| View previous topic :: View next topic |
| Author |
Message |
pappy_mcfae Watchman


Joined: 27 Dec 2007 Posts: 5806 Location: Pomona, California.
|
|
| Back to top |
|
 |
NeddySeagoon Administrator


Joined: 05 Jul 2003 Posts: 29965 Location: 56N 3W
|
Posted: Wed Apr 11, 2012 9:46 pm Post subject: |
|
|
pappy_mcfae,
The firmware loading sequende goes like this.
The driver is loaded and does looking for its firmware. If the driver is in the kernel (=y) the firmware must be too as root is not yet mounted, so /lib/firmware cannot be read.
If the driver is modular (=m) then its loaded from /lib/modules/uname -r/, so /lib/firmware is available too. Any kernel space that was holding the firmware has been freed, so a modular driver and in kernel firmware fail too.
This applies to all drivers that need firmware, not just video cards. This is why is a really good idea to make wireless drivers as modules. Many of them need firmware. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
 |
pappy_mcfae Watchman


Joined: 27 Dec 2007 Posts: 5806 Location: Pomona, California.
|
|
| Back to top |
|
 |
Tariella Guru

Joined: 10 Dec 2005 Posts: 377 Location: Europe/Austria
|
Posted: Thu Apr 12, 2012 10:43 am Post subject: |
|
|
I entered the following in make menuconfig
| Code: | (RS780_pfp.bin) External firmware blobs to build into the kernel binary
(/lib/firmware/) Firmware blobs root directory
|
make gives me
| Code: | MK_FW firmware/RS780_pfp.bin.gen.S
make[1]: ***no rule to make target "/lib/firmware/RS780_pfp.bin", necessary to create "firmware/RS780_pfp.bin.gen.o" |
What did go wrong? |
|
| Back to top |
|
 |
NeddySeagoon Administrator


Joined: 05 Jul 2003 Posts: 29965 Location: 56N 3W
|
Posted: Thu Apr 12, 2012 6:01 pm Post subject: |
|
|
Tariella,
Sooo close!
| Code: | | (RS780_pfp.bin) External firmware blobs to build into the kernel binary | is not quite right.
You said your file was called /lib/fimware/radeon/RS780_pfp.bin.
You have /lib/fimware/ correct for the root of your firmware tree.
RS780_pfp.bin needs to be | Code: | | radeon/RS780_pfp.bin | as your file in in the radeon/ subdirectory.
My example I pasted earlier is from my media player. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
 |
Tariella Guru

Joined: 10 Dec 2005 Posts: 377 Location: Europe/Austria
|
Posted: Thu Apr 12, 2012 9:19 pm Post subject: |
|
|
I should have realized that myself. Of course it would need the full path.
(For the file R600_rlc.bin I had to emerge the package radeon-ucode. Info from this thread)
A few tries later I found all the firmware I need and everything works.
Now there is only one kernel message
| Code: | | SP5100 TCO timer: mmio address 0xfec000f0 already in use |
The rest seems to work just fine. |
|
| Back to top |
|
 |
pappy_mcfae Watchman


Joined: 27 Dec 2007 Posts: 5806 Location: Pomona, California.
|
|
| Back to top |
|
 |
pappy_mcfae Watchman


Joined: 27 Dec 2007 Posts: 5806 Location: Pomona, California.
|
|
| Back to top |
|
 |
Moriah Veteran


