On Wed, 2005-01-19 at 09:28, Sami Farin wrote: > On Wed, Jan 19, 2005 at 08:55:40AM -0800, Daniel McNeil wrote: > > On Tue, 2005-01-18 at 18:31, Sami Farin wrote: > ... > > > I have Linux 2.6.10-ac9 + bio clone memory corruption -patch, > > > and dio_bug does not give errors (without your patch). > > > > I should have mentioned that my testing was on ext3 with 4k > > block size. The bio clone patch might affect this by merging > > the i/o into a single iovec. Here's an updated test program > > that uses 2 different buffers allocated separately that might > > prevent the merging. See if this works on your system. > > I have reiserfs... and this version does not give errors, either.
Reisefs must handle allocations differently. The bug only shows up if there is an unallocated hole and a O_DIRECT write with multiple iovecs writes partially into the allocated space and completes at least 1 iovec worth of data and then finishes the i/o using buffer i/o to allocate space for the hole and write the remaining data. Daniel - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
