On Sun, 28 Sep 2008 14:54:15 +0200 Michal Januszewski <[email protected]> wrote:
> Currently, it is possible to set a graphics VESA mode at boot time via
> the vga= parameter even when no framebuffer driver supporting this is
> configured. This could lead to the system booting with a black screen,
> without a usable console.
>
> Fix this problem by only allowing to set graphics modes at boot time
> if a supporting framebuffer driver is configured.
>
> Signed-off-by: Michal Januszewski <[email protected]>
> Acked-by: Krzysztof Helt <[email protected]>
> ---
> This version of the patch replaces the FB_SUPPORTS_BOOT_VESA macro from
> screen-info.h with a config option in drivers/video/Kconfig.
>
> diff --git a/arch/x86/boot/video-vesa.c b/arch/x86/boot/video-vesa.c
> index 401ad99..47e5080 100644
> --- a/arch/x86/boot/video-vesa.c
> +++ b/arch/x86/boot/video-vesa.c
> @@ -88,14 +88,11 @@ static int vesa_probe(void)
> (vminfo.memory_layout == 4 ||
> vminfo.memory_layout == 6) &&
> vminfo.memory_planes == 1) {
> -#ifdef CONFIG_FB
> +#ifdef CONFIG_FB_BOOT_VESA_SUPPORT
> /* Graphics mode, color, linear frame buffer
> supported. Only register the mode if
> if framebuffer is configured, however,
> - otherwise the user will be left without a screen.
> - We don't require CONFIG_FB_VESA, however, since
> - some of the other framebuffer drivers can use
> - this mode-setting, too. */
> + otherwise the user will be left without a screen. */
> mi = GET_HEAP(struct mode_info, 1);
> mi->mode = mode + VIDEO_FIRST_VESA;
> mi->depth = vminfo.bpp;
> @@ -133,10 +130,12 @@ static int vesa_set_mode(struct mode_info *mode)
> if ((vminfo.mode_attr & 0x15) == 0x05) {
> /* It's a supported text mode */
> is_graphic = 0;
> +#ifdef CONFIG_FB_BOOT_VESA_SUPPORT
> } else if ((vminfo.mode_attr & 0x99) == 0x99) {
> /* It's a graphics mode with linear frame buffer */
> is_graphic = 1;
> vesa_mode |= 0x4000; /* Request linear frame buffer */
> +#endif
> } else {
> return -1; /* Invalid mode */
> }
> diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
> index 70d135e..651249d 100644
> --- a/drivers/video/Kconfig
> +++ b/drivers/video/Kconfig
> @@ -76,6 +76,14 @@ config FB_DDC
> select I2C
> default n
>
> +config FB_BOOT_VESA_SUPPORT
> + bool
> + depends on FB
> + default n
> + ---help---
> + If true, at least one selected framebuffer driver can take advantage
> + of VESA video modes set at an early boot stage via the vga= parameter.
> +
> config FB_CFB_FILLRECT
> tristate
> depends on FB
> @@ -679,6 +687,7 @@ config FB_VESA
> select FB_CFB_COPYAREA
> select FB_CFB_IMAGEBLIT
> select VIDEO_SELECT
> + select FB_BOOT_VESA_SUPPORT
> help
> This is the frame buffer device driver for generic VESA 2.0
> compliant graphic cards. The older VESA 1.2 cards are not supported.
> @@ -702,6 +711,7 @@ config FB_IMAC
> select FB_CFB_FILLRECT
> select FB_CFB_COPYAREA
> select FB_CFB_IMAGEBLIT
> + select FB_BOOT_VESA_SUPPORT
> help
> This is the frame buffer device driver for the Intel-based Macintosh
>
> @@ -1124,6 +1134,7 @@ config FB_INTEL
> select FB_CFB_FILLRECT
> select FB_CFB_COPYAREA
> select FB_CFB_IMAGEBLIT
> + select FB_BOOT_VESA_SUPPORT
> help
> This driver supports the on-board graphics built in to the Intel
> 830M/845G/852GM/855GM/865G/915G/915GM/945G/945GM/965G/965GM chipsets.
> @@ -1476,6 +1487,7 @@ config FB_SIS
> select FB_CFB_FILLRECT
> select FB_CFB_COPYAREA
> select FB_CFB_IMAGEBLIT
> + select FB_BOOT_VESA_SUPPORT
> help
> This is the frame buffer device driver for the SiS 300, 315, 330
> and 340 series as well as XGI V3XT, V5, V8, Z7 graphics chipsets.
-mm's efifb-imacfb-consolidation-hardware-support.patch removed FB_IMAC
and moved it into FB_EDD. But your patch changes FB_IMAC and not
FB_EDD, so I don't know what needs to be done here.
On Fri, Oct 3, 2008 at 08:40, Andrew Morton <[email protected]> wrote:
> -mm's efifb-imacfb-consolidation-hardware-support.patch removed FB_IMAC
> and moved it into FB_EDD. But your patch changes FB_IMAC and not
> FB_EDD, so I don't know what needs to be done here.
I had a look at that patch and it looks like simply dropping the FB_IMAC
part from my patch is correct (i.e. the version of
fbdev-ignore-vesa-modes-if-framebuffer-does-not-support-them
that is currently in -mm is fine).
Best regards,
--
Michal Januszewski, Gentoo Linux Developer
http://people.gentoo.org/spock