Tags give the ability to mark specific points in history as being important
-
-
-
for-6.9/dm-changes
65e8fbde · ·- Fix DM core's IO submission (which include dm-io and dm-bufio) such that a bio's IO priority is propagated. Work focused on enabling both DM crypt and verity targets to retain the appropriate IO priority. - Fix DM raid reshape logic to not allow an empty flush bio to be requeued due to false concern about the bio, which doesn't have a data payload, accessing beyond the end of the device. - Fix DM core's internal resume so that it properly calls both presume and resume methods, which fixes the potential for a postsuspend and resume imbalance. - Update DM verity target to set DM_TARGET_SINGLETON flag because it doesn't make sense to have a DM table with a mix of targets that include dm-verity. - Small cleanups in DM crypt, thin, and integrity targets. - Fix references to dm-devel mailing list to use latest list address.
-
nfsd-6.9
9b350d3e · ·NFSD 6.9 Release Notes The bulk of the patches for this release are optimizations, code clean-ups, and minor bug fixes. One new feature to mention is that NFSD administrators now have the ability to revoke NFSv4 open and lock state. NFSD's NFSv3 support has had this capability for some time. As always I am grateful to NFSD contributors, reviewers, and testers.
-
-
-
-
wq-for-6.9-bh-conversions
8fea0c8f · ·workqueue: BH workqueue conversions for v6.9 This pull request contains two patches that convert tasklet users to BH workqueue - backtractest and usb hcd. DM conversions are being routed through the respective subsystem tree. Hopefully, the next cycle will see a lot more conversions.
-
wq-for-6.9
1acd92d9 · ·workqueue: Changes for v6.9 This cycle, a lot of workqueue changes including some that are significant and invasive. - During v6.6 cycle, unbound workqueues were updated so that they are more topology aware and flexible, which among other things improved workqueue behavior on modern multi-L3 CPUs. In the process, 636b927eba5b ("workqueue: Make unbound workqueues to use per-cpu pool_workqueues") switched unbound workqueues to use per-CPU frontend pool_workqueues as a part of increasing front-back mapping flexibility. An unwelcome side effect of this change was that this made max concurrency enforcement per-CPU blowing up the maximum number of allowed concurrent executions. I incorrectly assumed that this wouldn't cause practical problems as most unbound workqueue users are self-regulate max concurrency; however, there definitely are which don't (e.g. on IO paths) and the drastic increase in the allowed max concurrency led to noticeable perf regressions in some use cases. This is now addressed by separating out max concurrency enforcement to a separate struct - wq_node_nr_active - which makes @max_active consistently mean system-wide max concurrency regardless of the number of CPUs or (finally) NUMA nodes. This is a rather invasive and, in places, a bit clunky; however, the clunkiness rises from the the inherent requirement to handle the disagreement between the execution locality domain and max concurrency enforcement domain on some modern machines. See 5797b1c18919 ("workqueue: Implement system-wide nr_active enforcement for unbound workqueues") for more details. - BH workqueue support is added. They are similar to per-CPU workqueues but execute work items in the softirq context. This is expected to replace tasklet. However, currently, it's missing the ability to disable and enable work items which is needed to convert many tasklet users. To avoid crowding this merge window too much, this will be included in the next merge window. A separate pull request will be sent for the couple conversion patches that are currently pending. - Waiman plugged a long-standing hole in workqueue CPU isolation where ordered workqueues didn't follow wq_unbound_cpumask updates. Ordered workqueues now follow the same rules as other unbound workqueues. - More CPU isolation improvements: Juri fixed another deficit in workqueue isolation where unbound rescuers don't respect wq_unbound_cpumask. Leonardo fixed delayed_work timers firing on isolated CPUs. - Other misc changes. -
-
-
-
-
for-6.9/dm-vdo
cb824724 · ·- Introduce the DM vdo target which provides block-level deduplication, compression, and thin provisioning. Please see both: Documentation/admin-guide/device-mapper/vdo.rst and Documentation/admin-guide/device-mapper/vdo-design.rst - The DM vdo target handles its concurrency by pinning an IO, and subsequent stages of handling that IO, to a particular VDO thread. This aspect of VDO is "unique" but its overall implementation is very tightly coupled to its mostly lockless threading model. As such, VDO is not easily changed to use more traditional finer-grained locking and Linux workqueues. Please see the "Zones and Threading" section of vdo-design.rst - The DM vdo target has been used in production for many years but has seen significant changes over the past ~6 years to prepare it for upstream inclusion. The codebase is still large but it is isolated to drivers/md/dm-vdo/ and has been made considerably more approachable and maintainable. - Matt Sakai has been added to the MAINTAINERS file to reflect that he will send VDO changes upstream through the DM subsystem maintainers.
-
for-6.9/dm-bh-wq
c375b223 · ·- Convert the DM verity and crypt targets from (ab)using tasklets to using BH workqueues. These changes were coordinated with Tejun and are based ontop of DM's 6.9 changes and Tejun's 6.9 workqueue tree.
-