brunner@bullhead.uucp (09/27/90)
Hello everyone, I've just isolated a bug in the 2.1z version of the Metaware High C compiler, and received a set of bug-exercising pascal programs for their 3.0 Professional Pascal compiler, so I'm starting what may be the final round of compiler back-and-fourths for both the C and Pascal compilers. I've sent in all the 2.1y reports I've received - those of you who got acks back from me know this - what I'm solliciting here are final 2.1y and pp3.0 bugs. What works best, especially if your problem code is part of an application with (re-)distribution restrictions (e.g., Pike's son-of-sam editor), is the preprocessor output - so run whatever makefile you have, catch the stdout in a bucket, edit the line specific to the file(s) where the compiler fails or generates bogus code - change the "-O" to a "-E" and run as a makescript. Now all the code dependencies (includes) are present in foo.c.E If the output is huge or consists of several files, send me a floppy or a tape at: Eric Brunner IBM AWD 1510 Page Mill Road, MS 35A Palo Alto, Calif. 94303 If the output size < sizeof(internet-mailer-reasonable), then send it via email. If you've internet access, use "brunner@monet.berkeley.edu" - this is the simplest address. If not, use relative bang addressing to uunet, and I'm at "...!uunet!ibmsupt!brunner". Thanks in advance! #include <std/disclaimer.h> Eric Brunner, Consultant, IBM AWD Palo Alto (415) 855-4486 inet: brunner@monet.berkeley.edu uucp: uunet!ibmsupt!brunner trying to understand multiprocessing is like having bees live inside your head.