Red Hat Bugzilla – Bug 112694
Severe data corruption in pdc20265 w/ LBA48 disk on channel 0
Last modified: 2015-01-04 17:04:15 EST
From Bugzilla Helper: User-Agent: Mozilla/5.0 (compatible; Konqueror/3.1; i686; , de_DE.ISO-8859-1, de_DE, de) Description of problem: The pdc202xx_old driver severly corrupts data on any device off a pdc20265 controller that needs LBA58 adressing on channel 0, and possibly on other devices on channel 1 because of unconditionally disabling LBA48 addressing on channel 0 and unconditionally enabling LBA48 addressing on channel 1 of any PDC20265 controller found in the system. (Just compare the "address" line in /proc/ide/ide?/hd?/settings for a large disk on channel 0 and channel 1 of the same PDC20265 controller. Version-Release number of selected component (if applicable): Any kernel before 2.4.23 How reproducible: Always Steps to Reproduce: 1. connect a HD >137GB to channel 0 2. make at least 2 filesystems one at the beginning and one crossing or after the 137GB mark. 3. mount filesystems and start copying data to the high one. 4. watch it shred the other filesystem. Actual Results: severe filesystem corruption Expected Results: no corruption at all Additional info: See http://testing.lkml.org/slashdot.php?mid=330272 for discussion of this bug on LKML.
Created attachment 96717 [details] Fix for pdc202xx_old.c data corruption issue
Thanks for the bug report. However, Red Hat no longer maintains this version of the product. Please upgrade to the latest version and open a new bug if the problem persists. The Fedora Legacy project (http://fedoralegacy.org/) maintains some older releases, and if you believe this bug is interesting to them, please report the problem in the bug tracker at: http://bugzilla.fedora.us/