faq.txt 34 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605
  1. 1. [1]What is Free Pascal (FPC)?
  2. 2. [2]Which versions exist, and which one should I use?
  3. 3. [3]Free Pascal and GNU Pascal - a comparison
  4. 4. [4]Where can I get the compiler ?
  5. 5. [5]What are the considerations in porting
  6. 6. [6]I tried to compile my Delphi code with the Free Pascal
  7. 7. [7]I have to write a program for homework. Can you help?
  8. 8. [8]How can I build a unit?
  9. 9. [9]Will Free Pascal support TV (Turbo Vision) in the future?
  10. 10. [10]How can I compile the system unit?
  11. 11. [11]I get an internal error 9999 or 10?
  12. 12. [12]How does function overloading work?
  13. 13. [13]How can I call C functions?
  14. 14. [14]How can I use the graph unit with Free Pascal?
  15. 15. [15]Why do I get wrong colors when using the graph unit?
  16. 16. [16]Integrated Assembler syntax
  17. 17. [17]How can I access DOS memory / How can I do graphics
  18. programming?
  19. 18. [18]How can I run Free Pascal without a math coprocessor?
  20. 19. [19]How do I reserve more than 2 megabytes of RAM?
  21. 20. [20]How can I access I/O ports?
  22. 21. [21]I'm using the Dos compiler under Windows 95
  23. 22. [22]I'm using OS/2
  24. 23. [23]INSTALL.EXE of Dos version 0.99.10 reports "Load error: no
  25. DPMI"
  26. 24. [24]INSTALL.EXE of version 1.0 for Dos returns an error (-2) in
  27. Windows NT 4.0
  28. 25. [25]I want a new version NOW
  29. 26. [26]Where can I find a text mode IDE?
  30. 27. [27]How do I configure the Dos IDE?
  31. 28. [28]Why are the generated binaries so big?
  32. 29. [29]Unit system, syslinux, sysos2 or syswin32 not found errors
  33. 30. [30]Known bugs
  34. 31. [31]How can I find where an error occurred using the addresses a
  35. crashed program prints?
  36. 1. What is Free Pascal (FPC)?
  37. Originally named FPK-Pascal, the Free Pascal compiler is a 32 bit
  38. Turbo Pascal compatible Pascal compiler for DOS, Linux, Win32,
  39. OS/2 and (based on an older version) the AmigaOS. More operating
  40. systems (BeOS and FreeBSD/ELF are in advanced stages of
  41. development) are in the works.
  42. The compiler is written in Pascal and is able to compile its own
  43. sources. The source files are included.
  44. Free Pascal is modest regarding its minimal system requirements
  45. (386-25 Mhz for the Intel version and ideally a 68020 processor
  46. for the Motorola version). At least 2 megabytes of RAM are
  47. required. To remake the compiler more than 16MB is recommended.
  48. Short history:
  49. + 6/1993: project start
  50. + 10/1993: first little programs work
  51. + 3/1995: the compiler compiles the own sources
  52. + 3/1996: released to the internet
  53. + 7/2000: 1.0 version
  54. 2. Which versions exist, and which one should I use?
  55. FPC's version numbering changed a few times over the years.
  56. Versions before 0.99.5 are considered archaic. After the release
  57. of 0.99.5 a system in version numbering was introduced, and that
  58. system was changed slightly changed after the 1.0 release.
  59. Versioning for versions 0.99.5 - 1.0
  60. Compilers with an even last number are release versions(e.g.
  61. 0.99.8, 0.99.10, 0.99.12, 0.99.14 1.0.0)
  62. Compilers and packages with an odd last number are development
  63. versions (e.g. 0.99.9, 0.99.11, 0.99.13, 0.99.15)
  64. 0.99.5 is an exception to this rule, since 0.99.5 IS a release (a
  65. release prior to the introduction of this odd/even system).
  66. Letters behind the version number (0.99.12b, 0.99.5d) indicate
  67. release versions with some bugs and problems in the original
  68. release (respectively 0.99.12 and 0.99.5) fixed.
