On Sun, 09 Dec 2018 15:06:49 +0100 Andreas Beckmann <a...@debian.org> wrote:
> Package: sl-modem-dkms
> Version: 2.9.11~20110321-13
> Severity: serious
> User: debian...@lists.debian.org
> Usertags: piuparts
> Control: found -1 2.9.11~20110321-12
> 
> Hi,
> 
> sl-modem-dkms fails to build a module for 4.18.0-3-amd64 and 4.9.0-8-amd64

In fact it fails to build for any kernel version beginning with a 4,
because it doesn't recognise this as a "new" (2.6+) version.

[...]
> DKMS make.log for sl-modem-2.9.11~20110321 for kernel 4.18.0-3-amd64 (x86_64)
> Sun Dec  9 13:53:21 UTC 2018
> make: Entering directory 
> '/var/lib/dkms/sl-modem/2.9.11~20110321/build/drivers'
>         doing %.o: %.c
> cc -Wall -pipe -O3 -fomit-frame-pointer -D__KERNEL__ -DMODULE -DEXPORT_SYMTAB 
> `test -f /lib/modules/4.18.0-3-amd64/build/include/linux/modversions.h && 
> echo -DMODVERSIONS --include 
> /lib/modules/4.18.0-3-amd64/build/include/linux/modversions.h 
> -I/lib/modules/4.18.0-3-amd64/build/include`  -I. -I./../modem   -o 
> old_st7554.o -c old_st7554.c
> old_st7554.c:49:10: fatal error: linux/init.h: No such file or directory
>  #include <linux/init.h>
>           ^~~~~~~~~~~~~~
> compilation terminated.
[...]

Here it's invoking the compiler directly rather than using Kbuild.

This should be fixable with something like:

--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -86,7 +86,7 @@ remove-devices:
            $(RM) ${DESTDIR}/dev/slusb$(minor) ; ) echo -n
 
 else
-ifeq ($(shell grep 'UTS.*"\(2\.6\|3\)' 
$(KERNEL_DIR)/include/generated/utsrelease.h 
$(KERNEL_DIR)/include/linux/utsrelease.h $(KERNEL_DIR)/include/linux/version.h 
2>/dev/null),)
+ifeq ($(shell grep 'UTS.*"\(2\.6\|[3-9]\)' 
$(KERNEL_DIR)/include/generated/utsrelease.h 
$(KERNEL_DIR)/include/linux/utsrelease.h $(KERNEL_DIR)/include/linux/version.h 
2>/dev/null),)
 
 slusb-objs:= old_st7554.o
 obj:=.
--- END ---

Ben.

-- 
Ben Hutchings
Power corrupts.  Absolute power is kind of neat. - John Lehman


Attachment: signature.asc
Description: This is a digitally signed message part

Reply via email to