hdparm is busted in the hd? -> sd? change

confirmation:
/dev/sda:
 IO_support = 0 (default 16-bit)
 readonly = 0 (off)
 readahead = 256 (on)
 geometry = 9729/255/63, sectors = 156301488, start = 0
root@T3:~# hdparm -d1 /dev/sda

/dev/sda:
 setting using_dma to 1 (on)
 HDIO_SET_DMA failed: Inappropriate ioctl for device

This is a real problem since the system slows down significantly during HD access.
This also is a serious regression since this worked much better before the change.

This bug was caused by the change in the kernel which made IDE devices appear as SCSI devices, so I've added the kernel package here.
I Hope this is ok.

DMA ought to be set automatically. If you have a device where this isn't the case, please provide /var/log/dmesg and the output of the lspci command.

DMA is enabled, disc access does slow down the computer noticeably though.

This is the reason why I started to look whether DMA was enabled.

Do you have anything else I could try to get back the original performance?

[ 53.018256] libata version 2.00 loaded.
[ 53.022606] pata_via 0000:00:11.1: version 0.2.0
[ 53.023078] ACPI: PCI Interrupt Link [ALKA] BIOS reported IRQ 0, using IRQ 20
[ 53.023083] ACPI: PCI Interrupt Link [ALKA] enabled at IRQ 20
[ 53.023093] ACPI: PCI Interrupt 0000:00:11.1[A] -> Link [ALKA] -> GSI 20 (level, low)
[ 53.023166] ata1: PATA max UDMA/133 cmd 0x1F0 ctl 0x3F6 bmdma 0xE400 irq 14
[ 53.023217] ata2: PATA max UDMA/133 cmd 0x170 ctl 0x376 bmdma 0xE408 irq 15
[ 53.023248] scsi1 : pata_via
[ 53.239025] scsi 0:0:5:0: CD-ROM TEAC CD-R58S 1.0H PQ: 0 ANSI
[ 53.239058] target0:0:5: Beginning Domain Validation
[ 53.247569] ata1.00: ATA-6, max UDMA/100, 390719855 sectors: LBA48
[ 53.247573] ata1.00: ata1: dev 0 multi count 16
[ 53.255264] target0:0:5: FAST-10 SCSI 10.0 MB/s ST (100 ns, offset 15)
[ 53.265984] target0:0:5: Domain Validation skipping write tests
[ 53.265987] target0:0:5: Ending Domain Validation
[ 53.269738] ata1.00: configured for UDMA/100

pata_via has since been reverted back to the ide via driver.