  69. Versioning after 1.0
  70. Together with the release of 1.0 the version numbering has been
  71. slightly changed, and a system in versioning resembling the Linux
  72. kernel's has been introduced. The main difference is that the
  73. difference between a release version is now in the second number
  74. (1.0.x vs 1.1.x) instead of the third number (0.99.14 vs 0.99.15),
  75. and the third number now becomes the patch level, replacing the
  76. postfixed letter in the old system.
  77. + Releases that only fix bugs in version 1.0 will be numbered
  78. 1.0.x
  79. + New development (the so called snapshots) have version number
  80. 1.1.x. The meaning of the third version number x in the new
  81. development branch is not defined yet, it could be used for
  82. test releases or to signal major changes.
  83. + Eventually the 1.1.x versions, when stabilized will be
  84. released as version 1.2. Fixes on the 1.2 release will be
  85. numbered 1.2.x
  86. + The new development after the 1.2 release will be numbered
  87. 1.3.x and so on
  88. + When really big changes are implemented, the version will be
  89. updated in the major number. This could be case with e.g. a
  90. codegenerator rewrite with support for other processors
  91. Normally you would want to use a release. Releases are considered
  92. stable, and easier to support (the bugs, quirks and unintended
  93. "features" are well known after a period of time, and workarounds
  94. exist).
  95. Development snapshots (which are generated daily) reflect the
  96. current status of the compiler. Development versions probably have
  97. new features and larger bugs fixed since the last release, but
  98. might have some temporary stability drawbacks (which are usually
  99. fixed by the next day).
  100. Most support for development snapshots are basically the advise to
  101. upgrade to newer snapshot in which the bugs are hopefully fixed.
  102. Since version 0.99.8 the stability of the compiler steadily
  103. increased and development snapshots are often quite useful for
  104. certain categories of users. Ask in the maillists if it is worth
  105. the trouble in your case if you're not sure.
  106. The current release version is 1.00 for the OS/2, Linux, Windows
  107. and Dos (Go32V2) targets and 0.99.5d for the 680x0 based systems
  108. (Amiga and Atari ST). The development versions (snapshots) are
  109. numbered 1.1.x at the moment
  110. We advise all users to upgrade to the newest version for their
  111. target. (1.0 for intel processors, and 0.99.5d for Motorola)
  112. 3. Free Pascal and GNU Pascal - a comparison
  113. Aim:
  114. Free Pascal tries to implement a Borland compatible
  115. pascal compiler on as many platforms as possible. GNU
  116. Pascal tries to implement a portable pascal compiler
  117. based on POSIX.
  118. Version:
  119. Currently, Free Pascal is at version 1.00 for the Intel
  120. version and version 0.99.5d for the Motorola/Intel
  121. version. Version 0.99.5d differs from version 0.99.5 in
  122. that all run time library fixes have been applied, as
  123. well as all known code generation bugs. Version 1.00
  124. differs from version 0.99.5d in that all parser bugfixes
  125. have also been applied and also a lot of Delphi 2 and
  126. Delphi 3 extensions have been implemented. GNU Pascal is
  127. at version 2.8.1 (but this numbering is not really an
  128. indication, it follows the GNU C numbering, since it is a
  129. derivation of it)
  130. Operating systems:
  131. Free pascal runs on a limited number of systems : DOS,
  132. Win32, Linux, OS/2 and AmigaOS and is for the moment
  133. limited to the Intel and Motorola architectures. GNU
  134. Pascal runs basically on any system that can run GNU C.
  135. Sources:
  136. Free Pascal is entirely written in Pascal (about 6 Mb of
  137. source code), while GNU Pascal is written in C (it's an
  138. adaptation of the GNU C compiler: 2.8 Mb code + 8 MB of
  139. GNU C code)
  140. Language:
  141. Free Pascal supports the Borland Pascal dialect Borland,
  142. and implements the Delphi Object Pascal language. GNU
  143. Pascal supports ISO 7185, ISO 10206, (most of) Borland
  144. Pascal 7.0
  145. Extensions:
  146. Free Pascal implements method, function and operator
  147. overloading. GNU Pascal implements operator overloading.
  148. License:
  149. Both compilers come under the GNU GPL.
  150. Author:
  151. Free Pascal was started by Florian Klaempfl, Germany
  152. ([email protected]), GNU Pascal was started by
  153. Jukka Virtanen, Finland ([email protected]).
  154. 4. Where can I get the compiler ?
