2.4.6 kernel on MCP750 board

2.4.6 kernel on MCP750 board

[Rudolf Ladyzhenskii]

Hi, all

I am trying to put the latest kernel on Motorola MCP750 card. 2.2 kernel
used to work fine, but 2.4.6 have some PCI problems.
It can not allocate resources to network interface and, as a result, I can
not boot.
Startup dump attached.

Any help is appreciated,
Thanks,
Rudolf


PCI: Probing PCI hardware
Scanning bus 00
Found 00:00 [1057/4801] 000600 00
Fixup res 1 (200) of dev 00:00.0: 3c000000 -> fc000000
Found 00:58 [1106/0586] 000601 00
Found 00:59 [1106/0571] 000101 00
Found 00:5a [1106/3038] 000c03 00
Found 00:5b [1106/3040] 000000 00
Found 00:70 [1011/0009] 000200 00
Fixup res 1 (200) of dev 00:0e.0: 3bffff00 -> fbffff00
PCI:00:0e.0 Resource 6 [fffc0000-ffffffff] is unassigned
Found 00:a0 [1011/0026] 000604 01
Fixups for bus 00
Memory resource not set for host bridge 0
Scanning behind PCI bridge 00:14.0, config 010100, pass 0
Scanning bus 01
Fixups for bus 01
Unknown bridge resource 0: assuming transparent
Unknown bridge resource 1: assuming transparent
PCI: Unable to handle 64-bit address space for
Bus scan for 01 returning with max=01
Scanning behind PCI bridge 00:14.0, config 010100, pass 1
Bus scan for 00 returning with max=01
Setting PCI interrupts for a "Mesquite cPCI (MCP750)"
PCI:00:00.0: Resource 1: fc000000-fc03ffff (f=200)
PCI: Cannot allocate resource region 1 of device 00:00.0
PCI:00:0b.1: Resource 0: 0000fff0-0000fff7 (f=101)
PCI: Cannot allocate resource region 0 of device 00:0b.1
PCI: parent is c01cf030: 00000000-ffffffff (f=100)
PCI:00:0b.1: Resource 1: 0000ffec-0000ffef (f=10d)
PCI: Cannot allocate resource region 1 of device 00:0b.1
PCI: parent is c01cf030: 00000000-ffffffff (f=100)
PCI:00:0b.1: Resource 2: 0000ffe0-0000ffe7 (f=101)
PCI: Cannot allocate resource region 2 of device 00:0b.1
PCI: parent is c01cf030: 00000000-ffffffff (f=100)
PCI:00:0b.1: Resource 3: 0000ffdc-0000ffdf (f=10d)
PCI: Cannot allocate resource region 3 of device 00:0b.1
PCI: parent is c01cf030: 00000000-ffffffff (f=100)
PCI:00:0b.1: Resource 4: 0000ffc0-0000ffcf (f=101)
PCI: Cannot allocate resource region 4 of device 00:0b.1
PCI: parent is c01cf030: 00000000-ffffffff (f=100)
PCI:00:0b.2: Resource 4: 0000ffa0-0000ffbf (f=101)
PCI: Cannot allocate resource region 4 of device 00:0b.2
PCI: parent is c01cf030: 00000000-ffffffff (f=100)
PCI:00:0e.0: Resource 0: 0fffff00-0fffff7f (f=101)
PCI: Cannot allocate resource region 0 of device 00:0e.0
PCI: parent is c01cf030: 00000000-ffffffff (f=100)
PCI:00:0e.0: Resource 1: fbffff00-fbffff7f (f=200)
PCI: Cannot allocate resource region 1 of device 00:0e.0
PCI: Failed to allocate resource 0(1000-1007) for 00:0b.1
PCI: Failed to allocate resource 1(1000-1003) for 00:0b.1
PCI: Failed to allocate resource 2(1000-1007) for 00:0b.1
PCI: Failed to allocate resource 3(1000-1003) for 00:0b.1
PCI: Failed to allocate resource 4(1000-100f) for 00:0b.1
PCI: Failed to allocate resource 4(1000-101f) for 00:0b.2
PCI: Failed to allocate resource 0(1000-107f) for 00:0e.0
PCI: Failed to allocate resource 1(c0000000-bfffffff) for 00:0e.0
Activating ISA DMA hang workarounds.
PCI: Via IRQ fixup for 00:0b.2, from 16 to 0
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Starting kswapd v1.8
initialize_kbd: Keyboard reset failed, no ACK
Serial driver version 5.05a (2001-03-20) with MANY_PORTS SHARE_IRQ
SERIAL_PCI en
abled
keyboard: Timeout - AT keyboard not present?
keyboard: Timeout - AT keyboard not present?
block: queued sectors max/low 20002kB/6667kB, 64 slots per queue
Linux Tulip driver version 0.9.15-pre6 (July 2, 2001)
PCI: Unable to reserve I/O region #1:80@1000 for device 00:0e.0



