linux-kernel - [PATCH 0/7 v3] Make wake_up

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20240925053405.3960701-1-neilb@suse.de>
Date: Wed, 25 Sep 2024 15:31:37 +1000
From: NeilBrown <neilb@...e.de>
To: Ingo Molnar <mingo@...hat.com>,
	Peter Zijlstra <peterz@...radead.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>
Cc: linux-kernel@...r.kernel.org
Subject: [PATCH 0/7 v3] Make wake_up_{bit,var} less fragile

Hi,

 this is a revised set of patches for cleaning up wake_up_bit and
 wake_up_var and related.  They are the result of more proof reading,
 and of trying to make use of the new interfaces anywhere in the kernel
 where these interfaces are used.

 This should apply after -rc1, or at least after the block tree is merged.

 Note that on current master these cause a build error in bcachefs as it
 hasn't been updated with the new i_state bit management.

 Apart from proof-reading fixes this series enhances the previous series:

 - adds wait_var_event_any_lock() which can be used with any lock for
   which there are foo_lock and foo_unlock functions.  This is useful to
   deal with places that want to call filemap_invalidate_unlock/lock
   while waiting for a var

 - removes might_sleep() for wait_var_event_spinlock() as the code won't
   sleep until after it take the lock

 - adds wait_var_event_io() for io_schedule waits - xfs can use this.

 - as a bonus, softirq gets some improvements to waiting.  Feel free to
   drop or delay that one if not completely comfortable.

Thanks,
NeilBrown