The Linux Kernel Mailing List
 help / color / mirror / Atom feed
* Linux 3.0+ Disk performance problem - wrong pdflush behaviour
@ 2012-10-09  9:41 Viktor Nagy
  2012-10-10 16:57 ` Jan Kara
  0 siblings, 1 reply; 10+ messages in thread
From: Viktor Nagy @ 2012-10-09  9:41 UTC (permalink / raw)
  To: linux-kernel

[-- Attachment #1: Type: text/plain, Size: 3428 bytes --]

Hello,

Since Kernel version 3.0 pdflush blocks writes even the dirty bytes are 
well below /proc/sys/vm/dirty_bytes or /proc/sys/vm/dirty_ratio.
The kernel 2.6.39 works nice.

How this hurt us in the real life: We have a very high performance game 
server where the MySQL have to do many writes along the reads. All 
writes and reads are very simple and have to be very quick. If we run 
the system with Linux 3.2 we get unacceptable performance. Now we are 
stuck with 2.6.32 kernel here because this problem.

I attach the test program wrote by me which shows the problem. The 
program just writes blocks continously to random position to a given big 
file. The write rate limited to 100 MByte/s. In a well-working kernel it 
have to run with constant 100 MBit/s speed for indefinite long. The test 
have to be run on a simple HDD.

Test steps:
1. You have to use an XFS, EXT2 or ReiserFS partition for the test, Ext4 
forces flushes periodically. I recommend to use XFS.
2. create a big file on the test partiton. For 8 GByte RAM you can 
create a 2 GByte file. For 2 GB RAM I recommend to create 500MByte file. 
File creation can be done with this command:  dd if=/dev/zero 
of=bigfile2048M.bin bs=1M count=2048
3. compile pdflushtest.c: (gcc -o pdflushtest pdflushtest.c)
4. run pdflushtest: ./pdflushtest --file=/where/is/the/bigfile2048M.bin

In the beginning there can be some slowness even on well-working 
kernels. If you create the bigfile in the same run then it runs usually 
smootly from the beginning.

I don't know a setting of /proc/sys/vm variables which runs this test 
smootly on a 3.2.29 (3.0+) kernel. I think this is a kernel bug, because 
if I have much more "/proc/sys/vm/dirty_bytes" than the testfile size 
the test program should never be blocked.

A sample result:
11:20:05: speed:   99.994 MiB/s, time usage:  4.60 %, avg. block 
time:    1.8 us, max. block time: 18 us
11:20:06: speed:   99.994 MiB/s, time usage:  4.60 %, avg. block 
time:    1.8 us, max. block time: 10 us
11:20:07: speed:   99.994 MiB/s, time usage:  4.62 %, avg. block 
time:    1.8 us, max. block time: 11 us
11:20:08: speed:   99.989 MiB/s, time usage:  4.59 %, avg. block 
time:    1.8 us, max. block time: 58 us
11:20:09: speed:   99.997 MiB/s, time usage:  4.55 %, avg. block 
time:    1.8 us, max. block time: 13 us
11:20:10: speed:   28.840 MiB/s, time usage: 96.47 %, avg. block time:  
130.7 us, max. block time: 114076 us
11:20:11: speed:   30.505 MiB/s, time usage: 98.14 %, avg. block time:  
125.7 us, max. block time: 135008 us
11:20:12: speed:   25.956 MiB/s, time usage: 99.71 %, avg. block time:  
150.1 us, max. block time: 129839 us
11:20:13: speed:   25.088 MiB/s, time usage: 96.43 %, avg. block time:  
150.1 us, max. block time: 149649 us
11:20:14: speed:   32.438 MiB/s, time usage: 98.64 %, avg. block time:  
118.8 us, max. block time: 145649 us
11:20:15: speed:   22.765 MiB/s, time usage: 99.11 %, avg. block time:  
170.1 us, max. block time: 159749 us

At 11:20:10 the pdflush started its work, based on the 
"dirty_expire_centisecs". The test file was 2GByte, the 
/proc/sys/vm/dirty_bytes/dirty_bytes was 4000000000. The system (i5-3.4 
GHz, 8G RAM, Kernel 3.2.29-amd64) was booted to run this test only so it 
had 2,2 GByte RAM for cache, and 5.1 GByte RAM free (totally unused).

Sorry if I not found the right place to report this, please advise me 
were to send.

