AMD/Linux vs Intel/Microsoft

  • Thread starter Thread starter E
  • Start date Start date
Ken said:
The article hypothesises that because NX breaks some apps, it may
soften up customers for DRM that breaks rather more. You are free
to buy that theory if you like, but I don't. (NX doesn't break
much, for one thing.)

Sidenote: The article claims that something like the NX bit is less
necessary for Linux or one of the BSDs. I can't follow that. Linux does use
PROT_EXEC, when available since quite some time (on non-x86 processors back
then). It is possible to patch Linux so that the stack by default is
non-executable. At least one of the BSDs makes this the standard behaviour.
All C code is plagued with buffer overflows, and having a non-executable
stack or heap is only a small improvement. Compromising the stack can still
allow exploits. You can change the return address to jump to some library
function, and an obvious entry point would be a function to mark your code
as executable.
 
| On Mon, 12 Jan 2004 22:20:14 +0000, Kevin Lawton wrote:
|
|| | On Sat, 10 Jan 2004 14:39:02 +1300, ~misfit~ wrote:
|| |
|| || chrisv wrote:
|| |||
|| ||| I'm the customer, ATI. I'm the guy with the money that YOU
|| want. ||| Make some effort to help me out!
|| ||
|| || **** 'em. nVidia make some good cards and drivers.
|| |
|| | Yeah, I may have to go that way with my next card. I've never
|| liked | nVidia or it's products all that much, but if their card
|| has better | Linux support...
|| |
|| | These companies must know that there's a lot people who are
|| going to | be checking-out Linux, and they should be ready. Tux
|| Racer @ 1fps is | NO GOOD! 8)
||
|| Exactly - so vote with your wallet ! ;-)
|| This brings me back to using Matrox video cards in systems which
|| run a non-Micro$oft OS - often Linux, maybe something else (BeOS ?).
|| The high-spec nVidia and ATI cards show their best performance
|| ratings used on Windoze systems - and the battle for top position
|| seems to change monthly. This isn't much of an indication of how the
|| card might perform under, say, Linux as the system for displaying
|| complex 3D graphics is a bit different. Those top-ranking cards rely
|| on Windoze drivers which have been carefully written to exploit
|| their features. In a Linux box they might not seem quite so clever.
|
| The Nvidia supplied drivers for Linux are quite good. I don't have any
| recent references but I seem to remember that some of their cards work
| *better* in Linux than in Windows.
|
|| Matrox have the good sense, decency and
| end-user committment to offer
|| Linux drivers for most of their cards - that is one of the reason
|| why I use them and recommend them. The other two reasons are image
|| quality and the ability to drive two monitors (dual head). I even
|| have a system running BeOS giving a dual-head display from its
|| Matrox card (G400).
|
| Nvidia supplies drivers for Linux for ALL their currently shipping x86
| chipsets.
|
| I have systems with Nvidia graphics that are dual head (dvi and vga
| outs
| to a flat panel and tube monitor respectively). I also have systems
| with both an Nvidia card and a Matrox card. My dual head Nvidia
| setup... even though using one card still does OpenGL perfectly. I
| get around 2500FPS
| from an older GeForce2 GTS card using the Nvidia drivers.
|
|| To me, these things are more important than being able to show
|| about 70
|| fps in the latest games. My eyes posses a feature called
|| 'persistance of vision' which prevents such speed being necessary -
|| and ceratinly not worth paying for.
|
| One can get a very decent card from Nvidia for under US$100 and one
| that
| is decent for under US$50. While it would be nice if we could get the
| specs from Nvidia or they would completely open-source their drivers I
| have no problem with the current ones. The stock XFree86 "nv" driver
| works just fine in 2D.

So, if you are running an 'alternative' Op System like Linux or BeOS, then
you can get decent drivers to rn your Matrox, nVidia or S3 Savage card no
problem.
Ultimately, won't this just leave ATI 'out in the cold' while
non-Mocro$oft users opt for better-supported hardware ?
Kevin.

|
| -DU-...etc...
 
["Followup-To:" header set to comp.os.linux.advocacy.]
Datagram from THX1138 incoming on netlink socket
If I find it later there is an article on AMD's site that states that
NX IS for DRM...It's a sales pitch for corp types that was link
public.

Either I seriously misunderstand what NX pagetable bit is, or the article
writter has no clue.

As I understand it, the if page has NX bit set, trying to execute code
from that page causes page fault.

In x86, one can simulate that behaviour either by reducing code segment
limit, or playing nasty games with caches, so they are inconsistent in
order to make data references succeed, but code references fail.

-Ilari
--
Nothing's truly free (in price) in this world... There are only
approximations of it. Free (in price) is unattainable idealization.
-- Ilari Liusvaara
Linux LK_Perkele_IV9 2.4.23-selinux1 #2 Mon Jan 5 20:12:55 EET 2004 i686 unknown
9:13pm up 8 days, 0:49, 5 users, load average: 0.06, 0.04, 0.07
 
["Followup-To:" header set to comp.os.linux.advocacy.]
Datagram from Ken Hagan incoming on netlink socket
The article hypothesises that because NX breaks some apps, it may
soften up customers for DRM that breaks rather more. You are free
to buy that theory if you like, but I don't. (NX doesn't break
much, for one thing.)

If OS is done properly, the only apps that get broken by NX are
broken already.

-Ilari
 
David said:
.... snip ...

One can get a very decent card from Nvidia for under US$100 and
one that is decent for under US$50. While it would be nice if we
could get the specs from Nvidia or they would completely
open-source their drivers I have no problem with the current
ones. The stock XFree86 "nv" driver works just fine in 2D.

Conceded that that sort of thing handles 98% of most needs today.
However I miss the ability to build cheap systems from commercial
hardware from the ground up. This enables one to correct faults,
avoid copyrights, etc.

As an example 20 years ago I could buy a complete Kaypro, replace
the EPROM, and be running entirely on my own code. No disk drives
were necessary for some applications. Nothing was hidden - every
component was fully documented and replaceable. I can't do the
equivalent today. This makes such systems unsuitable for critical
applications, such as medicine, aeronautics, voting, even
financial.
 
So, if you are running an 'alternative' Op System like Linux or BeOS, then
you can get decent drivers to rn your Matrox, nVidia or S3 Savage card no
problem.
Ultimately, won't this just leave ATI 'out in the cold' while
non-Mocro$oft users opt for better-supported hardware ?

Leaving aside that many people do use ATI's Linux drivers
successfully:

A BOTE calculation with reasonable assumptions about number of card
sales dependent on some particular alternate OS, revenue to the OEM, and
size and cost of a graphics driver engineering and support team focusing
on that OS, what pops out the back end is that oficially supporting the
major flavors of Linux is in the ballpark of a break-even proposition,
while supporting any other alternate OS is done at a loss. The BeOS,
BSD, etc. markets are too small to be relevant to companies shipping
tens-hundreds of millions of chipsets/year.

Jon
 
Still, Don't you find it interesting that MS was able to come up with
a 64 bit implimentation of Windows that happens to only run on
Itanium?

No, no I don't. Furthermore, I don't see why anyone SHOULD find it at
all abnormal that Windows for IA64 arrived before Windows for AMD64.
The Itanium was released three years ago and the instruction set was
documented long before then. AMD64 hasn't even been here a year.

What's more, Windows for IA64 was late as well! When the Itanium came
out (3 years late), it was still 6-9 months before Microsoft had a
final version of Windows for it. The first systems shipped either
with Linux or with a beta version of WinXP. Sound familiar?
This is very true but you know you are being obtuse if you think that
is what is driving this.

Intel ( AMD ) and especially MS don't give a wet fart about server
security because if they did they would have had at least 10 years to
fix the problem. They only care about money and security wasn't part
of the deal.

Of course all they care about is money, they're a company! But the
fact of the matter is that security problems are costing Microsoft,
Intel and AMD sales, people are choosing IBM AIX and Sun Solaris
instead of Intel/AMD on Windows because of security problems.
Security problems cost money, and money talks.

