Re: 2.6.20-rc3 regression? hdparm shows 1/2...1/3 the throughput
> Tim Schmielau wrote: >> See >> http://lkml.org/lkml/2007/1/2/75 >> for the solution. OK; this has already been committed a few days ago. I assume this closes the issue. I will post after -rc4 in the unlikely case that it doesn't. -- Stefan Richter -=-=-=== ---= --==- http://arcgraph.de/sr/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: 2.6.20-rc3 regression? hdparm shows 1/2...1/3 the throughput
Tim Schmielau wrote: > On Sat, 6 Jan 2007, Stefan Richter wrote: > >> Did anybody else notice this? The result of "hdparm -t" under 2.6.20-rc >> seems to be less than half of what you get on 2.6.19. However, disk I/O >> did *not* get slower according to bonnie++. > > yes. See > http://lkml.org/lkml/2007/1/2/75 > for the solution. > > Tim Thanks. I should have remembered that. # hdparm -tT /dev/hda /dev/hda: Timing cached reads: 10864 MB in 2.00 seconds = 5440.10 MB/sec Timing buffered disk reads: 58 MB in 3.06 seconds = 18.94 MB/sec # cat /sys/block/hda/queue/scheduler noop anticipatory deadline [cfq] # echo anticipatory > /sys/block/hda/queue/scheduler # hdparm -tT /dev/hda /dev/hda: Timing cached reads: 10680 MB in 2.00 seconds = 5347.20 MB/sec Timing buffered disk reads: 148 MB in 3.02 seconds = 49.03 MB/sec -- Stefan Richter -=-=-=== ---= --==- http://arcgraph.de/sr/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: 2.6.20-rc3 regression? hdparm shows 1/2...1/3 the throughput
On Sat, 6 Jan 2007, Stefan Richter wrote: > Did anybody else notice this? The result of "hdparm -t" under 2.6.20-rc > seems to be less than half of what you get on 2.6.19. However, disk I/O > did *not* get slower according to bonnie++. yes. See http://lkml.org/lkml/2007/1/2/75 for the solution. Tim - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
2.6.20-rc3 regression? hdparm shows 1/2...1/3 the throughput
Did anybody else notice this? The result of "hdparm -t" under 2.6.20-rc seems to be less than half of what you get on 2.6.19. However, disk I/O did *not* get slower according to bonnie++. That is, there is no harm beyond confusion of the user. hdparm is lying for some reason. Below is the output of hdparm v6.6 and bonnie++ 1.93c. I ran hdparm 3 times each and bonnie++ 2 times to ensure that I get typical readings, on an otherwise idle system. hdparm -tT /dev/???: 2.6.19.1, IDE: Timing cached reads: 10728 MB in 2.00 seconds = 5371.95 MB/sec Timing buffered disk reads: 154 MB in 3.01 seconds = 51.11 MB/sec 2.6.19.1, SATA: Timing cached reads: 8028 MB in 2.00 seconds = 4017.69 MB/sec Timing buffered disk reads: 220 MB in 3.01 seconds = 73.18 MB/sec 2.6.19.1 + latest 1394 drivers, FireWire 800: Timing cached reads: 10216 MB in 2.00 seconds = 5114.73 MB/sec Timing buffered disk reads: 214 MB in 3.03 seconds = 70.71 MB/sec 2.6.19.1 + latest 1394 drivers, FireWire 400: Timing cached reads: 8892 MB in 2.00 seconds = 4449.76 MB/sec Timing buffered disk reads: 74 MB in 3.08 seconds = 24.02 MB/sec 2.6.20-rc3, IDE: Timing cached reads: 11492 MB in 2.00 seconds = 5753.76 MB/sec Timing buffered disk reads: 56 MB in 3.10 seconds = 18.09 MB/sec 2.6.20-rc3, SATA: Timing cached reads: 10736 MB in 2.00 seconds = 5374.70 MB/sec Timing buffered disk reads: 102 MB in 3.00 seconds = 33.99 MB/sec 2.6.20-rc3 + latest 1394 drivers, FireWire 800: Timing cached reads: 9476 MB in 2.00 seconds = 4742.33 MB/sec Timing buffered disk reads: 70 MB in 3.08 seconds = 22.69 MB/sec 2.6.20-rc3 + latest 1394 drivers, FireWire 400: Timing cached reads: 10336 MB in 2.00 seconds = 5173.65 MB/sec Timing buffered disk reads: 40 MB in 3.09 seconds = 12.95 MB/sec bonnie++ on the SATA disk with a single 47% filled reiserfs partition: 2.6.19.1, SATA: > Version 1.93c --Sequential Output-- --Sequential Input- > --Random- > Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- > --Seeks-- > MachineSize