- Haskell-Cafe - Haskell.org

Dynamic typing of polymorphic functions
by Alfonso Acosta 28 Dec '07

28 Dec '07

Hi all, As some of you might remember I'm working on a EDSL which models process networks (A process can simply be viewed as a box which has a number of input and output signals and makes computations over them). A simple example of the processes implemented is MapSY, which is similar to Haskell's well-known map function. It takes a signal and a processing function as arguments and applies that function the signal. Following Lava's approach, a specifically designed Signal type hides and "secretly" forwards the structure of the network. Such structure is logically an untyped graph. For that reason, the processing functions associated to the different processes are kept in Dynamic form. A very simplified implementation could be: ------ import Data.Typeable import Data.Dynamic -- the phantom type parameter makes signal typing consistent newtype Signal a = Signal PrimSignal newtype PrimSignal = PrimSignal (Proc (PrimSignal)) data Proc input = MapSY Dynamic -- The processing function input -- The process input -- the rest of the processes are omitted eval :: Proc Dynamic -> Dynamic -- evaluates the output of a process for one input eval (MapSY dynF dynIn) = dynApp dynF dynIn -- the map process constructor mapSY :: (Typeable a, Typeable b) => (a -> b) -> Signal a -> Signal b mapSY f (Signal primSig) = Signal (PrimSignal (MapSY (toDyn f) primSig)) ----- OK, If you managed to read until this point, you might have noticed that, due to the monomorphism restriction implied by Data.Typeable, it is impossible to build polymorphic processes. The approach works nicely for monomorphic processes though: add1 :: Signal Int -> Signal Int add1 = mapSY ((+1) :: Int -> Int) mapSndInt :: Signal (Int, Int) -> Signal Int mapSndInt = mapSY (snd :: (Int, Int) -> Int) But it does not work for building polymorphic processes. The following process would be really useful but its compilation obviously fails: mapSnd :: Signal (a, a) -> Signal a mapSnd = mapSY snd Could not deduce (Typeable a) from the context () arising from a use of `mapSY' Possible fix: add (Typeable a) to the context of the type signature for `mapSnd' Again, due to the monomorphism restriction of the Typeable class, the compiler's suggestion of adding a Typeable context didn't work. (Well, strangely enough, adding the "Typeable a" constraint hushed GHC but an error was instead triggered at runtime). I must admit I'm totally stuck with this problem. My two options are: 1) Forcing the user to define mapSnd & friends for each combination of types (which can hardly be considered an acceptable solution) 2) Think of a change in the internal representation of signals which made polymorphic processes possible. Polymorphic processes don't have to be necessarily definable by the user. They should be happy enough with a few polymorphic primitives (mapSnd would be one of them). * Maybe an implementation of unsafe dynamics using unsafeCoerce which allowed applying polymorphic functions to monomorphic values for specific safe cases? I would really appreciate any suggestions or remarks regarding my problem/design. Thanks in advance, Alfonso Acosta

3 3

ambiguous types although 'asTypeOf'
by Henning Thielemann 28 Dec '07

28 Dec '07

I thought I understand monomorphism restriction. But it seems, I don't. I have boilt down my problem to the following test. Don't try to make any sense of it, it is just the smallest code I could come up with. test :: (Integral a, RealFrac a) => a test = let c = undefined in asTypeOf (round c) c When compiling I get: Compiling StorableInstance ( src/StorableInstance.hs, interpreted ) src/StorableInstance.hs:38:17: Warning: Defaulting the following constraint(s) to type `Double' `RealFrac a' arising from use of `round' at src/StorableInstance.hs:38:17-21 In the first argument of `asTypeOf', namely `(round c)' In the definition of `test1': test1 = let c = undefined in asTypeOf (round c) c

3 3

A Foldable binary search tree
by Brad Larsen 27 Dec '07

27 Dec '07

Mainly for pedagogical purposes, I'm writing a BST. The elements in a BST need to be ordered, so I define the new data type like this: data (Ord a) => BST a = Empty | BST (BST a) a (BST a) After some hacking and some reading, I came across Data.Foldable, and think it would be neat to make BST Foldable: instance Foldable BST where foldr f s Empty = s foldr f s (BST l v r) = foldr f (f v (foldr f s r)) l However, ghc tells me ``Could not deduce (Ord a) from the context (Foldable BST) arising from use of `BST' at bst.hs:19:13-21'', which is the second line of the foldr definition. If I remove the (Ord a) context from the BST data definition, ghc accepts my code. However, I'd like to keep that context. How does one do this? Thanks! Brad Larsen P.S. Does that definition of foldr look reasonable, i.e. is it associating in the ``right'' order for a binary tree?

4 6

DSL question -- was: New slogan for haskell.org
by Steve Lihn 27 Dec '07

27 Dec '07

Thanks for the explanation on DSL. It helped me understand how Haskell works compared to other popular languages out there. It is a programming methodology change. Or what is called paradigm change on how to design a software with Haskell. Haskell has its general-purpose features. Yet its strength is the ability and ease of defining DSL by utilizing the higher order \"stuff\" (like type class, composite function, monads, and arrows), which to say the least, can be quite abstract and challenging. I do come aross a question while reading the DSL page on Wikipedia. http://en.wikipedia.org/wiki/Domain-specific_programming_language In the Disadvantage section (near the end), there is an item -- hard or impossible to debug. Can anybody explain why or what it means? And how does it apply to Haskell? -- steve On 12/11/07, Brandon S. Allbery KF8NH <allbery(a)ece.cmu.edu> wrote: > > > * Domain Specific Language (who needs it? other than academics and > > Wall Streeter?) > > DSELs can be thought of as a programming methodology; as such, it has > wide applicability, but most programmers don\'t think that way. Tcl > was originally positioned as a \"DSEL enabler\" (write composable > functions in C, tie them together in Tcl), but most programmers > \"don\'t get it\" and so don\'t tend to use it as such. More recently, > Lua seems to be using a similar philosophy with a little more success > --- but mainly by limiting it to something which most programmers can > deal with. Also compare how Perl and Python monoliths have replaced > the original Unix philosophy in which the shell is a DSEL enabler > (composing single-function programs like \"cat\" and \"sort\"). > > When it comes down to it, Haskell\'s strengths are only realizeable if > you approach programming problems in a different way from most common > languages. It\'s a fairly big leap, and arguably Haskell won\'t > compete well in the mainstream until more programmers have made that > leap. >

6 7

CAF's in Haskell
by Neil Mitchell 27 Dec '07

27 Dec '07

Hi, Are CAF's specified in the Haskell report? I couldn't find them mentioned. If not, why do all Haskell compilers support them? Is there some paper which persuaded everyone they were a good idea, or some reference I could look up? Thanks Neil

6 7

Fwd: Re: [Haskell-cafe] Wikipedia on first-class object
by Cristian Baboi 27 Dec '07

27 Dec '07

------- Forwarded message ------- From: "Cristian Baboi" <cristi(a)ot.onrc.ro> To: "Yitzchak Gale" <gale(a)sefer.org> Cc: Subject: Re: [Haskell-cafe] Wikipedia on first-class object Date: Thu, 27 Dec 2007 12:21:44 +0200 I think I found the answer to why functions cannot be written to files. This is by design. Haskell must be free. Enabling writing functions to files, might make it ilegal in some countries. :-) On Thu, 27 Dec 2007 11:10:21 +0200, Yitzchak Gale <gale(a)sefer.org> wrote: > Like any type, only certain operations make > sense on functions. Strings can be compared to each > other for equality and written to a disk, and you > can take the logarithm of a float, but none of those > operations make sense for functions. In particular, > two functions are equal only if they produce > the same value for every input, and in general it is > impossible for a computer to check that. > > -Yitz > > > ________ Information from NOD32 ________ > This message was checked by NOD32 Antivirus System for Linux Mail > Servers. > part000.txt - is OK > http://www.eset.com ________ Information from NOD32 ________ This message was checked by NOD32 Antivirus System for Linux Mail Servers. part000.txt - is OK http://www.eset.com

2 1

Comments on reading two ints off Bytestring
by Paulo J. Matos 27 Dec '07

27 Dec '07

Hello all, It is either too difficult to get two integers of a bytestring, in which case something should be done to ease the process or I should learn much more Haskell. I guess the latter is the correct guess. I have a bytestring containing two naturals. I was to get them as efficiently as possible. Here's my code: parseHeader :: BS.ByteString -> (Int, Int) parseHeader bs = let first = BS.readInt $ BS.dropWhile (not . isDigit) bs in if(isNothing first) then error "Couldn't find first natural." else let second = BS.readInt $ BS.dropWhile (not . isDigit) $ snd $ fromJust first in if(isNothing second) then error "Couldn't find second natural." else (fst $ fromJust first, fst $ fromJust second) This seems to work: > parseHeader $ BS.pack "hello 252 359" (252,359) Is there a better way? Cheers, -- Paulo Jorge Matos - pocm at soton.ac.uk http://www.personal.soton.ac.uk/pocm PhD Student @ ECS University of Southampton, UK

9 15

Why does this blow the stack?
by Justin Bailey 27 Dec '07

27 Dec '07

Given this function: dropTest n = head . drop n $ [1..] I get a stack overflow when n is greater than ~ 550,000 . Is that inevitable behavior for large n? Is there a better way to do it? Justin

10 22

Fwd: Re: [Haskell-cafe] Functions are first class values in C
by Cristian Baboi 27 Dec '07

27 Dec '07

On Sat, 22 Dec 2007 16:48:51 +0200, Peter Verswyvelen <bf3(a)telenet.be> wrote: > Cristian Baboi wrote >> Lazy constant in C: >> int C1 (){ return 7; } > > Not really, this is not lazy, since it always recomputes the value "7". > > To have "lazy" values in C you would have to do something like: > > struct ValueInt > { > int IsComputed; > union > { > int Value; > struct > { > int (*ComputeValue)(void *args); > void* Args; > }; > }; > }; > > int GetLazyInt (ValueInt* v) > { > if( !v->IsComputed ) > { > v->Value = v->ComputeValue(v->Args); > v->IsComputed = true; > } > return v->Value; > } > > But this of course, is totally useless in C and very bulky. It's also > impossible to know when to call freemem on the Args (hence garbage > collection in FP), when *not* to use lazy values but strict values > instead (hence strictness analysis in FP), etc... I know FP have automatic garbage collection. I know FP compilers use strictness analysis. In C++ one can isolate memory management in constructors and destructors. There are C compilers that are also able to do some optimizations. > I must say I had the same opinion as you had for many many years. I > always thought "functions as first class values" where just function > pointers, so what is it these Haskell/FP guys are so excited about? But > if you dig deeper, you'll see the magic... Notice you will have to give > yourself some time; it is very hard to get out of the imperative blob. > E.g. I'm still being sucked into the imperative blob after my first year > of Haskell hacking :) > PS: As I'm relatively new to Haskell, don't take the above C code too > seriously; it certainly will not reflect the way a real Haskell system > works. I am new to Haskell, but not new to declarative programming. I programmed in Prolog for several years, and I tryed LISP, but I don't liked the LISP syntax. I don't take my C example seriously either. The thing is I think that for a language to have "first-class" functions, it must be "homoiconic" if I understand the terms correctly. Have you tryed to write a Haskell program that manipulate Haskell programs ? Please don't tell me that Haskell compiler is written in Haskell, because there are C compilers written in C.

8 12

Trouble with types
by Konstantin Vladimirov 25 Dec '07

25 Dec '07

Colleagues! The essence of trouble may be given by code: [haskell] module TypeTrouble where class FirstClass a where firstFunction :: (SecondClass b) => a -> b class SecondClass a where secondFunction :: a -> Double data FirstData = FirstData Double data SecondData = SecondData Double instance SecondClass SecondData where secondFunction (SecondData d) = d instance FirstClass FirstData where firstFunction (FirstData d) = SecondData d [/haskell] I need, the firstFunction of FirstClass types to return a value of a SecondClass type. For example SecondData for FirstData, but for some FirstClass ThirdData, some SecondClass FourthData, etc. GHC 6.8.1 produces an error: [quote] typetrouble.hs:17:31: Couldn't match expected type `b' against inferred type `SecondData' `b' is a rigid type variable bound by the type signature for `firstFunction' at typetrouble.hs:4:31 In the expression: SecondData d In the definition of `firstFunction': firstFunction (FirstData d) = SecondData d In the definition for method `firstFunction' [/quote] How can I make this idea work? -- With best regards, Konstantin

5 4

Peyton Jones' "Beautiful Concurrency" .. i.e. Software Transactional Memory ...
by Galchin Vasili 25 Dec '07

25 Dec '07

Hello, My brain is a "out to lunch". I have read the paper "Beautiful Concurrency" (as well as a bunch of "gaming" papers regarding multi cores). I am playing with the "Account" example in the paper. In the paper, the alias "type Account = TVar Int" is used. I want to actually apply the function "withdraw" to an example "Account" parameter. I keep getting a type check error. Can someone give me a concrete example of > withdraw ...... ?? Kind regards, Vasya

2 1

FW: Treating command-line arguments as a Haskell expression
by Simon Peyton-Jones 25 Dec '07

25 Dec '07

Would someone familiar with the command-line-parsing libraries care to help Krassimir? Thanks Simon -----Original Message----- From: Krassimir Krustev [mailto:krassimir.krustev@gmail.com] Sent: 23 December 2007 11:38 To: Simon Peyton-Jones Subject: Treating command-line arguments as a Haskell expression Dear Dr. Jones: I've been pulling my hair over the Christmas Holidays trying to figure the following: I want to call a function from within Haskell module so that the name of this function corresponds to my first command-line argument and the rest rest of the command-line arguments are would become themselves the argument to this function. Coundn't figure this out. It is my first week with Haskell. Yours, Krassimir

5 4

Re: [Haskell-cafe] Functions are first class values in C
by Miguel Mitrofanov 24 Dec '07

24 Dec '07

> It is true, however, that preprocessing used to be (in pre-standard > days) separate from the language. This has not been true for decades. Well, I've seen cpp to be used as a preprocessor not for C sources but for something else.

2 1

Fwd: Re: [Haskell-cafe] Functions are first class values in C
by Cristian Baboi 24 Dec '07

24 Dec '07

------- Forwarded message ------- From: "Cristian Baboi" <cristian.baboi(a)gmail.com> To: "Jules Bean" <jules(a)jellybean.co.uk> Cc: Subject: Re: [Haskell-cafe] Functions are first class values in C Date: Sat, 22 Dec 2007 15:58:50 +0200 On Sat, 22 Dec 2007 14:09:13 +0200, Jules Bean <jules(a)jellybean.co.uk> wrote: > Cristian Baboi wrote: >> Let me show you an example to prove it. > That's not C. > That's the C preprocessor, which is a textual substitution macro > language. Well, the preprocessor is part of the language in a way. These two come together. > Macros certainly aren't first class (you can't pass a macro to a > function, only its expansion). In Haskell I cannot pass a function to a function, only its expansion. > C does support function pointers, which are something like first class > functions. The main things C lacks which people associate with true > first-class function is: > The ability to construct anonymous/local functions. If you look at the example you will see I've done that. > The ability to capture local variables and return a function with some > variables bound. If I can construct "anonymous" functions and "constants", I can construct functions with some variables bound. > The ability to write type-safe functions with polymorphic arguments. I didn't know this must be a property of first-class functions. C is staticaly typed, so type errors will be detected. Haskell is just C with some syntactic sugar :-)

4 10

Fwd: Re: [Haskell-cafe] more on the "Beautiful Concurrency" function "withdraw" application ...
by Dean Herington 24 Dec '07

24 Dec '07

>Date: Mon, 24 Dec 2007 01:44:16 -0500 >To: "Galchin Vasili" <vigalchin(a)gmail.com>, haskell-cafe(a)haskell.org >From: Dean Herington <heringtonlacey(a)mindspring.com> >Subject: Re: [Haskell-cafe] more on the "Beautiful Concurrency" >function "withdraw" application ... >Cc: >Bcc: >X-Attachments: > >At 12:19 AM -0600 12/24/07, Galchin Vasili wrote: >>module Main where >>import Control.Concurrent.STM >>import Control.Concurrent >>import System.Random >> >>type Account = TVar Int >> >>transfer :: Account -> Account -> Int -> IO () >>transfer from to amount >> = atomically (do {deposit to amount; >> withdraw from amount}) >> >>deposit :: Account -> Int -> STM () >>deposit acc amount = withdraw acc (- amount) >> >>withdraw :: Account -> Int -> STM () >>withdraw acc amount >> = do { bal <- readTVar acc; >> writeTVar acc (bal - amount) } >> >> >> >>When I try >> >>withdraw (TVar 10) .. >> >>I get "Not in scope: data constructor 'TVar'. OK .. "Tvar" is a >>class and not a Haskell data type ... so now what? > >You first need to create an account: > > > acct <- newTVar 100 Sorry, that would be: > acct <- atomically (newTVar 100) >Then you can withdraw from it: > > > atomically (withdraw acct 10) > >Dean

1 0

more on the "Beautiful Concurrency" function "withdraw" application ...
by Galchin Vasili 24 Dec '07

24 Dec '07

module Main where import Control.Concurrent.STM import Control.Concurrent import System.Random type Account = TVar Int transfer :: Account -> Account -> Int -> IO () transfer from to amount = atomically (do {deposit to amount; withdraw from amount}) deposit :: Account -> Int -> STM () deposit acc amount = withdraw acc (- amount) withdraw :: Account -> Int -> STM () withdraw acc amount = do { bal <- readTVar acc; writeTVar acc (bal - amount) } When I try withdraw (TVar 10) .. I get "Not in scope: data constructor 'TVar'. OK .. "Tvar" is a class and not a Haskell data type ... so now what? Vasya

2 1

Re: ANNOUNCE: GHC version 6.8.2
by Judah Jacobson 24 Dec '07

24 Dec '07

I neglected to CC the below email to haskell-cafe; apologies if anyone gets this twice. ---------- Forwarded message ---------- From: Judah Jacobson <judah.jacobson(a)gmail.com> Date: Dec 21, 2007 2:12 PM Subject: Re: [Haskell-cafe] Re: ANNOUNCE: GHC version 6.8.2 To: John Dorsey <haskell(a)colquitt.org> On Dec 21, 2007 12:48 PM, John Dorsey <haskell(a)colquitt.org> wrote: > On a related topic, I've been trying to build 6.8.2 on Leopard lately. > I've been running up against the infamous OS X readline issues. I know > some builders here have hacked past it, but I'm looking for a good > workaround... ideally one that works without changes outside the GHC > build area (besides installing a real readline). > > Here's what I noticed before I started drowning in the build platform. > (I'm no gnu-configure expert nor GHC insider.) > > I can get gnu-readline installed from Macports, no problem. > > The top-level configure in GHC doesn't respond to my various attempts: > > o using --with-readline-libraries and --with-readline-includes > (Although it looks like the libraries/readline/configure script > might recognize these, I can't get an option to pass through.) Actually, this is supposed to work. When running the top-level ghc configure, you should be able to just say ./configure --with-readline-libraries=/opt/local/lib --with-readline-includes=/opt/local/include and have those arguments automatically passed to the readline configure script. I just checked that this works on my machine (Tiger x86) when building 6.8.2 against MacPorts' readline. If it doesn't work for you, what errors are you getting? Also, a couple of other generic questions: - Are you on x86 or PPC? (The latter is not working right yet with Leopard.) - How are you bootstrapping the build of 6.8.2: are you using a pre-built binary of ghc-6.8.1, or some other way? Best, -Judah

2 1

Gtk2Hs installation problem with ghc 6.8.2 on Ubuntu 7.10
by Yin Wang 24 Dec '07

24 Dec '07

Hi, I'm installing Gtk2Hs on my Ubuntu 7.10. I have ghc 6.8.2 compiled with the extra libs in my home directory with the configuration "./configure --prefix=/home/wy/Programs" and it worked well. Then I configured gtk2hs in the same way and it compiled and installed with no problem. But when I tried to compile the demo program with any of the following command line: # ghc --make World.hs # ghc --make World.hs `pkg-config --cflags --libs gtk+-2.0` # ghc --make World.hs `pkg-config --cflags --libs gtk+-2.0` -L/home/wy/Programs/lib/gtk2hs It give some linking errors: Linking World ... World.o: In function `r2JR_info': (.text+0x46): undefined reference to `gtkzm0zi9zi12zi1_GraphicsziUIziGtkziAbstractziWidget_widgetShowAll_closure' World.o: In function `r2JR_info': (.text+0x4d): undefined reference to `gtkzm0zi9zi12zi1_GraphicsziUIziGtkziTypes_zdf550_closure' World.o: In function `r2JT_info': (.text+0xaa): undefined reference to `gtkzm0zi9zi12zi1_GraphicsziUIziGtkziAbstractziContainer_containerChild_closure' World.o: In function `r2JT_info': (.text+0xb1): undefined reference to `gtkzm0zi9zi12zi1_GraphicsziUIziGtkziTypes_zdf662_closure' World.o: In function `r2JT_info': (.text+0xb8): undefined reference to `gtkzm0zi9zi12zi1_GraphicsziUIziGtkziTypes_zdf550_closure' ... collect2: ld returned 1 exit status The first command worked well with the ghc 6.6.1 came with ubuntu. I tried to uninstall my old gtk package, but this doesn't help. Any ideas? -- Yin Wang

1 0

A question about functional dependencies
by alpheccar 24 Dec '07

24 Dec '07

I have defined a function on types thanks to functional dependencies. I use it like that: myFunction :: (TypeFunction MyComplexType f) => f -> f f is unique for MyComplexType. Is there a way to name f ? I would like to write: myFunction :: SimplifiedType -> SimplifiedType where TypeFunction MyComplexType SimplifiedType is an instance of the class TypeFunction. Of course, I don't want to write explicitly SimplifiedType otherwise I could just do type SimplifiedType = ... The idea is that I build a MyComplexType using type synonyms and type constructors. But, the result has a shape which is too complex. I simplify it into a canonical form that I can use in some algorithms at the type level. I suspect I will have to use the type constraint and that there is no way to name the result of a computation at the type level ? Thanks, Christophe.

1 0

Fwd: stg_uncheckedShiftRL64 unresolved when compiling with GHC 6 8 2
by golubovsky＠gmail.com 23 Dec '07

23 Dec '07

Hi, I posted this to the HAppS discussion group, but maybe somebody is able to answer in the Cafe, at least where this unresolved symbol belongs to. It looks like the same problem was discussed earlier related to GHC 6.7, but I got this problem right now, compiling HPaste with HAppS (using SearchPath). The linker message is: =================================== During interactive linking, GHCi couldn't find the following symbol: stg_uncheckedShiftRL64 This may be due to you not asking GHCi to load extra object files, archives or DLLs needed by your current session. Restart GHCi, specifying the missing library using the -L/path/to/object/dir and - lmissinglibname flags, or simply by naming the relevant files on the GHCi command line. Alternatively, this link failure might indicate a bug in GHCi. If you suspect the latter, please send a bug report to: glasgow-haskell-b...@haskell.org =================================== Output from ghc-pkg -l: =================================== /home2/dima/install//lib/ghc-6.8.2/package.conf: Cabal-1.2.3.0, array-0.1.0.0, base-3.0.1.0, binary-0.4.1, bytestring-0.9.0.1, containers-0.1.0.1, directory-1.0.0.0, filepath-1.1.0.0, (ghc-6.8.2), haskell-src-1.0.1.1, haskell98-1.0.1.0, hpc-0.5.0.0, hscolour-1.6, html-1.0.1.1, mtl-1.1.0.0, network-2.1.0.0, old-locale-1.0.0.0, old- time-1.0.0.0, packedstring-0.1.0.0, parsec-2.1.0.0, pretty-1.0.0.0, process-1.0.0.0, random-1.0.0.0, readline-1.0.1.0, regex-base-0.72.0.1, regex-compat-0.71.0.1, regex-posix-0.72.0.2, rts-1.0, stm-2.1.1.0, template-haskell-2.2.0.0, unix-2.3.0.0, xhtml-3000.0.2.1, zlib-0.4.0.1 =================================== To save time I compiled only GHC without extralibs, and later installed packages that were msissng during compilation from extralibs or hackage. Are there any modules shown whose versions raise any suspicions? Thanks.

2 1

MPC 2008: SECOND CALL FOR PAPERS
by Shin-Cheng Mu 23 Dec '07

23 Dec '07

SECOND CALL FOR PAPERS 9th International Conference on Mathematics of Program Construction (MPC'08) Marseille (Luminy), France, July 15-18th 2008 http://mpc08.lri.fr BACKGROUND The biennial MPC conferences aim to promote the development of mathematical principles and techniques that are demonstrably practical and effective in the process of constructing computer programs. Topics of interest range from algorithmics to support for program construction in programming languages and systems. The previous conferences were held in Twente, The Netherlands (1989), Oxford, UK (1992), Kloster Irsee, Germany (1995), Marstrand, Sweden (1998), Ponte de Lima, Portugal (2000), Dagstuhl, Germany (2002), Stirling, UK (2004, colocated with AMAST '04) and Kuressaare, Estonia (2006, colocated with AMAST '06). The 2008 conference will be held in Marseille, France at the International Center for Mathematical Meetings (http://www.cirm.univ-mrs.fr/web.ang) INVITED SPEAKERS To be announced. IMPORTANT DATES * Submission of abstracts: 14 January 2008 * Submission of full papers: 21 January 2008 * Notification of authors: 10 March 2008 * Camera-ready version: 10 April 2008 TOPICS Papers are solicited on mathematical methods and tools put to use in program construction. Topics of interest range from algorithmics to support for program construction in programming languages and systems. Some typical areas are type systems, program analysis and transformation, programming-language semantics, program logics. Theoretical contributions are welcome provided their relevance for program construction is clear. Reports on applications are welcome provided their mathematical basis is evident. SUBMISSION Submission is in two stages. Abstracts (plain text, 10 to 20 lines) must be submitted by 14 January 2008. Full papers (pdf) adhering to the llncs style must be submitted by 21 January 2008. There is no official page limit, but authors should strive for brevity. The web-based submission system is now open http://www.easychair.org/conferences/?conf=mpc08. Papers must report previously unpublished work and not be submitted concurrently to another conference with refereed proceedings. Accepted papers must be presented at the conference by one of the authors. The proceedings of MPC'08 will be published in the Lecture Notes in Computer Science series of Springer-Verlag. After the conference, the authors of the best papers will be invited to submit revised versions to a special issue of the Science of Computer Programming journal of Elsevier. PROGRAMME COMMITTEE Christine Paulin-Mohring INRIA-UniversitÈ Paris-Sud, France (chair) Philippe Audebaud Ecole Normale SupÈrieure Lyon, France (co-chair) Ralph-Johan Back Abo Akademi University, Finland Eerke Boiten University of Kent, UK Venanzio Capretta University of Nijmegen, Netherlands Sharon Curtis Oxford Brookes University, UK Jules Desharnais UniversitÈ Laval, QuÈbec, Canada Peter Dybjer Chalmers University of Technology, Sweden Jeremy Gibbons University of Oxford, UK Lindsay Groves Victoria University of Wellington, New Zealand Ian Hayes University of Queensland, Australia Eric Hehner University of Toronto, Canada Johan Jeuring Utrecht University, Netherlands Dexter Kozen Cornell University, USA Christian Lengauer Universit‰t Passau, Germany Lambert Meertens University of Utrecht, Netherlands Bernhard Mˆller Universit‰t Augsburg, Germany Carroll Morgan University of New South Wales, Australia Shin-Cheng Mu Academia Sinica, Taiwan Jose Nuno Oliveira Universidade do Minho, Portugal Tim Sheard Portland State University, USA Tarmo Uustalu Institute of Cybernetics Tallin, Estonia VENUE The conference will be held in Marseille, the second largest city in France next to Paris. Its port is the most important in France, and opens the city to the world through the Mediterranean Sea. MPC'08 will be hosted by the International Center for Mathematical Meetings. The center is located inside the Campus of Luminy Faculty. It is close to the "Calanques", an astounding wild coastline composed of creeks stretching from Marseille to Cassis. LOCAL ORGANIZERS MPC 2008 is organized with the support of INRIA. The local organizers are Philippe Audebaud and Christine Paulin-Mohring with the help of Marie-RenÈe Fleury. Enquiries regarding the programme (submission etc.) should be addressed to mpc08(at)lri.fr

1 0

[15/16] SBM: Predictions compared to the measurements
by Peter Firefly Brodersen Lund 23 Dec '07

23 Dec '07

Don Stewart and Duncan Coutts were nice enough to give me some predictions regarding the performance of the programs in my teaser email. When I answered them, I hadn't yet written the code to merge reports so I hadn't yet noticed the general pattern of much worse memory performance on 6.9. Also, Don Stewart hadn't yet released bytestring 0.9.0.2 (he did so in order to fix performance problems pointed out by my teaser benchmarks, actually). >From Duncan Coutts: >I'll try guessing at some ratios: > >1.0 space-bslc8-foldlx-1 >1.1 space-bslc8-acc-1 >2.0 space-bs-c8-foldlx-1 >2.1 space-bs-c8-acc-1 >4.0 space-xxxxx-acc-1 >15 space-xxxxx-foldl I would put them in two groups (I disregard the lenfil benchmark because it was for "extra credit"): space-bslc8-foldlx-1 around 1.4/1.5s (1.2/1.1s with bytestring 0.9.0.2) space-bs-c8-foldlx-1 around 1.2s space-bs-c8-acc-1 around 1.1s (1.3s on 6.6.1) space-bslc8-acc-1 around 3.4s (6.4s on 6.6.1!) space-xxxxx-acc-1 around 5.0s (5.4s on 6.8.2) space-xxxxx-foldl around 5.0s (5.3/5.4s on 6.8.2) >From Duncan Coutts (on the "extra credit" program hs/space-bslc8-lenfil-1): >Hmm. So that should work in constant memory, a few 64 chunks at once. >I'd expect this to be pretty fast. Ok, this was mean of me. I knew there was a memory bug here :) >From Don Stewart: >Summary: the suspicious lazy bytestring program works now. (constant >space, and fastest overall, as expected originally) > >Program 1, lazy bytestring length . filter > > Yesterday: > ./A +RTS -sstderr < 150M 1.01s user 0.10s system 98% cpu 1.123 total > 40M allocated > > * Today (fixed!): > ./A +RTS -sstderr < 150M 0.26s user 0.06s system 96% cpu 0.332 total > 2M allocated > > Reason, deprecated array fusion mucking up the optimiser. > >I think we can close this regression. Nope. Look at the memory graphs: hs/space-bslc8-lenfil-1: 38632 ██████████▌ | -- 38644 ██████████▌ | -- 1940 ▌ | -- 109404 █████████████████████████████▋ | -- 82324 ██████████████████████▍ | -- 109388 █████████████████████████████▋ | -- 82304 ██████████████████████▎ | It is fixed for ghc 6.8.2 running bytestring 0.9.0.2 but not for ghc 6.9.20071119 and head (as of noon 2007-12-19), no matter the bytestring version. There are lots of memory performance bugs in ghc 6.9. I've reordered the time bar charts to use the same order as my teaser email. hs/space-bslc8-lenfil-1: 1.521 7‰ 0.2 ███████▊ | -- 1.295 1‰ 0.2 ██████▌ | -- 0.429 0‰ 0.2 ██▏ | -- 1.323 1‰ 0.1 ██████▊ | -- 0.483 2‰ 0.4 ██▌ | -- 1.327 1‰ 0.2 ██████▊ | -- 0.482 2‰ 0.4 ██▌ | (this was the "extra credit" program) hs/space-bs-c8-acc-1: 1.287 20‰ 0.2 ██████▌ | -- 1.131 2‰ 0.1 █████▊ | -- 1.132 2‰ 0.3 █████▊ | -- 1.145 1‰ 0.2 █████▊ | -- 1.150 1‰ 0.2 █████▉ | -- 1.147 2‰ 0.1 █████▊ | -- 1.147 2‰ 0.2 █████▊ | hs/space-bslc8-acc-1: 6.367 17‰ 0.2 ████████████████████████████████▎ | -- 3.396 7‰ 0.1 █████████████████▎ | -- 3.508 4‰ 0.1 █████████████████▊ | -- 3.388 10‰ 0.1 █████████████████▏ | -- 3.441 10‰ 0.1 █████████████████▍ | -- 3.438 3‰ 0.0 █████████████████▍ | -- 3.434 13‰ 0.1 █████████████████▍ | hs/space-xxxxx-acc-1: 5.036 7‰ 0.1 █████████████████████████▌ | -- 5.381 6‰ 0.0 ███████████████████████████▎ | -- 5.358 8‰ 0.1 ███████████████████████████▏ | -- 4.991 1‰ 0.1 █████████████████████████▎ | -- 5.047 7‰ 0.1 █████████████████████████▌ | -- 5.051 3‰ 0.1 █████████████████████████▋ | -- 5.049 4‰ 0.1 █████████████████████████▋ | hs/space-bs-c8-foldlx-1: 1.272 1‰ 0.2 ██████▌ | -- 1.205 2‰ 0.2 ██████▏ | -- 1.164 3‰ 0.0 █████▉ | -- 1.221 1‰ 0.1 ██████▏ | -- 1.176 1‰ 0.3 ██████ | -- 1.259 1‰ 0.1 ██████▍ | -- 1.176 1‰ 0.2 ██████ | hs/space-bslc8-foldlx-1: 1.430 1‰ 0.1 ███████▎ | -- 1.400 1‰ 0.1 ███████▏ | -- 1.197 1‰ 0.0 ██████ | -- 1.458 0‰ 0.1 ███████▍ | -- 1.060 1‰ 0.0 █████▍ | -- 1.457 1‰ 0.1 ███████▍ | -- 1.060 1‰ 0.2 █████▍ | hs/space-xxxxx-foldl: 5.024 5‰ 0.0 █████████████████████████▌ | -- 5.371 9‰ 0.0 ███████████████████████████▎ | -- 5.301 9‰ 0.1 ██████████████████████████▉ | -- 5.035 2‰ 0.1 █████████████████████████▌ | -- 5.037 3‰ 0.0 █████████████████████████▌ | -- 5.046 4‰ 0.0 █████████████████████████▌ | -- 5.042 4‰ 0.0 █████████████████████████▌ | -Peter

2 1

GtkMathView wrapper for Haskell / Gtk2HS?
by Peter Verswyvelen 23 Dec '07

23 Dec '07

Did anybody already wrap http://helm.cs.unibo.it/mml-widget so it can be used with Haskell / Gtk2HS? If not, would it be good project for me to learn Haskell's FFI? It involves C++, so I guess that would be harded to wrap? Thanks, Peter

2 1

Reducing the executable output filesize with GHC (on Windows)
by Peter Verswyvelen 23 Dec '07

23 Dec '07

I remember having read how to do this a while ago, but I can't google it back. How can I force the linker to exclude all the sections that are not referenced? I thought it had to do something with -split-objs? I've written a short consolemode program, but it compiles to 1.3MB. Since I will be creating a lot of these small apps, surely this filesize can be reduced? Thanks, Peter PS: The program I wrote is 250 lines of code. It is a simple vectormath exercise generator for my students which I wrote in 5 hours, something I would never been able to do in C++/C#, even though I have much more experience in the latter languages. Haskell's sooo nice, and I'm just scratching the surface of it. Thank to all of you guys for the hard work J

2 1

Is a type synonym declaration really a synonym ?
by alpheccar 23 Dec '07

23 Dec '07

Can someone confirm me that: type TA = A :+: B type TB = C :+: D type T = TA :+: TB is not equivalent to type T = A :+: B :+: C :+: D where I have defined infixr 6 :+: data (f :+: g) data A data B data C data D I have a computation at type level which is working with the later definition of T but not with the former (ghc 6.8.1) Thanks.

3 3

MonadFix
by Joost Behrends 23 Dec '07

23 Dec '07

Hi, since about three weeks i am learning Haskell now. One of my first excercises is to decompose an Integer into its primefactors. I already posted discussion on the solution to the problem 35 in "99 excercises". My simple algorithm uses a datatype DivIter of 4 named fields together with the core iteration divstep :: DivIter -> DivIter divstep x | divisor x > bound x = x | ximod > 0 = x { divisor = (divisor x) +2 } | otherwise = x {dividend=xidiv, bound=intsqrt(xidiv), result = result x ++ [divisor x] } where (xidiv, ximod) = divMod (dividend x) (divisor x) (dividend x is already odd, when this is called). The problem to solve for really large Integers is how to call divstep iterated without not accumulating billions of stack frames. Here is one possibility: divisions = do y <- get if divisor y <= bound y then do put ( divstep y ) divisions else return y (this for a version of divstep without the first guard) called from res = execState divisions (DivIter { dividend = o1, divisor = 3, bound = intsqrt(o1), result = o2 }) ( where o1 "the odd part" of the number to decompose, o2 a list of its "contained" twos). This computes the primefactors of 2^120+1 in 17 minutes after all. But i cannot help feeling that this is an abuse of the State monad. The MonadFix has a function fix (a -> a) -> a and i added the first guard in divstep above for making this a fixpoint problem. For me the signature looks, as if using fix doesn't afford to create explicitely a variable of a MonadFix instance and a simple twoliner for divisions could do the job. What i do not understand at all from the documentation of fix is: "fix f is the least fixed point of the function f, i.e. the least defined x such that f x = x." What does "least" mean here ? There is nothing said about x being a variable of an instance of Ord. And why fix has not the type a -> (a -> a) -> a, means: How can i provide a starting point of the iteration x ==> f x ==> f (f x) ==> ...?

7 17

Hoogle search scope question -- was: [HXT] Simple question
by Steve Lihn 23 Dec '07

23 Dec '07

I have a hoogle question. While I was reading the HXT discussion (below), I tried to search "runX" and "readString" in Hoogle (since I am new to HXT and Arrows). But neither search yielded any result and I had to use google to find the Haskell docs. So I am wondering what is the scope of Hoogle that people can search? Only things in the standard library? How much does it cover in the Hackage? I tried to find the scope description in the Tutorial but in vain. It is fair to say if someone needs to use Hoogle, he/she probably has no idea what he is looking for, needless to say, what is in scope of Hoogle or what is not. Thanks, Steve On Dec 19, 2007 7:48 AM, Miguel Mitrofanov <miguelimo38(a)yandex.ru> wrote: > Seems rather strange for me, I've just installed HXT and got this: > > Prelude Text.XML.HXT.Arrow> runX $ readString [(a_validate,"0")] "<this> </this>" >>> writeDocumentToString [] > ["<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<this> </this>"] > > > Everything works fine except for the fact that all the nodes <this> > > </this> > > (that is, a space (an XML text node whose contents are a single space > > character) > > within a <this> element node) get transformed to a <this/> element node, > > _______________________________________________ > Haskell-Cafe mailing list > Haskell-Cafe(a)haskell.org > http://www.haskell.org/mailman/listinfo/haskell-cafe >

2 1

Functions are first class values in C
by Cristian Baboi 22 Dec '07

22 Dec '07

Let me show you an example to prove it. The example is limited to composition of unary functions defined on int u::Int->Int v::Int->Int o ::(Int->Int)->(Int->Int)->(Int->Int) o u v= \x->u(v(x)) #include <stdio.h> #include <conio.h> #include "functional.h" int f1(int x){ return x+x; } int f2(int x){ return 2*x; } int g1(int x){ return x+1; } int g2(int x){ return x-1; } #define P1 P0 int main(){ #define P2 P1 printf("%d,%d,%d\n",2, #define P3 O(f1,f2,P2)(2), #define P4 O(g1,g2,P3)(3)); #define P5 P4 getch(); #define P6 P5 } MAIN P6 Here is the file functional.h #define FUNC2(x,y) x##y #define FUNC1(x,y) FUNC2(x,y) #define FUNC(x) FUNC1(x,__COUNTER__) #define COMP(c,f,g,p) \ int c (int x) { return f(g(x)); }; \ p \ c #define O(f,g,p) COMP( FUNC(a), f, g, p) #define P0 #define MAIN

6 9

Dynamic typing of polymorphic functions
by oleg＠okmij.org 22 Dec '07

22 Dec '07

Alfonso Acosta wrote: > mapSY :: (Typeable a, Typeable b) => (a -> b) -> Signal a -> Signal b > mapSY f (Signal primSig) = Signal (PrimSignal (MapSY (toDyn f) primSig)) > > The following process would be really useful but its compilation > obviously fails: > > mapSnd :: Signal (a, a) -> Signal a > mapSnd = mapSY snd > > > Could not deduce (Typeable a) from the context () arising from a > use of `mapSY' > Possible fix: > add (Typeable a) to the context of the type signature for `mapSnd' It seems the compiler's complaint is reasonable. The signature of the mapSY function says that mapSY may only be applied _provided_ that type variables 'a' and 'b' are instantiated to the types that are members of Typeable. That is, mapSY has a condition on its use. When you write > mapSndInt :: Signal (Int, Int) -> Signal Int > mapSndInt = mapSY (snd :: (Int, Int) -> Int) the condition is satisfied: 'a' and 'b' are instantiated to Int, and Int is a member of Typeable. The definition of mapSnd has no constraint. The compiler is upset: mapSY requires a condition, and mapSnd does not provide any, and there is no obvious way how an obligation Typeable a could have been satisfied otherwise. So, writing > mapSnd :: Typeable a => Signal (a, a) -> Signal a > mapSnd = mapSY snd is the logical thing to do. > Well, strangely enough, adding the "Typeable a" constraint hushed GHC > but an error was instead triggered at runtime. Perhaps the latter is the real problem. If one switches to dynamic typing, the type errors show up as run-time errors. I believe the typing of eval is a bit odd (and also not very useful). The following code seems to work. It also shows how to apply a polymorphic function, pairing, to to signals of any type. Here's a test: > signal3 = cons const0 (cons const0 const1) *Foo> :t signal3 signal3 :: Signal (Int, (Int, Float)) > test1 = mapSnd signal3 test1 :: Signal (Int, Float) > test12 = beval test1 *Foo> :t test12 test12 :: (Int, Float) *Foo> test12 (0,1.0) {-# OPTIONS -fglasgow-exts #-} module Foo where import Data.Typeable import Data.Dynamic -- the phantom type parameter makes signal typing consistent newtype Signal a = Signal PrimSignal newtype PrimSignal = PrimSignal (Proc (PrimSignal)) data Proc input = MapSY Dynamic -- The processing function input -- The process input -- the rest of the processes are omitted | Const Dynamic | Cons Dynamic input input eval :: PrimSignal -> Dynamic -- evaluates the output of a process for one input eval (PrimSignal (MapSY dynF dynIn)) = dynApp dynF (eval dynIn) eval (PrimSignal (Cons cns a1 a2)) = dynApp (dynApp cns (eval a1)) (eval a2) eval (PrimSignal (Const inp)) = inp -- better eval beval :: Typeable a => Signal a -> a beval (Signal s) = maybe undefined id (fromDynamic (eval s)) -- sample signals const0 :: Signal Int const0 = Signal (PrimSignal (Const (toDyn (0::Int)))) const1 :: Signal Float const1 = Signal (PrimSignal (Const (toDyn (1::Float)))) -- the map process constructor mapSY :: (Typeable a, Typeable b) => (a -> b) -> Signal a -> Signal b mapSY f (Signal primSig) = Signal (PrimSignal (MapSY (toDyn f) primSig)) add1 :: Signal Int -> Signal Int add1 = mapSY ((+1) :: Int -> Int) mapSndInt :: Signal (Int, Int) -> Signal Int mapSndInt = mapSY (snd :: (Int, Int) -> Int) -- it is important to give the signature to (,) below: we pack the cons -- function of the right type! cons :: forall a b. (Typeable a, Typeable b) => Signal a -> Signal b -> Signal (a,b) cons (Signal sig1) (Signal sig2) = Signal (PrimSignal (Cons (toDyn ((,)::a->b->(a,b))) sig1 sig2)) mapSnd :: (Typeable a, Typeable b) => Signal (b, a) -> Signal a mapSnd = mapSY snd signal3 = cons const0 (cons const0 const1) -- *Foo> :t signal3 -- signal3 :: Signal (Int, (Int, Float)) test1 = mapSnd signal3 -- test1 :: Signal (Int, Float) test11 = let Signal s = test1 in eval s -- *Foo> test11 -- <<(Int,Float)>> -- Too bad. But we can do better. test12 = beval test1 {- *Foo> :t test12 test12 :: (Int, Float) *Foo> test12 (0,1.0) -}

2 2

[14/16] SBM: Behind the measurements (rationale)
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

// I am getting sick and tired of working on this project and it's probably // better to get it fired off than polishing it any further. // // This email could benefit from being rewritten from a rough draft into a // well-crafted letter but that would take a couple of hours. // // So here it is, a lot rougher than I'd like -- but it *IS* :) why so big input files? easiest way to spot non-linearity and bad memory behaviour. Anyway, files should be big enough to overflow caches and kick the gc in. (short files interesting, too, but the big ones cause more complex behaviour of run-time system and CPU. If complex behaviour is behaved then simple behaviour probably is too -- but can still get its constant factors improved. If complex behaviour bad, then shouldn't that be fixed in any case?) waitpid4() has a struct w/ info about the child program's resource usage. unfortunately, the peakrss field is not filled in. Seems to be a general Unix problem. I've seen complaints on the net that Solaris doesn't fill it in, either. Other solution needed. pause-at-end, /proc/self/maps + /proc/self/status. VmmHWM = peak of VmmRSS, which is Resident (working) Set Size. It doesn't say what is shared with other processes or the operating system, though. In our case, we don't expect to share anything but some libraries -- which nobody else wants to share with us anyway (except for the C library). We are the only user of them. Discovered about a week ago that I could probably have used waitid() w/ WNOWAIT flag but didn't know. Was quick to write pause-at-end, anyway. It took about 15 minutes from the desire to know the peak memory use to having written and tested the first cut of it. Pause-at-end not completely bullet-proof in case of dyn libraries that get unloaded before the end of the program has been reached. On the other hand, plenty good enough for these tests + can conceivably allow more intricate poking around than waitid() solution. getting good measurements - eatmem, dd, probably should also dd library. good to have a "sacrificial run". Good to measure how good the measurements are (rel. std.dev + user/sys/real check). why average -- disturbances are mostly interrupts, daemons that everybody have anyway, slightly luckier/unluckier physical pages. These are real effects that nobody can control anyway. I'm not interested in the best possible times on an ideal, undisturbed machine with a helpful kernel. I'm interested in clean times under realistic circumstances. Therefore average instead of minimum. why I use real and not user/sys -- handling of blocking reads vs. mmap vs. madvise/fadvise vs. reading in separate thread in the future. User+sys would probably give me better numbers at the moment and I could change to real later. Still, I choose to stay with real (and the difference is marginal, anyway). Funny that the exact distribution of time between sys and user fluctuates a lot. In space-bslc8-lenfil-2 sys varies between 0.160s and 0.244s. Real is completely stable with 5x 1.396s and 1x 1.397s. look at /proc/interrupts, perhaps copy before/after to .intr? Warn if more than 100 (or 1000) Hz + 10%? write date/time + runlevel to platforminfo and/or sysinfo. barcharts why barcharts. should the time/mem barcharts be equal length? don't think so (hard to colour them in a text file. Would work with less -r and the console but not in an email or a text editor. Visual difference is good). But should perhaps not be /that/ different. visible markers if measurements bad. (5% real/user/sys check, typically within 0.1% on old laptop when doing a quick or thorough benchmark. Occasionally up to 1% - and 3% on c/byte-4k because it only takes 56ms in total.) prints out how tight the user/sys/real thing is. microarchitecture -- performance counters. Would be interesting to look at once the obvious performance problems have been handled. Let's fix the memory usage of bytestrings, the performance of lazy bytestrings, and start using registers in the machine code first. regularity of input file probably means that branch predictor on all three CPUs can remember pattern of spaces vs. non-spaces (or at least part of the pattern). Branch predictors not only use two-bit saturating counter for strongly non-taken/weakly non-taken/weakly taken/strongly taken. They also try to remember the pattern of jumps/non-jumps. A more realistic test would have less regular input file. This effect is very small given the current performance limiters, though. cache -- turned out to be pretty regular (by eyeballing cachegrind reports). Go up a factor of 10 in filesize and the number of access also went up a factor of 10. The miss ratios stayed the same. The miss ratios differed a bit between the benchmarks but I don't think it's time to look into that yet. The data are available, though, for those who can't wait to look into that. minor page faults we gather that through /usr/bin/time -- and could also get the same info from dumping the right file inside /proc/self/. Probably not important yet. Probably will be once all the low-lying fruit has been gathered up from the ground. More of a factor on slower OS'es than Linux. C files. Buffer size. reading it all in one go is slower than (re)using a small buffer. Cache effects, both in the operating system when copying (because the destination will be cached with a small buffer but non-cached with a big buffer) and in the application (everything will be cached with the small buffer, nothing with the big buffer). Note that at least the Core and the Athlon64 have automatic prefetchers that tries to fill the cache in advance so we don't have to wait for the cache misses. Doesn't quite seem to work. Older caches had a different write behaviour, they were write-through instead of the modern (lazy) write-back. For those caches the writing to the user-space buffer should be slow even when a small buffer is reused all the time (because we would have to wait for all the writes to be flushed out to main memory). C files. getchar/getchar_unlocked. A comparison with getchar() is NOT what the simple haskell program does [grammar!]. Thread-safe by default (cause of libraries). getchar_unlocked() is what the haskell programs do. getchar() and getchar_unlocked() use a single buffer for stdin. getwchar() and getwchar_unlocked() included at the insistence of wli. Much slower, because it is run-time dependent on locale (to choose encoding). Therefore, can't be a macro like getchar_unlocked() is. With an indirect jump, should be same speed as getchar() on Core and Athlon64 -- but curiously isn't. C and Haskell integer sizes and other limitations. Haskell uses unboxed 32-bit signed integers, except in lazy lenfil tests. Most of the C programs are simple and just use an int for the space count. One of them (space-megabuf)is more complicated. off_t is 64-bit. ssize_t is 32-bit. Potential overflow in c/space-megabuf. Potential 32-bit wrap-around in all my C tests. Same problem with all Haskell tests, except for the two lenfil tests that use lazy bytestrings, because they use a 64-bit int for the length of the intermediate string of just the spaces filtered out from stdin. Also potentially out of memory. In practice, they have almost the same limit because they use about 107MB for the 143MB input file. In other words, it will run out of virtual address space or RAM or swap at about the same time that the others will run out of bits in a 32-bit signed integer. -Peter

1 0

[13/16] SBM: Graphs that show the infidelity of -sstderr
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

Everybody seems to use +RTS -sstderr to measure the memory performance of their Haskell (ghc) code. That's what I started out doing, too. However, top seemed to occasionally disagree with the run-time system. I decided to dig deeper and ask the operating system at process exit time. I use LD_PRELOAD to get my code injected into each benchmarked process where it gets activated when the process is about to exit and copies /proc/self/maps and /proc/self/status to stderr where the test harness captures them. Unfortunately, -sstderr really *IS* unreliable! tools/regensstderr.sh and tools/sstderr.pl are two scripts that together generate bar graphs for an existing measurement tarball. I used the following shell code to generate graphs for 7 different ghc/bytestring combinations: (for F in ghc-thorough-6.6.1.tgz \ ghc-thorough-6.8.2.tgz ghc-thorough-6.8.2-bs0902.tgz \ ghc-thorough-6.9.tgz ghc-thorough-6.9-bs0902.tgz \ ghc-thorough-head.tgz ghc-thorough-head-bs0902.tgz; \ do \ printf "%s\n-------\n" $F; \ tools/regensstderr.sh $F; \ printf "\n\n"; \ done \ ) > xx The graphs clearly show that -sstderr sometimes severely underestimates the memory use and sometimes somewhat overestimates it. Moreover, the errors are not always for the same programs. This holds for all ghc versions I've tested: 6.6.1, 6.8.2, 6.9.20071119, and head as of noon 2007-12-19. -Peter ghc-thorough-6.6.1.tgz ------- hs/byte-bs----acc: 1 147448 ▏ |██████████████████▎ | hs/byte-bs----foldlx: 145 147468 ██████████▎ |██████████████████▎ | hs/byte-bs----foldrx: 1 147452 ▏ |██████████████████▎ | hs/byte-bsl---acc: 2 2872 ▏ |▍ | hs/byte-xxxxx-acc-1: 1 1572 ▏ |▎ | hs/byte-xxxxx-acc-2: 1 1568 ▏ |▎ | hs/byte-xxxxx-foldl: 1 1572 ▏ |▎ | hs/space-bs-c8-acc-1: 1 147448 ▏ |██████████████████▎ | hs/space-bs-c8-count: 1 147448 ▏ |██████████████████▎ | hs/space-bs-c8-foldlx-1: 145 147468 ██████████▎ |██████████████████▎ | hs/space-bs-c8-foldlx-2: 145 147468 ██████████▎ |██████████████████▎ | hs/space-bs-c8-foldrx: 1 147448 ▏ |██████████████████▎ | hs/space-bs-c8-lenfil: 145 147464 ██████████▎ |██████████████████▎ | hs/space-bslc8-acc-1: 2 2872 ▏ |▍ | hs/space-bslc8-acc-2: 2 2872 ▏ |▍ | hs/space-bslc8-acc-3: 2 2868 ▏ |▍ | hs/space-bslc8-count: 2 1476 ▏ |▏ | hs/space-bslc8-foldl: 2 1552 ▏ |▎ | hs/space-bslc8-foldlx-1: 2 1556 ▏ |▎ | hs/space-bslc8-foldlx-2: 2 1552 ▏ |▎ | hs/space-bslc8-foldr-1: 297 169480 █████████████████████|█████████████████████| hs/space-bslc8-foldr-2: 297 169480 █████████████████████|█████████████████████| hs/space-bslc8-lenfil-1: 38 38632 ██▋ |████▊ | hs/space-bslc8-lenfil-2: 38 38628 ██▋ |████▊ | hs/space-bsl---foldlx: 2 1560 ▏ |▎ | hs/space-xxxxx-acc-1: 1 1568 ▏ |▎ | hs/space-xxxxx-acc-2: 1 1572 ▏ |▎ | hs/space-xxxxx-foldl: 1 1572 ▏ |▎ | hs/space-xxxxx-lenfil: 1 1556 ▏ |▎ | c/byte-getchar: - - c/byte-getchar-u: - - c/byte-4k: - - c/space-getchar: - - c/space-getchar-u: - - c/space-4k: - - c/space-megabuf: - - c/space-getwchar: - - c/space-getwchar-u: - - c/space-32k: - - c/space-32k-8: - - ghc-thorough-6.8.2.tgz ------- hs/byte-bs----acc: 1 147500 ▏ |██████████████████▎ | hs/byte-bs----foldlx: 145 147528 ██████████▎ |██████████████████▎ | hs/byte-bs----foldrx: 1 147500 ▏ |██████████████████▎ | hs/byte-bsl---acc: 2 2920 ▏ |▍ | hs/byte-xxxxx-acc-1: 1 1632 ▏ |▎ | hs/byte-xxxxx-acc-2: 1 1636 ▏ |▎ | hs/byte-xxxxx-foldl: 1 1632 ▏ |▎ | hs/space-bs-c8-acc-1: 1 147492 ▏ |██████████████████▎ | hs/space-bs-c8-count: 1 147500 ▏ |██████████████████▎ | hs/space-bs-c8-foldlx-1: 145 147524 ██████████▎ |██████████████████▎ | hs/space-bs-c8-foldlx-2: 145 147520 ██████████▎ |██████████████████▎ | hs/space-bs-c8-foldrx: 1 147496 ▏ |██████████████████▎ | hs/space-bs-c8-lenfil: 145 147524 ██████████▎ |██████████████████▎ | hs/space-bslc8-acc-1: 2 2920 ▏ |▍ | hs/space-bslc8-acc-2: 2 2924 ▏ |▍ | hs/space-bslc8-acc-3: 2 2924 ▏ |▍ | hs/space-bslc8-chunk-1: 2 1512 ▏ |▏ | hs/space-bslc8-chunk-2: 2 1512 ▏ |▏ | hs/space-bslc8-chunk-3: 2 1548 ▏ |▎ | hs/space-bslc8-chunk-4: 2 1512 ▏ |▏ | hs/space-bslc8-count: 2 1520 ▏ |▎ | hs/space-bslc8-foldl: 2 1620 ▏ |▎ | hs/space-bslc8-foldlx-1: 2 1620 ▏ |▎ | hs/space-bslc8-foldlx-2: 2 1616 ▏ |▎ | hs/space-bslc8-foldr-1: 297 169544 █████████████████████|█████████████████████| hs/space-bslc8-foldr-2: 297 169544 █████████████████████|█████████████████████| hs/space-bslc8-lenfil-1: 38 38644 ██▋ |████▊ | hs/space-bslc8-lenfil-2: 38 38648 ██▋ |████▊ | hs/space-bsl---foldlx: 2 1616 ▏ |▎ | hs/space-xxxxx-acc-1: 1 1636 ▏ |▎ | hs/space-xxxxx-acc-2: 1 1636 ▏ |▎ | hs/space-xxxxx-foldl: 1 1636 ▏ |▎ | hs/space-xxxxx-lenfil: 1 1608 ▏ |▎ | c/byte-getchar: - - c/byte-getchar-u: - - c/byte-4k: - - c/space-getchar: - - c/space-getchar-u: - - c/space-4k: - - c/space-megabuf: - - c/space-getwchar: - - c/space-getwchar-u: - - c/space-32k: - - c/space-32k-8: - - ghc-thorough-6.8.2-bs0902.tgz ------- hs/byte-bs----acc: 1 147616 ▏ |██████████████████▌ | hs/byte-bs----foldlx: 1 147624 ▏ |██████████████████▌ | hs/byte-bs----foldrx: 1 147624 ▏ |██████████████████▌ | hs/byte-bsl---acc: 2 3040 ▎ |▍ | hs/byte-xxxxx-acc-1: 1 1636 ▏ |▎ | hs/byte-xxxxx-acc-2: 1 1636 ▏ |▎ | hs/byte-xxxxx-foldl: 1 1636 ▏ |▎ | hs/space-bs-c8-acc-1: 1 147608 ▏ |██████████████████▌ | hs/space-bs-c8-count: 1 147612 ▏ |██████████████████▌ | hs/space-bs-c8-foldlx-1: 1 147608 ▏ |██████████████████▌ | hs/space-bs-c8-foldlx-2: 1 147608 ▏ |██████████████████▌ | hs/space-bs-c8-foldrx: 1 147608 ▏ |██████████████████▌ | hs/space-bs-c8-lenfil: 21 167176 ███ |█████████████████████| hs/space-bslc8-acc-1: 2 3048 ▎ |▍ | hs/space-bslc8-acc-2: 2 3048 ▎ |▍ | hs/space-bslc8-acc-3: 2 3052 ▎ |▍ | hs/space-bslc8-chunk-1: 2 1636 ▎ |▎ | hs/space-bslc8-chunk-2: 2 1640 ▎ |▎ | hs/space-bslc8-chunk-3: 2 1644 ▎ |▎ | hs/space-bslc8-chunk-4: 2 1636 ▎ |▎ | hs/space-bslc8-count: 2 1656 ▎ |▎ | hs/space-bslc8-foldl: 2 1632 ▎ |▎ | hs/space-bslc8-foldlx-1: 2 1636 ▎ |▎ | hs/space-bslc8-foldlx-2: 2 1636 ▎ |▎ | hs/space-bslc8-foldr-1: 149 149076 █████████████████████|██████████████████▊ | hs/space-bslc8-foldr-2: 149 149076 █████████████████████|██████████████████▊ | hs/space-bslc8-lenfil-1: 2 1940 ▎ |▎ | hs/space-bslc8-lenfil-2: 2 1940 ▎ |▎ | hs/space-bsl---foldlx: 2 1620 ▎ |▎ | hs/space-xxxxx-acc-1: 1 1636 ▏ |▎ | hs/space-xxxxx-acc-2: 1 1636 ▏ |▎ | hs/space-xxxxx-foldl: 1 1636 ▏ |▎ | hs/space-xxxxx-lenfil: 1 1604 ▏ |▎ | c/byte-getchar: - - c/byte-getchar-u: - - c/byte-4k: - - c/space-getchar: - - c/space-getchar-u: - - c/space-4k: - - c/space-megabuf: - - c/space-getwchar: - - c/space-getwchar-u: - - c/space-32k: - - c/space-32k-8: - - ghc-thorough-6.9.tgz ------- hs/byte-bs----acc: 1 147752 ▏ |█████████████████▍ | hs/byte-bs----foldlx: 145 147748 █████████▊ |█████████████████▍ | hs/byte-bs----foldrx: 1 147744 ▏ |█████████████████▍ | hs/byte-bsl---acc: 2 3172 ▏ |▍ | hs/byte-xxxxx-acc-1: 1 1904 ▏ |▎ | hs/byte-xxxxx-acc-2: 1 1904 ▏ |▎ | hs/byte-xxxxx-foldl: 1 1900 ▏ |▎ | hs/space-bs-c8-acc-1: 1 147736 ▏ |█████████████████▍ | hs/space-bs-c8-count: 1 147744 ▏ |█████████████████▍ | hs/space-bs-c8-foldlx-1: 145 147756 █████████▊ |█████████████████▍ | hs/space-bs-c8-foldlx-2: 145 147752 █████████▊ |█████████████████▍ | hs/space-bs-c8-foldrx: 1 147732 ▏ |█████████████████▍ | hs/space-bs-c8-lenfil: 145 147756 █████████▊ |█████████████████▍ | hs/space-bslc8-acc-1: 2 3192 ▏ |▍ | hs/space-bslc8-acc-2: 2 3192 ▏ |▍ | hs/space-bslc8-acc-3: 2 3196 ▏ |▍ | hs/space-bslc8-chunk-1: 64 65192 ████▎ |███████▋ | hs/space-bslc8-chunk-2: 64 65192 ████▎ |███████▋ | hs/space-bslc8-chunk-3: 74 75240 █████ |████████▉ | hs/space-bslc8-chunk-4: 75 76188 █████ |█████████ | hs/space-bslc8-count: 78 79204 █████▎ |█████████▍ | hs/space-bslc8-foldl: 85 85868 █████▊ |██████████▏ | hs/space-bslc8-foldlx-1: 85 85864 █████▊ |██████████▏ | hs/space-bslc8-foldlx-2: 85 85868 █████▊ |██████████▏ | hs/space-bslc8-foldr-1: 297 169700 ████████████████████|████████████████████| hs/space-bslc8-foldr-2: 297 169700 ████████████████████|████████████████████| hs/space-bslc8-lenfil-1: 115 109404 ███████▊ |████████████▉ | hs/space-bslc8-lenfil-2: 115 109404 ███████▊ |████████████▉ | hs/space-bsl---foldlx: 85 85836 █████▊ |██████████▏ | hs/space-xxxxx-acc-1: 1 1904 ▏ |▎ | hs/space-xxxxx-acc-2: 1 1904 ▏ |▎ | hs/space-xxxxx-foldl: 1 1900 ▏ |▎ | hs/space-xxxxx-lenfil: 1 1892 ▏ |▎ | hand/byte-bs----acc-a: 1 147764 ▏ |█████████████████▍ | hand/byte-bs----acc-b: 1 147764 ▏ |█████████████████▍ | hand/byte-bs----acc-c: 1 147760 ▏ |█████████████████▍ | hand/byte-bs----acc-d: 1 147760 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-a: 1 147752 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-b: 1 147752 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-c: 1 147752 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-d: 1 147756 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-e: 1 147752 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-f: 1 147748 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-g: 1 147752 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-h: 1 147748 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-i: 1 147748 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-j: 1 147748 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-k: 1 147752 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-l: 1 147752 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-m: 1 147748 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-n: 1 147752 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-o: 1 147752 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-p: 1 147752 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-q: 1 147748 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-r: 1 147748 ▏ |█████████████████▍ | hand/space-bs-c8-acc-1-s: 1 147744 ▏ |█████████████████▍ | c/byte-getchar: - - c/byte-getchar-u: - - c/byte-4k: - - c/space-getchar: - - c/space-getchar-u: - - c/space-4k: - - c/space-megabuf: - - c/space-getwchar: - - c/space-getwchar-u: - - c/space-32k: - - c/space-32k-8: - - ghc-thorough-6.9-bs0902.tgz ------- hs/byte-bs----acc: 1 147732 ▏ |██████████████████▌ | hs/byte-bs----foldlx: 1 147736 ▏ |██████████████████▌ | hs/byte-bs----foldrx: 1 147740 ▏ |██████████████████▌ | hs/byte-bsl---acc: 2 3184 ▎ |▍ | hs/byte-xxxxx-acc-1: 1 1904 ▏ |▎ | hs/byte-xxxxx-acc-2: 1 1900 ▏ |▎ | hs/byte-xxxxx-foldl: 1 1904 ▏ |▎ | hs/space-bs-c8-acc-1: 1 147740 ▏ |██████████████████▌ | hs/space-bs-c8-count: 1 147756 ▏ |██████████████████▌ | hs/space-bs-c8-foldlx-1: 1 147744 ▏ |██████████████████▌ | hs/space-bs-c8-foldlx-2: 1 147748 ▏ |██████████████████▌ | hs/space-bs-c8-foldrx: 1 147748 ▏ |██████████████████▌ | hs/space-bs-c8-lenfil: 21 167300 ███ |█████████████████████| hs/space-bslc8-acc-1: 2 3228 ▎ |▍ | hs/space-bslc8-acc-2: 2 3228 ▎ |▍ | hs/space-bslc8-acc-3: 2 3232 ▎ |▍ | hs/space-bslc8-chunk-1: 65 66216 █████████▏ |████████▎ | hs/space-bslc8-chunk-2: 65 66216 █████████▏ |████████▎ | hs/space-bslc8-chunk-3: 65 66208 █████████▏ |████████▎ | hs/space-bslc8-chunk-4: 78 79232 ███████████ |██████████ | hs/space-bslc8-count: 81 82248 ███████████▍ |██████████▍ | hs/space-bslc8-foldl: 78 79224 ███████████ |██████████ | hs/space-bslc8-foldlx-1: 78 79224 ███████████ |██████████ | hs/space-bslc8-foldlx-2: 78 79224 ███████████ |██████████ | hs/space-bslc8-foldr-1: 149 149704 █████████████████████|██████████████████▊ | hs/space-bslc8-foldr-2: 149 149708 █████████████████████|██████████████████▊ | hs/space-bslc8-lenfil-1: 81 82324 ███████████▍ |██████████▍ | hs/space-bslc8-lenfil-2: 81 82328 ███████████▍ |██████████▍ | hs/space-bsl---foldlx: 78 79184 ███████████ |██████████ | hs/space-xxxxx-acc-1: 1 1904 ▏ |▎ | hs/space-xxxxx-acc-2: 1 1900 ▏ |▎ | hs/space-xxxxx-foldl: 1 1904 ▏ |▎ | hs/space-xxxxx-lenfil: 1 1892 ▏ |▎ | c/byte-getchar: - - c/byte-getchar-u: - - c/byte-4k: - - c/space-getchar: - - c/space-getchar-u: - - c/space-4k: - - c/space-megabuf: - - c/space-getwchar: - - c/space-getwchar-u: - - c/space-32k: - - c/space-32k-8: - - ghc-thorough-head.tgz ------- hs/byte-bs----acc: 1 147772 ▏ |██████████████████▎ | hs/byte-bs----foldlx: 145 147796 ██████████▎ |██████████████████▎ | hs/byte-bs----foldrx: 1 147772 ▏ |██████████████████▎ | hs/byte-bsl---acc: 2 3212 ▏ |▍ | hs/byte-xxxxx-acc-1: 1 1900 ▏ |▎ | hs/byte-xxxxx-acc-2: 1 1900 ▏ |▎ | hs/byte-xxxxx-foldl: 1 1900 ▏ |▎ | hs/space-bs-c8-acc-1: 1 147752 ▏ |██████████████████▎ | hs/space-bs-c8-count: 1 147764 ▏ |██████████████████▎ | hs/space-bs-c8-foldlx-1: 145 147796 ██████████▎ |██████████████████▎ | hs/space-bs-c8-foldlx-2: 145 147796 ██████████▎ |██████████████████▎ | hs/space-bs-c8-foldrx: 1 147760 ▏ |██████████████████▎ | hs/space-bs-c8-lenfil: 145 147788 ██████████▎ |██████████████████▎ | hs/space-bslc8-acc-1: 2 3180 ▏ |▍ | hs/space-bslc8-acc-2: 2 3180 ▏ |▍ | hs/space-bslc8-acc-3: 2 3172 ▏ |▍ | hs/space-bslc8-chunk-1: 64 65204 ████▌ |████████▏ | hs/space-bslc8-chunk-2: 64 65204 ████▌ |████████▏ | hs/space-bslc8-chunk-3: 74 75216 █████▎ |█████████▎ | hs/space-bslc8-chunk-4: 75 76176 █████▎ |█████████▍ | hs/space-bslc8-count: 78 79220 █████▌ |█████████▊ | hs/space-bslc8-foldl: 85 85852 ██████ |██████████▋ | hs/space-bslc8-foldlx-1: 85 85852 ██████ |██████████▋ | hs/space-bslc8-foldlx-2: 85 85852 ██████ |██████████▋ | hs/space-bslc8-foldr-1: 297 169680 █████████████████████|█████████████████████| hs/space-bslc8-foldr-2: 297 169676 █████████████████████|█████████████████████| hs/space-bslc8-lenfil-1: 115 109388 ████████▏ |█████████████▌ | hs/space-bslc8-lenfil-2: 115 109392 ████████▏ |█████████████▌ | hs/space-bsl---foldlx: 85 85864 ██████ |██████████▋ | hs/space-xxxxx-acc-1: 1 1900 ▏ |▎ | hs/space-xxxxx-acc-2: 1 1904 ▏ |▎ | hs/space-xxxxx-foldl: 1 1904 ▏ |▎ | hs/space-xxxxx-lenfil: 1 1888 ▏ |▎ | c/byte-getchar: - - c/byte-getchar-u: - - c/byte-4k: - - c/space-getchar: - - c/space-getchar-u: - - c/space-4k: - - c/space-megabuf: - - c/space-getwchar: - - c/space-getwchar-u: - - c/space-32k: - - c/space-32k-8: - - ghc-thorough-head-bs0902.tgz ------- hs/byte-bs----acc: 1 147780 ▏ |██████████████████▌ | hs/byte-bs----foldlx: 1 147780 ▏ |██████████████████▌ | hs/byte-bs----foldrx: 1 147784 ▏ |██████████████████▌ | hs/byte-bsl---acc: 2 3216 ▎ |▍ | hs/byte-xxxxx-acc-1: 1 1900 ▏ |▎ | hs/byte-xxxxx-acc-2: 1 1896 ▏ |▎ | hs/byte-xxxxx-foldl: 1 1896 ▏ |▎ | hs/space-bs-c8-acc-1: 1 147752 ▏ |██████████████████▌ | hs/space-bs-c8-count: 1 147776 ▏ |██████████████████▌ | hs/space-bs-c8-foldlx-1: 1 147752 ▏ |██████████████████▌ | hs/space-bs-c8-foldlx-2: 1 147756 ▏ |██████████████████▌ | hs/space-bs-c8-foldrx: 1 147756 ▏ |██████████████████▌ | hs/space-bs-c8-lenfil: 21 167320 ███ |█████████████████████| hs/space-bslc8-acc-1: 2 3196 ▎ |▍ | hs/space-bslc8-acc-2: 2 3200 ▎ |▍ | hs/space-bslc8-acc-3: 2 3208 ▎ |▍ | hs/space-bslc8-chunk-1: 65 66224 █████████▏ |████████▎ | hs/space-bslc8-chunk-2: 65 66228 █████████▏ |████████▎ | hs/space-bslc8-chunk-3: 65 66224 █████████▏ |████████▎ | hs/space-bslc8-chunk-4: 78 79204 ███████████ |██████████ | hs/space-bslc8-count: 81 82224 ███████████▍ |██████████▍ | hs/space-bslc8-foldl: 78 79196 ███████████ |██████████ | hs/space-bslc8-foldlx-1: 78 79196 ███████████ |██████████ | hs/space-bslc8-foldlx-2: 78 79196 ███████████ |██████████ | hs/space-bslc8-foldr-1: 149 149680 █████████████████████|██████████████████▊ | hs/space-bslc8-foldr-2: 149 149676 █████████████████████|██████████████████▊ | hs/space-bslc8-lenfil-1: 81 82304 ███████████▍ |██████████▍ | hs/space-bslc8-lenfil-2: 81 82300 ███████████▍ |██████████▍ | hs/space-bsl---foldlx: 78 79208 ███████████ |██████████ | hs/space-xxxxx-acc-1: 1 1904 ▏ |▎ | hs/space-xxxxx-acc-2: 1 1900 ▏ |▎ | hs/space-xxxxx-foldl: 1 1904 ▏ |▎ | hs/space-xxxxx-lenfil: 1 1888 ▏ |▎ | c/byte-getchar: - - c/byte-getchar-u: - - c/byte-4k: - - c/space-getchar: - - c/space-getchar-u: - - c/space-4k: - - c/space-megabuf: - - c/space-getwchar: - - c/space-getwchar-u: - - c/space-32k: - - c/space-32k-8: - -

