| Age | Commit message (Collapse) | Author |
|---|
|
We only need the dev_t and not the entire struct stat in mntinfo.
Delete some superfluous casts.
|
|
Debug message consistency: format opts with %#x and modes with %#04o
|
|
If the network is unreachable when ntpd starts and host_dns fails, be sure
that we still close the HOST_DNS imsg.
Thanks to Paul de Weerd <weerd at weirdnet dot nl> for reporting this.
ok beck@
|
|
|
|
for a couple decades. Keep the OSIOCGIFCONF ioctl to support COMPAT_LINUX
but move the rest of the Linux-specific ioctl() handling into linux_socket.c
This lets struct osockaddr finally move from sys/socket.h to protocols/talkd.h
ok krw@ deraadt@ mpi@
|
|
|
|
and reduces groff/mandoc differences in base by about 1%
|
|
There were four bugs fixed by this patch:
* dtls1_buffer_record() now frees rdata->rbuf.buf on error. Since
s->s3->rbuf was memset, rdata->rbuf is the only pointer left which
points to the old rbuf. On error, rdata is freed so there will not
be any way of freeing this memory unless we do it here.
* Changed the return code of dtls1_buffer_record() to differentiate
between queue full (0) and error (-1). See below as this differs
from upstream.
* Handle errors if calls to dtls1_buffer_record() fail with -1.
Previously, it did not check the return value.
* Changed the way receipts are recorded. Previously, it was recorded
when processed successfully (whether buffered or not) in
dtls1_process_record(). Now, it records when it is handled in
dtls1_get_record(): either when it is entered into the queue to buffer
for the next epoch or when it is processed directly. Processing
buffered records does not add a receipt because it needed one in
order to get into the queue.
The above bugs combined contributed to an eventual DoS through memory
exhaustion. The memory leak came from dtls1_buffer_record()'s error
handling. The error handling can be triggered by a duplicate record
or malloc failure. It was possible to add duplicate records because
they were not being dropped. The faulty receipts logic did not detect
replays when dealing with records for the next epoch. Additionally,
dtls1_buffer_record()'s return value was not checked so an attacker
could send repeated replay records for the next epoch.
Reported to OpenSSL by Chris Mueller.
Patch based on OpenSSL commit 103b171d8fc282ef435f8de9afbf7782e312961f
and BoringSSL commit 44e2709cd65fbd2172b9516c79e56f1875f60300.
Our patch matches BoringSSL's commit. OpenSSL returns 0 when the queue
is full or when malloc() or pitem_new() fails. They return -1 on error
including !ssl3_setup_buffers() which is another failure to allocate
memory.
BoringSSL and LibreSSL changed the return code for dtls1_buffer_record()
to be 1 on success, 0 when the queue is full and -1 on error.
input + ok bcook@, jsing@
|
|
ok djm markus
|
|
work
|
|
|
|
OK jmc@
|
|
but safer not to wraparound by accident.
ok deraadt millert miod
|
|
|
|
using mandoc is better than using groff) and -Wunsupp (manual using
unsupported low-level roff(7) feature, probably using groff is better
than using mandoc). Once this feature is complete, it is intended
to help porting, making the decision whether to USE_GROFF easier.
As a first step, distinguish four classes of roff(7) requests:
1. Supported (currently 24 requests)
2. Currently ignored because unimportant (120) -> no message
3. Ignored for good because insecure (14) -> -Werror
4. Currently unsupported (68) -> these trigger the new -Wunsupp messages
|
|
|
|
|
|
so that the remaining information is more visible
ok kettenis miod
|
|
per-process value, and therefpore turns the VM_PSSTRINGS sysctl into a
per-process one as well. This gets rid of a pointer to the bottom of the
stack at a fixed location. Also clears the road for unmapping the stackgap.
ok deraadt@
|
|
|
|
|
|
directory is not contained in OpenBSD base, and because even people
having the directory often don't understand that they need to run
makewhatis(8) - and instead complain about the resulting warnings.
This commit reverts revisions 1.17 and 1.21.
