On mySimon: Deadwood - The Complete Series
BNET Business Network:
BNET
TechRepublic
ZDNet
TalkBack 30 of 126:
Next »
« Previous
The SVr4 src has definitely left the building.
``If I don't badly recall, the allegation of SCO was that the SMP code used in Linux was handed over by IBM when Linux' SMP support simply sucked.''

I agree that the SMP and NUMA support was sucking in the past. However in the recent years things have changed wildly. There are OSS projects out there (e.g., Linux Scalability Project, Kernel Performance Project, NUMA Project, etc.) that are dedicated in making Linux SMP and NUMA efficient.

Actually, the 2.6 kernel comes with several NUMA enhancements, including the O(1) scheduler, CPU sets, etc. And ALL these have NO connection to init. UNIX source. I know of at least one vendor (SGI) which has ported many of these decent NUMA mechanisms to their 2.4 kernel (ALtix).

I think now that Linux 2.6 has started moving passed older UNIX kernels or it is very close at least (IRIX, AIX, etc.)

Nevertheless, it makes very little sense to copy verbatim code that is 20+ years OLD and is thus severely outdated. Even the regular UNIX src tree has changed so much from the original AT&T licensed code that it must be almost unencumbered. I suspect that only the STREAMS pipelining code maybe still there from SVr4 but even this one was implemented by Mentat and is used by several vendors.

Ideas, algorithms, APIs and behavior models were used from UNIX but the actual src is simply nonsense.

-m
Posted by: michael_t   Posted on: 07/14/05 You are currently: a Guest | Members login | Terms of Use

