@max Yeah, that what i have when set debug output to serial with level 1:
Vendor Specific Data Block
it8212ide.device 53.20 (24.9.2014)
[it8212ide/dev_init] Nothing started, bailing out
lsi53c8xx.device 53.20 (24.9.2014)
[lsi53c8xx/dev_init] Nothing started, bailing out
sam460sata.device 53.20 (24.9.2014)
[sam460sata/dev_init] Running on PPC460ex
[sam460sata/dev_init] not in SATA mode
[sam460sata/dev_init] Nothing started, bailing out
sii0680ide.device 53.20 (24.9.2014)
[sii0680ide/dev_init] Nothing started, bailing out
sii3112ide.device 53.20 (24.9.2014)
[sii3112ide/dev_init] Nothing started, bailing out
sii3114ide.device 53.20 (24.9.2014)
[sii3114ide/dev_init] Found chip #0
[sii3114ide/init_port] ---> Port 0
[sii3114ide/init_port] IOBase 00001000, AltBase 0000100A
[sii3114ide/init_port] bmcr_base 00001020
[sii3114ide/init_port] MMIOBase 80000000
[sii3114ide/init_port] Config not forced, scanning port 0...
[sii3114ide/reg_config] assuming device 0 is TYPE_NONE
[sii3114ide/reg_config] assuming device 1 is TYPE_NONE
[sii3114ide/reg_config] there seems to be a device 0, assuming TYPE_UNKN
[sii3114ide/reg_config] there is no device 1 (sc 00 sn 00)
[sii3114ide/reg_config] device 0 is ATA
[sii3114ide/init_port] 1 device(s) on port
[sii3114ide/port_list] 0 Master : 'ATA'
[sii3114ide/port_list] 1 Slave : 'no device'
[sii3114ide/init_port] Starting 'sii3114ide.device - chip 0 port 0' task
[sii3114ide/int_install_handler] Installing handler for irq 48
[sii3114ide/init_unit] Trying to configure unit 0
[sii3114ide/irq_wait] timed out
[sii3114ide/exec_pio_data_in_cmd] <- here
[sii3114ide/ata_read_drive_properties] Unit 0 returned error 255, failbits 00000000h, timeout 0
[sii3114ide/init_port] Port 0 task started
[sii3114ide/init_port] ---> Port 1
[sii3114ide/init_port] IOBase 00001010, AltBase 0000101A
[sii3114ide/init_port] bmcr_base 00001028
[sii3114ide/init_port] MMIOBase 80000000
[sii3114ide/init_port] Config not forced, scanning port 1...
[sii3114ide/reg_config] assuming device 0 is TYPE_NONE
[sii3114ide/reg_config] assuming device 1 is TYPE_NONE
[sii3114ide/reg_config] there is no device 0 (sc 00 sn 00)
[sii3114ide/reg_config] there is no device 1 (sc 00 sn 00)
[sii3114ide/init_port] 0 device(s) on port
[sii3114ide/dev_init] All done
sii3512ide.device 53.20 (24.9.2014)
[sii3512ide/dev_init] Nothing started, bailing out
[sii3114ide/find_unit] unitnum 4 invalid (!chip)
[sii3114ide/find_unit] unitnum 5 invalid (!chip)
[sii3114ide/find_unit] unitnum 6 invalid (!chip)
[sii3114ide/find_unit] unitnum 7 invalid (!chip)
[sii3114ide/find_unit] unitnum 8 invalid (!chip)
[sii3114ide/find_unit] unitnum 9 invalid (!chip)
[sii3114ide/find_unit] unitnum 10 invalid (!chip)
[sii3114ide/find_unit] unitnum 11 invalid (!chip)
[sii3114ide/find_unit] unitnum 12 invalid (!chip)
[sii3114ide/find_unit] unitnum 13 invalid (!chip)
[sii3114ide/find_unit] unitnum 14 invalid (!chip)
[sii3114ide/find_unit] unitnum 15 invalid (!chip)
[OHCI] Enabled USB port on the 460ex processor.