1 0

[12/16] SBM: Graphs for 7 ghc/bytestring combinations on a 2GHz Athlon64 300+
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

This report combines the measurements from 7 ghc/bytestring combinations on a single machine, namely: ghc 6.6.1 ghc 6.8.2 ghc 6.8.2 + bytestring 0.9.0.2 ghc 6.9.20071119 ghc 6.9.20071119 + bytestring 0.9.0.2 ghc head-as-of-noon-2007-12-19 ghc head-as-of-noon-2007-12-19 + bytestring 0.9.0.2 This is the command I ran to generate the report: RESCALE=1 EXCLUDE='(^hand/|^c/)' \ tools/merge.pl \ ghc-thorough-6.6.1.tgz \ ghc-thorough-6.8.2.tgz ghc-thorough-6.8.2-bs0902.tgz \ ghc-thorough-6.9.tgz ghc-thorough-6.9-bs0902.tgz \ ghc-thorough-head.tgz ghc-thorough-head-bs0902.tgz \ > mergerep-amdy.txt We can readily see that there are some memory leaks with ghc 6.9.x. We can also see that Don Stewart's fix in bytestring 0.9.0.2 actually fixed things for hs/space-bslc8-lenfil-1 on ghc 6.8.2 but only improved things slightly for 6.9.x. It also makes things worse for hs/space-bs-c8-lenfil on 6.8.2 and both the 6.9.x's. Lazy bytestrings are slow, when used "manually", but quite acceptable, memory- wise. On the other hand, they cost dearly in memory on 6.9.x unless one does it all "by hand". Good old native strings seem to be the most predictable of the bunch but they are a bit slow. Surprisingly, they are not always /the/ slowest option measured! hs/space-bslc8-acc-1 and hs/space-bslc8-acc-2 are actually slower on ghc 6.6.1 than the native strings are in all combinations except hs/space-xxxxx-lenfil. -Peter PS: Note that this is the same report that I sent yesterday, except that it had the order of some of the benchmarks slightly wrong. charybdis ghc 6.6.1 AMD Athlon(tm) 64 Processor 3000+ 2009.160 MHz TESTKIND=THOROUGH SUFFIX= charybdis ghc 6.8.2 AMD Athlon(tm) 64 Processor 3000+ 2009.160 MHz TESTKIND=THOROUGH SUFFIX= charybdis ghc 6.8.2 AMD Athlon(tm) 64 Processor 3000+ 2009.160 MHz TESTKIND=THOROUGH SUFFIX=-bs0.9.0.2 charybdis ghc 6.9.20071119 AMD Athlon(tm) 64 Processor 3000+ 2009.160 MHz TESTKIND=THOROUGH SUFFIX= charybdis ghc 6.9.20071119 AMD Athlon(tm) 64 Processor 3000+ 2009.160 MHz TESTKIND=THOROUGH SUFFIX=-bs0.9.0.2 charybdis ghc 6.9.20071217 AMD Athlon(tm) 64 Processor 3000+ 2009.160 MHz TESTKIND=THOROUGH SUFFIX= charybdis ghc 6.9.20071217 AMD Athlon(tm) 64 Processor 3000+ 2009.160 MHz TESTKIND=THOROUGH SUFFIX=-bs0.9.0.2 Time (byte counting) std -------------------- avg dev slack hs/byte-bs----acc: 1.020 40‰ 0.0 █████▏ | -- 0.703 5‰ 0.4 ███▌ | -- 0.702 7‰ 0.3 ███▌ | -- 0.705 7‰ 0.1 ███▋ | -- 0.712 3‰ 0.1 ███▋ | -- 0.706 7‰ 0.1 ███▋ | -- 0.707 7‰ 0.9 ███▋ | hs/byte-bs----foldlx: 0.789 3‰ 0.3 ████ | -- 0.993 2‰ 0.2 █████ | -- 1.102 1‰ 0.2 █████▋ | -- 1.002 1‰ 0.5 █████▏ | -- 1.112 1‰ 0.3 █████▋ | -- 1.024 2‰ 0.2 █████▏ | -- 1.111 1‰ 0.1 █████▋ | hs/byte-bs----foldrx: 0.813 3‰ 0.1 ████▏ | -- 1.102 2‰ 0.3 █████▋ | -- 1.100 1‰ 0.1 █████▋ | -- 1.112 2‰ 0.1 █████▋ | -- 1.114 1‰ 0.4 █████▋ | -- 1.113 2‰ 0.5 █████▋ | -- 1.112 1‰ 0.2 █████▋ | hs/byte-bsl---acc: 3.599 13‰ 0.0 ██████████████████▎ | -- 2.609 17‰ 0.0 █████████████▎ | -- 2.560 15‰ 0.1 █████████████ | -- 2.595 14‰ 0.1 █████████████▏ | -- 2.574 16‰ 0.1 █████████████ | -- 2.613 12‰ 0.2 █████████████▎ | -- 2.656 44‰ 0.1 █████████████▌ | hs/byte-xxxxx-acc-1: 4.606 5‰ 0.0 ███████████████████████▍ | -- 4.730 3‰ 0.0 ████████████████████████ | -- 4.722 3‰ 0.1 ███████████████████████▉ | -- 4.436 5‰ 0.0 ██████████████████████▌ | -- 4.500 6‰ 0.1 ██████████████████████▊ | -- 4.462 4‰ 0.1 ██████████████████████▋ | -- 4.499 7‰ 0.1 ██████████████████████▊ | hs/byte-xxxxx-acc-2: 4.595 5‰ 0.0 ███████████████████████▎ | -- 4.704 1‰ 0.0 ███████████████████████▉ | -- 4.722 6‰ 0.0 ███████████████████████▉ | -- 4.473 8‰ 0.0 ██████████████████████▋ | -- 4.493 5‰ 0.0 ██████████████████████▊ | -- 4.477 3‰ 0.1 ██████████████████████▋ | -- 4.446 2‰ 0.0 ██████████████████████▌ | hs/byte-xxxxx-foldl: 4.566 9‰ 0.0 ███████████████████████▏ | -- 4.754 6‰ 0.1 ████████████████████████▏ | -- 4.735 7‰ 0.0 ████████████████████████ | -- 4.412 10‰ 0.0 ██████████████████████▍ | -- 4.486 3‰ 0.0 ██████████████████████▊ | -- 4.473 3‰ 0.0 ██████████████████████▋ | -- 4.511 10‰ 0.1 ██████████████████████▉ | Memory: Peak ------- KB hs/byte-bs----acc: 147448 ████████████████████████████████████████ | -- 147500 ████████████████████████████████████████ | -- 147616 ████████████████████████████████████████ | -- 147752 ████████████████████████████████████████ | -- 147732 ████████████████████████████████████████ | -- 147772 ████████████████████████████████████████ | -- 147780 ████████████████████████████████████████ | hs/byte-bs----foldlx: 147468 ████████████████████████████████████████ | -- 147528 ████████████████████████████████████████ | -- 147624 ████████████████████████████████████████ | -- 147748 ████████████████████████████████████████ | -- 147736 ████████████████████████████████████████ | -- 147796 ████████████████████████████████████████▏ | -- 147780 ████████████████████████████████████████ | hs/byte-bs----foldrx: 147452 ████████████████████████████████████████ | -- 147500 ████████████████████████████████████████ | -- 147624 ████████████████████████████████████████ | -- 147744 ████████████████████████████████████████ | -- 147740 ████████████████████████████████████████ | -- 147772 ████████████████████████████████████████ | -- 147784 ████████████████████████████████████████ | hs/byte-bsl---acc: 2872 ▊ | -- 2920 ▊ | -- 3040 ▉ | -- 3172 ▉ | -- 3184 ▉ | -- 3212 ▉ | -- 3216 ▉ | hs/byte-xxxxx-acc-1: 1572 ▍ | -- 1632 ▌ | -- 1636 ▌ | -- 1904 ▌ | -- 1904 ▌ | -- 1900 ▌ | -- 1900 ▌ | hs/byte-xxxxx-acc-2: 1568 ▍ | -- 1636 ▌ | -- 1636 ▌ | -- 1904 ▌ | -- 1900 ▌ | -- 1900 ▌ | -- 1896 ▌ | hs/byte-xxxxx-foldl: 1572 ▍ | -- 1632 ▌ | -- 1636 ▌ | -- 1900 ▌ | -- 1904 ▌ | -- 1900 ▌ | -- 1896 ▌ | Time (space counting) std --------------------- avg dev slack hs/space-bs-c8-acc-1: 1.287 20‰ 0.2 ██████▌ | -- 1.131 2‰ 0.1 █████▊ | -- 1.132 2‰ 0.3 █████▊ | -- 1.145 1‰ 0.2 █████▊ | -- 1.150 1‰ 0.2 █████▉ | -- 1.147 2‰ 0.1 █████▊ | -- 1.147 2‰ 0.2 █████▊ | hs/space-bs-c8-count: 0.770 2‰ 0.0 ███▉ | -- 0.711 1‰ 0.4 ███▋ | -- 0.513 2‰ 0.2 ██▋ | -- 0.521 1‰ 0.2 ██▋ | -- 0.523 3‰ 0.2 ██▋ | -- 0.522 5‰ 0.2 ██▋ | -- 0.523 3‰ 0.6 ██▋ | hs/space-bs-c8-foldlx-1: 1.272 1‰ 0.2 ██████▌ | -- 1.205 2‰ 0.2 ██████▏ | -- 1.164 3‰ 0.0 █████▉ | -- 1.221 1‰ 0.1 ██████▏ | -- 1.176 1‰ 0.3 ██████ | -- 1.259 1‰ 0.1 ██████▍ | -- 1.176 1‰ 0.2 ██████ | hs/space-bs-c8-foldlx-2: 1.270 1‰ 0.2 ██████▍ | -- 1.207 1‰ 0.2 ██████▏ | -- 1.165 2‰ 0.1 █████▉ | -- 1.219 2‰ 0.2 ██████▏ | -- 1.176 1‰ 0.3 ██████ | -- 1.258 1‰ 0.2 ██████▍ | -- 1.177 1‰ 0.2 ██████ | hs/space-bs-c8-foldrx: 1.064 2‰ 0.1 █████▍ | -- 1.168 2‰ 0.2 █████▉ | -- 1.164 1‰ 0.3 █████▉ | -- 1.172 2‰ 0.3 █████▉ | -- 1.177 1‰ 0.2 ██████ | -- 1.176 1‰ 0.2 ██████ | -- 1.176 1‰ 0.1 ██████ | hs/space-bs-c8-lenfil: 1.271 1‰ 0.1 ██████▍ | -- 1.218 1‰ 0.2 ██████▏ | -- 0.541 3‰ 0.2 ██▊ | -- 1.223 1‰ 0.2 ██████▎ | -- 0.549 1‰ 0.4 ██▊ | -- 1.225 1‰ 0.2 ██████▎ | -- 0.551 5‰ 0.4 ██▊ | hs/space-bslc8-acc-1: 6.367 17‰ 0.2 ████████████████████████████████▎ | -- 3.396 7‰ 0.1 █████████████████▎ | -- 3.508 4‰ 0.1 █████████████████▊ | -- 3.388 10‰ 0.1 █████████████████▏ | -- 3.441 10‰ 0.1 █████████████████▍ | -- 3.438 3‰ 0.0 █████████████████▍ | -- 3.434 13‰ 0.1 █████████████████▍ | hs/space-bslc8-acc-2: 6.150 21‰ 0.0 ███████████████████████████████▏ | -- 3.470 5‰ 0.0 █████████████████▋ | -- 3.439 10‰ 0.1 █████████████████▍ | -- 3.386 5‰ 0.1 █████████████████▏ | -- 3.479 11‰ 0.1 █████████████████▋ | -- 3.467 6‰ 0.1 █████████████████▌ | -- 3.470 10‰ 0.1 █████████████████▋ | hs/space-bslc8-acc-3: 4.041 21‰ 0.1 ████████████████████▌ | -- 2.959 13‰ 0.0 ███████████████ | -- 2.920 16‰ 0.1 ██████████████▊ | -- 2.832 4‰ 0.1 ██████████████▍ | -- 2.925 14‰ 0.1 ██████████████▊ | -- 2.987 9‰ 0.0 ███████████████▏ | -- 2.996 20‰ 0.1 ███████████████▏ | hs/space-bslc8-chunk-1: n/a -- 1.054 0‰ 0.3 █████▍ | -- 1.053 1‰ 0.1 █████▍ | -- 1.094 0‰ 0.3 █████▌ | -- 1.088 0‰ 0.0 █████▌ | -- 1.096 1‰ 0.1 █████▌ | -- 1.089 0‰ 0.6 █████▌ | hs/space-bslc8-chunk-2: n/a -- 1.055 0‰ 0.3 █████▍ | -- 1.053 0‰ 0.1 █████▍ | -- 1.094 1‰ 0.3 █████▌ | -- 1.088 1‰ 0.2 █████▌ | -- 1.096 1‰ 0.2 █████▌ | -- 1.089 0‰ 0.1 █████▌ | hs/space-bslc8-chunk-3: n/a -- 1.190 4‰ 0.2 ██████ | -- 1.199 1‰ 0.3 ██████▏ | -- 1.087 1‰ 0.3 █████▌ | -- 1.051 1‰ 0.4 █████▍ | -- 1.077 1‰ 0.1 █████▌ | -- 1.053 1‰ 0.2 █████▍ | hs/space-bslc8-chunk-4: n/a -- 1.054 0‰ 0.2 █████▍ | -- 1.054 0‰ 0.2 █████▍ | -- 1.099 0‰ 0.3 █████▋ | -- 1.097 1‰ 0.1 █████▌ | -- 1.094 1‰ 0.2 █████▌ | -- 1.096 0‰ 0.0 █████▌ | hs/space-bslc8-count: 0.713 1‰ 0.3 ███▋ | -- 0.735 1‰ 0.4 ███▊ | -- 0.431 0‰ 0.7 ██▏ | -- 0.477 2‰ 0.4 ██▍ | -- 0.482 1‰ 0.4 ██▌ | -- 0.479 1‰ 0.6 ██▍ | -- 0.482 2‰ 0.4 ██▌ | hs/space-bslc8-foldl: 1.431 1‰ 0.3 ███████▎ | -- 1.398 1‰ 0.2 ███████▏ | -- 1.197 1‰ 0.2 ██████ | -- 1.458 1‰ 0.3 ███████▍ | -- 1.059 0‰ 0.0 █████▍ | -- 1.459 1‰ 0.1 ███████▍ | -- 1.061 0‰ 0.4 █████▍ | hs/space-bslc8-foldlx-1: 1.430 1‰ 0.1 ███████▎ | -- 1.400 1‰ 0.1 ███████▏ | -- 1.197 1‰ 0.0 ██████ | -- 1.458 0‰ 0.1 ███████▍ | -- 1.060 1‰ 0.0 █████▍ | -- 1.457 1‰ 0.1 ███████▍ | -- 1.060 1‰ 0.2 █████▍ | hs/space-bslc8-foldlx-2: 1.431 1‰ 0.1 ███████▎ | -- 1.399 1‰ 0.1 ███████▏ | -- 1.196 1‰ 0.1 ██████ | -- 1.457 1‰ 0.1 ███████▍ | -- 1.060 1‰ 0.3 █████▍ | -- 1.458 1‰ 0.1 ███████▍ | -- 1.060 1‰ 0.0 █████▍ | hs/space-bslc8-foldr-1: 1.748 1‰ 0.0 ████████▉ | -- 1.579 2‰ 0.1 ████████ | -- 1.220 2‰ 0.2 ██████▏ | -- 1.312 1‰ 0.2 ██████▋ | -- 1.213 1‰ 0.2 ██████▏ | -- 1.325 4‰ 0.2 ██████▊ | -- 1.213 2‰ 0.2 ██████▏ | hs/space-bslc8-foldr-2: 1.745 1‰ 0.1 ████████▉ | -- 1.579 2‰ 0.1 ████████ | -- 1.220 1‰ 0.2 ██████▏ | -- 1.315 1‰ 0.2 ██████▋ | -- 1.210 1‰ 0.2 ██████▏ | -- 1.317 4‰ 0.2 ██████▋ | -- 1.213 2‰ 0.2 ██████▏ | hs/space-bslc8-lenfil-1: 1.521 7‰ 0.2 ███████▊ | -- 1.295 1‰ 0.2 ██████▌ | -- 0.429 0‰ 0.2 ██▏ | -- 1.323 1‰ 0.1 ██████▊ | -- 0.483 2‰ 0.4 ██▌ | -- 1.327 1‰ 0.2 ██████▊ | -- 0.482 2‰ 0.4 ██▌ | hs/space-bslc8-lenfil-2: 1.532 6‰ 0.1 ███████▊ | -- 1.296 0‰ 0.0 ██████▌ | -- 0.429 0‰ 0.2 ██▏ | -- 1.325 1‰ 0.1 ██████▊ | -- 0.483 1‰ 0.6 ██▌ | -- 1.327 1‰ 0.2 ██████▊ | -- 0.482 2‰ 0.4 ██▌ | hs/space-bsl---foldlx: 1.288 1‰ 0.2 ██████▌ | -- 1.333 1‰ 0.0 ██████▊ | -- 1.168 1‰ 0.2 █████▉ | -- 1.391 1‰ 0.1 ███████ | -- 1.006 1‰ 0.3 █████▏ | -- 1.395 1‰ 0.1 ███████▏ | -- 1.008 1‰ 0.2 █████▏ | hs/space-xxxxx-acc-1: 5.036 7‰ 0.1 █████████████████████████▌ | -- 5.381 6‰ 0.0 ███████████████████████████▎ | -- 5.358 8‰ 0.1 ███████████████████████████▏ | -- 4.991 1‰ 0.1 █████████████████████████▎ | -- 5.047 7‰ 0.1 █████████████████████████▌ | -- 5.051 3‰ 0.1 █████████████████████████▋ | -- 5.049 4‰ 0.1 █████████████████████████▋ | hs/space-xxxxx-acc-2: 5.009 7‰ 0.1 █████████████████████████▍ | -- 5.437 4‰ 0.1 ███████████████████████████▌ | -- 5.406 2‰ 0.0 ███████████████████████████▍ | -- 4.985 1‰ 0.0 █████████████████████████▎ | -- 5.057 4‰ 0.1 █████████████████████████▋ | -- 5.048 6‰ 0.1 █████████████████████████▌ | -- 5.089 6‰ 0.0 █████████████████████████▊ | hs/space-xxxxx-foldl: 5.024 5‰ 0.0 █████████████████████████▌ | -- 5.371 9‰ 0.0 ███████████████████████████▎ | -- 5.301 9‰ 0.1 ██████████████████████████▉ | -- 5.035 2‰ 0.1 █████████████████████████▌ | -- 5.037 3‰ 0.0 █████████████████████████▌ | -- 5.046 4‰ 0.0 █████████████████████████▌ | -- 5.042 4‰ 0.0 █████████████████████████▌ | hs/space-xxxxx-lenfil: 7.042 4‰ 0.0 ███████████████████████████████████▋ | -- 7.505 3‰ 0.0 ██████████████████████████████████████| -- 7.498 3‰ 0.0 ██████████████████████████████████████| -- 7.106 8‰ 0.1 ████████████████████████████████████ | -- 7.092 3‰ 0.1 ███████████████████████████████████▉ | -- 7.083 3‰ 0.0 ███████████████████████████████████▉ | -- 7.121 5‰ 0.1 ████████████████████████████████████ | Memory: Peak ------- KB hs/space-bs-c8-acc-1: 147448 ████████████████████████████████████████ | -- 147492 ████████████████████████████████████████ | -- 147608 ████████████████████████████████████████ | -- 147736 ████████████████████████████████████████ | -- 147740 ████████████████████████████████████████ | -- 147752 ████████████████████████████████████████ | -- 147752 ████████████████████████████████████████ | hs/space-bs-c8-count: 147448 ████████████████████████████████████████ | -- 147500 ████████████████████████████████████████ | -- 147612 ████████████████████████████████████████ | -- 147744 ████████████████████████████████████████ | -- 147756 ████████████████████████████████████████ | -- 147764 ████████████████████████████████████████ | -- 147776 ████████████████████████████████████████ | hs/space-bs-c8-foldlx-1: 147468 ████████████████████████████████████████ | -- 147524 ████████████████████████████████████████ | -- 147608 ████████████████████████████████████████ | -- 147756 ████████████████████████████████████████ | -- 147744 ████████████████████████████████████████ | -- 147796 ████████████████████████████████████████▏ | -- 147752 ████████████████████████████████████████ | hs/space-bs-c8-foldlx-2: 147468 ████████████████████████████████████████ | -- 147520 ████████████████████████████████████████ | -- 147608 ████████████████████████████████████████ | -- 147752 ████████████████████████████████████████ | -- 147748 ████████████████████████████████████████ | -- 147796 ████████████████████████████████████████▏ | -- 147756 ████████████████████████████████████████ | hs/space-bs-c8-foldrx: 147448 ████████████████████████████████████████ | -- 147496 ████████████████████████████████████████ | -- 147608 ████████████████████████████████████████ | -- 147732 ████████████████████████████████████████ | -- 147748 ████████████████████████████████████████ | -- 147760 ████████████████████████████████████████ | -- 147756 ████████████████████████████████████████ | hs/space-bs-c8-lenfil: 147464 ████████████████████████████████████████ | -- 147524 ████████████████████████████████████████ | -- 167176 █████████████████████████████████████████████▍| -- 147756 ████████████████████████████████████████ | -- 167300 █████████████████████████████████████████████▍| -- 147788 ████████████████████████████████████████ | -- 167320 █████████████████████████████████████████████▍| hs/space-bslc8-acc-1: 2872 ▊ | -- 2920 ▊ | -- 3048 ▉ | -- 3192 ▉ | -- 3228 ▉ | -- 3180 ▉ | -- 3196 ▉ | hs/space-bslc8-acc-2: 2872 ▊ | -- 2924 ▊ | -- 3048 ▉ | -- 3192 ▉ | -- 3228 ▉ | -- 3180 ▉ | -- 3200 ▉ | hs/space-bslc8-acc-3: 2868 ▊ | -- 2924 ▊ | -- 3052 ▉ | -- 3196 ▉ | -- 3232 ▉ | -- 3172 ▉ | -- 3208 ▉ | hs/space-bslc8-chunk-1: n/a -- 1512 ▍ | -- 1636 ▌ | -- 65192 █████████████████▋ | -- 66216 ██████████████████ | -- 65204 █████████████████▋ | -- 66224 ██████████████████ | hs/space-bslc8-chunk-2: n/a -- 1512 ▍ | -- 1640 ▌ | -- 65192 █████████████████▋ | -- 66216 ██████████████████ | -- 65204 █████████████████▋ | -- 66228 ██████████████████ | hs/space-bslc8-chunk-3: n/a -- 1548 ▍ | -- 1644 ▌ | -- 75240 ████████████████████▍ | -- 66208 ██████████████████ | -- 75216 ████████████████████▍ | -- 66224 ██████████████████ | hs/space-bslc8-chunk-4: n/a -- 1512 ▍ | -- 1636 ▌ | -- 76188 ████████████████████▋ | -- 79232 █████████████████████▌ | -- 76176 ████████████████████▋ | -- 79204 █████████████████████▌ | hs/space-bslc8-count: 1476 ▍ | -- 1520 ▍ | -- 1656 ▌ | -- 79204 █████████████████████▌ | -- 82248 ██████████████████████▎ | -- 79220 █████████████████████▌ | -- 82224 ██████████████████████▎ | hs/space-bslc8-foldl: 1552 ▍ | -- 1620 ▌ | -- 1632 ▌ | -- 85868 ███████████████████████▎ | -- 79224 █████████████████████▌ | -- 85852 ███████████████████████▎ | -- 79196 █████████████████████▌ | hs/space-bslc8-foldlx-1: 1556 ▍ | -- 1620 ▌ | -- 1636 ▌ | -- 85864 ███████████████████████▎ | -- 79224 █████████████████████▌ | -- 85852 ███████████████████████▎ | -- 79196 █████████████████████▌ | hs/space-bslc8-foldlx-2: 1552 ▍ | -- 1616 ▌ | -- 1636 ▌ | -- 85868 ███████████████████████▎ | -- 79224 █████████████████████▌ | -- 85852 ███████████████████████▎ | -- 79196 █████████████████████▌ | hs/space-bslc8-foldr-1: 169480 ██████████████████████████████████████████████| -- 169544 ██████████████████████████████████████████████| -- 149076 ████████████████████████████████████████▍ | -- 169700 ██████████████████████████████████████████████| -- 149704 ████████████████████████████████████████▋ | -- 169680 ██████████████████████████████████████████████| -- 149680 ████████████████████████████████████████▋ | hs/space-bslc8-foldr-2: 169480 ██████████████████████████████████████████████| -- 169544 ██████████████████████████████████████████████| -- 149076 ████████████████████████████████████████▍ | -- 169700 ██████████████████████████████████████████████| -- 149708 ████████████████████████████████████████▋ | -- 169676 ██████████████████████████████████████████████| -- 149676 ████████████████████████████████████████▋ | hs/space-bslc8-lenfil-1: 38632 ██████████▌ | -- 38644 ██████████▌ | -- 1940 ▌ | -- 109404 █████████████████████████████▋ | -- 82324 ██████████████████████▍ | -- 109388 █████████████████████████████▋ | -- 82304 ██████████████████████▎ | hs/space-bslc8-lenfil-2: 38628 ██████████▌ | -- 38648 ██████████▌ | -- 1940 ▌ | -- 109404 █████████████████████████████▋ | -- 82328 ██████████████████████▍ | -- 109392 █████████████████████████████▋ | -- 82300 ██████████████████████▎ | hs/space-bsl---foldlx: 1560 ▍ | -- 1616 ▌ | -- 1620 ▌ | -- 85836 ███████████████████████▎ | -- 79184 █████████████████████▌ | -- 85864 ███████████████████████▎ | -- 79208 █████████████████████▌ | hs/space-xxxxx-acc-1: 1568 ▍ | -- 1636 ▌ | -- 1636 ▌ | -- 1904 ▌ | -- 1904 ▌ | -- 1900 ▌ | -- 1904 ▌ | hs/space-xxxxx-acc-2: 1572 ▍ | -- 1636 ▌ | -- 1636 ▌ | -- 1904 ▌ | -- 1900 ▌ | -- 1904 ▌ | -- 1900 ▌ | hs/space-xxxxx-foldl: 1572 ▍ | -- 1636 ▌ | -- 1636 ▌ | -- 1900 ▌ | -- 1904 ▌ | -- 1904 ▌ | -- 1904 ▌ | hs/space-xxxxx-lenfil: 1556 ▍ | -- 1608 ▍ | -- 1604 ▍ | -- 1892 ▌ | -- 1892 ▌ | -- 1888 ▌ | -- 1888 ▌ |

