Note, that driver author (from ALI) suggested that for M5229 rev. <= 0x20
UDMA is "not stable", so it should be disabled by the driver...
> So far, so good?
> Here are my hdparm -tT /dev/hda results,
>
> /dev/hda:
> Timing buffer-cache reads: 128 MB in 3.04 seconds =42.11 MB/sec
> Timing buffered disk reads: 64 MB in 9.73 seconds = 6.87 MB/sec
Quite standard results for mword DMA mode 2 ...
> I am disappointed by the 6.87MB/sec. Shouldn't it be much more? With
> using_dma=0 I get 4-5MB/sec. NB: The Maxtor HD is a UDMA/Mode2 device and
> its specs say,
>
> DATA TRANSFER RATE:
> To/From Interface: ....................... Up to 33.0 MBs
> To/From Media: ........................... Up to 18.6 MBs
In UDMA/Mode2, yes. But you must use apropriate (UDMA-capable) controller
...
> A couple of supplementary points/questions: My kernel is reporting
> 'ALI15X3: MultiWord DMA enabled' rather than 'ALI15X3: Ultra DMA enabled'
> as Meelis' does. Any ideas why?
This chipset revision is not UDMA capable...
> According to Western Digital's UDMA info webpage, 'multiword DMA mode 2'
> should give a max. data-rate of 16.6MB/sec - much more than I get.
My WD drive with same chipset revision reports about 8.5 MB/s.
On unloaded AXP system (CMD646) with IBM drive I observe ~12 MB/s.
I observe, that for many BIOSes timings are calculated basing on maximum
possible PCI clock (42.5 MHz ?) rather than on currently being set one.
It effects in a bit worse transfer results. To test it you can try:
1. put the idebus=... parameter to the kernel during boot
(I use idebus=30, as set on my MoBo)
2. hdparm -d0 ...
3. hdpatm -p ... # calculate new timings
4. hdparm -X34 ...
5. hdparm -d1 ...
Difference shouldn't be meaningfull, but results slightly better.
Cheers
-- ======================================================================= Andrzej M. Krzysztofowicz ankry@mif.pg.gda.pl tel. (0-58) 347 14 61 Wydz.Fizyki Technicznej i Matematyki Stosowanej Politechniki Gdanskiej- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/