[PATCH] PM/console: Fix the black screen issue

[PATCH] PM/console: Fix the black screen issue

[tuhaowen]

When the computer enters sleep status without a monitor
connected, the system switches the console to the virtual
terminal tty63(SUSPEND_CONSOLE).

If a monitor is subsequently connected before waking up,
the system skips the required VT restoration process
during wake-up, leaving the console on tty63 instead of
switching back to tty1.

Signed-off-by: tuhaowen <tuhaowen@uniontech.com>
---
 kernel/power/console.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/kernel/power/console.c b/kernel/power/console.c
index fcdf0e14a47d..832e04bf5439 100644
--- a/kernel/power/console.c
+++ b/kernel/power/console.c
@@ -16,6 +16,7 @@
 #define SUSPEND_CONSOLE	(MAX_NR_CONSOLES-1)
 
 static int orig_fgconsole, orig_kmsg;
+static int vt_switch_done;
 
 static DEFINE_MUTEX(vt_switch_mutex);
 
@@ -136,15 +137,19 @@ void pm_prepare_console(void)
 	if (orig_fgconsole < 0)
 		return;
 
+	vt_switch_done = 1;
+
 	orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE);
 	return;
 }
 
 void pm_restore_console(void)
 {
-	if (!pm_vt_switch())
+	if (!pm_vt_switch() && !vt_switch_done)
 		return;
 
+	vt_switch_done = 0;
+
 	if (orig_fgconsole >= 0) {
 		vt_move_to_console(orig_fgconsole, 0);
 		vt_kmsg_redirect(orig_kmsg);
-- 
2.20.1

Re: [PATCH] PM/console: Fix the black screen issue

[Rafael J. Wysocki]

On Fri, May 16, 2025 at 5:47 AM tuhaowen <tuhaowen@uniontech.com> wrote:
>
> When the computer enters sleep status without a monitor
> connected, the system switches the console to the virtual
> terminal tty63(SUSPEND_CONSOLE).
>
> If a monitor is subsequently connected before waking up,
> the system skips the required VT restoration process
> during wake-up, leaving the console on tty63 instead of
> switching back to tty1.

Please add some information here on how exactly the issue is being
fixed.  Reverse-engineering code changes is hard.

> Signed-off-by: tuhaowen <tuhaowen@uniontech.com>
> ---
>  kernel/power/console.c | 7 ++++++-
>  1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/power/console.c b/kernel/power/console.c
> index fcdf0e14a47d..832e04bf5439 100644
> --- a/kernel/power/console.c
> +++ b/kernel/power/console.c
> @@ -16,6 +16,7 @@
>  #define SUSPEND_CONSOLE        (MAX_NR_CONSOLES-1)
>
>  static int orig_fgconsole, orig_kmsg;
> +static int vt_switch_done;
>
>  static DEFINE_MUTEX(vt_switch_mutex);
>
> @@ -136,15 +137,19 @@ void pm_prepare_console(void)
>         if (orig_fgconsole < 0)
>                 return;
>
> +       vt_switch_done = 1;
> +
>         orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE);
>         return;
>  }
>
>  void pm_restore_console(void)
>  {
> -       if (!pm_vt_switch())
> +       if (!pm_vt_switch() && !vt_switch_done)
>                 return;
>
> +       vt_switch_done = 0;
> +
>         if (orig_fgconsole >= 0) {
>                 vt_move_to_console(orig_fgconsole, 0);
>                 vt_kmsg_redirect(orig_kmsg);
> --

[PATCH v2] PM/console: Fix the black screen issue

[tuhaowen]

When the computer enters sleep status without a monitor
connected, the system switches the console to the virtual
terminal tty63(SUSPEND_CONSOLE).

If a monitor is subsequently connected before waking up,
the system skips the required VT restoration process
during wake-up, leaving the console on tty63 instead of
switching back to tty1.

To fix this issue, a global flag vt_switch_done is introduced
to record whether the system has successfully switched to
the suspend console via vt_move_to_console() during suspend.

If the switch was completed, vt_switch_done is set to 1.
Later during resume, this flag is checked to ensure that
the original console is restored properly by calling
vt_move_to_console(orig_fgconsole, 0).

