| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
page rounding! Disgusting. Instead, define local copies of those
macros.
|
|
won't be pulling in the uvm side of the kitchen.
|
|
this inside #ifdef _KERNEL in any case, so nothing really changes.
|
|
|
|
uvm_extern.h, uvm_param.h, and sysctl.h will no longer gaurantee that
_KERNEL code gets it.
|
|
pulls in the universe). occasionally they need sys/systm.h
|
|
|
|
|
|
ok mpi
|
|
via if_rxring things. this effectively deprecates the third argument
for MCLGETI and m_clget and makes the mbuf layer no longer care about
interfaces and simplifies the allocation paths.
the timeout used to measure livelock has been moved to net/if.c.
ok mpi@
|
|
to using if_rxr.
cut the reporting systat did over to the rxr ioctl.
tested as much as i can on alpha, amd64, and sparc64.
mpi@ has run it on macppc.
ok mpi@
|
|
out of the mbuf layer, and break the assumption that an interface will
only have a single ring per mbuf cluster size.
mpi@ is ok with moving this forward
|
|
argument to MCLGETI to make that clear.
|
|
rx descriptor.
slightly tweak by matthew
tested on alpha and amd64
|
|
right value out of the softc. then its assigned the same value again
after the rest of the var decls.
well, it used to be. not after this commit.
|
|
set. em_rxeof doesnt have to check that flag again.
|
|
entire tx path. if we try to bus_dmamap_load a very fragmented packet
m_defrag it and try again.
this is just like if_bge.c r1.355.
|
|
|
|
A little while back, guenther moved user credentials to a per-process
property, but also kept a per-thread cache that would remain stable
for the duration of a thread's system call. These system calls now
only access the thread's cached credentials, so they're safe to run
without the kernel lock.
ok kettenis, guenther
|
|
unconditionaly instead of only doing it if invoked indirectly through syscall
or __syscall.
While there, tweak stack argument fetching to only perform one large copyin()
call, rather than a loop of small ones.
ok kettenis@
|
|
request and ok deraadt
|
|
Now that file descriptor tables have moved from a per-thread resource
to per-process, it's impossible to fork a thread without sharing the
file descriptor table.
ok guenther
|
|
|
|
print a message and ignore the interrupt. There are BIOSen out there with
random garbage in NMI entries for the non-BP CPUs.
|
|
ok matthew@
|
|
(found the hard way by building with -fstack-shuffle)
|
|
From Jerome Glisse
16d30071c6cd4b72a950d265b0d7d08fb4220b92 in ubuntu 3.8
24f47acc78b0ab5e2201f859fe1f693ae90c7c83 in mainline linux
|
|
From Alex Deucher
6a6170926060797bdf25e5d6609a3168466f2e6c in ubuntu 3.8
d8ade3526b2aa0505132c404c05a38b73ea15490 in mainline linux
|
|
From Christian Koenig
72568eb7e4e40096fd80c7c8abd04ac630e45bb9 in ubuntu 3.8
544092596e8ac269f70e70961b5e9381909c9b1e in mainline linux
|
|
http://mail-index.netbsd.org/source-changes/2009/10/18/msg002090.html
|
|
such that curcpu() works. The extent code creates pools, and if we want to
add any sort of locking to the pool code we need that working. Found out the
hard way by guenther@.
ok dlg@, guenther@
|
|
PS_{ZOMBIE,EMBRYO} on the process instead of peeking into the process's
thread data. This eliminates the need for the thread-level SDEAD state.
Change kvm_getprocs() (both the sysctl() and kvm backends) to report the
"most active" scheduler state for the process's threads.
tweaks kettenis@
feedback and ok matthew@
|
|
ok mlarkin
|
|
ok jsg@
|
|
in bge_encap to fail because the dmamap lacks space, not necessarily
because the ring is full. however, bge_encap failure sets the OACTIVE
flag on the interface and keeps the packet at the start of the send
queue.
the next time we try to fill the tx ring we'll try to load the same
packet and fail. an empty tx ring means bge_txeof hasnt got anything
which is where the OACTIVE condition is cleared.
this diff adds handling of fragmented packets via m_defrag.
this might fix the issues landry@ has been complaining about on his
bulk build machines. i can reproduce the above problem in contrived
circumstances here and this diff fixes it, so its going in so landry@
is forced to test it.
|
|
wakeup and clearing the PG_BUSY and PG_WANTED flags, so try to keep those
bits as close together and defenitely avoid calling random code in between.
ok guenther@, tedu@
|
|
used on some archs.
|
|
It will no longer be pulled by uvm_extern.h in the short future.
ok jsg@
|
|
precise exceptions, as the actual data the FPU gives us is a 35-bit number,
with the exponent sign-extended to the double-precision exponent width.
Make sure we shrink it and fetch the remaining three low bits of mantissa from
the LS registers.
|
|
is `not comparable'. gcc relies upon `not equal' being set when comparing
numbers to infinities.
|
|
ptrace operation on MP kernels.
|
|
actually part of the GIO address range; these obviously come from speculative
R10000 execution and are completely harmless (except for the need to acknowledge
them, which we have to do anyway).
|
|
mode and reception of multicast traffic.
ok matthieu@ "looks good to me" rapha@
|
|
Theo approves, ok mpi@
|
|
ok mpi@ deraadt@
|
|
|
|
|
|
|
