Linux and NexGen/Cyrix chips

Linux and NexGen/Cyrix chips

Post by Arthur Blai » Tue, 30 Jan 1996 04:00:00



Has anyone tried Linux on the new Cyrix 6x86 chip or
any of the NexGen chips? Did it work?
Art.
 
 
 

Linux and NexGen/Cyrix chips

Post by Michael B. Mart » Wed, 31 Jan 1996 04:00:00


: Has anyone tried Linux on the new Cyrix 6x86 chip or
: any of the NexGen chips? Did it work?

According to Cyrix, it should.  Their Web page (http://www.cyrix.com)
has a "compatibility" section which listed a large number of OS's they
say work fine on their chip.  Linux was on the list.

-Michael

 
 
 

Linux and NexGen/Cyrix chips

Post by Chris Maso » Wed, 31 Jan 1996 04:00:00


I get it to boot on my nexgen P100 motherboard, which I also run
Windows 95 on.  But my SCSI cdrom isn't recognized.  Like a fool,
I bought an unsupported AHA-2920 SCSI adapter.

It sees that I have a PCI device (the mother board) and the adapter
but doesn't recognize that it is a SCSI host...

But I like my nexgen motherboard.  It is much faster than my old
386 25 mhz board...

--
Chris Mason                                                              

"The Unknown COBOL Programmer"                                          

The opinions expressed are mine, not my Employers.                      


 
 
 

Linux and NexGen/Cyrix chips

Post by roo » Fri, 02 Feb 1996 04:00:00



>Has anyone tried Linux on the new Cyrix 6x86 chip or
>any of the NexGen chips? Did it work?
>Art.

Just setup a Cyrix 6x86 system - Triton MB, 16 MB Ram, 256K PB cache.
It seems quite fast - kernel compile took 10 mins 2 sec.
Ran the old Byte benchmarks and it achieved an index of 14.1. This
is the same index a P5/120 achieved with 32M Ram.

Interestingly, the individual benchmark tests varied greatly comparing
between the Cyrix and P5/120. Some tests were much slower and others
were much faster.

Cheers,

Geo

 
 
 

Linux and NexGen/Cyrix chips

Post by roo » Sat, 03 Feb 1996 04:00:00


Ignore my previous post as the bechmarks I ran are in no way comparable
to the P5/120 benchmark from varesearch (www.varesearch.com) that I was
compareing it too.

Here is a better test:

Conditions:  Redhat 2.1, kernel 1.3.59, gcc-2.7.2, single user mode
             Cyrix 6x86/100Mhz, Triton MB, 16MB Ram, 256k PB cache
             Maxtor 1260 EIDE HD

  BYTE UNIX Benchmarks (Version 3.11)
  System -- Linux earth.GEO-NET.com 1.3.59 #1 Thu Feb 1 15:32:43 EST 1996 i486
  Start Benchmark Run: Fri Feb  2 07:26:51 EST 1996
   0 interactive users.
Dhrystone 2 without register variables   174213.2 lps   (10 secs, 6 samples)
Dhrystone 2 using register variables     174638.1 lps   (10 secs, 6 samples)
Arithmetic Test (type = arithoh)         3320749.9 lps   (10 secs, 6 samples)
Arithmetic Test (type = register)         27551.3 lps   (10 secs, 6 samples)
Arithmetic Test (type = short)            31511.5 lps   (10 secs, 6 samples)
Arithmetic Test (type = int)              27500.9 lps   (10 secs, 6 samples)
Arithmetic Test (type = long)             27500.4 lps   (10 secs, 6 samples)
Arithmetic Test (type = float)            14491.3 lps   (10 secs, 6 samples)
Arithmetic Test (type = double)           14488.9 lps   (10 secs, 6 samples)
System Call Overhead Test                 80069.4 lps   (10 secs, 6 samples)
Pipe Throughput Test                      78178.8 lps   (10 secs, 6 samples)
Pipe-based Context Switching Test         23512.8 lps   (10 secs, 6 samples)
Process Creation Test                      1220.6 lps   (10 secs, 6 samples)
Execl Throughput Test                       716.5 lps   (9 secs, 6 samples)
File Read  (10 seconds)                  253760.0 KBps  (10 secs, 6 samples)
File Write (10 seconds)                   13009.0 KBps  (10 secs, 6 samples)
File Copy  (10 seconds)                    3834.0 KBps  (10 secs, 6 samples)
File Read  (30 seconds)                  257404.0 KBps  (30 secs, 6 samples)
File Write (30 seconds)                   24308.0 KBps  (30 secs, 6 samples)
File Copy  (30 seconds)                    3538.0 KBps  (30 secs, 6 samples)
C Compiler Test                             102.1 lpm   (60 secs, 3 samples)
Shell scripts (1 concurrent)                156.6 lpm   (60 secs, 3 samples)
Shell scripts (2 concurrent)                 81.3 lpm   (60 secs, 3 samples)
Shell scripts (4 concurrent)                 41.6 lpm   (60 secs, 3 samples)
Shell scripts (8 concurrent)                 20.3 lpm   (60 secs, 3 samples)
Dc: sqrt(2) to 99 decimal places           5935.6 lpm   (60 secs, 6 samples)
Recursion Test--Tower of Hanoi             2669.8 lps   (10 secs, 6 samples)

                     INDEX VALUES            
TEST                                        BASELINE     RESULT      INDEX

Arithmetic Test (type = double)               2541.7    14488.9        5.7
Dhrystone 2 without register variables       22366.3   174213.2        7.8
Execl Throughput Test                           16.5      716.5       43.4
File Copy  (30 seconds)                        179.0     3538.0       19.8
Pipe-based Context Switching Test             1318.5    23512.8       17.8
Shell scripts (8 concurrent)                     4.0       20.3        5.1
                                                                 =========
     SUM of  6 items                                                  99.6
     AVERAGE                                                          16.6

BogoMips - Calibrating delay loop.. ok - 99.94 BogoMIPS

Interestingly, /proc/cpuinfo thinks this is a 486 !

Cheers,

Geo

 
 
 

Linux and NexGen/Cyrix chips

Post by Stephen Knila » Fri, 09 Feb 1996 04:00:00



>Ignore my previous post as the bechmarks I ran are in no way comparable
>to the P5/120 benchmark from varesearch (www.varesearch.com) that I was
>compareing it too.

>Here is a better test:

>Conditions:  Redhat 2.1, kernel 1.3.59, gcc-2.7.2, single user mode
>             Cyrix 6x86/100Mhz, Triton MB, 16MB Ram, 256k PB cache
>             Maxtor 1260 EIDE HD

What brand motherboard/bios are you using?  My disk drive is the maxtor
1626, which should be FASTER than yours, yet it is HALF the speed!
I ALSO have the triton 16MB 256k PB cache, albeit with a P5/75