Re: 2.6.28-git8: tg3 doesn't work due to firmware not loading (-git7 is ok)

On Wed, Jan 7, 2009 at 3:24 PM, Alessandro Suardi
[off-list ref] wrote:

From dmesg:

[asuardi@sandman src]$ dmesg | grep tg3
tg3.c:v3.97 (December 10, 2008)
tg3 0000:02:00.0: PCI INT A -> Link[LNKA] -> GSI 11 (level, low) -> IRQ 11
tg3 0000:02:00.0: setting latency timer to 64
tg3 0000:02:00.0: wake-up capability disabled by ACPI
tg3 0000:02:00.0: PME# disabled
tg3 0000:02:00.0: firmware: requesting tigon/tg3_tso.bin
tg3: Failed to load firmware "tigon/tg3_tso.bin"
tg3 0000:02:00.0: PCI INT A disabled
tg3: probe of 0000:02:00.0 failed with error -2


But /lib/firmware/tigon/tg3_tso.bin is actually present:

[asuardi@sandman src]$ ls -l /lib/firmware/tigon/tg3_tso.bin
-rw-r--r-- 1 root root 7004 2009-01-06 22:58 /lib/firmware/tigon/tg3_tso.bin



Diffing the .configs one can see CONFIG_KMOD disappeared; and
 the output of the kernel build in -git8 also has the IHEX parts about
 tg3 firmware which are new (I don't recall seeing those in the last
 1000 builds or so ;)

[asuardi@sandman src]$ diff .config-2.6.28-git[78]
3,4c3,4
< # Linux kernel version: 2.6.28-git7
< # Mon Jan  5 17:34:11 2009
---

quoted

# Linux kernel version: 2.6.28-git8
# Tue Jan  6 22:06:19 2009

133d132
< CONFIG_KMOD=y
1324a1324

quoted

# CONFIG_TWL4030_CORE is not set

1591d1590
< CONFIG_HID_BRIGHT=y
1595d1593
< CONFIG_HID_DELL=y
1602a1601

quoted

CONFIG_HID_NTRIG=y

1608a1608,1609

quoted

# CONFIG_GREENASIA_FF is not set
CONFIG_HID_TOPSEED=y

1821c1822
< # CONFIG_OCFS2_COMPAT_JBD is not set
---

quoted

CONFIG_OCFS2_FS_POSIX_ACL=y

1825c1826,1832
< # CONFIG_QUOTA is not set
---

quoted

CONFIG_QUOTA=y
CONFIG_QUOTA_NETLINK_INTERFACE=y
# CONFIG_PRINT_QUOTA_WARNING is not set
CONFIG_QUOTA_TREE=y
# CONFIG_QFMT_V1 is not set
CONFIG_QFMT_V2=y
CONFIG_QUOTACTL=y

[asuardi@sandman src]$ diff /tmp/make-kernel-2.6.28-git[78].out
669a670

quoted

  CC [M]  fs/ocfs2/blockcheck.o

692a694,695

quoted

  CC [M]  fs/ocfs2/quota_local.o
  CC [M]  fs/ocfs2/quota_global.o

693a697

quoted

  CC [M]  fs/ocfs2/acl.o

810a815,818

quoted

  CC      fs/dquot.o
  CC      fs/quota_v2.o
  CC      fs/quota_tree.o
  CC      fs/quota.o

1045c1053
< /share/src/linux-2.6.28-git7/arch/x86/include/asm/string_32.h:75:
warning: array subscript is above array bounds
---

quoted

/share/src/linux-2.6.28-git8/arch/x86/include/asm/string_32.h:75: warning: array subscript is above array bounds

1286d1293
<   CC      drivers/hid/hid-bright.o
1290d1296
<   CC      drivers/hid/hid-dell.o
1295a1302

quoted

  CC      drivers/hid/hid-ntrig.o

1300a1308

quoted

  CC      drivers/hid/hid-topseed.o

1521c1529
< /share/src/linux-2.6.28-git7/arch/x86/include/asm/string_32.h:75:
warning: array subscript is above array bounds
---

quoted

/share/src/linux-2.6.28-git8/arch/x86/include/asm/string_32.h:75: warning: array subscript is above array bounds

1983d1990
< net/rfkill/rfkill.c:59: warning: 'rfkill_led_trigger' defined but not used
2234,2235c2241,2242
< System is 2000 kB
< CRC 8dd05de2
---

quoted

System is 2013 kB
CRC 37411210

2646a2654,2656

quoted

  IHEX    firmware/tigon/tg3.bin
  IHEX    firmware/tigon/tg3_tso.bin
  IHEX    firmware/tigon/tg3_tso5.bin

So which is it, whatever the IHEX part did or the disappearance of
 CONFIG_KMOD (which is in the -git8 patch, indeed) to cause
 this issue ?

Note that I don't have CONFIG_FIRMWARE_IN_KERNEL, but on
 the other hand I didn't earlier, either... and yes, CONFIG_MODULES=y.


thanks in advance, ciao,

Still broken in -git10, cc'ing netdev... h/w is the onboard
 Gigabit ethernet on my Dell Latitude D610, that is

02:00.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5751
Gigabit Ethernet PCI Express (rev 01)

thanks,

--alessandro

 "Sun keeps rising in the west / I keep on waking fully confused"

   (The Replacements, "Within Your Reach")