Jameson Graef Rollins wrote: > If I keep everything else the same (same version of udev) I do *not* > experience the problem with kernel: > > 2.6.38-2-amd64 > > but I *do* experience the problem with kernels: > > 2.6.39-2-amd64 > 3.0.0-1-amd64 > > If you really think it's worthwhile to try downgrading udev let me know,
Nah, that shouldn't be necessary. Thanks for figuring out the kernel regression range. The only potentially problematic commit I could find is the following. It might be worth testing with it backed out. commit bf2253a6f00e Author: Tejun Heo <[email protected]> Date: Fri Apr 29 10:15:14 2011 +0200 cdrom: always check_disk_change() on open cdrom_open() called check_disk_change() after the rest of open path succeeded which leads to the following bizarre behavior. * After media change, if the device opened without O_NONBLOCK, open_for_data() naturally fails with -ENOMEDIA and check_disk_change() is never called. The media is known to be gone and the open failure makes it obvious to the userland but device invalidation never happens. * But if the device is opened with O_NONBLOCK, all the checks are bypassed and cdrom_open() doesn't notice that the media is not there and check_disk_change() is called and invalidation happens. There's nothing to be gained by avoiding calling check_disk_change() on open failure. Common cases end up calling check_disk_change() anyway. All we get is inconsistent behavior. Fix it by moving check_disk_change() invocation to the top of cdrom_open() so that it always gets called regardless of how the rest of open proceeds. Stable: 2.6.38 Signed-off-by: Tejun Heo <[email protected]> Reported-by: Amit Shah <[email protected]> Tested-by: Amit Shah <[email protected]> Cc: [email protected] Signed-off-by: Jens Axboe <[email protected]> diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c index 514dd8efaf73..75fb965b8f72 100644 --- a/drivers/cdrom/cdrom.c +++ b/drivers/cdrom/cdrom.c @@ -986,6 +986,9 @@ int cdrom_open(struct cdrom_device_info *cdi, struct block_device *bdev, fmode_t cdinfo(CD_OPEN, "entering cdrom_open\n"); + /* open is event synchronization point, check events first */ + check_disk_change(bdev); + /* if this was a O_NONBLOCK open and we should honor the flags, * do a quick open without drive/disc integrity checks. */ cdi->use_count++; @@ -1012,9 +1015,6 @@ int cdrom_open(struct cdrom_device_info *cdi, struct block_device *bdev, fmode_t cdinfo(CD_OPEN, "Use count for \"/dev/%s\" now %d\n", cdi->name, cdi->use_count); - /* Do this on open. Don't wait for mount, because they might - not be mounting, but opening with O_NONBLOCK */ - check_disk_change(bdev); return 0; err_release: if (CDROM_CAN(CDC_LOCK) && cdi->options & CDO_LOCK) { -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
