Re: Seinfeld's Contribution to the The Principles of Operation
Actually, the POPS is full of "may or may not". I think the reason for th at is that it gives the designers of specific machines latitude to do things differently, not bei ng constrained too much by the architecture. This in turn should lower the cost of the machine. The reason the "may or may not" stuff is in the POPS is to warn programme rs that they should not expects results that they determine by trial-and-error to be the same on every machine. Despite the different wording here, this is an example of that concept. You CANNOT trust the bit to be set only when the storage is referenced. S ome instructions even SAY that what happens may vary from machine to machine. The instruction I found this time in the manual is TPROT (TEST PROTECTION), but I think I have read others. Conversely, when a look-ahead references a page, but then the code on tha t page is not executed. does that count a referenced or not? Yes, the machine had to reference it , but no, it never got executed. YMWV (Your Mileage WILL Vary).. Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com On Thu, 7 Oct 2010 17:44:26 -0500, Gary M. Dennis wrote: >There is only one place in z Architecture Principles of Operation where the >word substantially is used. Even more surprising than its use is the fa ct >that it appears immediately before the word "accurate". My interest in t he >reference bit, though passing, is sincere since I would like to hang my hat >on the validity of that indicator. Substantially means " to a great ext ent >or degree". In system z architecture, I am unfamiliar with this level of >accuracy. > >For your reference (no pun intended) the paragraph in z POPS is: > >"The record provided by the reference bit is substantially accurate. The >reference bit may be set to one by fetching data or instructions that ar e >neither designated nor used by the program, and, under certain condition s, a >reference may be made without the reference bit being set to one. Under >certain unusual circumstances, a reference bit may be set to zero by oth er >than explicit program action." > >Although Seinfeld had to have written the previous 67 words, it is possi ble >(age 21 at the time) for him to have contributed the original paragraph >from the 1975 POPS and simply updated the paragraph for System z. The 1 975 >paragraph follows. > >"The record of references provided by the reference bit is substantially >accurate. The reference bit may be turned on by fetching data or >instructions that are neither designated nor used by the program, and, u nder >certain conditions, a reference may be made without the reference bit be ing >turned on. Under certain unusual conditions, a reference bit that is on may >be turned off by other than explicit program action. > >"Turned on" and "Turned off" apparently gave way to the more precise "se t to >one" and "set to zero" for those so technically challenged that on/off h ad >lost all meaning. Think about it; the technical writers were striving f or >more precision in their description than the "thing" being described >actually provided. > >Does anyone have a clue what sentences 2 and 3 (substantially) mean in >either paragraph and would anyone venture a guess as to why, after 40 ye ars >of architectural excellence, the reference bit can't be more than >"substantially accurate"? My curiosity is killing me. > > >--. .- .-. -.-- > >Gary Dennis > = == ==
Re: Seinfeld's Contribution to the The Principles of Operation
On Fri, Oct 8, 2010 at 7:33 PM, Bill Holder wrote: > I believe the reason for allowing some leeway has to do > with hardware design concepts for things like pipelining, > "reach ahead" and "out-of-order execution" - with the > hardware in essence "speculatively" performing certain > operations ahead of time or in parallel (to maximize > utilization and throughput), the potential exists for the > processor "thinking" something is going to be referenced, > and performing reference recording on the fetch from > storage, only to find out later that that path wasn't > really taken. To correct such apparent overindication of > reference would mean tracking "de-referencing" and having > to worry about (and serializing against) reference > recording actions from other processors. You could well > ask "why not defer reference recording until the path > decision is made?", but I would answer that doing so would > break the pipeline and constrain the design of the > processor / storage interface in ways that would defeat > the whole purpose. I would even say it's better from a paging perspective. If the code makes a U-turn just at the end of the page and causes the CPU to decode the next instruction already, it would be annoying if the OS would take that 2nd page away because nobody needs it anyway... assuming that such a case would also cause a page fault? Just don't bring your reference bits into court as evidence... ;-) | Rob
Re: Seinfeld's Contribution to the The Principles of Operation
