Re: [PATCH] x86: remove NexGen support (fwd)

Previous thread: [git pull] x86 fixes by Ingo Molnar on Wednesday, April 30, 2008 - 5:24 pm. (1 message)

Next thread: [PATCH] ia64 TS_RESTORE_SIGMASK by Roland McGrath on Wednesday, April 30, 2008 - 5:40 pm. (1 message)
To: <dmitri.vorobiev@...>, <mingo@...>, <tglx@...>, <hpa@...>, <linux-kernel@...>
Date: Wednesday, April 30, 2008 - 5:34 pm

Are you sure it never shipped? Wikipedia seems to suggest it did ship.

http://www.cpu-info.com/index2.php?mainid=Nx586&page=5

...FPU for nexgen was never shipped, that implies that CPU did ship.

Pavel

--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
--

To: Pavel Machek <pavel@...>
Cc: <mingo@...>, <tglx@...>, <hpa@...>, <linux-kernel@...>
Date: Wednesday, April 30, 2008 - 6:15 pm

The main reason for removing the NexGen support was that the latter
had effectively been broken for G-d knows how many years. As a person
somehow involved into embedded development, I simply could not resist
to trim a few dozen bytes of dead binary code off vmlinux. Besides,
the change is trivially revertable (just checked that), and in case
that somebody owns a NexGen box and complains, that can be addressed
in a matter of minutes.

--

To: Pavel Machek <pavel@...>
Cc: <dmitri.vorobiev@...>, <mingo@...>, <tglx@...>, <hpa@...>, <linux-kernel@...>
Date: Wednesday, April 30, 2008 - 5:45 pm

Exactly. Nx686 is the one that was turned into AMD K6.

--
Ondrej Zary
--

To: Pavel Machek <pavel@...>
Cc: <dmitri.vorobiev@...>, <mingo@...>, <tglx@...>, <linux-kernel@...>
Date: Wednesday, April 30, 2008 - 5:41 pm

Doesn't mean it shipped in anything other than sample or sample-like
quantities.

-hpa
--

To: H. Peter Anvin <hpa@...>
Cc: <dmitri.vorobiev@...>, <mingo@...>, <tglx@...>, kernel list <linux-kernel@...>
Date: Wednesday, April 30, 2008 - 5:51 pm

http://www.cpu-info.com/index2.php?mainid=Nx586&page=7

I do not know how many were really shipped, and yes, they seem to be
pretty uncommon, but it sounds like more than samples. Actually, some
seem to be for sale on ebay ;-).

Pavel

...
The first NxPCI AT motherboards were shipped at the beginning of
September 1995 after they were first announced in fall
1994. Motherboards featured support for EDO and Fast Page memory up to
768MB DRAM, up to four IDE devices, an on-board floppy controller, two
serial ports and one parallel. It had four ISA slots and three PCI
slots. Prices for a complete configuration, motherboard, CPU and
heatsink were at $339 in combination with the Nx586 P90 and $439 with
a P100.

...
With a big player like Intel, getting a foothold in the processor
business is hard. To gain the buyers attention continues price drops
had to be made. By the time the Nx586 was actually for sale, the entry
Nx586-P75 cost $404, the P80 $477, $539 for the P90 and the P100 was
priced at $777. With new price cuts in mind Nexgen sought ways to
reduce costs. One way was to make production of the Nx586 more
efficient, fit more die's on a wafer. A die shrink was announced in
the first half of 1995.
...

Nexgen discontinued its Nx586 P75 and P80 around August 1995. Demand
had shifted to higher performance parts and NexGen needed the capacity
for the imminent rollout of its new P120 and P133. Prices were once
again reduced from $299 to $245 for the P90 and from $399 to $340 for
the P100. In the mean time Nexgen was still trying to close additional
deals for the manufacturing of motherboards and chipsets. To date
motherboards were only manufactured by Alaris and chipsets by Fujitsu.

For over a year now Nexgen was shipping FPU less Nx586's, but this was
about to end. In November 1995 Nexgen announced the Nx586 with
integrated FPU, the Nx586 Pf100 and the faster Nx586 Pf120. The CPU
and FPU would packaged together on a Multichip Module and manufactured
by IBM. The Nx5...

To: Pavel Machek <pavel@...>
Cc: <dmitri.vorobiev@...>, <mingo@...>, <tglx@...>, kernel list <linux-kernel@...>
Date: Wednesday, April 30, 2008 - 5:56 pm

Cool. If someone is willing to pick one up and maintain this fairly
trivial amount of code we can resurrect it; given that the code had been
broken for several releases seem to indicate noone really cares.

(Also note that the removal will not prevent the kernel from actually
running on these chips; in fact, it might actually make them work.)

-hpa
--

Previous thread: [git pull] x86 fixes by Ingo Molnar on Wednesday, April 30, 2008 - 5:24 pm. (1 message)

Next thread: [PATCH] ia64 TS_RESTORE_SIGMASK by Roland McGrath on Wednesday, April 30, 2008 - 5:40 pm. (1 message)