  155. Free Pascal is available for download from all [32]official
  156. mirrors
  157. 5. What are the considerations in porting code to other processors?
  158. Because the compiler now supports processors other than the Intel,
  159. it is important to take a few precautions so that your code will
  160. execute correctly on all processors.
  161. + Limit your use of asm statements unless it is time critical
  162. code
  163. + Don't use the packed directive unless you know exactly what
  164. you are doing. Most processors require alignment of data, and
  165. using packed on objects,classes and records may break this
  166. requirement. If this is the case your code will simply crash
  167. on the target processors.
  168. + Clean up at the end of your program, i.e. close all files on
  169. exit, as some operating systems don't like it when some files
  170. are left opened.
  171. + Try not to rely on the endian of the specific machines when
  172. doing arithmetic operations. Furthermore, reading and writing
  173. of binary data to/from files will probably require byte swaps
  174. across different endian machines (swap is your friend in this
  175. case). This is even more important if you write binary data
  176. to files.
  177. + Try limiting your local variables in subroutines to 32K, as
  178. this is the limit of some processors, use dynamic allocation
  179. instead.
  180. + Try limiting the size of parameters passed to subroutines to
  181. 32K, as this is the limit of some processors, use const or
  182. var parameters instead.
  183. 6. I tried to compile my Delphi code with the Free Pascal Compiler,
  184. but it seems that it doesn't recognize Delphi style OOP.
  185. The compiler supports the Delphi OOP. Make sure you use the -S2 or
  186. -Sd switches (see the manuals for the meaning of these switches).
  187. For a list of Delphi incompabilities also check the manual.
  188. 7. I have to write a program for homework. Can you help?
  189. No. Please, don't send us mail about homework, we are no teachers.
  190. The Free Pascal development team tries to give good support for
  191. the Free Pascal compiler and are trying to always reply to emails.
  192. If we get emails like this, this becomes harder and harder.
  193. 8. How can I build a unit?
  194. It works like in Turbo Pascal. The first keyword in the file must
  195. be UNIT (not case sensitive). The compiler will generate two
  196. files: XXX.PPU and XXX.O. The PPU file contains the interface
  197. information for the compiler and the O-file the machine code (an
  198. object file, whose precise structure depends on the assembler you
  199. used). To use this unit in another unit or program, you must
  200. include its name in the USES clause of your program.
  201. 9. Will Free Pascal support TV (Turbo Vision) in the future?
  202. A Turbo Vision port, called Free Vision, has progressed nicely
  203. lately. It's already very usable, we are even writing an IDE in
  204. it. Due to copyrights problem the FreeVision source code is not
  205. available at the moment. You can download the IDE from the
  206. [33]development page. and get an idea of the look and feel though.
  207. 10. How can I compile the system unit?
  208. To recompile the system unit, it is recommended to have GNU make
  209. installed. typing 'make' in the rtl source directory will then
  210. recompile all RTL units including the system unit. You may choose
  211. to descend into the directory of your OS (e.g. rtl/go32v2) and do
  212. a 'make' there.
  213. It is possible to do all this manually, but you need more detailed
  214. knowledge of the RTL tree structure for that.
  215. 11. I get an internal error 9999 or 10?
  216. The latest versions of the Free Pascal Compiler come with an error
  217. handling routine which catches the segmentation fault and lets the
  218. compiler to exit gracefully. This is reported as an internal error
  219. 9999. Please try to reproduce the error and send [34]us a bug
  220. report.
  221. (For the curious, IE 9999 is not a specific bug. It is a safety
  222. measure which terminates if during compiling a certain condition
  223. is not met, which can be caused by several bugs. So if you report
  224. the bug, and get IE 9999 later in a different piece or part of
  225. sourcecode, it could be a completely different bug. IE 10 is
  226. something similar. It is a safety measure that is triggered when
  227. the estimated number of registers needed to evaluate an expression
  228. proves wrong. Just like IE 9999, two IE 10 problems are often
  229. independant of eachother.)
  230. 12. How does function overloading work?
