[haiku-development] Re: [GSoC] x86-64 port

  • From: Duane Bailey <bailey.d.r@xxxxxxxxx>
  • To: haiku-development@xxxxxxxxxxxxx
  • Date: Thu, 8 Apr 2010 14:14:58 -0400

On Apr 8, 2010, at 6:00 AM, Simon Taylor wrote:

> Hi Duane,
> Sorry in advance for pedantry...
>> [...] the Haiku OS [...]
> It's just "Haiku".

Duly noted.

> x86_64 would be very cool to have. Although there aren't many Haiku apps
> yet that need >4 GB of memory one of the other things about x86_64 mode
> is having access to more registers for SSE code which I expect would be
> useful for some of the video codecs and app_server drawing code.

> (^ from wikipedia: "AMD's implementation of SSE2 on the AMD64 (x86-64)
> platform includes an additional eight registers, doubling the total
> number to 16 (XMM0 through XMM15). These additional registers are only
> visible when running in 64-bit mode. Intel adopted these additional
> registers as part of their support for x86-64 architecture (or in
> Intel's parlance, "Intel 64") in 2004.")

This comes up from not posting my abstract as well; the features of x86-64 may 
not seem enough to justify initiating a large port, but together, they make a 
compelling environment. Along with the added memory, the extra GPRs allow the 
compiler some more room in which is can breathe, slightly increased security 
tools via the NX bit, etc. Most importantly, I just want to initiate the port 
and provide a platform on which future projects, even future GSoC projects, can 
build and extend the kernel and start on porting the userland.

Also, future processor features (virtualization, SSE5+, other architectures a 
la larrabee's descendents, etc) may depend on long mode. I'd rather port to an 
architecture that not everybody sees the need for rather than people all of a 
sudden needing an architecture that Haiku has no support for.

Anyway, cheers! Thanks for the input. Application has been clarified.


Other related posts: