On Mon, Oct 26, 2020 at 1:31 AM David Hildenbrand <david(a)redhat.com> wrote:
On 23.10.20 21:44, Axel Rasmussen wrote:
> On Fri, Oct 23, 2020 at 12:29 PM David Rientjes <rientjes(a)google.com> wrote:
>> On Wed, 21 Oct 2020, kernel test robot wrote:
>>> FYI, we noticed a 87.8% improvement of vm-scalability.throughput due to
>>> commit: 7fef431be9c9ac255838a9578331567b9dba4477 ("mm/page_alloc: place
pages to tail in __free_pages_core()")
>>> in testcase: vm-scalability
>>> on test machine: 192 threads Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz
with 192G memory
>>> with following parameters:
>>> runtime: 300s
>>> size: 512G
>>> test: anon-wx-rand-mt
>>> cpufreq_governor: performance
>>> ucode: 0x5002f01
>>> test-description: The motivation behind this suite is to exercise functions
and regions of the mm/ of the Linux kernel which are of interest to us.
>> I'm curious why we are not able to reproduce this improvement on Skylake
>> and actually see a slight performance degradation, at least for
>> Axel Rasmussen <axelrasmussen(a)google.com> can provide more details on our
> Right, our results show a slight regression on a Skylake machine ,
> and a slight performance increase on a Rome machine . For these
> tests, I used Linus' v5.9 tag as a baseline, and then applied this
> patchset onto that tag as a test kernel (the patches applied cleanly
> besides one comment, I didn't have to do any code fixups). This is
> running the same anon-wx-rand-mt test defined in the upstream
> lkp-tests job file:
looking at the yaml, am I right that each test is run after a fresh boot?
Yes-ish. For the results I posted, the larger context would have been
- Kernel installed, machine freshly rebooted.
- Various machine management daemons start by default, some are
stopped so as not to interfere with the test.
- Some packages are installed on the machine (the thing which
orchestrates the testing in particular).
- The test is run.
So, the machine is somewhat fresh in the sense that it hasn't been
e.g. serving production traffic just before running the test, but it's
also not as clean as it could be. It seems plausible this difference
explains the difference in the results (I'm not too familiar with how
the Intel kernel test robot is implemented).
As I already replied to David, this patch merely changes the initial
order of the freelists. The general end result is that lower memory
addresses will be allocated before higher memory addresses will be
allocated - within a zone, the first time memory is getting allocated.
Before, it was the other way around. Once a system ran for some time,
freelists are randomized.
There might be benchmarks/systems where this initial system state might
now be better suited - or worse. It doesn't really tell you that core-mm
is behaving better/worse now - it merely means that the initial system
state under which the benchmark was started affected the benchmark.
Looks like so far there is one benchmark+system where it's really
beneficial, there is one benchmark+system where it's slightly
beneficial, and one benchmark+system where there is a slight regression.
Something like the following would revert to the previous behavior:
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 23f5066bd4a5..fac82420cc3d 100644
@@ -1553,7 +1553,9 @@ void __free_pages_core(struct page *page, unsigned
* Bypass PCP and place fresh pages right to the tail, primarily
* relevant for memory onlining.
- __free_pages_ok(page, order, FPI_TO_TAIL);
+ __free_pages_ok(page, order,
+ system_state < SYSTEM_RUNNING ? FPI_NONE :
(Or better, passing the expected behavior via MEMINIT_EARLY/... to
But then, I am not convinced we should perform that change: having a
clean (initial) state might be true for these benchmarks, but it's far
from reality. The change in numbers doesn't show you that core-mm is
operating better/worse, just that the baseline for you tests changed due
to a changed initial system state.
Not to put words in David's mouth :) but at least from my perspective,
our original interest was "wow, an 87% improvement! maybe we should
deploy this patch to production!", and I'm mostly sharing my results
just to say "it actually doesn't seem to be a huge *general*
improvement", rather than to advocate for further changes / fixes.
IIUC the original motivation for this patch was to fix somewhat of an
edge case, not to make a very general improvement, so this seems fine.
> David / dhildenb