[OHCI] Forcing powering down before a reset
Seems that sixxxxx.device just can't do things properly ? I see that in version 53.21 of sixxxxx.device there was this error fixed:
Quote:
sii3114ide.device 53.21 (27.6.2017) <cmincea>
- Replaced ReadConfigLong(PCI_BASE_ADDRESS_x) calls with their PCI interface variant GetResourceRange(x)->BaseAddress to get the correct base address instead of the *bus* base address.
[sii3512ide/dev_init] Config forced in $sii3512ide_conf : 10 [sii3512ide/dev_init] Found chip #0 [sii3512ide/init_port] ---> Port 0 [sii3512ide/init_port] IOBase 00001000, AltBase 0000100A [sii3512ide/init_port] bmcr_base 00001020 [sii3512ide/init_port] MMIOBase 80000000 [sii3512ide/init_port] Config forced, won't scan port 0 [sii3512ide/init_port] 1 device(s) on port [sii3512ide/port_list] 0 Master : 'ATA' [sii3512ide/init_port] Starting 'sii3512ide.device - chip 0 port 0' task [sii3512ide/int_install_handler] Installing handler for irq 48 [sii3512ide/init_unit] Trying to configure unit 0 [sii3512ide/ata_read_drive_properties] multi_count 1 [sii3512ide/ata_read_drive_properties] Adjusting unit 0 to xfer mode 70 [sii3512ide/init_unit] Unit configured, flags $0187 [sii3512ide/init_port] Port 0 task started [sii3512ide/init_port] ---> Port 1 [sii3512ide/init_port] IOBase 00001010, AltBase 0000101A [sii3512ide/init_port] bmcr_base 00001028 [sii3512ide/init_port] MMIOBase 80000000 [sii3512ide/init_port] Config forced, won't scan port 1 [sii3512ide/init_port] 0 device(s) on port [sii3512ide/dev_init] All done [sii3512ide/find_unit] unitnum 2 invalid (!chip) [sii3512ide/find_unit] unitnum 3 invalid (!chip) [sii3512ide/find_unit] unitnum 4 invalid (!chip) [sii3512ide/find_unit] unitnum 5 invalid (!chip) [sii3512ide/find_unit] unitnum 6 invalid (!chip) [sii3512ide/find_unit] unitnum 7 invalid (!chip)
Unit configured, flags $0187! Yeah! Now I can see it in the MediaToolBox!
Interesting through why 3114 didn't work like that? I for sake of tests tried to put 3114 back again, and set variable for + reboot, no, still no luck. But 3512 is fine always with this "sii3512ide_conf 10"
@All Ok, another issue. Installed SLB_V2, created 2 partitions 100 GB for system and 400 GB for work, start installing. Installation fast enough, but not that fast (maybe because something needs to be done about PIO/etc modes after?), and when I am about to be on 50% have python exceptions with words:
Quote:
Exception exceptions.OSError: 'Cannot show page' in 'garbage collection' ignored Fatal Python error: unexpected exception during garbage collection ***Command 'Installation Utility' returned with unfreed signals 10000000!
At first, I think that this may be because I didn't have a fan over the CPU attached for the first tests, so I attach it, but it changes nothing - always the same issue, on exactly 50% of the installation
(click for full image)
Maybe issue with Installation Utility?
And issue always and 100% reproducible. So it's not random.
Edited by kas1e on 2021/4/29 8:37:22 Edited by kas1e on 2021/4/29 8:42:53 Edited by kas1e on 2021/4/29 8:43:51
That what the issue was: By default, SAM460 (and not only, but all other installation CDs) have an older version of usbhid driver, so new mouses may work wrong and cause issues. So I manually update it to the latest coming with fe-update2, and all the time Python crashes on 50% of installation because files were read-protected. But Python and InstallationUtility didn't handle it well with those versions coming with SAM460 FE install CD.
