Floorplan images show physical locations

Article: Silvermont, Intel's Low Power Architecture
By: rwessel (robertwessel.delete@this.yahoo.com), August 8, 2013 1:03 pm
Room: Moderated Discussions
Sebastian Soeiro (sebastian_2896.delete@this.hotmail.com) on August 7, 2013 3:33 pm wrote:
> rwessel (robertwessel.delete@this.yahoo.com) on August 4, 2013 10:58 pm wrote:
> > Sebastian Soeiro (sebastian_2896.delete@this.hotmail.com) on August 4, 2013 5:42 pm wrote:
> > > Melody Speer (pseudo.delete@this.nym.net) on August 4, 2013 4:11 am wrote:
> > > > Sebastian Soeiro (sebastian_2896.delete@this.hotmail.com) on August 3, 2013 8:23 am wrote:
> > > > > So an instruction asks the LSU to get ahold of "#12345678",
> > > > > and it asks the LSU to translate the virtual address
> > > > > into a physical address (here is a blank. How does it do
> > > > > this? What unit translates the virtual address into
> > > > > a physical address, and how?)
> > > >
> > > > Er... no. The LSU is part of the CPU core proper, and it has two jobs:
> > > >
> > > > 1. Compute the address #12345678
> > > > 2. Stall the processor until the result is ready.
> > > >
> > > > The address #12345678 is then sent to two places. The lower 12 bits (#678)
> > > > are sent to the L1 cache and used to look up several possible cache lines.
> > > >
> > > > At the same time, the upper 20 bits (#12345) are sent to the TLB. I'm going to assume a
> > > > processor with a 64 GB physical address space (like 32-bit x86 processors with PAE), so
> > > > if we get lucky and get a TLB hit, we discover that this corresponds to the physical address
> > > > #abcdef. (If we get a TLB miss, we need to do a page table walk and try again.)
> > > >
> > > > Then we take those upper 24 physical address bits and compare them to the set of candidates found in the
> > > > L1 cache. If we are lucky, one of them matches (#abcdef789 is found in the L1 cache) and we're done.
> > > >
> > > > If the L1 cache misses, we have the full physical address #abcde789 to
> > > > send to higher levels in the cache hierarchy: L2, L3, and main memory.
> > > >
> > > >
> > > > A typical L1 cache is made up of 64-byte cache lines. In addition to the data,
> > > > each line has a "cache tag": upper address bits (24 bits in my example), a validity
> > > > bit, a dirty bit, and possibly some LRU bits to control cache replacement.
> > > >
> > > > 64 of those lines (4096 bytes) make up a cache way. These are direct-mapped: line 0 starts
> > > > at address 0, line 1 starts at address 64, line 2 starts at address 128, and so on. Line
> > > > 63 starts at address 4032. Thus, for any given low address, there's only one possible cache
> > > > line that could contain that data. (Our example #789 is line 30, which covers addresses #780
> > > > through #7bf.) You can fetch that line and check the high-order bits for a match.
> > > >
> > > > Then (again, I'm using typical numbers) 8 of those ways make up the full 32K L1 cache. When
> > > > given an address, the 8 ways find a candidate cache line in parallel. When they're done,
> > > > the TLB has the upper address bits, and those are checked against the 8 candidates to find
> > > > the correct one. Then the lowest-order bits choose the correct part of that line.
> > > >
> > > > L1 cache latency is critical and there are a huge number of tricks that have been tried
> > > > to speed it up. For example, the L1 cache may return some data to the LSU saying "I think
> > > > this is the data you want, but I may be wrong". The CPU can then start processing with
> > > > that data, but back up and retry if it turns out that the cache's first guess was wrong.
> > >
> > > Wow those were definitely some... Amazing reads. I may have bitten off a little more than I can chew.
> > > I read through the explanations a handful of times each and think I have a fair grip on it, but still
> > > some questions (as usual. I apologize for missing probably very, very obvious information.)
> > >
> > > So, if I understand correctly (this is with address translation ON as well as with a TLB, as
> > > it makes most sense to learn with these intricacies included as most modern CPUs have them);
> > >
> > > An instruction requests from the LSU an line of data (I suppose we'll use data for the example?), the
> > > request is handed along with a "challenge" made up in order to calculate the virtual address.
> >
> >
> > I'd not call it a challenge, rather we compute an index into the cache from some of the address bits.
> >
> >
> > > At this point in time, the LSU does two things simultaneously;
> > > it will search both the TLB for an entry of when
> > > the translation has already been done in the past, and it will
> > > also search the L1 cache itself for the tag corresponding
> > > to the line of data needed; this can vary depending on if the
> > > cache is direct mapped, fully associative, or set-way
> > > associative. (the explanations on those things were top-notch, might I add. Thank you!)
> >
> >
> > I'd say it was searching for a cache entry with a tag matching the required address.
> >
> > The cache is ultimately a kind of memory. The index to the line in the cache is computed,
> > then the entire line is read from the cache, and then the tag for each entry in that line
> > is checked against the desired address. If one of them matches, you have a hit.
> >
> >
> > > If there is a TLB hit, then the translation is brought in along with the cache line if the tag is found
> > > and they are compared. If they match, the data is found. If there is a cache miss, the translation found
> > > in the TLB is used to search the other caches until the data is found. If there is a TLB miss, it wont
> > > matter if there is an L1 hit. If there is a TLB miss, the challenge is given to the AGU which calculates
> > > the challenge into a physical address, and that physical address is used to find the data entry.
> >
> >
> > No, the AGU is usually the thing that figures out what logical address the instruction is requesting.
> > In the example we're using, the AGU computes that "25(ebx,4*edx)" is pointing to logical address
> > 0x12345678. On some processors, it may do some additional duties (on x86, for example, the AGU is
> > what usually handles segmentation, but it's still producing the required logical address).
