- Fix a core dump when smtp_policy_maps specifies an invalid TLS level.
- Fix a missing " in \%s\", in postconf(1) fatal error messages, which
violated the C language spec. Reported by Iain Hibbert.
- Stop excessive recursion in the cleanup server while recovering from a
virtual alias expansion loop. Problem found at Two Sigma.
- Stop exponential memory allocation with virtual alias expansion loops.
This came to light after fixing the previous problem.
netbsd-7 if it is configured by wdogctl=YES in rc.conf, and also
avoid a hangup duaring shutdown.
- don't put an entry which is already in the queue.
- read a first entry properly to handle the queue.
- check the lom status if input buffer is empty before writing.
- rename ext2fs_checksb() -> ext2fs_sbcheck(): more consistent
- in ext2fs_sbcheck(), add a check to ensure e2fs_inode_size!=0,
otherwise division by zero
- add ext2fs_sbcompute(), to compute dynamic values of the superblock.
It is done twice in _reload() and _mountfs(), so put it in a function.
- reorder the code in charge of loading the superblock: now, read the
superblock, swap it directly, and *then* pass it to ext2fs_sbcheck().
It is similar to what ffs now does. It is better since the fields don't
need to be swapped on the fly in ext2fs_sbcheck().
Tested on amd64.
Add security model to specify the difference between cprng_strong and
cprng_fast.
Fix code references. cprng_fast now uses ChaCha8, not RC4.
XXX Would have been nice if they had been called cprng and cprng_weak
to reduce confusion about which one to use, or even random and
weakrandom. Too late for that now, though.
Tested on O2 and Indy.
Things that are compile-tested only for lack of hardware:
- GIO ethernet cards with PCI bridges
- IP2x hardware not found on Indy
- IP1x
aarch64elf.em. Original commit message:
Recursively add DT_NEEDED entries from shared libraries if symbols are
used indirectly. This is more in line with the old GNU ld behavior, but
not exactly the desired semantic.
Patch from Martin Husemann.
checks. Compared to OpenBSD's reallocarray, makes it easier to avoid memory
leaks on allocation failures and it doesn't depend on malloc(0) != NULL
for correct error checking. Compared to plain realloc, it also avoids
the problem of intermediate integer overflows. The trade-off is the use
of void * to side step C type system with regard to generic pointer to
pointer.
