Problem with mounting a device second time.

[ali nasir]

Hi,

1. Before calling the ioctl, we ensure that all the files are closed using the fclose.
2. If we do the ioctl, the unmount() is still succesfull.
3. How do i find the remaining user of the disk? Is there a struct where i can see something?

Regards,
Ali

------------------------------
On Thu 19 Apr, 2012 12:14 PM IST Sebastian Huber wrote:

>On 04/19/2012 08:10 AM, ali nasir wrote:
>> What we do is the following:
>> 1. With the SD card inserted, we write a set of files (4 to 5 files) on the card. After all the fwrite(..) has returned, we call the ioctl for the RTEMS_BLKIO_SYNCDEV to ensure that the data is written to the card.
>
>You should close all files before the ioctl to make sure that all buffers of 
>the IO library are flushed.
>
>> 2. We then unmount the card by calling the below set of functions:
>> unmount
>> rmdir
>> rtems_bdbuf_syncdev
>> rtems_disk_delete
>> rtems_bdbuf_purge_dev
>> rtems_io_unregister_driver
>>
>> When we step in in the rtems_disk_delete function, the free_disk_device() is not called as the dd->uses equals 1.
>
>This means someone is still using this disk.  You should figure out this user.
>
>> 3. After this, if we insert the second card, after removing the first, then the second card fails in the create_phys by returning 12
>
>Yes, this is because the first disk is still in use.
>
>>
>> 4. When we do not call the ioctl for the RTEMS_BLKIO_SYNCDEV, then we are succesfully able to unmount and also mount the second card.
>
>Is the unmount() not successful if you do the ioctl?
>
>>
>> Should we remove the call to the ioctl for the RTEMS_BLKIO_SYNCDEV? or is there some other solution to the problem?
>> Please guide.
>
>The solution is to track down the remaining disk user.
>
>-- 
>Sebastian Huber, embedded brains GmbH
>
>Address : Obere Lagerstr. 30, D-82178 Puchheim, Germany
>Phone   : +49 89 18 90 80 79-6
>Fax     : +49 89 18 90 80 79-9
>E-Mail  : sebastian.huber at embedded-brains.de
>PGP     : Public key available on request.
>
>Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.
>_______________________________________________
>rtems-users mailing list
>rtems-users at rtems.org
>http://www.rtems.org/mailman/listinfo/rtems-users