> >
> > If we're running DAT of, the logical address can be used directly as a physical address.
> >
> > If we're running DAT on, the logical address is a virtual address, and must be translated,
> > via the page translation mechanism, to a physical address. In DAT on mode only, the virtual
> > address is given to the TLB for translation to a physical address, or if that fails is passed
> > to the page table walker (in whatever form that exists) to generate the translation.
> >
> >
> > > I know I got atleast one thing wrong as I see a "blank". Specifically;
> > >
> > > rwessel said in this quote: "Now if we go back to the LSU, and instead we’re running with DAT
> > > on. Now the LSU generates logical address 0x12345678, and since we’re running DAT on, it’s
> > > actually a virtual address. So the LSU it passes the virtual address to the address translation
> > > unit, which translates it to a physical address. Let’s say it ends up 0x88888678."
> > >
> > > Now, isnt the challenge (when I refer to the challenge, I mean ""mov eax,25(ebx,4*edx)"") supposed
> > > to be directly given to the AGU? Or does the LSU change the challenge into a virtual address and the
> > > AGU calculates it into a physical address (with no calculations though? I don't understand.)
> >
> >
> > Again, the AGU (acknowledging that the terminology here is not quite standardized) does the "25(ebx,4*edx)"
> > to 0x12345678 thing. Some later stage in the LSU (again, there's not a formal set of boundaries) is
> > what translates that logical address to a physical address (if we're running DAT on). After that happen
> > is when you start to hit caches (with the exception of the overlap done in many L1 caches).
> >
> >
> > > Also, heres on bit of information I'm missing that is probably dreadfully obvious but...
> > > What is a "cache line"? It is an entry or a part of an entry? I don't understand that.
> >
> >
> > The definition makes the most sense for a set-associative cache. A cache line is what's referred to
> > by the index used to do the first part of the lookup in the cache. A cache line contains several (again,
> > ignoring direct mapped or fully associative caches) cache entries. Let's say the we're using the scheme
> > previously described, where the index is the seventh through twelfth lowest bits of the address. IOW,
> > if we have a 32 bit (physical) address we can consider it split as follows (in bits):
> >
> > ttttttttttttttttiiiiiioooooo
> >
> > The oooooooo is the offset within the 64 bytes cache entry. The iiiiii is the
> > index into the cache, and the ttt..ttt is the sixteen bit tag comparator.
> >
> > Thus any address with the same iiiiii bits will index to the same cache line. If we allow more
> > than one cache entry in a cache line we have a N-way set associative cache. So if we have a four
> > way set associative cache, each cache line will contain four cache entries, each of which can hold
> > a different block of 64 bytes (although all the blocks/entries will have the same index).
> >
> > If we're talking about caches in generic terms (from a program visible
> > perspective), we often use cache line as a synonym for cache entry.
> >
> >
> > > Lastly, just to verify, the page walker is the component that does the "searching"
> > > with the physical address in the event of a TLB miss, correct?
> >
> > The page walker searches *for* the physical address in the page tables, given a virtual
> > address as input. It should generate the same result as a TLB hit (which also translates
> > a virtual address to a physical address), but does the process the hard way - instead of
> > just looking at recent translation to see if we've done it already (which is what the TLB
> > does), the page walker has to step through the actual page tables to get the result.
> >
>
> Sorry for the late reply, I've been awfully busy lately. I really do appreciate you
> taking the time to educate me though, this is really, really interesting stuff!
>
> So when the LSU searches both the TLB and the L1 cache, it is looking at ALL the tags in the L1 cache, and
> once found, it compares the ENTIRE entry that was found in the L1 cache to the translation in the TLB? How
> often do these not match? It'd seem like they would 100% of the time... Maybe I'm not seeing something.


If you had a fully associative cache, the lookup would search all the tags (one for each cache entry) for one matching the desired address. If you had a direct-mapped cache, the index developed from the address would be used to select and read a single cache entry, and then that one entries tag would be compared. For a set-associative cache, one entire line of cache entries would be selected (and read), and then the tag for each of those cache entries would be compared.


> So basically, the AGU converts the calculations for the proposed virtual
> (or physical with DAT off?) and hands the tag off to the LSU?


No, the tag is in the cache entry - it says what part of memory is held by the cache line. It is compared to part of the address being looked up. For example, with 64 byte cache entries, you don't need to compare the low six bits (since those would just select a particular byte within the cache entry). With a direct-mapped or set-associative cache, the index does not need to be stored in the tag or compared either, since that part of the comparison happens as a side effect of the hashing process - IOW, all addresses with 010111 in the second six bits (from the low end), are going to go into cache line 39, so if you're looking for an address with 010111 in those bits, you could only ever find it in cache line 39 - so you so have to store those bits in the tag.

The AGU generates a logical address, which is then either treated as a physical address, and used as is, or as a virtual address, and then translated to a physical address.


> So basically, cache lines are a list of cache entries (the tags atleast) consisting of one
> of each entry from however large the associativeness is? That would assume that a directly
> mapped cache has a cache line with only one entry in it at a time and a fully associative
> cache would have the tag of all the entries of the cache in it at one time, correct?


A cache line is (almost always) several complete cache entries. If you had a four way set associative cache with 64 byte entries and 40 bits (five bytes) of additional info (tag, validity flags, protection bits, and whatever else the implementation decided to toss in there), a read of one cache line would actually read some 276* bytes out of the cache, all at once. It's reasonable to consider that four-way set associative cache to be a 2208 bit wide memory.

It is technically correct to consider a fully associative cache as having one line (although you would use a different physical implementation - there is not a reason to ever "read" the whole "line" in a fully associative cache, it's effectively already available), and a direct mapped cache as having cache lines with a single cache entry in them, but no one ever uses the terms like that. Similarly no one ever calls sailplanes "zero engine airplanes" (although that's certainly true).


