1 <Sect1 id="options-debugging">
2 <Title>Debugging the compiler
6 <IndexTerm><Primary>debugging options (for GHC)</Primary></IndexTerm>
10 HACKER TERRITORY. HACKER TERRITORY.
14 <Sect2 id="replacing-phases">
15 <Title>Replacing the program for one or more phases.
19 <IndexTerm><Primary>GHC phases, changing</Primary></IndexTerm>
20 <IndexTerm><Primary>phases, changing GHC</Primary></IndexTerm>
21 You may specify that a different program be used for one of the phases
22 of the compilation system, in place of whatever the driver <Command>ghc</Command> has
23 wired into it. For example, you might want to try a different
25 <Option>-pgm<phase-code><program-name></Option><IndexTerm><Primary>-pgm<phase><stuff>
26 option</Primary></IndexTerm> option to <Command>ghc</Command> will cause it to use <Literal><program-name></Literal>
27 for phase <Literal><phase-code></Literal>, where the codes to indicate the phases are:
33 <ColSpec Align="Left" Colsep="0">
34 <ColSpec Align="Left" Colsep="0">
37 <Entry><Emphasis>code</Emphasis> </Entry>
38 <Entry><Emphasis>phase</Emphasis> </Entry>
44 <Entry> literate pre-processor </Entry>
49 <Entry> C pre-processor (if -cpp only) </Entry>
54 <Entry> Haskell compiler </Entry>
59 <Entry> C compiler</Entry>
64 <Entry> assembler </Entry>
69 <Entry> linker </Entry>
74 <Entry> Makefile dependency generator </Entry>
85 <Sect2 id="forcing-options-through">
86 <Title>Forcing options to a particular phase.
90 <IndexTerm><Primary>forcing GHC-phase options</Primary></IndexTerm>
94 The preceding sections describe driver options that are mostly
95 applicable to one particular phase. You may also <Emphasis>force</Emphasis> a
96 specific option <Option><option></Option> to be passed to a particular phase
97 <Literal><phase-code></Literal> by feeding the driver the option
98 <Option>-opt<phase-code><option></Option>.<IndexTerm><Primary>-opt<phase><stuff>
99 option</Primary></IndexTerm> The codes to indicate the phases are the same as in the
104 So, for example, to force an <Option>-Ewurble</Option> option to the assembler, you
105 would tell the driver <Option>-opta-Ewurble</Option> (the dash before the E is
110 Besides getting options to the Haskell compiler with <Option>-optC<blah></Option>,
111 you can get options through to its runtime system with
112 <Option>-optCrts<blah></Option><IndexTerm><Primary>-optCrts<blah> option</Primary></IndexTerm>.
116 So, for example: when I want to use my normal driver but with my
117 profiled compiler binary, I use this script:
121 exec /local/grasp_tmp3/simonpj/ghc-BUILDS/working-alpha/ghc/driver/ghc \
122 -pgmC/local/grasp_tmp3/simonpj/ghc-BUILDS/working-hsc-prof/hsc \
132 <Sect2 id="dumping-output">
133 <Title>Dumping out compiler intermediate structures
137 <IndexTerm><Primary>dumping GHC intermediates</Primary></IndexTerm>
138 <IndexTerm><Primary>intermediate passes, output</Primary></IndexTerm>
145 <Term><Option>-noC</Option>:</Term>
148 <IndexTerm><Primary>-noC option</Primary></IndexTerm>
149 Don't bother generating C output <Emphasis>or</Emphasis> an interface file. Usually
150 used in conjunction with one or more of the <Option>-ddump-*</Option> options; for
151 example: <Command>ghc -noC -ddump-simpl Foo.hs</Command>
156 <Term><Option>-hi</Option>:</Term>
159 <IndexTerm><Primary>-hi option</Primary></IndexTerm>
160 <Emphasis>Do</Emphasis> generate an interface file. This would normally be used in
161 conjunction with <Option>-noC</Option>, which turns off interface generation;
162 thus: <Option>-noC -hi</Option>.
