Messages in this thread | | | Date | Sat, 2 Dec 2006 19:35:03 -0800 | From | "Kurtis D. Rader" <> | Subject | Re: data corruption with nvidia chipsets and IDE/SATA drives |
| |
On Sat, 2006-12-02 17:17:37, Kurtis D. Rader wrote: > The same disks attached to a Promise TX2 SATA controller (in the same > system) experience no corruption.
I spoke too soon. Corruption is occurring with the disks attached to the Promise TX2 SATA controller but much less frequently. With the drives attached to the nVidia controller copying certain 2 GiB files would result in at least five bytes, and as many as thirty, being corrupted every single time. On the Promise controller a given copy is likely to be good. And when corruption does occur fewer bytes are being affected --- as little as a single byte in a 2 GiB file. But still, some files never show corruption while others do.
The Promise controller in a PCI slot is measurably slower than the nVidia on the baseboard so the speed of the transfers appears to be a factor. In addition to the pattern of data. My hunch is this is a nVidia nForce 4 chipset design defect involving buss crosstalk or something similar. Which may be why I'm not seeing it when writing to my relatively slow PATA disks.
-- Kurtis D. Rader, Linux level 3 support email: krader@us.ibm.com IBM Integrated Technology Services DID: +1 503-578-3714 15300 SW Koll Pkwy, MS RHE2-O2 service: 800-IBM-SERV Beaverton, OR 97006-6063 http://www.ibm.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |