On Sun, 2009-02-22 at 15:10 +0000, Ben Hutchings wrote: > On Sun, 2009-02-22 at 11:37 +0100, Bastian Blank wrote: [...] > > Why is that a global variable? > > This was written by Jaswinder Singh; ask him. ;-) This was clearly wrong given that firmware was loaded in the probe function, not the module init function. Given that firmware is not needed outside of the probe function, I have removed the static firmware variable from this and the tehuti driver patches. > > > +index efaf84d..dec67e0 100644 > > > +--- a/drivers/net/tehuti.h > > > ++++ b/drivers/net/tehuti.h > > > +@@ -29,6 +29,7 @@ > > > + #include <linux/if_vlan.h> > > > + #include <linux/interrupt.h> > > > + #include <linux/vmalloc.h> > > > ++#include <linux/firmware.h> > > > + #include <asm/byteorder.h> > > > > This looks wrong. Nothing from firmware.h is used in this header. > > Right, this seems to be a holdover from an earlier version of the patch > which kept referenes to the firmware in struct bdx_priv. In fact the #include is here because tehuti.c only #includes tehuti.h. This is strange but I don't see any point in changing this. > > > + [ Ben Hutchings ] > > > + * Remove firmware from drivers and make them use request_firmware(): > > > + - mga (closes: #502666) > > > + - qla1280 (closes: #502667) > > > + - r128 (closes: #494007) > > > + - radeon (closes: #494009) > > > + - tehuti (closes: #501153) > > > + - typhoon (closes: #502669) > > > > The patches still needs to be accepted upstream. > > The patch for qla1280 has been. The others, I'll try to push again. I have now submitted the other patches to drm and net maintainers and lists. Ben.
Attachment:
signature.asc
Description: This is a digitally signed message part