On 01/06/2014 04:26 PM, Jan Kara wrote:
On Mon 06-01-14 15:51:55, Raghavendra K T wrote:
Currently, max_sane_readahead returns zero on the cpu with empty numa node,
fix this by checking for potential empty numa node case during calculation.
We also limit the number of readahead pages to 4k.
Signed-off-by: Raghavendra K T <raghavendra.kt@xxxxxxxxxxxxxxxxxx>
---
The current patch limits the readahead into 4k pages (16MB was suggested
by Linus). and also handles the case of memoryless cpu issuing readahead
failures. We still do not consider [fm]advise() specific calculations
here. I have dropped the iterating over numa node to calculate free page
idea. I do not have much idea whether there is any impact on big
streaming apps.. Comments/suggestions ?
As you say I would be also interested what impact this has on a streaming
application. It should be rather easy to check - create 1 GB file, drop
caches. Then measure how long does it take to open the file, call fadvise
FADV_WILLNEED, read the whole file (for a kernel with and without your
patch). Do several measurements so that we get some meaningful statistics.
Resulting numbers can then be part of the changelog. Thanks!
Hi Honza,
Thanks for the idea. (sorry for the delay, spent my own time to do some
fadvise and other benchmarking). Here is the result on my x240 machine
with 32 cpu (w/ HT) 128GB ram.
Below test was for 1gb test file as per suggestion.
x base_result
+ patched_result
N Min Max Median Avg Stddev
x 12 7.217 7.444 7.2345 7.2603333 0.06442802
+ 12 7.24 7.431 7.243 7.2684167 0.059649672
From the result we could see that there is not much impact with the
patch.
I shall include the result in changelog when I resend/next version
depending on the others' comment.
---
test file looked something like this:
char buf[4096];
int main()
{
int fd = open("testfile", O_RDONLY);
unsigned long read_bytes = 0;
int sz;
posix_fadvise(fd, 0, 0, POSIX_FADV_DONTNEED);
do {
sz = read(fd, buf, 4096);
read_bytes += sz;
} while (sz > 0);
close(fd);
printf (" Total bytes read = %lu \n", read_bytes);
return 0;
}
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>