Best regards
Viktor Nagy

[-- Attachment #2: pdflushtest.c --]
[-- Type: text/x-csrc, Size: 5798 bytes --]


#include <stdio.h>
#include <string.h>
#include <getopt.h>
#include <inttypes.h>
#include <sys/time.h>
#include <time.h>
#include <stdlib.h>
#include <unistd.h>
#include <sys/stat.h>
#include <fcntl.h> 

// parameters

char * filename = NULL;
enum testermode {READ, WRITE} mode = WRITE;
float speedlimit = 100; // MiB / s
int burstcount = 10;

// internals

int blocksize = 4096;
int blockcount = 0;

int fh = 0;

char filebuffer[65536];

static struct option long_options[] = 
{
	{"mode",     required_argument, 0,  0 },
	{"m",     required_argument, 0,  0 },
	{"file",    required_argument, 0,  0 },
	{"f",    required_argument, 0,  0 },
	{"limit",    required_argument, 0,  0 },
	{"l",    required_argument, 0,  0 },
	{"burst",    required_argument, 0,  0 },
	{"b",    required_argument, 0,  0 },
	{0,         0,                 0,  0 }
};

void parse_commandline(int argc, char **argv)
{
	int r;
	int option_index = 0;
    
    while (0 <= (r = getopt_long_only(argc, argv, "", long_options, &option_index)))
    {
        const char * optname = long_options[option_index].name;
		if (r == 0)
		{			
			if (strcmp("file", optname) == 0 || strcmp("f", optname) == 0)
			{
				filename = strdup(optarg);
			}
			else if (strcmp("mode", optname) == 0 || strcmp("m", optname) == 0)
			{
				if (strcmp("reader", optarg) == 0 || strcmp("read", optarg) == 0)
				{
					mode = READ;
				}
				else if (strcmp("writer", optarg) == 0 || strcmp("write", optarg) == 0)
				{
					mode = WRITE;
				}
				else
				{
					printf("Invalid mode: \"%s\"\n", optarg);
				}
			}
			else if (strcmp("burst", optname) == 0 || strcmp("b", optname) == 0)
			{
				burstcount = atoi(optarg);
				if (burstcount < 1)  burstcount = 1;
			}
			else if (strcmp("limit", optname) == 0 || strcmp("l", optname) == 0)
			{
				speedlimit = atoi(optarg);
				if (speedlimit < 1)  speedlimit = 1;
			}
			else
			{
				printf("Unhandled option: \"%s\"\n", optname);
			}
		}
		else
		{
			exit(1);
			break;
		}
	}
}

char * getmodename(enum testermode amode)
{
    if (READ == amode)  return "reader";
    if (WRITE == amode)  return "writer";
    return "unknown";
}

int64_t getusecs()
{
	struct timeval tv;
	struct timezone tz;
	gettimeofday(&tv, &tz);
	return (tv.tv_sec * 1000000 + tv.tv_usec);
}

void print_timestamp()
{
	struct tm *tmp;
    time_t t;
    char outstr[200];
    
	t = time(NULL);
    tmp = localtime(&t);
    
	if (strftime(outstr, sizeof(outstr), "%H:%M:%S", tmp) == 0) 
	{
        return;
    }	
    printf("%s", outstr);
}

void read_block(int blockindex)
{
	lseek(fh, blockindex * blocksize, SEEK_SET);
	
	int64_t r = read(fh, &filebuffer, blocksize);
	
	if (r <= 0)
	{
		printf("Error reading block %i\n", blockindex);
		exit(1);
	}
}

void write_block(int blockindex)
{
	lseek(fh, blockindex * blocksize, SEEK_SET);
	
	int64_t r = write(fh, &filebuffer, blocksize);
	
	if (r <= 0)
	{
		printf("Error writing block %i\n", blockindex);
		exit(1);
	}
}

void run_worker()  // generates disk writes
{
	printf("Running %s...\n", getmodename(mode));
	
	int bcnt = 0;
	int64_t bytecnt = 0;
	int n;
	
	int64_t t1 = getusecs();
	int64_t t2 = 0;
	float interval = 0;
	float speed = 0;
	
	int64_t worktime = 0;
	
	int64_t maxblocktime = 0;
	
	while (1)
	{		
		int64_t bwt1 = getusecs();
		
		// some burst read
		for (n = 1; n <= burstcount; n++)
		{
			int bi = rand() % blockcount;			
			int64_t sbt1 = getusecs();
			if (WRITE == mode)
			{
				write_block(bi);			
			}
			else
			{
				read_block(bi);
			}
			int64_t sblocktime = getusecs() - sbt1;
			if (maxblocktime < sblocktime)  maxblocktime = sblocktime;
			bcnt++;
			bytecnt += blocksize;
		}
		
		worktime += (getusecs() - bwt1);
				
