131072 * 4 = 524288.
A bigger alignment by multiple of... doesn't mean its not good.
If its not a trouble, use the 524288, to be safe.
Besides, on other hardware platform they probably use the same assumption of 512
alignment so it will be better to conform to this alignment.
Regards,
tzahi.
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On Behalf Of Shachar Shemesh
> Sent: Thursday, September 29, 2005 2:55 PM
> To: Israel Linux Mailing List
> Subject: open(2) in O_DIRECT of a block device
>
> Hi all,
>
>
> The manual page for open is a bit unclear on the matter. It says:
>
> "Under Linux 2.4 transfer sizes, and the alignment of user buffer and
> file offset must all be multiples of the logical block size
> of the file
> system. Under Linux 2.6 alignemnet to 512-byte boundaries suffices."
>
>
> How does this apply for block devices? After all, it's unlikely that I
> need to conform to the 131072 bytes block that reiserFs reports when
> accessing the /dev/hda. On the other hand, I have not managed to find
> what the native access boundry for block devices is (an ioctl?).
>
>
> Thanks,
>
>
> Shachar
>
>
> =================================================================
> To unsubscribe, send mail to [EMAIL PROTECTED] with
> the word "unsubscribe" in the message body, e.g., run the command
> echo unsubscribe | mail [EMAIL PROTECTED]
>
>
>
=================================================================
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
