sweet thanks...not that waiting a few hours wouldve killed me...dice wrote:Not yet. Maybe by morningjimlynch11 wrote:is it in portage...its not showing up in mine...
Edit: If you rename mm-sorces-2.5.68-r2.ebuild to mm-sources-2.5.68-r3.ebuild it works just dandy
Under "File systems"greg32 wrote:I'm having problems getting the 2.5.68 dev kernel working properly. It's all compiled, and I am in it now. I added the devpts stuff to my fstab, but it fails to mount. This would probably be cause I could not find anywhere in the kernel where I should enable devpts, and so I left it for now. But as a result, I am having problems getting any terminal to work....
where in the kernel is the devpts stuff? I enable all the usual devfs stuff in there, but nothing mention devpts.
regards Greg
Code: Select all
b44: eth0: Link is downDo you have ACPI installed? With ACPI enabled my network card driver loads and I can see it with ifconfig but it won't transfer anything. Add APCI=off and everything works fine.poisson wrote:I finally make 2.5.68 boot, the main problem is network.
My Broadcom 4401 works with 2.4.20 (bcm440 module), but not with new kernel driver (b44 module):"ifconfig" recognise eth0 device, but "route" refuses to add an entry in it's table.Code: Select all
b44: eth0: Link is down
Have I to install iproute2 package?
Thanks
Thanks for those news, I'll give it a try and report my results here...darktux wrote:Well.... I'm using 2.5.68-mm3 and it lloks like the problem with browsing/xmms is much better, although not 100% solved
I thought this patch was in the mm3 kernel allready? Does it suffice to compile the mm3 kernel or do I need to reverse this patch in the mm3 kernel. tnx and regards...aleph-3 wrote:yokem55:
ftp://ftp.kernel.org/pub/linux/kernel/p ... nounce.txt
says:
synaptics-mouse-support.patch
Add Synaptics touchpad tweaking to psmouse driver
I tried "acpi=off" as boot paramenter, acpi effectively has been disabled, but network is still not working.grover wrote:Do you have ACPI installed? With ACPI enabled my network card driver loads and I can see it with ifconfig but it won't transfer anything. Add APCI=off and everything works fine.
If you ever find a solution to this, let us know. The last kernel I tried that did not have these symptoms was mm-sources-2.5.66-r2, and it's been running fine for me till now, a part from some pcmcia issues...lurid wrote:Yeah, I have the same problem with sound skipping thats reported all over the place. With the plain dev-sources there is no problem at all. With mm-sources any time anything accesses the disk at all, music stops. Even switchinging workspaces in Fluxbox makes it skip. Are these problems being worked on? Is there something in the kernel that can be enabled/disabled to fix this?
bah, its been that way for ages. it seems to work for "everybody" except those with A7V motherboards (which is probably a lot of us). I'm no kernel hacker so I've been hoping that somebody who knows how to fix such things will catch it, but nothing has changed since I've been monitoring the problem more-or-less since the earliest days of the b44 driver.poisson wrote:I finally make 2.5.68 boot, the main problem is network.
My Broadcom 4401 works with 2.4.20 (bcm440 module), but not with new kernel driver (b44 module):"ifconfig" recognise eth0 device, but "route" refuses to add an entry in it's table.Code: Select all
b44: eth0: Link is down
Have I to install iproute2 package?
Thanks
unfortunately, the problem persists in 2.5.70. However, I'm happy to report that I have found a way to "port" the old bcm4400 driver to 2.5.70, and it seems to be working fine (am posting from it right now). Here is a patch against the 1.0.1 bcnet sources from Broadcomm (or are they from Asus?):darktux wrote:Lets just drop this. 2.5.70 is out!
