If we have multiple interrupt resources, like for Z8530 clones on the
mac-io bus, we cannot setup FAST interrupt handlers. This because we
use spinlocks to protect the hardware and all interrupt resources are
assigned the same interrupt handler. When the interrupt handler is
invoked for interrupt X, it could be preempted for interrupt Y while
it was holding the lock (where X and Y are the interrupt resources
corresponding a single instance of this driver). This is a deadlock.
By only using a MPSAFE handler in that case we prevent preemption.
Remove non-socket buffer routines from uipc_sockbuf.c, and socket buffer
specific routines from uipc_socket2.c following repo-copy. We might
rethink the location of one or two at some point, but the division was
relatively clean. uipc_sockbuf.c is now the home of routines that
manipulate socket buffers.
Rather than calling mircotime() in catchpacket(), make catchpacket()
take a timeval indicating when the packet was captured. Move
microtime() to the calling functions and grab the timestamp as soon
as we know that we're going to call catchpacket at least once.
This means that we call microtime() once per matched packet, as
opposed to once per matched packet per bpf listener. It also means
that we return the same timestamp to all bpf listeners, rather than
slightly different ones.
It would be more accurate to call microtime() even earlier for all
packets, as you have to grab (1+#listener) locks before you can
determine if the packet will be logged. You could always grab a
timestamp before the locks, but microtime() can be costly, so this
didn't seem like a good idea.
(I guess most ethernet interfaces will have a bpf listener these
days because of dhclient. That means that we could be doing two bpf
locks on most packets going through the interface.)
soreceive_generic(), and sopoll_generic(). Add new functions sosend(),
soreceive(), and sopoll(), which are wrappers for pru_sosend,
pru_soreceive, and pru_sopoll, and are now used univerally by socket
consumers rather than either directly invoking the old so*() functions
or directly invoking the protocol switch method (about an even split
prior to this commit).
This completes an architectural change that was begun in 1996 to permit
protocols to provide substitute implementations, as now used by UDP.
Consumers now uniformly invoke sosend(), soreceive(), and sopoll() to
perform these operations on sockets -- in particular, distributed file
systems and socket system calls.
sos [Mon, 24 Jul 2006 10:44:50 +0000 (10:44 +0000)]
Finally fix support for the newer MCP51/MCP55 nVidia chipsets.
The register layout has changed since the original NV4 - sigh.
Hotplug support has been fixed for all nVidia chipsets that supports it
(including the MCP51/55).
Remove old kern.malloc sysctl, which generated a text representation of
the kernel malloc(9) state for vmstat -m. libmemstat is now used to
generate a machine-readable version which is converged by vmstat -m
into a human-readable version.
Add two new unpcb flags, UNP_BINDING and UNP_CONNECTING, which will be
used to mark UNIX domain sockets as being in the process of binding or
connecting. Use these to prevent simultaneous bind or connect
operations by multiple threads or processes on the same socket at the
same time, which closes race conditions present in the UNIX domain
socket implementation since inception.
Implement support for HMAC/SHA1 and HMAC/SHA256 acceleration found in
new VIA CPUs.
For older CPUs HMAC/SHA1 and HMAC/SHA256 (and others) will still be done
in software.
Move symmetric cryptography (currently only AES-CBC 128/192/256) to
padlock_cipher.c file. Move HMAC cryptography to padlock_hash.c file.
Set ses_ictx and ses_octx to NULL after freeing them, so we won't free
them twice.
This is possible for example in situation when session is used in
authentication context, then freed and then used in encryption context
and freed - in encryption context ses_ictx and ses_octx are not touched
at newsession time, but padlock_freesession could still try to free them
when they are not NULL.
The KAME project ceased work on IPv6 and IPSec in March of 2006.
Remove the README file which warns against cosmetic or local only
changes. FreeBSD committers should now feel free to work on the
IPv6 and IPSec code without fetters. The KAME mailing lists still
exist and it is always a good idea to ask questions about this code
on the snap-users@kame.net mailing list.
Clean up the svr4 socket cache and streams code some to make it more easily
locked.
- Move all the svr4 socket cache code into svr4_socket.c, specifically
move svr4_delete_socket() over from streams.c. Make the socket cache
entry structure and svr4_head private to svr4_socket.c as a result.
- Add a mutex to protect the svr4 socket cache.
- Change svr4_find_socket() to copy the sockaddr_un struct into a
caller-supplied sockaddr_un rather than giving the caller a pointer to
our internal one. This removes the one case where code outside of
svr4_socket.c could access data in the cache.
- Add an eventhandler for process_exit and process_exec to purge the cache
of any entries for the exiting or execing process.
- Add methods to init and destroy the socket cache and call them from the
svr4 ABI module's event handler.
- Conditionally grab Giant around socreate() in streamsopen().
- Use fdclose() instead of inlining it in streamsopen() when handling
socreate() failure.
- Only allocate a stream structure and attach it to a socket in
streamsopen(). Previously, if a svr4 program performed a stream
operation on an arbitrary socket not opened via the streams device,
we would attach streams state data to it and change f_ops of the
associated struct file while it was in use. The latter was especially
not safe, and if a program wants a stream object it should open it via
the streams device anyway.
- Don't bother locking so_emuldata in the streams code now that we only
touch it right after creating a socket (in streamsopen()) or when
tearing it down when the file is closed.
- Remove D_NEEDGIANT from the streams device as it is no longer needed.
Add conditional VFS Giant locking to svr4_sys_fchroot() and mark it MPSAFE.
