From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with archive (Exim 4.43) id 1Knc5a-0002Bv-Tx for mharc-grub-devel@gnu.org; Wed, 08 Oct 2008 12:40:42 -0400 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1Knc5Y-0002B4-GD for grub-devel@gnu.org; Wed, 08 Oct 2008 12:40:40 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1Knc5W-00029p-JR for grub-devel@gnu.org; Wed, 08 Oct 2008 12:40:39 -0400 Received: from [199.232.76.173] (port=56480 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1Knc5W-00029b-Dq for grub-devel@gnu.org; Wed, 08 Oct 2008 12:40:38 -0400 Received: from mta-out.inet.fi ([195.156.147.13]:48501 helo=kirsi1.inet.fi) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1Knc5W-0002ky-HU for grub-devel@gnu.org; Wed, 08 Oct 2008 12:40:38 -0400 Received: from [127.0.0.1] (84.248.105.254) by kirsi1.inet.fi (8.5.014) id 48DA2F8900C5C349 for grub-devel@gnu.org; Wed, 8 Oct 2008 19:40:34 +0300 Message-ID: <48ECE286.9090701@nic.fi> Date: Wed, 08 Oct 2008 19:40:38 +0300 From: =?UTF-8?B?VmVzYSBKw6TDpHNrZWzDpGluZW4=?= User-Agent: Thunderbird 2.0.0.17 (Windows/20080914) MIME-Version: 1.0 To: The development of GRUB 2 References: <1460730204.89441223439151710.JavaMail.root@aczmb1> In-Reply-To: <1460730204.89441223439151710.JavaMail.root@aczmb1> X-Enigmail-Version: 0.95.7 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-detected-operating-system: by monty-python.gnu.org: GNU/Linux 2.6 (newer, 3) Subject: Re: No scrolling for long input lines X-BeenThere: grub-devel@gnu.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: The development of GRUB 2 List-Id: The development of GRUB 2 List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Oct 2008 16:40:40 -0000 Andy Goth wrote: > "Andy Goth" wrote: >> I'll test the current SVN this Friday, or perhaps Thursday night at >> the earliest. > > Actually I did spend time tonight investigating. I needed to get the latest SVN for another reason (about which I have questions I'll ask in a few days), so I went ahead and looked into my scrolling input problem. > > Yup, still there. Today's SVN is affected. > > The problem is present when actually booting using GRUB (ordinary VGA text mode), but it all works fine in grub-emu. > > Wait, I take that back. In grub-emu, typing very long lines (that wrap and *cause the screen to scroll*) results in the text wrapping at column 80 instead of column 79, so that there is *not* a white space character in the rightmost column. Pressing Ctrl-U from such a line will only erase only the bottom line of text, except for the first six characters (which presumably correspond to "grub> "). > > Something's flaky! However, this grub-emu stuff might be unrelated to the problem I've been having with the actual boot loader. > > I also add that in grub-emu, the carriage doesn't return after printing an error message, so I get a stairstep effect. With such an indented prompt, issuing a command like "ls" causes output to be printed to the *left* of the prompt. > > Enough about grub-emu. In the actual boot loader, I noticed that the pattern seems to be: it will scroll the screen if inserting text will cause the last character to overflow the right edge of the screen. That sounds like the way it should be, right? The problem is that only currently visible characters appear to be checked for overflow. This doesn't include the character being typed or characters that previously have failed to scroll onscreen. > > Test cases: > > 1. Near the top of the screen, type a bunch of text and overflow the edge. Works fine. > 2. Hold down enter until the screen starts scrolling. Works fine. > 3. Type a bunch of text and overflow the edge. Fails to scroll! > 4. Use the left arrow key one or two times and type text. Fails to scroll! > 5. Use the left arrow key until the cursor is onscreen, and type text. Everything scrolls into view. > > I'd absolutely love to debug all this and provide patches, but I really don't have time right now. In fact I didn't have time to do the research I did; I should have been sleeping. :^( This seems to be BIOS issue in a way. In startup.S grub_console_real_putchar there is quite good comment about what are BIOS limitations. Choices are to modify this code here, or make better bios console terminal.