On Mon, 2007-04-09 at 20:48 -0500, Larry Finger wrote:
> > Speaking of the build system, you may want to have a warning in case if
> > anything that is compiled as a module (either bcm43xx or softmac) is
> > already available and linked into the kernel. Make sure the checks
> > don't affect "make clean" and similar maintenance commands.
>
> I'm not sure how to do this, but will do some more investigation.
Actually, your makefiles only rely on the kernel build system for
compilation, so checking is not an issue.
This patch would do the check. but it's a bit on the strict side. It
will disallow compilation of the driver only if ieee80211 is in the
kernel. But it may be a good thing because ieee80211 may be without
SoftMAC or incompatible, and distros won't link it statically.
Signed-off-by: Pavel Roskin <[EMAIL PROTECTED]>
--- bcm43xx/Makefile
+++ bcm43xx/Makefile
@@ -32,6 +32,10 @@
ifneq ($(KERNELRELEASE),)
# call from kernel build system
+ifeq ($(CONFIG_BCM43XX),y)
+$(error CONFIG_BCM43XX is already in the kernel)
+endif
+
obj-m := bcm43xx.o
bcm43xx-objs := bcm43xx_main.o bcm43xx_dma.o $(DEBUGFS_OBJ) \
bcm43xx_radio.o bcm43xx_phy.o \
--- ieee80211/Makefile
+++ ieee80211/Makefile
@@ -20,6 +20,10 @@
ifneq ($(KERNELRELEASE),)
# call from kernel build system
+ifeq ($(CONFIG_IEEE80211),y)
+$(error CONFIG_IEEE80211 is already in the kernel)
+endif
+
obj-m := ieee80211.o
obj-m += ieee80211_crypt.o
obj-m += ieee80211_crypt_wep.o
> There is a new version on the FTP site. I would appreciate seeing if that
> will work for you.
Yes, it's compiling just fine! You'll hear from me again if there are
any runtime issues.
--
Regards,
Pavel Roskin
_______________________________________________
Bcm43xx-dev mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev
