Bug 23011
| Summary: | UDMA 66 not working on 4.1.1-Release ? | ||
|---|---|---|---|
| Product: | Base System | Reporter: | Barry Tigner <tigner> |
| Component: | kern | Assignee: | freebsd-bugs (Nobody) <bugs> |
| Status: | Closed FIXED | ||
| Severity: | Affects Only Me | ||
| Priority: | Normal | ||
| Version: | Unspecified | ||
| Hardware: | Any | ||
| OS: | Any | ||
Same problem appears on FBSD 4.2-RELEASE. Using Promise 66 paddle board, IBM fast IDE drive, proper UDMA ribbon cable (including those supplied by Promise Inc and IBM). dmesg excerpt: ata2-master: DMA limited to UDMA33, non-ATA66 compliant cable ad4: 29314MB <IBM-DTLA-307030> [59560/16/63] at ata2-master UDMA33 acd0: CDROM <TOSHIBA CD-ROM XM-6402B> at ata2-slave using PIO4 Mounting root from ufs:/dev/ad4s1a Motherboard is an older Tyan TomCat IV. IDE DMA support is turned on in the <rebuilt> kernel. Joe D. From: gArgraves@earthlink.net I have experiance the same problems. hardware: MB = EPOX EP-MVP3G2 w/VIA Apollo MVP3 AGPset CPU = AMD K6-2 500Mhz RAM = 128 SDRAM 2 Linksys Fast ethernet 10/100 HD = MAXSTOR 20G (ATA66) and also tried with a WD 45G (ATA66) I have also tried other modes like various PIO. Nothing seem to work correctly. Always get a variaty of disk errors. If i put a non-ATA (UDMA) drive in system I can install freeBSD with sucessfully. This seems to be true for ver 4.0 from CD. 4.1.1 from ftp server and ver 4.2 from ftp server. PS this URL may give in-site to problem: http://www.uwsg.indiana.edu/hypermail/linux/kernel/9812.0/0621.html Gary Hi, I'm experiencing the same problem in getting UDMA66 out of FreeBSD. dmesg snippet follows: ... atapci0: <VIA 82C686 ATA66 controller> port 0xd000-0xd00f at device 7.1 on pci0 ata0: at 0x1f0 irq 14 on atapci0 ata1: at 0x170 irq 15 on atapci0 ... ad0: 9773MB <FUJITSU MPF3102AT> [19857/16/63] at ata0-master UDMA66 ad1: 4110MB <QUANTUM FIREBALL ST4.3A> [14848/9/63] at ata0-slave UDMA33 acd0: CD-RW <YAMAHA CRW8824E> at ata1-master using WDMA2 acd1: DVD-ROM <Pioneer DVD-ROM ATAPIModel DVD-104S 0118> at ata1-slave using UDMA33 I believe I've got round the problem by editing ata-dma.c and commenting out the code that sets udmamode back to 2 (which the above dmesg shows). Anyone knowing a good I/O benchmark to confirm this please let me know. State Changed From-To: open->closed We only support uptil UDMA33 on the Sis chips. |
dmesg reports HD is running at ata33, cdrom at PIO4. Both work ok in Win98, checked bios settings , recompiled kernel with ata dma enabled, still get the same message. I AM using an 80 wire ata66 cable on ATA0 and a 40 wire ata33 cable on ATA1. I've tried replacing the cables with new ones even though it appears to work ok in windows, and still get the same message. dmesg excerpt follows here. > dmesg |grep ata atapci0: <SiS 5591 ATA33 controller> port 0xd000-0xd00f,0xd400-0xd403,0xd800-0xd807,0xe000-0xe003,0xe400-0xe407 irq 14 at device 0.1 on pci0 ata0: at 0x1f0 irq 14 on atapci0 ata1: at 0x170 irq 15 on atapci0 ata0-master: DMA limited to UDMA33, non-ATA66 compliant cable ad0: 19531MB <Maxtor 92048U8> [39683/16/63] at ata0-master using UDMA33 acd0: CDROM <CD-ROM 56X> at ata1-master using PIO4 afd0: 120MB <LS-120 COSM 02 UHD Floppy> [963/8/32] at ata1-slave using PIO2 Mainboard=Asus p5A-B, Ali chipset, builtin snd,agp2x , shared mem for vga CPU=AMD K6-2 3DNow/400mhz. MEM=64MB PC100 SDRAM HD=Maxtor 7200rpm 2MB cache ata66 20GB ATA0-master CDrom=56x ATA33 MagicSpin ATA1-master nofloppy LS-120 superfloppy = ATA1-slave Fix: no clue at all. I have enabled ata dma in the kernel and recompiled. I've tried recompiling with optimizations on, still same problem. How-To-Repeat: it repeats all the time.