Prev: van Emde Boas tree implementation
Next: Bron-Kerbosh algorithm for clique detection
From: Randy Howard on 10 Jan 2007 16:33
On Wed, 10 Jan 2007 15:21:42 -0600, globalsk wrote
(in article <1168464102.176197.99380(a)k58g2000hse.googlegroups.com>):

>
> user923005 ha escrito:
>
>> globalsk wrote:
>>> I can't, this is for a university project and I have to use
>>> simplescalar's cache. It's code is kinda of documented, but not enough
>>> in the areas I'm looking and dont seem to find any information about
>>> this, not even in the main web page. That's why I ask here as I'm
>>> someone might have the knowledge about all this maskings.
>>
>> Did you read this stuff:
>> http://simplescalar.com/docs/users_guide_v2.pdf
>>
>> A search on "cache" shows 78 hits in that document.
>>
>> And for this one:
>> http://simplescalar.com/docs/hack_guide_v2.pdf
>>
>> A search on "cache" shows 105 hits.
>>
>> 35 hits in here:
>> http://simplescalar.com/docs/simple_tutorial_v4.pdf
>>
>> 159 in here:
>> http://simplescalar.com/docs/simple_tutorial_v2.pdf
>>
>> If (after reading these things) you do not get your answers, why not
>> send an email to info(a)simplescalar.com with your specific questions.
>
> Of course, I did look those up, but they dont give answers to my
> cuestions.

So you sent an email to that address with the specific questions you
don't yet have answers for and are waiting on a response?



--
Randy Howard (2reply remove FOOBAR)
"The power of accurate observation is called cynicism by those
who have not got it." - George Bernard Shaw





From: globalsk on 10 Jan 2007 18:43

Randy Howard ha escrito:

> On Wed, 10 Jan 2007 15:21:42 -0600, globalsk wrote
> (in article <1168464102.176197.99380(a)k58g2000hse.googlegroups.com>):
>
> >
> > user923005 ha escrito:
> >
> >> globalsk wrote:
> >>> I can't, this is for a university project and I have to use
> >>> simplescalar's cache. It's code is kinda of documented, but not enough
> >>> in the areas I'm looking and dont seem to find any information about
> >>> this, not even in the main web page. That's why I ask here as I'm
> >>> someone might have the knowledge about all this maskings.
> >>
> >> Did you read this stuff:
> >> http://simplescalar.com/docs/users_guide_v2.pdf
> >>
> >> A search on "cache" shows 78 hits in that document.
> >>
> >> And for this one:
> >> http://simplescalar.com/docs/hack_guide_v2.pdf
> >>
> >> A search on "cache" shows 105 hits.
> >>
> >> 35 hits in here:
> >> http://simplescalar.com/docs/simple_tutorial_v4.pdf
> >>
> >> 159 in here:
> >> http://simplescalar.com/docs/simple_tutorial_v2.pdf
> >>
> >> If (after reading these things) you do not get your answers, why not
> >> send an email to info(a)simplescalar.com with your specific questions.
> >
> > Of course, I did look those up, but they dont give answers to my
> > cuestions.
>
> So you sent an email to that address with the specific questions you
> don't yet have answers for and are waiting on a response?
>
>
>
> --
> Randy Howard (2reply remove FOOBAR)
> "The power of accurate observation is called cynicism by those
> who have not got it." - George Bernard Shaw

Hello Randy,

I did write an email to the address but don't know (or hope) the will
answer me. Thats why I'm trying in a more active sources of information
like this one.

