linux-kernel - [PATCH v3] serial: core: Call device_set_awake

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>] [day] [month] [year] [list]

Message-Id: <20240430095930.2806067-1-claudiu.beznea.uj@bp.renesas.com>
Date: Tue, 30 Apr 2024 12:59:30 +0300
From: Claudiu <claudiu.beznea@...on.dev>
To: gregkh@...uxfoundation.org,
	jirislaby@...nel.org,
	tony@...mide.com,
	andriy.shevchenko@...ux.intel.com,
	l.sanfilippo@...bus.com,
	tglx@...utronix.de,
	geert+renesas@...der.be,
	ulf.hansson@...aro.org,
	peng.fan@....com
Cc: linux-kernel@...r.kernel.org,
	linux-serial@...r.kernel.org,
	linux-pm@...r.kernel.org,
	claudiu.beznea@...on.dev,
	Claudiu Beznea <claudiu.beznea.uj@...renesas.com>,
	Geert Uytterhoeven <geert@...ux-m68k.org>,
	Tony Lindgren <tony.lindgren@...ux.intel.com>
Subject: [PATCH v3] serial: core: Call device_set_awake_path() for console port

From: Claudiu Beznea <claudiu.beznea.uj@...renesas.com>

In case the UART port is used as a console, no_console_suspend is
available in bootargs and UART port is part of a software-controlled power
domain we need to call device_set_awake_path(). This lets the power
domain core code know that this domain should not be powered off
during system suspend. Otherwise, the UART port power domain is turned off,
nothing is printed while suspending and the suspend/resume process is
blocked. This was detected on the Renesas RZ/G3S SoC while adding support
for power domains.

Based on code investigation (on v6.9-rc5), this issue is present on other
SoCs (e.g., Renesas R-Mobile A1 [1], IMX8QXP [2]) and different SoCs have
particular implementation to handle it. Due to this the patch added the
call of device_set_awake_path() in uart_suspend_port() instead of having
it in the platform specific UART driver.

[1] drivers/pmdomain/renesas/rmobile-sysc.c:116
[2] drivers/pmdomain/imx/scu-pd.c:357

Suggested-by: Ulf Hansson <ulf.hansson@...aro.org>
Suggested-by: Geert Uytterhoeven <geert@...ux-m68k.org>
Reviewed-by: Ulf Hansson <ulf.hansson@...aro.org>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Reviewed-by: Tony Lindgren <tony.lindgren@...ux.intel.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@...der.be>
Tested-by: Geert Uytterhoeven <geert+renesas@...der.be>
Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@...renesas.com>
---

Changes in v3:
- remove https links from commit description
- collected tags
- removed RFT tag

Changes in v2:
- used device_set_awake_path() instead of device_set_wakeup_path()
- moved the support in uart_suspend_port() to make it generic for
  other drivers
- fixed typos in commit description
- updated the commit description to reflect the new changes and the fact
  that support may be applied to other SoCs
- added Suggested-by tag; this was initially proposed by Ulf to move it
  in the serial driver then Geert propose to have it more generic in
  uart_suspend_port()

 drivers/tty/serial/serial_core.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c
index 8a76f378c445..84ae70d52b05 100644
--- a/drivers/tty/serial/serial_core.c
+++ b/drivers/tty/serial/serial_core.c
@@ -2408,6 +2408,7 @@ int uart_suspend_port(struct uart_driver *drv, struct uart_port *uport)
 			uport->ops->stop_rx(uport);
 			uart_port_unlock_irq(uport);
 		}
+		device_set_awake_path(uport->dev);
 		goto unlock;
 	}
 
-- 
2.39.2