*276 = 4*(64+5)


> So the page walker uses the virtual address to locate the physical address of the needed cache entry?


The page table walker finds the physical address associated with the virtual address. It's used only when the TLB does not already know that. In either case the resulting physical* address is used to search the cache.


*Again, ignoring the existence/possibility of virtual addressing in caches.
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
Silvermont architecture in-depth articleDavid Kanter2013/05/06 02:30 PM
  Silvermont architecture in-depth articleSHK2013/05/06 03:19 PM
    Silvermont architecture in-depth articletianbing2013/05/18 09:08 PM
      Silvermont architecture in-depth articleKlimax2013/05/18 11:04 PM
        Silvermont architecture in-depth articleEduardoS2013/05/19 12:20 AM
          Silvermont architecture in-depth articleKlimax2013/05/19 02:34 AM
            Silvermont architecture in-depth articleMichael S2013/05/19 04:02 AM
              Silvermont architecture in-depth articleKlimax2013/05/19 04:36 AM
                Silvermont architecture in-depth articleMichael S2013/05/19 05:42 AM
                  Silvermont architecture in-depth articleMichael S2013/05/19 05:45 AM
                Silvermont architecture in-depth articleWilco2013/05/19 06:00 AM
                  Silvermont architecture in-depth articleMichael S2013/05/19 06:30 AM
                    Silvermont architecture in-depth articleDavid Kanter2013/05/19 01:28 PM
      Silvermont architecture in-depth articleRicardo B2013/05/19 07:20 AM
        Silvermont architecture in-depth articleLinus Torvalds2013/05/19 11:24 AM
          Silvermont architecture in-depth articleMichael S2013/05/19 11:39 AM
            Silvermont architecture in-depth articleLinus Torvalds2013/05/19 12:19 PM
              Silvermont architecture in-depth articleanon2013/05/19 10:11 PM
                Silvermont architecture in-depth articleVincent Diepeveen2013/05/21 08:34 AM
                I am not sure I agreeDavid Kanter2013/05/21 01:54 PM
                  I am not sure I agreeanon2013/05/21 04:54 PM
                    I am not sure I agreebakaneko2013/05/22 03:42 AM
                Silvermont architecture in-depth articlebakaneko2013/05/22 03:44 AM
          Silvermont architecture in-depth articleRicardo B2013/05/19 04:17 PM
        Silvermont architecture in-depth articleMaynard Handley2013/05/19 05:44 PM
  Silvermont architecture in-depth articlenone2013/05/06 03:32 PM
    Another use for huge page TLB entriesPaul A. Clayton2013/05/06 09:27 PM
      Another use for huge page TLB entriesanon2013/05/06 09:35 PM
        Lack of paging structure cache informationPaul A. Clayton2013/05/07 06:51 AM
      Another use for huge page TLB entriesMaynard Handley2013/05/14 05:49 PM
        Another use for huge page TLB entriesUngo2013/05/15 12:15 PM
          Another use for huge page TLB entriesMaynard Handley2013/05/15 03:33 PM
            Another use for huge page TLB entriesUngo2013/05/16 03:54 PM
    Silvermont architecture in-depth articleanon2013/05/07 12:11 AM
      Silvermont architecture in-depth articlenone2013/05/07 02:27 AM
        Silvermont architecture in-depth articleJames2013/05/07 05:18 AM
  Silvermont architecture in-depth articleMichael S2013/05/06 03:59 PM
    Silvermont architecture in-depth articleDavid Kanter2013/05/06 05:48 PM
      Silvermont architecture in-depth articleMichael S2013/05/07 12:49 AM
        Silvermont architecture in-depth articleWilco2013/05/07 05:37 AM
          Silvermont architecture in-depth articleMichael S2013/05/07 05:49 AM
            Silvermont architecture in-depth articleWilco2013/05/07 08:17 AM
              Silvermont architecture in-depth articleExophase2013/05/07 09:00 AM
                Silvermont architecture in-depth articleExophase2013/05/07 09:05 AM
                  Resources allocated for different templatesDavid Kanter2013/05/07 09:38 AM
                  Silvermont architecture in-depth articleEduardoS2013/05/07 10:18 AM
  Silvermont architecture in-depth articleEduardoS2013/05/06 06:31 PM
    Silvermont architecture in-depth articleDavid Kanter2013/05/06 07:26 PM
  Some editing suggestionsPaul A. Clayton2013/05/06 09:00 PM
    Some editing suggestionsGabriele Svelto2013/05/08 05:27 AM
      Some editing suggestionsEduardoS2013/05/08 08:27 AM
        Original T series more web-oriented?Paul A. Clayton2013/05/08 11:01 AM
    SMT is a dodo birdBrett2013/05/09 08:26 PM
      SMT is a dodo birdExophase2013/05/09 08:57 PM
      SMT is a dodo birdDavid Kanter2013/05/10 12:34 AM
        SMT is a dodo birdMichael S2013/05/10 03:08 AM
        SMT is a dodo birdGabriele Svelto2013/05/10 03:45 AM
          SMT is a dodo birdRichardC2013/05/11 07:02 AM
            SMT is a dodo birdRicardo B2013/05/11 08:07 AM
              SMT is a dodo birdRichardC2013/05/11 04:39 PM
                SMT is a dodo birdRicardo B2013/05/11 06:55 PM
                  SMT is a dodo birdRichardC2013/05/12 12:36 AM
                    quad-core since Nov 2006RichardC2013/05/12 06:41 AM
                      quad-core since Nov 2006mpx2013/05/13 01:34 AM
                        quad-core since Nov 2006RichardC2013/05/13 03:48 AM
                          quad-core since Nov 2006Jukka Larja2013/05/14 11:34 AM
                            quad-core since Nov 2006RichardC2013/05/14 12:23 PM
                            quad-core since Nov 2006RichardC2013/05/14 12:24 PM
                    SMT is a dodo birdRicardo B2013/05/12 05:48 PM
                      * browsers are moving to a process per tab model (NT)Ricardo B2013/05/12 05:54 PM
                        * browsers are moving to a process per tab modelDoug S2013/05/12 08:27 PM
                          * browsers are moving to a process per tab modelMichael S2013/05/13 12:43 AM