  231. function overloading is implemented, like in C++:
  232. procedure a(i : integer);
  233. begin
  234. end;
  235. procedure a(s : string);
  236. begin
  237. end;
  238. begin
  239. a('asdfdasf');
  240. a(1234);
  241. end.
  242. You must be careful. If one of your overloaded functions is in the
  243. interface part of your unit, then all overloaded functions must be
  244. in the interface part. If you leave one out, the compiler will
  245. complain with a 'This overloaded function can't be local' message.
  246. Overloaded functions must differ in their parameters, it's not
  247. enough if their return types are different.
  248. 13. How can I call C functions?
  249. C calling convention is implemented as follows: The compiler
  250. pushes the parameters from right to left, but the procedure has to
  251. clear the stack. For calling the C function strcmp declare the
  252. following:
  253. function strcmp(s1 : pchar;s2 : pchar) : integer;cdecl;external;
  254. Since 0.99.5, the older [C]; won't work!
  255. 14. How can I use the graph unit with Free Pascal?
  256. Since 0.99.12, the graph unit is available both for Dos and Linux.
  257. Under Dos, it only supported VESA modes though. Since version
  258. 0.99.14, a new more system independant graph unit is included
  259. (although the only extra supported OS is Win32 and this is only
  260. rudimentary support) which also supports standard VGA.
  261. Since version 1.0, we also have a completely platform independent
  262. way of selecting resolutions and bitdepths. You are strongly
  263. encouraged to use it, because other ways will probably fail on one
  264. or other platform. See the documentation of the graph unit for
  265. more information.
  266. 15. Why do I get wrong colors when using the graph unit?
  267. If you use detect as graphdriver, you will end up with the highest
  268. supported bitdepth. Since the graph unit currently only supports
  269. up to 16 bits per pixel modes and since this bitdepth is supported
  270. by all graphics cards made in at least the last 5 years, you will
  271. most likely get a 16 bit mode.
  272. The main problem is that in 16 (and 15, 24, 32, ...) bit modes,
  273. the colors aren't set anymore using an index in a palette (the
  274. palettized way is called "indexed color"). In these modes, the
  275. color number itself determines what color you get on screen and
  276. you can't change this color. The color is encoded as follows (for
  277. most graphics cards on PC's at least):
  278. + 15 bit color: lower 5 bits are blue intensity, next come 5
  279. bits of green and then 5 bits of red. The highest bit of the
  280. word is ignored.
  281. + 16 bit color: lower 5 bits are blue intensite, next come *6*
  282. bits of green and then 5 bits of red.
  283. This means that either you have to rewrite your program so it can
  284. work with this so-called "direct color" scheme, or that you have
  285. to use D8BIT as graphdriver and DetectMode as graphmode. This will
  286. ensure that you end up with a 256 (indexed) color mode. If there
  287. are no 256 color modes supported, then graphresult will contain
  288. the value GrNotDetected after you called InitGraph and you can
  289. retry with graphdriver D4BIT. Make sure you use the constant names
  290. (D8BIT, D4BIT, ...) and not their actual numeric values, because
  291. those values can change with the next release! That the very
  292. reason why such symbolic constants exist.
  293. 16. Integrated Assembler syntax
  294. The default assembler syntax (AT&T style) is different from the
  295. one in Borland Pascal (Intel style).
  296. However, as of version 0.99.0, the compiler supports Intel style
  297. assembly syntax. See the documentation for more info on how to use
  298. different assembler styles.
  299. A description of the AT&T syntax can be found in the DJGPP FAQ
  300. [35]http://www.delorie.com/djgpp/v2faq/faq102.html#Syntax or in
  301. Brennan's Guide to Inline Assembly
  302. [36]http://www.rt66.com/%7Ebrennan/djgpp/djgpp asm.html. The
  303. documentation also contains a chapter where the difference between
  304. the Intel and AT&T style assembly is explained.
  305. Or you can use the convertor program at
  306. [37]http://rcs.urz.tu-dresden.de/schoenfu/zip/asmtrans.zip .
  307. 17. How can I access DOS memory / How can I do graphics programming?
  308. You can do like in TP, via absolute or mem[]. For larger memory
  309. blocks use the dosmemput/dosmemget routines in Go32 unit.
  310. 18. How can I run Free Pascal without a math coprocessor?
  311. On the Intel version the emulator is automatically loaded by the
  312. compiler if you add the following commands to your autoexec.bat:
  313. SET 387=N
  314. SET EMU386=C:\PP\BIN\GO32V2\WEMU387.DXE
  315. (don't forget to replace the C:\PP with the directory where you
  316. installed FPC)
  317. 19. How do I reserve more than 2 megabytes of RAM?
  318. By default Free Pascal allocates only 2MB of RAM for your
  319. application. If it just allocated all it could get, people running
  320. Windows would have problems as Windows would increase the swap
  321. file size to give the program more memory on and on, until the
  322. swap file drive would be full.
  323. You can specify the size of the heap with -Chxxxx. The default
  324. value is -Ch4000000. Try -Ch10000000, provided you got enough swap
  325. space.
  326. However, the heap size doesn't really matter anymore, since the
  327. Heap is able to grow: if you've used all the available heap space,
  328. the program will try to get more memory from the OS, so the heap
  329. is limited to the maximum amount of free memory provided by the
  330. OS.
  331. It is only handy if you know you will need at least a certain
  332. amount of memory. You can then specify this value using the -Ch
  333. parameter, so your program will allocate it at once on startup.
  334. This is slightly faster than growing the heap a number of times.
  335. 20. How can I access I/O ports?
  336. With versions before 0.99.10: if you're under DOS you can use the
  337. outport* and inport* procedures of the go32 unit.
  338. Since version 0.99.8, the Port array is supported like in TP, as
  339. long as you use the ports unit in your program (not available
  340. under Win32).
  341. I/O port access is possible under Linux, but that requires root
  342. privileges. Check the manuals for the IOPerm, ReadPort and
  343. WritePort procedures. (Unit Linux)
  344. 21. I'm using the Dos compiler under Windows 95
  345. There is a problem with the Dos compiler and Win 95 on computers
  346. with less than 16 MB. First set in the properties of the DOS box
  347. the DPMI memory size to max value. Now try to start a demo program
  348. in the DOS box, e.g. HELLO (starting takes some time). If this
  349. works you will be able to get the compiler to work by recompiling
  350. it with a smaller heap size, perhaps 2 or 4 MB (option -Chxxxx).
  351. 22. I'm using OS/2
  352. Problems have been reported that the GO32v2 compiler does not run
  353. on some OS/2 installations. You can use the native OS/2 compiler
  354. (strongly preferred solution) or maybe compile a GO32v1 compiler
  355. yourself. However, the GO32v2 version should generally work under
  356. OS/2 as well.
  357. 23. INSTALL.EXE of Dos version 0.99.10 reports "Load error: no DPMI"
  358. The file cwsdpmi.exe is missing in the main directory of the zip
  359. archive. The above message pops up if no other DPMI services are
  360. available. Such services are for example available in a Dos window
  361. of Windows. You can either extract that file from basego32.zip or
  362. download it from
  363. [38]http://www.brain.uni-freiburg.de/%7Eklaus/cwsdpmi.exe. Put it
  364. into the same directory as install.exe and run install again.
  365. 24. INSTALL.EXE of version 1.0 for Dos returns an error (-2) in
  366. Windows NT 4.0
  367. This is caused by long file names in some of the .ZIPs of the
  368. dosversion. A new installer will be generated that ignores the
  369. packages with long file names in it. Currently it is still being
  370. tested. Alternatively, one could use the installer from the Win32
  371. 1.0 version under NT. This has the additional benefit that the
  372. archives with long filenames can be selected and installed too.
  373. The exact cause of this problem is that a NT 4.0 dosbox doesn't
  374. support long file names for dos programs. Windows 95,98 and 2000
  375. don't exhibit this problem.
  376. + The current ZIPs on ftp have been updated with the new
  377. installer.
  378. + Dosw32100.zip, has now default the win32 installer, and the
  379. go32v2 installer packaged as installd.exe.