Requested by deraadt@ millert@ kettenis@ who argue that people
using /usr/ports/infrastructure/bin/ already need to set PATH,
so editing man.conf (or, though more fragile, setting MANPATH)
should not be a big deal for them.
|
|
the name of the macro being described, and look(1).
ok jmc
|
|
been one char too big when it was defined in terms of MAXHOSTNAMELEN.
While here, NI_MAXHOST is big enough for getnameinfo(3).
OK benno@
|
|
This brings bus_space_mmap(9) to socppc and change its bus_space_map(9)
implementation to use kernel_map instead of phys_map like macppc and
everybody else.
|
|
delete <sys/param.h> if now possible
ok guenther
|
|
delete <sys/param.h> if now possible
ok guenther
|
|
Do not append an SQL clause looking into the large "keys" table.
Instead, filter the result of the SQL query in buildnames() where
equivalent data from the much smaller "mlinks" table is already
available for free.
This is relevant because man(1) uses the equivalent of "-S ${MACHINE}"
by default since main.c rev. 1.122, to make sure that manuals for
the current architecture are shown. With many ports installed, this
patch can speed up man(1) by a factor of more than a hundred.
Slowness reported by Theo Buehler <theo at math dot ethz dot ch>, thanks!
|
|
FSCALE still needs to be exported .. unless someone fixes a few ports
that need it.
ok guenther, tested by sthen
|
|
unused typedef & external definitions.
|
|
memory for the mbuf layer up front. These days it only matters for
sizing of macros in <sys/mbuf.h>, so move it there. MCLSHIFT, MCLBYTES,
and MCLOFSET can move also (a decade ago, architectures had different
mbuf sizes. you don't want to know more)
ok guenther, ports fallout checked for by sthen
|
|
int to an aligned variable before operating on it.
ok claudio
|
|
used in disklabel type tools, newfs, fsck, when related to ufs or
ufs-derived filesystems, but also in a struct. Those things always include
<ufs/ffs/fs.h>. Move it there
ok guenther, tested in ports by sthen
|
|
each, never in portable software. Mop it up.
ok guenther, tested in ports tree by sthen
|
|
each .c file, and out of the .h files. To avoid overinclude.
ok gilles, in principle. If this has been done right, -portable should
become easier to maintain.
|
|
Analysis and patch by Richard Kojedzinszky, thanks! ok henning
|
|
ok mikeb
|
|
ok kettenis@
|
|
for the kernel pmap and kill pmap_kremove_pg(). Finally guard the hash
lock code under "MULTIPROCESSOR" to explicit which part of the code
received some MP love.
ok kettenis@
|
|
sendmail flags on the command line.
Also allow "from" to be set in mailrc.
Use sendmail's "-t" flag when executing sendmail instead of specifying
the list of recipients in argv. The "-f" flag will be used to set
the from address if specified.
|
|
realloc() which is expensive for large blacklists.
|
|
|
|
Suggested by deraadt@
|
|
Suggested by deraadt@
|
|
Suggested by deraadt@
|
|
hardware is fairly retarded. While it allows receive buffers with an
ETHER_ALIGN offset, it only allows the size of the buffers to be specified in
multiples of 1K. This means that if we want to use standard mbuf clusters
we will waste 1024 - ETHER_ALIGN bytes per cluster, which is a lot for the
2K clusters we use now. Compromise a bit by using 4K clusters on strict
alignment architectures and tell the hardware to use 3K of those, reducing
the spillage a bit. While this isn't optimal, at least on sparc64 where we
have 8K pages, the pool page allocation overhead should be the same as on
amd64/i386 where we have 4K pages and continue to use 2K mbuf clusters.
ok mikeb@, dlg@
|
|
OK jmc@ guenther@
|
|
|
|
Fix from Alexey Suslikov <alexey.suslikov at gmail.com>. Thanks!
tested by many, ok armani
|
|
rdist/client.h and rdistd/server.h
Only put #includes in .h files that are necessary for the .h to be used;
all other #includes go in the .c files
Move all extern variable declarations to the .h files, renaming local
variables to avoid shadowing
Replace me_type member of mntent_t with an "is NFS?" flag bit
|
