Re: [PATCH 2.5.62-mm3] objrmap fix for X

["Martin J. Bligh" <mbligh@aracnet.com>]

>>> Keep the flag for now, find the escaped page under X, remove the flag
>>> later?
>> 
>> It occurred to me I'm already using (abusing?) the flag for nonlinear
>> pages, so I have to keep it.  I'll chase solutions for X.
> 
> Ok, the vm_ops->nopage function is set in drivers like drm and agp.  I
> don't think it's reasonable to require all of them to set PageAnon.  So
> here's a patch that tests the page on do_no_page and sets the flag
> appropriately.

Well, it runs fine, but I get truly freaky performance results. My machine
might have gone wacko on me or something - the patch seems perfectly simple
to me. Kernbench is all over the map - user and elapsed way up, system is
down. Ummm .. probably all too strange to be true, and I've made a mistake,
but if some more sane person than I could run a quick test, would help.

Thanks,

M.

Kernbench-2: (make -j N vmlinux, where N = 2 x num_cpus)
                              Elapsed        User      System         CPU
              2.5.62-mjb3       43.92      557.65       94.12     1483.50
                     test       68.61      923.78       90.19     1477.33

Kernbench-16: (make -j N vmlinux, where N = 16 x num_cpus)
                              Elapsed        User      System         CPU
              2.5.62-mjb3       45.21      560.46      114.58     1492.67
                     test       69.04      927.20      100.56     1488.17

DISCLAIMER: SPEC(tm) and the benchmark name SDET(tm) are registered
trademarks of the Standard Performance Evaluation Corporation. This 
benchmarking was performed for research purposes only, and the run results
are non-compliant and not-comparable with any published results.

Results are shown as percentages of the first set displayed

SDET 1  (see disclaimer)
                                Throughput    Std. Dev
                   2.5.62-mjb3       100.0%         3.1%
                          test        83.9%         1.7%

SDET 2  (see disclaimer)
                                Throughput    Std. Dev
                   2.5.62-mjb3       100.0%         4.0%
                          test        87.0%         3.8%

SDET 4  (see disclaimer)
                                Throughput    Std. Dev
                   2.5.62-mjb3       100.0%         2.0%
                          test        90.3%         2.1%

SDET 8  (see disclaimer)
                                Throughput    Std. Dev
                   2.5.62-mjb3       100.0%         4.6%
                          test        95.6%         6.3%

SDET 16  (see disclaimer)
                                Throughput    Std. Dev
                   2.5.62-mjb3       100.0%         2.7%
                          test       103.0%         3.4%

SDET 32  (see disclaimer)
                                Throughput    Std. Dev
                   2.5.62-mjb3       100.0%         0.9%
                          test        96.6%         1.0%

SDET 64  (see disclaimer)
                                Throughput    Std. Dev
                   2.5.62-mjb3       100.0%         1.1%
                          test        94.8%         0.6%

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org">aart@kvack.org</a>