Tor created a while ago, removes the raw I/O piece (that has cache coherency
problems), and adds a buffer cache / VM freeing piece.
Essentially this patch causes O_DIRECT I/O to not be left in the cache, but
does not prevent it from going through the cache, hence the 80%. For
the last 20% we need a method by which the I/O can be issued directly to
buffer supplied by the user process and bypass the buffer cache entirely,
but still maintain cache coherency.
I also have the code working under -stable but the changes made to sys/file.h
may not be MFCable, so an MFC is not on the table yet.
Submitted by: tegge, dillon
o If the class is PCIC_BRIDGE, subclass is PCIS_BRIDGE_PCMCIA and
programming interface is 0, assume that it is a generic PCMCIA PCI
chip we can program. I don't think there are any of these that
we don't know about, but you never know.
o If the class is PCIC_BRIDGE, subclass is PCIS_BRIDGE_CARDBUS and
programming interface is 0, assume that it is a YENTA cardbus bridge
that we know how to cope with. There are likely some cardbus bridges
that haven't it made it in here yet.
pcic_{get,put}b_io. There are some pci bridges (the CL-PD6729 and
maybe others) that do not have memory mapped registers, so we'll need
these in both places. Declare them in pcicvar.h.
- Assert Giant in vm_pageout_scan() for the vnode hacking that it does.
- Don't hold vm_mtx around vget() or vput().
- Lock Giant when calling vm_pageout_scan() from the pagedaemon. Also,
lock curproc while setting the P_BUFEXHAUST flag.
- For now we still hold Giant for all of the vm_daemon. When process
limits are locked we will be only need Giant for swapout_procs().
- Restore the previous order of setting up a new vm_object. The previous
had a small bug where we zero'd out the flags after we set the
OBJ_ONEMAPPING flag.
- Add several asserts of vm_mtx.
- Assert Giant is held rather than locking and unlocking it in a few
places.
- Add in some #ifdef objlocks code to lock individual vm objects when
vm objects each have their own lock someday.
- Don't bother acquiring the allproc lock for a ddb command. If DDB
blocked on the lock, that would be worse than having an inconsistent
allproc list.
- Add a few KTR tracepoints to track the addition and removal of
vm_map_entry's and the creation adn free'ing of vmspace's.
- Adjust a few portions of code so that we update the process' vmspace
pointer to its new vmspace before freeing the old vmspace.
- Don't lock Giant in the scheduler() function except for when calling
faultin().
- In swapout_procs(), lock the VM before the proccess to avoid a lock order
violation.
- In swapout_procs(), release the allproc lock before calling swapout().
We restart the process scan after swapping out a process.
- In swapout_procs(), un #if 0 the code to bump the vmspace reference count
and lock the process' vm structures. This bug was introduced by me and
could result in the vmspace being free'd out from under a running
process.
- Fix an old bug where the vmspace reference count was not free'd if we
failed the swap_idle_threshold2 test.
acquired.
- Assert Giant is held in the strategy, getpages, and putpages methods and
the getchainbuf, flushchainbuf, and waitchainbuf functions.
- Always call flushchainbuf() w/o the VM lock.
- Always call vfs_setdirty() with vm_mtx held.
- Fix an old comment: vm_hold_unload_pages is called vm_hold_free_pages()
nowadays.
- Always call vm_hold_free_pages() w/o vm_mtx held.
vnodes.
- Fix an old bug that would leak a reference to a fd if the vnode being
mmap'd wasn't of type VREG or VCHR.
- Lock Giant in vm_mmap() around calls into the VM that can call into
pager routines that need Giant or into other VM routines that need
Giant.
- Replace code that used a goto to jump around the else branch of a test
to use an else branch instead.
the built-in 1000baseX interface in the Level 1 LXT1001 chip. The Level 1
PHY comes up with the isolate bit in the control register set by default,
but it also has the autonegotiate bit set. When you tell the xmphy driver
to select IFM_AUTO mode, it sees that the autoneg bit is already on, and
thus doesn't bother updating the control register. However this means that
the isolate bit is never turned off (unless you manually select 1000baseSX
full or half duplex mode, which does result in the control register being
modified and the ISO bit being turned off).
This subtle and unusual behavioral difference stopped me from being able
to receive packets on the SMC9462TX card for several days, since isolating
the PHY disconnects it from the MAC's data interface. The fix is to omit
the 'is the autoneg big set?' test, since it doesn't really provide much
of an optimization anyway.
This commit also updates the xmphy driver to support the Jato/Level 1
internal PHY. (I'm not sure how Jato Technologies is related to Level 1:
all I know is the OUI from the PHY ID registers maps to Jato in the OUI
database.) This will be used once I add the if_lge driver to support
the LXT10010 chip.
- Don't bother releasing Giant while doing a lookup on the vm_map of
initproc while starting up init. We have to grab it again right after
the lookup anyways.
the chipset. This is already how the multi-hose systems handle resource
allocation and it fixes a bug where dense and bwx memory allocations were
not handled properly.
Reviewed by: gallatin
