| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
data in it. m_getptr() hops over empty buffers and points to the first
allocated data byte. Because of this the m_dup_pkthdr() call done by
m_copym0() can panic because not the first mbuf is passed.
Found the hard way by myself, diff by blambert@ commiitting for him since
he is not around. Tested and OK myself
|
|
* Comment why we call the video interrupt in the mmap interface now in
uvideo.c
|
|
for both structs, the new members are 'bps' and 'msb', which
describe the number of bytes per sample and data alignment in the
sample, respectively. drivers must properly set these fields in
the 'query_encoding', 'set_parameters' and 'get_default_params'
hardware interface methods.
discussed with ratchov, deraadt
|
|
is a variable on sparc. This should be revisited... after the arguments
for pagesize vs 4K complete :)
|
|
go back to something more like the previous design, and have the thread do
the heavy lifting. solves vmmaplk panics.
ok deraadt oga thib
[and even simple diffs are hard to get perfect. help from mdempsky and deraadt]
|
|
ok mglocker
|
|
Learned the hard way. dlg@ confirmed that it is save to just remove them,
the desync will still work but the reader needs to dequeue all packets first.
|
|
no point in copying before having done the final sanity check (we copied
before the last one).
ok armani
|
|
get rid of it.
ok armani
|
|
From FreeBSD via Brad.
|
|
|
|
mysterious and unused mbtypes[] array in mbuf.h.
ok tedu@, deraadt@
|
|
By trying to do the latter, ioprbs(4) was inadvertently always
signalling to the SCSI mid and upper layers that read/write requests
succeeded, even if they failed or only partially completed.
Light testing by claudio@, which is still way more effort than this
driver deserved.
|
|
condition that prevents us from queuing it, which in turn means that
processes listening on the routing socket for changes to the kernel state
will get out of sync. currently this is handled by the following comment:
/* should notify about lost packet */
this change introduces a new rtsock message called RTM_DESYNC that
notifies about lost packets and uses it instead of this comment.
when we detect loss we flush all the message in the routing socket and
attempt to queue an RTM_DESYNC message instead. to guarantee that we will
enqueue DESYNC we keep trying it when an attempt to enqueue or dequeue
any messages is made, and in the worst case a timeout tries to guarantee
that desync is added to the socket.
ive been running this in production for 2 or 3 weeks.
tested by sthen@
ok sthen@ claudio@ deraadt@
code written by andrew sallaway at the univeristy of queensland.
|
|
|
|
From SAITOH Masanobu.
Confirmed to fix pr 6427, and should fix 6301 as well.
Tested by a bunch of people, and discussed with claudio
during c2k10.
|
|
|
|
|
|
|
|
later, when we trust it)
ok matthew
|
|
|
|
pools, sized by powers of 2, which are constrained to dma memory.
ok matthew tedu thib
|
|
ok krw@, claudio@
|
|
+ find.fp_wsize *= find.fp_mss;
From: Ed Schouten <ed at FreeBSD.org>
|
|
Tested on Thinkpad T510.
|
|
|
|
|
|
ok claudio@
|
|
grabbing allproclk in proc_zap(); skip such processes in sysctl(KERN_PROC*)
and handle the NULL pointer in ddb's ps.
ok tedu@
|
|
Although with aditional 20us delays added just before returning
from miibus read/writes as required by the hardware.
Based on the Linux r8169 driver; from Brad.
|
|
without modifying the u-boot environment.
|
|
|
|
sun4c, as it has a sun4c OpenPROM but a sun4 8KB pagesize. VME devices are
not supported yet.
ok deraadt@
|
|
|
|
|
|
addresses line up correctly. Newer i386 gcc4 - for some reason -- is
prompting the linker with -N to not do this alignment which it used to do
(perhaps because the alignment of all .o's have dropped to very small
amounts??). Using -Z wastes around a page of bootblock space which matters
even more on the small install media, so this definately needs revisiting.
Workaround from toby
|
|
|
|
Works on Dell x9xx and Studio 1555
|
|
their own buf to physio(9).
compiler tested by nick@ and miod@; ok miod@
general approval thib@, deraadt@
|
|
bytes worth of data. Also, pass M_CANFAIL to malloc(9) and handle
allocation failures by returning ENOMEM.
Prevents user space from causing a kernel panic by passing
ridiculously large datalen values.
ok dlg@, krw@
|
|
it basically has the poll code swap the ccb completion path out for
something that clears the flag the poll code is looping on. it just
runs the interrupt handler to try and get that ccb back off the
hardware.
the reply queue is now protected by a semaphore, meaning that if
one cpu is already pulling stuff off the chip, another cpu may try
to also pull stuff off, but will return immediately cos the first
cpu will end up doing the work it wants anyway.
i did it this way so the ccb completion paths wouldnt be run while
a mutex was held, and to avoid reordering of ccb completions and
updates to the reply queue consumer that could be caused by allowing
concurrent runs of the interrupt handler to step on each other.
there is a big change here in that polling doesnt time out anymore.
it keeps trying until it succeeds, or it keeps trying. i think it
is dangerous to complete a ccb until th echip has returned it.
the only safe alternative is to do a full reset of the chip, but
that is a massive amount of work.
reviewed by mikeb
ok krw@
testing by me on real and logical disks
|
|
|
|
registers in the prologue to realign the stack.
ok marco@, deraadt@, guenther@
|
|
should be 100% identical.
ok toby
|
|
This allows to run isakmpd/iked/ipsecctl in multiple rdomains
independently (with "route exec"); the kernel will pickup the rdomain
from the process context of the pfkey socket and load the flows and
SAs into the matching rdomain encap routing table. The network stack
also needs to pass the rdomain to the ipsec stack to lookup the
correct rdomain that belongs to an interface/mbuf/... You can now run
individual IPsec configs per rdomain or create IPsec VPNs between
multiple rdomains on the same machine ;). Note that a primary enc(4)
in addition to enc0 interface is required per rdomain, eg. enc1 rdomain 1.
Test by some people, mostly on existing "rdomain 0" setups. Was in
snaps for some days and people didn't complain.
ok claudio@ naddy@
|
|
the callback functions. This fixes a problem where dynamic routes in
different tables would not get deleted because the callback was doing
the remove on the wrong table.
OK henning@
|
|
actually have a protocol/address family set. Messages like RTM_IFINFO or
RTM_IFANNOUNCE are family independent and should not be filtered but
it is possible to use a ROUTE_MSGFILTER to filter these messages.
This allows to reduce the messages sent to AF_INET or AF_INET6 only daemons.
OK henning@, deraadt@
|
|
sending a bulk update, only say we're using the states referenced
by the next and last pointers into the table. this means the pf
timeout thread can keep trimming states during a bulk update.
|