/************************/
Rudolf Ladyzhenskii
Senior DSP Engineer
Advanced Communication Technologies
ph.  +61 3 8080 8215
fax. +61 3 9672 8800
Level 9, 190 Queen Street,
Melbourne, 3000
Australia
/***********************/


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: 2.4.6 kernel on MCP750 board

[Matt Porter]

On Wed, Jul 18, 2001 at 09:54:41AM +1000, Rudolf Ladyzhenskii wrote:
>
> Hi, all
>
> I am trying to put the latest kernel on Motorola MCP750 card. 2.2 kernel
> used to work fine, but 2.4.6 have some PCI problems.
> It can not allocate resources to network interface and, as a result, I can
> not boot.
> Startup dump attached.

The PCI subsystem is shooting itself in the foot when it sees a P2P
bridge with I/O and Mem windows disabled.  That's the case when you
plug in a system master (that uses a transparent bridge to cPCI) in
by itself.  A peripheral card in any slot will avoid the issue.

Apply the following patch (current linuxppc_2_4_devel / 2.4.7pre8) to
cure the problem:

===== drivers/pci/pci.c 1.16 vs edited =====
--- 1.16/drivers/pci/pci.c	Wed Jul  4 18:46:09 2001
+++ edited/drivers/pci/pci.c	Thu Jul 19 14:47:38 2001
@@ -957,13 +957,9 @@
 		res->start = base;
 		res->end = limit + 0xfff;
 		res->name = child->name;
-	} else {
-		/*
-		 * Ugh. We don't know enough about this bridge. Just assume
-		 * that it's entirely transparent.
-		 */
-		printk(KERN_ERR "Unknown bridge resource %d: assuming transparent\n", 0);
-		child->resource[0] = child->parent->resource[0];
+	} else if (base > limit) {
+		/* Disabled bridge windows get no resources */
+		printk(KERN_DEBUG "PCI: Found disabled bridge I/O window\n");
 	}

 	res = child->resource[1];
@@ -976,10 +972,9 @@
 		res->start = base;
 		res->end = limit + 0xfffff;
 		res->name = child->name;
-	} else {
-		/* See comment above. Same thing */
-		printk(KERN_ERR "Unknown bridge resource %d: assuming transparent\n", 1);
-		child->resource[1] = child->parent->resource[1];
+	} else if (base > limit) {
+		/* Disabled bridge windows get no resources */
+		printk(KERN_DEBUG "PCI: Found disabled bridge memory window\n");
 	}

 	res = child->resource[2];
@@ -1003,10 +998,9 @@
 		res->start = base;
 		res->end = limit + 0xfffff;
 		res->name = child->name;
-	} else {
-		/* See comments above */
-		printk(KERN_ERR "Unknown bridge resource %d: assuming transparent\n", 2);
-		child->resource[2] = child->parent->resource[2];
+	} else if (base > limit) {
+		/* Disabled bridge windows get no resources */
+		printk(KERN_NOTICE "PCI: Found disabled bridge prefetch window\n");
 	}
 }


--
Matt Porter
MontaVista Software, Inc.
mporter@mvista.com

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: 2.4.6 kernel on MCP750 board

[Olaf Hering]

On Thu, Jul 19, Matt Porter wrote:

>
> On Wed, Jul 18, 2001 at 09:54:41AM +1000, Rudolf Ladyzhenskii wrote:
> >
> > Hi, all
> >
> > I am trying to put the latest kernel on Motorola MCP750 card. 2.2 kernel
> > used to work fine, but 2.4.6 have some PCI problems.
> > It can not allocate resources to network interface and, as a result, I can
> > not boot.
> > Startup dump attached.
>
> The PCI subsystem is shooting itself in the foot when it sees a P2P
> bridge with I/O and Mem windows disabled.  That's the case when you
> plug in a system master (that uses a transparent bridge to cPCI) in
> by itself.  A peripheral card in any slot will avoid the issue.
>
> Apply the following patch (current linuxppc_2_4_devel / 2.4.7pre8) to
> cure the problem:

That fixed my mtx+, thanks :)
Changeset 1.152 in _2_4 broke it.



Gruss Olaf

--
 $ man clone

BUGS
       Main feature not yet implemented...

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

RE: 2.4.6 kernel on MCP750 board

[Rudolf Ladyzhenskii]

Thanks, Matt

Your patch did make things a bit better. Now it tries to initialize the
tulip, but I do not think it is very happy.
Kernel tries to send BOOTP requests, but they never reach the server.
And tulip driver complains about missing EEPROM. Old kernel was finding the
EEPROM OK. Also, MAC address is changing between two kernels somehow. I
assume it is due to the detection of an EEPROM.

I've included updated start up dump.

<Skipped>

Linux/PPC load: console=ttyS0,9600 console=tty0 root=/dev/nfs
nfsroot=/usr/local
/opt/hardhat/devkit/ppc/7xx/target ip=on
Uncompressing Linux...done.
Now booting the kernel
Total memory = 32MB; using 128kB for hash table (at c01e0000)
Linux version 2.4.6 (marka@actmel042.act.domain) (gcc version 2.95.2
19991030 (2
.95.3 prerelease/franzo)) #39 Mon Jul 23 10:18:32 EST 2001
PReP architecture
On node 0 totalpages: 8192
zone(0): 8192 pages.
zone(1): 0 pages.
zone(2): 0 pages.
Kernel command line: console=ttyS0,9600 console=tty0 root=/dev/nfs
nfsroot=/usr/
local/opt/hardhat/devkit/ppc/7xx/target ip=on
OpenPIC Version 1.3 (2 CPUs and 16 IRQ sources) at fc000000
OpenPIC timer frequency is 8.333848 MHz
time_init: decrementer frequency = 16.666442 MHz
Calibrating delay loop... 465.30 BogoMIPS
Memory: 30228k available (1140k kernel code, 576k data, 100k init, 0k
highmem)
Dentry-cache hash table entries: 4096 (order: 3, 32768 bytes)
Inode-cache hash table entries: 2048 (order: 2, 16384 bytes)
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
Buffer-cache hash table entries: 1024 (order: 0, 4096 bytes)
Page-cache hash table entries: 8192 (order: 3, 32768 bytes)
POSIX conformance testing by UNIFIX
PCI: Probing PCI hardware
Fixup res 1 (200) of dev 00:00.0: 3c000000 -> fc000000
Fixup res 1 (200) of dev 00:0e.0: 3bffff00 -> fbffff00
PCI:00:0e.0 Resource 6 [fffc0000-ffffffff] is unassigned
Memory resource not set for host bridge 0
Fixup res 0 (1208) of dev 01:0d.0: 3bef0000 -> fbef0000
Fixup res 1 (200) of dev 01:0d.0: 3beef000 -> fbeef000
Fixup res 2 (200) of dev 01:0d.0: 3bc00000 -> fbc00000
Fixup res 3 (200) of dev 01:0d.0: 3bb80000 -> fbb80000
PCI: Unable to handle 64-bit address space for
Setting PCI interrupts for a "Mesquite cPCI (MCP750)"
PCI:00:00.0: Resource 1: fc000000-fc03ffff (f=200)
PCI: Cannot allocate resource region 1 of device 00:00.0
PCI:  parent is c01ce04c: c0000000-ffffffff (f=200)
PCI:00:0b.1: Resource 0: 0000fff0-0000fff7 (f=101)
PCI: Cannot allocate resource region 0 of device 00:0b.1
PCI:  parent is c01ce030: 00000000-ffffffff (f=100)
PCI:00:0b.1: Resource 1: 0000ffec-0000ffef (f=10d)
PCI: Cannot allocate resource region 1 of device 00:0b.1
PCI:  parent is c01ce030: 00000000-ffffffff (f=100)
PCI:00:0b.1: Resource 2: 0000ffe0-0000ffe7 (f=101)
PCI: Cannot allocate resource region 2 of device 00:0b.1
PCI:  parent is c01ce030: 00000000-ffffffff (f=100)
PCI:00:0b.1: Resource 3: 0000ffdc-0000ffdf (f=10d)
PCI: Cannot allocate resource region 3 of device 00:0b.1
PCI:  parent is c01ce030: 00000000-ffffffff (f=100)
PCI:00:0b.1: Resource 4: 0000ffc0-0000ffcf (f=101)
PCI: Cannot allocate resource region 4 of device 00:0b.1
