>> > >[Jack Wang] >> > >I don't think the statement below is error: >> > >child->linkrate = min(parent_phy->linkrate, child->max_linkrate); >> >> The parent_phy is same to child_phy, I don't think it's right. >> >[Jack Wang] The parent_phy->linkrate is report from DISCOVER response as >negotiated link rate(parent phy with child phy), so same is right. Parent_phy = &parent->ex_dev.ex_phy[phy_id]; (sas_ex_discover_expander) struct expander_device *parent_ex = &parent->ex_dev;(sas_ex_get_linkrate) struct ex_phy *phy = &parent_ex->ex_phy[i]; so, parent_phy and phy are same one variable. >> >> Yes, you can test like this. >> >> >Then expander1's linkrate will set to 3G, and expander2's linkrate will >> set >> >> >to 6G, that is correct. >> >> yes >> >> > But the connection rate will be 3G from hba to >> >> >expander2. >> >> How to configure the connection rate? >> >> Now, MVSAS driver get 6G from sas_ex_get_linkrate function, and set >> linkrate >> >> of OPEN address frame to the value. right? >> >> >> >> -- >> >[Jack Wang] >> >Connection rate is auto, link layer will insert deletable primitives to do >> >rate matching (sas2r15 7.14). >> >> >Yes, you should set the linkrate to that value . >> >> OK.MVSAS driver seem right. But, the problem is command fail because of link >> error, that is why I commit this patch. And my patch work fine. >> You can reproduce the issue and fix it. Thanks! >[Jack Wang] I re-consider it , you maybe should compare the HBA linkrate >with the child phy linkrate and chose a minor. Yes, how about this: - child->linkrate = min(parent_phy->linkrate, child->max_linkrate); + child->linkrate = min(parent->max_linkrate, child->max_linkrate); The max link rate of parent have HBA link rate info: dev->max_linkrate = port->linkrate; (in sas_get_port_device function). -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
