1 |
2010/2/14 Willie Wong <wwong@××××××××××××××.edu>: |
2 |
> On Mon, Feb 15, 2010 at 01:48:01AM +0100, Frank Steinmetzger wrote: |
3 |
<SNIP> |
4 |
>> |
5 |
>> action SS (1st) SS (2nd) SS+2 SS+4 SS+6 SS+8 |
6 |
>> -------------+----------+----------+----------+----------+----------+---------- |
7 |
>> untar portage 3m12.517 2m55.916 1m46.663 1m35.341 1m47.829 1m43.677 |
8 |
>> rm portage 4m11.109 3m54.950 3m18.820 3m11.378 3m21.804 3m12.433 |
9 |
>> cp 1GB file 0m21.383 0m13.558 0m14.920 0m12.813 0m13.407 0m13.681 |
10 |
|
11 |
|
12 |
|
13 |
|
14 |
> |
15 |
> Instead of guessing using this rather imprecise metric, why not just |
16 |
> look up the serial number of your drive and see what the physical |
17 |
> sector size is? If you don't want to open your box, you can usually |
18 |
> get the information from dmesg. |
19 |
|
20 |
|
21 |
hdparm capital eye works very nicely: |
22 |
|
23 |
gandalf ~ # hdparm -I /dev/sda |
24 |
|
25 |
/dev/sda: |
26 |
|
27 |
ATA device, with non-removable media |
28 |
Model Number: WDC WD10EARS-00Y5B1 |
29 |
Serial Number: WD-WCAV55464493 |
30 |
Firmware Revision: 80.00A80 |
31 |
Transport: Serial, SATA 1.0a, SATA II Extensions, |
32 |
SATA Rev 2.5, SATA Rev 2.6 |
33 |
Standards: |
34 |
Supported: 8 7 6 5 |
35 |
Likely used: 8 |
36 |
<SNIP> |
37 |
|
38 |
> |
39 |
> Only caveat: don't trust the harddrive to report accurate geometry. |
40 |
> This whole issue is due to the harddrives lying about their physical |
41 |
> geometry to be compatible with older versions of Windows. So the |
42 |
> physical sector size listed in dmesg may not be the real one. Which is |
43 |
> why you are advised to look up the model number on the vendor's |
44 |
> website yourself to determine the physical sector size. |
45 |
> |
46 |
> W |
47 |
> -- |
48 |
> Willie W. Wong wwong@××××××××××××××.edu |
49 |
|
50 |
Very true... |
51 |
|
52 |
Since this thread started and you help (me at least1) understand what |
53 |
I was dealing with I got in contact with Mark Lord - the developer and |
54 |
maintainer of the hdparm program. I was interested in seeing if we |
55 |
could get hdparm to recognize this aspect of the drive. He was very |
56 |
interested and asked me to send along additional info which he then |
57 |
analyzed and decided that, at least at this time, even drives that we |
58 |
__know__ are 4K sector sizes are not implementing any way of reading |
59 |
it from the drive's firmware which is supported, at least in the newer |
60 |
SATA specs. With that he decided that even for his own new 4K drives |
61 |
he cannot do anything except either assume they are 4K and partition |
62 |
appropriately or look up specs specifically as you suggest. |
63 |
|
64 |
Currently I'm partial to the idea that all my sector starting |
65 |
addresses will end in '000'. It's easy to remember and at most that |
66 |
wastes (I think) 512K bytes between sectors so it's not much in terms |
67 |
of the overall disk space. Just a couple of megabyte on a drive with 4 |
68 |
partitions. |
69 |
|
70 |
= Mark |