Also, call change_dir() instead of doing part of it inline (this now adds
a mac_check_vnode_chdir() call) to match fchdir() and call
mac_check_vnode_chroot() to match chroot(). Also, use the change_root()
function to do the actual change root to match chroot().
- Pass the MPSAFE flag to namei() in linux_uselib() and handle conditional
Giant VFS locking in that function.
- Remove bogus code to handle the case where namei() returns success but a
NULL vnode pointer.
- Note that this code duplicates exec_check_permissions() and annotate
where it differs.
- Hold the vnode lock longer to protect the write to set VV_TEXT in
v_vflag.
- Mark linux_uselib() MPSAFE.
If we get an error w/o atapi sense information, just print a newline
to terminate the message we started. I get non-terminated messages
when reading audio tracks w/o this patch.
clean-up: rewrote explore_null and explore_numeric without using sentinel.
we do not need it since we make (at most) a single addrinfo entry in these
cases.
Change semantics of socket close and detach. Add a new protocol switch
function, pru_close, to notify protocols that the file descriptor or
other consumer of a socket is closing the socket. pru_abort is now a
notification of close also, and no longer detaches. pru_detach is no
longer used to notify of close, and will be called during socket
tear-down by sofree() when all references to a socket evaporate after
an earlier call to abort or close the socket. This means detach is now
an unconditional teardown of a socket, whereas previously sockets could
persist after detach of the protocol retained a reference.
This faciliates sharing mutexes between layers of the network stack as
the mutex is required during the checking and removal of references at
the head of sofree(). With this change, pru_detach can now assume that
the mutex will no longer be required by the socket layer after
completion, whereas before this was not necessarily true.
Import from OpenBSD 1.168, dhartmei:
fix a bug in the input sanity check of DIOCCHANGERULE (not used by pfctl,
but third-party tools). a rule must have a non-empty replacement address
list when it's a translation rule but not an anchor call (i.e. "nat ...
->" needs a replacement address, but "nat-anchor ..." doesn't). the check
confused "rule is an anchor call" with "rule is defined within an anchor".
report from Michal Mertl, Max Laier.
Fix ALT_BREAK_TO_DEBUGGER on the AT91 :
The core uart code expects the receive method to actually puts the
characters read into its buffers. For AT91, it's done in the ipend routine,
so also check if we have the alternate break sequence here.
Add pmap_clear_write() to the interface between the virtual memory
system's machine-dependent and machine-independent layers. Once
pmap_clear_write() is implemented on all of our supported
architectures, I intend to replace all calls to pmap_page_protect() by
calls to pmap_clear_write(). Why? Both the use and implementation of
pmap_page_protect() in our virtual memory system has subtle errors,
specifically, the management of execute permission is broken on some
architectures. The "prot" argument to pmap_page_protect() should
behave differently from the "prot" argument to other pmap functions.
Instead of meaning, "give the specified access rights to all of the
physical page's mappings," it means "don't take away the specified
access rights from all of the physical page's mappings, but do take
away the ones that aren't specified." However, owing to our i386
legacy, i.e., no support for no-execute rights, all but one invocation
of pmap_page_protect() specifies VM_PROT_READ only, when the intent
is, in fact, to remove only write permission. Consequently, a
faithful implementation of pmap_page_protect(), e.g., ia64, would
remove execute permission as well as write permission. On the other
hand, some architectures that support execute permission have
basically ignored whether or not VM_PROT_EXECUTE is passed to
pmap_page_protect(), e.g., amd64 and sparc64. This change represents
the first step in replacing pmap_page_protect() by the less subtle
pmap_clear_write() that is already implemented on amd64, i386, and
sparc64.
- Remove hardcoded /etc/ntp.conf configuration file from ntpdate rc.d script
and replace it with a new ntpdate_config variable.
- Document it in defaults/rc.conf and rc.conf.5.
- Document ntpdate_hosts in defaults/rc.conf.
Since resetting hardware takes a very long time and results in link
renegotiation, we only initialize the hardware only when it is
absolutely required. Process SIOCGIFADDR ioctl in em(4) when we know
an IPv4 address is added. Handling SIOCGIFADDR in a driver is
layering violation but it seems that there is no easy way without
rewritting hardware initialization code to reduce settle time after
reset.
This should fix a long standing bug which didn't send ARP packet when
interface address is changed or an alias address is added. Another
effect of this fix is it doesn't need additional delays anymore when
adding an alias address to the interface.
While I'm here add a new if_flags into softc which remembers current
prgroammed interface flags and make use of it when we have to program
promiscuous mode.
Make svr4_sys_waitsys() a lot less ugly and mark it MPSAFE.
- If the WNOWAIT flag isn't specified and either of WEXITED or WTRAPPED is
set, then just call kern_wait() and let it do all the work. This means
that this function no longer has to duplicate the work to teardown
zombies that is done in kern_wait(). Instead, if the above conditions
aren't true, then it uses a simpler loop to implement WNOWAIT and/or
tracing for only stopped or continued processes. This function still
has to duplicate code from kern_wait() for the latter two cases, but
those are much simpler.
- Sync the code to handle the WCONTINUED and WSTOPPED cases with the
equivalent code in kern_wait().
- Fix several places that would return with the proctree lock still held.
- Lock the current process to prevent lost wakeup races when blocking.
Add a mutex to protect the list of interrupt config hooks. We do assume
that the only remove hook operation that can occur while processing the
hooks is to remove the currently executing hook. This should be safe as
the existing code has assumed this already for a long time now.
Don't free the sockaddr in kern_bind() and kern_connect() as not all
callers pass a sockaddr allocated via malloc() from M_SONAME anymore.
Instead, free it in the callers when necessary.