Joined: 27 Mar 2004 Posts: 1815 Location: Warsaw KY US
|
Posted: Sat Apr 14, 2012 8:03 pm Post subject: |
|
|
Pappy, HELP! I am messing around with that old AMD quad core box again -- the one that seems to stall out until I tap a key or move the mouse to create an interrupt, or whatever.
I just completely rebuilt the software for it using your 3.2.1-gentoo-r2-x86_64-08.config and the appropriate kernel release. This box has a 4-core AMD cpu and 8 GB of RAM, along with 3 1.5 TB SATA drives in a hot-swap SATA backplane running Linux software RAID-1 as a 3-way mirror on /dev/md0. /dev/md0 is then encrypted with LUKS and then is made a physical volume under LVM2. The PV has a single volume group on it, and that VG has a single logical volume on it, which is the root filesystem under XFS. The whole mess is booted with an initramfs from USB flash stick as the boot device.
I know this sounds messy, but it works great on several other systems I am running around here, including my trusty laptop, which has been running this way for over 2 years.
My problem is that this particular motherboard seems to be my nemesis!
I had hoped that by starting completely over, and using the series 3.x.x kernel, that all my woes would just dissapear into thin air. No such luck!
The thing runs fine under the live-CD, so it is possible to make it work; I just can't figure out how to do it.
It boots from the stick, but never gets to the point where it starts running the init script in the initramfs. Something probably has the wrong driver configured would be my guess.
The boot screen looks like this when it stalls: http://www.elilabs.com/~rj/foo.gif
The lspci -n output is:
| Code: |
00:00.0 0600: 1022:9600
00:01.0 0604: 1022:9602
00:0a.0 0604: 1022:9609
00:11.0 0106: 1002:4391
00:12.0 0c03: 1002:4397
00:12.1 0c03: 1002:4398
00:12.2 0c03: 1002:4396
00:13.0 0c03: 1002:4397
00:13.1 0c03: 1002:4398
00:13.2 0c03: 1002:4396
00:14.0 0c05: 1002:4385 (rev 3a)
00:14.1 0101: 1002:439c
00:14.3 0601: 1002:439d
00:14.4 0604: 1002:4384
00:14.5 0c03: 1002:4399
00:18.0 0600: 1022:1200
00:18.1 0600: 1022:1201
00:18.2 0600: 1022:1202
00:18.3 0600: 1022:1203
00:18.4 0600: 1022:1204
01:05.0 0300: 1002:9610
01:05.1 0403: 1002:960f
02:00.0 0200: 10ec:8168 (rev 02)
03:06.0 0200: 10ec:8139 (rev 10)
03:07.0 0200: 10ec:8139 (rev 10)
03:0e.0 0c00: 104c:8024
|
The .config is at: http://www.elilabs.com/~rj/foo.config
And the /proc/cpuinfo is:
| Code: |
processor : 0
vendor_id : AuthenticAMD
cpu family : 16
model : 2
model name : AMD Phenom(tm) 9650 Quad-Core Processor
stepping : 3
cpu MHz : 2304.857
cache size : 512 KB
physical id : 0
siblings : 4
core id : 0
cpu cores : 4
apicid : 0
initial apicid : 0
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs
bogomips : 4609.70
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate
processor : 1
vendor_id : AuthenticAMD
cpu family : 16
model : 2
model name : AMD Phenom(tm) 9650 Quad-Core Processor
stepping : 3
cpu MHz : 2304.857
cache size : 512 KB
physical id : 0
siblings : 4
core id : 1
cpu cores : 4
apicid : 1
initial apicid : 1
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs
bogomips : 4609.98
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate
processor : 2
vendor_id : AuthenticAMD
cpu family : 16
model : 2
model name : AMD Phenom(tm) 9650 Quad-Core Processor
stepping : 3
cpu MHz : 2304.857
cache size : 512 KB
physical id : 0
siblings : 4
core id : 2
cpu cores : 4
apicid : 2
initial apicid : 2
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs
bogomips : 4609.85
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate
processor : 3
vendor_id : AuthenticAMD
cpu family : 16
model : 2
model name : AMD Phenom(tm) 9650 Quad-Core Processor
stepping : 3
cpu MHz : 2304.857
cache size : 512 KB
physical id : 0
siblings : 4
core id : 3
cpu cores : 4
apicid : 3
initial apicid : 3
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs
bogomips : 4609.86
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate
|
Let me know if you need anything else. Thanks for always being so helpful!  _________________ The MyWord KJV Bible tool is at http://www.elilabs.com/~myword
Foghorn Leghorn is a Warner Bros. cartoon character. |
|
| Back to top |
|
 |
NeddySeagoon Administrator


Joined: 05 Jul 2003 Posts: 29965 Location: 56N 3W
|
Posted: Sat Apr 14, 2012 9:27 pm Post subject: |
|
|
Moriah,
If you want a WAG from me, turn off Message Signalled Interrupts, if you can. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
 |
Moriah Veteran


Joined: 27 Mar 2004 Posts: 1815 Location: Warsaw KY US
|
Posted: Sat Apr 14, 2012 9:35 pm Post subject: |
|
|
Where do I find that in the "make menuconfig" tool?
Or is that a BIOS configuration item? _________________ The MyWord KJV Bible tool is at http://www.elilabs.com/~myword
Foghorn Leghorn is a Warner Bros. cartoon character. |
|
| Back to top |
|
 |
NeddySeagoon Administrator


Joined: 05 Jul 2003 Posts: 29965 Location: 56N 3W
|
Posted: Sat Apr 14, 2012 9:41 pm Post subject: |
|
|
Moriah,
Its a kernel option, under | Code: | | Bus options (PCI etc.) | on my system | Code: | | -*- Message Signaled Interrupts (MSI and MSI-X) | its forced on by something else.
The help says | Code: | | Selected by: AMD_IOMMU [=y] && IOMMU_SUPPORT [=y] && X86_64 [=y] && PCI [=y] && ACPI [=y] |
_________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
 |
Moriah Veteran


Joined: 27 Mar 2004 Posts: 1815 Location: Warsaw KY US
|
Posted: Sat Apr 14, 2012 9:58 pm Post subject: |
|
|
That's ugly! I need AMD_IOMMU, IOMMU_SUPPORT, X86_64, and PCI because its an AMD 4-core cpu with 8 GB ram. The question is, "What's ACPI?" I don't know if I need that or not. _________________ The MyWord KJV Bible tool is at http://www.elilabs.com/~myword
Foghorn Leghorn is a Warner Bros. cartoon character. |
|
| Back to top |
|
 |
NeddySeagoon Administrator


Joined: 05 Jul 2003 Posts: 29965 Location: 56N 3W
|
Posted: Sat Apr 14, 2012 10:43 pm Post subject: |
|
|
Moriah,
ACPI is power management. You can do without it for a test. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
 |
Moriah Veteran


Joined: 27 Mar 2004 Posts: 1815 Location: Warsaw KY US
|
Posted: Sat Apr 14, 2012 10:58 pm Post subject: |
|
|
Duh! I got confused because of AHCI looks a lot like ACPI, and I am doing several things at once right now, so my concentration is divided. Yes, I can turn off ACPI for a test. Thanks for the suggestion. I will let you know how it works. _________________ The MyWord KJV Bible tool is at http://www.elilabs.com/~myword
Foghorn Leghorn is a Warner Bros. cartoon character. |
|
| Back to top |
|
 |
Moriah Veteran


Joined: 27 Mar 2004 Posts: 1815 Location: Warsaw KY US
|
Posted: Sun Apr 15, 2012 4:14 am Post subject: |
|
|
OK, that worked when I got rid of ACPI.
I also had to correct an unrelated typo, but that was not related.
So what do I do now? Just run without power management? Does that affect cpu throttling or fan speed control? I don't really care about sleep and wake-up, etc. as this is a server that will be running continuously, but I do care about heat. High temperatures kill electronics components, and run up my air conditioning bill too.
But at least it boots, so now I can get on with the rest of the configuration -- tomorrow.  _________________ The MyWord KJV Bible tool is at http://www.elilabs.com/~myword
Foghorn Leghorn is a Warner Bros. cartoon character. |
|
| Back to top |
|
 |
pappy_mcfae Watchman


Joined: 27 Dec 2007 Posts: 5806 Location: Pomona, California.
|
Posted: Sun Apr 15, 2012 6:43 am Post subject: |
|
|
Moriah,
I think it's very likely that your MSI issue was caused by the selection of an incorrect driver, or two. I went through the .config, and did find some incorrect drivers. I replaced them with the proper ones. I also temporarily turned off MSI. If this .config works, and it should, turn MSI back on, and see if the issue is now gone. Don't worry about ACPI in this kernel, it is properly set, and unaffected by MSI.
I also set you up for ATI KMS (Kernel Mode Setting), which is basically a different kind of frame buffer. It will build firmware, and that firmware should be properly installed by the make modules_install command. If that doesn't work, try make firmware_install. This command will create all in-kernel firmware blobs. That's actually a bit of overkill, but if the firmware isn't generated with the former command, the latter will definitely make it happen.
Assuming the coming severe storms don't shred the place where I'm living, I'll stop by to check out how this went later on today.
Click here for the new .config. Use the standard rules for setting things up, and post /var/log/dmesg, assuming you get a proper boot. If you don't, I'm fairly sure it won't be because of an MSI quirk.
Cheers,
Pappy _________________ SITE LIST:
Main: http://www.kernel-seeds.org
Mirror: http://kernel-seeds.grytpype-thynne.org/
Mirror 2: http://kernel-seeds.gentoostudio.org/
Mirror 3: http://www.elilabs.com/~pappy/
Mirror 4: http://62.3.120.142/~seeds/ |
|
| Back to top |
|
 |
Moriah Veteran


Joined: 27 Mar 2004 Posts: 1815 Location: Warsaw KY US
|
Posted: Sun Apr 15, 2012 12:11 pm Post subject: |
|
|
Pappy: I would like to know which drivers were incorrect, if you remember. Even with the Debian web page, it is not always completely obvious which driver one needs all the time.
Hopefully the storms will be a blessing only, in that they will provide rain to relieve the drought down there, without the damaging winds or flooding.
I am busy this morning with church, but I will try to play with your new config this afternoon. Thanks!  _________________ The MyWord KJV Bible tool is at http://www.elilabs.com/~myword
Foghorn Leghorn is a Warner Bros. cartoon character. |
|
| Back to top |
|
 |
pappy_mcfae Watchman


Joined: 27 Dec 2007 Posts: 5806 Location: Pomona, California.
|
Posted: Sun Apr 15, 2012 8:43 pm Post subject: |
|
|
The I2C was definitely incorrect, That would render the SMBUS and all devices under it useless. The CPU was set for generic x86_64. While this is technically correct, it's not as correct as setting for the AMD option. The PCMCIA options were also enabled. While this shouldn't cause an issue, it's better to not have options enabled that don't exist. The PATA interface wasn't set at all. While you may not be using it, it's probably better to have it using its resources so other devices don't try to step on them. Most importantly, I changed from the UHCI USB host to the OHCI.
The latter may well be the cause of the worst of your issues. In general, if the mobo has an Intel or Via chipset, UHCI is the order of the day, along with EHCI for machines with USB2, which is pretty much everything. If the mobo isn't, and has any other chipset (AMD/ATI, etc), then OHCI is a must, and EHCI should be there as well for UBS2. Because you're using a USB key to boot the machine, the fact that the machine can't switch to OHCI to continue to boot could easily cause the hang you experienced.
Fortunately, while those storms did pass here, they were very weak. This is probably because they were early in the day, and didn't have solar heating to give them the power they were showing in Oklahoma yesterday. The condition of the web server tells me we didn't even lose power.
Anyway, let me know how things are booting at this point.
Cheers,
Pappy _________________ SITE LIST:
Main: http://www.kernel-seeds.org
Mirror: http://kernel-seeds.grytpype-thynne.org/
Mirror 2: http://kernel-seeds.gentoostudio.org/
Mirror 3: http://www.elilabs.com/~pappy/
Mirror 4: http://62.3.120.142/~seeds/ |
|
| Back to top |
|
 |
NeddySeagoon Administrator


Joined: 05 Jul 2003 Posts: 29965 Location: 56N 3W
|
Posted: Sun Apr 15, 2012 9:12 pm Post subject: |
|
|
Moriah,
What pappy says is correct but incomplete. Some chipsets have USB Transaction Translation hardware, this means that you can turn off both OHCI and UHCI, turn on Transaction Translation Support and USB1 devices still work. This gives you wierd dmesg like | Code: | [ 2.048109] usb 2-1: new high-speed USB device number 2 using ehci_hcd
[ 2.296126] usb 2-6: new high-speed USB device number 3 using ehci_hcd
[ 2.528915] usb 1-6.1: new low-speed USB device number 3 using ehci_hcd
[ 2.720522] usb 1-6.2: new full-speed USB device number 4 using ehci_hcd
[ 2.968998] usb 1-6.4: new low-speed USB device number 5 using ehci_hcd | notice how low-speed (1.5Mbit) full-speed (12Mbit) and high-speed (480Mbit) devices are all using the ehci_hcd (USB2) driver.
The only way to tell if you have Transaction Translation hardware is to try it. The downside is that if you don't, no USB1 device will work, so its not something you do if you only have a usb keyboard. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
 |
pappy_mcfae Watchman