This prevents scenarios where the resume logic skips console
restoration due to incorrect detection of the console state,
especially when a monitor is reconnected before waking up.

Signed-off-by: tuhaowen <tuhaowen@uniontech.com>
---
Changes in v2:
- Added explanation in the commit message on how the issue is fixed.
- Link to v1: https://lore.kernel.org/all/20250516034643.22355-1-tuhaowen@uniontech.com
---
 kernel/power/console.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/kernel/power/console.c b/kernel/power/console.c
index fcdf0e14a47d..832e04bf5439 100644
--- a/kernel/power/console.c
+++ b/kernel/power/console.c
@@ -16,6 +16,7 @@
 #define SUSPEND_CONSOLE	(MAX_NR_CONSOLES-1)
 
 static int orig_fgconsole, orig_kmsg;
+static int vt_switch_done;
 
 static DEFINE_MUTEX(vt_switch_mutex);
 
@@ -136,15 +137,19 @@ void pm_prepare_console(void)
 	if (orig_fgconsole < 0)
 		return;
 
+	vt_switch_done = 1;
+
 	orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE);
 	return;
 }
 
 void pm_restore_console(void)
 {
-	if (!pm_vt_switch())
+	if (!pm_vt_switch() && !vt_switch_done)
 		return;
 
+	vt_switch_done = 0;
+
 	if (orig_fgconsole >= 0) {
 		vt_move_to_console(orig_fgconsole, 0);
 		vt_kmsg_redirect(orig_kmsg);
-- 
2.20.1

[PATCH v2 RESEND] PM/console: Fix the black screen issue

[tuhaowen]

When the computer enters sleep status without a monitor
connected, the system switches the console to the virtual
terminal tty63(SUSPEND_CONSOLE).

If a monitor is subsequently connected before waking up,
the system skips the required VT restoration process
during wake-up, leaving the console on tty63 instead of
switching back to tty1.

To fix this issue, a global flag vt_switch_done is introduced
to record whether the system has successfully switched to
the suspend console via vt_move_to_console() during suspend.

If the switch was completed, vt_switch_done is set to 1.
Later during resume, this flag is checked to ensure that
the original console is restored properly by calling
vt_move_to_console(orig_fgconsole, 0).

This prevents scenarios where the resume logic skips console
restoration due to incorrect detection of the console state,
especially when a monitor is reconnected before waking up.

Signed-off-by: tuhaowen <tuhaowen@uniontech.com>
---
Changes in v2:
- Added explanation in the commit message on how the issue is fixed.
- Link to v1: https://lore.kernel.org/all/20250516034643.22355-1-tuhaowen@uniontech.com
---
 kernel/power/console.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/kernel/power/console.c b/kernel/power/console.c
index fcdf0e14a47d..832e04bf5439 100644
--- a/kernel/power/console.c
+++ b/kernel/power/console.c
@@ -16,6 +16,7 @@
 #define SUSPEND_CONSOLE	(MAX_NR_CONSOLES-1)
 
 static int orig_fgconsole, orig_kmsg;
+static int vt_switch_done;
 
 static DEFINE_MUTEX(vt_switch_mutex);
 
@@ -136,15 +137,19 @@ void pm_prepare_console(void)
 	if (orig_fgconsole < 0)
 		return;
 
+	vt_switch_done = 1;
+
 	orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE);
 	return;
 }
 
 void pm_restore_console(void)
 {
-	if (!pm_vt_switch())
+	if (!pm_vt_switch() && !vt_switch_done)
 		return;
 
+	vt_switch_done = 0;
+
 	if (orig_fgconsole >= 0) {
 		vt_move_to_console(orig_fgconsole, 0);
 		vt_kmsg_redirect(orig_kmsg);
-- 
2.20.1

Re: [PATCH v2 RESEND] PM/console: Fix the black screen issue

[Rafael J. Wysocki]

