* update pointers in struct hc_driver
* make ssb_ehci_detach more look like the pci version
* remove resume and suspend as they are never activated and not compiling
* replace ssb_ehci_start with ehci_run as it only calls this method
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@21427 3c298f89-4303-0410-b956-a3cf2f4a3e73
Disable some calls the WRT54G3G implementation does not like
and enable interrupts to allow hotplugging.
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@20240 3c298f89-4303-0410-b956-a3cf2f4a3e73
The defined offset is wrong and the fixup-code overrides it
later on so that it never gets used for most PCI devices.
Unfortunately the yenta-socket allocates its own resources
and crashes because of the wrong mem_offset.
It seems that the offset and fixup code came from 2.4 where
resource allocation was handled differently.
This patch removes the unneeded parts and thus enables
the yenta_socket on the WRT54G3G platform.
It was tested on Asus WL500G-Premium (v1 and v2), Linksys
WRT54G3G, Netgear WGT634U
Signed-off-by: Michael Buesch <mb@bu3sch.de>
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@20239 3c298f89-4303-0410-b956-a3cf2f4a3e73
The patch commited in r18413 was wrong.
This patch prevents prom_init_mem from scanning over 128MB ram.
This is from #6765 and #3177
Refresh all patches
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@20072 3c298f89-4303-0410-b956-a3cf2f4a3e73