Joined: 27 Dec 2007 Posts: 5806 Location: Pomona, California.
|
Posted: Mon Apr 16, 2012 5:26 am Post subject: |
|
|
I never knew that. I love when I learn new things.
I have also noted that at times, on certain motherboards, turning UHCI in lieu of OHCI, or vice versa, can cause hangs. Sometimes, mixing UHCI and OHCI can do the same thing, especially when only one is represented in the hardware setup. I have also seen mobos that have not only OHCI, UHCI, and EHCI, but also add in XHCI, the driver for USB-3.
In the final analysis, it is definitely better to set for what you have hardware-wise. That allows you to at least know that any issues you have that are driver related are caused by bugs, not interaction between needed and unneeded drivers.
I am definitely curious as to whether the kernel will now fully boot. Please keep us informed.
Cheers,
Pappy _________________ SITE LIST:
Main: http://www.kernel-seeds.org
Mirror: http://kernel-seeds.grytpype-thynne.org/
Mirror 2: http://kernel-seeds.gentoostudio.org/
Mirror 3: http://www.elilabs.com/~pappy/
Mirror 4: http://62.3.120.142/~seeds/ |
|
| Back to top |
|
 |
Moriah Veteran


Joined: 27 Mar 2004 Posts: 1815 Location: Warsaw KY US
|
Posted: Mon Apr 16, 2012 2:57 pm Post subject: |
|
|
I will keep you informed. I am learning a bunch of new things here too.
I had a drive die in a web/email server yesterday afternoon, so I need to replace that and get that server back up and running, then I will be able to get back on this "nemesis" machine we have been talking about. I first need to locate a PATA drive to put in the dead server that is 160 GB or bigger, as these beasts are getting scarce. I will check a couple of local sources. Hopefully this will not be a mail order. _________________ The MyWord KJV Bible tool is at http://www.elilabs.com/~myword
Foghorn Leghorn is a Warner Bros. cartoon character. |
|
| Back to top |
|
 |
Moriah Veteran


Joined: 27 Mar 2004 Posts: 1815 Location: Warsaw KY US
|
Posted: Mon Apr 16, 2012 11:23 pm Post subject: |
|
|
I found a 250 GB PATA drive, so I swapped out the dead drive and put the 250 GB one in. It is running badblocks -wv right now. After that finished -- assuming it is good -- I will partition and set up LVM and restore files from the backup server to reincarnate that web/email server. Meanwhile, I am starting to look over the config you sent me to the "nemesis" machine.  _________________ The MyWord KJV Bible tool is at http://www.elilabs.com/~myword
Foghorn Leghorn is a Warner Bros. cartoon character. |
|
| Back to top |
|
 |
pappy_mcfae Watchman


Joined: 27 Dec 2007 Posts: 5806 Location: Pomona, California.
|
|
| Back to top |
|
 |
Moriah Veteran


Joined: 27 Mar 2004 Posts: 1815 Location: Warsaw KY US
|
Posted: Wed Apr 18, 2012 4:05 am Post subject: |
|
|
Pappy: I built a new kernel with the config you gave me, and it barfed; it wouldn't boot all the way. In fact, it died at the same place as the snapshot of the screen I sent with my original request a few days ago. I haven't looked yet, but I suspect you turned ACPI back on, and it doesn't like that. It's midnite here, and I am going to bed. I will reboot from the liveCD and take a look at the ACPI setting with make menuconfig tomorrow.  _________________ The MyWord KJV Bible tool is at http://www.elilabs.com/~myword
Foghorn Leghorn is a Warner Bros. cartoon character. |
|
| Back to top |
|
 |
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|