From: Barry on 10 Jan 2007 20:16

"globalsk" <jmilanasue(a)gmail.com> wrote in message
news:1168464692.827813.289410(a)i39g2000hsf.googlegroups.com...
>
> Barry ha escrito:
>
> > "globalsk" <jmilanasue(a)gmail.com> wrote in message
> > news:1168392800.832029.100260(a)o58g2000hsb.googlegroups.com...
> > > More insight about this topic and the core of the question relies in
> > > this part of code also:
> > >
> > > cp->blk_mask = bsize-1;
> > > cp->set_shift = log_base2(bsize);
> > > cp->set_mask = nsets-1;
> > > cp->tag_shift = cp->set_shift + log_base2(nsets);
> > > cp->tag_mask = (1 << (32 - cp->tag_shift))-1;
> > > cp->tagset_mask = ~cp->blk_mask;
> > >
> > > Don't understand how this masking works.
> > > I know all this is later used for testing cache's hit or miss in a
> > > block:
> > >
> > > if (CACHE_TAGSET(cp, addr) == cp->last_tagset)
> > > {
> > > /* hit in the same block */
> > > blk = cp->last_blk;
> > > goto cache_fast_hit;
> > > }),
> > >
> > > But if someone can give examples with addresses and bits it will be
> > > clearer.
> > >
> > > All this code is for cache simulation as you can see, where you can
> > > define associativity and all the regular parameters for a cache.
> > > globalsk ha escrito:
> > >
> > > > Hello,
> > > >
> > > > I'm trying to understand this fragment of code in Simplescalar's
> > > > cache.c code (I've search and look over dozens sites, including
> > > > simplescalar.com and there's no help about the coding):
> > > >
> > > > /* cache access macros */
> > > > #define CACHE_TAG(cp, addr) ((addr) >> (cp)->tag_shift)
> > > > #define CACHE_SET(cp, addr) (((addr) >> (cp)->set_shift) &
> > > > (cp)->set_mask)
> > > > #define CACHE_BLK(cp, addr) ((addr) & (cp)->blk_mask)
> > > > #define CACHE_TAGSET(cp, addr) ((addr) & (cp)->tagset_mask)
> > > >
> > > > I know is about shifting and moving bits around, but don't know how
and
> > > > specially why is done in this way and for what.
> > > >
> > > > cp is an structure for a cache
> > > > addr is a typedef qword_t (64 bits)
> > > > md_addr_t blk_mask; idem, typedef qword_t
> > > > int set_shift;
> > > > md_addr_t set_mask; /* use *after* shift */
> > > > int tag_shift;
> > > > md_addr_t tag_mask; /* use *after* shift */
> > > > md_addr_t tagset_mask; /* used for fast hit detection */
> > > >
> > > > Maybe this is little information, but maybe someone with the
knowledge
> > > > about caches and this type of coding can help out. Please if someone
> > > > can give a helping hand about this and links where they explain in
more
> > > > detail simplescalar's code will be very appreciated.
> > > >
> > > > Please forgive my syntax mistakes, english is not my mother tongue.
> > >
> >
> > I don't think you have clearly expressed whether you do not
> > understand the C code, or how the code is attempting to simulate
> > a cache.
> >
> > I thought it was the C code in which case you can find many
> > sources on bit shifting and masking with a simple web search.
> > Of course you can search on Kernighan & Ritchie and learn
> > 101 more things in the process. The chances that you are
> > programming in C at a university and someone doesn't have a
> > copy of K&R you could borrow are miniscule. But if you
> > intend to do any C programming you should just by one.
> >
> > How the simulator does is job is certainly a question for the author(s)
> > unless you are willing to pay someone to take the time to do it for
> > you. Although, if it were a common tool you would likely find
> > someone here willing to give help.
>
> Hello is more about how the code is attempting to simulate a cache and
> how it does it with all this masking and shifting (I now what the
> instructions do but don't understand why). Of course the author could
> answer it but is not likely, more likely is to find someone with a
> little knowledge on cache, which for whom it will be pretty easy to
> answer I think, don't have to be an expert. I myself is the first time
> I have to touch something related with a cache and is taking some
> effort fo find answers.
>
> Thanks
>

Well you are looking at a few lines of code and talking about
"cache." I can think of at least three levels of cache on some
systems, before you get to I/O.


From: globalsk on 11 Jan 2007 05:28

Barry ha escrito:

> "globalsk" <jmilanasue(a)gmail.com> wrote in message
> news:1168464692.827813.289410(a)i39g2000hsf.googlegroups.com...
> >
> > Barry ha escrito:
> >
> > > "globalsk" <jmilanasue(a)gmail.com> wrote in message
> > > news:1168392800.832029.100260(a)o58g2000hsb.googlegroups.com...
> > > > More insight about this topic and the core of the question relies in
> > > > this part of code also:
> > > >
> > > > cp->blk_mask = bsize-1;
> > > > cp->set_shift = log_base2(bsize);
> > > > cp->set_mask = nsets-1;
> > > > cp->tag_shift = cp->set_shift + log_base2(nsets);
> > > > cp->tag_mask = (1 << (32 - cp->tag_shift))-1;
> > > > cp->tagset_mask = ~cp->blk_mask;
> > > >
> > > > Don't understand how this masking works.
> > > > I know all this is later used for testing cache's hit or miss in a
> > > > block:
> > > >
> > > > if (CACHE_TAGSET(cp, addr) == cp->last_tagset)
> > > > {
> > > > /* hit in the same block */
> > > > blk = cp->last_blk;
> > > > goto cache_fast_hit;
> > > > }),
> > > >
> > > > But if someone can give examples with addresses and bits it will be
> > > > clearer.
> > > >
> > > > All this code is for cache simulation as you can see, where you can
> > > > define associativity and all the regular parameters for a cache.
> > > > globalsk ha escrito:
> > > >
> > > > > Hello,
> > > > >
> > > > > I'm trying to understand this fragment of code in Simplescalar's
> > > > > cache.c code (I've search and look over dozens sites, including
> > > > > simplescalar.com and there's no help about the coding):
> > > > >
> > > > > /* cache access macros */
> > > > > #define CACHE_TAG(cp, addr) ((addr) >> (cp)->tag_shift)
> > > > > #define CACHE_SET(cp, addr) (((addr) >> (cp)->set_shift) &
> > > > > (cp)->set_mask)
> > > > > #define CACHE_BLK(cp, addr) ((addr) & (cp)->blk_mask)
> > > > > #define CACHE_TAGSET(cp, addr) ((addr) & (cp)->tagset_mask)
> > > > >
> > > > > I know is about shifting and moving bits around, but don't know how
> and
> > > > > specially why is done in this way and for what.
> > > > >
> > > > > cp is an structure for a cache
> > > > > addr is a typedef qword_t (64 bits)
> > > > > md_addr_t blk_mask; idem, typedef qword_t
> > > > > int set_shift;
> > > > > md_addr_t set_mask; /* use *after* shift */
> > > > > int tag_shift;
> > > > > md_addr_t tag_mask; /* use *after* shift */
> > > > > md_addr_t tagset_mask; /* used for fast hit detection */
> > > > >
> > > > > Maybe this is little information, but maybe someone with the
> knowledge
> > > > > about caches and this type of coding can help out. Please if someone
> > > > > can give a helping hand about this and links where they explain in
> more
> > > > > detail simplescalar's code will be very appreciated.
> > > > >
> > > > > Please forgive my syntax mistakes, english is not my mother tongue.
> > > >
> > >
> > > I don't think you have clearly expressed whether you do not
> > > understand the C code, or how the code is attempting to simulate
> > > a cache.
> > >
> > > I thought it was the C code in which case you can find many
> > > sources on bit shifting and masking with a simple web search.
> > > Of course you can search on Kernighan & Ritchie and learn
> > > 101 more things in the process. The chances that you are
> > > programming in C at a university and someone doesn't have a
> > > copy of K&R you could borrow are miniscule. But if you
> > > intend to do any C programming you should just by one.
> > >
> > > How the simulator does is job is certainly a question for the author(s)
> > > unless you are willing to pay someone to take the time to do it for
> > > you. Although, if it were a common tool you would likely find
> > > someone here willing to give help.
> >
> > Hello is more about how the code is attempting to simulate a cache and
> > how it does it with all this masking and shifting (I now what the
> > instructions do but don't understand why). Of course the author could
> > answer it but is not likely, more likely is to find someone with a
> > little knowledge on cache, which for whom it will be pretty easy to
> > answer I think, don't have to be an expert. I myself is the first time
> > I have to touch something related with a cache and is taking some
> > effort fo find answers.
> >
> > Thanks
> >
>
> Well you are looking at a few lines of code and talking about
> "cache." I can think of at least three levels of cache on some
> systems, before you get to I/O.

In simplescalar simultation you can define levels and sizes of cache,
in the project we'll be using 2 levels, L1: il1 & dL1 and L2: unified.

My question is why they do the masking and shifting in this way. The
answer will be to put a general examples/theory with and address the
cpu is looking up in a cache and the corresponding hit with data in a
line in a block.

Please forgive me, I'm only looking for help!

First  |  Prev  | 
Pages: 1 2 3
Prev: van Emde Boas tree implementation
Next: Bron-Kerbosh algorithm for clique detection