Also, most device functions related to IRQ handling no longer take a device
pointer. We make so many assumptions about the machine's layout in irq.c that
this just seemed silly.
The architecture here follows that of the vax port -- each device has its
evcnt in its softc, but defers actually incrementing it to the IRQ
dispatcher. This way, devices can attach sub-counts (e.g. Rx and Tx counts
for Ethernet interfaces), but don't all have to have code to increment the
counters.
Drivers deliberately call evcnt_attach_dynamic() before establishing their
interrupt handler so that the establish routine can attach a parent event if
that's appropriate. At present, it isn't.
The architecture here follows that of the vax port -- each device has its
evcnt in its softc, but defers actually incrementing it to the IRQ
dispatcher. This way, devices can attach sub-counts (e.g. Rx and Tx counts
for Ethernet interfaces), but don't all have to have code to increment the
counters.
Drivers deliberately call evcnt_attach_dynamic() before establishing their
interrupt handler so that the establish routine can attach a parent event if
that's appropriate. At present, it isn't.
This is not unprecedented, as we do it in >100 places in the tree.
If you disagree with this philosophy, take it to tech-kern for a discussion
FIRST before reverting; TNF, not one particular person, owns this file.
amap_free(): Assert that the amap is locked.
amap_share_protect(): Assert that the amap is locked.
amap_wipeout(): Assert that the amap is locked.
uvm_anfree(): Assert that the anon has a reference count of 0 and is
not locked.
uvm_anon_lockloanpg(): Assert that the anon is locked.
anon_pagein(): Assert that the anon is locked.
uvmfault_anonget(): Assert that the anon is locked.
uvm_pagealloc_strat(): Assert that the uobj or the anon is locked
And fix the problems these have uncovered:
amap_cow_now(): Lock the new anon after allocating it, and unref and
unlock it (rather than lock!) before freeing it in case
of an error condition. This should fix a problem reported
by Dan Carosone using cdrecord on an i386 MP kernel.
uvm_fault(): Case1B -- Lock the new anon afer allocating it, and unlock
it later when we unlock the old anon.
Case2 -- Lock the new anon after allocating it, and unlock
it later by passing it to uvmfault_unlockall() (we set anon
to NULL if we're not doing a promote fault).