1 0

[11/16] SBM: Graphs for hand-tweaked assembly benchmarks
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

This report compares the hand-tweaked assembly programs with the original untweaked programs on two vastly different microarchitectures. This is the command I ran to generate the report: EXCLUDE='(xxxx|-bsl|chunk|count|acc-[23]|fold|lenfil|^c/)' \ tools/merge.pl \ ghc-armada-thorough-6.9.tgz \ ghc-thorough-6.9.tgz \ > xx I cut out the memory sections manually since we've already seen them and inserted a few newlines for grouping purposes. The first one should note is that not all tweaks are better than the originals! The second is that the sequence of tweaks is not quite monotonically decreasing in run-time. The improvements don't really start until -e on the Athlon64 and -f on both. Not until then have the load pressure been sufficiently relieved on the L1 cache that the code actually runs faster. Note also how the two microarchitectures seem to have plateaus in different places. The Athlon64 seems to have the number 3 built into its silicon (efg, jkl, mno) which fits very well with what we know about it from AMD's documentation (the front end splits the instructions up into smaller pieces which then get distributed to three different "pipelines", each with its own out-of-order execution engine). The Pentium III seems to have trouble with the simple MMX code but does very well with the more advanced MMX code that keeps 8 space counters in a single MMX register for many iterations. The code I used to add those counters horizontally is the same in both -q and -r. Perhaps operations on both MMX and normal registers are slow? Loop unrolling (-s) doesn't seem to matter, in this case. -Peter ls-search ghc 6.9.20071119 Pentium III (Coppermine) 596.932 MHz TESTKIND=THOROUGH SUFFIX= charybdis ghc 6.9.20071119 AMD Athlon(tm) 64 Processor 3000+ 2009.160 MHz TESTKIND=THOROUGH SUFFIX= Time (byte counting) std -------------------- avg dev slack hs/byte-bs----acc: 3.274 1‰ 0.1 ███████████████████████████ | -- 0.705 7‰ 0.1 █████████████████████▋ | hand/byte-bs----acc-a: 3.511 1‰ 0.0 ████████████████████████████▉ | -- 0.639 2‰ 0.2 ███████████████████▋ | hand/byte-bs----acc-b: 1.998 2‰ 0.1 ████████████████▌ | -- 0.414 2‰ 0.5 ████████████▊ | hand/byte-bs----acc-c: 1.876 2‰ 0.1 ███████████████▌ | -- 0.414 3‰ 0.2 ████████████▊ | hand/byte-bs----acc-d: 1.876 1‰ 0.1 ███████████████▌ | -- 0.415 3‰ 0.2 ████████████▊ | Time (space counting) std --------------------- avg dev slack hs/space-bs-c8-acc-1: 4.318 1‰ 0.0 ███████████████████████████████████▋ | -- 1.145 1‰ 0.2 ███████████████████████████████████▏ | hand/space-bs-c8-acc-1-a: 4.318 1‰ 0.0 ███████████████████████████████████▋ | -- 1.177 2‰ 0.3 ████████████████████████████████████▏| hand/space-bs-c8-acc-1-b: 4.331 1‰ 0.0 ███████████████████████████████████▋ | -- 1.104 1‰ 0.2 █████████████████████████████████▉ | hand/space-bs-c8-acc-1-c: 4.492 1‰ 0.1 █████████████████████████████████████| -- 1.207 1‰ 0.3 █████████████████████████████████████| hand/space-bs-c8-acc-1-d: 4.354 1‰ 0.0 ███████████████████████████████████▉ | -- 1.191 1‰ 0.2 ████████████████████████████████████▌| hand/space-bs-c8-acc-1-e: 4.424 0‰ 0.1 ████████████████████████████████████▌| -- 0.937 1‰ 0.2 ████████████████████████████▊ | hand/space-bs-c8-acc-1-f: 4.164 1‰ 0.0 ██████████████████████████████████▎ | -- 0.921 1‰ 0.2 ████████████████████████████▎ | hand/space-bs-c8-acc-1-g: 4.309 1‰ 0.1 ███████████████████████████████████▌ | -- 0.927 2‰ 0.4 ████████████████████████████▍ | hand/space-bs-c8-acc-1-h: 4.202 1‰ 0.1 ██████████████████████████████████▋ | -- 0.886 2‰ 0.2 ███████████████████████████▏ | hand/space-bs-c8-acc-1-i: 3.820 1‰ 0.1 ███████████████████████████████▌ | -- 0.803 3‰ 0.4 ████████████████████████▋ | hand/space-bs-c8-acc-1-j: 3.472 1‰ 0.0 ████████████████████████████▋ | -- 0.706 2‰ 0.1 █████████████████████▋ | hand/space-bs-c8-acc-1-k: 3.474 1‰ 0.0 ████████████████████████████▋ | -- 0.705 1‰ 0.0 █████████████████████▋ | hand/space-bs-c8-acc-1-l: 3.498 1‰ 0.1 ████████████████████████████▉ | -- 0.710 2‰ 0.1 █████████████████████▊ | hand/space-bs-c8-acc-1-m: 3.397 1‰ 0.1 ████████████████████████████ | -- 0.642 6‰ 0.3 ███████████████████▋ | hand/space-bs-c8-acc-1-n: 3.373 1‰ 0.0 ███████████████████████████▊ | -- 0.636 4‰ 0.5 ███████████████████▌ | hand/space-bs-c8-acc-1-o: 3.118 1‰ 0.1 █████████████████████████▋ | -- 0.626 2‰ 0.0 ███████████████████▎ | hand/space-bs-c8-acc-1-p: 2.935 2‰ 0.0 ████████████████████████▏ | -- 0.565 3‰ 0.4 █████████████████▍ | hand/space-bs-c8-acc-1-q: 3.477 1‰ 0.1 ████████████████████████████▋ | -- 0.418 6‰ 0.7 ████████████▉ | hand/space-bs-c8-acc-1-r: 1.674 1‰ 0.1 █████████████▊ | -- 0.334 5‰ 0.6 ██████████▎ | hand/space-bs-c8-acc-1-s: 1.627 1‰ 0.2 █████████████▍ | -- 0.335 4‰ 0.9 ██████████▎ |

1 0

[10/16] SBM: Graphs for 6.9.x across four cpus
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

This is what you get if you merge the previous four reports (and filter out the hand-tweaked assembly benchmarks). I generated the report with the following command: EXCLUDE='^hand/' \ tools/merge.pl \ ghc-armada-thorough-6.9.tgz \ ghc-fischer-thorough-6.9.tgz \ ghc-albatross-thorough-6.9.tgz \ ghc-thorough-6.9.tgz \ > xx This graph shows the memory usage to be (almost) exactly the same. The difference can easily be explained with slightly different versions of the C library (there was also a security update or two from Ubuntu in the last few days - the kernel was definitely updated and I think the C library was too). We are also talking about two versions of Ubuntu and one (old) version of SuSE. And we are talking about three not quite identical versions of ghc. The fischer and albatross runs were made before the inclusion of the three benchmarks from Don Stewart (hs/space-bs-c8-count, hs/space-bslc8-count, and hs/space-bslc8-chunk-4) so they are a couple of holes in the graphs. The speed pattern is more fun. It makes no sense to compare absolute times here so the graphs were not rescaled. One would naïvely expect the bars to be about the same when taken in groups of four but that really turns out to be far from the case! The explanation is most likely that we are looking at four very different microarchitectures. ********** This should hammer home the point that benchmarking on any single machine isn't enough! ********** Two probably aren't enough either... -Peter ls-search ghc 6.9.20071119 Pentium III (Coppermine) 596.932 MHz TESTKIND=THOROUGH SUFFIX= Fischer's machine ghc 6.9.20071124 AMD Duron(tm) processor 1200.089 MHz TESTKIND=THOROUGH SUFFIX= albatross ghc 6.9.20071208 (or thereabouts) Genuine Intel(R) CPU T2300 @ 1.66GHz (Core Duo) 1667.000 MHz TESTKIND=THOROUGH SUFFIX= charybdis ghc 6.9.20071119 AMD Athlon(tm) 64 Processor 3000+ 2009.160 MHz TESTKIND=THOROUGH SUFFIX= Time (byte counting) std -------------------- avg dev slack hs/byte-bs----acc: 3.274 1‰ 0.1 ███▏ | -- 1.892 21‰ 0.4 ██▊ | -- 0.918 4‰ 0.1 █████▍ | -- 0.705 7‰ 0.1 ███▋ | hs/byte-bs----foldlx: 4.027 0‰ 0.0 ███▊ | -- 2.258 3‰ 0.1 ███▎ | -- 1.014 2‰ 0.2 ██████ | -- 1.002 1‰ 0.5 █████▎ | hs/byte-bs----foldrx: 4.184 1‰ 0.0 ███▉ | -- 2.933 0‰ 0.1 ████▎ | -- 0.999 2‰ 0.1 █████▉ | -- 1.112 2‰ 0.1 █████▊ | hs/byte-bsl---acc: 28.005 10‰ 0.0 ██████████████████████████▍ | -- 14.319 45‰ 0.1 ████████████████████▌ | -- 1.957 1‰ 0.1 ███████████▌ | -- 2.595 14‰ 0.1 █████████████▌ | hs/byte-xxxxx-acc-1: 25.852 4‰ 0.0 ████████████████████████▎ | -- 20.915 17‰ 4.0 █████████████████████████████▉ | -- 3.591 2‰ 0.1 █████████████████████▏ | -- 4.436 5‰ 0.0 ███████████████████████▏ | hs/byte-xxxxx-acc-2: 25.622 5‰ 0.0 ████████████████████████▏ | -- 20.691 8‰ 0.1 █████████████████████████████▌ | -- 3.598 4‰ 0.1 █████████████████████▏ | -- 4.473 8‰ 0.0 ███████████████████████▎ | hs/byte-xxxxx-foldl: 25.803 5‰ 0.0 ████████████████████████▎ | -- 20.610 5‰ 1.4 █████████████████████████████▍ | -- 3.594 1‰ 0.1 █████████████████████▏ | -- 4.412 10‰ 0.0 ███████████████████████ | c/byte-getchar: 13.016 0‰ 0.0 ████████████▎ | -- 9.042 0‰ 0.1 ████████████▉ | -- 3.761 2‰ 0.1 ██████████████████████▏ | -- 2.422 27‰ 0.1 ████████████▋ | c/byte-getchar-u: 1.662 1‰ 0.1 █▋ | -- 1.314 3‰ 0.2 █▉ | -- 0.719 2‰ 0.3 ████▎ | -- 0.632 3‰ 0.3 ███▎ | c/byte-4k: 0.543 2‰ 0.2 ▌ | -- 0.419 5‰ 0.5 ▋ | -- 0.101 10‰ 3.1 ▋ | -- 0.094 26‰ 2.2 ▌ | Memory: Peak ------- KB hs/byte-bs----acc: 147752 ████████████████████████████████████████ | -- 147492 ████████████████████████████████████████ | -- 147784 ████████████████████████████████████████▏ | -- 147752 ████████████████████████████████████████ | hs/byte-bs----foldlx: 147756 ████████████████████████████████████████ | -- 147492 ████████████████████████████████████████ | -- 147812 ████████████████████████████████████████▏ | -- 147748 ████████████████████████████████████████ | hs/byte-bs----foldrx: 147760 ████████████████████████████████████████ | -- 147488 ████████████████████████████████████████ | -- 147788 ████████████████████████████████████████▏ | -- 147744 ████████████████████████████████████████ | hs/byte-bsl---acc: 3180 ▉ | -- 2896 ▊ | -- 3228 ▉ | -- 3172 ▉ | hs/byte-xxxxx-acc-1: 1916 ▌ | -- 1612 ▌ | -- 1904 ▌ | -- 1904 ▌ | hs/byte-xxxxx-acc-2: 1912 ▌ | -- 1612 ▌ | -- 1908 ▌ | -- 1904 ▌ | hs/byte-xxxxx-foldl: 1912 ▌ | -- 1612 ▌ | -- 1908 ▌ | -- 1900 ▌ | c/byte-getchar: 436 ▏ | -- 384 ▏ | -- 432 ▏ | -- 440 ▏ | c/byte-getchar-u: 432 ▏ | -- 384 ▏ | -- 440 ▏ | -- 440 ▏ | c/byte-4k: 436 ▏ | -- 380 ▏ | -- 436 ▏ | -- 440 ▏ | Time (space counting) std --------------------- avg dev slack hs/space-bs-c8-acc-1: 4.318 1‰ 0.0 ████ | -- 2.467 1‰ 0.3 ███▌ | -- 1.313 0‰ 0.2 ███████▊ | -- 1.145 1‰ 0.2 ██████ | hs/space-bs-c8-count: 3.118 1‰ 0.1 ██▉ | -- n/a -- n/a -- 0.521 1‰ 0.2 ██▊ | hs/space-bs-c8-foldlx-1: 4.631 1‰ 0.0 ████▍ | -- 2.585 2‰ 0.1 ███▊ | -- 1.326 1‰ 0.1 ███████▊ | -- 1.221 1‰ 0.1 ██████▍ | hs/space-bs-c8-foldlx-2: 4.632 1‰ 0.0 ████▍ | -- 2.576 2‰ 0.3 ███▋ | -- 1.328 1‰ 0.1 ███████▉ | -- 1.219 2‰ 0.2 ██████▍ | hs/space-bs-c8-foldrx: 4.678 0‰ 0.0 ████▍ | -- 2.982 8‰ 2.3 ████▎ | -- 1.128 1‰ 0.2 ██████▋ | -- 1.172 2‰ 0.3 ██████▏ | hs/space-bs-c8-lenfil: 4.634 1‰ 0.1 ████▍ | -- 2.599 1‰ 0.2 ███▊ | -- 1.327 2‰ 0.2 ███████▊ | -- 1.223 1‰ 0.2 ██████▍ | hs/space-bslc8-acc-1: 32.733 7‰ 0.0 ██████████████████████████████▊ | -- 15.228 8‰ 0.1 █████████████████████▊ | -- 3.002 1‰ 0.1 █████████████████▋ | -- 3.388 10‰ 0.1 █████████████████▋ | hs/space-bslc8-acc-2: 33.112 11‰ 0.0 ███████████████████████████████▏ | -- 15.855 38‰ 0.0 ██████████████████████▋ | -- 3.010 2‰ 0.0 █████████████████▊ | -- 3.386 5‰ 0.1 █████████████████▋ | hs/space-bslc8-acc-3: 30.294 12‰ 0.0 ████████████████████████████▌ | -- 14.980 14‰ 0.0 █████████████████████▍ | -- 2.586 1‰ 0.1 ███████████████▎ | -- 2.832 4‰ 0.1 ██████████████▊ | hs/space-bslc8-chunk-1: 3.888 0‰ 0.1 ███▋ | -- 2.443 2‰ 0.2 ███▌ | -- 1.192 0‰ 0.0 ███████ | -- 1.094 0‰ 0.3 █████▊ | hs/space-bslc8-chunk-2: 3.887 1‰ 0.0 ███▋ | -- 2.449 1‰ 0.3 ███▌ | -- 1.189 0‰ 0.1 ███████ | -- 1.094 1‰ 0.3 █████▊ | hs/space-bslc8-chunk-3: 3.881 0‰ 0.1 ███▋ | -- 2.534 3‰ 0.3 ███▋ | -- 1.272 1‰ 0.2 ███████▌ | -- 1.087 1‰ 0.3 █████▋ | hs/space-bslc8-chunk-4: 3.910 1‰ 0.0 ███▋ | -- n/a -- n/a -- 1.099 0‰ 0.3 █████▊ | hs/space-bslc8-count: 2.355 1‰ 0.1 ██▎ | -- n/a -- n/a -- 0.477 2‰ 0.4 ██▌ | hs/space-bslc8-foldl: 4.674 0‰ 0.0 ████▍ | -- 2.938 1‰ 0.2 ████▎ | -- 1.517 0‰ 0.1 ████████▉ | -- 1.458 1‰ 0.3 ███████▋ | hs/space-bslc8-foldlx-1: 4.676 0‰ 0.0 ████▍ | -- 2.928 1‰ 0.0 ████▏ | -- 1.516 0‰ 0.0 ████████▉ | -- 1.458 0‰ 0.1 ███████▋ | hs/space-bslc8-foldlx-2: 4.676 0‰ 0.0 ████▍ | -- 2.937 2‰ 0.2 ████▎ | -- 1.511 0‰ 0.0 ████████▉ | -- 1.457 1‰ 0.1 ███████▋ | hs/space-bslc8-foldr-1: 5.222 0‰ 0.0 ████▉ | -- 4.043 6‰ 0.1 █████▊ | -- 1.205 1‰ 0.2 ███████▏ | -- 1.312 1‰ 0.2 ██████▉ | hs/space-bslc8-foldr-2: 5.224 1‰ 0.1 ████▉ | -- 4.007 4‰ 0.1 █████▊ | -- 1.202 3‰ 0.1 ███████▏ | -- 1.315 1‰ 0.2 ██████▉ | hs/space-bslc8-lenfil-1: 4.712 1‰ 0.0 ████▍ | -- 3.240 1‰ 0.2 ████▋ | -- 1.400 0‰ 0.0 ████████▎ | -- 1.323 1‰ 0.1 ██████▉ | hs/space-bslc8-lenfil-2: 4.714 1‰ 0.0 ████▍ | -- 3.236 1‰ 0.2 ████▋ | -- 1.396 0‰ 0.0 ████████▎ | -- 1.325 1‰ 0.1 ██████▉ | hs/space-bsl---foldlx: 4.522 0‰ 0.1 ████▎ | -- 2.821 1‰ 0.1 ████ | -- 1.476 1‰ 0.1 ████████▋ | -- 1.391 1‰ 0.1 ███████▎ | hs/space-xxxxx-acc-1: 27.529 6‰ 0.0 █████████████████████████▉ | -- 21.002 4‰ 0.1 ██████████████████████████████ | -- 4.245 1‰ 0.0 █████████████████████████ | -- 4.991 1‰ 0.1 ██████████████████████████ | hs/space-xxxxx-acc-2: 27.200 3‰ 0.0 █████████████████████████▋ | -- 21.270 22‰ 4.7 ██████████████████████████████▍ | -- 4.241 2‰ 0.1 █████████████████████████ | -- 4.985 1‰ 0.0 ██████████████████████████ | hs/space-xxxxx-foldl: 27.436 8‰ 0.0 █████████████████████████▊ | -- 20.934 1‰ 0.1 █████████████████████████████▉ | -- 4.239 2‰ 0.1 █████████████████████████ | -- 5.035 2‰ 0.1 ██████████████████████████▎ | hs/space-xxxxx-lenfil: 39.343 3‰ 0.0 █████████████████████████████████████| -- 25.915 3‰ 0.0 █████████████████████████████████████| -- 6.287 2‰ 0.0 █████████████████████████████████████| -- 7.106 8‰ 0.1 █████████████████████████████████████| c/space-getchar: 13.268 0‰ 0.0 ████████████▌ | -- 9.354 0‰ 0.0 █████████████▍ | -- 4.031 18‰ 0.0 ███████████████████████▊ | -- 2.465 15‰ 0.1 ████████████▉ | c/space-getchar-u: 2.170 0‰ 0.1 ██ | -- 1.676 2‰ 0.2 ██▍ | -- 0.609 2‰ 0.7 ███▋ | -- 0.631 1‰ 0.5 ███▎ | c/space-4k: 1.815 1‰ 0.1 █▊ | -- 1.293 2‰ 0.5 █▉ | -- 0.465 2‰ 0.6 ██▊ | -- 0.319 3‰ 0.6 █▋ | c/space-megabuf: 2.763 2‰ 0.1 ██▋ | -- 1.830 3‰ 0.5 ██▋ | -- 0.613 1‰ 0.3 ███▋ | -- 0.515 3‰ 0.4 ██▋ | c/space-getwchar: 15.105 29‰ 0.0 ██████████████▎ | -- 14.721 1‰ 0.1 █████████████████████ | -- 4.584 1‰ 0.1 ███████████████████████████ | -- 3.956 6‰ 0.1 ████████████████████▋ | c/space-getwchar-u: 8.979 1‰ 0.0 ████████▌ | -- 4.814 0‰ 0.0 ██████▉ | -- 2.437 0‰ 0.1 ██████████████▍ | -- 2.217 1‰ 0.1 ███████████▌ | c/space-32k: 1.786 9‰ 0.1 █▋ | -- 1.276 2‰ 0.6 █▉ | -- 0.445 1‰ 0.5 ██▋ | -- 0.303 3‰ 1.0 █▋ | c/space-32k-8: 1.777 1‰ 0.1 █▋ | -- 1.275 2‰ 0.2 █▉ | -- 0.448 0‰ 0.0 ██▋ | -- 0.304 2‰ 0.3 █▋ | Memory: Peak ------- KB hs/space-bs-c8-acc-1: 147748 ████████████████████████████████████████ | -- 147488 ████████████████████████████████████████ | -- 147772 ████████████████████████████████████████ | -- 147736 ████████████████████████████████████████ | hs/space-bs-c8-count: 147760 ████████████████████████████████████████ | -- n/a -- n/a -- 147744 ████████████████████████████████████████ | hs/space-bs-c8-foldlx-1: 147760 ████████████████████████████████████████ | -- 147492 ████████████████████████████████████████ | -- 147800 ████████████████████████████████████████▏ | -- 147756 ████████████████████████████████████████ | hs/space-bs-c8-foldlx-2: 147764 ████████████████████████████████████████ | -- 147492 ████████████████████████████████████████ | -- 147804 ████████████████████████████████████████▏ | -- 147752 ████████████████████████████████████████ | hs/space-bs-c8-foldrx: 147748 ████████████████████████████████████████ | -- 147488 ████████████████████████████████████████ | -- 147776 ████████████████████████████████████████ | -- 147732 ████████████████████████████████████████ | hs/space-bs-c8-lenfil: 147764 ████████████████████████████████████████ | -- 147492 ████████████████████████████████████████ | -- 147796 ████████████████████████████████████████▏ | -- 147756 ████████████████████████████████████████ | hs/space-bslc8-acc-1: 3196 ▉ | -- 2896 ▊ | -- 3192 ▉ | -- 3192 ▉ | hs/space-bslc8-acc-2: 3200 ▉ | -- 2896 ▊ | -- 3192 ▉ | -- 3192 ▉ | hs/space-bslc8-acc-3: 3204 ▉ | -- 2896 ▊ | -- 3188 ▉ | -- 3196 ▉ | hs/space-bslc8-chunk-1: 65204 █████████████████▋ | -- 65892 █████████████████▉ | -- 65208 █████████████████▋ | -- 65192 █████████████████▋ | hs/space-bslc8-chunk-2: 65200 █████████████████▋ | -- 65892 █████████████████▉ | -- 65212 █████████████████▋ | -- 65192 █████████████████▋ | hs/space-bslc8-chunk-3: 75248 ████████████████████▍ | -- 76472 ████████████████████▊ | -- 75232 ████████████████████▍ | -- 75240 ████████████████████▍ | hs/space-bslc8-chunk-4: 76204 ████████████████████▋ | -- n/a -- n/a -- 76188 ████████████████████▋ | hs/space-bslc8-count: 79224 █████████████████████▌ | -- n/a -- n/a -- 79204 █████████████████████▌ | hs/space-bslc8-foldl: 85880 ███████████████████████▎ | -- 86772 ███████████████████████▋ | -- 85856 ███████████████████████▎ | -- 85868 ███████████████████████▎ | hs/space-bslc8-foldlx-1: 85880 ███████████████████████▎ | -- 86772 ███████████████████████▋ | -- 85852 ███████████████████████▎ | -- 85864 ███████████████████████▎ | hs/space-bslc8-foldlx-2: 85880 ███████████████████████▎ | -- 86772 ███████████████████████▋ | -- 85856 ███████████████████████▎ | -- 85868 ███████████████████████▎ | hs/space-bslc8-foldr-1: 169704 ██████████████████████████████████████████████| -- 169360 ██████████████████████████████████████████████| -- 169684 ██████████████████████████████████████████████| -- 169700 ██████████████████████████████████████████████| hs/space-bslc8-foldr-2: 169700 ██████████████████████████████████████████████| -- 169360 ██████████████████████████████████████████████| -- 169684 ██████████████████████████████████████████████| -- 169700 ██████████████████████████████████████████████| hs/space-bslc8-lenfil-1: 109408 █████████████████████████████▋ | -- 110704 ██████████████████████████████▏ | -- 109404 █████████████████████████████▋ | -- 109404 █████████████████████████████▋ | hs/space-bslc8-lenfil-2: 109404 █████████████████████████████▋ | -- 110704 ██████████████████████████████▏ | -- 109404 █████████████████████████████▋ | -- 109404 █████████████████████████████▋ | hs/space-bsl---foldlx: 85844 ███████████████████████▎ | -- 86776 ███████████████████████▋ | -- 85880 ███████████████████████▎ | -- 85836 ███████████████████████▎ | hs/space-xxxxx-acc-1: 1912 ▌ | -- 1612 ▌ | -- 1916 ▌ | -- 1904 ▌ | hs/space-xxxxx-acc-2: 1912 ▌ | -- 1612 ▌ | -- 1916 ▌ | -- 1904 ▌ | hs/space-xxxxx-foldl: 1912 ▌ | -- 1612 ▌ | -- 1916 ▌ | -- 1900 ▌ | hs/space-xxxxx-lenfil: 1904 ▌ | -- 1588 ▍ | -- 1892 ▌ | -- 1892 ▌ | c/space-getchar: 436 ▏ | -- 384 ▏ | -- 436 ▏ | -- 444 ▏ | c/space-getchar-u: 436 ▏ | -- 384 ▏ | -- 436 ▏ | -- 444 ▏ | c/space-4k: 472 ▏ | -- 412 ▏ | -- 468 ▏ | -- 468 ▏ | c/space-megabuf: 146968 ███████████████████████████████████████▉ | -- 146904 ███████████████████████████████████████▉ | -- 146968 ███████████████████████████████████████▉ | -- 146968 ███████████████████████████████████████▉ | c/space-getwchar: 484 ▏ | -- 440 ▏ | -- 484 ▏ | -- 496 ▏ | c/space-getwchar-u: 484 ▏ | -- 440 ▏ | -- 480 ▏ | -- 500 ▏ | c/space-32k: 500 ▏ | -- 436 ▏ | -- 496 ▏ | -- 496 ▏ | c/space-32k-8: 496 ▏ | -- 436 ▏ | -- 500 ▏ | -- 500 ▏ |

