hardware prefeth, branch predictor in the context of multitasking
|
Prev: Online Exams for Certification, Free Practice Exams, Study Material, Dumps
Next: Motherboard unusuable because of 1 millimeter of missing plastic ?!?!?!?!
From: Del Cecchi on 28 Nov 2009 21:20 "Andy "Krazy" Glew" <ag-news(a)patten-glew.net> wrote in message news:4AF8D67F.6060604(a)patten-glew.net... > Terje Mathisen wrote: >> I'm somewhat in love with the idea of a multi-level table: >> >> A large shared but simple (2-bit counter or similar) augmented with >> a small exception table per core which only stores branch info for >> branches that have missed in the large table. >> >> Is this even feasible? :-) > > Yes. > > But when I submitted papers on such designs, 1996-2000, they got > rejected. > > Daniel Jimenez eventually broke through, with the second level table > being a great big neural net predictor. The east Anglia hack has revealed the glory of peer review. Your papers may have been rejected because they offended one of the reviewers pre conceived notions.
From: Robert Myers on 28 Nov 2009 23:56 On Nov 28, 9:20 pm, "Del Cecchi" <delcecchioftheno...(a)gmail.com> wrote: > The east Anglia hack has revealed the glory of peer review. Your > papers may have been rejected because they offended one of the > reviewers pre conceived notions. Every field where prestige is money is going to have similar problems. This one had to be coming. Physical reality is never so simple as the popular theology we've been hearing. Robert.
From: Bernd Paysan on 29 Nov 2009 12:33
Robert Myers wrote: > On Nov 28, 9:20 pm, "Del Cecchi" <delcecchioftheno...(a)gmail.com> > wrote: > >> The east Anglia hack has revealed the glory of peer review. Your >> papers may have been rejected because they offended one of the >> reviewers pre conceived notions. > > Every field where prestige is money is going to have similar > problems. This one had to be coming. Physical reality is never so > simple as the popular theology we've been hearing. The problems Wikipedia (especially the German one) is having lately with people rather deleting articles than writing new ones and thereby driving out volunteers shows that it doesn't even need money or prestige for having these problems. Peer review means that only non- controversial articles will be published. This means progress is only allowed into the direction where the participants in peer review think it should happen. -- Bernd Paysan "If you want it done right, you have to do it yourself" http://www.jwdt.com/~paysan/ |