| To: | "Hen, Shmulik" <shmulik.hen@xxxxxxxxx> |
|---|---|
| Subject: | major bonding bug? |
| From: | Jeff Garzik <jgarzik@xxxxxxxxx> |
| Date: | Tue, 30 Dec 2003 11:31:15 -0500 |
| Cc: | "Noam, Amir" <amir.noam@xxxxxxxxx>, "Marom, Noam" <noam.marom@xxxxxxxxx>, netdev@xxxxxxxxxxx, fubar@xxxxxxxxxx |
| In-reply-to: | <E791C176A6139242A988ABA8B3D9B38A014C965E@xxxxxxxxxxxxxxxxxxxxxxx> |
| References: | <E791C176A6139242A988ABA8B3D9B38A014C965E@xxxxxxxxxxxxxxxxxxxxxxx> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Mutt/1.3.28i |
hmmm...
It looks like a lot of code now depends on global variable "bond_mode".
This looks very wrong... bonding mode should be per-interface, not
global to the entire driver. What happens when a user wants
BOND_MODE_ROUNDROBIN on bond0, and BOND_MODE_TLB on bond1?
Jeff
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH 1/3][bonding 2.4] Cannot remove and re-enslave the original active slave, Jeff Garzik |
|---|---|
| Next by Date: | Re: major bonding bug?, Amir Noam |
| Previous by Thread: | Re: [PATCH 11/19] bonding cleanup 2.4 - Re-org struct bonding members, Jeff Garzik |
| Next by Thread: | Re: major bonding bug?, Amir Noam |
| Indexes: | [Date] [Thread] [Top] [All Lists] |