1 0

[9/16] SBM: 6.9.20071119 Athlon64
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

This set of measurements was captured by me on my slightly old noname 2GHz Athlon64 3000+ (in 32-bit mode on a 32-bit kernel). It has 1GB RAM and runs Ubuntu 7.04. The benchmarks were run two days ago (2007-12-19) using ghc 6.9.20071119 in runlevel 1 (single-user mode). I did leave the network cable in, which might have produced some interrupts (not very likely -- 1) why would it run in promiscuous mode and 2) who broadcasts apart from my wired/wireless access point which broadcasts a packet per second?). Still, the quality of the timings proved to be good, with a standard deviation of at most 2.7% of the average run-time. This run includes the hand-tweaked assembly benchmarks (as does the Pentium III run in the previous email). Note how much the assembler tweaks help! -Peter charybdis ghc 6.9.20071119 AMD Athlon(tm) 64 Processor 3000+ 2009.160 MHz TESTKIND=THOROUGH SUFFIX= Time (byte counting) std -------------------- avg dev slack hs/byte-bs----acc: 0.705 7‰ 0.1 ███▋ | hs/byte-bs----foldlx: 1.002 1‰ 0.5 █████▎ | hs/byte-bs----foldrx: 1.112 2‰ 0.1 █████▊ | hs/byte-bsl---acc: 2.595 14‰ 0.1 █████████████▌ | hs/byte-xxxxx-acc-1: 4.436 5‰ 0.0 ███████████████████████▏ | hs/byte-xxxxx-acc-2: 4.473 8‰ 0.0 ███████████████████████▎ | hs/byte-xxxxx-foldl: 4.412 10‰ 0.0 ███████████████████████ | hand/byte-bs----acc-a: 0.639 2‰ 0.2 ███▍ | hand/byte-bs----acc-b: 0.414 2‰ 0.5 ██▏ | hand/byte-bs----acc-c: 0.414 3‰ 0.2 ██▏ | hand/byte-bs----acc-d: 0.415 3‰ 0.2 ██▏ | c/byte-getchar: 2.422 27‰ 0.1 ████████████▋ | c/byte-getchar-u: 0.632 3‰ 0.3 ███▎ | c/byte-4k: 0.094 26‰ 2.2 ▌ | Memory: Peak ------- KB hs/byte-bs----acc: 147752 ███████████████████████████████████████▏ | hs/byte-bs----foldlx: 147748 ███████████████████████████████████████▏ | hs/byte-bs----foldrx: 147744 ███████████████████████████████████████▏ | hs/byte-bsl---acc: 3172 ▉ | hs/byte-xxxxx-acc-1: 1904 ▌ | hs/byte-xxxxx-acc-2: 1904 ▌ | hs/byte-xxxxx-foldl: 1900 ▌ | hand/byte-bs----acc-a: 147764 ███████████████████████████████████████▏ | hand/byte-bs----acc-b: 147764 ███████████████████████████████████████▏ | hand/byte-bs----acc-c: 147760 ███████████████████████████████████████▏ | hand/byte-bs----acc-d: 147760 ███████████████████████████████████████▏ | c/byte-getchar: 440 ▏ | c/byte-getchar-u: 440 ▏ | c/byte-4k: 440 ▏ | Time (space counting) std --------------------- avg dev slack hs/space-bs-c8-acc-1: 1.145 1‰ 0.2 ██████ | hs/space-bs-c8-count: 0.521 1‰ 0.2 ██▊ | hs/space-bs-c8-foldlx-1: 1.221 1‰ 0.1 ██████▍ | hs/space-bs-c8-foldlx-2: 1.219 2‰ 0.2 ██████▍ | hs/space-bs-c8-foldrx: 1.172 2‰ 0.3 ██████▏ | hs/space-bs-c8-lenfil: 1.223 1‰ 0.2 ██████▍ | hs/space-bslc8-acc-1: 3.388 10‰ 0.1 █████████████████▋ | hs/space-bslc8-acc-2: 3.386 5‰ 0.1 █████████████████▋ | hs/space-bslc8-acc-3: 2.832 4‰ 0.1 ██████████████▊ | hs/space-bslc8-chunk-1: 1.094 0‰ 0.3 █████▊ | hs/space-bslc8-chunk-2: 1.094 1‰ 0.3 █████▊ | hs/space-bslc8-chunk-3: 1.087 1‰ 0.3 █████▋ | hs/space-bslc8-chunk-4: 1.099 0‰ 0.3 █████▊ | hs/space-bslc8-count: 0.477 2‰ 0.4 ██▌ | hs/space-bslc8-foldl: 1.458 1‰ 0.3 ███████▋ | hs/space-bslc8-foldlx-1: 1.458 0‰ 0.1 ███████▋ | hs/space-bslc8-foldlx-2: 1.457 1‰ 0.1 ███████▋ | hs/space-bslc8-foldr-1: 1.312 1‰ 0.2 ██████▉ | hs/space-bslc8-foldr-2: 1.315 1‰ 0.2 ██████▉ | hs/space-bslc8-lenfil-1: 1.323 1‰ 0.1 ██████▉ | hs/space-bslc8-lenfil-2: 1.325 1‰ 0.1 ██████▉ | hs/space-bsl---foldlx: 1.391 1‰ 0.1 ███████▎ | hs/space-xxxxx-acc-1: 4.991 1‰ 0.1 ██████████████████████████ | hs/space-xxxxx-acc-2: 4.985 1‰ 0.0 ██████████████████████████ | hs/space-xxxxx-foldl: 5.035 2‰ 0.1 ██████████████████████████▎ | hs/space-xxxxx-lenfil: 7.106 8‰ 0.1 █████████████████████████████████████| hand/space-bs-c8-acc-1-a: 1.177 2‰ 0.3 ██████▏ | hand/space-bs-c8-acc-1-b: 1.104 1‰ 0.2 █████▊ | hand/space-bs-c8-acc-1-c: 1.207 1‰ 0.3 ██████▎ | hand/space-bs-c8-acc-1-d: 1.191 1‰ 0.2 ██████▎ | hand/space-bs-c8-acc-1-e: 0.937 1‰ 0.2 ████▉ | hand/space-bs-c8-acc-1-f: 0.921 1‰ 0.2 ████▊ | hand/space-bs-c8-acc-1-g: 0.927 2‰ 0.4 ████▉ | hand/space-bs-c8-acc-1-h: 0.886 2‰ 0.2 ████▋ | hand/space-bs-c8-acc-1-i: 0.803 3‰ 0.4 ████▏ | hand/space-bs-c8-acc-1-j: 0.706 2‰ 0.1 ███▋ | hand/space-bs-c8-acc-1-k: 0.705 1‰ 0.0 ███▋ | hand/space-bs-c8-acc-1-l: 0.710 2‰ 0.1 ███▊ | hand/space-bs-c8-acc-1-m: 0.642 6‰ 0.3 ███▍ | hand/space-bs-c8-acc-1-n: 0.636 4‰ 0.5 ███▎ | hand/space-bs-c8-acc-1-o: 0.626 2‰ 0.0 ███▎ | hand/space-bs-c8-acc-1-p: 0.565 3‰ 0.4 ███ | hand/space-bs-c8-acc-1-q: 0.418 6‰ 0.7 ██▏ | hand/space-bs-c8-acc-1-r: 0.334 5‰ 0.6 █▊ | hand/space-bs-c8-acc-1-s: 0.335 4‰ 0.9 █▊ | c/space-getchar: 2.465 15‰ 0.1 ████████████▉ | c/space-getchar-u: 0.631 1‰ 0.5 ███▎ | c/space-4k: 0.319 3‰ 0.6 █▋ | c/space-megabuf: 0.515 3‰ 0.4 ██▋ | c/space-getwchar: 3.956 6‰ 0.1 ████████████████████▋ | c/space-getwchar-u: 2.217 1‰ 0.1 ███████████▌ | c/space-32k: 0.303 3‰ 1.0 █▋ | c/space-32k-8: 0.304 2‰ 0.3 █▋ | Memory: Peak ------- KB hs/space-bs-c8-acc-1: 147736 ███████████████████████████████████████▏ | hs/space-bs-c8-count: 147744 ███████████████████████████████████████▏ | hs/space-bs-c8-foldlx-1: 147756 ███████████████████████████████████████▏ | hs/space-bs-c8-foldlx-2: 147752 ███████████████████████████████████████▏ | hs/space-bs-c8-foldrx: 147732 ███████████████████████████████████████▏ | hs/space-bs-c8-lenfil: 147756 ███████████████████████████████████████▏ | hs/space-bslc8-acc-1: 3192 ▉ | hs/space-bslc8-acc-2: 3192 ▉ | hs/space-bslc8-acc-3: 3196 ▉ | hs/space-bslc8-chunk-1: 65192 █████████████████▎ | hs/space-bslc8-chunk-2: 65192 █████████████████▎ | hs/space-bslc8-chunk-3: 75240 ████████████████████ | hs/space-bslc8-chunk-4: 76188 ████████████████████▎ | hs/space-bslc8-count: 79204 █████████████████████ | hs/space-bslc8-foldl: 85868 ██████████████████████▊ | hs/space-bslc8-foldlx-1: 85864 ██████████████████████▊ | hs/space-bslc8-foldlx-2: 85868 ██████████████████████▊ | hs/space-bslc8-foldr-1: 169700 █████████████████████████████████████████████| hs/space-bslc8-foldr-2: 169700 █████████████████████████████████████████████| hs/space-bslc8-lenfil-1: 109404 █████████████████████████████ | hs/space-bslc8-lenfil-2: 109404 █████████████████████████████ | hs/space-bsl---foldlx: 85836 ██████████████████████▊ | hs/space-xxxxx-acc-1: 1904 ▌ | hs/space-xxxxx-acc-2: 1904 ▌ | hs/space-xxxxx-foldl: 1900 ▌ | hs/space-xxxxx-lenfil: 1892 ▌ | hand/space-bs-c8-acc-1-a: 147752 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-b: 147752 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-c: 147752 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-d: 147756 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-e: 147752 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-f: 147748 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-g: 147752 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-h: 147748 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-i: 147748 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-j: 147748 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-k: 147752 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-l: 147752 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-m: 147748 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-n: 147752 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-o: 147752 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-p: 147752 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-q: 147748 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-r: 147748 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-s: 147744 ███████████████████████████████████████▏ | c/space-getchar: 444 ▏ | c/space-getchar-u: 444 ▏ | c/space-4k: 468 ▏ | c/space-megabuf: 146968 ███████████████████████████████████████ | c/space-getwchar: 496 ▏ | c/space-getwchar-u: 500 ▏ | c/space-32k: 496 ▏ | c/space-32k-8: 500 ▏ |

1 0

[8/16] SBM: 6.9.20071119 Pentium III
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

This set of measurements was captured by me on my old Compaq Armada E500 from around the year 2000. It has an Intel Coppermine Pentium III running at 600 MHz with 384 MB RAM and running Ubuntu 7.10. The benchmarks were run today (2007-12-21) using ghc 6.9.20071119 (freshly downloaded and reinstalled) in runlevel 1 (single-user mode). I did leave the wireless card in, which might have produced some interrupts. Still, the quality of the timings proved to be good, with a standard deviation of at most 1.2% of the average run-time. This run includes the hand-tweaked assembly benchmarks (as does the Athlon64 run in the next email). Note how much less the assembler tweaks help here (until one gets down to the wicked MMX tweaks) compared to the situation on the Athlon64 in the next mail. They still help, though. -Peter ls-search ghc 6.9.20071119 Pentium III (Coppermine) 596.932 MHz TESTKIND=THOROUGH SUFFIX= Time (byte counting) std -------------------- avg dev slack hs/byte-bs----acc: 3.274 1‰ 0.1 ███ | hs/byte-bs----foldlx: 4.027 0‰ 0.0 ███▋ | hs/byte-bs----foldrx: 4.184 1‰ 0.0 ███▉ | hs/byte-bsl---acc: 28.005 10‰ 0.0 █████████████████████████▋ | hs/byte-xxxxx-acc-1: 25.852 4‰ 0.0 ███████████████████████▋ | hs/byte-xxxxx-acc-2: 25.622 5‰ 0.0 ███████████████████████▌ | hs/byte-xxxxx-foldl: 25.803 5‰ 0.0 ███████████████████████▋ | hand/byte-bs----acc-a: 3.511 1‰ 0.0 ███▎ | hand/byte-bs----acc-b: 1.998 2‰ 0.1 █▉ | hand/byte-bs----acc-c: 1.876 2‰ 0.1 █▊ | hand/byte-bs----acc-d: 1.876 1‰ 0.1 █▊ | c/byte-getchar: 13.016 0‰ 0.0 ███████████▉ | c/byte-getchar-u: 1.662 1‰ 0.1 █▌ | c/byte-4k: 0.543 2‰ 0.2 ▌ | Memory: Peak ------- KB hs/byte-bs----acc: 147752 ███████████████████████████████████████▏ | hs/byte-bs----foldlx: 147756 ███████████████████████████████████████▏ | hs/byte-bs----foldrx: 147760 ███████████████████████████████████████▏ | hs/byte-bsl---acc: 3180 ▉ | hs/byte-xxxxx-acc-1: 1916 ▌ | hs/byte-xxxxx-acc-2: 1912 ▌ | hs/byte-xxxxx-foldl: 1912 ▌ | hand/byte-bs----acc-a: 147772 ███████████████████████████████████████▏ | hand/byte-bs----acc-b: 147776 ███████████████████████████████████████▏ | hand/byte-bs----acc-c: 147772 ███████████████████████████████████████▏ | hand/byte-bs----acc-d: 147776 ███████████████████████████████████████▏ | c/byte-getchar: 436 ▏ | c/byte-getchar-u: 432 ▏ | c/byte-4k: 436 ▏ | Time (space counting) std --------------------- avg dev slack hs/space-bs-c8-acc-1: 4.318 1‰ 0.0 ████ | hs/space-bs-c8-count: 3.118 1‰ 0.1 ██▉ | hs/space-bs-c8-foldlx-1: 4.631 1‰ 0.0 ████▎ | hs/space-bs-c8-foldlx-2: 4.632 1‰ 0.0 ████▎ | hs/space-bs-c8-foldrx: 4.678 0‰ 0.0 ████▎ | hs/space-bs-c8-lenfil: 4.634 1‰ 0.1 ████▎ | hs/space-bslc8-acc-1: 32.733 7‰ 0.0 ██████████████████████████████ | hs/space-bslc8-acc-2: 33.112 11‰ 0.0 ██████████████████████████████▎ | hs/space-bslc8-acc-3: 30.294 12‰ 0.0 ███████████████████████████▊ | hs/space-bslc8-chunk-1: 3.888 0‰ 0.1 ███▌ | hs/space-bslc8-chunk-2: 3.887 1‰ 0.0 ███▌ | hs/space-bslc8-chunk-3: 3.881 0‰ 0.1 ███▌ | hs/space-bslc8-chunk-4: 3.910 1‰ 0.0 ███▋ | hs/space-bslc8-count: 2.355 1‰ 0.1 ██▏ | hs/space-bslc8-foldl: 4.674 0‰ 0.0 ████▎ | hs/space-bslc8-foldlx-1: 4.676 0‰ 0.0 ████▎ | hs/space-bslc8-foldlx-2: 4.676 0‰ 0.0 ████▎ | hs/space-bslc8-foldr-1: 5.222 0‰ 0.0 ████▊ | hs/space-bslc8-foldr-2: 5.224 1‰ 0.1 ████▊ | hs/space-bslc8-lenfil-1: 4.712 1‰ 0.0 ████▎ | hs/space-bslc8-lenfil-2: 4.714 1‰ 0.0 ████▍ | hs/space-bsl---foldlx: 4.522 0‰ 0.1 ████▏ | hs/space-xxxxx-acc-1: 27.529 6‰ 0.0 █████████████████████████▎ | hs/space-xxxxx-acc-2: 27.200 3‰ 0.0 ████████████████████████▉ | hs/space-xxxxx-foldl: 27.436 8‰ 0.0 █████████████████████████▏ | hs/space-xxxxx-lenfil: 39.343 3‰ 0.0 ████████████████████████████████████| hand/space-bs-c8-acc-1-a: 4.318 1‰ 0.0 ████ | hand/space-bs-c8-acc-1-b: 4.331 1‰ 0.0 ████ | hand/space-bs-c8-acc-1-c: 4.492 1‰ 0.1 ████▏ | hand/space-bs-c8-acc-1-d: 4.354 1‰ 0.0 ████ | hand/space-bs-c8-acc-1-e: 4.424 0‰ 0.1 ████ | hand/space-bs-c8-acc-1-f: 4.164 1‰ 0.0 ███▊ | hand/space-bs-c8-acc-1-g: 4.309 1‰ 0.1 ████ | hand/space-bs-c8-acc-1-h: 4.202 1‰ 0.1 ███▉ | hand/space-bs-c8-acc-1-i: 3.820 1‰ 0.1 ███▌ | hand/space-bs-c8-acc-1-j: 3.472 1‰ 0.0 ███▏ | hand/space-bs-c8-acc-1-k: 3.474 1‰ 0.0 ███▏ | hand/space-bs-c8-acc-1-l: 3.498 1‰ 0.1 ███▎ | hand/space-bs-c8-acc-1-m: 3.397 1‰ 0.1 ███▏ | hand/space-bs-c8-acc-1-n: 3.373 1‰ 0.0 ███▏ | hand/space-bs-c8-acc-1-o: 3.118 1‰ 0.1 ██▉ | hand/space-bs-c8-acc-1-p: 2.935 2‰ 0.0 ██▋ | hand/space-bs-c8-acc-1-q: 3.477 1‰ 0.1 ███▏ | hand/space-bs-c8-acc-1-r: 1.674 1‰ 0.1 █▌ | hand/space-bs-c8-acc-1-s: 1.627 1‰ 0.2 █▌ | c/space-getchar: 13.268 0‰ 0.0 ████████████▏ | c/space-getchar-u: 2.170 0‰ 0.1 ██ | c/space-4k: 1.815 1‰ 0.1 █▋ | c/space-megabuf: 2.763 2‰ 0.1 ██▌ | c/space-getwchar: 15.105 29‰ 0.0 █████████████▉ | c/space-getwchar-u: 8.979 1‰ 0.0 ████████▎ | c/space-32k: 1.786 9‰ 0.1 █▋ | c/space-32k-8: 1.777 1‰ 0.1 █▋ | Memory: Peak ------- KB hs/space-bs-c8-acc-1: 147748 ███████████████████████████████████████▏ | hs/space-bs-c8-count: 147760 ███████████████████████████████████████▏ | hs/space-bs-c8-foldlx-1: 147760 ███████████████████████████████████████▏ | hs/space-bs-c8-foldlx-2: 147764 ███████████████████████████████████████▏ | hs/space-bs-c8-foldrx: 147748 ███████████████████████████████████████▏ | hs/space-bs-c8-lenfil: 147764 ███████████████████████████████████████▏ | hs/space-bslc8-acc-1: 3196 ▉ | hs/space-bslc8-acc-2: 3200 ▉ | hs/space-bslc8-acc-3: 3204 ▉ | hs/space-bslc8-chunk-1: 65204 █████████████████▎ | hs/space-bslc8-chunk-2: 65200 █████████████████▎ | hs/space-bslc8-chunk-3: 75248 ████████████████████ | hs/space-bslc8-chunk-4: 76204 ████████████████████▎ | hs/space-bslc8-count: 79224 █████████████████████ | hs/space-bslc8-foldl: 85880 ██████████████████████▊ | hs/space-bslc8-foldlx-1: 85880 ██████████████████████▊ | hs/space-bslc8-foldlx-2: 85880 ██████████████████████▊ | hs/space-bslc8-foldr-1: 169704 █████████████████████████████████████████████| hs/space-bslc8-foldr-2: 169700 █████████████████████████████████████████████| hs/space-bslc8-lenfil-1: 109408 █████████████████████████████ | hs/space-bslc8-lenfil-2: 109404 █████████████████████████████ | hs/space-bsl---foldlx: 85844 ██████████████████████▊ | hs/space-xxxxx-acc-1: 1912 ▌ | hs/space-xxxxx-acc-2: 1912 ▌ | hs/space-xxxxx-foldl: 1912 ▌ | hs/space-xxxxx-lenfil: 1904 ▌ | hand/space-bs-c8-acc-1-a: 147760 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-b: 147764 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-c: 147764 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-d: 147764 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-e: 147764 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-f: 147760 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-g: 147760 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-h: 147764 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-i: 147764 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-j: 147764 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-k: 147764 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-l: 147760 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-m: 147760 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-n: 147764 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-o: 147760 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-p: 147760 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-q: 147760 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-r: 147760 ███████████████████████████████████████▏ | hand/space-bs-c8-acc-1-s: 147760 ███████████████████████████████████████▏ | c/space-getchar: 436 ▏ | c/space-getchar-u: 436 ▏ | c/space-4k: 472 ▏ | c/space-megabuf: 146968 ███████████████████████████████████████ | c/space-getwchar: 484 ▏ | c/space-getwchar-u: 484 ▏ | c/space-32k: 500 ▏ | c/space-32k-8: 496 ▏ |

1 0

[7/16] SBM: 6.9.20071208 Core Duo
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

This set of measurements was captured by me on a friend's machine at his place, namely an Intel Core Duo laptop running Ubuntu 7.10. The benchmarks were run on 2007-12-09 using ghc 6.9.20071208 (approx) in runlevel 1 (single-user mode). I used an early version of the benchmark suite which didn't capture the exact ghc version used and an early version of my ghc install scripts that didn't try for a precise version. It ran on what was ghc head at the time. -Peter albatross ghc 6.9.20071208 (or thereabouts) Genuine Intel(R) CPU T2300 @ 1.66GHz (Core Duo) 1667.000 MHz TESTKIND=THOROUGH SUFFIX= Time (byte counting) std -------------------- avg dev slack hs/byte-bs----acc: 0.918 4‰ 0.1 █████▌ | hs/byte-bs----foldlx: 1.014 2‰ 0.2 ██████▏ | hs/byte-bs----foldrx: 0.999 2‰ 0.1 ██████ | hs/byte-bsl---acc: 1.957 1‰ 0.1 ███████████▉ | hs/byte-xxxxx-acc-1: 3.591 2‰ 0.1 █████████████████████▊ | hs/byte-xxxxx-acc-2: 3.598 4‰ 0.1 █████████████████████▊ | hs/byte-xxxxx-foldl: 3.594 1‰ 0.1 █████████████████████▊ | c/byte-getchar: 3.761 2‰ 0.1 ██████████████████████▊ | c/byte-getchar-u: 0.719 2‰ 0.3 ████▍ | c/byte-4k: 0.101 10‰ 3.1 ▋ | Memory: Peak ------- KB hs/byte-bs----acc: 147784 ████████████████████████████████████████▏ | hs/byte-bs----foldlx: 147812 ████████████████████████████████████████▏ | hs/byte-bs----foldrx: 147788 ████████████████████████████████████████▏ | hs/byte-bsl---acc: 3228 ▉ | hs/byte-xxxxx-acc-1: 1904 ▌ | hs/byte-xxxxx-acc-2: 1908 ▌ | hs/byte-xxxxx-foldl: 1908 ▌ | c/byte-getchar: 432 ▏ | c/byte-getchar-u: 440 ▏ | c/byte-4k: 436 ▏ | Time (space counting) std --------------------- avg dev slack hs/space-bs-c8-acc-1: 1.313 0‰ 0.2 ███████▉ | hs/space-bs-c8-foldlx-1: 1.326 1‰ 0.1 ████████ | hs/space-bs-c8-foldlx-2: 1.328 1‰ 0.1 ████████ | hs/space-bs-c8-foldrx: 1.128 1‰ 0.2 ██████▉ | hs/space-bs-c8-lenfil: 1.327 2‰ 0.2 ████████ | hs/space-bslc8-acc-1: 3.002 1‰ 0.1 ██████████████████▏ | hs/space-bslc8-acc-2: 3.010 2‰ 0.0 ██████████████████▎ | hs/space-bslc8-acc-3: 2.586 1‰ 0.1 ███████████████▋ | hs/space-bslc8-chunk-1: 1.192 0‰ 0.0 ███████▎ | hs/space-bslc8-chunk-2: 1.189 0‰ 0.1 ███████▏ | hs/space-bslc8-chunk-3: 1.272 1‰ 0.2 ███████▊ | hs/space-bslc8-foldl: 1.517 0‰ 0.1 █████████▏ | hs/space-bslc8-foldlx-1: 1.516 0‰ 0.0 █████████▏ | hs/space-bslc8-foldlx-2: 1.511 0‰ 0.0 █████████▏ | hs/space-bslc8-foldr-1: 1.205 1‰ 0.2 ███████▎ | hs/space-bslc8-foldr-2: 1.202 3‰ 0.1 ███████▎ | hs/space-bslc8-lenfil-1: 1.400 0‰ 0.0 ████████▌ | hs/space-bslc8-lenfil-2: 1.396 0‰ 0.0 ████████▌ | hs/space-bsl---foldlx: 1.476 1‰ 0.1 ████████▉ | hs/space-xxxxx-acc-1: 4.245 1‰ 0.0 █████████████████████████▋ | hs/space-xxxxx-acc-2: 4.241 2‰ 0.1 █████████████████████████▋ | hs/space-xxxxx-foldl: 4.239 2‰ 0.1 █████████████████████████▋ | hs/space-xxxxx-lenfil: 6.287 2‰ 0.0 ██████████████████████████████████████| c/space-getchar: 4.031 18‰ 0.0 ████████████████████████▍ | c/space-getchar-u: 0.609 2‰ 0.7 ███▋ | c/space-4k: 0.465 2‰ 0.6 ██▊ | c/space-megabuf: 0.613 1‰ 0.3 ███▊ | c/space-getwchar: 4.584 1‰ 0.1 ███████████████████████████▊ | c/space-getwchar-u: 2.437 0‰ 0.1 ██████████████▊ | c/space-32k: 0.445 1‰ 0.5 ██▊ | c/space-32k-8: 0.448 0‰ 0.0 ██▊ | Memory: Peak ------- KB hs/space-bs-c8-acc-1: 147772 ████████████████████████████████████████ | hs/space-bs-c8-foldlx-1: 147800 ████████████████████████████████████████▏ | hs/space-bs-c8-foldlx-2: 147804 ████████████████████████████████████████▏ | hs/space-bs-c8-foldrx: 147776 ████████████████████████████████████████ | hs/space-bs-c8-lenfil: 147796 ████████████████████████████████████████▏ | hs/space-bslc8-acc-1: 3192 ▉ | hs/space-bslc8-acc-2: 3192 ▉ | hs/space-bslc8-acc-3: 3188 ▉ | hs/space-bslc8-chunk-1: 65208 █████████████████▋ | hs/space-bslc8-chunk-2: 65212 █████████████████▋ | hs/space-bslc8-chunk-3: 75232 ████████████████████▍ | hs/space-bslc8-foldl: 85856 ███████████████████████▎ | hs/space-bslc8-foldlx-1: 85852 ███████████████████████▎ | hs/space-bslc8-foldlx-2: 85856 ███████████████████████▎ | hs/space-bslc8-foldr-1: 169684 ██████████████████████████████████████████████| hs/space-bslc8-foldr-2: 169684 ██████████████████████████████████████████████| hs/space-bslc8-lenfil-1: 109404 █████████████████████████████▋ | hs/space-bslc8-lenfil-2: 109404 █████████████████████████████▋ | hs/space-bsl---foldlx: 85880 ███████████████████████▎ | hs/space-xxxxx-acc-1: 1916 ▌ | hs/space-xxxxx-acc-2: 1916 ▌ | hs/space-xxxxx-foldl: 1916 ▌ | hs/space-xxxxx-lenfil: 1892 ▌ | c/space-getchar: 436 ▏ | c/space-getchar-u: 436 ▏ | c/space-4k: 468 ▏ | c/space-megabuf: 146968 ███████████████████████████████████████▉ | c/space-getwchar: 484 ▏ | c/space-getwchar-u: 480 ▏ | c/space-32k: 496 ▏ | c/space-32k-8: 500 ▏ |

1 0

[6/16] SBM: 6.9.20071124 Athlon Duron
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

