Date: Mon, 5 Sep 2005 12:41:05 -0500 From: Jason Harmening <jason.harmening@gmail.com> To: freebsd-hardware@freebsd.org, freebsd-stable@freebsd.org Subject: atapicam + MO drive weirdness in 6.0 Message-ID: <200509051241.05303.Jason.Harmening@gmail.com>
next in thread | raw e-mail | index | archive | help
On my amd64 desktop, I have a Fujitsu DynaMO 2.3G ATAPI MO drive, and I've always had to use atapicam with it because there's no native ATA driver. The drive is genuinely capable of UDMA33, and I have hw.ata.atapi_dma turned on. In the past (with 5.x) the SCSI device listing that comes from atapicam at boot time has always shown the full transfer rate, as follows: da0 at ata0 bus 0 target 0 lun 0 da0: <FUJITSU MCJ3230AP 0030> Removable Optical SCSI-4 device da0: 33.000MB/s transfers But since I upgraded to 6.0-BETA a month and a half ago, it doesn't look as if DMA is being enabled: da0 at ata0 bus 0 target 0 lun 0 da0: <FUJITSU MCJ3230AP 0030> Removable Optical SCSI-4 device da0: 3.300MB/s transfers My UDMA33 CD burner still shows 33 MB/s in its atapicam SCSI listing, so I know UDMA transfers aren't completely broken with atapicam. And I can't seem to use atacontrol to manually change the transfer mode, because atacontrol has changed and requires an actual ata device file, as opposed to the old "atacontrol <channel#> <mode0> <mode1>" syntax. I can still use the MO drive, but I'd like to be able to tap into its full transfer rate. I'm running the latest BETA4 now, and the problem is still present. It's not a showstopper for me by any means, but both this problem and the removal of the old manual atacontrol syntax seem to be unfortunate regressions in the new ATA subsystem. Any help or further explanation would be greatly appreciated. Thanks, Jason Harmening
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200509051241.05303.Jason.Harmening>