On 21/03/17 21:49, Michael Brown wrote:
What is the purpose of this patch? The "desired result" in the commit
log message appears to be undefined.
Thanks,
Michael
Hi Michael,
Extremely sorry, I completely missed this email.
Purpose of this patch is to fix a problem, when there are two different
DHCPACK, one from DHCP server and other from proxy DHCP server, and both
DHCP contain different Vendor Class Identifier (option 60).
After processing both DHCPACK, there will two settings registered, one
with DHCP_SETTING_NAME and another with PROXYDHCP_SETTING_NAME. Both
these settings will have DHCP_VENDOR_CLASS_ID in them, with different
values.
When a query to get DHCP_VENDOR_CLASS_ID, using
fetch_string_setting(NULL, &data) is made, most of time
DHCP_VENDOR_CLASS_ID from setting DHCP_SETTING_NAME would be returned,
which would not have "PXEClient" as value.
During pxeboot discovery, have_pxe_menu() need to look for
DHCP_VENDOR_CLASS_ID from PROXYDHCP_SETTING_NAME first (if
PROXYDHCP_SETTING_NAME is present, else look for DHCP_VENDOR_CLASS_ID
from all other setting). The patch does this by first finding setting
using find_setting(PROXYDHCP_SETTING_NAME) then passes that as an arg to
fetch_string_setting(). If there are no PROXYDHCP_SETTING_NAME setting,
then find_setting would return NULL and fetch_string_setting(NULL) will
try to lookup from all settings.
If the commit message of this patch is not clear I can modify it as you
want.
Sorry for the delay again,
Anoob.
_______________________________________________
ipxe-devel mailing list
ipxe-devel@lists.ipxe.org
https://lists.ipxe.org/mailman/listinfo.cgi/ipxe-devel