  380. + If you already downloaded one of the large Dos zips, repeated
  381. downloading is not necessary, just download a new installer:
  382. o [39]Plain dos installer. For dos without a 32-bit
  383. windows loaded or OS/2
  384. o [40]Win32 installer, for all win32 targets (win 95,98,NT
  385. en 2000) including their dosboxes
  386. + If you downloaded an OS/2 version, and experience problems,
  387. you can try to download the new dos installer
  388. 25. I want a new version NOW
  389. In the time between the release of new official versions, you can
  390. have a look at and test developer versions (so-called
  391. "snapshots"). Be warned though: this is work under progress, so in
  392. addition to old bugs fixed and new features added, this may also
  393. contain new bugs.
  394. Snapshots are generated automatically each night from the current
  395. source at that moment. Sometimes this may fail due to bigger
  396. changes not yet fully implemented. If your version doesn't work,
  397. try again one or two days later. You're advised not to download
  398. the GO32v1 version for Dos, since it's not supported any more.
  399. The latest snapshot can always be downloaded from the
  400. [41]development web page.
  401. To install a snapshot, extract the zip archive into the existing
  402. program directory of the last official version of Free Pascal
  403. (after making a backup of the original of course). You can also
  404. extract it into an empty directory and then move the files to the
  405. program directory, overwriting existing files. Make sure that you
  406. extract the ZIP archive such that the included directory structure
  407. remains intact. For example if you use PKUNZIP, use "pkunzip -d"
  408. instead of just "pkunzip". Note that snpashots also contain a new
  409. RTL which most likely can't be used with the previous release
  410. version, so backup your old RTL as well.
  411. 26. Where can I find a text mode IDE?
  412. The development of the IDE (integrated development environment) is
  413. not yet finished. However a working test version of the IDE is
  414. available as snapshot. It requires the latest compiler snapshot be
  415. installed on top of the current official version for your
  416. particular platform (1.00 for GO32v2 or Win32). So if you have not
  417. already done that, first install the latest official version (e.g.
  418. file dos100.zip or dos100full.zip, you find these in the
  419. [42]download section).
  420. Then get and extract the latest snapshot for your platform (e.g.
  421. snapshot.zip) into the directory containing the official version.
  422. Next, do the same with one of the IDE snapshots. For more details
  423. on where to find and how to install a snapshot, please see the
  424. previous FAQ item. For additional instructions for required IDE
  425. configuration please also read the next FAQ item.
  426. 27. How do I configure the Dos IDE?
  427. Once you have installed the IDE (see the previous FAQ item), it
  428. requires two configuration changes before it can compile. This is
  429. due to the fact that the IDE includes its own compiler; it does
  430. not use ppc386.exe and thus it also does not use the configuration
  431. in the file ppc386.cfg.
  432. Start fp.exe, select Target from the Compile menu and then check
  433. GO32v2. Next, choose Directories in the Otions menu and in the
  434. line "Unit directories" enter the path to your copy of the rtl
  435. directory, usually c:\pp\rtl\go32v2. If you have done everything
  436. correct and it still doesn't work, you may have grabbed a snapshot
  437. that has a bug; in this case try again one or two days later or
  438. ask for help on one of the [43]mailing lists.
  439. 28. Why are the generated binaries so big?
  440. There are several reasons and remedies for this:
  441. 1. If you are using 0.99.12: Due to some problems with the
  442. binary writer, 0.99.12 wasn't released with smartlinkable
  443. RTLs. Smartlinking causes only actually used procedures,
  444. functions and constants to be linked in.
  445. You can remedy this by using a development version and
  446. creating a smartlinking RTL. See the [44]make cycle faq or
  447. use a later release if available (0.99.14 and later do
  448. include a smartlinkable RTL). To turn on the generation of
  449. smartlinkable units, use the -Cx command line option when
  450. compiling your units. To turn on the linking of previously
  451. generated smarlinkable units, use the -XX (-XS in 0.99.12 and
  452. earlier) command line option when compiling a program.
  453. 2. Normally, all symbol information is included in the resulting
  454. program (for easier debugging). You can remove this by using
  455. the -Xs command line option when compiling your program (it
  456. won't do anything when compiling units)
  457. 3. You can use UPX to pack the .EXEs (just like e.g. pklite) for
  458. Dos (GO32v2) and Windows targets. Look [45]here for more
  459. info.