On Fri, May 23, 2025 at 10:57 AM tuhaowen <tuhaowen@uniontech.com> wrote:
>
> When the computer enters sleep status without a monitor
> connected, the system switches the console to the virtual
> terminal tty63(SUSPEND_CONSOLE).
>
> If a monitor is subsequently connected before waking up,
> the system skips the required VT restoration process
> during wake-up, leaving the console on tty63 instead of
> switching back to tty1.
>
> To fix this issue, a global flag vt_switch_done is introduced
> to record whether the system has successfully switched to
> the suspend console via vt_move_to_console() during suspend.
>
> If the switch was completed, vt_switch_done is set to 1.
> Later during resume, this flag is checked to ensure that
> the original console is restored properly by calling
> vt_move_to_console(orig_fgconsole, 0).
>
> This prevents scenarios where the resume logic skips console
> restoration due to incorrect detection of the console state,
> especially when a monitor is reconnected before waking up.
>
> Signed-off-by: tuhaowen <tuhaowen@uniontech.com>
> ---
> Changes in v2:
> - Added explanation in the commit message on how the issue is fixed.
> - Link to v1: https://lore.kernel.org/all/20250516034643.22355-1-tuhaowen@uniontech.com
> ---
>  kernel/power/console.c | 7 ++++++-
>  1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/power/console.c b/kernel/power/console.c
> index fcdf0e14a47d..832e04bf5439 100644
> --- a/kernel/power/console.c
> +++ b/kernel/power/console.c
> @@ -16,6 +16,7 @@
>  #define SUSPEND_CONSOLE        (MAX_NR_CONSOLES-1)
>
>  static int orig_fgconsole, orig_kmsg;
> +static int vt_switch_done;

Better use bool for this.

Thanks!

>
>  static DEFINE_MUTEX(vt_switch_mutex);
>
> @@ -136,15 +137,19 @@ void pm_prepare_console(void)
>         if (orig_fgconsole < 0)
>                 return;
>
> +       vt_switch_done = 1;
> +
>         orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE);
>         return;
>  }
>
>  void pm_restore_console(void)
>  {
> -       if (!pm_vt_switch())
> +       if (!pm_vt_switch() && !vt_switch_done)
>                 return;
>
> +       vt_switch_done = 0;
> +
>         if (orig_fgconsole >= 0) {
>                 vt_move_to_console(orig_fgconsole, 0);
>                 vt_kmsg_redirect(orig_kmsg);
> --
> 2.20.1
>

[PATCH v3] PM/console: Fix the black screen issue

[tuhaowen]

When the computer enters sleep status without a monitor
connected, the system switches the console to the virtual
terminal tty63(SUSPEND_CONSOLE).

If a monitor is subsequently connected before waking up,
the system skips the required VT restoration process
during wake-up, leaving the console on tty63 instead of
switching back to tty1.

To fix this issue, a global flag vt_switch_done is introduced
to record whether the system has successfully switched to
the suspend console via vt_move_to_console() during suspend.

If the switch was completed, vt_switch_done is set to 1.
Later during resume, this flag is checked to ensure that
the original console is restored properly by calling
vt_move_to_console(orig_fgconsole, 0).

This prevents scenarios where the resume logic skips console
restoration due to incorrect detection of the console state,
especially when a monitor is reconnected before waking up.

Signed-off-by: tuhaowen <tuhaowen@uniontech.com>
---
Changes in v3:
- Changed the type of `vt_switch_done` from `int` to `bool`.
- Link to v2: https://lore.kernel.org/all/20250516084011.29309-1-tuhaowen@uniontech.com
- Link to v1: https://lore.kernel.org/all/20250516034643.22355-1-tuhaowen@uniontech.com
---
 kernel/power/console.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/kernel/power/console.c b/kernel/power/console.c
index fcdf0e14a47d..2a0722038f02 100644
--- a/kernel/power/console.c
+++ b/kernel/power/console.c
@@ -16,6 +16,7 @@
 #define SUSPEND_CONSOLE	(MAX_NR_CONSOLES-1)
 
 static int orig_fgconsole, orig_kmsg;
+static bool vt_switch_done;
 
 static DEFINE_MUTEX(vt_switch_mutex);
 
@@ -136,15 +137,19 @@ void pm_prepare_console(void)
 	if (orig_fgconsole < 0)
 		return;
 
+	vt_switch_done = true;
+
 	orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE);
 	return;
 }
 
 void pm_restore_console(void)
 {
-	if (!pm_vt_switch())
+	if (!pm_vt_switch() && !vt_switch_done)
 		return;
 
+	vt_switch_done = false;
+
 	if (orig_fgconsole >= 0) {
 		vt_move_to_console(orig_fgconsole, 0);
 		vt_kmsg_redirect(orig_kmsg);
-- 
2.20.1

[PATCH v4] PM/console: Fix the black screen issue

[tuhaowen]

When the computer enters sleep status without a monitor
connected, the system switches the console to the virtual
terminal tty63(SUSPEND_CONSOLE).

