[SeaBIOS] [PATCH] Ctrl-Break support (INT 1B, etc)

[Kevin O'Connor]

On Thu, Jul 21, 2011 at 02:49:08AM +0300, Nikolay Nikolov wrote:
> Hi, all,
> 
> I made a patch for proper Ctrl-Break handling. This is useful for
> MS-DOS or compatible operating systems like FreeDOS.
> 
> When Ctrl-Break is detected, it clears the keyboard buffer, sets the
> Ctrl-Break flag at [0040h:0071h], invokes int 1Bh and then adds
> 0000h to the keyboard buffer. If a non-enhanced keyboard is present,
> Ctrl-Scroll Lock works just like Ctrl-Break. This seems to be the
> "standard" BIOS behaviour, at least on the PCs I tested.
> 
> The patch was tested with MS-DOS 6.22, running under qemu.

Thanks.  Please see my comments below.

> 
> I'm new to git (and this is my first SeaBIOS patch ever), so I'm not
> sure what's the preferred way to submit patches. I just did a "git
> diff" and attached the result. Sorry if that's not the best way :)

Best way is to run "git format-patch", but "git diff" is also fine.
Please do include a "Signed-off-by" line in your commit message or
email message though.

> diff --git a/src/kbd.c b/src/kbd.c
> index 1977c5d..e85f92d 100644
> --- a/src/kbd.c
> +++ b/src/kbd.c
> @@ -406,7 +406,6 @@ __process_key(u8 scancode)
>      }
>  
>      // XXX - PrtScr should cause int 0x05
> -    // XXX - Ctrl+Break should cause int 0x1B
>      // XXX - SysReq should cause int 0x15/0x85
>  
>      switch (scancode) {
> @@ -474,12 +473,31 @@ __process_key(u8 scancode)
>          flags1 &= ~KF1_NUM;
>          break;
>  
> -    case 0x46: /* Scroll Lock press */
> -        flags1 |= KF1_SCROLL;
> -        flags0 ^= KF0_SCROLLACTIVE;
> +    case 0x46: /* Scroll Lock or Ctrl-Break press */
> +        if (flags2 & KF2_LAST_E0
> +            || (!(flags2 & KF2_101KBD) && (flags0 & KF0_CTRLACTIVE))) {
> +            // Ctrl-Break press
> +            flags2 &= ~KF2_LAST_E0;
> +            SET_BDA(kbd_flag2, flags2);
> +            SET_BDA(break_flag, 0x80);
> +            SET_BDA(kbd_buf_tail, GET_BDA(kbd_buf_head));
> +            asm volatile ("int $0x1B":::"memory");

Can this use call16_simpint() instead?  If not, calling an interrupt
should probably backup/restore the register state.

> +            enqueue_key(0, 0);

RBIL implies the 0000 should be added to the keyboard buffer before
the int 0x1B - though I'm not really familiar enough to say which is
really correct.

-Kevin