Alert moderator to an offensive message

Subscribe to this discussion via Email or RSS

Was there ever any question?  Sabz5150 | 07/14/05
And the MS funded SCO FUD descends into a joke  Richard Flude | 07/14/05
i believe a couple people are eating their words  Monkey_MCSE | 07/14/05
I never got this accusation  Mark Miller | 07/14/05
OK add Sun if it makes you feel better  Richard Flude | 07/14/05
OS Companies  __howard__ | 07/14/05
I don't like the self-centered paranoid drivel  Mark Miller | 07/16/05
I thought Baystar pulled out  Mark Miller | 07/16/05
Yes, but I don't see SCO laughing... ? (nt)  michael_t | 07/14/05
We have a say in my country that  michael_t | 07/14/05
Raising questions  John L. Ries | 07/14/05
Yeah, they did it...  Overcast | 07/15/05
Don't get me wrong  John L. Ries | 07/15/05
Who will sue the gov. who without reasons supports SCO in favor of MS? (NT)  Vily Clay | 07/14/05
Vily give me a source for this accusation  Squawkbox | 07/15/05
Since when the DOJ does not belong to the government? (NT)  Vily Clay | 07/15/05
Vily that is pure speculation and conjecture  Squawkbox | 07/15/05
So why does antitrust work for any corp. except MS? (NT)  Vily Clay | 07/16/05
Right hand not knowing what the left is doing.  __howard__ | 07/14/05
So what you're saying is...  Cardinal_Bill | 07/14/05
Not a "Get out of jail free card"  __howard__ | 07/14/05
Highly unlikely  monolith | 07/15/05
Lawyers aren't all the bright  voska | 07/15/05
Bright = Commonsense/NOT  eknoid@... | 07/15/05
Lawyers not all that bright  leelitch | 07/15/05
Why lawyers don't get IT  cyberian_z | 07/16/05
SCO is behaving irrationally  __howard__ | 07/15/05
Why SCO ? Why?  michael_t | 07/14/05
Well  thetargos | 07/14/05
The SVr4 src has definitely left the building.  michael_t | 07/14/05
That was my point.  thetargos | 07/15/05
Solaris and SVr4  michael_t | 07/14/05
Thanks for the pointer.  thetargos | 07/15/05
SCO didn't invent IX SMP  maldain | 07/15/05
The only aspect in SVr4 relevant to SMP  michael_t | 07/17/05
SCO needs to keep the faith...  Mike Cox | 07/14/05
haha 9.0 Mikey  Monkey_MCSE | 07/14/05
You left out the tanned skin? nt  michael_t | 07/14/05
Say What?!  timn1969 | 07/15/05
You May Be A Good Programmer ...  PMC-CON | 07/15/05
Spelling  RickyF | 07/15/05
Is there an archive of Mike Cox posts available online anywhere?  Joseph Friedman | 07/15/05
Why? It's getting old already.  dstinson_z | 07/15/05
ah  Joseph Friedman | 07/15/05
Quit reading his posts then.  linux_skynyrd | 07/15/05
Wait a damn minute  Squawkbox | 07/15/05
This certainly does not bode well for SCO  Mark Miller | 07/14/05
Its worse than that  Richard Flude | 07/14/05
But that's not what IBM's being sued for  John L. Ries | 07/15/05
Indeed  Mark Miller | 07/16/05
Where is Don Rupert?  TechnicalSmurf | 07/14/05
Didn't he morph to No_Ax_no... ? (nt)  michael_t | 07/17/05
Lawyers, on retainer, with nothing better to do.  pokey6660 | 07/15/05
Reverse engineering highly unlikely to produce IDENTICAL code.  __howard__ | 07/15/05
But then SCO isn't alleging identical code  maldain | 07/15/05
incredible  __howard__ | 07/15/05
I found some things intriguing  Mark Miller | 07/16/05
BAH-HAHAHAHAHAHAHAHA!!!!!  James T. Kirk | 07/15/05
No, no, no  Chad_z | 07/15/05
Proof is not enough!  Reverend MacFellow | 07/15/05
Re: proof is not enough  Otto_Delete | 07/15/05
Put a fork in SCO she is DONE!!!  Laff | 07/15/05
Well, now that explains a lot...  JonathonDoe | 07/15/05
Nope, no pass for SCO  Otto_Delete | 07/15/05
150 people can produce a variety of opinons  __howard__ | 07/15/05
SCO e-mail: No 'smoking gun' in Linux code  Loverock Davidson | 07/15/05
Reputable Sources  tic swayback | 07/15/05
Here, here...  Physco Dude | 07/15/05
Where, where?  mggordon | 07/15/05
"Everything I say is a lie"  Mark Miller | 07/16/05
Corection...  Physco Dude | 07/15/05
Correction...  Loverock Davidson | 07/15/05
RTFA  tic swayback | 07/15/05
I did  Loverock Davidson | 07/15/05
And yet...  tic swayback | 07/15/05
That means...  John L. Ries | 07/15/05
Considering the source  John L. Ries | 07/15/05
Hmmm....  bchesmer | 07/15/05
Huh????  bchesmer | 07/15/05
Ignorence is bliss for Loverock Davidson.  B.O.F.H. | 07/17/05
Still bad for linsux  Loverock Davidson | 07/15/05
are you crazy  Reverend MacFellow | 07/15/05
he's not crazy...  stormdoor | 07/15/05
Never ascribe  Yagotta B. Kidding | 07/15/05
You must be on drugs or have a sense of humor...  Physco Dude | 07/15/05
OR he skipped taking his drugs (medication) wink nt  michael_t | 07/17/05
Dude, don't inflame the mob.  dstinson_z | 07/15/05
There are mobs on both sides  daver_z | 07/15/05
Microsoft, not Linux, steals code all the time  daver_z | 07/15/05
You forgot...  bchesmer | 07/15/05
Don't forget  rmerts@... | 07/15/05
You and  bchesmer | 07/15/05
Special Prosecutor time  dzash2000 | 07/15/05
Survival: District Court Edition  gardoglee | 07/15/05
SCO will declare bankrupcy soon!  Reverend MacFellow | 07/15/05
nothing here, folks ...  code_flogger | 07/15/05
Exactly...  jasonp@... | 07/15/05
One less SCO Engineer  daver_z | 07/15/05
GE should sue SCO  daver_z | 07/15/05
Grab a tinny  monolith | 07/15/05
Does it really matter?  Art Royce | 07/15/05
Dismissal  Yagotta B. Kidding | 07/15/05
so the big kahuna himself is gonna testify. is he gonna flame SCO? (nt)  wessonjoe | 07/15/05
IBM did ask for dismissal  John L. Ries | 07/15/05
No....  bchesmer | 07/16/05
IBM did  bchesmer | 07/15/05
Hmmmmm  Overcast | 07/15/05
The problem with code...  ArtMac | 07/15/05
if the court uses the same rule of measure with SCO that they use with song  wessonjoe | 07/15/05
Over 100s, 1000s of lines of code?  __howard__ | 07/15/05
On the otherhand...  daver_z | 07/15/05
Burden of proof on SCO  __howard__ | 07/15/05
Good point  daver_z | 07/15/05
No matter win or loose. Microsoft saved sometime for Longhorn  tyn | 07/15/05
Your assumption is wrong  daver_z | 07/15/05
Bonk, bonk, thump, thump, bang bang bang  Squawkbox | 07/15/05
This does explain...  John L. Ries | 07/15/05
follow the money  __howard__ | 07/15/05
Depends on what you're trying to do...  John L. Ries | 07/15/05
pump & dump  tldwg04011 | 07/15/05
I wouldn't be surprised  Mark Miller | 07/16/05
"almost no evidence" does not equal "or complete lack thereof"  hawkeyeaz1 | 07/15/05
So?  CobraA1 | 07/16/05
Code fragments??? Gimme a break!  rmerts@... | 07/15/05
Fragments?  __howard__ | 07/17/05
on what plannet re you living  NemesisNL | 07/15/05

What do you think?

SponsoredWhite Papers, Webcasts, and Downloads