Ever since 5.19.0-rc1 the serial device has taken almost a second longer in both suspend and resume. This effect is witnessed in half the machines in our lab (~10 machines). It occurs on dell, asus, gigabyte and other machines so it's not hardware specific. I opened a bugzilla issue which includes the git bisect: https://bugzilla.kernel.org/show_bug.cgi?id=216216 Bad commit: commit 3b604ca81202eea2a917eb6491e90f610fba0ec7 Author: John Ogness <john.ogness@xxxxxxxxxxxxx> Date: Thu Apr 21 23:28:46 2022 +0206 printk: add pr_flush() Provide a might-sleep function to allow waiting for console printers to catch up to the latest logged message. Use pr_flush() whenever it is desirable to get buffered messages printed before continuing: suspend_console(), resume_console(), console_stop(), console_start(), console_unblank(). Signed-off-by: John Ogness <john.ogness@xxxxxxxxxxxxx> Reviewed-by: Petr Mladek <pmladek@xxxxxxxx> Signed-off-by: Petr Mladek <pmladek@xxxxxxxx> Link: https://lore.kernel.org/r/20220421212250.565456-12-john.ogness@xxxxxxxxxxxxx include/linux/printk.h | 7 +++++ kernel/printk/printk.c | 83 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 90 insertions(+)
