Mark Kirkwood
2005-Feb-10 21:52 UTC
atapci VIA 82C596B UDMA66 controller: problem for 5.X ?
Rob, I am currently running 5.3 Release on a Tyan Trinity 400 (VIA 82C596B UDMA66) with 2 x Seagate Barracude IV (UDMA100) and have no stability issues. e.g from dmesg: atapci0: <VIA 82C596B UDMA66 controller> port 0xe000-0xe00f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 at device 7.1 on pci0 ata0: channel #0 on atapci0 ata1: channel #1 on atapci0 ad0: 19092MB <ST320011A/3.10> [38792/16/63] at ata0-master UDMA66 ad1: 19092MB <ST320011A/3.10> [38792/16/63] at ata0-slave UDMA66 acd0: CDRW <CREATIVE CD-RW RW8439E/1R09> at ata1-master PIO4 What does your dmesg show if you dont force ata_dma=0? Hmmmm - you have an IBM disk in one pc and WDs in the other, hard to believe that the disks are the issue. Do both pcs have the same mobo? (if so what is it). Something else - did you upgrade from 4.x or fresh install? With respect to attempting a fix, you could try $ atacontrol mode 0 UDMA66 UDMA66 immediately after boot, to see if you can force a sensible transfer mode setting. regards Mark Rob wrote:> Hi, > > I have two PCs, with apparently a somewhat old > motherboard, since there's a maximum support for > UDMA66 by the harddisk controller. > The attached harddisks are newer and allow a > higher speed of UDMA100. > > This combination of older motherboard and newer > harddisk causes great problems with 5.X (it never > did with 4.X). I have irregular crashes of X and > even crashes at initial bootup. All this is solved > by using hw.ata.ata_dma="0", which forces the > harddisk to use the very low speed of PIO4 :(. > > Both PCs have a VIA 82C596B UDMA66 controller. > I wonder if that is the problem for 5.X ? > > On these two PCs, the output is of > grep -i ata /var/run/dmesg.boot > is: > > -- PC1 ------------------- > atapci0: <VIA 82C596B UDMA66 controller> port > 0xe000-0xe00f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 > at device 7.1 on pci0 > ata0: channel #0 on atapci0 > ata1: channel #1 on atapci0 > ad0: 43979MB <IBM-DTLA-307045/TX6OA50C> [89355/16/63] > at ata0-master PIO4 > acd0: CDROM <CRD-8520B/1.00> at ata1-master PIO4 > > > -- PC2 ------------------- > atapci0: <VIA 82C596B UDMA66 controller> port > 0xc000-0xc00f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 > at device 7.1 on pci0 > ata0: channel #0 on atapci0 > ata1: channel #1 on atapci0 > ad0: 76319MB <WDC WD800JB-00CRA1/17.07W17> > [155061/16/63] at ata0-master PIO4 > ad1: 16448MB <WDC WD172AA/05.05B05> [33420/16/63] at > ata0-slave PIO4 > acd0: CDRW <HL-DT-ST RW/DVD GCC-4320B/1.00> at > ata1-master UDMA33 > > ------------------------------------------------- > > I consider this a serious drawback of 5.3. > > Any idea how this can be resolved? > > Best regards, > Rob. > > > > __________________________________ > Do you Yahoo!? > Yahoo! Mail - Helps protect you from nasty viruses. > http://promotions.yahoo.com/new_mail > _______________________________________________ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org" > >
Mars Trading
2005-Feb-11 01:57 UTC
atapci VIA 82C596B UDMA66 controller: problem for 5.X ?
Hey Rob, I'm on the same hardware - 82C596B (UDMA66) - but using a Seagate 20GB UDMA100 Drive. Relevant info from dmesg: atapci0: <VIA 82C596B UDMA66 controller> port 0xd000-0xd00f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 at device 7.1 on pci0 ata0: channel #0 on atapci0 ad0: 19092MB <ST320014A/3.07> [38792/16/63] at ata0-master UDMA66 sysctl -a shows: hw.ata.ata_dma: 1 hw.ata.wc: 1 hw.ata.atapi_dma: 1 I'm using 5-stable now but didn't have problems when using 5.3-release. Flakey hardware? Dennis On Thu, 10 Feb 2005 19:18:26 -0800 (PST), Rob <spamrefuse@yahoo.com> wrote:> > Hi, > > I have two PCs, with apparently a somewhat old > motherboard, since there's a maximum support for > UDMA66 by the harddisk controller. > The attached harddisks are newer and allow a > higher speed of UDMA100. > > This combination of older motherboard and newer > harddisk causes great problems with 5.X (it never > did with 4.X). I have irregular crashes of X and > even crashes at initial bootup. All this is solved > by using hw.ata.ata_dma="0", which forces the > harddisk to use the very low speed of PIO4 :(. > > Both PCs have a VIA 82C596B UDMA66 controller. > I wonder if that is the problem for 5.X ? > > On these two PCs, the output is of > grep -i ata /var/run/dmesg.boot > is: > > -- PC1 ------------------- > atapci0: <VIA 82C596B UDMA66 controller> port > 0xe000-0xe00f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 > at device 7.1 on pci0 > ata0: channel #0 on atapci0 > ata1: channel #1 on atapci0 > ad0: 43979MB <IBM-DTLA-307045/TX6OA50C> [89355/16/63] > at ata0-master PIO4 > acd0: CDROM <CRD-8520B/1.00> at ata1-master PIO4 > > -- PC2 ------------------- > atapci0: <VIA 82C596B UDMA66 controller> port > 0xc000-0xc00f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 > at device 7.1 on pci0 > ata0: channel #0 on atapci0 > ata1: channel #1 on atapci0 > ad0: 76319MB <WDC WD800JB-00CRA1/17.07W17> > [155061/16/63] at ata0-master PIO4 > ad1: 16448MB <WDC WD172AA/05.05B05> [33420/16/63] at > ata0-slave PIO4 > acd0: CDRW <HL-DT-ST RW/DVD GCC-4320B/1.00> at > ata1-master UDMA33 > > ------------------------------------------------- > > I consider this a serious drawback of 5.3. > > Any idea how this can be resolved? > > Best regards, > Rob. > > __________________________________ > Do you Yahoo!? > Yahoo! Mail - Helps protect you from nasty viruses. > http://promotions.yahoo.com/new_mail > _______________________________________________ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org" >-- Mars Trading 9620 Kamagong St., Makati, Manila, Philippines Phone: (632) 8952241 FAX: (632) 8996124 email: marstrade@gmail.com
Mark Kirkwood wrote:> Rob, > > I am currently running 5.3 Release on a Tyan > Trinity 400 (VIA 82C596B > UDMA66) with 2 x Seagate Barracude IV (UDMA100) > and have no stability issues. e.g from dmesg: > > atapci0: <VIA 82C596B UDMA66 controller> port > 0xe000-0xe00f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 > at device 7.1 on pci0 > ata0: channel #0 on atapci0 > ata1: channel #1 on atapci0 > ad0: 19092MB <ST320011A/3.10> [38792/16/63] > at ata0-master UDMA66 > ad1: 19092MB <ST320011A/3.10> [38792/16/63] > at ata0-slave UDMA66 > acd0: CDRW <CREATIVE CD-RW RW8439E/1R09> > at ata1-master PIO4 > > What does your dmesg show if you dont force > ata_dma=0?At initial bootup, I get some UDMA error messages, several lines with something like: READ_DMA UDMA ... ICRC error ... LBA.... then the kernel panics and the system reboots; the same error messages occur again etc. etc. ad infinitum.> With respect to attempting a fix, you could try > $ atacontrol mode 0 UDMA66 UDMA66I have put this into the file '/etc/rc.early', and that works. So then, after bootup, I get: # sysctl hw.ata.ata_dma hw.ata.ata_dma: 0 # atacontrol mode 0 Master = UDMA66 Slave = BIOSPIO What does that mean? UDMA66 and hw.ata.ata_dma=0 ? And why then is UDMA66 not automatically choosen at bootup? Thanks, Rob. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
Mark Kirkwood
2005-Feb-12 19:45 UTC
atapci VIA 82C596B UDMA66 controller: problem for 5.X ?