                        * browsers are moving to a process per tab modelMaynard Handley2013/05/13 06:52 PM
                          * browsers are moving to a process per tab modelgallier22013/05/14 01:31 AM
                            * browsers are moving to a process per tab modelDoug S2013/05/14 11:58 AM
                          * browsers are moving to a process per tab modelRichardC2013/05/14 03:54 AM
                            * browsers are moving to a process per tab modelRicardo B2013/05/14 05:06 AM
                              * browsers are moving to a process per tab modelMichael S2013/05/14 05:45 AM
                                * browsers are moving to a process per tab modelRicardo B2013/05/14 07:48 AM
                                How about anti-virus, security, utility, and encryption cpu usage and delays?David Ball2013/05/14 07:48 AM
                                  How about anti-virus, security, utility, and encryption cpu usage and delays?RichardC2013/05/14 10:34 AM
                                * browsers are moving to a process per tab modelUngo2013/05/14 04:50 PM
                      SMT is a dodo birdEduardoS2013/05/12 07:21 PM
                        SMT is a dodo birdJouni Osmala2013/05/12 09:57 PM
                        SMT is a dodo birdRicardo B2013/05/13 02:40 AM
                          SMT is a dodo birdMichael S2013/05/13 03:50 AM
                            SMT is a dodo birdRicardo B2013/05/13 06:13 AM
                    SMT is a dodo birdPatrick Chase2013/05/13 08:33 PM
                      SMT is a dodo birdRichardC2013/05/14 10:38 AM
                        SMT is a dodo birdRicardo B2013/05/14 11:15 AM
                          SMT is a dodo birdRicardo B2013/05/14 11:18 AM
                            SMT is a dodo birdRichardC2013/05/14 12:09 PM
                              SMT is a dodo birdRicardo B2013/05/14 12:50 PM
                                SMT is a dodo birdRichardC2013/05/14 01:09 PM
                                  SMT is a dodo birdRicardo B2013/05/14 02:47 PM
                                  SMT is a dodo birdMaynard Handley2013/05/14 05:02 PM
                                    SMT is a dodo birdRichardC2013/05/14 05:40 PM
                                      SMT is a dodo birdBrendan2013/05/15 02:26 AM
                                        SMT is a dodo birdRichardC2013/05/15 03:14 AM
                                        SMT is a dodo birdRichardC2013/05/15 03:20 AM
                                          SMT is a dodo birdBrendan2013/05/15 04:10 AM
                                            SMT is a dodo birdGabriele Svelto2013/05/15 06:40 AM
                                              SMT is a dodo birdBrendan2013/05/15 07:29 AM
                                            SMT is a dodo birdRichardC2013/05/15 07:29 AM
                                              SMT is a dodo birdRicardo B2013/05/15 09:16 AM
                                                SMT is a dodo birdRichardC2013/05/15 11:06 AM
                                                  SMT is a dodo birdRicardo B2013/05/15 12:19 PM
                                            SMT is a dodo birdEduardoS2013/05/15 04:51 PM
                                              SMT is a dodo birdBrendan2013/05/16 12:29 AM
                                                The Problem With ThreadsRichardC2013/05/16 06:57 AM
                                                  The Problem With Threadsanon2013/05/16 08:55 AM
                                                  The Problem With ThreadsMaynard Handley2013/05/16 10:22 AM
                                                  The Problem With ThreadsBrendan2013/05/16 11:54 AM
                                                    The Problem With ThreadsRichardC2013/05/16 12:19 PM
                                                      The Problem With ThreadsSymmetry2013/05/16 01:14 PM
                                                        The Problem With ThreadsMaynard Handley2013/05/16 02:50 PM
                                                          The Problem With ThreadsSymmetry2013/05/16 04:22 PM
                                                            The Problem With ThreadsRichardC2013/05/17 07:22 AM
                                                        The Problem With ThreadsRichardC2013/05/16 05:49 PM
                                                      The Problem With ThreadsRicardo B2013/05/16 02:12 PM
                                                        The Problem With ThreadsRichardC2013/05/17 07:49 AM
                                                    The Problem With ThreadsBrendan2013/05/17 03:03 AM
                                                SMT is a dodo birdKoby2013/05/16 06:58 AM
                                          SMT is a dodo birdmpx2013/05/15 02:09 PM
                                            SMT is a dodo birdEduardoS2013/05/15 04:55 PM
                               4C/4T 4.0GHz was significantly faster than 4C/4T 3.8GHz??Mark Roulo2013/05/14 05:37 PM
                                 4C/4T 4.0GHz was significantly faster than 4C/4T 3.8GHz??RichardC2013/05/14 05:54 PM
                                  Thanks. Mark Roulo2013/05/14 06:05 PM
                                   4C/4T 4.0GHz was significantly faster than 4C/4T 3.8GHz??Exophase2013/05/14 08:10 PM
                                     More tests of HT and gamesExophase2013/05/14 08:24 PM
                                       More tests of HT and gamesRichardC2013/05/15 03:26 AM
                                         More tests of HT and gamesExophase2013/05/15 06:43 AM
                                           More tests of HT and gamesMichael S2013/05/15 12:35 PM
                                             More tests of HT and gamesRicardo B2013/05/15 01:16 PM
                                               More tests of HT and gamesRicardo B2013/05/15 01:19 PM