PCI:  parent is c01ce030: 00000000-ffffffff (f=100)
PCI:00:0b.2: Resource 4: 0000ffa0-0000ffbf (f=101)
PCI: Cannot allocate resource region 4 of device 00:0b.2
PCI:  parent is c01ce030: 00000000-ffffffff (f=100)
PCI:00:0e.0: Resource 0: 0fffff00-0fffff7f (f=101)
PCI: Cannot allocate resource region 0 of device 00:0e.0
PCI:  parent is c01ce030: 00000000-ffffffff (f=100)
PCI:00:0e.0: Resource 1: fbffff00-fbffff7f (f=200)
PCI: Cannot allocate resource region 1 of device 00:0e.0
PCI:  parent is c01ce04c: c0000000-ffffffff (f=200)
PCI:01:0d.0: Resource 0: fbef0000-fbefffff (f=1208)
PCI: Cannot allocate resource region 0 of device 01:0d.0
PCI:  parent is c02ba938: fbb00000-fbefffff (f=200)
PCI:01:0d.0: Resource 1: fbeef000-fbeeffff (f=200)
PCI: Cannot allocate resource region 1 of device 01:0d.0
PCI:  parent is c02ba938: fbb00000-fbefffff (f=200)
PCI:01:0d.0: Resource 2: fbc00000-fbdfffff (f=200)
PCI: Cannot allocate resource region 2 of device 01:0d.0
PCI:  parent is c02ba938: fbb00000-fbefffff (f=200)
PCI:01:0d.0: Resource 3: fbb80000-fbbfffff (f=200)
PCI: Cannot allocate resource region 3 of device 01:0d.0
PCI:  parent is c02ba938: fbb00000-fbefffff (f=200)
PCI: Failed to allocate resource 0(1000-1007) for 00:0b.1
PCI: Failed to allocate resource 1(1000-1003) for 00:0b.1
PCI: Failed to allocate resource 2(1000-1007) for 00:0b.1
PCI: Failed to allocate resource 3(1000-1003) for 00:0b.1
PCI: Failed to allocate resource 4(1000-100f) for 00:0b.1
PCI: Failed to allocate resource 4(1000-101f) for 00:0b.2
PCI: Failed to allocate resource 0(1080-10ff) for 00:0e.0
  got res[c0000000:c000007f] for resource 1 of PCI device 1011:0009
PCI: Failed to allocate resource 0(10000000-1000ffff) for 01:0d.0
PCI: Failed to allocate resource 1(10000000-10000fff) for 01:0d.0
PCI: Failed to allocate resource 2(10000000-101fffff) for 01:0d.0
PCI: Failed to allocate resource 3(10280000-102fffff) for 01:0d.0
Activating ISA DMA hang workarounds.
PCI: Via IRQ fixup for 00:0b.2, from 16 to 0
isapnp: Scanning for PnP cards...
isapnp: No Plug & Play device found
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Starting kswapd v1.8
initialize_kbd: Keyboard reset failed, no ACK
Serial driver version 5.05a (2001-03-20) with MANY_PORTS SHARE_IRQ
SERIAL_PCI IS
APNP enabled
keyboard: Timeout - AT keyboard not present?
keyboard: Timeout - AT keyboard not present?
ttyS00 at 0x03f8 (irq = 4) is a 16550A
ttyS01 at 0x02f8 (irq = 3) is a 16550A
block: queued sectors max/low 20002kB/6667kB, 64 slots per queue
Linux Tulip driver version 0.9.15-pre6 (July 2, 2001)
tulip0:  Missing EEPROM, this interface may not work correctly!
eth0: Digital DS21140 Tulip rev 34 at 0x1080, EEPROM not present,
00:4C:69:6E:75
:79, IRQ 18.
PPP generic driver version 2.4.1
PPP Deflate Compression module registered
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP, IGMP
IP: routing cache hash table of 512 buckets, 4Kbytes
TCP: Hash tables configured (established 2048 bind 2048)
Sending BOOTP requests .<7>eth0: Media selection tick, 100baseTx, status
fffffff
f mode ffffffff SIA ffffffff ffffffff ffffffff ffffffff.
.....<7>eth0: Media selection tick, 100baseTx, status ffffffff mode ffffffff
SIA
 ffffffff ffffffff ffffffff ffffffff.
 timed out!
IP-Config: Retrying forever (NFS root)...
..


