Linux - HardwareThis forum is for Hardware issues.
Having trouble installing a piece of hardware? Want to know if that peripheral is compatible with Linux?
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
I've turned raid off while fighting with DMA modes. I have bios for RAID functions of 3112 and I even updated it. It was recommended because of mistakes in previous version of that bios.
Version 1.0.0.11 will crash if you installed any PCI card on SE7505VB2 motherboard....I have tried, it work perfectly. During installation of Red Hat 7.3 it detects my mirror raid set and happilly installed and boot. However, it just simply cannot tolerate any PCI card. Please tell me if you manage to overcome this trouble.
First of all I was one of those virgin-window users. Now I am running Mandrake 9.1 on a XP 2500+, AN78X Deluxe, Seagate SATA 80MB, 512 Kingston 2700, Asus Geforce 5200, and I guess the rest is irrelevant.
Anyway, ran HDPARM -Tt /dev/hda and got the following answer:
---------------------------------------------
[root@localhost /]# hdparm -Tt /dev/hda
/dev/hda:
read() failed: Input/output error
Timing buffered disk reads: 0 MB in 0.00 seconds = nan MB/sec
Hmm.. suspicious results: probably not enough free memory for a proper test.
--------------------------------------------
Where do we go from there? I am lost
Thanks!
Salvador.
Sorry guys, I guesse I used the wrong letter at the end of the command...
Now it goes like this:
-------------------------------------------------
[root@localhost /]# hdparm -Tt /dev/hde
/dev/hde:
Timing buffer-cache reads: 128 MB in 0.26 seconds =492.31 MB/sec
Timing buffered disk reads: 64 MB in 46.28 seconds = 1.38 MB/sec
[root@localhost /]#
---------------------------------------------------
However, I have seen some guys posting buffer-cache reads at 512 MB/sec and reading at 32MB/sec.
I gave up on Seagate SATA (4 ST3120026AS drives RAID 1/0 on ABIT IS7-G @ ~$600) a few weeks ago and sent them back. Switched to un-of-the-mill ATA (4 WD1200JB + 3ware 7500-4LP RAID 1/0 on ABIT IS7 @ $640).
I'm getting:
hdparm -Tt /dev/sda
/dev/sda:
Timing buffer-cache reads: 128 MB in 0.15 seconds =853.33 MB/sec
Timing buffered disk reads: 64 MB in 0.98 seconds = 65.31 MB/sec
So I paid $40 for no headache, full redundancy and fast disk reads/writes out-of-the-box under Mandrake or RedHat.
Just a consideration if you're looking into hardware. It's not the 80 - 110 MB/sec sustained I was hoping for (silly me) with RAID 1/0 SATA, but its infinately better than spending a few days to try and get better than 1-3 MB/sec out of my Seagate SATA, and if you don't have SATA RAID on your MoBo, the SATA RAID cards are rediculously priced.
Maybe I'll try again in 6 months for the next round of server upgrades
Originally posted by arxik Right now Im' testing P4G8X with Seagate SATA drive under Suse 8.2 linux. [...] Anyway with hdparm -X69 -u1 -d1 I managed to get 41 Mb/sec speed and no problems.
I have the same (deluxe) board, and I'm using Suse 8.2 too. If I use the commandline after booting it works fine, but if I put it to /etc/init.d/boot.local (which the last executed bootscript in SuSE 8.2) it fails, with a system crash.
The /etc/sysconfig/hardware differs very much from the mandrake config file, and you can't set the parameters needed for a succesfull configuration during startup.
My question: does anybody know how to change the SuSE 8.2 startup scripts/configuration to automatically set udma5 on the SATA?
but just checking /var/log/messages.log (dmesg), It notice that is filled with lines of : hde: sata_error = 0x00000000, watchdog = 0, siimage_mmio_ide_dma_test_irq
whereas the disk behaves normarly and noticeably faster.
Does anybody know if is is likely to be a problem ?
BTW: 1-what is the use of the command echo max_kb_per_request:15 > /proc/ide/hde/settings ?
2- what is the best param for hdparm -c1 or -c3 ?
