Re: Small bug in 1.3.76

Andrew C. Esh (andrewes@cnt.com)
Thu, 21 Mar 1996 08:58:39 -0600


Boris Tobotras wrote:
>
> In arch/i386/kernel/setup.c there should be "break" in cases inside
> the switch where i486model and i586model gets called (sorry, haven't
> sources handy).

Hi Tobotras! Saw you in #linux yesterday. :)

I have a bug too. While compiling module support for the FAT
file system, I got some errors pertaining to the "_basic_version"
macro while I was compiling "fs/fat/fatfs_syms.c". It seems that
the "X(a)" macro call (defined in "include/linux/symtab_begin.h")
has been changed to call basic_version, which takes two arguments.
Older tables are set up with single arguments to the "X()" macro,
and most of them were converted. The two of the three X() calls
that weren't compiling are "fat_code2uni", and "fat_uni2code",
which are defined in "fs/fat/tables.h". the third is "fat_readdirx",
which is defined as a function in "fs/fat/dir.c", but not as a table.

This bug causes me to have to drop the FAT module from my kernel. It
won't compile.

BTW: Is there a Web page and a form I should fill out to report this?
Is this the appropriate way to report these?

-- 
Andrew C. Esh			mailto:andrew_esh@cnt.com
IRC: DrChandra
Computer Network Technology	andrewes@mtn.org (finger for PGP key)
6500 Wedgwood Road		612.550.8000 (main)
Maple Grove MN 55311		612.550.8229 (direct)
http://www.cnt.com - CNT Inc. Home Page
http://www.mtn.org/~andrewes - ACE Home Page