From vikasnkumar at users.sourceforge.net Sun Oct 22 23:16:52 2006 From: vikasnkumar at users.sourceforge.net (Vikas N Kumar) Date: Sun, 22 Oct 2006 19:16:52 -0400 Subject: [movitz-devel] building image Message-ID: <792c60620610221616m271d0e76j9fb2c57b2c9cb95f@mail.gmail.com> Hi How does one build a fresh image of movitz using CLISP ? Based on the earlier emails from last year, I tried (load "load.lisp") (movitz:create-image) After ignoring some errors and not compiling pci.lisp and ip4.lisp and los0-gc.lisp, in the end i see SYMBOLIC-IMAGE followed by a number . But all that is created is .lib and .fas files. How does one create the final .img file ? Thanks and regards, Vikas -- http://www.vikaskumar.org/ From ffjeld at common-lisp.net Mon Oct 23 06:37:44 2006 From: ffjeld at common-lisp.net (Frode Vatvedt Fjeld) Date: 23 Oct 2006 08:37:44 +0200 Subject: [movitz-devel] Re: building image References: <792c60620610221616m271d0e76j9fb2c57b2c9cb95f@mail.gmail.com> Message-ID: <2hwt6ra66f.fsf@vserver.cs.uit.no> Hi Vikas, "Vikas N Kumar" writes: > How does one create the final .img file ? Use the function dump-image, which takes as arguments the pathname of the image file (default "los0-image"), and the (symbolic) image you want to dump, which is typically the thingy that was just returned from create-image. Create-image will by default set the variable movitz:*image* to the same value as it returns. Many movitz functions subsequently use this variable as the (default) image to manipulate, including dump-image. There are functions that do useful manipulations of an existing image. Most notable of these is movitz-compile-file, which compiles the contents of a file pretty much the way you expect. That is, if the file contains "(defun foo (x) (1+ x))", then the current *image*'s function definition for foo will be updated to reflect this. Then, the next image generated by dump-image will contain the new foo. The typical task, then, of the Movitz "IDE" (in whatever form, there are I believe two variants of Emacs integration in CVS) is to have a keybinding M-C-x or C-c C-c that places the current top-level form in a temporary file and applies movitz-compile-file to that file. Regards, -- Frode Vatvedt Fjeld From vikasnkumar at users.sourceforge.net Mon Oct 23 13:28:57 2006 From: vikasnkumar at users.sourceforge.net (Vikas N Kumar) Date: Mon, 23 Oct 2006 09:28:57 -0400 Subject: [movitz-devel] Re: building image In-Reply-To: <2hwt6ra66f.fsf@vserver.cs.uit.no> References: <792c60620610221616m271d0e76j9fb2c57b2c9cb95f@mail.gmail.com> <2hwt6ra66f.fsf@vserver.cs.uit.no> Message-ID: <792c60620610230628l6d7eb592s6c17350623dc6021@mail.gmail.com> Thanks Frode. But am I doing the right thing by ignoring the errors that are occuring while compiling los0-gc.lisp ? Regards, Vikas On 23 Oct 2006 08:37:44 +0200, Frode Vatvedt Fjeld wrote: > Hi Vikas, > > "Vikas N Kumar" writes: > > > How does one create the final .img file ? > > Use the function dump-image, which takes as arguments the pathname of > the image file (default "los0-image"), and the (symbolic) image you > want to dump, which is typically the thingy that was just returned > from create-image. > > Create-image will by default set the variable movitz:*image* to the > same value as it returns. Many movitz functions subsequently use this > variable as the (default) image to manipulate, including dump-image. > > There are functions that do useful