Rob wrote:> > > What does that mean? UDMA66 and hw.ata.ata_dma=0 ? >Hmmm, that is interesting, 2 suggestion to determine what is going on: - run diskinfo -t on your disks and let us know the results - use sysctl to set hw.ata.ata_dma=1 and see what happens> And why then is UDMA66 not automatically choosen > at bootup?At a guess I would suggest that dma is broken for your motherboards in 5.3 (what were they again?). regards Mark P.s : My diskinfo results FYI: zmori# diskinfo -t /dev/ad1s1a /dev/ad1s1a 512 # sectorsize 18946621952 # mediasize in bytes (18G) 37005121 # mediasize in sectors 36711 # Cylinders according to firmware. 16 # Heads according to firmware. 63 # Sectors according to firmware. Seek times: Full stroke: 250 iter in 5.024736 sec = 20.099 msec Half stroke: 250 iter in 3.989959 sec = 15.960 msec Quarter stroke: 500 iter in 6.864865 sec = 13.730 msec Short forward: 400 iter in 3.010462 sec = 7.526 msec Short backward: 400 iter in 1.247889 sec = 3.120 msec Seq outer: 2048 iter in 0.490103 sec = 0.239 msec Seq inner: 2048 iter in 0.457750 sec = 0.224 msec Transfer rates: outside: 102400 kbytes in 2.418307 sec = 42344 kbytes/sec middle: 102400 kbytes in 2.571397 sec = 39823 kbytes/sec inside: 102400 kbytes in 4.105673 sec = 24941 kbytes/sec
--- Mars Trading <marstrade@gmail.com> wrote:> This idea may seem useless but what have you got to > lose? > > Have you tried changing bios setting for hard drive > mode to "auto" or > something other than "LBA"? Maybe "LARGE" or "CHS"?Is there a risk that I lose all data on my disk, when changing this in the BIOS? BTW: During the fresh FreeBSD install, I have never encountered a choice for formatting with or without LBA. In the Fdisk window, I choose 'use entire disk for FreeBSD', and in the partition window I have set 'newfs' for all partitions. Thanks, Rob. __________________________________ Do you Yahoo!? Meet the all-new My Yahoo! - Try it today! http://my.yahoo.com
Hi, I have two PCs, with apparently a somewhat old motherboard, since there's a maximum support for UDMA66 by the harddisk controller. The attached harddisks are newer and allow a higher speed of UDMA100. This combination of older motherboard and newer harddisk causes great problems with 5.X (it never did with 4.X). I have irregular crashes of X and even crashes at initial bootup. All this is solved by using hw.ata.ata_dma="0", which forces the harddisk to use the very low speed of PIO4 :(. Both PCs have a VIA 82C596B UDMA66 controller. I wonder if that is the problem for 5.X ? On these two PCs, the output is of grep -i ata /var/run/dmesg.boot is: -- PC1 ------------------- atapci0: <VIA 82C596B UDMA66 controller> port 0xe000-0xe00f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 at device 7.1 on pci0 ata0: channel #0 on atapci0 ata1: channel #1 on atapci0 ad0: 43979MB <IBM-DTLA-307045/TX6OA50C> [89355/16/63] at ata0-master PIO4 acd0: CDROM <CRD-8520B/1.00> at ata1-master PIO4 -- PC2 ------------------- atapci0: <VIA 82C596B UDMA66 controller> port 0xc000-0xc00f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 at device 7.1 on pci0 ata0: channel #0 on atapci0 ata1: channel #1 on atapci0 ad0: 76319MB <WDC WD800JB-00CRA1/17.07W17> [155061/16/63] at ata0-master PIO4 ad1: 16448MB <WDC WD172AA/05.05B05> [33420/16/63] at ata0-slave PIO4 acd0: CDRW <HL-DT-ST RW/DVD GCC-4320B/1.00> at ata1-master UDMA33 ------------------------------------------------- I consider this a serious drawback of 5.3. Any idea how this can be resolved? Best regards, Rob. __________________________________ Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. http://promotions.yahoo.com/new_mail _______________________________________________ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"