K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec > %CP > stein2G 203 99 57389 18 26943 9 856 93 66032 17 190.8 > 6 > Latency 48553us 295ms1130ms 196ms 18075us 500ms > Version 1.93c --Sequential Create-- Random > Create > stein -Create-- --Read--- -Delete-- -Create-- --Read--- > -Delete-- > files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec > %CP > 16 16541 55 + +++ 18794 71 22933 79 + +++ 24005 > 95 > Latency 107ms 129us 175ms 16613us 76us 236us That is, 56.0 MB/s write and 64.5 MB/s read performance of sequential block I/O. 2.6.20-rc3, SATA: > Version 1.93c --Sequential Output-- --Sequential Input- > --Random- > Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- > --Seeks-- > MachineSize K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec > %CP > stein2G 210 99 57340 18 26920 10 1166 89 66805 17 195.3 > 5 > Latency 47794us 284ms 869ms 249ms 36229us 533ms > Version 1.93c --Sequential Create-- Random > Create > stein -Create-- --Read--- -Delete-- -Create-- --Read--- > -Delete-- > files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec > %CP > 16 19645 67 + +++ 21767 83 26951 94 + +++ 24859 > 98 > Latency 231ms 85us 120us 469us 69us 118us That is, 56.0 MB/s write and 65.2 MB/s read performance of sequential block I/O. It's a 32bit kernel on a dual core x86 PC: > $ uname -a > Linux stein 2.6.20-rc3 #7 SMP PREEMPT Sat Jan 6 17:07:30 CET 2007 i686 > Intel(R) Core(TM)2 CPU T7200 @ 2.00GHz GenuineIntel GNU/Linux The motherboard is i945GT based. > $ /usr/sbin/lspci > 00:00.0 Host bridge: Intel Corporation Mobile Memory Controller Hub (rev 03) > 00:01.0 PCI bridge: Intel Corporation Mobile PCI Express Graphics Port (rev > 03) > 00:02.0 VGA compatible controller: Intel Corporation Mobile Integrated > Graphics Controller (rev 03) > 00:1b.0 Class 0403: Intel Corporation 82801G (ICH7 Family) High Definition > Audio Controller (rev 01) > 00:1c.0 PCI bridge: Intel Corporation 82801G (ICH7 Family) PCI Express Port 1 > (rev 01) > 00:1c.4 PCI bridge: Intel Corporation 82801GR/GH/GHM
2.6.20-rc3 regression? hdparm shows 1/2...1/3 the throughput
Did anybody else notice this? The result of hdparm -t under 2.6.20-rc seems to be less than half of what you get on 2.6.19. However, disk I/O did *not* get slower according to bonnie++. That is, there is no harm beyond confusion of the user. hdparm is lying for some reason. Below is the output of hdparm v6.6 and bonnie++ 1.93c. I ran hdparm 3 times each and bonnie++ 2 times to ensure that I get typical readings, on an otherwise idle system. hdparm -tT /dev/???: 2.6.19.1, IDE: Timing cached reads: 10728 MB in 2.00 seconds = 5371.95 MB/sec Timing buffered disk reads: 154 MB in 3.01 seconds = 51.11 MB/sec 2.6.19.1, SATA: Timing cached reads: 8028 MB in 2.00 seconds = 4017.69 MB/sec Timing buffered disk reads: 220 MB in 3.01 seconds = 73.18 MB/sec 2.6.19.1 + latest 1394 drivers, FireWire 800: Timing cached reads: 10216 MB in 2.00 seconds = 5114.73 MB/sec Timing buffered disk reads: 214 MB in 3.03 seconds = 70.71 MB/sec 2.6.19.1 + latest 1394 drivers, FireWire 400: Timing cached reads: 8892 MB in 2.00 seconds = 4449.76 MB/sec Timing buffered disk reads: 74 MB in 3.08 seconds = 24.02 MB/sec 2.6.20-rc3, IDE: Timing cached reads: 11492 MB in 2.00 seconds = 5753.76 MB/sec Timing buffered disk reads: 56 MB in 3.10 seconds = 18.09 MB/sec 2.6.20-rc3, SATA: Timing cached reads: 10736 MB in 2.00 seconds = 5374.70 MB/sec Timing buffered disk reads: 102 MB in 3.00 seconds = 33.99 MB/sec 2.6.20-rc3 + latest 1394 drivers, FireWire 800: Timing cached reads: 9476 MB in 2.00 seconds = 4742.33 MB/sec Timing buffered disk reads: 70 MB in 3.08 seconds = 22.69 MB/sec 2.6.20-rc3 + latest 1394 drivers, FireWire 400: Timing cached reads: 10336 MB in 2.00 seconds = 5173.65 MB/sec Timing buffered disk reads: 40 MB in 3.09 seconds = 12.95 MB/sec bonnie++ on the SATA disk with a single 47% filled reiserfs partition: 