--------------------------------------------------------------------------------------
Hardware Info
Motherboard: Abit NF7S with
Sil3112A Serial ATA with Sil3112A Serial ATA controller:
01:0b.0 RAID bus controller: CMD Technology Inc Silicon Image SiI 3112 SATARaid Controller (rev 02)
OS: Mandrake Linux 9.1 (Kernel: 2.4.21-0.13mdk)
I have an ASUS A7N8X deluxe mobo and a SATA Maxtor hardisk.
I tried everything in these posts with no success, only system failures hard lockups...
What do I need to do to get this working? Wait for kernel 2.6 or what?
Last edited by Jormungand; 07-31-2003 at 04:48 PM.
Originally posted by Jormungand I have an ASUS A7N8X deluxe mobo and a SATA Maxtor hardisk.
I tried everything in these posts with no success, only system failures hard lockups...
What do I need to do to get this working? Wait for kernel 2.6 or what?
For checking out hardware problems I can't recommend the knoppix 3.2 live CD highly enough. It has an up to date 2.4.21 kernel and awesome hardware detection. Just put it in the drive and boot. It also gives you the chance to set boot parameters to fine-tune things.
It should detect all drives on the system (in my case a SATA hd, a SCSI CD-writer, a DVD reader and an IDE zip drive). Existing partitions and devices show up as desktop icons. Right-click on one to mount read-only. Also you can try
If you like what you see there is a script to install knoppix to HD (you basically get a Debian/KDE system). At least you will have a better idea where the blame lies.
Also, doesn't your BIOS let you set the SATA disk to compatible mode? If I do that knoppix thinks its a plain ATA and puts it at /dev/hda (v. /dev/hde in SATA mode). Have you tried that?
Is SATA enhanced mode any better than compatible mode?
I have a new P4 m/b (Asus P4P800-VM) and SATA drive (Barracuda V 120G) for a dual boot Linux/WinME system. Looking through the BIOS settings I see a choice of 'enhanced' or 'compatible' mode for the SATA drive. WinME would not install in enhanced mode, but goes onto the first partition OK in compatible mode. But this system is mainly for Linux and 'enhanced' has to better than 'compatible' doesn't it? Read on...
I have knoppix 3.2 installed on the HD, but the live knoppix CD seems to give similar hdparm readings (where I have compared) and makes it easy to fiddle with BIOS settings. Have to turn DMA on tho', which is automatic for the HD install.
In 'enhanced' mode the SATA disk shows up as /dev/hde:
/dev/hde:
Timing buffer-cache reads: 128 MB in 0.16 seconds =800.00 MB/sec
Timing buffered disk reads: 64 MB in 1.15 seconds = 55.65 MB/sec
In 'compatible' mode the same disk is at /dev/hda and gives exactly the same timings! In fact one cache read was even higher (914Mb/s) but that seems to be an anomaly.
In both cases it made little difference whether the IO_support flag was set to 16 or 32 bits.
The downside to 'enhanced' mode is that you probably have to reset the BIOS to boot WinME/98/95 (XP and 2000 are supposed to be OK). I haven't even tried to set up LILO to figure that one out. If you are dual booting to XP or 2000 I suppose you would want to be in enhanced mode.
Maybe there is some important dimension to disk performance that hdparm is missing, but otherwise why would you care?
I just wanted to say thanks to everyone for keeping this thread active and interesting! It's been a while since I started this thread and it amazes me how much interest there is in the Linux community about serial ATA drives.
I hope there is going to be a solid resolution sometime this year or early next year, I heard that developers are definitely putting in their efforts to get things moving on SATA support.
I am also impressed that a google search of "Linux SATA support" yields this thread as the first result. That's impressive.
May be the best approach for you would be to merge the patch to the kernel source tree and rebuild the kernel. You don't need to do much to configure, since you could just see if there's a file called config-<kernel version> on the /boot partition, copy that file to the /usr/src/linux-<version> directory as .config, make a menuconfig or an xconfig and see what the settings are, then apply the patch and see what was changed... normally a patch also installs some documentation... so check also the contents of the Documentation directory under the source tree main directory.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.