  460. 4. You can use LXLITE for packing EMX binaries, but you won't be
  461. able to run them under DOS (with extender) any more then. It
  462. might even not be possible to use them on lower OS/2 versions
  463. (like 2.x) depending on chosen type of compression. LXLITE
  464. can be found e.g. on [46]Hobbes, search for LXLITE.
  465. 5. Turn on optimalisations, both for supplied packages (RTL,
  466. API, FV, FCL) and for your own code, this will also decrease
  467. the code size.
  468. 29. Unit system, syslinux, sysos2 or syswin32 not found errors
  469. System (syslinux, sysos2 or syswin32, depending on platform) is
  470. Pascal's base unit which is implicitely used in all programs. This
  471. unit defines several standard procedures and structures, and must
  472. be found to be able to compile any pascal program by FPC.
  473. The location of the system.ppu and syslinux.o files are determined
  474. by the -Fu switch which can be specified commandline, but is
  475. usually in the ppc386.cfg configuration file.
  476. If the compiler can't find this unit there are three possible
  477. causes:
  478. 1. The ppc386.cfg isn't in the same path as the compiler
  479. executable (go32v2, win32 and OS/2) or can't be found as
  480. "/etc/ppc386.cfg" or ".ppc386.cfg" in your homedirectory
  481. (Linux).
  482. 2. The ppc386.cfg doesn't contain the -Fu line, or a wrong one.
  483. See the [47]make cycle faq, especially the chapters about the
  484. ppc386.cfg and the directory structure.
  485. 3. The files ARE found but the wrong version or platform.
  486. Correct ppc386.cfg to point to the right versions or
  487. reinstall the right versions (this can happen if you try to
  488. use a [48]snapshot compiler while the -Fu statemnt in the
  489. used ppc386.cfg still points to the RTL that came with the
  490. official release compiler).
  491. A handy trick can be executing "ppc386 programname -vt", this
  492. shows where the compiler is currently looking for the system
  493. unit's files. You might want to pipe this through more (Dos, OS/2,
  494. Windows) or less (Linux), since it can generate more than one
  495. screen information:
  496. Dos, OS/2, Windows:
  497. ppc386 programname -vt |more
  498. Linux:
  499. ppc386 programname -vt |less
  500. 30. Known bugs
  501. Go to the [49]bugs page
  502. 31. How can I find where an error occurred using the addresses a
  503. crashed program prints?
  504. 1. Starting with version 1.00, the easiest possibility is to
  505. recompile your program with -gl debugging option. This way
  506. unit LineInfo is automatically linked in, and the printout
  507. after a program crash then contains source line numbers in
  508. addition to addresses. To see RTL functions in the backtrace
  509. with their real name, you have to recompile the RTL with -gl
  510. too.
  511. 2. For older versions, or more comprehensive checking, compile
  512. the program with debugging information (use the -g command
  513. line option)
  514. 3. Load the program in the debugger (gdb(w) for 0.99.12b and
  515. earlier, gdbpas(w) for 0.99.14 and later) using
  516. gdb(pas)(w) --directory=<src dirs> myprog.exe
  517. Notes:
  518. o Under Linux, don't add the ".exe" after myprog
  519. o "src dirs" is a list of directories containing the
  520. source code files of myprog and the units it uses
  521. seperated by semi-colons (";"). The current directory is
  522. automatically included.
  523. 4. Once inside the debugger, you can (optionally) set the
  524. command line options that will be passed to your program
  525. using the command "set args <option1 option2 ...>"
  526. 5. To start the program, type "run" and press enter
  527. 6. After the program has crashed, the address of the instruction
  528. where the crash occurred will be shown. The debugger will try
  529. to display the source code line corresponding with this
  530. address. Note that this can be inside a procedure of the RTL,
  531. so the source may not always be available and most likely the
  532. RTL wasn't compiled with debugging information.
  533. 7. If you then type "bt" (BackTrace), the addreses in the call
  534. stack will be shown (the addresses of the procedures which
  535. were called before the program got to the current address).
