so the last change has us comparing pages to bytes instead of pages
to buffers! The consequence was to try to free radically less memory
than UVM wanted us to -- though always at least one buffer, which is
probably why the results weren't dire.
This does suggest that buf_canrelease() could be a *lot* more
conservative about how much to release than "2 * page deficit". In
fact, serious trouble seems to ensue if it's not -- when anything
else on the system demands enough pages, we slam down to the low
water mark nd stay there. I've adjusted it to use min(page defecit,
buffer memory / 16), which still isn't quite right but seems better.
Another change: consider the case of an infinite loop that does
"tar xzf pkgsrc.tar.gz ; rm -rf pkgsrc". Each time the rm runs,
all the dead metadata will go on the AGE list -- and, until we hit
the high-water mark, stay there, at which point it may be slowly
recycled. Two adjustments seem to solve this: 1) whack buf_lotsfree()
to return 0 if there's anything on the AGE list; 2) whack buf_canrelease()
to count the memory used by the AGE list and always return at least
that much.
This basically turns the AGE list into a "delayed free" list, since we
can't entirely eliminate it as we can't free pool items from interrupt
context (e.g. from biodone()).
To consider: with the bookkeeping corrected, should buf_drain() move
back to the _end_ of the pagedaemon, and should the calculation then
try to give back at least the current defecit?
- Disable symbol versioning (for now)
- Make sure that libiberty knows its being configured with a
cross compiler.
- The CXX_* variables are no longer needed/used.
- LIB2FUNCS_EXTRA gets pulled in via LIB2ADD
- Get LIB1ASMFUNCS and LIB2ASMSRC
- MAYBE_USE_COLLECT2 got renamed to USE_COLLECT2 (but might not
be used)
- Get EXTRA_HEADERS so that we get generate the right paths for
CPPFLAGS
- Get some variables related to shared libgcc
sys/dev/pckbc/files.pckbc." This is true of sparc, so define __HAVE_NWSCONS,
and remove the code in pckbc_machdep_cnattach() that was conditional on its
not being defined and that is no longer needed (since with __HAVE_NWSCONS,
pckbc_cnattach() will call pckbd_cnattach() itself if necessary).
of bounds, then complain and recover by freeing all of the descriptors.
That will usually provoke additional complaints---see the next
paragraph.
If the firmware returns to us an RSS descriptor that is in-bounds
but out of service (id_node == NULL), then gripe and get out of
the interrupt handler. ***This will still leak RSS descriptors.***
I'm working on a better solution. Possibly I will just free all of
the descriptors.
bug report from Simon Burge: the firmware seems to lock up.
If rate-adaptation is disabled on a Lucent card, then do all the
rate-adaptation work *except* for writing the new transmit rate,
since I suspect that is what locks-up the firmware.