167 <Term><Option>-dshow-passes</Option>:</Term>
170 <IndexTerm><Primary>-dshow-passes option</Primary></IndexTerm>
171 Prints a message to stderr as each pass starts. Gives a warm but
172 undoubtedly misleading feeling that GHC is telling you what's
178 <Term><Option>-ddump-<pass></Option>:</Term>
181 <IndexTerm><Primary>-ddump-<pass> options</Primary></IndexTerm>
182 Make a debugging dump after pass <Literal><pass></Literal> (may be common enough to
183 need a short form…). You can get all of these at once (<Emphasis>lots</Emphasis> of
184 output) by using <Option>-ddump-all</Option>, or most of them with <Option>-ddump-most</Option>.
185 Some of the most useful ones are:
192 <Term><Option>-ddump-parsed</Option>:</Term>
201 <Term><Option>-ddump-rn</Option>:</Term>
210 <Term><Option>-ddump-minimal-imports</Option>:</Term>
213 Dump to the file "M.imports" (where M is the module being compiled)
214 a "minimal" set of import declarations. You can safely replace
215 all the import declarations in "M.hs" with those found in "M.imports".
216 Why would you want to do that? Because the "minimal" imports (a) import
217 everything explicitly, by name, and (b) import nothing that is not required.
218 It can be quite painful to maintain this property by hand, so this flag is
219 intended to reduce the labour.
225 <Term><Option>-ddump-hi-diffs</Option>:</Term>
228 Dump to stdout a summary of the differences between the existing interface file (if any)
229 for this module, and the new one.
235 <Term><Option>-ddump-tc</Option>:</Term>
244 <Term><Option>-ddump-types</Option>:</Term>
247 Dump a type signature for each value defined at the top level
248 of the module. The list is sorted alphabetically.
249 Using <Option>-dppr-debug</Option> dumps a type signature for
250 all the imported and system-defined things as well; useful
251 for debugging the compiler.
257 <Term><Option>-ddump-deriv</Option>:</Term>
265 <Term><Option>-ddump-ds</Option>:</Term>
273 <Term><Option>-ddump-spec</Option>:</Term>
276 output of specialisation pass
281 <Term><Option>-ddump-rules</Option>:</Term>
284 dumps all rewrite rules (including those generated by the specialisation pass)
289 <Term><Option>-ddump-simpl</Option>:</Term>
292 simplifer output (Core-to-Core passes)
297 <Term><Option>-ddump-usagesp</Option>:</Term>
300 UsageSP inference pre-inf and output
305 <Term><Option>-ddump-cpranal</Option>:</Term>
313 <Term><Option>-ddump-stranal</Option>:</Term>
316 strictness analyser output
321 <Term><Option>-ddump-workwrap</Option>:</Term>
324 worker/wrapper split output
329 <Term><Option>-ddump-occur-anal</Option>:</Term>
332 `occurrence analysis' output
337 <Term><Option>-ddump-stg</Option>:</Term>
340 output of STG-to-STG passes
345 <Term><Option>-ddump-absC</Option>:</Term>
348 <Emphasis>un</Emphasis>flattened Abstract C
353 <Term><Option>-ddump-flatC</Option>:</Term>
356 <Emphasis>flattened</Emphasis> Abstract C
361 <Term><Option>-ddump-realC</Option>:</Term>
364 same as what goes to the C compiler
369 <Term><Option>-ddump-asm</Option>:</Term>