This set of measurements was captured by Daniel Fischer on one of his older machines, running SuSE 8.2, which has a Linux 2.4 kernel! The benchmarks were run on 2007-12-16 using ghc 6.9.20071124. Unfortunately, the ghc version is not quite the same as the one I've used for most measurements (6.9.20071119) so things may be a little different just for that reason. The results seemed a bit off at first, but now that I have graphs of all the runs on all the machines they don't seem strange at all. First of all, the memory use is about the same as on the other machines. Secondly, the timing differences for the C getchar/getwchar might be partly due to different versions of the C library. The remaining differences (a "steeper" profile than on the core duo and the Athlon64) may be due to different microarchitectures. -Peter Fischer's machine ghc 6.9.20071124 AMD Duron(tm) processor 1200.089 MHz TESTKIND=THOROUGH SUFFIX= Time (byte counting) std -------------------- avg dev slack hs/byte-bs----acc: 1.892 21‰ 0.4 ██▊ | hs/byte-bs----foldlx: 2.258 3‰ 0.1 ███▎ | hs/byte-bs----foldrx: 2.933 0‰ 0.1 ████▎ | hs/byte-bsl---acc: 14.319 45‰ 0.1 ████████████████████▌ | hs/byte-xxxxx-acc-1: 20.915 17‰ 4.0 █████████████████████████████▉ | hs/byte-xxxxx-acc-2: 20.691 8‰ 0.1 █████████████████████████████▌ | hs/byte-xxxxx-foldl: 20.610 5‰ 1.4 █████████████████████████████▍ | c/byte-getchar: 9.042 0‰ 0.1 ████████████▉ | c/byte-getchar-u: 1.314 3‰ 0.2 █▉ | c/byte-4k: 0.419 5‰ 0.5 ▋ | Memory: Peak ------- KB hs/byte-bs----acc: 147492 ████████████████████████████████████████ | hs/byte-bs----foldlx: 147492 ████████████████████████████████████████ | hs/byte-bs----foldrx: 147488 ████████████████████████████████████████ | hs/byte-bsl---acc: 2896 ▊ | hs/byte-xxxxx-acc-1: 1612 ▌ | hs/byte-xxxxx-acc-2: 1612 ▌ | hs/byte-xxxxx-foldl: 1612 ▌ | c/byte-getchar: 384 ▏ | c/byte-getchar-u: 384 ▏ | c/byte-4k: 380 ▏ | Time (space counting) std --------------------- avg dev slack hs/space-bs-c8-acc-1: 2.467 1‰ 0.3 ███▌ | hs/space-bs-c8-foldlx-1: 2.585 2‰ 0.1 ███▊ | hs/space-bs-c8-foldlx-2: 2.576 2‰ 0.3 ███▋ | hs/space-bs-c8-foldrx: 2.982 8‰ 2.3 ████▎ | hs/space-bs-c8-lenfil: 2.599 1‰ 0.2 ███▊ | hs/space-bslc8-acc-1: 15.228 8‰ 0.1 █████████████████████▊ | hs/space-bslc8-acc-2: 15.855 38‰ 0.0 ██████████████████████▋ | hs/space-bslc8-acc-3: 14.980 14‰ 0.0 █████████████████████▍ | hs/space-bslc8-chunk-1: 2.443 2‰ 0.2 ███▌ | hs/space-bslc8-chunk-2: 2.449 1‰ 0.3 ███▌ | hs/space-bslc8-chunk-3: 2.534 3‰ 0.3 ███▋ | hs/space-bslc8-foldl: 2.938 1‰ 0.2 ████▎ | hs/space-bslc8-foldlx-1: 2.928 1‰ 0.0 ████▏ | hs/space-bslc8-foldlx-2: 2.937 2‰ 0.2 ████▎ | hs/space-bslc8-foldr-1: 4.043 6‰ 0.1 █████▊ | hs/space-bslc8-foldr-2: 4.007 4‰ 0.1 █████▊ | hs/space-bslc8-lenfil-1: 3.240 1‰ 0.2 ████▋ | hs/space-bslc8-lenfil-2: 3.236 1‰ 0.2 ████▋ | hs/space-bsl---foldlx: 2.821 1‰ 0.1 ████ | hs/space-xxxxx-acc-1: 21.002 4‰ 0.1 ██████████████████████████████ | hs/space-xxxxx-acc-2: 21.270 22‰ 4.7 ██████████████████████████████▍ | hs/space-xxxxx-foldl: 20.934 1‰ 0.1 █████████████████████████████▉ | hs/space-xxxxx-lenfil: 25.915 3‰ 0.0 █████████████████████████████████████| c/space-getchar: 9.354 0‰ 0.0 █████████████▍ | c/space-getchar-u: 1.676 2‰ 0.2 ██▍ | c/space-4k: 1.293 2‰ 0.5 █▉ | c/space-megabuf: 1.830 3‰ 0.5 ██▋ | c/space-getwchar: 14.721 1‰ 0.1 █████████████████████ | c/space-getwchar-u: 4.814 0‰ 0.0 ██████▉ | c/space-32k: 1.276 2‰ 0.6 █▉ | c/space-32k-8: 1.275 2‰ 0.2 █▉ | Memory: Peak ------- KB hs/space-bs-c8-acc-1: 147488 ████████████████████████████████████████ | hs/space-bs-c8-foldlx-1: 147492 ████████████████████████████████████████ | hs/space-bs-c8-foldlx-2: 147492 ████████████████████████████████████████ | hs/space-bs-c8-foldrx: 147488 ████████████████████████████████████████ | hs/space-bs-c8-lenfil: 147492 ████████████████████████████████████████ | hs/space-bslc8-acc-1: 2896 ▊ | hs/space-bslc8-acc-2: 2896 ▊ | hs/space-bslc8-acc-3: 2896 ▊ | hs/space-bslc8-chunk-1: 65892 █████████████████▉ | hs/space-bslc8-chunk-2: 65892 █████████████████▉ | hs/space-bslc8-chunk-3: 76472 ████████████████████▊ | hs/space-bslc8-foldl: 86772 ███████████████████████▋ | hs/space-bslc8-foldlx-1: 86772 ███████████████████████▋ | hs/space-bslc8-foldlx-2: 86772 ███████████████████████▋ | hs/space-bslc8-foldr-1: 169360 ██████████████████████████████████████████████| hs/space-bslc8-foldr-2: 169360 ██████████████████████████████████████████████| hs/space-bslc8-lenfil-1: 110704 ██████████████████████████████▏ | hs/space-bslc8-lenfil-2: 110704 ██████████████████████████████▏ | hs/space-bsl---foldlx: 86776 ███████████████████████▋ | hs/space-xxxxx-acc-1: 1612 ▌ | hs/space-xxxxx-acc-2: 1612 ▌ | hs/space-xxxxx-foldl: 1612 ▌ | hs/space-xxxxx-lenfil: 1588 ▍ | c/space-getchar: 384 ▏ | c/space-getchar-u: 384 ▏ | c/space-4k: 412 ▏ | c/space-megabuf: 146904 ███████████████████████████████████████▉ | c/space-getwchar: 440 ▏ | c/space-getwchar-u: 440 ▏ | c/space-32k: 436 ▏ | c/space-32k-8: 436 ▏ |

1 0

[5/16] SBM: Support scripts and scriptlets
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

Some of the scripts warrant a closer look. 'make zipdata' creates a nice tarball with all the data necessary to recreate a report AND to merge that report together with other reports, possible with rescaled bar charts. Very handy. All the files in the tarball are inside the 'ghc-measurements/' directory so the risk of things going wrong when unpacking the tarball is less. The names of the benchmarks are put in ghc-measurements/progs, mainly to ensure they end up in the right order when regenerating and merging reports. tools/genreport.pl [list of benchmarks to put in the report] It doesn't parse the command-line in any way because life is too short for command-line parsing. Instead, it is controlled via (too many) environment variables. ASCII - set to avoid using UTF-8 for bar charts and "per mille" character. NOSRC - the tool normally creates */*.srctimemem files containing the source code for each benchmark with bar charts for time/mem appended to the end. Setting this variable switches that off (necessary when regenerating and merging reports). EXCLUDE - disregard some of the benchmarks on the command line. Why is this necessary? Because it makes regenerating and merging reports easier. And because I was too lazy to filter the command line in tools/regenreport.sh and tools/merge.pl. FINDMAX - used by tools/merge.pl when rescaling. Outputs max time and max mem to stdout instead of the normal report. MAX_FILEWIDTH - used by tools/merge.pl to make merged reports look nice MAX_TIME, MAX_PEAKMEM - used by tools/merge.pl when rescaling Note that strictly speaking, there is a bug in the script(s) because it conflates the width of time/mem measurement represented as numbers (which you always want to take into account when merging) and MAX_TIME/MAX_PEAKMEM (which you only care about when rescaling). [FIXED now - 2007-12-21] tools/regenreport.sh unpacks a measurement tarball into a tmp directory and runs tools/genreport.pl to generate the report. Takes care not to disturb the normal files. tools/merge.pl [tarballs] Uses tools/regenreport.sh on each tarball in turn to generate a report which it reads in and stores on a benchmark-by-benchmark basis. At the end, synthetically combine all the pieces it cut out of the original report(s) into a brand-spanking new, merged report. Even the headers and the platforminfo at the top of each report is cut out and stored in data structures until they get spit out again at the end. The reading magic is in the state machine in gather(). It is not as bad as it looks. Some of the complications arise from marking repeated benchmark names as ' -- ', which improves the readability of the merged reports immensely. Another part of the complications arise due to the fact that not all tarballs contain the exact same benchmarks! Those that don't get a nice 'n/a' instead of numbers and a bar. And finally, the benchmarks should be in the right order. That is trickier than it sounds... When rescaling, tools/regenreport.sh is first run once for each tarball with the FINDMAX environment variable set. This results in tools/regenreport.sh outputting the maximum filename width, time, and peakmem for each tarball. ASCII - use ASCII instead of UTF-8 RESCALE - sometimes you want to rescale and sometimes you don't MAX_FILEWIDTH - if you want to force a specific width MAX_TIME, MAX_PEAKMEM - if you want to force a specific max -Peter

1 0

[4/16] SBM: How to use the Makefile (how to run benchmarks etc.)
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

Introduction ------------ Most of the smarts of the benchmark harness is in the Makefile. If you want to rerun the benchmarks (or a single benchmark) or look at the intermediate code for a benchmark or the I/O trace or the memory consumption or the time spent or ... then you use the makefile. There are some support scripts in shell and Perl (and two C programs) that the Makefile uses to do its job. And there are some that you, the user, will want to interact directly with. The benchmarks are only expected to work on Linux. They have been tested on SuSE 8.2 (from 2003, with a 2.4 kernel), Ubuntu 7.04, and Ubuntu 7.10. Quick howto ----------- make phase1 -- compiles, generates test files, measures memory use. Safe to run on a busy machine if there's no active memory pressure. make phase2 -- timing runs. NOT safe to run on a busy machine. Should be run in runlevel 1 (= no X, no daemons, single-user mode) for best measurements. Outputs report at end. This is where you check the quality of the measurements. If you don't like them, run 'make redophase2' (or delete the .time and .stat files with low quality and run 'make phase2' again.) make zipdata -- make a tarball with all the measurements, suitable for emailing or putting on a website. The Makefile will beep after phase 1 and 2. The above will run a "NORMAL" run, which is fine during development if you want to see if you nailed a performance bug. It runs reasonably fast (about 43 seconds on my Athlon64). If you want better measurements, you should use: make TESTKIND=THOROUGH phase1 phase2 zipdata This will use a 150MB data file instead of a 15MB one and it will run the timing measurements 6 times (before throwing the first away) instead of 4 times (before throwing the first away). If you don't want to use single-user mode, you can improve the measurements by piping the output to a file (or run the test from the console) instead of involving a terminal and an X server (the screen update may kick in in the middle of a timing run and disturb things if for no other reason than their polluting the CPU caches). Filesystem layout ----------------- The benchmarks are in: hs/*.hs c/*.c hand/*.s hand/*.hs and hand/*.c are not compiled. The two *.hs files are the originals from which the tweaked assembly code has been derived. The two *.c files are sketches of how the MMX tweaks work (because MMX code by itself can be a bit off-putting). These are the support scripts: tools/genfiles.pl -- generate the test input files. tools/cutmem.pl tools/cutpid.pl -- both are used to disentangle the outputs of strace and pause-at-end (see below). I combine strace, memory info, and +RTS -sstderr into a single run to save time. This means that things end up in fewer files than I'd like. tools/cut.pl -- cut out main loop from disassembly ('make discut') tools/stat.pl -- looks at all timings for a single benchmark and calculates average and standard deviation and "time slack", that is the discrepancy between user+sys and real. It optionally throws away the first run. tools/eatmem.c -- allocates a chunk of memory and makes damn sure it really is in RAM! tools/pause-at-end.c -- part of a hack that copies /proc/self/maps and /proc/self/status to stderr just before a benchmark exits. tools/iosummary.pl -- takes an strace and sums up the I/O tools/genreport.pl -- generate a nice report with bar charts. Takes way too many options in the form of environment variables. tools/regenreport.sh -- regenerates the report from ANY measurement tarball. tools/merge.pl -- merge data from many measurement tarballs, with or without rescaling. Generated files: hs/*.core hs/*.stg hs/*.cmm hs/*.s -- intermediate code hs/*.hi -- "Haskell Interface" */*.o -- object code */* (the files in $(HSPROGS) $(CPROGS) $(HANDPROGS)) -- programs */*.dis */*.discut -- disassembled programs (and inner loops) */*.doc -- source + intermediate code + inner loops + timings */*.mem -- output from '+RTS -sstderr' + /proc/self/status + /proc/self/maps + output from /usr/bin/time (where the number of minor page faults is most interesting datum) */*.strace -- complete strace, taken together with */*.mem */*.iotrace -- only I/O operations from the strace (read/write/ select) */*.iosum -- summary of I/O operations */*.time -- time measurements */*.stat -- average + std.dev. + "time slack" */*.srctimespace -- source code + time/mem barchart (in ASCII) sysinfo -- description of the platform (uname, ghc, gcc, etc) platforminfo -- short description of the platform report.txt [8-16K] docs [1MB] -- sysinfo + all */*.doc concatenated Makefile targets ---------------- This is taken from 'make help': phase1 -- preparation + measurements that can run in background phase2 -- measurements that should run on unloaded machine redophase2 -- rerun phase2 doc, [ASCII=1] report, lastreport - reports zipdata -- zip up measurements (to ghc-measurements.tar.gz) prog,core,stg,cmm,asm,dis,discut -- compile, compile to core/stg/cmm/asm, disassemble, cut out main loop time,stat,mem,strace,iotrace,iosum,cache -- measure run-time, GHC heap + OS mem, syscalls, I/O patterns, cache cleartime, clean, distclean -- delete measurements etc TESTKIND=(SMOKETEST,NORMAL,THOROUGH), defaults to NORMAL STRACE=OLD, defaults to NEW Necessary tools --------------- Perl, sed, /usr/bin/time, bash (doesn't have to be the default shell as long as it its in PATH), strace, GNU Make, objdump (from the binutils package), gcc. Other things that could come in handy: A console and/or terminal that understands UTF-8. A less that understands UTF-8. An editor that understands UTF-8. A %!&# printing program that understands both UTF-8 and fonts. A2ps doesn't do UTF-8. Uniprint used 1) a proportional font which 2) didn't even have all the fractional-width blocks. U2ps used a with none of the block characters. I ended up resorting to gedit's print function :( That's enough for this email. -Peter

1 0

[3/16] SBM: The Makefile
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

This is the entire Makefile. It perhaps ought to be sent as an attachment but my hacky mailer script wouldn't like it. A few of the lines are wider than 80 columns, unfortunately. -Peter # GHC benchmarks of parsing (bytestring, basic code generation, I/O). # # Copyright 2007 Peter Lund <firefly(a)vax64.dk>, licensed under GPLv2. # # # You will need the following tools: # perl, strace, /usr/bin/time, bash, a gcc that uses shared libraries and libc # (not dietlibc, klibc, uclibc), objdump (usually found in the binutils # package). # # A benchmark run on a new platform can be split into two phases: # # phase1: Compiles, dices, and slices the code in various ways. If this # completes you can be pretty sure that everything works all right. # Performs non-timing sensitive measurements. # # phase2: Performs timing sensitive measurements. It is a good idea to run # this phase on an idle machine, preferably without using X. # For example, you can log out of X and run "telinit 1" to get to # single-user mode. If you do so, please remember to set the correct # path to your ghc compiler. # Make a few bits of the makefile less noisy. Q:=@ ######################################################### # Ask ghc to optimize and warn GHCFLAGS= -O2 -W # Some newer versions of gcc prefer -Wextra -Wall GCCWARNFLAGS=-W -Wall # Default compilers CC=gcc GHC=ghc GHCPKG=ghc-pkg ######################################################### HSPROGS=hs/byte-bs----acc \ hs/byte-bs----foldlx \ hs/byte-bs----foldrx \ hs/byte-bsl---acc \ hs/byte-xxxxx-acc-1 \ hs/byte-xxxxx-acc-2 \ hs/byte-xxxxx-foldl \ \ hs/space-bs-c8-acc-1 \ hs/space-bs-c8-count \ hs/space-bs-c8-foldlx-1 \ hs/space-bs-c8-foldlx-2 \ hs/space-bs-c8-foldrx \ hs/space-bs-c8-lenfil \ hs/space-bslc8-acc-1 \ hs/space-bslc8-acc-2 \ hs/space-bslc8-acc-3 \ hs/space-bslc8-chunk-1 \ hs/space-bslc8-chunk-2 \ hs/space-bslc8-chunk-3 \ hs/space-bslc8-chunk-4 \ hs/space-bslc8-count \ hs/space-bslc8-foldl \ hs/space-bslc8-foldlx-1 \ hs/space-bslc8-foldlx-2 \ hs/space-bslc8-foldr-1 \ hs/space-bslc8-foldr-2 \ hs/space-bslc8-lenfil-1 \ hs/space-bslc8-lenfil-2 \ hs/space-bsl---foldlx \ hs/space-xxxxx-acc-1 \ hs/space-xxxxx-acc-2 \ hs/space-xxxxx-foldl \ hs/space-xxxxx-lenfil # RMPROGS keeps track of programs that are not always included in the tests. # We do want 'make clean' to delete them even when they are not currently # part of the build (they may be left over from a previous build). # stack overflow with long4. #HSPROGS:=$(HSPROGS) hs/byte-xxxxx-foldr-1 RMPROGS:=$(RMPROGS) hs/byte-xxxxx-foldr-1 # stack overflow with long4. #HSPROGS:=$(HSPROGS) hs/byte-xxxxx-foldr-2 RMPROGS:=$(RMPROGS) hs/byte-xxxxx-foldr-2 # stack overflow with long4. #HSPROGS:=$(HSPROGS) hs/space-xxxxx-foldr-1 RMPROGS:=$(RMPROGS) hs/space-xxxxx-foldr-1 # stack overflow with long4. #HSPROGS:=$(HSPROGS) hs/space-xxxxx-foldr-2 RMPROGS:=$(RMPROGS) hs/space-xxxxx-foldr-2 HANDPROGS= hand/byte-bs----acc-a \ hand/byte-bs----acc-b \ hand/byte-bs----acc-c \ hand/byte-bs----acc-d \ \ hand/space-bs-c8-acc-1-a \ hand/space-bs-c8-acc-1-b \ hand/space-bs-c8-acc-1-c \ hand/space-bs-c8-acc-1-d \ hand/space-bs-c8-acc-1-e \ hand/space-bs-c8-acc-1-f \ hand/space-bs-c8-acc-1-g \ hand/space-bs-c8-acc-1-h \ hand/space-bs-c8-acc-1-i \ hand/space-bs-c8-acc-1-j \ hand/space-bs-c8-acc-1-k \ hand/space-bs-c8-acc-1-l \ hand/space-bs-c8-acc-1-m \ hand/space-bs-c8-acc-1-n \ hand/space-bs-c8-acc-1-o \ hand/space-bs-c8-acc-1-p \ hand/space-bs-c8-acc-1-q \ hand/space-bs-c8-acc-1-r \ hand/space-bs-c8-acc-1-s RMPROGS:=$(RMPROGS) $(HANDPROGS) ifeq ($(shell $(GHCPKG) list | grep bytestring),) # ghc 6.6.1 with an old version of bytestring in 'base' but without its own # module name HSPROGS:=$(shell printf "%s\n" $(HSPROGS) | grep -v '.*-chunk-.*') endif HANDTEXT:=including hand-tweaked assembly ifeq ($(shell $(GHC) --version | grep 6.9.20071119),) HANDPROGS:= HANDTEXT:=no hand-tweaked assembly endif ifneq ($(SUFFIX),) HANDPROGS:= HANDTEXT:=no hand-tweaked assembly endif CPROGS= c/byte-getchar c/byte-getchar-u c/byte-4k \ \ c/space-getchar c/space-getchar-u c/space-4k \ c/space-megabuf c/space-getwchar c/space-getwchar-u \ c/space-32k c/space-32k-8 ######################################################### # The benchmarks can be run in three modes. The default can be overridden from # command line: # # make TESTKIND=SMOKETEST phase1 # # just tests the test suite, as fast as possible #TESTKINDDEFAULT=SMOKETEST # small test TESTKINDDEFAULT=NORMAL # very thorough test #TESTKINDDEFAULT=THOROUGH TESTKIND=$(TESTKINDDEFAULT) ifeq ($(TESTKIND),THOROUGH) TESTFILE= testfiles/long4 TESTFILECACHE= testfiles/long3 else ifeq ($(TESTKIND),NORMAL) TESTFILE= testfiles/long3 TESTFILECACHE= testfiles/long2 else ifeq ($(TESTKIND),SMOKETEST) TESTFILE= testfiles/long2 TESTFILECACHE= testfiles/long1 endif endif endif # Older versions of strace don't support the -E parameter which we use to # set LD_PRELOAD before running the straced command (so we can get by with # a single run of each benchmark in phase 1 instead of 2). # # Override with STRACE=OLD on the command line if you need to work with an # old strace. STRACE=NEW ######################################################### .PHONY: XXXXFIRST testfiles core stg cmm asm dis discut prog \ time stat strace iotrace iosum mem cache doc phase1 phase2 redophase2 XXXXFIRST: help testfiles: testfiles/long1 testfiles/long2 testfiles/long3 testfiles/long4 core: $(addsuffix .core ,$(HSPROGS) ) stg: $(addsuffix .stg ,$(HSPROGS) ) cmm: $(addsuffix .cmm ,$(HSPROGS) ) asm: $(addsuffix .s ,$(HSPROGS) $(CPROGS)) dis: $(addsuffix .dis ,$(HSPROGS) $(HANDPROGS) $(CPROGS)) discut: $(addsuffix .discut ,$(HSPROGS) $(HANDPROGS) $(CPROGS)) prog: $(HSPROGS) $(HANDPROGS) $(CPROGS) time: prog \ testfiles \ $(addsuffix .time ,$(HSPROGS) $(HANDPROGS) $(CPROGS)) stat: $(addsuffix .stat ,$(HSPROGS) $(HANDPROGS) $(CPROGS)) strace: $(addsuffix .strace ,$(HSPROGS) $(HANDPROGS) $(CPROGS)) iotrace:$(addsuffix .iotrace,$(HSPROGS) $(HANDPROGS) $(CPROGS)) iosum: $(addsuffix .iosum ,$(HSPROGS) $(HANDPROGS) $(CPROGS)) mem: $(addsuffix .mem ,$(HSPROGS) $(HANDPROGS) $(CPROGS)) cache: $(addsuffix .cache ,$(HSPROGS) $(HANDPROGS) $(CPROGS)) doc: $(addsuffix .doc ,$(HSPROGS) $(HANDPROGS) $(CPROGS)) ### phase1: testfiles tools/eatmem prog iosum mem printf "Done!\07" # beep phase2: time report printf "Done!\07" # beep redophase2: cleartime rm -f */*.srctimespace report.txt $(MAKE) phase2 ######################################################### testfiles/long1: mkdir -p testfiles tools/genfiles.pl 10000 > "$@" testfiles/long2: mkdir -p testfiles tools/genfiles.pl 100000 > "$@" testfiles/long3: mkdir -p testfiles tools/genfiles.pl 1000000 > "$@" testfiles/long4: mkdir -p testfiles tools/genfiles.pl 10000000 > "$@" ######################################################### tools/eatmem: tools/eatmem.c $(CC) $(GCCWARNFLAGS) -O2 "$<" -o "$@" tools/pause-at-end.so: tools/pause-at-end.c $(CC) $(GCCWARNFLAGS) -shared -ldl "$<" -o "$@" ######################################################### docs: core stg cmm asm discut time iotrace doc sysinfo rm -f docs cat */*.doc sysinfo > docs hs/%.doc: hs/%.core hs/%.stg hs/%.cmm hs/%.s hs/%.discut hs/%.time (export F="$(basename $@)" ; \ printf "\n" ; \ printf "*********************************************\n" ; \ printf "****\n" ; \ printf "**** %s:\n" "$$F" ; \ printf "****\n" ; \ printf "*********************************************\n\n" ; \ printf "Haskell code:\n\n" ; \ cat "$$F.hs" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.core" ; \ cat "$$F.core" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.stg" ; \ cat "$$F.stg" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.cmm" ; \ cat "$$F.cmm" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.s" ; \ cat "$$F.s" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.discut" ; \ cat "$$F.discut" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.time" ; \ cat "$$F.time" ; \ printf -- "-------------------------------\n" ; \ printf "\014" ; \ ) >> "$@" hand/%.doc: hand/%.discut hand/%.time (export F="$(basename $@)" ; \ printf "\n" ; \ printf "*********************************************\n" ; \ printf "****\n" ; \ printf "**** %s:\n" "$$F" ; \ printf "****\n" ; \ printf "*********************************************\n\n" ; \ printf "Haskell code:\n\n" ; \ export X=`echo $$F | sed -e 's/[a-z]$$//'` ; \ cat "$$X.hs" ; \ printf -- "-------------------------------\n" ; \ printf "%s (hand tweaked):\n" "$$F.s" ; \ cat "$$F.s" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.discut" ; \ cat "$$F.discut" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.time" ; \ cat "$$F.time" ; \ printf -- "-------------------------------\n" ; \ printf "\014" ; \ ) >> "$@" c/%.doc: c/%.s c/%.discut c/%.time (export F="$(basename $@)" ; \ printf "\n" ; \ printf "*********************************************\n" ; \ printf "****\n" ; \ printf "**** %s:\n" "$$F" ; \ printf "****\n" ; \ printf "*********************************************\n\n" ; \ printf "C code:\n\n" ; \ cat "$$F.c" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.s" ; \ cat "$$F.s" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.discut" ; \ cat "$$F.discut" ; \ printf -- "-------------------------------\n" ; \ printf "%s:\n" "$$F.time" ; \ cat "$$F.time" ; \ printf -- "-------------------------------\n" ; \ printf "\014" ; \ ) >> "$@" ######################################################### .PHONY: report.txt lastreport report: report.txt $(Q)cat report.txt report.txt \ $(addsuffix .srctimespace,$(HSPROGS) $(HANDPROGS) $(CPROGS)): \ iosum mem time stat platforminfo $(Q)tools/genreport.pl $(HSPROGS) $(HANDPROGS) $(CPROGS) > report.txt lastreport: $(Q)cat report.txt ######################################################### # Probably all or most of the targets (left-hand sides) in these rules should # be mentioned in a .SECONDARY rule so make won't delete them behind our backs, # in its infinite wisdom. This is sometimes necessary when using pattern # rules (i.e. rules with '%' wildcards in them). # # For some reason, it doesn't seem to be all that necessary, although I had to # insert a couple of those .SECONDARY things earlier to make make behave. For # example, I had to insert this rule at some point but now things keep working # even when it's commented out: # # .SECONDARY: $(HSPROGS) $(HANDPROGS) $(CPROGS) # hs/%: hs/%.hs $(GHC) $(GHCFLAGS) --make -fforce-recomp "$<" -o "$@" hand/%: hand/%.s $(GHC) -no-hs-main "$<" -o "$@" -package bytestring c/%: c/%.c $(CC) $(GCCWARNFLAGS) -O2 "$<" -o "$@" ### %.dis: % @# Limit the disassembly for speed reasons (10x+ difference) and @# file size reasons (20x-30x difference). @# The stuff we are interested in comes early in the .text segment so @# there's no reason to disassemble the entire runtime system, which @# comes afterwards in case of hand/ and hs/ binaries. objdump -M intel -D --stop-address=0x08060000 "$<" > "$@" ### %.discut: %.dis tools/cut.pl < "$<" > "$@" ### %.core: %.hs $(GHC) $(GHCFLAGS) -c -ddump-simpl "$<" > "$@" ### %.stg: %.hs $(GHC) $(GHCFLAGS) -c -ddump-stg "$<" > "$@" ### %.cmm: %.hs $(GHC) $(GHCFLAGS) -c -ddump-cmm "$<" > "$@" ### %.s: %.hs $(GHC) $(GHCFLAGS) -c -fforce-recomp -keep-s-files "$<" .SECONDARY: $(addsuffix .s,$(CPROGS)) %.s: %.c $(CC) $(GCCWARNFLAGS) -O2 -S $< -o $@ ######################################################### # The first run is sacrificial, except when smoketesting where there only is # one run. ifeq ($(TESTKIND),THOROUGH) TIME= bash -c "time $<" < $(TESTFILE); \ bash -c "time $<" < $(TESTFILE); \ bash -c "time $<" < $(TESTFILE); \ bash -c "time $<" < $(TESTFILE); \ bash -c "time $<" < $(TESTFILE); \ bash -c "time $<" < $(TESTFILE) NOSKIP:= else ifeq ($(TESTKIND),NORMAL) TIME= bash -c "time $<" < $(TESTFILE); \ bash -c "time $<" < $(TESTFILE); \ bash -c "time $<" < $(TESTFILE); \ bash -c "time $<" < $(TESTFILE) NOSKIP:= else ifeq ($(TESTKIND),SMOKETEST) TIME= bash -c "time $<" < $(TESTFILE) NOSKIP:= NOSKIP=1 endif endif endif # In order to reduce the risk of swapping during the time test, we try to make # sure there's twice the test file size free (and a bit). # NEEDFREE is measured in kilobytes. NEEDFREE=$(shell expr 22 '*' `ls -s $(TESTFILE) | cut -f1 -d' '` / 10) %.time: % tools/eatmem $(TESTFILE) printf "%s\n" "$<" > "$@" tools/eatmem $(NEEDFREE) dd if=$(TESTFILE) of=/dev/null dd if="$<" of=/dev/null ($(TIME)) >>"$@" 2>&1 printf "%s\n\n" "-----" >> "$@" %.stat: %.time $(NOSKIP) tools/stat.pl < "$<" > "$@" %.mem %.strace: % tools/pause-at-end.so $(TESTFILE) ifeq ($(STRACE),OLD) strace -o tmp.strace -f \ /usr/bin/time "$<" +RTS -sstderr < $(TESTFILE) > $(basename $@).mem 2>&1 LD_PRELOAD=tools/pause-at-end.so \ "$<" +RTS -sstderr < $(TESTFILE) >> $(basename $@).mem 2>&1 else strace -o tmp.strace -ELD_PRELOAD=tools/pause-at-end.so -f \ /usr/bin/time "$<" +RTS -sstderr < $(TESTFILE) > $(basename $@).mem 2>&1 endif tools/cutmem.pl < $(basename $@).mem > tmp mv tmp $(basename $@).mem tools/cutpid.pl < tmp.strace > $(basename $@).strace rm -f tmp.strace #%.strace: %.mem # @echo > /dev/null %.iotrace: %.strace grep '^$read\|write\|select$' "$<" > "$@" %.iosum: %.iotrace tools/iosummary.pl < "$<" > "$@" %.cache: % $(TESTFILECACHE) valgrind --tool=cachegrind "$<" < $(TESTFILECACHE) 2> "$@" ######################################################### .PHONY: zipdata help cleartime clean distclean sysinfo: hostname > sysinfo cat /etc/*release >> sysinfo @echo >> sysinfo uname -a >> sysinfo @echo >> sysinfo cat /proc/cpuinfo >> sysinfo $(GHC) --version >> sysinfo echo >> sysinfo $(CC) --version >> sysinfo # This variable makes testing with weird /proc/cpuinfo files easier CPUINFO=/proc/cpuinfo platforminfo: hostname > platforminfo (printf 'ghc '; ($(GHC) --version | sed -ne 's/^The.*version //p')) >> platforminfo cat $(CPUINFO) | sed -ne '/model name.*:/ { s/model name.*: //p; q}' >> platforminfo printf "%s MHz\n" `cat $(CPUINFO) | sed -ne '/cpu MHz.*:/ { s/cpu MHz.*: //p; q}'` >> platforminfo printf "TESTKIND=$(TESTKIND)\n" >> platforminfo printf "SUFFIX=$(SUFFIX)\n" >> platforminfo zipdata: time stat mem strace iotrace iosum sysinfo report.txt rm -f ghc-measurements.tar.gz rm -rf ghc-measurements mkdir -p ghc-measurements cp --parents \ $(addprefix */*, .time .stat .mem .iosum) sysinfo report.txt platforminfo \ ghc-measurements printf "%s " "$(HSPROGS)" "$(HANDPROGS)" "$(CPROGS)" > ghc-measurements/progs tar -zcf ghc-measurements.tar.gz ghc-measurements rm -rf ghc-measurements help: @echo 'Measurements of very simple string I/O and parsing.' @printf ' (%d benchmarks, %s)\n' `echo $(HSPROGS) $(HANDPROGS) $(CPROGS) | wc -w` "$(HANDTEXT)" @echo '' @echo ' phase1 -- preparation + measurements that can run in background' @echo ' phase2 -- measurements that should run on unloaded machine' @echo ' redophase2 -- rerun phase2' @echo '' @echo ' doc, [ASCII=1] report, lastreport - reports' @echo ' zipdata -- zip up measurements (to ghc-measurements.tar.gz)' @echo '' @echo ' prog,core,stg,cmm,asm,dis,discut' @echo ' -- compile, compile to core/stg/cmm/asm, disassemble, cut out main loop' @echo ' time,stat,mem,strace,iotrace,iosum,cache' @echo ' -- measure run-time, GHC heap + OS mem, syscalls, I/O patterns, cache' @echo '' @echo ' cleartime, clean, distclean -- delete measurements etc' @echo '' @echo ' TESTKIND=(SMOKETEST,NORMAL,THOROUGH), defaults to $(TESTKINDDEFAULT)' @echo ' STRACE=OLD, defaults to NEW' cleartime: rm -f */*.time clean: # keep and hand/*.s ! rm -rf */*.hi */*.o *.o \ */*.core */*.stg */*.cmm hs/*.s c/*.s */*.dis */*.discut \ */*.hcr \ */*.time */*.stat \ */*.real \ */*.strace */*.iotrace */*.iosum \ */*.mem */*.cache cachegrind.out.* \ */*.doc */*.srctimespace \ tmp.strace tmp \ tools/eatmem tools/pause-at-end.so \ $(HSPROGS) $(CPROGS) $(HANDPROGS) $(RMPROGS) a.out \ testfiles/ \ ghc-measurements/ \ sysinfo platforminfo docs xx.ps distclean: clean rm -f *~ */*~ report.txt ghc-measurements.tar.gz

