Hi, This is v2 of a series working towards threaded/atomic console printing. This series is only a subset of the original v1 [0]. That series began with various cleanups before adding the threaded/atomic code. Since none of those cleanups made it into the recent 6.5 merge window, I have decided to post them as their own series. I hope this helps to get them accepted without being attached to any threaded/atomic discussions.
Changes since v1: - drop patches 5-18 (they will return in a follow-up series) - console_unblank() aborts if called from NMI context - console_flush_on_panic() directly flushes rather than using console_lock/_unlock dance (because console_lock/_unlock is not NMI-safe) - remove @console_suspended and rely only on CON_SUSPENDED flag - rename abandon_console_lock_in_panic() to other_cpu_in_panic() - console_trylock() and console_lock() will fail and block, respectively, while another CPU is in panic - adjust various comments and whitespace as suggested John Ogness [0] https://lore.kernel.org/lkml/20230302195618.156940-1-john.ogn...@linutronix.de John Ogness (5): kdb: do not assume write() callback available printk: Add NMI safety to console_flush_on_panic() and console_unblank() printk: Consolidate console deferred printing printk: Add per-console suspended state printk: Rename abandon_console_lock_in_panic() to other_cpu_in_panic() include/linux/console.h | 3 + kernel/debug/kdb/kdb_io.c | 2 + kernel/printk/internal.h | 2 + kernel/printk/printk.c | 189 +++++++++++++++++++++++------------- kernel/printk/printk_safe.c | 9 +- 5 files changed, 132 insertions(+), 73 deletions(-) base-commit: 7ec85f3e089aa423a69559bf4555b6218b5a2ef7 -- 2.30.2 _______________________________________________ Kgdb-bugreport mailing list Kgdb-bugreport@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/kgdb-bugreport