372 assembly language from the native-code generator
377 <IndexTerm><Primary>-ddump-all option</Primary></IndexTerm>
378 <IndexTerm><Primary>-ddump-most option</Primary></IndexTerm>
379 <IndexTerm><Primary>-ddump-parsed option</Primary></IndexTerm>
380 <IndexTerm><Primary>-ddump-rn option</Primary></IndexTerm>
381 <IndexTerm><Primary>-ddump-tc option</Primary></IndexTerm>
382 <IndexTerm><Primary>-ddump-deriv option</Primary></IndexTerm>
383 <IndexTerm><Primary>-ddump-ds option</Primary></IndexTerm>
384 <IndexTerm><Primary>-ddump-simpl option</Primary></IndexTerm>
385 <IndexTerm><Primary>-ddump-cpranal option</Primary></IndexTerm>
386 <IndexTerm><Primary>-ddump-workwrap option</Primary></IndexTerm>
387 <IndexTerm><Primary>-ddump-rules option</Primary></IndexTerm>
388 <IndexTerm><Primary>-ddump-usagesp option</Primary></IndexTerm>
389 <IndexTerm><Primary>-ddump-stranal option</Primary></IndexTerm>
390 <IndexTerm><Primary>-ddump-occur-anal option</Primary></IndexTerm>
391 <IndexTerm><Primary>-ddump-spec option</Primary></IndexTerm>
392 <IndexTerm><Primary>-ddump-stg option</Primary></IndexTerm>
393 <IndexTerm><Primary>-ddump-absC option</Primary></IndexTerm>
394 <IndexTerm><Primary>-ddump-flatC option</Primary></IndexTerm>
395 <IndexTerm><Primary>-ddump-realC option</Primary></IndexTerm>
396 <IndexTerm><Primary>-ddump-asm option</Primary></IndexTerm>
401 <Term><Option>-dverbose-simpl</Option> and <Option>-dverbose-stg</Option>:</Term>
404 <IndexTerm><Primary>-dverbose-simpl option</Primary></IndexTerm>
405 <IndexTerm><Primary>-dverbose-stg option</Primary></IndexTerm>
406 Show the output of the intermediate Core-to-Core and STG-to-STG
407 passes, respectively. (<Emphasis>Lots</Emphasis> of output!) So: when we're
411 % ghc -noC -O -ddump-simpl -dverbose-simpl -dcore-lint Foo.hs
418 <Term><Option>-ddump-simpl-iterations</Option>:</Term>
421 <IndexTerm><Primary>-ddump-simpl-iterations option</Primary></IndexTerm>
422 Show the output of each <Emphasis>iteration</Emphasis> of the simplifier (each run of
423 the simplifier has a maximum number of iterations, normally 4). Used
424 when even <Option>-dverbose-simpl</Option> doesn't cut it.
429 <Term><Option>-dppr-{user,debug</Option>}:</Term>
432 <IndexTerm><Primary>-dppr-user option</Primary></IndexTerm>
433 <IndexTerm><Primary>-dppr-debug option</Primary></IndexTerm>
434 Debugging output is in one of several “styles.” Take the printing
435 of types, for example. In the “user” style, the compiler's internal
436 ideas about types are presented in Haskell source-level syntax,
437 insofar as possible. In the “debug” style (which is the default for
438 debugging output), the types are printed in with
439 explicit foralls, and variables have their unique-id attached (so you
440 can check for things that look the same but aren't).
445 <Term><Option>-ddump-simpl-stats</Option>:</Term>
448 <IndexTerm><Primary>-ddump-simpl-stats option</Primary></IndexTerm>
449 Dump statistics about how many of each kind
450 of transformation too place. If you add <Option>-dppr-debug</Option> you get more detailed information.
455 <Term><Option>-ddump-raw-asm</Option>:</Term>
458 <IndexTerm><Primary>-ddump-raw-asm option</Primary></IndexTerm>
459 Dump out the assembly-language stuff, before the “mangler” gets it.
464 <Term><Option>-ddump-rn-trace</Option>:</Term>
467 <IndexTerm><Primary>-ddump-rn-trace</Primary></IndexTerm>
468 Make the renamer be *real* chatty about what it is upto.
473 <Term><Option>-dshow-rn-stats</Option>:</Term>
476 <IndexTerm><Primary>-dshow-rn-stats</Primary></IndexTerm>
477 Print out summary of what kind of information the renamer had to bring
483 <Term><Option>-dshow-unused-imports</Option>:</Term>
486 <IndexTerm><Primary>-dshow-unused-imports</Primary></IndexTerm>
487 Have the renamer report what imports does not contribute.
496 <Sect2 id="checking-consistency">
497 <Title>Checking for consistency
501 <IndexTerm><Primary>consistency checks</Primary></IndexTerm>
502 <IndexTerm><Primary>lint</Primary></IndexTerm>
509 <Term><Option>-dcore-lint</Option>:</Term>
512 <IndexTerm><Primary>-dcore-lint option</Primary></IndexTerm>
513 Turn on heavyweight intra-pass sanity-checking within GHC, at Core
514 level. (It checks GHC's sanity, not yours.)
519 <Term><Option>-dstg-lint</Option>:</Term>
522 <IndexTerm><Primary>-dstg-lint option</Primary></IndexTerm>
528 <Term><Option>-dusagesp-lint</Option>:</Term>
531 <IndexTerm><Primary>-dstg-lint option</Primary></IndexTerm>
532 Turn on checks around UsageSP inference (<Option>-fusagesp</Option>). This verifies
533 various simple properties of the results of the inference, and also
534 warns if any identifier with a used-once annotation before the
535 inference has a used-many annotation afterwards; this could indicate a
536 non-worksafe transformation is being applied.
546 <Title>How to read Core syntax (from some <Option>-ddump-*</Option> flags)</Title>
549 <IndexTerm><Primary>reading Core syntax</Primary></IndexTerm>
550 <IndexTerm><Primary>Core syntax, how to read</Primary></IndexTerm>
554 Let's do this by commenting an example. It's from doing
555 <Option>-ddump-ds</Option> on this code:
558 skip2 m = m : skip2 (m+2)
564 Before we jump in, a word about names of things. Within GHC,
565 variables, type constructors, etc., are identified by their
566 “Uniques.” These are of the form `letter' plus `number' (both
567 loosely interpreted). The `letter' gives some idea of where the
568 Unique came from; e.g., <Literal>_</Literal> means “built-in type variable”;
569 <Literal>t</Literal> means “from the typechecker”; <Literal>s</Literal> means “from the
570 simplifier”; and so on. The `number' is printed fairly compactly in
571 a `base-62' format, which everyone hates except me (WDP).
575 Remember, everything has a “Unique” and it is usually printed out
576 when debugging, in some form or another. So here we go…
582 Main.skip2{-r1L6-} :: _forall_ a$_4 =>{{Num a$_4}} -> a$_4 -> [a$_4]
584 --# `r1L6' is the Unique for Main.skip2;
585 --# `_4' is the Unique for the type-variable (template) `a'
586 --# `{{Num a$_4}}' is a dictionary argument
590 --# `_NI_' means "no (pragmatic) information" yet; it will later
591 --# evolve into the GHC_PRAGMA info that goes into interface files.
594 /\ _4 -> \ d.Num.t4Gt ->
597 +.t4Hg :: _4 -> _4 -> _4
599 +.t4Hg = (+{-r3JH-} _4) d.Num.t4Gt
601 fromInt.t4GS :: Int{-2i-} -> _4
603 fromInt.t4GS = (fromInt{-r3JX-} _4) d.Num.t4Gt
605 --# The `+' class method (Unique: r3JH) selects the addition code
606 --# from a `Num' dictionary (now an explicit lamba'd argument).
607 --# Because Core is 2nd-order lambda-calculus, type applications
608 --# and lambdas (/\) are explicit. So `+' is first applied to a
609 --# type (`_4'), then to a dictionary, yielding the actual addition
610 --# function that we will use subsequently...
