unroll the loop in ENTRY(pagezero)
acc' to the submitter this results in a reproducible 1% perf
improvement under buildworld like workload
I validated correctness and run-testing, but not performance impact
Submitted by: lidl@pix.net
Reviewed by: adrian
PR: 199151
MFC After: 1 month
These are similar to the mips24k performance counters - some are
available on perfcnt0/3, some are available on perfcnt1/4.
However, the events aren't all the same.
* Add the events, named the same as from Linux oprofile.
* Verify they're the same as "MIPS32(R) 74KTM Processor Core Family
Software User's Manual"; Document Number: MD00519; Revision 01.05.
* Rename INSTRUCTIONS to something else, so it doesn't clash with
the alias INSTRUCTIONS. I'll try to tidy this up later; there
are a few other aliases to add and shuffle around.
Tested:
* QCA9558 SoC (AP135 board) - MIPS74Kc core (no FPU.)
* make universe; where it didn't fail for other reasons.
TODO:
* It'd be nice to support the four performance counters
in at least this hardware, rather than just two.
Reviewed by: bsdimp ("looks good; don't break world".)
Summary:
Book-E and AIM trap.c are almost identical, except for a few bits. This is step
1 in unifying them.
This also renumbers EXC_DEBUG, to not conflict with AIM vector numbers. Since
this is the only one thus far that is used in the switch statement in trap(),
it's the only one renumbered. If others get added to the switch, which conflict
with AIM numbers, they should also be renumbered.
Reviewers: #powerpc, marcel, nwhitehorn
Reviewed By: marcel
Subscribers: imp
Differential Revision: https://reviews.freebsd.org/D2215
that performs the equivalent of an automatic madvise(..., MADV_DONTNEED).
The current heuristic, even with the improvements that I made a few years
ago, is a good example of making the wrong trade-off, or optimizing for
the infrequent case. The infrequent case being reading a single file that
is much larger than memory using mmap(2). And, in this case, the page
daemon isn't the bottleneck; it's the I/O.
In all other cases, the current heuristic has too many false positives,
i.e., it caches too many pages that are later reused. To give one
example, thousands of pages are cached by the current heuristic during a
buildworld and all of them are reactivated before the buildworld
completes. In particular, clang reads source files using mmap(2) and
there are some relatively large source files in our source tree, e.g.,
sqlite, that are read multiple times. With the new heuristic, I see fewer
false positives and they have a much lower cost.
I actually tried something like this more than two years ago and it
didn't perform as well as the cache behind heuristic. However, that was
before the changes to the page daemon in late summer of 2013 and the
existence of pmap_advise(). In particular, with the page daemon doing
its work more frequently and in smaller batches, it now completes its
work while the application accessing the file is blocked on I/O.
Whereas previously, the page daemon appeared to hog the CPU for so long
that it caused "hiccups" in the application's execution.
Finally, I'll add that the elimination of cache pages is a prerequisite
for NUMA support.
Reviewed by: jeff, kib
Sponsored by: EMC / Isilon Storage Division
- Add macros to handle the differences in accessing these registers on arm
and arm64.
- Use the fdt data to detect if we are on an ARMv7 or ARMv8.
- Use the virtual timer by default on arm64, we may not have access to
the physical timer.
Differential Revision: https://reviews.freebsd.org/D2208
Reviewed by: emaste
Sponsored by: The FreeBSD Foundation
Previously, the driver was trying to blink the LED in the newstate
function, but that only gets called once (unlike OpenBSD's net80211
stack). Move the LED blinking to set_channel().
While there, don't try to set the channel when we switch to the SCAN
state. This is already accomplished by the set_channel() function.
MFC after: 1 week
It's necessary to reset the screen to make sure any vendor pixels are
gone when we start boot1. In the Lenovo X1 (3rd gen), this is the
only way to clear the screen. Previously, the Lenovo logo would only
disappear after the kernel started scrolling the display.
After resetting the screen, EFI could put us in the worst LCD mode
(oversized characters), so we now find the largest mode we can use and
hope it's the most appropriate one (it's not trivial to tell what's
the correct LCD resolution at this point). It's worth noting that the
final stage loader has a 'mode' command that can be used to switch
text modes.
While there, enable the software cursor, just like in the legacy boot
mode.
MFC after: 1 week
Even on Illumos, with its much larger KVA, ZFS ARC steps back if KVA usage
reaches certain threshold (3/4 on i386 or 16/17 otherwise). FreeBSD has
even less KVA, but had no such limit on archs with direct map as amd64.
As result, on machines with a lot of RAM, during load with very small user-
space memory pressure, such as `zfs send`, it was possible to reach state,
when there is enough both physical RAM and KVA (I've seen up to 25-30%),
but no continuous KVA range to allocate even single 128KB I/O request.
Address this situation from two sides:
- restore KVA usage limitations in a way the most close to Illumos;
- introduce new requirement for KVA fragmentation, specifying that we
should have at least one sequential KVA range of zfs_max_recordsize bytes.
Experiments show that first limitation done alone is not sufficient. On
machine with 64GB of RAM it is sometimes needed to drop up to half of ARC
size to get at leats one 1MB KVA chunk. Statically limiting ARC to half
of KVA/RAM is too strict, so second limitation makes it to work in cycles:
accumulate trash up to certain critical mass, do massive spring-cleaning,
and then start littering again. :)
MFC after: 1 month