I guess I can say I "have a clue", in that I can offer my personal interpretation, as an operating system developer and designer, though I cannot claim to officially speak for the architects or IBM (so take this all with the prerequisite grain of salt): In my opinion, that paragraph is in the context of the entire Reference Recording section, and in particular, the very first sentence: "Reference recording provides information for use in selecting pages for replacement." Based on this, my interpretation of "substantially accurate" in this context is that reference recording is "good enough" for an operating system to use it for page replacement selection, that is, managing the overcommitment of main storage in support of larger virtual storage. I believe the reason for allowing some leeway has to do with hardware design concepts for things like pipelining, "reach ahead" and "out-of-order execution" - with the hardware in essence "speculatively" performing certain operations ahead of time or in parallel (to maximize utilization and throughput), the potential exists for the processor "thinking" something is going to be referenced, and performing reference recording on the fetch from storage, only to find out later that that path wasn't really taken. To correct such apparent overindication of reference would mean tracking "de-referencing" and having to worry about (and serializing against) reference recording actions from other processors. You could well ask "why not defer reference recording until the path decision is made?", but I would answer that doing so would break the pipeline and constrain the design of the processor / storage interface in ways that would defeat the whole purpose. Again, these are only my interpretations as an operating system developer, I am not officially speaking for architecture, hardware design, or IBM as a whole. - Bill Holder, z/VM Development, IBM
Re: Seinfeld's Contribution to the The Principles of Operation
In sentence two the reference is discussing how a program can either Access records in read or write access I think Regards -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Gary M. Dennis Sent: Thursday, October 07, 2010 6:44 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Seinfeld's Contribution to the The Principles of Operation There is only one place in z Architecture Principles of Operation where the word substantially is used. Even more surprising than its use is the fact that it appears immediately before the word "accurate". My interest in the reference bit, though passing, is sincere since I would like to hang my hat on the validity of that indicator. Substantially means " to a great extent or degree". In system z architecture, I am unfamiliar with this level of accuracy. For your reference (no pun intended) the paragraph in z POPS is: "The record provided by the reference bit is substantially accurate. The reference bit may be set to one by fetching data or instructions that are neither designated nor used by the program, and, under certain conditions, a reference may be made without the reference bit being set to one. Under certain unusual circumstances, a reference bit may be set to zero by other than explicit program action." Although Seinfeld had to have written the previous 67 words, it is possible (age 21 at the time) for him to have contributed the original paragraph from the 1975 POPS and simply updated the paragraph for System z. The 1975 paragraph follows. "The record of references provided by the reference bit is substantially accurate. The reference bit may be turned on by fetching data or instructions that are neither designated nor used by the program, and, under certain conditions, a reference may be made without the reference bit being turned on. Under certain unusual conditions, a reference bit that is on may be turned off by other than explicit program action. "Turned on" and "Turned off" apparently gave way to the more precise "set to one" and "set to zero" for those so technically challenged that on/off had lost all meaning. Think about it; the technical writers were striving for more precision in their description than the "thing" being described actually provided. Does anyone have a clue what sentences 2 and 3 (substantially) mean in either paragraph and would anyone venture a guess as to why, after 40 years of architectural excellence, the reference bit can't be more than "substantially accurate"? My curiosity is killing me. --. .- .-. -.-- Gary Dennis
Seinfeld's Contribution to the The Principles of Operation
There is only one place in z Architecture Principles of Operation where the word substantially is used. Even more surprising than its use is the fact that it appears immediately before the word "accurate". My interest in the reference bit, though passing, is sincere since I would like to hang my hat on the validity of that indicator. Substantially means " to a great extent or degree". In system z architecture, I am unfamiliar with this level of accuracy. For your reference (no pun intended) the paragraph in z POPS is: "The record provided by the reference bit is substantially accurate. The reference bit may be set to one by fetching data or instructions that are neither designated nor used by the program, and, under certain conditions, a reference may be made without the reference bit being set to one. Under certain unusual circumstances, a reference bit may be set to zero by other than explicit program action." Although Seinfeld had to have written the previous 67 words, it is possible (age 21 at the time) for him to have contributed the original paragraph from the 1975 POPS and simply updated the paragraph for System z. The 1975 paragraph follows. "The record of references provided by the reference bit is substantially accurate. The reference bit may be turned on by fetching data or instructions that are neither designated nor used by the program, and, under certain conditions, a reference may be made without the reference bit being turned on. Under certain unusual conditions, a reference bit that is on may be turned off by other than explicit program action. "Turned on" and "Turned off" apparently gave way to the more precise "set to one" and "set to zero" for those so technically challenged that on/off had lost all meaning. Think about it; the technical writers were striving for more precision in their description than the "thing" being described actually provided. Does anyone have a clue what sentences 2 and 3 (substantially) mean in either paragraph and would anyone venture a guess as to why, after 40 years of architectural excellence, the reference bit can't be more than "substantially accurate"? My curiosity is killing me. --. .- .-. -.-- Gary Dennis