Code: Select all
--- ../src/Makefile 2002-08-23 18:53:23.000000000 -0500
+++ Makefile 2003-05-27 17:57:25.000000000 -0500
@@ -2,10 +2,13 @@
# Makefile for building Linux Broadcom Gigabit ethernet driver as a module.
# $id$
+# set this to match your kernel compile
+CUSTOMCFLAGS=-O2 -mpreferred-stack-boundary=2 -march=athlon
+
# PREFIX may be set by the RPM build to set the effective root.
PREFIX=
ifeq ($(shell ls /lib/modules/`uname -r`/build > /dev/null 2>&1 && echo build),)
-LINUX=/usr/src/linux
+LINUX=/usr/src/linux-beta
else
LINUX=/lib/modules/`uname -r`/build
endif
@@ -16,7 +19,8 @@
CC = kgcc
endif
LD = ld
-CFLAGS=-DMODULE -D__KERNEL__ -DDBG=0 -Wall -Wstrict-prototypes -O6 -I$(LINUX)/include
+CFLAGS=-DMODULE -D__KERNEL__ -DDBG=0 -Wp,-MD,.bcm4400.o.d -Wstrict-prototypes -Wall -Wno-trigraphs -fno-strict-aliasing -fno-common -pipe $(CUSTOMCFLAGS) -I$(LINUX)/include -I$(LINUX)/include/asm-i386/mach-default -nostdinc -iwithprefix include -DKBUILD_BASENAME=bcm4400 -DKBUILD_MODNAME=bcm4400 -c
+CFLAGS2=-DMODULE -D__KERNEL__ -DDBG=0 -Wp,-MD,.bcm4400.mod.o.d -Wstrict-prototypes -Wall -Wno-trigraphs -fno-strict-aliasing -fno-common -pipe $(CUSTOMCFLAGS) -I$(LINUX)/include -I$(LINUX)/include/asm-i386/mach-default -nostdinc -iwithprefix include -DKBUILD_BASENAME=bcm4400 -DKBUILD_MODNAME=bcm4400 -c
ifeq ($(wildcard ~/rpmdir),)
rpmdir = /usr/src/redhat
@@ -40,22 +44,31 @@
CFLAGS += -DBCM_SMALL_DRV=1
endif
-all: bcm4400.o
+all: bcm4400.ko
tar: bcm4400.tgz
bcm4400.o: b44um.o b44proc.o b44lm.o
$(LD) -i $^ -o $@
-install: bcm4400.o
+bcm4400.mod.c: bcm4400.o
+ $(LINUX)/scripts/modpost $(LINUX)/vmlinux bcm4400.o
+
+bcm4400.mod.o: bcm4400.mod.c
+ $(CC) $(CFLAGS2) -o bcm4400.mod.o bcm4400.mod.c
+
+bcm4400.ko: bcm4400.o bcm4400.mod.o
+ $(LD) -m elf_i386 -r -o bcm4400.ko $^
+
+install: bcm4400.ko
@if [ -d $(PREFIX)/lib/modules/`uname -r`/kernel/drivers/addon/bcm4400 ];\
- then install -m 444 bcm4400.o $(PREFIX)/lib/modules/`uname -r`/kernel/drivers/addon/bcm4400;\
+ then install -m 444 bcm4400.ko $(PREFIX)/lib/modules/`uname -r`/kernel/drivers/addon/bcm4400;\
elif [ -d $(PREFIX)/lib/modules/`uname -r`/kernel/drivers/net/bcm ];\
- then install -m 444 bcm4400.o $(PREFIX)/lib/modules/`uname -r`/kernel/drivers/net/bcm;\
+ then install -m 444 bcm4400.ko $(PREFIX)/lib/modules/`uname -r`/kernel/drivers/net/bcm;\
elif [ -d $(PREFIX)/lib/modules/`uname -r`/kernel ];\
then mkdir -p $(PREFIX)/lib/modules/`uname -r`/kernel/drivers/net;\
- install -m 444 bcm4400.o $(PREFIX)/lib/modules/`uname -r`/kernel/drivers/net;\
+ install -m 444 bcm4400.ko $(PREFIX)/lib/modules/`uname -r`/kernel/drivers/net;\
else mkdir -p $(PREFIX)/lib/modules/`uname -r`/net;\
- install -m 444 bcm4400.o $(PREFIX)/lib/modules/`uname -r`/net/;\
+ install -m 444 bcm4400.ko $(PREFIX)/lib/modules/`uname -r`/net/;\
fi
@if [ "$(PREFIX)" = "" ]; then /sbin/depmod -a ;\
else echo " *** Run '/sbin/depmod -a' to update the module database.";\
@@ -82,4 +95,4 @@
@make SMALL=1
clean:
- -rm -f bcm4400.o b44um.o b44proc.o b44lm.o
+ -rm -f bcm4400.o b44um.o b44proc.o b44lm.o bcm4400.ko .tmp_bcm4400.o .bcm4400.o.cmd .bcm4400.mod.o.d bcm4400.mod.c bcm4400.mod.o .bcm4400.o.d
--- ../src/b44mm.h 2002-08-23 18:53:19.000000000 -0500
+++ b44mm.h 2003-05-27 16:31:44.000000000 -0500
@@ -28,7 +28,7 @@
#include <linux/version.h>
#ifdef MODULE
#ifdef MODVERSIONS
-#include <linux/modversions.h>
+#include <config/modversions.h>
#endif
#include <linux/module.h>
#else
--- ../src/b44um.c 2002-08-23 18:53:19.000000000 -0500
+++ b44um.c 2003-05-27 20:41:18.000000000 -0500
@@ -205,7 +205,7 @@
STATIC void bcm4400_timer(unsigned long data);
STATIC void bcm4400_tx_timeout(struct net_device *dev);
STATIC int bcm4400_start_xmit(struct sk_buff *skb, struct net_device *dev);
-STATIC void bcm4400_interrupt(int irq, void *dev_instance, struct pt_regs *regs);
+STATIC irqreturn_t bcm4400_interrupt(int irq, void *dev_instance, struct pt_regs *regs);
#ifdef BCM_TASKLET
STATIC void bcm4400_tasklet(unsigned long data);
#endif
@@ -611,7 +611,7 @@
}
-STATIC void
+STATIC irqreturn_t
bcm4400_interrupt(int irq, void *dev_instance, struct pt_regs *regs)
{
struct net_device *dev = (struct net_device *)dev_instance;
@@ -622,16 +622,16 @@
#endif
if (!pDevice->InitDone)
- return;
+ return IRQ_HANDLED;
if (atomic_read(&pUmDevice->intr_sem))
- return;
+ return IRQ_HANDLED;
if (test_and_set_bit(0, (void*)&pUmDevice->interrupt)) {
printk(KERN_ERR "%s: Duplicate entry of the interrupt handler by "
"processor %d.\n",
dev->name, hard_smp_processor_id());
- return;
+ return IRQ_HANDLED;
}
@@ -665,7 +665,7 @@
pUmDevice->tx_queued = 0;
netif_wake_queue(dev);
}
- return;
+ return IRQ_HANDLED;
}