Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 22 Jul 2015 15:47:12 +0300
From:      Dmitry Sivachenko <trtrmitya@gmail.com>
To:        hackers@freebsd.org
Subject:   Re: Strange memory management with mmap()
Message-ID:  <15DE3B94-3C09-4855-A274-D5655B049403@gmail.com>
In-Reply-To: <3434ED75-7994-4E9E-9B06-FACCD7DC90FF@gmail.com>
References:  <FDB6E0F9-A3FF-4194-83C1-A3121CBAE407@gmail.com> <3434ED75-7994-4E9E-9B06-FACCD7DC90FF@gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help

> On 16 =D0=B8=D1=8E=D0=BB=D1=8F 2015 =D0=B3., at 21:19, Dmitry =
Sivachenko <trtrmitya@gmail.com> wrote:
>=20
>>=20
>> On 16 =D0=B8=D1=8E=D0=BB=D1=8F 2015 =D0=B3., at 18:42, Dmitry =
Sivachenko <trtrmitya@gmail.com> wrote:
>>=20
>> Hello!
>>=20
>> I am using FreeBSD-10-stable and writing a program that uses large =
data file via mmap() in read only mode.
>> To be specific, I have 256GB RAM machine and typical size of data =
file is ~160GB (more than 1/2 of RAM and less that the whole RAM).
>> There is no other programs running during the test.
>>=20
>> Consider the following use case: I have two files on disk.  I mmap() =
the first one and prefetch data to RAM (touch every page of the file).
>> After that I expect all data to be cached in RAM and subsequent =
access will be fast.
>>=20
>> Next I do munmap() on the first file, mmap() the second one and do =
the same test: prefetch data and expect it to be cached in RAM (and some =
of the pages belonging to the first file to be purged out, because =
size_of(file1)+size_of(file2) > size_of(RAM).
>>=20
>> Please find my test program attached.
>>=20
>> I run the program with 2 files provided via command line (both about =
160GB).
>> What I observe in real is:
>> -- before I run the program all RAM is in FREE state as reported by =
top(1).
>> -- after first prefetch() of the first file, all it's data goes to =
"Cache" state, RES column of the process remains the same (small)
>> -- second prefetch() works fast as expected, memory goes from Cache =
to Active state, RES column of the process grows up to match file size =
(SIZE=3D=3DRES now)
>> -- now first prefetch() for second file starts: the remaining Free =
memory goes to Cache state, Active size still equals to first file size.
>> -- second prefetch() for second file works as slow as first one, like =
if nothing was cached in memory during the first prefetch() run, RES =
column does not change.
>>=20
>>=20
>> Here is the output:
>> % /tmp/a.out file1.dat file2.dat
>> file1.dat... First prefault time: 1235.747351 seconds
>> Second prefault time: 74.893323 seconds
>> Ok.
>> file2.dat... First prefault time: 1316.405527 seconds
>> Second prefault time: 1311.491842 seconds
>> Ok.
>>=20
>=20
>=20


I tried the same test program on Linux machine with similar hardware.  =
It behaves like expected (second prefetch works very fast):

file1.dat... First prefault time: 2664.621088 seconds
Second prefault time: 1.969283 seconds
Ok.
file2.dat... First prefault time: 2917.009003 seconds
Second prefault time: 34.128762 seconds
Ok.





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?15DE3B94-3C09-4855-A274-D5655B049403>