| Age | Commit message (Collapse) | Author |
|---|
|
test and ok gonzalo
|
|
PMAP_CANFAIL flag is set. Return ENOMEM instead.
|
|
if the setting mode is failed.
reported and tested by Joe Gidi
|
|
|
|
ok mlarkin@
|
|
last VM is terminated. This allows to remove the explicit "vmm
enable" / "vmm disable" (VMM_IOC_START / VMM_IOC_STOP) ioctls. You'll
have to update kernel and userland for this change, as the kernel ABI
changes.
OK mpi@ mlarkin@
|
|
|
|
|
|
|
|
|
|
there are two things shared between the network stack and drivers
in the send path: the send queue and the IFF_OACTIVE flag. the send
queue is now protected by a mutex. this diff makes the oactive
functionality mpsafe too.
IFF_OACTIVE is part of if_flags. there are two problems with that.
firstly, if_flags is a short and we dont have any MI atomic operations
to manipulate a short. secondly, while we could make the IFF_OACTIVE
operates mpsafe, all changes to other flags would have to be made
safe at the same time, otherwise a read-modify-write cycle on their
updates could clobber the oactive change.
instead, this moves the oactive mark into struct ifqueue and provides
an API for changing it. there's ifq_set_oactive, ifq_clr_oactive,
and ifq_is_oactive. these are modelled on ifsq_set_oactive,
ifsq_clr_oactive, and ifsq_is_oactive in dragonflybsd.
this diff includes changes to all the drivers manipulating IFF_OACTIVE
to now use the ifsq_{set,clr_is}_oactive API too.
ok kettenis@ mpi@ jmatthew@ deraadt@
|
|
|
|
|
|
reported by Stefan Kempf with supplied patch, thanks.
|
|
ok mlarkin
|
|
|
|
|
|
|
|
Incoherency brought by jmc@ and schwarze@ because the SYNOPSIS of the
corresponding manual do not match its NAME.
This is one step further in the zs(4) unification... Go dokey go!
Tested by jmatthew@, ok sebastia@, deraadt@
|
|
Return EAGAIN in this case:
# vmmctl -d
vmmctl: disable VMM command failed (35) - Resource temporarily unavailable
ok mlarkin@
|
|
Currently one MUST do "vmmctl -e" before creating a vm with "vmm -S ...".
Lately this could be done automagically by vmd(8) but the kernel should
not allow things that wont fly.
While here, disable vmm mode in error path if at least one of the CPUs
failed to enable it.
ok mlarkin@
|
|
ok mlarkin@
|
|
|
|
found by deraadt@
|
|
link is down.
ok mpi@
|
|
the code is refactored so the IFQ macros call newly implemented ifq
functions. the ifq code is split so each discipline (priq and hfsc
in our case) is an opaque set of operations that the common ifq
code can call. the common code does the locking, accounting (ifq_len
manipulation), and freeing of the mbuf if the disciplines enqueue
function rejects it. theyre kind of like bufqs in the block layer
with their fifo and nscan disciplines.
the new api also supports atomic switching of disciplines at runtime.
the hfsc setup in pf_ioctl.c has been tweaked to build a complete
hfsc_if structure which it attaches to the send queue in a single
operation, rather than attaching to the interface up front and
building up a list of queues.
the send queue is now mutexed, which raises the expectation that
packets can be enqueued or purged on one cpu while another cpu is
dequeueing them in a driver for transmission. a lot of drivers use
IFQ_POLL to peek at an mbuf and attempt to fit it on the ring before
committing to it with a later IFQ_DEQUEUE operation. if the mbuf
gets freed in between the POLL and DEQUEUE operations, fireworks
will ensue.
to avoid this, the ifq api introduces ifq_deq_begin, ifq_deq_rollback,
and ifq_deq_commit. ifq_deq_begin allows a driver to take the ifq
mutex and get a reference to the mbuf they wish to try and tx. if
there's space, they can ifq_deq_commit it to remove the mbuf and
release the mutex. if there's no space, ifq_deq_rollback simply
releases the mutex. this api was developed to make updating the
drivers using IFQ_POLL easy, instead of having to do significant
semantic changes to avoid POLL that we cannot test on all the
hardware.
the common code has been tested pretty hard, and all the driver
modifications are straightforward except for de(4). if that breaks
it can be dealt with later.
ok mpi@ jmatthew@
|
|
the resulting code a little.
|
|
reception of short non-IP packets which I broke in revision 1.28.
|
|
and convert all gets() users.
ok deraadt@
|
|
- Add $OpenBSD$ tag,
- constify "struct cfattach",
- Use <uvm/uvm_extern.h> rather than <uvm/uvm.h>, it's enough.
- Keep the "struct vm" private. This allows us to not pull <uvm/uvm_extern.h>
in <macine/vmmvar.h>
- Prefer DPRINTF() for debug macro as dprintf(3) is a standard function name.
- Add vmm_debug and fix VMM_DEBUG build
- Remove unneeded <sys/rwlock.h> from <machine/vmmvar.h>
- Kill whitespaces
ok mlarkin@
|
|
|
|
followed by an isync.
ok mpi@
|
|
Now that "struct bpf_d" depends on <sys/srp.h> this is one of the offender
for removing the header from <sys/param.h>
|
|
|
|
From Stefan Kempf sn.kempf at t-online.de
|
|
ok mpi@
|
|
circulated on hackers@, no objections. Disabled by default.
|
|
ok mpi@
|
|
IFQ_POLL(&ifp->if_snd, m);
if (m == NULL)
return;
IFQ_DEQUEUE(&ifp->if_snd, m);
is the same as
IFQ_DEQUEUE(&ifp->if_snd, m);
if (m == NULL)
return;
ok mpi@
|
|
of <uvm/uvm_pmap.h>.
|
|
track type of pmap and various conversion and pte bit manipulation
functions for EPT.
|
|
problem reported by semarie@
|
|
during locore, information based on this will be returned to guest VMs
issuing cpuid instructions later, under certain circumstances.
|
|
particular don't attempt to calcuate the header checksum using a
size that will cause a crash.
Tested & ok yasuoka@
|
|
changes for vmm.
|
|
|
|
|
|
|
|
This may fixes the issue when loading a compressed kernel on macbook
reported by gonzalo, jung, Joel Roberts, Bryan Vyhmeist and many.
|
|
IF_PREPEND only checks priq state, so it wont work out well if
you're using hfsc.
tested by miod@
ok mpi@ miod@
|
