Martin[3] at request of Qu Wenruo > > looks promising[4], and I hope we can get this to land upstream soon. > > > > This is an update on the initial RFC patch[5], which addresses the > > documentation feedback from Anand Jain. > > > > [0]: https://fedoraproject.org/wiki/SIGs/Asahi > > [1]: https://asahi-fedora-remix.org/ > > [2]: https://vt.social/@lina/110060963422545117 > > [3]: https://lore.kernel.org/linux-btrfs/fdffeecd-964f-0c69-f869-eb9ceca20263@suse.com/T/#m11d7939de96c43b3a7cdabc7c568d8bcafc7ca83 > > [4]: https://lore.kernel.org/linux-btrfs/fdffeecd-964f-0c69-f869-eb9ceca20263@suse.com/T/#mf382b78a8122b0cb82147a536c85b6a9098a2895 > > [5]: https://lore.kernel.org/linux-btrfs/fdffeecd-964f-0c69-f869-eb9ceca20263@suse.com/T/#t > > > > This all looks good to me, you can add > > Reviewed-by: Josef Bacik <josef@toxicpanda.com> > > I've got a mac studio on the way, I'll add this to our CI testing to make sure > we've got good coverage of this code going forward. Thanks, > Since we've got three Reviewed-by tags (you, Qu, and Anand), do I need to send another version of this to land or can this land as-is?