                                               More tests of HT and gamesJukka Larja2013/05/16 06:12 AM
                                         More tests of HT and gamesStubabe2013/05/16 12:24 PM
                                       More tests of HT and gamesmpx2013/05/19 02:56 AM
                          SMT is a dodo birdAction_Parsnip2013/05/14 01:28 PM
                          SMT is a dodo birdJukka Larja2013/05/15 09:02 AM
                        SMT is a dodo birdPatrick Chase2013/05/17 09:43 AM
                          SMT is a dodo birdJouni Osmala2013/05/17 10:27 AM
                          SMT is a dodo birdBrett2013/05/17 07:26 PM
                The cost of multithreadingPaul A. Clayton2013/05/11 07:51 PM
                  The cost of multithreadingRichardC2013/05/12 12:20 AM
                SMT benefits high end desktop.Jouni Osmala2013/05/11 10:04 PM
                SMT is a dodo birdHeikki Kultala2013/05/12 01:02 AM
                  SMT is a dodo birdRichardC2013/05/12 06:57 AM
                SMT is a dodo birdJukka Larja2013/05/12 03:33 AM
                SMT is a dodo birdStubabe2013/05/14 12:09 PM
                  SMT is a dodo birdRichardC2013/05/14 01:43 PM
                    SMT is a dodo birdRicardo B2013/05/14 03:00 PM
                      SMT is a dodo birdRichardC2013/05/14 06:27 PM
                        SMT is a dodo birdRicardo B2013/05/15 03:50 AM
                          SMT is a dodo birdRichardC2013/05/15 08:53 AM
                            SMT is a dodo birdRicardo B2013/05/15 09:31 AM
                              SMT is a dodo birdRichardC2013/05/15 11:13 AM
                                SMT is a dodo birdRicardo B2013/05/15 12:43 PM
                                  SMT is a dodo birdRichardC2013/05/15 01:15 PM
                                    SMT is a dodo birdRicardo B2013/05/15 01:40 PM
                                      SMT is a dodo birdRichardC2013/05/15 02:00 PM
                                        SMT is a dodo birdMaynard Handley2013/05/15 03:55 PM
                                        SMT is a dodo birdJouni Osmala2013/05/16 03:16 AM
                    SMT is a dodo birdStubabe2013/05/16 11:54 AM
                  Using the available die areaDoug S2013/05/14 05:51 PM
                    Using the available die areaRicardo B2013/05/15 04:15 AM
                      Using the available die areaStubabe2013/05/16 11:26 AM
                        Using the available die areaRicardo B2013/05/16 02:05 PM
      SMT is a dodo birdmpx2013/05/10 11:51 PM
        SMT is a dodo birdFoo_2013/05/11 05:21 AM
          SMT is a dodo birdEduardoS2013/05/11 09:28 AM
        SMT is a dodo birdDoug S2013/05/11 10:49 AM
          SMT is a dodo birdmpx2013/05/12 12:04 PM
            SMT is a dodo birdLinus Torvalds2013/05/12 02:22 PM
              Mediocre SMT implementations?Paul A. Clayton2013/05/12 05:06 PM
              SMT is a dodo birdmpx2013/05/12 06:13 PM
                Thanks for the actual research effort! (NT)Paul A. Clayton2013/05/13 05:43 AM
              How Does Silvermont Compare To A15?Ashraf Eassa2013/05/15 11:59 AM
                How Does Silvermont Compare To A15?Maynard Handley2013/05/15 04:08 PM
                  How Does Silvermont Compare To A15?TREZA2013/05/16 03:17 AM
                How Does Silvermont Compare To A15?Wilco2013/05/15 05:37 PM
                  How Does Silvermont Compare To A15?David Kanter2013/05/17 08:00 AM
                    How Does Silvermont Compare To A15?Exophase2013/05/17 08:52 AM
                      How Does Silvermont Compare To A15?David Kanter2013/05/17 09:38 PM
                        How Does Silvermont Compare To A15?Exophase2013/05/17 10:33 PM
                          FYI Tegra4 has A15 in it also, so 3 devices (NT)S. Rao2013/05/22 12:59 AM
                            FYI Tegra4 has A15 in it also, so 3 devicesMichael S2013/05/22 03:21 AM
                              FYI Tegra4 has A15 in it also, so 3 devicesS. Rao2013/05/22 10:06 AM
                        How Does Silvermont Compare To A15?Wilco2013/05/19 05:19 AM
                          How Does Silvermont Compare To A15?Patrick Chase2013/05/20 09:29 AM
                            Virtualization was also mostly server-orientedPaul A. Clayton2013/05/20 02:37 PM
                    How Does Silvermont Compare To A15?Wilco2013/05/17 12:22 PM
                      How Does Silvermont Compare To A15?David Kanter2013/05/17 09:29 PM
                        How Does Silvermont Compare To A15?Exophase2013/05/17 10:41 PM
                          How Does Silvermont Compare To A15?Gabriele Svelto2013/05/18 05:48 AM
                            How Does Silvermont Compare To A15?Exophase2013/05/18 09:57 AM
                          How Does Silvermont Compare To A15?David Kanter2013/05/18 08:02 AM
                            How Does Silvermont Compare To A15?Exophase2013/05/18 10:07 AM
                            How Does Silvermont Compare To A15?Wilco2013/05/18 12:04 PM
                        How Does Silvermont Compare To A15?anon2013/05/17 11:03 PM
                        How Does Silvermont Compare To A15?Maynard Handley2013/05/19 06:27 PM
                          How Does Silvermont Compare To A15?David Kanter2013/05/20 05:30 PM
                            How Does Silvermont Compare To A15?Exophase2013/05/20 06:06 PM
                              big.LITTLE and low-leakage processesPatrick Chase2013/05/20 09:26 PM
                                big.LITTLE and low-leakage processesExophase2013/05/20 09:56 PM
                                  big.LITTLE and low-leakage processesPatrick Chase2013/05/20 10:16 PM
                      Clock frequency comparisons and processDavid Kanter2013/05/18 08:50 AM
                        Hard cores from ARM?Patrick Chase2013/05/19 12:38 PM