		// some wait						
		while (1)
		{		
			t2 = getusecs();
			
			interval = (float) (t2 - t1) / 1000000.;

			speed = (float) bytecnt / (interval * 1024. * 1024.);
			
			if ((speed > speedlimit) && (interval < 1))
			{
				usleep(100);
			}  
			else
			{
				break;
			}
		} 
		
		if (interval >= 1)
		{
			float avgblocktime = (float) worktime / (float) bcnt;
					
			print_timestamp();
			printf(": speed: %8.3f MiB/s, time usage: %5.2f %%, avg. block time: %6.1f us, max. block time: %ld us\n", 
			   speed,
			   100 * ((float) worktime / 1000000.) / interval, 
			   avgblocktime,
			   maxblocktime 
			);
			
			if (interval > 1.1)
			{
				printf("  Work was blocked for a long time! Test interval: %8.3f s\n", interval);
			}
			
			t1 = getusecs();
			bcnt = 0;
			bytecnt = 0;
			worktime = 0;
			maxblocktime = 0;
		}
	}
}

int main(int argc, char **argv)
{
	struct stat sst;
	
	srand(time(NULL));
	
	printf("Linux kernel pdflush tester\n\n");
	printf("Usage: pdflushtest --mode read|write --file filename\n");
	printf("Options:\n  --limit x : limit the maximal speed to x MiB/s\n");
	printf("  --burst x : do the work with x block bursts\n\n");
	
    parse_commandline(argc, argv);
           
    printf("Current parameters: \n");    
    printf("  mode: %s\n", getmodename(mode));
    printf("  file: %s\n", filename);
    printf("  burst length: %i\n", burstcount);
    printf("  speed limit: %f MiB/s\n", speedlimit);
    printf("\n");
    
    // getting scratchfile parameters
    
    printf("Working file \"%s\":\n", filename);
    
    if (stat(filename, &sst) == 0)
    {
		blocksize = sst.st_blksize;
		blockcount = sst.st_size / blocksize;
		
		printf("  size: %ld\n", sst.st_size);
		printf("  blocksize: %i\n", blocksize);
		printf("  number of blocks: %i\n", blockcount);
	}
	else
	{
		printf("Error getting info for file \"%s\"\n", filename);
		exit(1);
	}
	
    printf("\n");
    
	fh = open(filename, O_RDWR);	
	if (fh < 0)
	{
		printf("Error opening file \"%s\"\n", filename);
		exit(1);
	}
	
	run_worker();
        
    printf("finished.\n");
       
    return 0;
}


^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: Linux 3.0+ Disk performance problem - wrong pdflush behaviour
@ 2013-03-06 14:35 Harshana Ranmuthu
  2013-03-07 15:48 ` Jan Kara
  0 siblings, 1 reply; 10+ messages in thread
From: Harshana Ranmuthu @ 2013-03-06 14:35 UTC (permalink / raw)
  To: linux-kernel

>   I don't know how to fix the issue without reverting the patch. Sorry.

with reference to post URL
http://marc.info/?l=linux-fsdevel&m=134997043820759&w=2

I was going through this post, as we are also having problems with the
same commit. In our case, we are appending to a file rather than
updating.

Let me explain the issue as I understand and the solution I think
which can fix the issue.

If a write() writes to a page (for the second time) which is being
flushed (because the page was dirty from the fist write), data written
to disk may have part from the first write and  rest from the second
write. Although, after the second write, page will have the correct
data, but disk will have corrupted data, ie, part from first write and
rest from the second write. After flush(), dirty flag on the page will
be cleared. If there has been a third write on the same page (this
time assuming a clean write), then accurate data will be written to
disk on the next flush. If there's no third write, then file on disk
will remain to have corrupted data. If we can make sure, after the
second write(), page remains dirty (because flush and write() happened
at the same time), then even there's no third write, that page will be
flushed again with accurate data, correcting the corruption on the
disk.
In summery, solution is to make sure the conflicting (write and flush
happen at the same time) pages kept dirty after the flush. This will
make sure these pages will be flushed again even there's no subsequent
write()s to them.

I don't know how easy / difficult this change is. Hope you'll consider this.

Harshana

^ permalink raw reply	[flat|nested] 10+ messages in thread
end of thread, other threads:[~2013-03-07 15:48 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-10-09  9:41 Linux 3.0+ Disk performance problem - wrong pdflush behaviour Viktor Nagy
2012-10-10 16:57 ` Jan Kara
2012-10-10 20:44   ` Viktor Nagy
2012-10-10 21:27     ` Jan Kara
2012-10-11 10:52       ` Viktor Nagy
2012-10-11 10:10         ` Jan Kara
2012-10-11 12:58           ` Viktor Nagy
2012-10-11 15:47             ` Jan Kara
  -- strict thread matches above, loose matches on Subject: below --
2013-03-06 14:35 Harshana Ranmuthu
2013-03-07 15:48 ` Jan Kara

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox