Description of Problem: With kernel-2.4.18-0.22 (Athlon), IDE performance on an Abit KR7A133 motherboard is quite poor: /dev/hda: Timing buffer-cache reads: 128 MB in 0.48 seconds =266.67 MB/sec Timing buffered disk reads: 64 MB in 10.36 seconds = 6.18 MB/sec (The drive is a Western Digital WD400BB on an 80pin cable. It is certainly capable of higher speeds than this.) From dmidecode: Handle 0x0002 DMI type 2, 8 bytes. Board Information Block Vendor: ABIT <http://www.abit.com.tw> Product: VT8366A-8233 (KR7A) Version: v1.0 ~ Serial Number: From dmesg: VP_IDE: IDE controller on PCI bus 00 dev 89 PCI: No IRQ known for interrupt pin A of device 00:11.1. Please try using pci=biosirq. VP_IDE: chipset revision 6 VP_IDE: not 100% native mode: will probe irqs later VP_IDE: Unknown VIA SouthBridge, contact Vojtech Pavlik <vojtech> I will attach lspci output. There is no /proc/ide/via. Note that a KR7A board (no 133, which indicates that it has the previous revision of the chipset which did not support ATA133) identically configured with the same processor and hard drive works fine. I suspect that a table somewhere in the IDE driver needs updating (or perhaps FUTURE_BRIDGES needs to be set).
Created attachment 55488 [details] Output from lspci -vvxxx
Created attachment 55545 [details] Small patch which corrects the problem
I have attached a small patch which I extracted and adapted from the larger VIA IDE 3.33 patch. With this patch, I get the following in dmesg: Uniform Multi-Platform E-IDE driver Revision: 6.31 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx VP_IDE: IDE controller on PCI bus 00 dev 89 PCI: No IRQ known for interrupt pin A of device 00:11.1. Please try using pci=biosirq. VP_IDE: chipset revision 6 VP_IDE: not 100% native mode: will probe irqs later ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx VP_IDE: VIA vt8233a (rev 00) IDE UDMA100 controller on pci00:11.1 ide0: BM-DMA at 0xd800-0xd807, BIOS settings: hda:DMA, hdb:pio ide1: BM-DMA at 0xd808-0xd80f, BIOS settings: hdc:DMA, hdd:pio and the following cheap performance result: /dev/hda: Timing buffer-cache reads: 128 MB in 0.48 seconds =266.67 MB/sec Timing buffered disk reads: 64 MB in 1.87 seconds = 34.22 MB/sec While the full VIA IDE patch would be nice (for proper Ultra133 speed support) something like this patch is somewhat essential as the VIA Ultra133 chipsets work poorly without it.
This issue is still extant in Valhalla. The previously attached patch still applies and still solves the problem.
added for the next build; thanks!
The patch seems to work for my Asus A7V333 running RHL7.3 with Red Hat's 2.4.18-4. A useful stopgap. Thanks. From dmesg, after the patch: Uniform Multi-Platform E-IDE driver Revision: 6.31 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx VP_IDE: IDE controller on PCI bus 00 dev 89 PCI: No IRQ known for interrupt pin A of device 00:11.1. Please try using pci=biosirq. VP_IDE: chipset revision 6 VP_IDE: not 100% native mode: will probe irqs later ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx VP_IDE: VIA vt8233a (rev 00) IDE UDMA100 controller on pci00:11.1 ide0: BM-DMA at 0xb400-0xb407, BIOS settings: hda:DMA, hdb:pio ide1: BM-DMA at 0xb408-0xb40f, BIOS settings: hdc:DMA, hdd:pio