[Openmcl-devel] trouble tracking trunk on linuxx86

[Gary Byers]

What (if anything) happens if you quit from that lisp session and run 
the newly-built image ?

I believe that this problem likely stems from the impact of hash-table 
changes made in the trunk
several months ago.  When those changes were introduced, it was 
necessary to call (REBUILD-CCL)
twice in order to ensure that all code was compiled with the most recent 
versions of affected code.

Many people found that requrement confusing, and I changed things a few 
weeks ago to remove
the need to rebuild twice.  I may have introduced the problem you 
describe now at that point,  which
may be caused by older hash-table-sensitive code and newer code trying 
to coexist in the same image.


In the past, we often tried to avoid exposing these kinds of 
bootstrapping issues by providing new
binaries when changes ike this were made, but that wasn't done in this case.

On 05/19/2015 05:57 PM, Mark H. David wrote:
> QutDid
>
> ./lx86cl --no-init
> Welcome to Clozure Common Lisp Version 1.10-dev-r16128M-trunk  (LinuxX8632)!
> (ccl:rebuild-ccl :full t)
> =>
> .... a lot of output ....
> ;Building lisp-kernel ...
> #<error printing #<Recursive printing error  #x19B1D441> #x19B1F68E>
>> While executing: #<error printing #<Recursive printing error  #x1400984E> #x1423966E>, in process #<error printing #<Recursive printing error  #x19B1B9E1> #x1063A40E>(1).
>> Type :POP to abort, :R for a list of available restarts.
>> Type :? for other options.
> 1 > :b
>   (F6FFEC58) : 0 (FUNCALL #'#<REBUILD-CCL> :UPDATE NIL :FULL T :CLEAN T :KERNEL T :FORCE NIL :RELOAD T :EXIT NIL :RELOAD-ARGUMENTS NIL :VERBOSE NIL :OPTIONAL-FEATURES NIL :SAVE-SOURCE-LOCATIONS :NO-TEXT :ALLOW-CONSTANT-REDEFINITION T) 2375
>   (F6FFED40) : 1 (CALL-CHECK-REGS REBUILD-CCL :FULL T) 247
>   (F6FFED5C) : 2 (TOPLEVEL-EVAL (REBUILD-CCL :FULL T) NIL) 711
>   (F6FFED98) : 3 (READ-LOOP :INPUT-STREAM #<SYNONYM-STREAM to *TERMINAL-IO* #x18176216> :OUTPUT-STREAM #<SYNONYM-STREAM to *TERMINAL-IO* #x1817613E> :BREAK-LEVEL 0 :PROMPT-FUNCTION #<Compiled-function (:INTERNAL CCL::READ-LOOP) (Non-Global)  #x1070EA06>) 2039
>   (F6FFEEBC) : 4 (RUN-READ-LOOP :BREAK-LEVEL 0) 183
>   (F6FFEED0) : 5 (TOPLEVEL-LOOP) 71
>   (F6FFEED8) : 6 (FUNCALL #'#<(:INTERNAL (TOPLEVEL-FUNCTION (CCL::LISP-DEVELOPMENT-SYSTEM T)))>) 71
>   (F6FFEEE8) : 7 (FUNCALL #'#<(:INTERNAL CCL::MAKE-MCL-LISTENER-PROCESS)>) 551
>   (F6FFEF34) : 8 (RUN-PROCESS-INITIAL-FORM #<TTY-LISTENER listener(1) [Active] #x18175806> (#<COMPILED-LEXICAL-CLOSURE # #x1817553E>)) 639
>   (F6FFEF78) : 9 (FUNCALL #'#<(:INTERNAL (CCL::%PROCESS-PRESET-INTERNAL (PROCESS)))> #<TTY-LISTENER listener(1) [Active] #x18175806> (#<COMPILED-LEXICAL-CLOSURE # #x1817553E>)) 519
>   (F6FFEFCC) : 10 (FUNCALL #'#<(:INTERNAL CCL::THREAD-MAKE-STARTUP-FUNCTION)>) 255
> 1 >
>
>
>
> Note: pretty much the same thing happened with 64-bit when I tried that.
>
> Thanks,
>
> Mark
>
> _______________________________________________
> Openmcl-devel mailing list
> Openmcl-devel at clozure.com
> https://lists.clozure.com/mailman/listinfo/openmcl-devel