If a monitor is subsequently connected before waking up,
the system skips the required VT restoration process
during wake-up, leaving the console on tty63 instead of
switching back to tty1.

To fix this issue, a global flag vt_switch_done is introduced
to record whether the system has successfully switched to
the suspend console via vt_move_to_console() during suspend.

If the switch was completed, vt_switch_done is set to 1.
Later during resume, this flag is checked to ensure that
the original console is restored properly by calling
vt_move_to_console(orig_fgconsole, 0).

This prevents scenarios where the resume logic skips console
restoration due to incorrect detection of the console state,
especially when a monitor is reconnected before waking up.

Signed-off-by: tuhaowen <tuhaowen@uniontech.com>
---
Changes in v4:
- Moved `vt_switch_done = false;` below the `if (orig_fgconsole >= 0)`
  block to ensure it is only reset after a console switch has occurred.
- Link to v3: https://lore.kernel.org/all/20250526010854.7834-1-tuhaowen@uniontech.com
- Link to v2: https://lore.kernel.org/all/20250516084011.29309-1-tuhaowen@uniontech.com
- Link to v1: https://lore.kernel.org/all/20250516034643.22355-1-tuhaowen@uniontech.com
---
 kernel/power/console.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/kernel/power/console.c b/kernel/power/console.c
index fcdf0e14a47d..19c48aa5355d 100644
--- a/kernel/power/console.c
+++ b/kernel/power/console.c
@@ -16,6 +16,7 @@
 #define SUSPEND_CONSOLE	(MAX_NR_CONSOLES-1)
 
 static int orig_fgconsole, orig_kmsg;
+static bool vt_switch_done;
 
 static DEFINE_MUTEX(vt_switch_mutex);
 
@@ -136,17 +137,21 @@ void pm_prepare_console(void)
 	if (orig_fgconsole < 0)
 		return;
 
+	vt_switch_done = true;
+
 	orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE);
 	return;
 }
 
 void pm_restore_console(void)
 {
-	if (!pm_vt_switch())
+	if (!pm_vt_switch() && !vt_switch_done)
 		return;
 
 	if (orig_fgconsole >= 0) {
 		vt_move_to_console(orig_fgconsole, 0);
 		vt_kmsg_redirect(orig_kmsg);
 	}
+
+	vt_switch_done = false;
 }
-- 
2.20.1

[PATCH v4 RESEND] PM/console: Fix the black screen issue

[tuhaowen]

When the computer enters sleep status without a monitor
connected, the system switches the console to the virtual
terminal tty63(SUSPEND_CONSOLE).

If a monitor is subsequently connected before waking up,
the system skips the required VT restoration process
during wake-up, leaving the console on tty63 instead of
switching back to tty1.

To fix this issue, a global flag vt_switch_done is introduced
to record whether the system has successfully switched to
the suspend console via vt_move_to_console() during suspend.

If the switch was completed, vt_switch_done is set to 1.
Later during resume, this flag is checked to ensure that
the original console is restored properly by calling
vt_move_to_console(orig_fgconsole, 0).

This prevents scenarios where the resume logic skips console
restoration due to incorrect detection of the console state,
especially when a monitor is reconnected before waking up.