2.6.19.1, SATA: Version 1.93c --Sequential Output-- --Sequential Input- --Random- Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- MachineSize K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP stein2G 203 99 57389 18 26943 9 856 93 66032 17 190.8 6 Latency 48553us 295ms1130ms 196ms 18075us 500ms Version 1.93c --Sequential Create-- Random Create stein -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete-- files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP 16 16541 55 + +++ 18794 71 22933 79 + +++ 24005 95 Latency 107ms 129us 175ms 16613us 76us 236us That is, 56.0 MB/s write and 64.5 MB/s read performance of sequential block I/O. 2.6.20-rc3, SATA: Version 1.93c --Sequential Output-- --Sequential Input- --Random- Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- MachineSize K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP stein2G 210 99 57340 18 26920 10 1166 89 66805 17 195.3 5 Latency 47794us 284ms 869ms 249ms 36229us 533ms Version 1.93c --Sequential Create-- Random Create stein -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete-- files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP 16 19645 67 + +++ 21767 83 26951 94 + +++ 24859 98 Latency 231ms 85us 120us 469us 69us 118us That is, 56.0 MB/s write and 65.2 MB/s read performance of sequential block I/O. It's a 32bit kernel on a dual core x86 PC: $ uname -a Linux stein 2.6.20-rc3 #7 SMP PREEMPT Sat Jan 6 17:07:30 CET 2007 i686 Intel(R) Core(TM)2 CPU T7200 @ 2.00GHz GenuineIntel GNU/Linux The motherboard is i945GT based. $ /usr/sbin/lspci 00:00.0 Host bridge: Intel Corporation Mobile Memory Controller Hub (rev 03) 00:01.0 PCI bridge: Intel Corporation Mobile PCI Express Graphics Port (rev 03) 00:02.0 VGA compatible controller: Intel Corporation Mobile Integrated Graphics Controller (rev 03) 00:1b.0 Class 0403: Intel Corporation 82801G (ICH7 Family) High Definition Audio Controller (rev 01) 00:1c.0 PCI bridge: Intel Corporation 82801G (ICH7 Family) PCI Express Port 1 (rev 01) 00:1c.4 PCI bridge: Intel Corporation 82801GR/GH/GHM (ICH7 Family) PCI Express Port 5 (rev 01) 00:1d.0
Re: 2.6.20-rc3 regression? hdparm shows 1/2...1/3 the throughput
On Sat, 6 Jan 2007, Stefan Richter wrote: Did anybody else notice this? The result of hdparm -t under 2.6.20-rc seems to be less than half of what you get on 2.6.19. However, disk I/O did *not* get slower according to bonnie++. yes. See http://lkml.org/lkml/2007/1/2/75 for the solution. Tim - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: 2.6.20-rc3 regression? hdparm shows 1/2...1/3 the throughput
Tim Schmielau wrote: On Sat, 6 Jan 2007, Stefan Richter wrote: Did anybody else notice this? The result of hdparm -t under 2.6.20-rc seems to be less than half of what you get on 2.6.19. However, disk I/O did *not* get slower according to bonnie++. yes. See http://lkml.org/lkml/2007/1/2/75 for the solution. Tim Thanks. I should have remembered that. # hdparm -tT /dev/hda /dev/hda: Timing cached reads: 10864 MB in 2.00 seconds = 5440.10 MB/sec Timing buffered disk reads: 58 MB in 3.06 seconds = 18.94 MB/sec # cat /sys/block/hda/queue/scheduler noop anticipatory deadline [cfq] # echo anticipatory /sys/block/hda/queue/scheduler # hdparm -tT /dev/hda /dev/hda: Timing cached reads: 10680 MB in 2.00 seconds = 5347.20 MB/sec Timing buffered disk reads: 148 MB in 3.02 seconds = 49.03 MB/sec -- Stefan Richter -=-=-=== ---= --==- http://arcgraph.de/sr/ - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: 2.6.20-rc3 regression? hdparm shows 1/2...1/3 the throughput
Tim Schmielau wrote: See http://lkml.org/lkml/2007/1/2/75 for the solution. OK; this has already been committed a few days ago. I assume this closes the issue. I will post after -rc4 in the unlikely case that it doesn't. -- Stefan Richter -=-=-=== ---= --==- http://arcgraph.de/sr/ - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/