Skip to contentSkip to conversation

Build 64bit on RaQ2

DISCUSSION Jim Gifford 3 participants 7 messages Mar 25, 2005 – Apr 2, 2005

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

Has anyone had any luck compiling a 64 bit version on the RaQ2. I can
get it to compile but, it locks up during boot up.

elf64: 00080000 - 0042fd3f (ffffffff,803e6000) (ffffffff,8000000)
elf64: ffffffff,80080000 (8008000) 3731589t + 134331t

That's all I got during bootup, no error messages or anything.

Signature · 3 lines
----
Jim Gifford
[email protected]
#1 lore raw
Quoted 7 lines · Jim Gifford

Has anyone had any luck compiling a 64 bit version on the RaQ2. I can
get it to compile but, it locks up during boot up.

elf64: 00080000 - 0042fd3f (ffffffff,803e6000) (ffffffff,8000000)
elf64: ffffffff,80080000 (8008000) 3731589t + 134331t

That's all I got during bootup, no error messages or anything.

Peter Horton had some experimental code in the kernel at one point to try this. I managed to get a 2.6.9 (I think, maybe 2.6.10) mips64 kernel to boot on cobalt, but it was pretty useless. Poor machine was slower than molasses uphill in winter.

I believe cobalt would run better using the o64 kernel hack than pure n64 kernel (-mabi=64), but cobalt needs some fixups to its spaces.h before an o64 kernel will build on it.

--Kumba

Signature · 1 line
"Such is oft the course of deeds that move the wheels of the world: small hands do them because they must, while the eyes of the great are elsewhere." --Elrond
#2 lore raw
In reply to Jim Gifford
Quoted 8 lines · Jim Gifford

Has anyone had any luck compiling a 64 bit version on the RaQ2. I can
get it to compile but, it locks up during boot up.

elf64: 00080000 - 0042fd3f (ffffffff,803e6000) (ffffffff,8000000)
elf64: ffffffff,80080000 (8008000) 3731589t + 134331t

That's all I got during bootup, no error messages or anything.

As a starting point you need to ensure the "cpu_has_llsc" is false for 64-bit Cobalt kernels. LLD/SCD is broken on RM5230/5231. There is an experimental patch for 2.6.9 on
http://www.colonel-panic.org/cobalt-mips.

P.

#3 lore raw
Quoted 14 lines · Peter Horton
Quoted 10 lines · Jim Gifford

Has anyone had any luck compiling a 64 bit version on the RaQ2. I can
get it to compile but, it locks up during boot up.

elf64: 00080000 - 0042fd3f (ffffffff,803e6000) (ffffffff,8000000)
elf64: ffffffff,80080000 (8008000) 3731589t + 134331t

That's all I got during bootup, no error messages or anything.

Quoted 8 lines

As a starting point you need to ensure the "cpu_has_llsc" is false for 64-bit Cobalt kernels. LLD/SCD is broken on RM5230/5231. There is an experimental patch for 2.6.9 on
http://www.colonel-panic.org/cobalt-mips.

P.

Peter,
Got it to compile, but the tulip driver is giving me fits. I built
it as a module and into the kernel

0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed

Signature · 3 lines
----
Jim Gifford
[email protected]
#4 lore raw
Quoted 5 lines · Jim Gifford

Got it to compile, but the tulip driver is giving me fits. I built
it as a module and into the kernel

Quoted 7 lines

0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed

Tulip driver gave me problems also. I landed up inserting a printk() which made it work, see the patch. I didn't get round to debugging it any further, sorry.

P.

#5 lore raw
Quoted 17 lines · Peter Horton
Quoted 4 lines · Jim Gifford

Got it to compile, but the tulip driver is giving me fits. I built
it as a module and into the kernel

Quoted 9 lines

0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed
0000:00:07.0: tulip_stop_rxtx() failed

Quoted 8 lines

Tulip driver gave me problems also. I landed up inserting a printk() which made it work, see the patch. I didn't get round to debugging it any further, sorry.

P.

I posted a simliar question to the linux-net list, a lot of people don't understand why the driver doesn't work in 64 bit, but works perfectly in 32 bit.

Signature · 3 lines
----
Jim Gifford
[email protected]
#6 lore raw
In reply to Peter Horton
Quoted 9 lines · Peter Horton

Tulip driver gave me problems also. I landed up inserting a printk() which made it work, see the patch. I didn't get round to debugging it any further, sorry.

P.