                  Architecture isn't everythingPatrick Chase2013/05/17 05:49 PM
                    Architecture isn't everythingBjörn R. Björnsson2013/05/17 06:34 PM
                      Architecture isn't everythingaaron spink2013/05/17 08:55 PM
                        Load delay slots need not be a problemPaul A. Clayton2013/05/18 05:41 AM
                          Load delay slots need not be a problemExophase2013/05/18 10:33 AM
                            Did cache misses disrupt load delay?Paul A. Clayton2013/05/18 03:48 PM
                              Did cache misses disrupt load delay?Exophase2013/05/18 07:45 PM
                            TI C6x load delaysPatrick Chase2013/05/19 04:19 PM
                              Hexagon documentation?David Kanter2013/05/23 11:19 AM
                                Hexagon documentation?Exophase2013/05/23 11:45 AM
                                Hexagon documentation?Patrick Chase2013/05/23 03:55 PM
                        Architecture isn't everythingLinus Torvalds2013/05/18 11:31 AM
                How Does Silvermont Compare To A15?David Kanter2013/05/17 08:12 AM
                  How Does Silvermont Compare To A15?Ashraf Eassa2013/05/18 09:06 PM
                    How Does Silvermont Compare To A15?Ashraf Eassa2013/05/18 09:06 PM
            SMT is a dodo birdDoug S2013/05/12 08:22 PM
              SMT is a dodo birdGabriele Svelto2013/05/13 12:53 AM
                SMT is a dodo birdMichael S2013/05/13 01:13 AM
              SMT is a dodo birdMichael S2013/05/13 01:30 AM
  Silvermont architecture in-depth articlerwessel2013/05/06 09:04 PM
    Two different RSBsPaul A. Clayton2013/05/06 09:35 PM
  Decode cacheanon2013/05/07 04:53 AM
    Decode cachenone2013/05/07 04:59 AM
      Decode cacheanon2013/05/07 06:08 AM
    There is also predecode information in IcachePaul A. Clayton2013/05/07 07:18 AM
      There is also predecode information in IcacheMaynard Handley2013/05/13 09:23 PM
    Decode cachenona2013/05/07 11:25 PM
  Silvermont architecture in-depth articlekashing ho2013/05/07 12:18 PM
    Silvermont architecture in-depth articleEduardoS2013/05/07 01:41 PM
      Silvermont architecture in-depth articleLinus Torvalds2013/05/07 03:07 PM
        Silvermont architecture in-depth articleEduardoS2013/05/07 07:18 PM
          Silvermont architecture in-depth articleY2013/05/08 04:23 AM
  Silvermont architecture in-depth articleChris Rodinis2013/05/22 03:56 PM
    Silvermont architecture in-depth articleDavid Kanter2013/05/23 01:17 AM
  Silvermont architecture in-depth articleSebastian Soeiro2013/05/28 09:00 AM
    Silvermont architecture in-depth articleDavid Kanter2013/05/28 12:14 PM
      Silvermont architecture in-depth articleSebastian Soeiro2013/05/28 12:45 PM
        Silvermont architecture in-depth articleRicardo B2013/05/28 03:58 PM
          Silvermont architecture in-depth articleSebastian Soeiro2013/05/28 04:27 PM
            Silvermont architecture in-depth articleRicardo B2013/05/29 02:41 PM
              Silvermont architecture in-depth articleSebastian Soeiro2013/05/29 10:16 PM
                Silvermont architecture in-depth articleRicardo B2013/05/30 08:05 AM
                  Silvermont architecture in-depth articleSebastian Soeiro2013/05/30 09:48 AM
                    Silvermont architecture in-depth articleRicardo B2013/05/30 11:18 AM
                      Silvermont architecture in-depth articleSebastian Soeiro2013/05/30 01:13 PM
                        Silvermont architecture in-depth articleSebastian Soeiro2013/05/30 03:39 PM
                          Silvermont architecture in-depth articleRicardo B2013/05/30 04:43 PM
                            Silvermont architecture in-depth articleSebastian Soeiro2013/05/30 07:52 PM
                              Silvermont architecture in-depth articleRicardo B2013/05/31 04:34 AM
                                Silvermont architecture in-depth articleSebastian Soeiro2013/05/31 06:59 AM
                                  Silvermont architecture in-depth articleKlimax2013/05/31 07:15 AM
                                    Silvermont architecture in-depth articleSebastian Soeiro2013/05/31 08:01 AM
                                      Silvermont architecture in-depth articlerwessel2013/05/31 10:59 AM
                                        Silvermont architecture in-depth articleSebastian Soeiro2013/05/31 02:26 PM
                                          Silvermont architecture in-depth articlerwessel2013/05/31 03:11 PM
                                            Silvermont architecture in-depth articleSebastian Soeiro2013/06/03 09:51 AM
                                              Silvermont architecture in-depth articlerwessel2013/06/03 10:45 AM
                                              Silvermont architecture in-depth articlegallier22013/06/04 02:06 AM
                                                Silvermont architecture in-depth articleSebastian Soeiro2013/06/08 08:30 AM
                                  Silvermont architecture in-depth articleRicardo B2013/05/31 12:22 PM
                                    Silvermont architecture in-depth articleSebastian Soeiro2013/05/31 02:22 PM
                                      Silvermont architecture in-depth articleDavid Kanter2013/05/31 03:08 PM
                                        Silvermont architecture in-depth articleSebastian Soeiro2013/05/31 09:04 PM
                                          Silvermont architecture in-depth articlerwessel2013/05/31 09:53 PM
                                            Silvermont architecture in-depth articleSebastian Soeiro2013/05/31 10:59 PM
                                              Silvermont architecture in-depth articleDavid Kanter2013/06/02 03:02 AM
                                                Silvermont architecture in-depth articleSebastian Soeiro2013/06/02 08:24 AM
