behaviour of splhigh() was changed. The new behaviour raised the level to
IPL_HIGH if it was lower, but left it alone if the same or higher. Apparently
the firmware transfers to NetBSD with all interrupts blocked, or something
else blocks all interrupts (including machine checks) before the code
is really started. This meant that machine checks were blocked during
device probes, leading to false device probes _and_ an unexpected (and
therefore fatal) machine check at the spl0() after autoconfiguration.
"smallest amount to put in cluster", and (MHLEN + MLEN) is the largest amount
amount of data that can be stored without clustering.
Pointed out by Thorsten Frueauf <frueauf@ira.uka.de> with W. R. Stevens'
TCP/IP Illustrated, Vol. 2, at hand.
(1) it was using 'max', and some functions use a variable
of that name (*sigh*), and
(2) that makes it easier to be a bit trickier, and only call
swpipl if changing the IPL.
Un-munge pkg_install/lib/Makefile which I'd overwritten on libftpio commit.
Make it compile on NetBSD.
Install pkg_* programs in /usr/sbin
XXX agc - Still to do XXX
Cleanup call to mktemp (for a tmp dir) in pkg_install/lib/pen.c
Create ftperr.c in the correct directory.
filesystem. This uses fts(3) to access the directory structure (and
not the raw device), so the standard access permissions are adhered
to (unlike dumping an entire filesystem, which just requires read
access to the raw disk device).
* Support SIGINFO status reporting.
* Remove now unused variables that previously stored the (e)uid.
* Be more informative in a couple of error messages.
- Add NetBSD autoconfiguration support.
- Rearrange code slightly to minimize the number of #ifdefs.
- Don't use a structure to access CSRs. Use macros that DTRT for
the NetBSD and FreeBSD cases.
- Deal with alignment contraint on Alpha - add 2-byte padding at the
beginning of the RFA, so that the data will be 4-byte aligned, after
the 14-byte Ethernet header.
Thanks to Matthias Drochner for the testing, and David Greenman for
the feedback on the changes.
kernels, at the same time getting rid of up to 3 conditional branches and a
bit over one cacheline fetch (for the 68060; the saving is a bit smaller for
040 and yet smaller for the 020/30).
While we're here, also get rid of an redundant lea (using SP-relative
addressing) and of two redundant pushes.
While we're here, also fix a panic which would tear us down on 68060 machines
if a branch prediction error ever occured.
mellon