| 1 |
On Wed, Aug 14, 2013 at 12:18:41PM +0700, Pandu Poluan wrote: |
| 2 |
> Hello list! |
| 3 |
> |
| 4 |
> My company has 2 HP DL585 G5 servers and 5 Dell R... something servers. All |
| 5 |
> using AMD processors. They currently are acting as XenServer hosts. |
| 6 |
> |
| 7 |
> How do I determine the 'least common denominator' for Gentoo VMs (running |
| 8 |
> as XenServer guests), especially for gcc flags? |
| 9 |
> |
| 10 |
> I know that the (theoretical) best performance is to use -march=native , |
| 11 |
> but since the processors of the HP servers are not exactly the same as the |
| 12 |
> Dell's, I'm concerned that compiling with -march=native will render the VMs |
| 13 |
> unable to migrate between the different hosts. |
| 14 |
> |
| 15 |
> Note: Yes I know the HP servers are much older than the Dell ones, but if I |
| 16 |
> go -march=native then perform an emerge when the guest is on the Dell host, |
| 17 |
> the guest VM might not be able to migrate to the older HPs. |
| 18 |
|
| 19 |
To check what options CFLAGS set as -march=native would use: |
| 20 |
gcc -march=native -E -v - </dev/null 2>&1 | sed -n 's/.* -v - //p' |
| 21 |
(the first thing in the output is what CPU -march=native would enable) |
| 22 |
|
| 23 |
Then you can run: |
| 24 |
diff -u <(gcc -Q --help=target) <(gcc -march=native -Q --help=target) |
| 25 |
to display target-specific options, versus native ones. |
| 26 |
|
| 27 |
Assuming the 2 HP servers are identical, and the 5 Dell servers are identical, |
| 28 |
you then only need to get the commonality of two processors (HP and Dell). |
| 29 |
Since they're both AMD, you should have a good set of common features to help |
| 30 |
you determine that "least common denominator", or target. |
| 31 |
-- |
| 32 |
Happy Penguin Computers >') |
| 33 |
126 Fenco Drive ( \ |
| 34 |
Tupelo, MS 38801 ^^ |
| 35 |
support@×××××××××××××××××××××.com |
| 36 |
662-269-2706 662-205-6424 |
| 37 |
http://happypenguincomputers.com/ |
| 38 |
|
| 39 |
A: Because it messes up the order in which people normally read text. |
| 40 |
Q: Why is top-posting such a bad thing? |
| 41 |
A: Top-posting. |
| 42 |
Q: What is the most annoying thing in e-mail? |
| 43 |
|
| 44 |
Don't top-post: http://en.wikipedia.org/wiki/Top_post#Top-posting |
Archives