Re: [PATCH] [FAT] fix VFAT compat ioctls on 64-bit systems

2007-04-29 Thread OGAWA Hirofumi
Bart Oldeman <[EMAIL PROTECTED]> writes: > I found your reply on linux-kernel via an archive but am not subscribed. Sorry. I CCed actually, but sf.net rejected my email configuration. > Anyway, I like your approach but there was one small bug. Thanks for fixing my stupid bug. -- OGAWA Hirofumi

Re: [PATCH] [FAT] fix VFAT compat ioctls on 64-bit systems

2007-04-29 Thread Bart Oldeman
On 2007-04-29, OGAWA Hirofumi <[EMAIL PROTECTED]> wrote: > > I think we should use filldir() callback on compat_ioctl too, and it > should fix all bugs of compat code. > > The untested patch is below, although the code seems not clean > enough. What do you think? There is one problem: -

Re: [PATCH] [FAT] fix VFAT compat ioctls on 64-bit systems

2007-04-29 Thread OGAWA Hirofumi
Hi, Bart Oldeman <[EMAIL PROTECTED]> writes: > if you compile and run the below test case in an msdos or vfat directory > on an x86-64 system with -m32 you'll get garbage in the kernel_dirent > struct followed by a SIGSEGV. > > The patch below fixes this. It's a minimal patch -- I haven't conve