                                                  Silvermont architecture in-depth articleDavid Kanter2013/06/02 11:10 PM
                                                    Silvermont architecture in-depth articleSebastian Soeiro2013/06/03 08:25 AM
                                              Silvermont architecture in-depth articlerwessel2013/06/03 12:09 AM
                                                Silvermont architecture in-depth articleSebastian Soeiro2013/06/03 09:59 AM
                                                  Silvermont architecture in-depth articlerwessel2013/06/03 10:51 AM
                                                    Silvermont architecture in-depth articleSebastian Soeiro2013/06/04 06:31 PM
                                      Silvermont architecture in-depth articlerwessel2013/05/31 03:20 PM
                                        Silvermont architecture in-depth articleSebastian Soeiro2013/05/31 06:59 PM
                                          Silvermont architecture in-depth articlerwessel2013/05/31 09:02 PM
                                            Silvermont architecture in-depth articleSebastian Soeiro2013/05/31 09:13 PM
                                            Quibbling on multithreadingPaul A. Clayton2013/06/01 10:11 AM
                                              Quibbling on multithreadingrwessel2013/06/02 11:54 PM
                                                Exploiting for SMT was my "contribution"Paul A. Clayton2013/06/03 04:53 AM
                                            Silvermont architecture in-depth articleMichael S2013/06/01 10:49 AM
                        Silvermont architecture in-depth articleRicardo B2013/05/30 04:41 PM
                          Didn't notice how this was to be organized...Sebastian Soeiro2013/05/30 07:55 PM
          Silvermont, not Saltwell? (NT)papapapa2013/05/28 04:33 PM
            Of course. Not enough sleep. (NT)Ricardo B2013/05/29 02:27 PM
  No out-of-order in FP cluster of silvermont?anon2013/06/28 03:27 AM
    No out-of-order in FP cluster of silvermont?Michael S2013/06/28 04:52 AM
      No out-of-order in FP cluster of silvermont?Exophase2013/06/28 01:54 PM
        No out-of-order in FP cluster of silvermont?Michael S2013/06/29 11:21 AM
          No out-of-order in FP cluster of silvermont?Exophase2013/06/29 09:07 PM
      No out-of-order in FP cluster of silvermont?Linus Torvalds2013/06/29 12:22 PM
        No out-of-order in FP cluster of silvermont?anon2013/06/29 05:26 PM
          No out-of-order in FP cluster of silvermont?EduardoS2013/06/29 06:50 PM
            No out-of-order in FP cluster of silvermont?anon2013/06/29 09:15 PM
              No out-of-order in FP cluster of silvermont?EduardoS2013/06/29 09:31 PM
                No out-of-order in FP cluster of silvermont?anon2013/06/30 02:53 AM
                  No out-of-order in FP cluster of silvermont?EduardoS2013/06/30 05:24 AM
                    No out-of-order in FP cluster of silvermont?anon2013/06/30 10:41 AM
                      No out-of-order in FP cluster of silvermont?EduardoS2013/06/30 01:26 PM
                        No out-of-order in FP cluster of silvermont?anon2013/07/01 12:32 AM
                          No out-of-order in FP cluster of silvermont?EduardoS2013/07/01 05:40 PM
                            No out-of-order in FP cluster of silvermont?anon2013/07/01 06:44 PM
                            No out-of-order in FP cluster of silvermont?Michael S2013/07/02 05:40 AM
                              No out-of-order in FP cluster of silvermont?Exophase2013/07/02 09:03 AM
                              No out-of-order in FP cluster of silvermont?EduardoS2013/07/02 05:20 PM
                          No out-of-order in FP cluster of silvermont?Patrick Chase2013/07/02 09:25 AM
                      No out-of-order in FP cluster of silvermont?Etienne2013/07/02 04:36 AM
                        No out-of-order in FP cluster of silvermont?anon2013/07/02 07:13 AM
                          No out-of-order in FP cluster of silvermont?Patrick Chase2013/07/02 10:03 AM
                            No out-of-order in FP cluster of silvermont?anon2013/07/02 04:12 PM
                              No out-of-order in FP cluster of silvermont?Patrick Chase2013/07/02 04:43 PM
                                No out-of-order in FP cluster of silvermont?anon2013/07/02 05:47 PM
                                  No out-of-order in FP cluster of silvermont?Patrick Chase2013/07/02 07:36 PM
                                    No out-of-order in FP cluster of silvermont?anon2013/07/02 09:03 PM
                                      No out-of-order in FP cluster of silvermont?Symmetry2013/07/03 07:14 AM
                                        No out-of-order in FP cluster of silvermont?Patrick Chase2013/07/03 09:04 PM
                                          No out-of-order in FP cluster of silvermont?Patrick Chase2013/07/04 03:23 AM
                        No out-of-order in FP cluster of silvermont?Patrick Chase2013/07/02 10:01 AM
                          No out-of-order in FP cluster of silvermont?Etienne2013/07/03 03:26 AM
                            layer 2 cache behaviourEtienne2013/07/05 03:14 AM
                              layer 2 cache behaviourMichael S2013/07/05 06:41 AM
                                layer 2 cache behaviourEtienne2013/07/05 09:15 AM
                              layer 2 cache behaviourPatrick Chase2013/07/05 11:37 AM
                                layer 2 cache behaviourMichael S2013/07/06 10:57 AM
                                  layer 2 cache behaviourPatrick Chase2013/07/06 12:08 PM
                                    layer 2 cache behaviourPatrick Chase2013/07/06 12:19 PM
                                    layer 2 cache behaviourEtienne2013/07/08 03:10 AM
                                      layer 2 cache behaviourPatrick Chase2013/07/08 10:02 AM