Signed-off-by: tuhaowen <tuhaowen@uniontech.com>
---
Changes in v4:
- Moved `vt_switch_done = false;` below the `if (orig_fgconsole >= 0)`
  block to ensure it is only reset after a console switch has occurred.
- Link to v3: https://lore.kernel.org/all/20250526010854.7834-1-tuhaowen@uniontech.com
- Link to v2: https://lore.kernel.org/all/20250516084011.29309-1-tuhaowen@uniontech.com
- Link to v1: https://lore.kernel.org/all/20250516034643.22355-1-tuhaowen@uniontech.com
---
 kernel/power/console.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/kernel/power/console.c b/kernel/power/console.c
index fcdf0e14a47d..19c48aa5355d 100644
--- a/kernel/power/console.c
+++ b/kernel/power/console.c
@@ -16,6 +16,7 @@
 #define SUSPEND_CONSOLE	(MAX_NR_CONSOLES-1)
 
 static int orig_fgconsole, orig_kmsg;
+static bool vt_switch_done;
 
 static DEFINE_MUTEX(vt_switch_mutex);
 
@@ -136,17 +137,21 @@ void pm_prepare_console(void)
 	if (orig_fgconsole < 0)
 		return;
 
+	vt_switch_done = true;
+
 	orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE);
 	return;
 }
 
 void pm_restore_console(void)
 {
-	if (!pm_vt_switch())
+	if (!pm_vt_switch() && !vt_switch_done)
 		return;
 
 	if (orig_fgconsole >= 0) {
 		vt_move_to_console(orig_fgconsole, 0);
 		vt_kmsg_redirect(orig_kmsg);
 	}
+
+	vt_switch_done = false;
 }
-- 
2.20.1

Re: [PATCH v4] PM/console: Fix the black screen issue

[Rafael J. Wysocki]

On Tue, Jun 3, 2025 at 3:02 AM tuhaowen <tuhaowen@uniontech.com> wrote:
>
> When the computer enters sleep status without a monitor
> connected, the system switches the console to the virtual
> terminal tty63(SUSPEND_CONSOLE).
>
> If a monitor is subsequently connected before waking up,
> the system skips the required VT restoration process
> during wake-up, leaving the console on tty63 instead of
> switching back to tty1.
>
> To fix this issue, a global flag vt_switch_done is introduced
> to record whether the system has successfully switched to
> the suspend console via vt_move_to_console() during suspend.
>
> If the switch was completed, vt_switch_done is set to 1.
> Later during resume, this flag is checked to ensure that
> the original console is restored properly by calling
> vt_move_to_console(orig_fgconsole, 0).
>
> This prevents scenarios where the resume logic skips console
> restoration due to incorrect detection of the console state,
> especially when a monitor is reconnected before waking up.
>
> Signed-off-by: tuhaowen <tuhaowen@uniontech.com>
> ---
> Changes in v4:
> - Moved `vt_switch_done = false;` below the `if (orig_fgconsole >= 0)`
>   block to ensure it is only reset after a console switch has occurred.
> - Link to v3: https://lore.kernel.org/all/20250526010854.7834-1-tuhaowen@uniontech.com
> - Link to v2: https://lore.kernel.org/all/20250516084011.29309-1-tuhaowen@uniontech.com
> - Link to v1: https://lore.kernel.org/all/20250516034643.22355-1-tuhaowen@uniontech.com
> ---
>  kernel/power/console.c | 7 ++++++-
>  1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/power/console.c b/kernel/power/console.c
> index fcdf0e14a47d..19c48aa5355d 100644
> --- a/kernel/power/console.c
> +++ b/kernel/power/console.c
> @@ -16,6 +16,7 @@
>  #define SUSPEND_CONSOLE        (MAX_NR_CONSOLES-1)
>
>  static int orig_fgconsole, orig_kmsg;
> +static bool vt_switch_done;
>
>  static DEFINE_MUTEX(vt_switch_mutex);
>
> @@ -136,17 +137,21 @@ void pm_prepare_console(void)
>         if (orig_fgconsole < 0)
>                 return;
>
> +       vt_switch_done = true;
> +
>         orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE);
>         return;
>  }
>
>  void pm_restore_console(void)
>  {
> -       if (!pm_vt_switch())
> +       if (!pm_vt_switch() && !vt_switch_done)
>                 return;
>
>         if (orig_fgconsole >= 0) {
>                 vt_move_to_console(orig_fgconsole, 0);
>                 vt_kmsg_redirect(orig_kmsg);
>         }
> +
> +       vt_switch_done = false;
>  }
> --

Applied as 6.17 material, thanks!