612 --# We play the exact same game with the (non-standard) class method
613 --# `fromInt'. Unsurprisingly, the type `Int' is wired into the
623 } in fromInt.t4GS ds.d4Qz
625 --# `I# 2#' is just the literal Int `2'; it reflects the fact that
626 --# GHC defines `data Int = I# Int#', where Int# is the primitive
627 --# unboxed type. (see relevant info about unboxed types elsewhere...)
629 --# The `!' after `I#' indicates that this is a *saturated*
630 --# application of the `I#' data constructor (i.e., not partially
633 skip2.t3Ja :: _4 -> [_4]
637 let { ds.d4QQ :: [_4]
643 ds.d4QY = +.t4Hg m.r1H4 lit.t4Hb
644 } in skip2.t3Ja ds.d4QY
654 (“It's just a simple functional language” is an unregisterised
655 trademark of Peyton Jones Enterprises, plc.)
660 <Sect2 id="source-file-options">
661 <Title>Command line options in source files
665 <IndexTerm><Primary>source-file options</Primary></IndexTerm>
669 Sometimes it is useful to make the connection between a source file
670 and the command-line options it requires quite tight. For instance,
671 if a (Glasgow) Haskell source file uses <Literal>casm</Literal>s, the C back-end
672 often needs to be told about which header files to include. Rather than
673 maintaining the list of files the source depends on in a
674 <Filename>Makefile</Filename> (using the <Option>-#include</Option> command-line option), it is
675 possible to do this directly in the source file using the <Literal>OPTIONS</Literal>
676 pragma <IndexTerm><Primary>OPTIONS pragma</Primary></IndexTerm>:
681 {-# OPTIONS -#include "foo.h" #-}
689 <Literal>OPTIONS</Literal> pragmas are only looked for at the top of your source
690 files, upto the first (non-literate,non-empty) line not containing
691 <Literal>OPTIONS</Literal>. Multiple <Literal>OPTIONS</Literal> pragmas are recognised. Note
692 that your command shell does not get to the source file options, they
693 are just included literally in the array of command-line arguments
694 the compiler driver maintains internally, so you'll be desperately
695 disappointed if you try to glob etc. inside <Literal>OPTIONS</Literal>.
699 NOTE: the contents of OPTIONS are prepended to the command-line
700 options, so you *do* have the ability to override OPTIONS settings
701 via the command line.
705 It is not recommended to move all the contents of your Makefiles into
706 your source files, but in some circumstances, the <Literal>OPTIONS</Literal> pragma
707 is the Right Thing. (If you use <Option>-keep-hc-file-too</Option> and have OPTION
708 flags in your module, the OPTIONS will get put into the generated .hc
715 <title>Unregisterised compilation</title>
716 <indexterm><primary>unregisterised compilation</primary></indexterm>
718 <para>The term "unregisterised" really means "compile via vanilla
719 C", disabling some of the platform-specific tricks that GHC
720 normally uses to make programs go faster. When compiling
721 unregisterised, GHC simply generates a C file which is compiled
724 <para>Unregisterised compilation can be useful when porting GHC to
725 a new machine, since it reduces the prerequisite tools to
726 <command>gcc</command>, <command>as</command>, and
727 <command>ld</command> and nothing more, and furthermore the amount
728 of platform-specific code that needs to be written in order to get
729 unregisterised compilation going is usually fairly small.</para>
733 <term><option>-unreg</option>:</term>
734 <indexterm><primary><option>-unreg</option></primary></indexterm>
736 <para>Compile via vanilla ANSI C only, turning off
737 platform-specific optimisations. NOTE: in order to use
738 <option>-unreg</option>, you need to have a set of libraries
739 (including the RTS) built for unregisterised compilation.
740 This amounts to building GHC with way "u" enabled.</para>
749 ;;; Local Variables: ***
751 ;;; sgml-parent-document: ("users_guide.sgml" "book" "chapter" "sect1") ***