Of course there is the other issue I mentioned about MS wanting to
bring the media companies onto their side as well and drive sales that
way. It's a two-pronged strategy that MS is bundling together in one.
Microsoft talks all about the security aspects of it publicly and
doesn't say much about the DRM stuff, while the /. hordes yell and
scream about DRM and totally ignore the security aspect of it (all the
while yelling and screaming about MS' poor security track record).
http://www.theregister.co.uk/content/4/33729.html
MS to intro hardware-linked security for AMD64, Itanium, future CPUs
By John Lettice

Good god that article is TERRIBLE! Even by The Registers standards
it's bad! They even managed to get a clueless (or possibly just
incorrect) quote to back up their own cluelessness.

"The 32-bit version of Windows currently leverages the NX processor
feature, as defined by the AMD64 Architecture Programmer's Manual." So
actually, it's not being introduced to the mass market with SP2 - it's
here already for AMD64 platforms."

If they had bothered to do a good 10 seconds worth of research they
would have discovered that the NX bit is NOT SUPPORTED IN 32-BIT MODE!
AMD publishes complete documentation for their AMD64 instruction set
and they flat out state:

"No Execute (NX) Bit. Bit 63. This bit is present in the translation
table entries defined for PAE paging, with the exception that the
legacy-mode PDPE does not contain this bit. This bit is not supported
by non-PAE paging."

I like how they complain about how the NX bit doesn't seem to be a
must-have for Linux... Yet again, 10 seconds worth of research
probably would have told them that it's already used in Linux for
AMD64! I believe that it's turned off by default because some poor
written programs break, but it's certainly in there.
If I find it later there is an article on AMD's site that states that
NX IS for DRM...It's a sales pitch for corp types that was link
public.

Please do, I'd be interested in seeing how they managed that, because
there is NO WAY to get any sort or DRM using this feature. At the
very best it might be of a tiny bit of help in making existing DRM a
tiny bit trickier to break, but that would be trivial to get around.

I would highly recommend that you read AMD's documentation on what
this bit does. You can find it at page 175 of Volume 2 of their AMD64
Programming Guides (System Programming). Here's a link to the pdf:

http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/24593.pdf
 
Leaving aside that many people do use ATI's Linux drivers
successfully:

A BOTE calculation with reasonable assumptions about number of card
sales dependent on some particular alternate OS, revenue to the OEM, and
size and cost of a graphics driver engineering and support team focusing
on that OS, what pops out the back end is that oficially supporting the
major flavors of Linux is in the ballpark of a break-even proposition,
while supporting any other alternate OS is done at a loss. The BeOS,
BSD, etc. markets are too small to be relevant to companies shipping
tens-hundreds of millions of chipsets/year.

I hear that Linux is surpassing (or pretty darn close) Apple in desktop
share. Apple seems to be split about 50-50 between Nvidia and ATI graphics
sets. I don't know how that factors in to BOTE calculations but it MUST be
putting Linux in a better position than it was.

Another factor that I found surprising the first time I heard about it is
that, for Nvidia at least, they are working closely with the movie
industry in providing the drivers they need for their really high end
cards. I heard a talk by a group of people from Dreamworks at Linux World
Expo-NYC (I think it was 2 years ago). There seems to be a big push in
Hollywood SFX houses not just for render farms but for Linux on the
desktop of their artists. They tend to buy the really high end graphics
stuff like Nvidias Fire GL. They buy them every time there is a new one
and they buy them by the hundreds if not thousands. How many gamers buy a
US$500+ Fire GL card? Probably not many. These SFX houses have a certain
amount of suction with the chipset makers since they are buying the high
margin stuff. I think the reason we have seen decent drivers from Nvidia
is largely thanks to the SFX market. I am not sure where ATI fits into
that picture... though I suppose it must try.

-DU-...etc...
 
In comp.arch Tom Morris said:
That may be true of the vendors that you deal with, but most application
vendors take testing pretty seriously because it has a direct effect on
customer satisfaction (revenue) and support costs (expense).

The commercial vendors that I'm familiar with would (and did) consider
a cross compilation environment with no opportunity for testing a
non-starter.

Heck, the CAD vendors only certify and support specific versions
of the graphics card drivers!

Once you read the changelogs you will fast get to the point where you
say "oh, I see why". GFX drivers seem to be fast moving into complexity
levels that the companies have trouble handling.
 
Back
Top