Subject: kern/15457: 1-5 branch missing cardbusdevs entry for Realtek 8139
To: None <gnats-bugs@gnats.netbsd.org>
From: None <John.P.Darrow@wheaton.edu>
List: netbsd-bugs
Date: 02/01/2002 14:57:29
>Number:         15457
>Category:       kern
>Synopsis:       1-5 branch missing cardbusdevs entry for Realtek 8139
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Feb 01 12:58:01 PST 2002
>Closed-Date:
>Last-Modified:
>Originator:     John Darrow
>Release:        release-1-5 branch supped as of 2002-01-30
>Organization:
Computing Services, Wheaton College, Wheaton, IL
>Environment:
NetBSD michal.wheaton.edu 1.5.3_ALPHA NetBSD 1.5.3_ALPHA (JDARROW) #0: Thu Jan 31 13:21:10 CST 2002     jdarrow@michal.wheaton.edu:/var/src/sys/arch/i386/compile/JDARROW i386


>Description:
Teemu Rinta-aho <teemu.rinta-aho@nomadiclab.com> wrote:
>How-To-Repeat:
Try to build GENERIC_LAPTOP kernel (e.g. via make release).  See
build die after a while.
	
>Fix:
Please apply the following patch to sys/dev/cardbus/cardbusdevs and
regenerate the generated files (make -f Makefile.cardbusdevs).  This
is basically a re-pullup of 1.19, but done by patch to avoid possible
fuzz issues (the line in question in the trunk revision is not at the
end of the file, like it is on the branch...)

Index: cardbusdevs
===================================================================
RCS file: /source/cvs/netbsd/release-1-5/src/sys/dev/cardbus/cardbusdevs,v
retrieving revision 1.10.2.2
diff -u -r1.10.2.2 cardbusdevs
--- cardbusdevs	2002/01/31 08:49:28	1.10.2.2
+++ cardbusdevs	2002/02/01 20:42:05
@@ -103,3 +103,4 @@
 
 /* Realtek (Creative Labs?) products */
 product REALTEK RT8138	0x8138	8138 Ethernet
+product REALTEK RT8139	0x8139	8139 Ethernet
>Release-Note:
>Audit-Trail:
>Unformatted:
 >I just updated the netbsd-1-5 branch from
 >anoncvs.netbsd.org but I can not compile the
 >kernel on my laptop. See the following:
 >
 >cc  -O2 -Werror -Wall -Wmissing-prototypes -Wstrict-prototypes
 >-Wpointer-arith -Wno-uninitialized -Wno-main -I. -I../../../../arch
 >-I../../../.. -nostdinc -DLKM -DMAXUSERS=32 -D_KERNEL -Di386  -c
 >../../../../dev/cardbus/if_rtk_cardbus.c
 >../../../../dev/cardbus/if_rtk_cardbus.c:110:
 >`CARDBUS_PRODUCT_REALTEK_RT8139' undeclared here (not in a function)
 >../../../../dev/cardbus/if_rtk_cardbus.c:110: initializer element for
 >`rtk_cardbus_devs[3].rtk_did' is not constant
 >*** Error code 1
 >
 >Stop.
 
 I encountered this too.  It's basically a pullup merge botch by CVS.
 What happened is that the latest pullup (1.10.2.2) included a pullup
 of 1.13, which, along with adding a couple of new devices, also
 deleted an extra blank line at the end of the file.  However, due to
 a previous pullup of a later trunk revision (1.19), that line was not
 blank, but contained the RT8139 identifier.  Unfortunately, CVS didn't
 catch the conflict, and happily deleted the line.