By default we have in sam460 install cd:
installation utility: 53.12 and python 53.35
Once I upgraded to Installation utility 53.18 and python 53.37, instead of an exception in InstallationUtility I can see the nice error window saying that my "hid.fdcalss" is read-protected.
So I make it be readable, and the installation goes fine.
What it all means in end? That to install it all with Radeon RX and all modern mouses and stuff, is not for the beginner, that for sure!
What there need to make a thing be easier:
From Uboot side:
1. UBoot need to be shown on Polaris cards 2. By default one of the boot options in the Uboot should be USB of course too. 3. UBoot should have that "sii3512ide_conf 10" set by default
And while it not possible probably anymore to do: Installation CD should contain all latest stuff (Polaris driver, latest USB drivers, etc). Or user forced to create his own installation CD with all stuff included: can be done, but 100% not for casual users.
But hey hey, it can be assembled and installed :)
PS. And I also noticed an annoying issue: the mouse often didn't work after reboot, and I need to unplug and plug it again into the same slot to make it works (at least that happens for those slots place on the back of the motherboard). Are there any settings maybe like timeouts or something I can change in Uboot so it will work always?
To add it's the same USB mouse and USB keyboard I use on x5000 and on pegasos2, and never had such an issue.
Interesting through why 3114 didn't work like that? I for sake of tests tried to put 3114 back again, and set variable for + reboot, no, still no luck. But 3512 is fine always with this "sii3512ide_conf 10"
But SiI3114 has four ports, maybe you need to include all four in the flags, e.g. "setenv sii3114ide_dev 1000" or something?
Don't know if that's the problem, it just looked odd to me.
Yeah! That did the trick! With sii3114 it works now once I set 1000 instead of 10. Cool :)
And seems once I install both update1 and update2, issues with the mouse not being recognized after boot is gone too.
@All Now after it works, how I can get out all the possible speed from? If I remember right, the drives b SATA devices by default work in PIO mode and should be changed to DMA, or that is not true anymore? Or in the case with SSD, it's all fine already?
And what kind of disk speed tool we still can use on the latest os4 to measure speed? The same "diskspeed" from os4depot?
EDIT: just running "idetool -l sii3114ide.device" bring me that:
Quote:
Flags: $00000187 - present, LBA48 used, supports DMA, ATA, supports cache flush, interrupts used. Xfer mode: best pio 12 (PIO 4, 16 MB/s) / best dma 70 (UDMA 6, 133 MB/s) / current 70 (UDMA 6, 133 MB/s) ...
So seems DMA enabled by default already and information in the sam460 guide from 2014 are a bit out of date?
Edited by kas1e on 2021/4/30 4:29:37 Edited by kas1e on 2021/4/30 4:37:43 Edited by kas1e on 2021/4/30 4:50:44 Edited by kas1e on 2021/4/30 4:51:18
@Max Is Sam460CR have on the board any kind of fan connector? I mean it can be not a Fan connector, just 4 pins which give +12? I see there on the plate J42 (while it didn't mention in the user's guide), maybe this is that one?
EDIT: just running "idetool -l sii3114ide.device" bring me that:
Quote:
Flags: $00000187 - present, LBA48 used, supports DMA, ATA, supports cache flush, interrupts used. Xfer mode: best pio 12 (PIO 4, 16 MB/s) / best dma 70 (UDMA 6, 133 MB/s) / current 70 (UDMA 6, 133 MB/s) ...
So seems DMA enabled by default already and information in the sam460 guide from 2014 are a bit out of date?
Yeah, if I remember correctly, the defaulting to PIO was done for the internal a1ide.device on the old AmigaOnes, where DMA mode was problematic and needed hardware fixes with other problems etc.
I think the whole series of "IDE" (SATA etc.) drivers have long been defaulting to DMA modes on anything else.
So unless it looks like it is too optimistic and defaults to a DMA mode too high, so errors occur, there is normally no need to use the *_xfer variables.
