bug-hurd
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

console server and kernel prints


From: Marcus Brinkmann
Subject: console server and kernel prints
Date: Fri, 8 Feb 2002 19:20:16 +0100
User-agent: Mutt/1.3.27i

Hi,

when we use a true user space console which accesses the gva graphic card at
the hardware level (I/O ports, mapped memory), we will face some interesting
races with the kernel.  I am not so worried about the normal operation
(setting cursor etc) although it would be neat to avoid races there, too,
but also about other, less common things.  For example, when I load a new
font, I have to stop the sequencer and map another page into the memory
window at 0x0a0000, and during that time a kernel printf could corrupt
fonts.  I don't think anything really serious can happen (like loosing the
state of the video card) as long as all the kernel does is poking into the
video memory and accessing the registers for cursor positioning. 

But in the future, when we will use a framebuffer, we will face a similar
problem in that the kernel messages will not appear but as dust in the
background picture or so.

I am not sure we can do anything reasonable about it.  Disabling the kernel
messages would be the cure for corruption, but it might make you miss out on
important messages from the kernel.  Maybe the kernel could check the
register settings before writing, if they appear to be sane (having page0
and page1 mapped at the right position in odd/even mode).  Not sure what it
could do if they are not sane, though.

Why doesn't every PC have a serial console for kernel messages built in?

Thanks,
Marcus

-- 
`Rhubarb is no Egyptian god.' Debian http://www.debian.org brinkmd@debian.org
Marcus Brinkmann              GNU    http://www.gnu.org    marcus@gnu.org
Marcus.Brinkmann@ruhr-uni-bochum.de
http://www.marcus-brinkmann.de



reply via email to

[Prev in Thread] Current Thread [Next in Thread]