Rudolf

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: 2.4.6 kernel on MCP750 board

[Matt Porter]

On Mon, Jul 23, 2001 at 11:03:12AM +1000, Rudolf Ladyzhenskii wrote:
>
> Thanks, Matt
>
> Your patch did make things a bit better. Now it tries to initialize the
> tulip, but I do not think it is very happy.
> Kernel tries to send BOOTP requests, but they never reach the server.
> And tulip driver complains about missing EEPROM. Old kernel was finding the
> EEPROM OK. Also, MAC address is changing between two kernels somehow. I
> assume it is due to the detection of an EEPROM.
>
> I've included updated start up dump.

Are you running linuxppc_2_4_devel or a kernel.org 2.4.6?  What rev. of
mcp750?

--
Matt Porter
MontaVista Software, Inc.
mporter@mvista.com

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

RE: 2.4.6 kernel on MCP750 board

[Rudolf Ladyzhenskii]

Hi,

>
> On Mon, Jul 23, 2001 at 11:03:12AM +1000, Rudolf Ladyzhenskii wrote:
> >
> > Thanks, Matt
> >
> > Your patch did make things a bit better. Now it tries to
> initialize the
> > tulip, but I do not think it is very happy.
> > Kernel tries to send BOOTP requests, but they never reach
> the server.
> > And tulip driver complains about missing EEPROM. Old kernel
> was finding the
> > EEPROM OK. Also, MAC address is changing between two
> kernels somehow. I
> > assume it is due to the detection of an EEPROM.
> >
> > I've included updated start up dump.
>
> Are you running linuxppc_2_4_devel or a kernel.org 2.4.6?
> What rev. of
> mcp750?
>

We are running 2.4.6 kernel from kernel.org.
The latest kernel opn Monta Vista side did not work at all on MCP750 and on
our custom 8260 based board. Kernel woud start, but none of rhe external
interfaces would work (console, Ethernet). Eventually (by flashing LEDs on
our board) I traced it down to the transmit section of console code. It will
sit in the endless loop waiting for Tx buffer go empty. Looks like
interrupts were disabled by some reason.

Anyway, we downloaded latest 2.4.6 from kernel.org This works fine on our
8260 board.
We use MCP750 REV.A board

Rudolf

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: 2.4.6 kernel on MCP750 board

[Matt Porter]

On Wed, Jul 25, 2001 at 09:48:24AM +1000, Rudolf Ladyzhenskii wrote:
> > Are you running linuxppc_2_4_devel or a kernel.org 2.4.6?
> > What rev. of
> > mcp750?
> >
>
> We are running 2.4.6 kernel from kernel.org.
> The latest kernel opn Monta Vista side did not work at all on MCP750 and on
> our custom 8260 based board. Kernel woud start, but none of rhe external
> interfaces would work (console, Ethernet). Eventually (by flashing LEDs on
> our board) I traced it down to the transmit section of console code. It will
> sit in the endless loop waiting for Tx buffer go empty. Looks like
> interrupts were disabled by some reason.

The kernel.org releases generally don't work on any given PPC platform
outside of pmacs.  I can't comment on 8260 systems since I don't have
any, but the patch I gave you was tested on my Rev. J MCP750 versus
a linuxppc_2_4_devel tree pulled from bitmover a few days ago.  I've
got to run now but I'll try my board tomorrow with the latest batch
of changesets just as a sanity check that things still work.  You
shouldn't be seeing all those resource explosions.

> Anyway, we downloaded latest 2.4.6 from kernel.org This works fine on our
> 8260 board.
> We use MCP750 REV.A board

Are you sure you didn't look at the Revision marking of the memory
mezzanine PCB?  Check the sticker next to the serial number on the
back of the MCP750 PCB.  For example, mine is 01J indicating a Rev. J
board.  I hope yours is later than a Rev. A. :)

--
Matt Porter
MontaVista Software, Inc.
mporter@mvista.com

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/