  536. You can see which source code lines these present using the
  537. command
  538. info line *<address>
  539. For example:
  540. info line *0x05bd8
  541. References
  542. 1. file://localhost/home/pfv/html/faq.htm#WhatIsFP
  543. 2. file://localhost/home/pfv/html/faq.htm#versions
  544. 3. file://localhost/home/pfv/html/faq.htm#FPandGNUPascal
  545. 4. file://localhost/home/pfv/html/faq.htm#WhereToGetFP
  546. 5. file://localhost/home/pfv/html/faq.htm#PortabilityTips
  547. 6. file://localhost/home/pfv/html/faq.htm#OOP
  548. 7. file://localhost/home/pfv/html/faq.htm#HOMEWORK
  549. 8. file://localhost/home/pfv/html/faq.htm#HowcanIbuildaunit
  550. 9. file://localhost/home/pfv/html/faq.htm#TurboVision
  551. 10. file://localhost/home/pfv/html/faq.htm#CompileSystemUnit
  552. 11. file://localhost/home/pfv/html/faq.htm#Internalerror9999
  553. 12. file://localhost/home/pfv/html/faq.htm#Howdoesfunctionoverloadingwork
  554. 13. file://localhost/home/pfv/html/faq.htm#HowToCallCFuncuntions
  555. 14. file://localhost/home/pfv/html/faq.htm#HowToUseGraph
  556. 15. file://localhost/home/pfv/html/faq.htm#WrongColors
  557. 16. file://localhost/home/pfv/html/faq.htm#IntegratedAssemblerSyntax
  558. 17. file://localhost/home/pfv/html/faq.htm#HowToAccessDosMemory
  559. 18. file://localhost/home/pfv/html/faq.htm#FPwithoutfpu
  560. 19. file://localhost/home/pfv/html/faq.htm#AccessingMoreThan4MB
  561. 20. file://localhost/home/pfv/html/faq.htm#accessioports
  562. 21. file://localhost/home/pfv/html/faq.htm#ImusingWin95
  563. 22. file://localhost/home/pfv/html/faq.htm#ImusingOS2
  564. 23. file://localhost/home/pfv/html/faq.htm#dpmi
  565. 24. file://localhost/home/pfv/html/faq.htm#instal10NT
  566. 25. file://localhost/home/pfv/html/faq.htm#snapshot
  567. 26. file://localhost/home/pfv/html/faq.htm#ideinst
  568. 27. file://localhost/home/pfv/html/faq.htm#ideconfig
  569. 28. file://localhost/home/pfv/html/faq.htm#binariesbig
  570. 29. file://localhost/home/pfv/html/faq.htm#systemnotfound
  571. 30. file://localhost/home/pfv/html/faq.htm#KnownBugs
  572. 31. file://localhost/home/pfv/html/faq.htm#ErrorPos
  573. 32. file://localhost/home/pfv/html/download.html
  574. 33. file://localhost/home/pfv/html/develop.html#snapshot
  575. 34. file://localhost/home/pfv/html/bugs.html
  576. 35. http://www.delorie.com/djgpp/v2faq/faq102.html#Syntax
  577. 36. http://www.rt66.com/%7Ebrennan/djgpp/djgpp%A0asm.html
  578. 37. http://rcs.urz.tu-dresden.de/schoenfu/zip/asmtrans.zip
  579. 38. http://www.brain.uni-freiburg.de/%7Eklaus/cwsdpmi.exe
  580. 39. ftp://ftp.freepascal.org/pub/fpc/dist/dos-1.00/separate/install.exe
  581. 40. ftp://ftp.freepascal.org/pub/fpc/dist/win32-1.00/separate/install.exe
  582. 41. file://localhost/home/pfv/html/develop.html#snapshot
  583. 42. file://localhost/home/pfv/html/download.html
  584. 43. file://localhost/home/pfv/html/maillist.html
  585. 44. file://localhost/home/pfv/html/makecyc.html
  586. 45. http://wildsau.idv.uni-linz.ac.at/mfx/upx.html
  587. 46. http://hobbes.nmsu.edu/
  588. 47. file://localhost/home/pfv/html/makecyc.html
  589. 48. file://localhost/home/pfv/html/faq.htm#snapshot
  590. 49. file://localhost/home/pfv/html/bugs.html