Re: 2.6.1-rc1

[Paolo Ornati]

On Wednesday 31 December 2003 09:36, Linus Torvalds wrote:
> Ok, I've merged a lot of pending patches into 2.6.1-rc1, and will now
> calm down for a while again, to make sure that the final 2.6.1 is ok.
>
> Most of the updates is for stuff that has been in -mm for a long while
> and is stable, along with driver updates (SCSI, network, i2c and USB).
>
> 		Linus
>
> ----

With 2.6.1-rc1 I have noticed a strange IDE performance change.

Results of "hdparm -t /dev/hda" with 2.6.0 kernel:
(readahead = 256):		~26.31 MB/s
(readahead = 128):		~31.82 MB/s

PS = readahead is set to 256 by default on my system, 128 seems to be the 
best value

Results of "hdparm -t /dev/hda" with 2.6.1-rc1 kernel:
(readahead = 256):		~26.41 MB/s
(readahead = 128):		~26.27 MB/s

Setting readahead to 128 doesn't have the same effect with the new kernel...

INFO on my HD:

/dev/hda:
 multcount    = 16 (on)
 IO_support   =  1 (32-bit)
 unmaskirq    =  1 (on)
 using_dma    =  1 (on)
 keepsettings =  0 (off)
 readonly     =  0 (off)
 readahead    = 128 (on)
 geometry     = 38792/16/63, sectors = 39102336, start = 0

/dev/hda:

 Model=WDC WD200BB-53AUA1, FwRev=18.20D18, SerialNo=WD-WMA6Y1501425
 Config={ HardSect NotMFM HdSw>15uSec SpinMotCtl Fixed DTR>5Mbs FmtGapReq }
 RawCHS=16383/16/63, TrkSize=57600, SectSize=600, ECCbytes=40
 BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=39102336
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio1 pio2 pio3 pio4
 DMA modes:  mdma0 mdma1 mdma2
 UDMA modes: udma0 udma1 udma2 udma3 *udma4 udma5
 AdvancedPM=no WriteCache=enabled
 Drive conforms to: device does not report version:  1 2 3 4 5


IDE controller:

00:04.1 IDE interface: VIA Technologies, Inc. VT82C586/B/686A/B PIPC Bus 
Master IDE (rev 10) (prog-if 8a
[Master SecP PriP])
        Flags: bus master, medium devsel, latency 32
        I/O ports at b800 [size=16]
        Capabilities: [c0] Power Management version 2


I don't understand how this happens... the only changes to IDE driver seems 
to be these:

>
> Summary of changes from v2.6.0 to v2.6.1-rc1
> ============================================
> Andrew Morton:
>   o Can't disable IDE DMA
>   o IDE MMIO fix
>   o IDE capability elevation fix
>
> Linus Torvalds:
>   o Make IDE DRQ and READY timeouts longer

For my tests I have used this stupid shell script:

#!/bin/bash

echo "HD test for linux `uname -r`"
echo

ra=8
for i in `seq 12`; do
    echo "READAHEAD = $ra";
    hdparm -a $ra /dev/hda;
    for j in `seq 3`; do
	hdparm -t /dev/hda;
    done;
    ra=$(($ra*2));
done

Results for 2.6.0 && 2.6.1-rc1 are attached.

Bye

-- 
	Paolo Ornati
	Linux v2.6.0

Attachment: 2.6.0.gz
Description: GNU Zip compressed data

Attachment: 2.6.1-rc1.gz
Description: GNU Zip compressed data