From mboxrd@z Thu Jan 1 00:00:00 1970 From: quanstro@quanstro.net (erik quanstrom) Date: Tue, 8 Mar 2011 10:52:36 -0500 Subject: [9fans] self modifying code in intel vga bios? In-Reply-To: <20110308152734.GH4355@fangle.proxima.alt.za> References: <80ac49ab88c41d7cef9e96285eeb4c1b@gmx.de> <20110308152734.GH4355@fangle.proxima.alt.za> Message-ID: Topicbox-Message-UUID: b90a31e0-ead6-11e9-9d60-3106f5b1d025 On Tue Mar 8 10:30:25 EST 2011, lucio at proxima.alt.za wrote: > On Tue, Mar 08, 2011 at 07:21:50AM -0800, Paul Lalonde wrote: > > The last time I poked at one of these self-modifying bits they were really > > just jitting a blit loop, in place. Drops register pressure a little bit, > > which has always been a bit of an issue in x86 land. > > > A bankrupt CPU architecture. I guess the day they all do it, it will > be decreed "the right way". In the meantime, I guess we can still > shop around. just like ata is no longer a physical spec, x86 is no longer a architecture. it is simply an instruction set. i can think of many different architectures that implement the x86 instruction set you can buy today, for example: bonnell (atom), athlon64, intel 5000 (conroe and friends), nehalem, sandy bridge, via c7, etc. the same thing goes for arm, and power. you can complain that it's hard to decode x86, but this is also a strength: x86 has good code density. imagine the chaos that we'd have if the cpu vendors released a new instruction set for each new architecture? it may be that instruction sets aren't very important any longer. - erik