[fd-dev] [ANNOUNCE] kernel 2025a
 From: Bart Oldeman 
 To: fd-dev@topica.com, fd-kernel@topica.com

Hereby I would like to thanks all testers. You know who you are.
There were quite a few small things to fix in kernel 2025a.

Get it shortly from

FAT32 is of course still experimental, though more stable than before.

Some of you were interested in md5sums - here you go:
248cb3eaf993e19aeabdba2783fd8e1a  ke2025asrc.zip
58996a4c5672fec6f70e2a5e46f0dacc  ke2025a16.zip
5d842f338e970c401d6f97438bd97cc5  ke2025a32.zip

+ Changes Tom
    * initialize LBA parameter struct to 0's - some BIOSes don't fill
      in everything. Thanks to Bernd Blaauw's bug report.
    * Minor boot sector optimization.
    * added some exeflat options
+ Changes Jeremy
    * changes to fdkrncfg.c: mostly cosmetic to the output
    * added sys.doc file (updated by Bart)
+ Changes Bart
    * let int 0 (divide by zero) and int 6 (invalid opcode) print a
      stack trace to aid in debugging
    * fixed floppy drive detection for XT's.
    * fixed bug in int21/ah=0x32 and friends - we should flush data before the
      forced media_check
    * Use CHS addressing if possible on partitions with the relevant ID,
      you can overide this using the new FORCELBA and GLOBALENABLELBASUPPORT
      sys config options.
    * Always use BIOS values for translating LBA to CHS for fixed disks -
      ignoring the boot sector values. For floppies the boot sector is used.
    * fixed "truename hello." to remove the trailing dot.
    * added options to SYS to write to a boot sector image file
    * set Critical Error Code for ioctl's
    * fake cluster size for 64k clusters just like FAT32 for INT21/AH=0x36
      and friends.
+ Changes Bart+Tom+Victor
    * make makefile system more robust
+ Changes Victor
    * cleaned up map_cluster and dir_read (bug fixed by Bart)
    * updated LFN API (in development, not used yet)