From SYS:Documentation/IDE/sii3114ide_dev.doc:
Quote:
Default : if you don't set this envvar, sii3114ide will use the best UDMA mode supported by the drive. This is only true for hardware platforms with stable UDMA support otherwise the default is best PIO mode.
@All Anyone who own sam460 on 1.1ghz , did you have any kind of fan+heatsynk on the top of the cpu ? Why i ask, because i remember that sam's cpu were positioned as ones which can works without active cooling, but what i found now, is that without cooling+heatsync, it just start to be too hot, and whole system just freezes. I.e. without cooling it is unpossible to use.
The next question i have : if you have active cooling, then how you install your heatsynk on cpu ? The motherboard do not have any holes for fasteners of any kind, so it seems that thermo-glue is the only way ?
Just my motherboard when i got it , seems have some kind of thermo-glue, but when you start use os, cpu start to be hot enough, and then all this glue becomes fluid and the heatsync together with fan just slides out from cpu. Maybe the thermo-glue which were used in my board wasn't good enough, or my CPU is strangely hot for no reassons..
Anyway, will be interesting to know how owners of sam boards care about. Thanks!
@TiredOfLife And without active cooling also freeze-too-hot-cpu-etc ? I just thinking maybe making some kind of addon which will hold the fan but use other motherboard's holes to mount to.
Because as i aware, if cpu is hot, then any thermo-glue may be fluid after a while and the fan slides.
@AmigaOldskooler Well.. their fan it just radiator + not very good fan on top. And the connection is the same : glue only, without any fixation to motherboard. Even not just glue, but some stiker connected by one side to the radiator, and another one is when you remove a paper and glut it on cpu. As i understand those glue-stickers offten happen to be fluid and radiator/fan just slides : at least that what i have with mine board when it arrive.
I think more about steel rods with which you can press the cooler to the processor, while screwing the other ends of the steel rods into the free holes of the motherboar. Because all this "glue" stuff is maybe ok, but not the strong enough imho.
@Spectre I apologize for not being clear enough in my first question about it. I'm referring to a heatsink rather than a fan. i.e., how Sam460 users attach heatsinks on top of the CPU. Of course, when the heatsink is on top of the CPU and attached to it like a stone, you can add anything you need on top of it, no problem.
The issue is that the fan and heatsync that came with my SAM460 were not attached to the motherboard but were instead glued to the top of the CPU with some glue or whatever that stick of paper is called. When the CPU reaches full capacity in first time, the heatsink, along with the fan, slides off the CPU and onto the motherboard. And whole glue was fluid.
My question is about how to add a heatsink on top of the CPU so that it does not slide off when the CPU is hot.
I'm not sure what kind of glue or other materials Acube uses for my board, but once I start using it, the heatsink and fan simply slide off the CPU and onto the motherboard. Now, what I want is to somehow attach it, like a stone. To put it another way, how attach a heatsink over the CPU, with no slides off when the CPU is hot. Then, of course, I can add any kind of fan, that is no problem. The issue is how to attach a heatsink to the CPU so this will not overheat glue and heatsink will not slide off.
Edited by kas1e on 2022/11/14 15:46:01 Edited by kas1e on 2022/11/14 15:47:07
Heat sink is well glued on on both my fired board and my working replacement (2014 batch). So they may have used a different glue on your batch or it was not properly glued on with your board . So probably you should Email A-Cube for directions .
I experienced the exact same thing with the fan coming off. I overcame it the lowest tech way possible ... I keep the motherboard facing up so my mini-tower must lie on its side.
News feed
OS4 Feedback forum
OS4Depot Feedback forum
Amigabounty forum
OpenAmiga forum
AmiCygnix forum
ABC shell forum
AmiKit forum
Cinnamon Writer forum
CodeBench forum
Digital Universe forum
Dopus 5 forum
E-UAE forum
Gnash forum
Ibrowse forum
JAmiga forum
Odyssey forum
OWB forum
Qt forum
SmartFileSystem forum
Timberwolf forum
TouchDevice forum
TuneNet forum
Unsatisfactory Software forum
Donate