manipulations of an existing > image. Most notable of these is movitz-compile-file, which compiles > the contents of a file pretty much the way you expect. That is, if the > file contains "(defun foo (x) (1+ x))", then the current *image*'s > function definition for foo will be updated to reflect this. Then, > the next image generated by dump-image will contain the new foo. > > The typical task, then, of the Movitz "IDE" (in whatever form, there > are I believe two variants of Emacs integration in CVS) is to have a > keybinding M-C-x or C-c C-c that places the current top-level form in > a temporary file and applies movitz-compile-file to that file. > > Regards, > -- > Frode Vatvedt Fjeld > > _______________________________________________ > movitz-devel site list > movitz-devel at common-lisp.net > http://common-lisp.net/mailman/listinfo/movitz-devel > -- http://www.vikaskumar.org/ From ffjeld at common-lisp.net Mon Oct 23 14:18:21 2006 From: ffjeld at common-lisp.net (Frode Vatvedt Fjeld) Date: 23 Oct 2006 16:18:21 +0200 Subject: [movitz-devel] Re: building image References: <792c60620610221616m271d0e76j9fb2c57b2c9cb95f@mail.gmail.com> <2hwt6ra66f.fsf@vserver.cs.uit.no> <792c60620610230628l6d7eb592s6c17350623dc6021@mail.gmail.com> Message-ID: <2hslhf9kuq.fsf@vserver.cs.uit.no> "Vikas N Kumar" writes: > But am I doing the right thing by ignoring the errors that are > occuring while compiling los0-gc.lisp ? Well, it is not the intention that there should be errors there. What do they look like? -- Frode Vatvedt Fjeld From vikasnkumar at users.sourceforge.net Thu Oct 26 22:34:30 2006 From: vikasnkumar at users.sourceforge.net (Vikas N Kumar) Date: Thu, 26 Oct 2006 18:34:30 -0400 Subject: [movitz-devel] Re: building image In-Reply-To: <2hslhf9kuq.fsf@vserver.cs.uit.no> References: <792c60620610221616m271d0e76j9fb2c57b2c9cb95f@mail.gmail.com> <2hwt6ra66f.fsf@vserver.cs.uit.no> <792c60620610230628l6d7eb592s6c17350623dc6021@mail.gmail.com> <2hslhf9kuq.fsf@vserver.cs.uit.no> Message-ID: <792c60620610261534xda98edcie054a8b3ec4c7f2a@mail.gmail.com> Hi Frode, I have attached the error in a text file. I ran Clisp 2.41 (latest stable version) on windows XP and tried building the latest CVS version of movitz. I tried copying all the warnings as much as I could. There is one part in the text file with the string "" where I missed the warning messages. If you scroll down to the end of the file, you will see the errors in los0-gc.lisp which I encountered. I typed "continue" and avoided compiling los0-gc.lisp. What is the error due to ? Regards, Vikas On 23 Oct 2006 16:18:21 +0200, Frode Vatvedt Fjeld wrote: > "Vikas N Kumar" writes: > > > But am I doing the right thing by ignoring the errors that are > > occuring while compiling los0-gc.lisp ? > > Well, it is not the intention that there should be errors there. What > do they look like? > > -- > Frode Vatvedt Fjeld > > _______________________________________________ > movitz-devel site list > movitz-devel at common-lisp.net > http://common-lisp.net/mailman/listinfo/movitz-devel > -- http://www.vikaskumar.org/ -------------- next part -------------- [10]> (movitz:create-image ) ;; NIL value: #xC5. WARNING in DEFMACRO-expander-LOAD-macro-106775 : variable VERBOSE is not used. Misspelled or missing IGNORE declaration? WARNING in DEFMACRO-expander-LOAD-macro-106775 : variable COMMON-LISP:PRINT is not used. Misspelled or missing IGNORE declaration? WARNING in DEFMACRO-expander-LOAD-macro-106775 : variable IF-DOES-NOT-EXIST is not used. Misspelled or missing IGNORE declaration? WARNING in DEFMACRO-expander-LOAD-macro-106775 : variable EXTERNAL-FORMAT is not used. Misspelled or missing IGNORE declaration? ;; While Movitz compiling EXPT in losp/muerte/integers.lisp: WARNING: compiling typep (COMMON-LISP:RATIONAL COMMON-LISP:* -1) [COMMON-LISP] ;; While Movitz compiling EXPT in losp/muerte/integers.lisp: WARNING: compiling typep (COMMON-LISP:RATIONAL COMMON-LISP:* -1) [COMMON-LISP] ;; While Movitz compiling ADJOIN in losp/muerte/cons.lisp: WARNING: Compiler-macro for MEMBER failed: Illegal keyword/value pair :TEST-NOT, TEST-NOT in argument list. The allowed keywords are (:TEST :KEY) ;; While Movitz compiling ADJUST-ARRAY in losp/muerte/arrays.lisp: WARNING: Unused variable: INITIAL-CONTENTS ;; While Movitz compiling ADJUST-ARRAY in losp/muerte/arrays.lisp: WARNING: Unused variable: ELEMENT-TYPE ;; While Movitz compiling SUBSEQ in losp/muerte/sequences.lisp: WARNING: Unused variable: #:|sub-vector-123592| Function CANONICALIZE-SPECIALIZERS is not defined WARNING in CANONICALIZE-SPECIALIZERS : Function CANONICALIZE-SPECIALIZER is not defined WARNING in PARSE-DEFMETHOD : Function EXTRACT-LAMBDA-LIST is not defined WARNING in PARSE-DEFMETHOD : Function EXTRACT-SPECIALIZERS is not defined WARNING in REQUIRED-PORTION : Function GF-REQUIRED-ARGLIST is not defined WARNING in GF-REQUIRED-ARGLIST : Function ANALYZE-LAMBDA-LIST is not defined WARNING in EXTRACT-LAMBDA-LIST : Function ANALYZE-LAMBDA-LIST is not defined WARNING in EXTRACT-SPECIALIZERS : Function ANALYZE-LAMBDA-LIST is not defined WARNING in ENSURE-METHOD : Function MAKE-INSTANCE-STANDARD-METHOD is not defined WARNING in ENSURE-METHOD : Function MOVITZ-ADD-METHOD is not defined WARNING in MAKE-INSTANCE-STANDARD-METHOD : Function STD-COMPUTE-METHOD-FUNCTION is not defined WARNING in MOVITZ-ADD-METHOD : Function MOVITZ-FIND-METHOD is not defined WARNING in MOVITZ-ADD-METHOD : Function MOVITZ-REMOVE-METHOD is not defined WARNING in ADD-READER-METHOD : Function TOP-LEVEL-ENVIRONMENT is not defined WARNING in ADD-WRITER-METHOD : Function TOP-LEVEL-ENVIRONMENT is not defined WARNING in SLOW-METHOD-LOOKUP : Function COMPUTE-APPLICABLE-METHODS-USING-CLASSES is not defin WARNING in SLOW-METHOD-LOOKUP : Function STD-COMPUTE-EFFECTIVE-METHOD-FUNCTION is not defined WARNING in SLOW-METHOD-LOOKUP : Function COMPUTE-EFFECTIVE-METHOD-FUNCTION is not defined WARNING in COMPUTE-APPLICABLE-METHODS-USING-CLASSES-2 : Function STD-METHOD-MORE-SPECIFIC-P is not defined WARNING in COMPUTE-APPLICABLE-METHODS-USING-CLASSES-2 : Function METHOD-MORE-SPECIFIC-P is not defined WARNING in STD-COMPUTE-METHOD-FUNCTION : Function COMPILE-IN-LEXICAL-ENVIRONMENT is not defined ;; While Movitz compiling CLONE-RUN-TIME-CONTEXT in losp/muerte/run-time-contex .lisp: WARNING: Unused variable: NAME WARNING in 1-1 : Function PRINT-LOOP-UNIVERSE is not defined WARNING in DEFMACRO-expander-WITH-MINIMAX-VALUE-macro-150170 : Function LOOP-TYPED-INIT is not defined WARNING in WITH-MINIMAX-VALUE : Function LOOP-TYPED-INIT is not defined WARNING in LOOP-MAKE-PSETQ : Function LOOP-MAKE-DESETQ is not defined WARNING in LOOP-CONSTANT-FOLD-IF-POSSIBLE : Function LOOP-WARN is not defined WARNING in DUPLICATABLE-CODE-P : Function ESTIMATE-CODE-SIZE is not defined WARNING in ESTIMATE-CODE-SIZE : Function ESTIMATE-CODE-SIZE-1 is not defined WARNING in LOOP-TRANSLATE : Function LOOP-ITERATION-DRIVER is not defined WARNING in LOOP-TRANSLATE : Function LOOP-BIND-BLOCK is not defined WARNING in LOOP-ITERATION-DRIVER : Function LOOP-POP-SOURCE is not defined WARNING in LOOP-ITERATION-DRIVER : Function LOOP-HACK-ITERATION is not defined WARNING in LOOP-MAKE-VARIABLE : Function LOOP-DECLARE-VARIABLE is not defined WARNING in LOOP-MAKE-VARIABLE : Function LOOP-DECLARE-VARIABLE is not defined WARNING in LOOP-DO-IF-GET-CLAUSE : Function LOOP-WHEN-IT-VARIABLE is not defined WARNING in LOOP-DO-THEREIS : Function LOOP-WHEN-IT-VARIABLE is not defined WARNING in LOOP-FOR-BEING : Function LOOP-COLLECT-PREPOSITIONAL-PHRASES is not defined ;; While Movitz compiling MAP-STACK-DIT in losp/muerte/scavenge.lisp: WARNING: Unused variable: OLD-CODE-VECTOR ** - Continuable Error COMMON-LISP:READ from # : must specify element of vector of length 250 If you continue (by typing 'continue'): Skip Movitz compilation of "losp/los0-gc .lisp". The following restarts are also available: RETRY :R1 Restart Movitz compilation of "losp/los0-gc.lisp". SKIP-TOPLEVEL-FORM :R2 Skip the compilation of top-level form REQUIRE. RETRY :R3 Restart Movitz compilation of #P"losp\\los0.lisp". CONTINUE :R4 Skip Movitz compilation of #P"losp\\los0.lisp". ABORT :R5 COMMON-LISP:ABORT Break 1 LOS0[11]> continue ;; While Movitz compiling TEST-CLOSURE in losp\los0.lisp: WARNING: dynamic-extent flet: CLOSURE ;; While Movitz compiling INSTALL-INTERNAL-TIME in losp\los0.lisp: WARNING: compiling typep (COMMON-LISP:RATIONAL 0 COMMON-LISP:*) [COMMON-LISP] ;; While Movitz compiling INSTALL-INTERNAL-TIME in losp\los0.lisp: WARNING: compiling typep (COMMON-LISP:RATIONAL 0 COMMON-LISP:*) [COMMON-LISP] 0 errors, 0 warnings # From vikasnkumar at users.sourceforge.net Thu Oct 26 22:42:54 2006 From: vikasnkumar at users.sourceforge.net (Vikas N Kumar) Date: Thu, 26 Oct 2006 18:42:54 -0400 Subject: [movitz-devel] Re: building image In-Reply-To: <792c60620610261534xda98edcie054a8b3ec4c7f2a@mail.gmail.com> References: <792c60620610221616m271d0e76j9fb2c57b2c9cb95f@mail.gmail.com> <2hwt6ra66f.fsf@vserver.cs.uit.no> <792c60620610230628l6d7eb592s6c17350623dc6021@mail.gmail.com> <2hslhf9kuq.fsf@vserver.cs.uit.no> <792c60620610261534xda98edcie054a8b3ec4c7f2a@mail.gmail.com> Message-ID: <792c60620610261542t7c8f3631ycffbda6ae765b577@mail.gmail.com> Hi Frode, When I run dump-image i get the following error. What am I supposed to do ? From the code in image.lisp, dump-image should be able to use all the default values. [21]> (movitz:dump-image ) ;; Doing initiating dump.. *** - The value of (HASH-TABLE-TEST MOVITZ::LISP-HASH) must be one of EQ, EQL, EQUAL The value is: FASTHASH-EQ The following restarts are available: ABORT :R1 ABORT Break 1 [22]> Regards, Vikas On 10/26/06, Vikas N Kumar wrote: > Hi Frode, > I have attached the error in a text file. > I ran Clisp 2.41 (latest stable version) on windows XP and tried > building the latest CVS version of movitz. I tried copying all the > warnings as much as I could. > There is one part in the text file with the string "" where I > missed the warning messages. > If you scroll down to the end of the file, you will see the errors in > los0-gc.lisp which I encountered. I typed "continue" and avoided > compiling los0-gc.lisp. > > What is the error due to ? > > Regards, > Vikas > > On 23 Oct 2006 16:18:21 +0200, Frode Vatvedt Fjeld > wrote: > > "Vikas N Kumar" writes: > > > > > But am I doing the right thing by ignoring the errors that are > > > occuring while compiling los0-gc.lisp ? > > > > Well, it is not the intention that there should be errors there. What > > do they look like? > > > > -- > > Frode Vatvedt Fjeld > > > > _______________________________________________ > > movitz-devel site list > > movitz-devel at common-lisp.net > > http://common-lisp.net/mailman/listinfo/movitz-devel > > > > > -- > http://www.vikaskumar.org/ > > > -- http://www.vikaskumar.org/ From ffjeld at common-lisp.net Fri Oct 27 06:23:53 2006 From: ffjeld at common-lisp.net (Frode Vatvedt Fjeld) Date: 27 Oct 2006 08:23:53 +0200 Subject: [movitz-devel] Re: building image References: <792c60620610221616m271d0e76j9fb2c57b2c9cb95f@mail.gmail.com> <2hwt6ra66f.fsf@vserver.cs.uit.no> <792c60620610230628l6d7eb592s6c17350623dc6021@mail.gmail.com> <2hslhf9kuq.fsf@vserver.cs.uit.no> <792c60620610261534xda98edcie054a8b3ec4c7f2a@mail.gmail.com> Message-ID: <2h3b9a9szq.fsf@vserver.cs.uit.no> "Vikas N Kumar" writes: > ** - Continuable Error > COMMON-LISP:READ from > # @339> > : must specify element of vector of length 250 This is in all likelihood because the CLisp reader doesn't accept the syntax "#250()". I believe that CLisp will accept e.g. "#250(nil)", so I checked in a fix like that just now. -- Frode Vatvedt Fjeld From ffjeld at common-lisp.net Fri Oct 27 06:58:37 2006 From: ffjeld at common-lisp.net (Frode Vatvedt Fjeld) Date: 27 Oct 2006 08:58:37 +0200 Subject: [movitz-devel] Re: building image References: <792c60620610221616m271d0e76j9fb2c57b2c9cb95f@mail.gmail.com> <2hwt6ra66f.fsf@vserver.cs.uit.no> <792c60620610230628l6d7eb592s6c17350623dc6021@mail.gmail.com> <2hslhf9kuq.fsf@vserver.cs.uit.no> <792c60620610261534xda98edcie054a8b3ec4c7f2a@mail.gmail.com> <792c60620610261542t7c8f3631ycffbda6ae765b577@mail.gmail.com> Message-ID: <2hy7r28cte.fsf@vserver.cs.uit.no> "Vikas N Kumar" writes: > Hi Frode, > When I run dump-image i get the following error. What am I supposed to > do ? From the code in image.lisp, dump-image should be able to use all > the default values. > > [21]> (movitz:dump-image ) > ;; Doing initiating dump.. > *** - The value of (HASH-TABLE-TEST MOVITZ::LISP-HASH) must be one of EQ, EQL, > EQUAL > The value is: FASTHASH-EQ > The following restarts are available: > ABORT :R1 ABORT > Break 1 [22]> Hi Vikas, this issue occurs when Movitz tries to convert a native (i.e. CLisp) hash-table object to its internal format. When it looks up the hash-table's test function, the code expects it to be one of eq, eql, or equal (there is an ecase statement about this in storage-types.lisp). However, CLisp returns things like fasthash-eql, which confuses Movitz. I have added code to deal with this now. Also, I verified that after these two fixes, CLisp will build Movitz Ok, at least on two systems I have access to (windows/cygwin and FreeBSD). -- Frode Vatvedt Fjeld From vikasnkumar at users.sourceforge.net Fri Oct 27 15:08:31 2006 From: vikasnkumar at users.sourceforge.net (Vikas N Kumar) Date: Fri, 27 Oct 2006 11:08:31 -0400 Subject: [movitz-devel] Re: building image In-Reply-To: <2hy7r28cte.fsf@vserver.cs.uit.no> References: <792c60620610221616m271d0e76j9fb2c57b2c9cb95f@mail.gmail.com> <2hwt6ra66f.fsf@vserver.cs.uit.no> <792c60620610230628l6d7eb592s6c17350623dc6021@mail.gmail.com> <2hslhf9kuq.fsf@vserver.cs.uit.no> <792c60620610261534xda98edcie054a8b3ec4c7f2a@mail.gmail.com> <792c60620610261542t7c8f3631ycffbda6ae765b577@mail.gmail.com> <2hy7r28cte.fsf@vserver.cs.uit.no> Message-ID: <792c60620610270808j258e2166q2fe8d61ab84246a0@mail.gmail.com> Hi Frode Thanks. I was able to build the image fine and dump it But now i get this error when i load that image with Bochs. Any ideas on debugging this issue ? Plex86/Bochs VGABios 0.5c 07 Jul 2005 This VGA/VBE Bios is released under the GNU LGPL Please visit : . http://bochs.sourceforge.net . http://www.nongnu.org/vgabios Bochs VBE Display Adapter enabled Bochs BIOS, 1 cpu, $Revision: 1.138.2.1 $ $Date: 2005/07/06 19:30:36 $ ata0 master: Generic 1234 ATA-2 Hard-Disk (499 MBytes) ata0 slave: Generic 1234 ATAPI-4 CD-Rom/DVD-Rom Booting from Hard Disk... Loading Movitz 2.. (Failed!) On 27 Oct 2006 08:58:37 +0200, Frode Vatvedt Fjeld wrote: > "Vikas N Kumar" writes: > > > Hi Frode, > > When I run dump-image i get the following error. What am I supposed to > > do ? From the code in image.lisp, dump-image should be able to use all > > the default values. > > > > [21]> (movitz:dump-image ) > > ;; Doing initiating dump.. > > *** - The value of (HASH-TABLE-TEST MOVITZ::LISP-HASH) must be one of EQ, EQL, > > EQUAL > > The value is: FASTHASH-EQ > > The following restarts are available: > > ABORT :R1 ABORT > > Break 1 [22]> > > Hi Vikas, this issue occurs when Movitz tries to convert a native > (i.e. CLisp) hash-table object to its internal format. When it looks > up the hash-table's test function, the code expects it to be one of > eq, eql, or equal (there is an ecase statement about this in > storage-types.lisp). However, CLisp returns things like fasthash-eql, > which confuses Movitz. I have added code to deal with this now. > > Also, I verified that after these two fixes, CLisp will build Movitz > Ok, at least on two systems I have access to (windows/cygwin and > FreeBSD). > > -- > Frode Vatvedt Fjeld > > _______________________________________________ > movitz-devel site list > movitz-devel at common-lisp.net > http://common-lisp.net/mailman/listinfo/movitz-devel > -- http://www.vikaskumar.org/ From ffjeld at common-lisp.net Fri Oct 27 17:46:30 2006 From: ffjeld at common-lisp.net (Frode Vatvedt Fjeld) Date: 27 Oct 2006 19:46:30 +0200 Subject: [movitz-devel] Re: building image References: <792c60620610221616m271d0e76j9fb2c57b2c9cb95f@mail.gmail.com> <2hwt6ra66f.fsf@vserver.cs.uit.no> <792c60620610230628l6d7eb592s6c17350623dc6021@mail.gmail.com> <2hslhf9kuq.fsf@vserver.cs.uit.no> <792c60620610261534xda98edcie054a8b3ec4c7f2a@mail.gmail.com> <792c60620610261542t7c8f3631ycffbda6ae765b577@mail.gmail.com> <2hy7r28cte.fsf@vserver.cs.uit.no> <792c60620610270808j258e2166q2fe8d61ab84246a0@mail.gmail.com> Message-ID: <2hfyd98xe1.fsf@vserver.cs.uit.no> "Vikas N Kumar" writes: > Hi Frode > Thanks. I was able to build the image fine and dump it > But now i get this error when i load that image with Bochs. > Any ideas on debugging this issue ? > [..] > Booting from Hard Disk... > Loading Movitz 2.. Hi Vikas, The image produced by dump-image includes a 512-byte bootloader that is only able to load from floppy. Apparently you are trying to use it as a hard-drive image. If you configure Bochs to use the image as floppya, it should work. Since this way you can use the output of dump-image directly, this is most convenient. It is also possible to use GRUB to construct a hard-drive or CDROM-compatible image, etc. -- Frode Vatvedt Fjeld