1 0

[2/16] SBM: Inner loops of the hand-tweaked assembly benchmarks
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

I've taken the two benchmarks byte-bs----acc and space-bs-c8-acc-1 and gradually tweaked their inner loops from something that used memory all the time to something that used registers more and more efficiently. I've done this gradually, pretty much one register at a time. Along the way, I've also done a simple common subexpression/loop hoisting thing in which I combined the pointer to the start of the string and the index into the string into a single pointer. Doing this in real life may cause bad problems with the garbage collector. At the end, I go a bit mad and start doing heroic optimizations (reading four bytes at a time, using MMX registers to read 8 bytes at a time, twisted MMX math to keep 8 space counters in an MMX register + a bit of loop unrolling). Here follows first the two original inner loops and then the 23 hand-tweaked versions. I used the following shell code to isolate the inner loops: (for F in hs/byte-bs----acc.s hs/space-bs-c8-acc-1.s hand/*.s ; \ do echo "------------------------------"; \ echo "$F:"; \ echo ; \ cat "$F" | perl -e 'while(<>){ if (/Main_zdwcnt_info:/ .. /.section .data/) { print; }}' | head -n-1; \ done; \ echo "=============================="; \ ) > xx.txt -Peter ------------------------------ hs/byte-bs----acc.s: Main_zdwcnt_info: .LcYL: cmpl $0,16(%ebp) jle .LcYO movl 12(%ebp),%eax incl %eax movl (%ebp),%ecx incl %ecx subl $1,16(%ebp) movl %eax,12(%ebp) movl %ecx,(%ebp) jmp Main_zdwcnt_info .LcYO: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) ------------------------------ hs/space-bs-c8-acc-1.s: Main_zdwcnt_info: .Lc16u: cmpl $0,16(%ebp) jle .Lc16x movl 4(%ebp),%eax movl 12(%ebp),%ecx movzbl (%eax,%ecx,1),%eax cmpl $32,%eax jne .Lc16F movl 12(%ebp),%eax incl %eax movl (%ebp),%ecx incl %ecx subl $1,16(%ebp) movl %eax,12(%ebp) movl %ecx,(%ebp) jmp Main_zdwcnt_info .Lc16x: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) .Lc16F: movl 12(%ebp),%eax incl %eax subl $1,16(%ebp) movl %eax,12(%ebp) jmp Main_zdwcnt_info ------------------------------ hand/byte-bs----acc-a.s: Main_zdwcnt_info: .LcYN: cmpl $0,16(%ebp) jle .LcYQ movl 00(%ebp),%ecx movl 12(%ebp),%eax movl 16(%ebp),%edx incl %ecx incl %eax decl %edx movl %ecx,00(%ebp) movl %eax,12(%ebp) movl %edx,16(%ebp) jmp Main_zdwcnt_info .LcYQ: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) ------------------------------ hand/byte-bs----acc-b.s: Main_zdwcnt_info: .LcYN: cmpl $0,16(%ebp) jle .LcYQ movl 00(%ebp),%ecx movl 12(%ebp),%eax movl 16(%ebp),%edx .L_again: cmpl $0,%edx jle .L_out incl %ecx incl %eax decl %edx jmp .L_again .L_out: movl %ecx,00(%ebp) movl %eax,12(%ebp) movl %edx,16(%ebp) jmp Main_zdwcnt_info .LcYQ: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) ------------------------------ hand/byte-bs----acc-c.s: Main_zdwcnt_info: .LcYN: cmpl $0,16(%ebp) jle .LcYQ movl 00(%ebp),%ecx movl 12(%ebp),%eax movl 16(%ebp),%edx cmpl $0,%edx jle .L_out .L_again: incl %ecx incl %eax decl %edx cmpl $0,%edx jg .L_again .L_out: movl %ecx,00(%ebp) movl %eax,12(%ebp) movl %edx,16(%ebp) jmp Main_zdwcnt_info .LcYQ: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) ------------------------------ hand/byte-bs----acc-d.s: Main_zdwcnt_info: .LcYN: cmpl $0,16(%ebp) jle .LcYQ movl 00(%ebp),%ecx movl 12(%ebp),%eax movl 16(%ebp),%edx cmpl $0,%edx jle .L_out .align 16 .L_again: incl %ecx incl %eax decl %edx cmpl $0,%edx jg .L_again .L_out: movl %ecx,00(%ebp) movl %eax,12(%ebp) movl %edx,16(%ebp) jmp Main_zdwcnt_info .LcYQ: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-a.s: Main_zdwcnt_info: .Lc16w: cmpl $0,16(%ebp) jle .Lc16z movl 4(%ebp),%eax movl 12(%ebp),%ecx movzbl (%eax,%ecx,1),%eax cmpl $32,%eax jne .Lc16H movl 12(%ebp),%eax incl %eax movl (%ebp),%ecx incl %ecx subl $1,16(%ebp) movl %eax,12(%ebp) movl %ecx,(%ebp) jmp Main_zdwcnt_info .Lc16H: movl 12(%ebp),%eax incl %eax subl $1,16(%ebp) movl %eax,12(%ebp) jmp Main_zdwcnt_info .Lc16z: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-b.s: Main_zdwcnt_info: .Lc16w: cmpl $0,16(%ebp) jle .Lc16z movl 4(%ebp),%eax movl 12(%ebp),%ecx movzbl (%eax,%ecx,1),%eax cmpl $32,%eax je .Lc16H movl 12(%ebp),%eax incl %eax subl $1,16(%ebp) movl %eax,12(%ebp) jmp Main_zdwcnt_info .Lc16H: movl 12(%ebp),%eax incl %eax movl (%ebp),%ecx incl %ecx subl $1,16(%ebp) movl %eax,12(%ebp) movl %ecx,(%ebp) jmp Main_zdwcnt_info .Lc16z: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-c.s: Main_zdwcnt_info: .Lc16w: cmpl $0,16(%ebp) jle .Lc16z movl 4(%ebp),%eax movl 12(%ebp),%ecx movzbl (%eax,%ecx,1),%eax cmpl $32,%eax jne .Lc16H movl (%ebp),%ecx incl %ecx movl 12(%ebp),%eax incl %eax movl %ecx,(%ebp) movl %eax,12(%ebp) subl $1,16(%ebp) jmp Main_zdwcnt_info .Lc16z: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) .Lc16H: movl 12(%ebp),%eax incl %eax movl %eax,12(%ebp) subl $1,16(%ebp) jmp Main_zdwcnt_info ------------------------------ hand/space-bs-c8-acc-1-d.s: Main_zdwcnt_info: .Lc16w: cmpl $0,16(%ebp) jle .Lc16z movl 4(%ebp),%eax movl 12(%ebp),%ecx movzbl (%eax,%ecx,1),%eax cmpl $32,%eax jne .Lc16H addl $1,(%ebp) addl $1,12(%ebp) subl $1,16(%ebp) jmp Main_zdwcnt_info .Lc16z: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) .Lc16H: addl $1,12(%ebp) subl $1,16(%ebp) jmp Main_zdwcnt_info ------------------------------ hand/space-bs-c8-acc-1-e.s: Main_zdwcnt_info: .Lc16w: cmpl $0,16(%ebp) jle .Lc16z movl 4(%ebp),%eax movl 12(%ebp),%ecx movzbl (%eax,%ecx,1),%eax cmpl $32,%eax jne .Lc16H movl 12(%ebp),%eax incl %eax incl %ecx movl (%ebp),%eax incl %eax subl $1,16(%ebp) movl %ecx,12(%ebp) movl %eax,(%ebp) jmp Main_zdwcnt_info .Lc16z: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) .Lc16H: incl %ecx subl $1,16(%ebp) movl %ecx,12(%ebp) jmp Main_zdwcnt_info ------------------------------ hand/space-bs-c8-acc-1-f.s: Main_zdwcnt_info: .Lc16w: cmpl $0,16(%ebp) jle .Lc16z movl 4(%ebp),%eax movl 12(%ebp),%ecx movzbl (%eax,%ecx,1),%eax cmpl $32,%eax jne .Lc16H incl %ecx subl $1,16(%ebp) addl $1,(%ebp) movl %ecx,12(%ebp) jmp Main_zdwcnt_info .Lc16z: movl (%ebp),%esi addl $20,%ebp jmp *(%ebp) .Lc16H: incl %ecx subl $1,16(%ebp) movl %ecx,12(%ebp) jmp Main_zdwcnt_info ------------------------------ hand/space-bs-c8-acc-1-g.s: Main_zdwcnt_info: movl (%ebp),%esi .Lc16w: cmpl $0,16(%ebp) jle .Lc16z movl 4(%ebp),%eax movl 12(%ebp),%ecx movzbl (%eax,%ecx,1),%eax cmpl $32,%eax jne .Lc16H incl %ecx subl $1,16(%ebp) inc %esi movl %ecx,12(%ebp) jmp .Lc16w .Lc16z: addl $20,%ebp jmp *(%ebp) .Lc16H: incl %ecx subl $1,16(%ebp) movl %ecx,12(%ebp) jmp .Lc16w ------------------------------ hand/space-bs-c8-acc-1-h.s: Main_zdwcnt_info: movl (%ebp),%esi movl 12(%ebp),%ecx .Lc16w: cmpl $0,16(%ebp) jle .Lc16z movl 4(%ebp),%eax movzbl (%eax,%ecx,1),%eax cmpl $32,%eax jne .Lc16H incl %ecx subl $1,16(%ebp) inc %esi jmp .Lc16w .Lc16z: addl $20,%ebp jmp *(%ebp) .Lc16H: incl %ecx subl $1,16(%ebp) jmp .Lc16w ------------------------------ hand/space-bs-c8-acc-1-i.s: Main_zdwcnt_info: movl (%ebp),%esi movl 12(%ebp),%ecx movl 16(%ebp),%edx .Lc16w: cmpl $0,%edx jle .Lc16z movl 4(%ebp),%eax movzbl (%eax,%ecx,1),%eax cmpl $32,%eax jne .Lc16H incl %ecx decl %edx inc %esi jmp .Lc16w .Lc16z: addl $20,%ebp jmp *(%ebp) .Lc16H: incl %ecx decl %edx jmp .Lc16w ------------------------------ hand/space-bs-c8-acc-1-j.s: Main_zdwcnt_info: movl (%ebp),%esi movl 4(%ebp),%ecx addl 12(%ebp),%ecx movl 16(%ebp),%edx .Lc16w: cmpl $0,%edx jle .Lc16z movzbl (%ecx),%eax cmpl $32,%eax jne .Lc16H incl %ecx decl %edx inc %esi jmp .Lc16w .Lc16z: addl $20,%ebp jmp *(%ebp) .Lc16H: incl %ecx decl %edx jmp .Lc16w ------------------------------ hand/space-bs-c8-acc-1-k.s: Main_zdwcnt_info: movl (%ebp),%esi movl 4(%ebp),%ecx addl 12(%ebp),%ecx movl 16(%ebp),%edx .Lc16w: cmpl $0,%edx jle .Lc16z movzbl (%ecx),%eax cmpl $32,%eax jne .Lc16H incl %ecx decl %edx inc %esi jmp .Lc16w .Lc16H: incl %ecx decl %edx jmp .Lc16w .Lc16z: addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-l.s: Main_zdwcnt_info: movl (%ebp),%esi movl 4(%ebp),%ecx addl 12(%ebp),%ecx movl 16(%ebp),%edx .Lc16w: cmpl $0,%edx jle .Lc16z movzbl (%ecx),%eax incl %ecx decl %edx cmpl $32,%eax jne .Lc16H inc %esi jmp .Lc16w .Lc16H: jmp .Lc16w .Lc16z: addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-m.s: Main_zdwcnt_info: movl (%ebp),%esi movl 4(%ebp),%ecx addl 12(%ebp),%ecx movl 16(%ebp),%edx .Lc16w: cmpl $0,%edx jle .Lc16z movzbl (%ecx),%eax incl %ecx decl %edx cmpl $32,%eax jne .Lc16w inc %esi jmp .Lc16w .Lc16z: addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-n.s: Main_zdwcnt_info: movl (%ebp),%esi movl 4(%ebp),%ecx addl 12(%ebp),%ecx movl 16(%ebp),%edx .Lc16w: cmpl $0,%edx jle .Lc16z .Lc16xx: movzbl (%ecx),%eax incl %ecx decl %edx cmpl $32,%eax jne .Lc16w inc %esi cmpl $0,%edx jg .Lc16xx .Lc16z: addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-o.s: Main_zdwcnt_info: movl (%ebp),%esi movl 4(%ebp),%ecx addl 12(%ebp),%ecx movl 16(%ebp),%edx .Lc16w: cmpl $0,%edx jle .Lc16z .Lc16xx: movzbl (%ecx),%eax incl %ecx decl %edx cmpl $32,%eax jne .Lc16w inc %esi cmpl $0,%edx jle .Lc16z movzbl (%ecx),%eax incl %ecx decl %edx cmpl $32,%eax jne .Lc16w inc %esi cmpl $0,%edx jg .Lc16xx .Lc16z: addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-p.s: Main_zdwcnt_info: movl (%ebp),%esi movl 4(%ebp),%ecx addl 12(%ebp),%ecx movl 16(%ebp),%edx .Lc16w4: cmpl $4,%edx jl .Lc16wxx movl (%ecx),%eax addl $4,%ecx subl $4,%edx cmpb $32,%al jne .Lc16wa incl %esi .Lc16wa: cmpb $32,%ah jne .Lc16wb incl %esi .Lc16wb: shrl $16,%eax cmpb $32,%al jne .Lc16wc incl %esi .Lc16wc: cmpb $32,%ah jne .Lc16w4 incl %esi jmp .Lc16w4 .Lc16w1: cmpl $0,%edx jle .Lc16z .Lc16wxx: movzbl (%ecx),%eax incl %ecx decl %edx cmpl $32,%eax jne .Lc16w1 inc %esi jmp .Lc16w1 .Lc16z: addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-q.s: Main_zdwcnt_info: movl (%ebp),%esi /* #spaces found */ movl 4(%ebp),%ecx /* ptr */ addl 12(%ebp),%ecx /* ... + idx */ movl 16(%ebp),%edx /* cnt of remaining bytes */ emms /* clear fp tags so we can use mmx instrs */ mov $0x20202020,%eax movd %eax,%mm1 /* mm1: 0000000020202020 */ movq %mm1,%mm0 /* mm0: 0000000020202020 */ psllq $32,%mm1 /* mm1: 2020202000000000 */ por %mm0,%mm1 /* mm1: 2020202020202020 */ mov $0x01010101,%eax movd %eax,%mm2 /* mm2: 0000000001010101 */ movq %mm2,%mm0 /* mm0: 0000000001010101 */ psllq $32,%mm2 /* mm2: 0101010100000000 */ por %mm0,%mm2 /* mm2: 0101010101010101 */ /* MMX loads can use any alignment (potentially at a speed-hit) */ /* this loop looks at 8 bytes at a time */ .Lc16w8: cmpl $8,%edx jl .Lc16w1 movq (%ecx),%mm0 /* mm0 holds 8 characters */ addl $8,%ecx subl $8,%edx pcmpeqb %mm1,%mm0 /* cmp byte for byte with ' ' */ /* the result flag is 00 or FF */ pand %mm2,%mm0 /* turn FF into 01, which is actually useful */ /* if we could just add the bytes up horizontally in %mm0, sigh.. .*/ movd %mm0,%eax push %eax add %ah, %al and $0x03,%eax add %eax,%esi pop %eax shr $16,%eax add %ah,%al and $0x03,%eax add %eax,%esi psrlq $32,%mm0 movd %mm0,%eax push %eax add %ah, %al and $0x03,%eax add %eax,%esi pop %eax shr $16,%eax add %ah,%al and $0x03,%eax add %eax,%esi jmp .Lc16w8 /* this loop looks at one byte at a time to handle the remainder */ .Lc16w1: cmpl $0,%edx jle .Lc16z movzbl (%ecx),%eax incl %ecx decl %edx cmpl $32,%eax jne .Lc16w1 inc %esi jmp .Lc16w1 /* done, remember to clear fp/mmx tags with emms */ .Lc16z: emms addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-r.s: Main_zdwcnt_info: movl (%ebp),%esi /* #spaces found */ movl 4(%ebp),%ecx /* ptr */ addl 12(%ebp),%ecx /* ... + idx */ movl 16(%ebp),%edx /* cnt of remaining bytes */ emms /* clear fp tags so we can use mmx instrs */ mov $0x20202020,%eax movd %eax,%mm1 /* mm1: 0000000020202020 */ movq %mm1,%mm0 /* mm0: 0000000020202020 */ psllq $32,%mm1 /* mm1: 2020202000000000 */ por %mm0,%mm1 /* mm1: 2020202020202020 */ mov $0x01010101,%eax movd %eax,%mm2 /* mm2: 0000000001010101 */ movq %mm2,%mm0 /* mm0: 0000000001010101 */ psllq $32,%mm2 /* mm2: 0101010100000000 */ por %mm0,%mm2 /* mm2: 0101010101010101 */ /* MMX loads can use any alignment (potentially at a speed-hit) */ /* therefore we don't have to try to read 1-7 bytes one at a time */ /* first in order to end up with an aligned %ecx. */ .Lc16_mainloop: cmpl $8,%edx jl .Lc16w1 movl %edx,%eax shr $3,%eax cmpl $127,%eax jle .Lc16_127 movl $127,%eax .Lc16_127: shl $3,%eax sub %eax,%edx shr $3,%eax pxor %mm3,%mm3 /* clear block of space counters */ /* loop up to 127 times in a loop that looks at 8 bytes at a time. */ /* Going above 255 could overflow the 8 counters in mm3. */ /* Going above 127 could overflow the horizontal summation code. */ .Lc16w8: cmpl $0,%eax jle .Lc16w8end movq (%ecx),%mm0 /* mm0 holds 8 characters */ addl $8,%ecx decl %eax pcmpeqb %mm1,%mm0 /* cmp byte for byte with ' ' */ /* the result flag is 00 or FF */ pand %mm2,%mm0 /* turn FF into 01, which is actually useful */ paddb %mm0,%mm3 /* add to the 8 space counters */ jmp .Lc16w8 .Lc16w8end: /* sum the 8 space counters in mm3 and add to %esi */ /* if only MMX had horizontal byte adds... */ movd %mm3,%eax push %eax add %ah, %al /* NOTE! potential overflow! */ and $0xFF,%eax add %eax,%esi pop %eax shr $16,%eax add %ah,%al /* NOTE! potential overflow! */ and $0xFF,%eax add %eax,%esi psrlq $32,%mm3 movd %mm3,%eax push %eax add %ah, %al /* NOTE! potential overflow! */ and $0xFF,%eax add %eax,%esi pop %eax shr $16,%eax add %ah,%al /* NOTE! potential overflow! */ and $0xFF,%eax add %eax,%esi jmp .Lc16_mainloop /* this loop looks at one byte at a time to handle the remainder */ .Lc16w1: cmpl $0,%edx jle .Lc16z movzbl (%ecx),%eax incl %ecx decl %edx cmpl $32,%eax jne .Lc16w1 inc %esi jmp .Lc16w1 /* done, remember to clear fp/mmx tags with emms */ .Lc16z: emms addl $20,%ebp jmp *(%ebp) ------------------------------ hand/space-bs-c8-acc-1-s.s: Main_zdwcnt_info: movl (%ebp),%esi /* #spaces found */ movl 4(%ebp),%ecx /* ptr */ addl 12(%ebp),%ecx /* ... + idx */ movl 16(%ebp),%edx /* cnt of remaining bytes */ emms /* clear fp tags so we can use mmx instrs */ mov $0x20202020,%eax movd %eax,%mm1 /* mm1: 0000000020202020 */ movq %mm1,%mm0 /* mm0: 0000000020202020 */ psllq $32,%mm1 /* mm1: 2020202000000000 */ por %mm0,%mm1 /* mm1: 2020202020202020 */ mov $0x01010101,%eax movd %eax,%mm2 /* mm2: 0000000001010101 */ movq %mm2,%mm0 /* mm0: 0000000001010101 */ psllq $32,%mm2 /* mm2: 0101010100000000 */ por %mm0,%mm2 /* mm2: 0101010101010101 */ /* MMX loads can use any alignment (potentially at a speed-hit) */ /* therefore we don't have to try to read 1-7 bytes one at a time */ /* first in order to end up with an aligned %ecx. */ .Lc16_mainloop: cmpl $8,%edx jl .Lc16w1 movl %edx,%eax shr $3,%eax cmpl $127,%eax jle .Lc16_127 movl $127,%eax .Lc16_127: shl $3,%eax sub %eax,%edx shr $3,%eax pxor %mm3,%mm3 /* clear block of space counters */ /* loop up to 127 times in a loop that looks at 8 bytes at a time. */ /* Going above 255 could overflow the 8 counters in mm3. */ /* Going above 127 could overflow the horizontal summation code. */ cmpl $0,%eax jle .Lc16w8end /* this is an unspeakably ugly and sloppy loop unrolling. Doesn't */ /* seem to help much on an Athlon64 3000+. */ test $1,%eax jz .Lc16w8 incl %eax jmp .Lc16w8x .Lc16w8: movq (%ecx),%mm0 /* mm0 holds 8 characters */ addl $8,%ecx pcmpeqb %mm1,%mm0 /* cmp byte for byte with ' ' */ /* the result flag is 00 or FF */ pand %mm2,%mm0 /* turn FF into 01, which is actually useful */ paddb %mm0,%mm3 /* add to the 8 space counters */ .Lc16w8x: movq (%ecx),%mm0 /* mm0 holds 8 characters */ addl $8,%ecx pcmpeqb %mm1,%mm0 /* cmp byte for byte with ' ' */ /* the result flag is 00 or FF */ pand %mm2,%mm0 /* turn FF into 01, which is actually useful */ paddb %mm0,%mm3 /* add to the 8 space counters */ subl $2,%eax jnz .Lc16w8 .Lc16w8end: /* sum the 8 space counters in mm3 and add to %esi */ /* if only MMX had horizontal byte adds... */ movd %mm3,%eax push %eax add %ah, %al /* NOTE! potential overflow! */ and $0xFF,%eax add %eax,%esi pop %eax shr $16,%eax add %ah,%al /* NOTE! potential overflow! */ and $0xFF,%eax add %eax,%esi psrlq $32,%mm3 movd %mm3,%eax push %eax add %ah, %al /* NOTE! potential overflow! */ and $0xFF,%eax add %eax,%esi pop %eax shr $16,%eax add %ah,%al /* NOTE! potential overflow! */ and $0xFF,%eax add %eax,%esi jmp .Lc16_mainloop /* this loop looks at one byte at a time to handle the remainder */ .Lc16w1: cmpl $0,%edx jle .Lc16z movzbl (%ecx),%eax incl %ecx decl %edx cmpl $32,%eax jne .Lc16w1 inc %esi jmp .Lc16w1 /* done, remember to clear fp/mmx tags with emms */ .Lc16z: emms addl $20,%ebp jmp *(%ebp) ==============================

1 0

[1/16] SBM: The Haskell and C benchmarks
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

Here are the 48 Haskell and C benchmarks. Don Stewart contributed three (although I had to fight a bit to make one of them compile). Jules Bean (quicksilver) contributed one. Bertram Felgenhauer (int-e) contributed three (in the form of a single file, which I untangled). Spencer Jannsen (sjannsen) contributed one. wli (William Lee Irwin III) inspired me to add the getwchar benchmarks. I used the following shell code to gather all the benchmarks: (for F in hs/*.hs c/*.c; \ do echo "------------------------------"; \ echo "$F:"; \ echo ; \ cat "$F"; \ done; \ echo "==============================" \ ) > xx.txt They are not in the same order as in the Makefile or in the reports, unfortunately. -Peter ------------------------------ hs/byte-bs----acc.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString as B cnt :: Int -> B.ByteString -> Int cnt !acc !bs = if B.null bs then acc else cnt (acc+1) (B.tail bs) main = do s <- B.getContents print (cnt 0 s) ------------------------------ hs/byte-bs----foldlx.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString as B cnt :: B.ByteString -> Int cnt !bs = B.foldl' (\sum _ -> sum+1) 0 bs main = do s <- B.getContents print (cnt s) ------------------------------ hs/byte-bs----foldrx.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString as B cnt :: B.ByteString -> Int cnt !bs = B.foldr' (\_ sum -> sum+1) 0 bs main = do s <- B.getContents print (cnt s) ------------------------------ hs/byte-bsl---acc.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Lazy as B cnt :: Int -> B.ByteString -> Int cnt !acc !bs = if B.null bs then acc else cnt (acc+1) (B.tail bs) main = do s <- B.getContents print (cnt 0 s) ------------------------------ hs/byte-xxxxx-acc-1.hs: {-# LANGUAGE BangPatterns #-} cnt :: Int -> String -> Int cnt !acc bs = if null bs then acc else cnt (acc+1) (tail bs) main = do s <- getContents print (cnt 0 s) ------------------------------ hs/byte-xxxxx-acc-2.hs: {-# LANGUAGE BangPatterns #-} cnt :: Int -> String -> Int cnt !acc !bs = if null bs then acc else cnt (acc+1) (tail bs) main = do s <- getContents print (cnt 0 s) ------------------------------ hs/byte-xxxxx-foldl.hs: {-# LANGUAGE BangPatterns #-} cnt :: String -> Int cnt !bs = foldl (\sum _ -> sum+1) 0 bs main = do s <- getContents print (cnt s) ------------------------------ hs/byte-xxxxx-foldr-1.hs: {-# LANGUAGE BangPatterns #-} cnt :: String -> Int cnt bs = foldr (\_ sum -> sum+1) 0 bs main = do s <- getContents print (cnt s) ------------------------------ hs/byte-xxxxx-foldr-2.hs: {-# LANGUAGE BangPatterns #-} cnt :: String -> Int cnt !bs = foldr (\_ sum -> sum+1) 0 bs main = do s <- getContents print (cnt s) ------------------------------ hs/space-bs-c8-acc-1.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Char8 as B cnt :: Int -> B.ByteString -> Int cnt !acc bs = if B.null bs then acc else cnt (if B.head bs == ' ' then acc+1 else acc) (B.tail bs) main = do s <- B.getContents print (cnt 0 s) ------------------------------ hs/space-bs-c8-count.hs: -- Don Stewart import qualified Data.ByteString.Char8 as B main = print . B.count ' ' =<< B.getContents ------------------------------ hs/space-bs-c8-foldlx-1.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Char8 as B cnt :: B.ByteString -> Int cnt bs = B.foldl' (\sum c -> if c == ' ' then sum+1 else sum) 0 bs main = do s <- B.getContents print (cnt s) ------------------------------ hs/space-bs-c8-foldlx-2.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Char8 as B main = do s <- B.getContents print $ B.foldl' (\v c -> if c == ' ' then v+1 else v :: Int) 0 s ------------------------------ hs/space-bs-c8-foldrx.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Char8 as B cnt :: B.ByteString -> Int cnt bs = B.foldr' (\c sum -> if c == ' ' then sum+1 else sum) 0 bs main = do s <- B.getContents print (cnt s) ------------------------------ hs/space-bs-c8-lenfil.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Char8 as B cnt :: B.ByteString -> Int cnt bs = B.length (B.filter (== ' ') bs) main = do s <- B.getContents print (cnt s) ------------------------------ hs/space-bslc8-acc-1.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Lazy.Char8 as B cnt :: Int -> B.ByteString -> Int cnt !acc bs = if B.null bs then acc else cnt (if B.head bs == ' ' then acc+1 else acc) (B.tail bs) main = do s <- B.getContents print (cnt 0 s) ------------------------------ hs/space-bslc8-acc-2.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Lazy.Char8 as B cnt :: Int -> B.ByteString -> Int cnt !acc !bs = if B.null bs then acc else cnt (if B.head bs == ' ' then acc+1 else acc) (B.tail bs) main = do s <- B.getContents print (cnt 0 s) ------------------------------ hs/space-bslc8-acc-3.hs: {-# LANGUAGE BangPatterns #-} -- this version by quicksilver import qualified Data.ByteString.Lazy.Char8 as B cnt :: Int -> B.ByteString -> Int cnt !acc bs | B.null bs = acc | B.head bs == ' ' = cnt (acc+1) (B.tail bs) | otherwise = cnt acc (B.tail bs) main = do s <- B.getContents print (cnt 0 s) ------------------------------ hs/space-bslc8-chunk-1.hs: {-# LANGUAGE BangPatterns #-} -- this version by int-e import qualified Data.ByteString.Lazy.Char8 as B import qualified Data.ByteString.Char8 as BS import Data.List (foldl') cntS :: Int -> BS.ByteString -> Int cntS !acc !bs = case BS.uncons bs of Nothing -> acc Just (hd, tl) | hd == ' ' -> cntS (acc+1) tl | otherwise -> cntS acc tl cnt :: Int -> B.ByteString -> Int cnt acc bs = foldl' cntS acc (B.toChunks bs) main = do s <- B.getContents print $ cnt 0 s ------------------------------ hs/space-bslc8-chunk-2.hs: {-# LANGUAGE BangPatterns #-} -- this version by int-e import qualified Data.ByteString.Lazy.Char8 as B import qualified Data.ByteString.Char8 as BS import Data.List (foldl') cntS' :: Int -> BS.ByteString -> Int cntS' !acc !bs | BS.null bs = acc | BS.head bs == ' ' = cntS' (acc+1) (BS.tail bs) | otherwise = cntS' acc (BS.tail bs) cnt :: Int -> B.ByteString -> Int cnt acc bs = foldl' cntS' acc (B.toChunks bs) main = do s <- B.getContents print $ cnt 0 s ------------------------------ hs/space-bslc8-chunk-3.hs: {-# LANGUAGE BangPatterns #-} -- this version by int-e import qualified Data.ByteString.Lazy.Char8 as B import qualified Data.ByteString.Char8 as BS import Data.List (foldl') cntS'' :: Int -> BS.ByteString -> Int cntS'' !acc !bs = BS.foldl' (\v c -> if c == ' ' then v+1 else v) acc bs cnt :: Int -> B.ByteString -> Int cnt acc bs = foldl' cntS'' acc (B.toChunks bs) main = do s <- B.getContents print $ cnt 0 s ------------------------------ hs/space-bslc8-chunk-4.hs: {-# LANGUAGE BangPatterns #-} -- Don Stewart import qualified Data.ByteString.Lazy.Char8 as BLC8 import qualified Data.ByteString.Lazy.Internal as BLI import qualified Data.ByteString as B import qualified Data.ByteString.Unsafe as BU import qualified Data.ByteString.Internal as BI cnt :: Int -> BLC8.ByteString -> Int cnt n BLI.Empty = n cnt n (BLI.Chunk x xs) = cnt (n + cnt_strict 0 x) xs -- process lazy spine where -- now we can process a chunk without checking for Empty cnt_strict !i !s -- then strict chunk | B.null s = i | c == ' ' = cnt_strict (i+1) t | otherwise = cnt_strict i t where (c,t) = (BI.w2c (BU.unsafeHead s), BU.unsafeTail s) -- no bounds check main = do s <- BLC8.getContents; print (cnt 0 s) ------------------------------ hs/space-bslc8-count.hs: -- Don Stewart import qualified Data.ByteString.Lazy.Char8 as B main = print . B.count ' ' =<< B.getContents ------------------------------ hs/space-bslc8-foldl.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Lazy.Char8 as B cnt :: B.ByteString -> Int cnt !bs = B.foldl (\sum c -> if c == ' ' then sum+1 else sum) 0 bs main = do s <- B.getContents print (cnt s) ------------------------------ hs/space-bslc8-foldlx-1.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Lazy.Char8 as B cnt :: B.ByteString -> Int cnt bs = B.foldl' (\sum c -> if c == ' ' then sum+1 else sum) 0 bs main = do s <- B.getContents print (cnt s) ------------------------------ hs/space-bslc8-foldlx-2.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Lazy.Char8 as B cnt :: B.ByteString -> Int cnt !bs = B.foldl' (\sum c -> if c == ' ' then sum+1 else sum) 0 bs main = do s <- B.getContents print (cnt s) ------------------------------ hs/space-bslc8-foldr-1.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Lazy.Char8 as B cnt :: B.ByteString -> Int cnt bs = B.foldr (\c sum -> if c == ' ' then sum+1 else sum) 0 bs main = do s <- B.getContents print (cnt s) ------------------------------ hs/space-bslc8-foldr-2.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Lazy.Char8 as B cnt :: B.ByteString -> Int cnt !bs = B.foldr (\c sum -> if c == ' ' then sum+1 else sum) 0 bs main = do s <- B.getContents print (cnt s) ------------------------------ hs/space-bslc8-lenfil-1.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Lazy.Char8 as B import GHC.Int (Int64) -- note that D.BS.Lazy.Char8.length is ByteString -> Int64 -- D.BS.C8.length is ByteString -> Int cnt :: B.ByteString -> Int64 cnt bs = B.length (B.filter (== ' ') bs) main = do s <- B.getContents print (cnt s) ------------------------------ hs/space-bslc8-lenfil-2.hs: {-# LANGUAGE BangPatterns #-} import qualified Data.ByteString.Lazy.Char8 as B import GHC.Int (Int64) -- note that D.BS.Lazy.Char8.length is ByteString -> Int64 -- D.BS.C8.length is ByteString -> Int cnt :: B.ByteString -> Int64 cnt !bs = B.length (B.filter (== ' ') bs) main = do s <- B.getContents print (cnt s) ------------------------------ hs/space-bsl---foldlx.hs: {-# LANGUAGE BangPatterns #-} -- this version by sjannsen import Data.ByteString.Lazy as B cnt :: B.ByteString -> Int cnt = B.foldl' f 0 where f !n 32 = n+1 f !n _ = n main = do s <- B.getContents print $ cnt s ------------------------------ hs/space-xxxxx-acc-1.hs: {-# LANGUAGE BangPatterns #-} cnt :: Int -> String -> Int cnt !acc bs = if null bs then acc else cnt (if head bs == ' ' then acc+1 else acc) (tail bs) main = do s <- getContents print (cnt 0 s) ------------------------------ hs/space-xxxxx-acc-2.hs: {-# LANGUAGE BangPatterns #-} cnt :: Int -> String -> Int cnt !acc !bs = if null bs then acc else cnt (if head bs == ' ' then acc+1 else acc) (tail bs) main = do s <- getContents print (cnt 0 s) ------------------------------ hs/space-xxxxx-foldl.hs: {-# LANGUAGE BangPatterns #-} cnt :: String -> Int cnt bs = foldl (\sum c -> if c == ' ' then sum+1 else sum) 0 bs main = do s <- getContents print (cnt s) ------------------------------ hs/space-xxxxx-foldr-1.hs: {-# LANGUAGE BangPatterns #-} cnt :: String -> Int cnt bs = foldr (\c sum -> if c == ' ' then sum+1 else sum) 0 bs main = do s <- getContents print (cnt s) ------------------------------ hs/space-xxxxx-foldr-2.hs: {-# LANGUAGE BangPatterns #-} cnt :: String -> Int cnt !bs = foldr (\c sum -> if c == ' ' then sum+1 else sum) 0 bs main = do s <- getContents print (cnt s) ------------------------------ hs/space-xxxxx-lenfil.hs: {-# LANGUAGE BangPatterns #-} cnt :: String -> Int cnt bs = length (filter (== ' ') bs) main = do s <- getContents print (cnt s) ------------------------------ c/byte-4k.c: #include <stdio.h> #include <unistd.h> #include <stdlib.h> #include <errno.h> int Main_cnt() { int cnt = 0; ssize_t sze; char buf[4*1024]; do { again: sze = read(fileno(stdin), buf, sizeof(buf)); if (sze < 0) { switch (errno) { case EAGAIN: goto again; default: perror("read() failed\n"); exit(1); } } cnt += sze; } while (sze != 0); return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ------------------------------ c/byte-getchar.c: #include <stdio.h> #include <stdlib.h> int Main_cnt() { int cnt = 0; int c; while ((c = getchar()) != EOF) cnt++; return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ------------------------------ c/byte-getchar-u.c: #include <stdio.h> #include <stdlib.h> int Main_cnt() { int cnt = 0; int c; while ((c = getchar_unlocked()) != EOF) cnt++; return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ------------------------------ c/space-32k-8.c: #include <stdio.h> #include <unistd.h> #include <stdlib.h> #include <errno.h> int Main_cnt() { int cnt = 0; ssize_t sze, left; char buf[32760]; char *p; printf("using a buffer of %g KB\n", sizeof(buf) / 1024.0); do { again: sze = read(fileno(stdin), buf, sizeof(buf)); if (sze < 0) { switch (errno) { case EAGAIN: goto again; default: perror("read() failed\n"); exit(1); } } for (p = buf, left=sze; left > 0; left--) if (*p++ == ' ') cnt++; } while (sze != 0); return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ------------------------------ c/space-32k.c: #include <stdio.h> #include <unistd.h> #include <stdlib.h> #include <errno.h> int Main_cnt() { int cnt = 0; ssize_t sze, left; char buf[32*1024]; char *p; printf("using a buffer of %g KB\n", sizeof(buf) / 1024.0); do { again: sze = read(fileno(stdin), buf, sizeof(buf)); if (sze < 0) { switch (errno) { case EAGAIN: goto again; default: perror("read() failed\n"); exit(1); } } for (p = buf, left=sze; left > 0; left--) if (*p++ == ' ') cnt++; } while (sze != 0); return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ------------------------------ c/space-4k.c: #include <stdio.h> #include <unistd.h> #include <stdlib.h> #include <errno.h> int Main_cnt() { int cnt = 0; ssize_t sze, left; char buf[4*1024]; char *p; printf("using a buffer of %g KB\n", sizeof(buf) / 1024.0); do { again: sze = read(fileno(stdin), buf, sizeof(buf)); if (sze < 0) { switch (errno) { case EAGAIN: goto again; default: perror("read() failed\n"); exit(1); } } for (p = buf, left=sze; left > 0; left--) if (*p++ == ' ') cnt++; } while (sze != 0); return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ------------------------------ c/space-getchar.c: #include <stdio.h> #include <stdlib.h> int Main_cnt() { int cnt = 0; int c; while ((c = getchar()) != EOF) if (c == ' ') cnt++; return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ------------------------------ c/space-getchar-u.c: #include <stdio.h> #include <stdlib.h> int Main_cnt() { int cnt = 0; int c; while ((c = getchar_unlocked()) != EOF) if (c == ' ') cnt++; return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ------------------------------ c/space-getwchar.c: #include <stdio.h> #include <wchar.h> #include <stdlib.h> int Main_cnt() { int cnt = 0; wint_t c; while ((c = getwchar()) != WEOF) if (c == ' ') cnt++; return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ------------------------------ c/space-getwchar-u.c: #define _GNU_SOURCE #include <stdio.h> #include <wchar.h> #include <stdlib.h> int Main_cnt() { int cnt = 0; wint_t c; while ((c = getwchar_unlocked()) != WEOF) if (c == ' ') cnt++; return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ------------------------------ c/space-megabuf.c: #include <stdio.h> #include <stdlib.h> #include <errno.h> #include <sys/stat.h> #include <unistd.h> int isfile(int handle) { struct stat buf; if (fstat(handle, &buf) == -1) { perror("fstat(stdin)\n"); exit(1); } return S_ISREG(buf.st_mode); } ssize_t getbufsize() { if (isfile(fileno(stdin))) { off_t x; x = lseek(fileno(stdin), 0, SEEK_END); if (x == -1) { perror("lseek(... SEEK_END)\n"); exit(1); } if (lseek(fileno(stdin), 0, SEEK_SET) == -1) { perror("lseek(... SEEK_SET)\n"); exit(1); } if (x > 1*1024*1024*1024LL) { x = 1024*1024*1024LL; } return x; /* file size for files */ } else { return 10*1024*1024; /* 10M for non-files */ } } int Main_cnt() { int cnt = 0, reads=0, retries=0; ssize_t sze, left, bufsize; char *buf; char *p; bufsize = getbufsize(); printf("using a buffer of %g MB\n", bufsize / (1024*1024.0)); buf = malloc(bufsize); if (!buf) { fprintf(stderr, "couldn't allocate %lld bytes\n", (long long) bufsize); } do { again: sze = read(fileno(stdin), buf, bufsize); if (sze < 0) { switch (errno) { case EAGAIN: retries++; goto again; default: perror("read() failed\n"); exit(1); } } reads++; for (p = buf, left=sze; left > 0; left--) if (*p++ == ' ') cnt++; } while (sze != 0); printf("%d reads, %d retries\n", reads, retries); return cnt; } int main() { printf("%d\n", Main_cnt()); return EXIT_SUCCESS; } ==============================

1 0

[0/16] SBM: Simple Bytestring Microbenchmarks, Overview and Introduction
by Peter Firefly Brodersen Lund 22 Dec '07

22 Dec '07

Table of contents: 0/16 This email 1/16 The Haskell and C benchmarks 2/16 Inner loops of the hand-tweaked assembly benchmarks 3/16 The Makefile 4/16 How to use the Makefile (how to run benchmarks etc.) 5/16 Support scripts and scriptlets 6/16 6.9.20071124 Athlon Duron 7/16 6.9.20071208 Core Duo 8/16 6.9.20071119 Pentium III 9/16 6.9.20071119 Athlon64 10/16 Graphs for 6.9.x across four cpus 11/16 Graphs for hand-tweaked assembly benchmarks 12/16 Graphs for 7 ghc/bytestring combinations on a 2GHz Athlon64 300+ 13/16 Graphs that show the infidelity of -sstderr 14/16 Behind the measurements (rationale) 15/16 Predictions compared to the measurements 16/16 Discussion and Conclusion Simple Bytestring Microbenchmarks --------------------------------- Introduction ------------ I love parsers. I have been writing parsers for fun for over twenty years. The nicest way to construct a parser used to be to write a recursive descent parser by hand. If you had to work with people who'd had the misfortune of a university education, you would resort to lex and yacc (flex and bison), despite their many shortcomings. Combinator parsers is the only real improvement over hand-written recursive descent parsers that I know of. They do tend to require features that not all languages provide. I don't know how to write a good one in C, for example. They do work very well in Haskell, though. So, I've started writing a parser in Haskell (ghc, really) for the programming language X++. X++ is not a nice language but that's beside the point. The challenge for me is to write an efficient compiler + provide good analysis tools for X++. I think I stand a better chance of doing that in Haskell (ghc) than in practically any other language. There are a few drawbacks, though. I love speed. And efficiency. String handling in Haskell -------------------------- Native strings are simple and generally work well, but they are slow, take up too much memory and there's the whole encoding mess that still needs to be sorted out. People have worked on other string representations and libraries for quite a while. I think packedstrings (as used in Darcs) was one of the first ones. Bytestrings is the current incarnation. It seems to be just the right thing, especially when combined with improved automatic fusion in the compiler so higher-order functions don't have to be expensive. I use Parsec as my parser combinator library at the moment, which uses native strings. I would dearly love Parsec to be faster and use less memory. I think bytestrings will be part of any substantial improvement in Parsec's resource consumption. Other performance concerns -------------------------- File I/O is also interesting for a compiler writer. I would like to have a program that is as fast as possible both when the source files are already cached by the operating system and when they are not. The former situation is best handled with mmap() and the latter is best handled by read(), preferably in combination with multi-threading so the compiler doesn't have to waste too much time waiting for disk seeks. Haskell seems to be very close to ideal for me because it has very good threading support and very accessible raw access to the operating system. File I/O is not my current bottleneck, though. I'll probably take a closer look at file I/O when the other performance problems have been solved. Then there's the general quality of the generated code. Having read just about every paper on ghc that was available back in the late nineties (when I first looked at Haskell), I'd thought that the quality was good and that the compiler also had extremely good high-level optimizations, in other words, that abstraction was free. I also read the C-- papers and thought that it was a very interesting and promising approach. I'd expected the C-- path to have matured and be well- optimized by now. Unfortunately, the backend is /the/ weak spot in ghc. The frontend is heroic, the typesystems are (too) abundant and rich, the language itself is nice -- but the backend is not. Looking at the generated code I'd say that it is slightly better than Turbo Pascal 3.x and about on par with Turbo Pascal 4.0, a compiler that didn't use any intermediate code at all, compiled each statement in isolation, was single-pass, and had a compilation speed of about 27000 lines per minute on an 8 MHz IBM PC AT. Ecosystem and culture --------------------- Haskell has a very good ecosystem. Probably the second best one amongst the modern functional languages. Ten years ago, I'd thought that Standard ML would win but the only MLish language with a good ecosystem and culture is OCaml, which unfortunately isn't really Standard ML. By ecosystem I mean things like access to raw operating system calls, access to libraries written in other languages, readily available libraries for graphical user interfaces, databases, XML processing, network I/O. Parsing is nice, too, but practically all functional programming languages have that -- and not much else. The culture is also good. People actually use this stuff. They care about it. And they don't hang around waiting for somebody to tell them what to do, they start on their own. And they actually seem to be interested in good performance :) Hackage and cabal are very promising already and may be what finally makes ghc real-world useful for more people, because most people are not interested in working with raw source packages and fiddling with compiler flags and weird error messages. They don't like chasing dependencies, either. So, what's the problem? ----------------------- The major problem with Haskell (ghc) is that its performance (in terms of both speed and memory use) is unpredictable. The second-worst problem is that the actual performance is not good enough. These benchmarks ---------------- I have written a bunch of microbenchmarks that either count all the bytes in stdin or all the spaces in stdin. And some supporting benchmarks in C. And I've also handtweaked the assembly of one byte-counting and one space-counting microbenchmark to illustrate what difference it would make if the backend could use registers in a less stupid^W^W more efficient manner. Homepage + source code ---------------------- I have put up a homepage for the benchmarks at: http://vax64.dk/ghc-bs-tests The raw measurements are in tarballs on that page. The source code for the benchmarks (+ support code) is in a mercurial repository at: http://vax64.dyndns.org/repo/hg/ghc-bs-tests I used scripts to install the various versions of ghc and bytestring both to avoid operator error and so you could me look over my shoulder. The scripts are in a mercurial repository at: http://vax64.dyndns.org/repo/hg/ghc-installations You can either follow the link and download any version you like as a tarball or you can (preferably) clone the repositories with: hg http://vax64.dyndns.org/repo/hg/ghc-bs-tests hg http://vax64.dyndns.org/repo/hg/ghc-installations All my code in those repositories is GPLv2. The text file 'text.txt' in the ghc-bs-tests repository is unfortunately partly in Danish and partly in very terse English. Acknowledgements ---------------- Daniel Fischer, for running the benchmarks on his SuSE 8.2 Athlon Duron 1200 MHz machine and for being helpful and patient while I made the scripts work on a 2.4 kernel and with unhelpful versions of GNU Make and strace. Erik van der Meer, for letting me run the benchmarks (and install ghc) on his Core Duo laptop. And for discussions on measurements over the years. Don Stewart, for playing along and for fixing a bytestring problem. And for contributing three benchmarks (one of which I had to change a bit before it would compile). Duncan Coutts, for playing along. Jules Bean (quicksilver), for contributing a benchmark. Bertram Felgenhauer (int-e), for contributing three benchmarks (in the form of a single file, which I untangled to three files). Spencer Jannsen (sjannsen), for contributing a benchmark. William Lee Irwin III (wli), for inspiring me to add the getwchar benchmarks. -Peter

1 0

readline problems building GHC on Mac OS X (was: Re: [Haskell-cafe] Re: ANNOUNCE: GHC version 6.8.2)
by Thorkil Naur 22 Dec '07

22 Dec '07

Hello, Although I have been building various GHC versions on various PPC Mac OS X systems for a while now, I'm afraid that I don't really have a good answer for your questions. However, your questions provide an excellect opportunity to discuss this, so that is what I am going to do. There are several questions here: (1) Which readline do we use? (2) Where do we store it? (3) What do we call it? (4) How do we make the Haskell readline library build process select the right one? And perhaps (5) How do we persuade the GHC build process to make the Haskell readline build that happens as part of building GHC select the right one? One at a time: 1. Which readline do we use? GNU readline, of course. As opposed to the readline installed as /usr/include/readline/*.h and /usr/lib/libreadline.dylib on our PPC Mac OS X machines which are said to be (and can even be observed to be) symbolic links to something called libedit and which, to me, never has managed to provide something suitable for use by GHC. But what is GNU readline, then? I don't exactly know, but my best guess is something like ftp://ftp.cwru.edu/pub/bash/readline-5.2.tar.gz. I never tried to install GNU readline directly from this file. On some occasions, I have installed readline from mac ports. Although I am fairly confident that what was installed was some version of the GNU readline, I am not sure. On other occasions, I have installed GNU readline from various sources related to GHC, some times known to me, at other times not. 2.Where do we store readline? I don't know where a readline based on the GNU download ftp://ftp.cwru.edu/pub/bash/readline-5.2.tar.gz would become installed (by default). The mac ports version installs by default at /opt/local/include/readline/*.h and /opt/local/lib/libreadline.*. Various readlines related to GHC have installed themselves (or were requested to become installed) as frameworks, this new and different Mac OS X mechanism for referring to a set of header files and corresponding library. So they have gone into /Library/Frameworks. 3. What do we call it? Here is where the interesting things start to happen: A central problem has been the ambiguity caused by Apple's decision to install symbolic links to the "edit" headers and "edit" library called "readline". And various mechanisms have been used to work around this problem: (a) If you have installed a mac ports readline at /opt/local/..., with GHC 6.6 at least, you were able to use the --with-readline-* options to direct GHC/the library build process to look in these directories first and thereby avoid the "edit" library; (b) At some point, a (possibly modified) version of the GNU readline library appeared, intended to be installed as a framework by the name of "GNUreadline" (as opposed to the bare "readline" name used earlier). This avoids the name clash caused by the Apple linking of "readline" to "edit". The problem that the Haskell readline library now needs to refer to a framework "GNUreadline" rather than ... (whatever it is that it refers to in a more Unix'y setting) is solvable. In addition, however, the readline library (or rather: The GNUreadline library derived from the readline library) refers to itself using the bare "readline" name, so that has to be changed also, leading to a need to maintain a complete and slightly modified version (GNUreadline) of the readline library. It seems to me that this situation is less than ideal. I mean, in theory, somebody may come along at some point with some library calling itself GNUreadline and then we would have to adapt, doing the whole thing all over again. This manner of avoiding the name clash problem does not seem tenable in the long run. Instead, what we should be able to do, is to specify, directly and to the point, that "readline", wherever we stored it, is what we want. That possibility does not exist, unfortunately, so we will have to make the best use that we can of the existing mechanisms, as far as we can figure out what they are, to get the desired effect. And if it turns out that the existing mechanisms do not allow us to do what we want, we need to request extensions and modifications of the mechanisms, until they are able to support our requirements. I am not quite sure that I am done with this subject, but let me go on with 4. How do we make the Haskell readline library build process select the right one? This is where I believe we can do something useful, making the Haskell readline library more capable in selecting its foundation readline library. I haven't worked out the details, some discussion is at http://hackage.haskell.org/trac/ghc/ticket/1395 and related tickets, but I am quite sure that methods can be found to select the desired readline library, without resorting to reissuing that library in a changed form and under a new name. And if this turns out to be absolutely impossible, I would much prefer pressing for the introduction of mechanisms that makes it possible to select the desired version of the library, removing this impossibility. Rather than issuing the library under a different name. Finally: 5. How do we persuade the GHC build process to make the Haskell readline build that happens as part of building GHC select the right one? Answer: I don't know. At some point, I did know, that was when the --with-readline-* options were introduced for the GHC ./configure. Nowadays, I am not sure. Generally, I believe that it is fine for the GHC build process (whatever phase) to pass parameters to the build process of some library. But at the same time, the fact that such passing of parameters takes place must be very explicitly reported somewhere, in the output of the build process, probably. Best regards Thorkil On Friday 21 December 2007 21:48, John Dorsey wrote: > (Moving to the cafe) > > On a related topic, I've been trying to build 6.8.2 on Leopard lately. > I've been running up against the infamous OS X readline issues. I know > some builders here have hacked past it, but I'm looking for a good > workaround... ideally one that works without changes outside the GHC > build area (besides installing a real readline). > > Here's what I noticed before I started drowning in the build platform. > (I'm no gnu-configure expert nor GHC insider.) > > I can get gnu-readline installed from Macports, no problem. > > The top-level configure in GHC doesn't respond to my various attempts: > > o using --with-readline-libraries and --with-readline-includes > (Although it looks like the libraries/readline/configure script > might recognize these, I can't get an option to pass through.) > o setting LDFLAGS and CPPFLAGS environment variables (with > -L/opt/local/lib and -I/opt/local/include resp.) in my shell > before running configure > o playing with the above settings and others in a mk/build.mk > > Until Apple fixes their broken-readline issue (maybe when the readline > compatibility of libedit improves)... maybe the top-level configure can > pass through flags or settings somehow? > > For those who've built with readline on OS X: have you had to resort to > blasting the existing readline library link, or is there a configuration > option within the GHC tree that you've gotten to work? > > Should I be filing a trac bug instead of asking here? > > Thanks for any help. There's no urgency for me; I'm just trying to get > a working environment at home; I'd prefer to be able to bootstrap from > the ground up; and I'd like to be able to contribute to testing/debugging > on OSX. > > John > > _______________________________________________ > Haskell-Cafe mailing list > Haskell-Cafe(a)haskell.org > http://www.haskell.org/mailman/listinfo/haskell-cafe >

2 1

Optimizing cellular automata & the beauty of unlifted types
by Justin Bailey 22 Dec '07

22 Dec '07

I'm back with another version of my cellular automata simulator. Since the last iteration I discovered GHC's unlifted types and the primitive operations that go with them. Using these types, rather than Ints, sped my code up by 2x or more. http://hpaste.org/4151#a2 -- first half of program http://hpaste.org/4151#a3 -- remaining (note 3 lines or so from first half are repeated) The key observation came from looking at the Core files, which showed a lot of int2word# and word2int# conversions going on. Figuring out how to remove those led me to the unlifted types. Coding with these types is not easy (for example, I couldn't see a way to write a Word# value directly - I had to write stuff like "int2Word# 1#"), but because I had an existing algorithm to guide me, combined with type checking, it was a fairly straightforward implementation. At first I felt kind of bad about using these operations, but then I noticed they are used pretty heavily by GHC itself. If it's good enough for GHC, it's good enough for me. The 2x performance gain didn't hurt either. Finally, the safety that comes from using the ST monad is just awesome. I've got low-level bit munging combined with high-level abstraction where I need it. So cool! I was disappointed to find early on that using higher-order functions in tight loops was a performance killer. It's unfortunate because I started with a very elegant, short implementation based on a simple Ring buffer and map. The current version is certainly harder to understand and has some weird limitations. However, having the simple implementation let me use quickcheck to compare their results on random rules and inputs, which gave me high confidence that my complex implemenation is correct. One thing I absolutely love about this program is its memory performance. It manages to stay within 1 - 10 MB of memory, depending on how much output is produced. How cool is that? On Dec 3, 2007 2:44 AM, Mirko Rahn <rahn(a)ira.uka.de > wrote: > It is interesting, that the naive implementation > ... is only 3 times slower than your quite complex, hard to follow and hard > to debug implementation. > Now the naive implementation is 100x slower, so I don't feel so bad about this comment any more. > > As always, I prefer to write most code in Haskell, quick, easy, nice, > reasonable fast, ... If speed matters, I switch to some lower level > language, as you did staying inside Haskell. > I have to take exception here - I *want* to write my code in Haskell. If Haskell isn't fast enough to beat the C implementation, I'd rather find a way to make my Haskell program faster than switch to some other language. Justin

4 5

eager/strict eval katas
by Thomas Hartman 22 Dec '07

22 Dec '07

I'm trying to get a better handle on eager/strict eval in haskell, and a great way to do this is by building up from simple exercises to harder exercises. So far I have exercise 1) add the integers [1..10^6] (stack overflows if you do a naive fold, as described on wiki) exercise 2) find the first integer such that average of [1..n] is > [10^6] (solution involves building an accum list of (average,listLength) tuples. again you can't do a naive fold due to stack overflow, but in this case even strict foldl' from data.list isn't "strict enough", I had to define my own custom fold to be strict on the tuples.) anybody got other suggestions, or links to places where eager eval is required to solve simply stated problems? or exercises that demystify doing eager IO/eager whatever monad, where that is required? Also am I correct that the terms eager and strict can be used more or less interchangeably in this problem space? Tired of this folk wisdom that haskell is only for the elite because getting around stack overflow from lazy eval is impossible to teach to newbies. t. --- This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden.

6 8

Re: ANNOUNCE: GHC version 6.8.2
by John Dorsey 22 Dec '07

22 Dec '07

(Moving to the cafe) On a related topic, I've been trying to build 6.8.2 on Leopard lately. I've been running up against the infamous OS X readline issues. I know some builders here have hacked past it, but I'm looking for a good workaround... ideally one that works without changes outside the GHC build area (besides installing a real readline). Here's what I noticed before I started drowning in the build platform. (I'm no gnu-configure expert nor GHC insider.) I can get gnu-readline installed from Macports, no problem. The top-level configure in GHC doesn't respond to my various attempts: o using --with-readline-libraries and --with-readline-includes (Although it looks like the libraries/readline/configure script might recognize these, I can't get an option to pass through.) o setting LDFLAGS and CPPFLAGS environment variables (with -L/opt/local/lib and -I/opt/local/include resp.) in my shell before running configure o playing with the above settings and others in a mk/build.mk Until Apple fixes their broken-readline issue (maybe when the readline compatibility of libedit improves)... maybe the top-level configure can pass through flags or settings somehow? For those who've built with readline on OS X: have you had to resort to blasting the existing readline library link, or is there a configuration option within the GHC tree that you've gotten to work? Should I be filing a trac bug instead of asking here? Thanks for any help. There's no urgency for me; I'm just trying to get a working environment at home; I'd prefer to be able to bootstrap from the ground up; and I'd like to be able to contribute to testing/debugging on OSX. John

1 0

functional maps
by Chad Scherrer 22 Dec '07

22 Dec '07

A while back I was playing with Data.Map was getting irritated about lookups that fail having the type of values, but wrapped in an extra monad. I decided to work around this by putting a default in the data type itself, so we have a "functional map" data FMap k a = FMap (k -> a) (Map k a) This has been really convenient - it's a monad, and failed lookups have the same type as successful ones. lookup :: (Ord k) => k -> FMap k a -> a lookup k (FMap f m)= Map.findWithDefault (f k) k m This also makes it much nicer to build a function that tabulates a list of pairs (nicer than I've found using Data.Map, anyway): fromPairs :: (Ord k) => [(k,a)] -> FMap k [a] fromPairs = foldl' (flip . uncurry $ insertWith (:)) $ return [] insertWith :: (Ord k) => (a -> b -> b) -> k -> a -> FMap k b -> FMap k b insertWith f k x m = case lookup k m of v -> insert k (f x v) m Ok, great, but fromPairs is blowing the stack. It does fine for a while, but today I was trying to use it for a few million pairs. It runs for a while, eats a couple gigs of ram, and then I get a stack overflow. Any advice for tracking this down? Thanks! Chad

1 1

How to make Prelude.read: no parse more verbose ...
by Georg Sauthoff 21 Dec '07

21 Dec '07

Hi, I try to debug some existing Haskell-Code. Out of the blue I get a 'progname: Prelude.read: no parse' error message from GHC. Great. Well, the code includes # grep '\<read\>' *| wc -l 23 (sic!) calls to the read fn. Well, how do I compile a Haskell program in such a way, that I get a useful error message from read? I mean, like the filename/linenumber of the calling expression for starters. Really crazy would be the possibility to print out a backtrace, if some (library) function calls error ... Best regards Georg -- Fortune : 'Real programmers don't comment their code. It was hard to write, it should be hard to understand.' ;)

6 6