                                  layer 2 cache behaviourPatrick Chase2013/07/06 02:15 PM
                                  layer 2 cache behaviourUngo2013/07/09 03:24 PM
                      No out-of-order in FP cluster of silvermont?Patrick Chase2013/07/02 08:43 AM
                  No out-of-order in FP cluster of silvermont?Patrick Chase2013/07/02 08:34 AM
                    No out-of-order in FP cluster of silvermont?anon2013/07/02 04:15 PM
            No out-of-order in FP cluster of silvermont?none2013/06/30 05:03 AM
              No out-of-order in FP cluster of silvermont?EduardoS2013/06/30 05:45 AM
                No out-of-order in FP cluster of silvermont?none2013/06/30 06:28 AM
                  No out-of-order in FP cluster of silvermont?EduardoS2013/06/30 01:27 PM
        No out-of-order in FP cluster of silvermont?x2013/06/30 03:54 AM
          No out-of-order in FP cluster of silvermont?Foo_2013/06/30 06:15 AM
        No out-of-order in FP cluster of silvermont?rwessel2013/06/30 10:11 PM
          No out-of-order in FP cluster of silvermont?Linus Torvalds2013/07/01 01:40 PM
            No out-of-order in FP cluster of silvermont?Patrick Chase2013/07/02 10:35 AM
              No out-of-order in FP cluster of silvermont?Linus Torvalds2013/07/02 11:18 AM
            FPU in integer codes2013/07/02 11:38 AM
              FPU in integer codesPatrick Chase2013/07/02 11:56 AM
              FPU in integer codesLinus Torvalds2013/07/02 12:12 PM
                FPU in integer codesPatrick Chase2013/07/02 12:34 PM
                  FPU in integer codesLinus Torvalds2013/07/02 01:55 PM
        No out-of-order in FP cluster of silvermont?2013/06/30 11:42 PM
          No out-of-order in FP cluster of silvermont?bakaneko2013/07/03 03:54 AM
        What about integer SIMD?Exophase2013/07/01 12:44 AM
          What about integer SIMD?anon2013/07/01 01:45 AM
            What about integer SIMD?Michael S2013/07/01 08:57 AM
              What about integer SIMD?rwessel2013/07/01 03:39 PM
                What about integer SIMD?Michael S2013/07/01 04:11 PM
                  What about integer SIMD?EduardoS2013/07/01 06:04 PM
                    What about integer SIMD?Michael S2013/07/02 12:32 AM
          What about integer SIMD?hi2013/07/01 08:15 AM
            What about integer SIMD?Exophase2013/07/01 10:50 AM
            What about integer SIMD?David Kanter2013/07/02 01:20 PM
              What about integer SIMD?hi2013/07/02 03:48 PM
                What about integer SIMD?David Kanter2013/07/04 03:19 PM
  Silvermont architecture in-depth articleSebastian Soeiro2013/07/30 09:34 AM
    TLBs are caches of page tablesPaul A. Clayton2013/07/30 11:03 AM
      TLBs are caches of page tablesSebastian Soeiro2013/07/30 01:18 PM
        TLBs are caches of page tablesPaul A. Clayton2013/07/30 03:59 PM
          TLBs are caches of page tablesSebastian Soeiro2013/07/30 05:27 PM
            TLBs are caches of page tablesrwessel2013/07/30 07:01 PM
              TLBs are caches of page tablesSebastian Soeiro2013/07/31 02:15 PM
                Floorplan images show physical locationsPaul A. Clayton2013/07/31 04:11 PM
                  Floorplan images show physical locationsSebastian Soeiro2013/07/31 09:12 PM
                    Floorplan images show physical locationsrwessel2013/08/01 12:21 AM
                      Floorplan images show physical locationsSebastian Soeiro2013/08/02 12:58 PM
                        Floorplan images show physical locationsAntti-Ville Tuunainen2013/08/02 02:09 PM
                        Floorplan images show physical locationsStubabe2013/08/02 07:49 PM
                          Floorplan images show physical locationsSebastian Soeiro2013/08/03 08:23 AM
                            Floorplan images show physical locationsrwessel2013/08/03 11:16 PM
                            Floorplan images show physical locationsMelody Speer2013/08/04 04:11 AM
                              Floorplan images show physical locationsSebastian Soeiro2013/08/04 05:42 PM
                                Floorplan images show physical locationsrwessel2013/08/04 10:58 PM
                                  Floorplan images show physical locationsSebastian Soeiro2013/08/07 03:33 PM
                                    Floorplan images show physical locationsrwessel2013/08/08 01:03 PM
                                      Floorplan images show physical locationsSebastian Soeiro2013/08/11 09:20 AM
                                        Page tables are set up by the OSPaul A. Clayton2013/08/11 01:25 PM
                                          Page tables are set up by the OSSebastian Soeiro2013/08/16 08:43 PM
                                            Multilevel page table formatPaul A. Clayton2013/08/17 11:18 AM
                                              Multilevel page table formatSebastian Soeiro2013/08/19 03:28 PM
                                                Multilevel page table formatrwessel2013/08/20 12:08 AM
                                                  Multilevel page table formatSebastian Soeiro2013/08/22 04:52 PM
                                                    Multilevel page table formatTREZA2013/08/22 05:30 PM
                                                    Multilevel page table formatrwessel2013/08/23 10:58 PM
                                                      Multilevel page table formatSebastian Soeiro2013/08/25 08:20 AM
                TLBs are caches of page tablesrwessel2013/08/01 12:41 AM
                  TLBs are caches of page tablesSebastian Soeiro2013/08/02 12:51 PM
                    TLBs are caches of page tablesrwessel2013/08/02 06:55 PM
  Silvermont instruction schedulingHugo Décharnes2013/10/02 03:25 AM
    Silvermont instruction schedulingMichael S2013/10/02 04:29 AM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell green?