bug#27884: parted reports the partition is not optimally aligned, even though it's 1MiB-aligned

[Alicia Boya García]

On 09/20/2017 02:55 PM, Phil Susi wrote:
> On 7/30/2017 6:46 PM, Alicia Boya García wrote:
>> Hi.
>>
>> Is there any explanation to this GNU parted output?
>>
>> GNU Parted 3.2
>> Using /dev/sde
>> Welcome to GNU Parted! Type 'help' to view a list of commands.
>> (parted) unit B                                                          
>> (parted) p                                                               
>> Model: Seagate M3 Portable (scsi)
>> Disk /dev/sde: 4000787029504B
>> Sector size (logical/physical): 512B/4096B
>> Partition Table: gpt
>> Disk Flags:
>>
>> Number  Start     End             Size            File system  Name  Flags
>>  1      1048576B  4000786153471B  4000785104896B  ext4
>>
>> (parted) align-check                                                     
>> alignment type(min/opt)  [optimal]/minimal?                              
>> Partition number? 1                                                      
>> 1 not aligned
>>
>> Why does it say it's not aligned? Both the the start and the end are
>> 1MB-aligned (and therefore trivially 4096B-aligned and 512B-aligned), am
>> I right?
> Maybe your drive is advertising a different optimal alignment.  What
> does /sys/block/sde/queue/optimal_io_size say?
>
Indeed.

This drive was advertising a ridiculous optimal_io_size that is not even
a power of two (2^16-1 512-byte sectors). Unfortunately this seems to be
the case with more Seagate drives:

http://lists.gnu.org/archive/html/bug-parted/2016-12/msg00002.html
https://patchwork.kernel.org/patch/5249211/