svn commit: r336805 - in stable/11: share/man/man4 sys/dev/vt/hw/vga

[Konstantin Belousov]

On Sat, Jul 28, 2018 at 07:37:01AM +0000, Eitan Adler wrote:
> Author: eadler
> Date: Sat Jul 28 07:37:01 2018
> New Revision: 336805
> URL: https://svnweb.freebsd.org/changeset/base/336805
> 
> Log:
>   MFC r335629:
>   
>   vt: add option to ignore NO_VGA flag in ACPI
>   
>   To workaround buggy firmware that sets this flag when there's actually
>   a VGA present.
>   
>   Ref D16003
>   PR:		229235
> 
> Modified:
>   stable/11/share/man/man4/vt.4
>   stable/11/sys/dev/vt/hw/vga/vt_vga.c
> Directory Properties:
>   stable/11/   (props changed)
> 
> Modified: stable/11/share/man/man4/vt.4
> ==============================================================================
> --- stable/11/share/man/man4/vt.4	Sat Jul 28 06:46:10 2018	(r336804)
> +++ stable/11/share/man/man4/vt.4	Sat Jul 28 07:37:01 2018	(r336805)
> @@ -44,6 +44,7 @@
>  In
>  .Xr loader.conf 5 :
>  .Cd hw.vga.textmode=1
> +.Cd hw.vga.acpi_ignore_no_vga=1
>  .Cd kern.vty=vt
>  .Cd kern.vt.color.<colornum>.rgb="<colorspec>"
>  .Cd kern.vt.fb.default_mode="<X>x<Y>"
> @@ -196,6 +197,10 @@ prompt or in
>  Set to 1 to use virtual terminals in text mode instead of graphics mode.
>  Features that require graphics mode, like loadable fonts, will be
>  disabled.
> +.It Va hw.vga.acpi_ignore_no_vga
> +Set to 1 to force the usage of the VGA driver regardless of whether
> +ACPI IAPC_BOOT_ARCH signals no VGA support.
> +Can be used to workaround firmware bugs in the ACPI tables.
>  .It Va kern.vty
>  Set this value to
>  .Ql vt
> 
> Modified: stable/11/sys/dev/vt/hw/vga/vt_vga.c
> ==============================================================================
> --- stable/11/sys/dev/vt/hw/vga/vt_vga.c	Sat Jul 28 06:46:10 2018	(r336804)
> +++ stable/11/sys/dev/vt/hw/vga/vt_vga.c	Sat Jul 28 07:37:01 2018	(r336805)
> @@ -1211,6 +1211,12 @@ vga_acpi_disabled(void)
>  	ACPI_TABLE_FADT *fadt;
>  	vm_paddr_t physaddr;
>  	uint16_t flags;
> +	int ignore;
> +
> +	TUNABLE_INT_FETCH("hw.vga.acpi_ignore_no_vga", &ignore);
> +
> +	if (ignore)
> +	    return (false);
If the tunable is not set, this code evaluates the stack garbage.
Why did you merged this without asking the author and not looking
at the followups ?

>  
>  	physaddr = acpi_find_table(ACPI_SIG_FADT);
>  	if (physaddr == 0)