Peter here is the fix for the linux-mips version of kernel and the tulip driver. Thanx again for you help

Signature · 3 lines
----
Jim Gifford
[email protected]
drivers/net/tulip/de2104x.c +1 -6
diff -Naurp linux.orig/drivers/net/tulip/de2104x.c linux/drivers/net/tulip/de2104x.c
@@ -1787,15 +1787,10 @@ static void __init de21041_get_srom_info
 	/* DEC now has a specification but early board makers
 	   just put the address in the first EEPROM locations. */
 	/* This does  memcmp(eedata, eedata+16, 8) */
-
-#ifndef CONFIG_MIPS_COBALT
-
 	for (i = 0; i < 8; i ++)
 		if (ee_data[i] != ee_data[16+i])
 			sa_offset = 20;
 
-#endif
-
 	/* store MAC address */
 	for (i = 0; i < 6; i ++)
 		de->dev->dev_addr[i] = ee_data[i + sa_offset];
@@ -1932,7 +1927,7 @@ bad_srom:
 	goto fill_defaults;
 }
 
-static int __init de_init_one (struct pci_dev *pdev,
+static int __devinit de_init_one (struct pci_dev *pdev,
 				  const struct pci_device_id *ent)
 {
 	struct net_device *dev;
drivers/net/tulip/de4x5.c +0 -1
diff -Naurp linux.orig/drivers/net/tulip/de4x5.c linux/drivers/net/tulip/de4x5.c
@@ -2124,7 +2124,6 @@ static struct eisa_driver de4x5_eisa_dri
                 .remove  = __devexit_p (de4x5_eisa_remove),
         }
 };
-MODULE_DEVICE_TABLE(eisa, de4x5_eisa_ids);
 #endif
 
 #ifdef CONFIG_PCI
drivers/net/tulip/media.c +20 -5
diff -Naurp linux.orig/drivers/net/tulip/media.c linux/drivers/net/tulip/media.c
@@ -44,8 +44,10 @@ static const unsigned char comet_miireg2
 
 /* MII transceiver control section.
    Read and write the MII registers using software-generated serial
-   MDIO protocol.  See the MII specifications or DP83840A data sheet
-   for details. */
+   MDIO protocol.
+   See IEEE 802.3-2002.pdf (Section 2, Chapter "22.2.4 Management functions")
+   or DP83840A data sheet for more details.
+   */
 
 int tulip_mdio_read(struct net_device *dev, int phy_id, int location)
 {
@@ -307,13 +309,29 @@ void tulip_select_media(struct net_devic
 				int reset_length = p[2 + init_length];
 				misc_info = (u16*)(reset_sequence + reset_length);
 				if (startup) {
+					int timeout = 10;	/* max 1 ms */
 					iowrite32(mtable->csr12dir | 0x100, ioaddr + CSR12);
 					for (i = 0; i < reset_length; i++)
 						iowrite32(reset_sequence[i], ioaddr + CSR12);
+
+					/* flush posted writes */
+					ioread32(ioaddr + CSR12);
+
+					/* Sect 3.10.3 in DP83840A.pdf (p39) */
+					udelay(500);
+
+					/* Section 4.2 in DP83840A.pdf (p43) */
+					/* and IEEE 802.3 "22.2.4.1.1 Reset" */
+					while (timeout-- &&
+						(tulip_mdio_read (dev, phy_num, MII_BMCR) & BMCR_RESET))
+						udelay(100);
 				}
 				for (i = 0; i < init_length; i++)
 					iowrite32(init_sequence[i], ioaddr + CSR12);
+
+				ioread32(ioaddr + CSR12);	/* flush posted writes */
 			}
+
 			tmp_info = get_u16(&misc_info[1]);
 			if (tmp_info)
 				tp->advertising[phy_num] = tmp_info | 1;
@@ -399,9 +417,6 @@ void tulip_select_media(struct net_devic
 	}
 
 	tp->csr6 = new_csr6 | (tp->csr6 & 0xfdff) | (tp->full_duplex ? 0x0200 : 0);
-
-	udelay(1000);
-
 	return;
 }
 
drivers/net/tulip/tulip_core.c +3 -3
diff -Naurp linux.orig/drivers/net/tulip/tulip_core.c linux/drivers/net/tulip/tulip_core.c
@@ -22,7 +22,7 @@
 #else
 #define DRV_VERSION	"1.1.13"
 #endif
-#define DRV_RELDATE	"May 11, 2002"
+#define DRV_RELDATE	"December 15, 2004"
 
 
 #include <linux/module.h>
@@ -1514,8 +1514,8 @@ static int __devinit tulip_init_one (str
                     (PCI_SLOT(pdev->devfn) == 12))) {
                        /* Cobalt MAC address in first EEPROM locations. */
                        sa_offset = 0;
-		       /* Ensure our media table fixup get's applied */
-		       memcpy(ee_data + 16, ee_data, 8);
+                       /* No media table either */
+                       tp->flags &= ~HAS_MEDIA_TABLE;
                }
 #endif
 #ifdef CONFIG_GSC
drivers/net/tulip/tulip.h +5 -2
diff -Naurp linux.orig/drivers/net/tulip/tulip.h linux/drivers/net/tulip/tulip.h
@@ -475,8 +475,11 @@ static inline void tulip_stop_rxtx(struc
 			udelay(10);
 
 		if (!i)
-			printk(KERN_DEBUG "%s: tulip_stop_rxtx() failed\n",
-					pci_name(tp->pdev));
+			printk(KERN_DEBUG "%s: tulip_stop_rxtx() failed"
+					" (CSR5 0x%x CSR6 0x%x)\n",
+					pci_name(tp->pdev),
+					ioread32(ioaddr + CSR5),
+					ioread32(ioaddr + CSR6));
 	}
 }