Amiga OS4.1 update 3 issues

This forum is for all AmigaOne 500, Sam440 and Sam460 specific issues.

Amiga OS4.1 update 3 issues

Postby jostein_aarbakk » Sun Sep 04, 2011 4:59 pm

First, I have to tell you who have been working on OS4 that there are a lot to like about the OS4 update 3:
- My Razer Copperhead laser mouse now works (it didn't on earlier OS-versions)
- the USB is faster
- the USB seems stable (I wrote/read a full DVD record via external DVD ROM without errors)
- It is possible to write a DVD via an external USB DVD writer with no negative sideeffects (multitasking works ok). Actually, I was amazed by how easy the OS recognized my external USB writer. It was just plug&play. Impressive!
- HD reads/writes are much faster than on earlier OS versions (10-15% speed improvement (!!)).
- SATA DVD drive consumes less CPU
- some 2D operations are much faster than update 2 according to GFXBench2D

All good improvements.
I really appreciate the continous development done with Amiga OS.

However, there are a few issues that needs to be sorted out.
Issue 1-5 below didn't exist in earlier OS4.1 versions, issue 6 did, and for what issues 7+ are concerned, I actually don't know.
Here it goes:

1) My "AEon Amiga mouse" gets deactivated at startup, and is hard to get working at all.
At 1st power on, it's "dead".
After a soft reboot, it might work (but not always).
After subsequent soft reboots, it doesn't work at all.

2) My "Razer Copperhead laser gaming mouse" is more stable, but gets deactivated now and then, sometimes at startup (happened when an USB card reader was connected before startup, but this might just be a coincidence).
Deactivation happens often when using Directory Opus, regardless of whether an USB memory card reader is connected or not (especially when using the scroll wheel on the mouse).

3) DVPlayer is less efficient with update 3 (consumes more CPU) compared to update 2, even though the DVPlayer binary is the same as I run on update 2. Version of DVPlayer used is 0.76 (the old one is 0.65).
This means the performance of v0.76 on OS4.1 update 3 is about similar as v0.65 on OS4.1 update 2, even though DVPlayer v0.76 itself is much, much faster than version 0.65.

4) RageMem shows me that the VIDEO BUS transfer speed is reduced (READ: 41->39 MB/s, WRITE: 63->53 MB/sec). Maybe this is the reason of DVPlayer's lowered performance? I don't know.

5) USB2 read speeds seems to be limited to around 6 MB/sec, even with devices that supports transfer speeds as high as 20 MB/sec. Write speeds are lower. I have tested reading from an external USB DVD-writer and USB memory cards; The max speed are the same.


6) I am experiencing the same issues with my "internal SATA DVD-writer" as on OS4.1 update 2:

a) When writing a DVD (via AmiDVD), max speed is still only 4xDVD, no matter what speed I set in AmiDVD (my DVD writer supports up to 20x and DVD records up to 16x). This is strange.

b) ..and it's almost impossible to access files on my disk partitions while burning. The CPU usage is low (which is good), but even though, the whole system seems to be "locked" with given intervals (if playing a video in the background, video freezes with given intervals).
PS: This is not an issue when burning via an external USB writer, and I am using AmiDVD in both cases.
So the reason is probably SATA related.

c) Problems with multitasking when using DVD ROM (this issue is probably related to the one above).
Here is what I do and the results:
- I start copying a file from DVD (1GB file),
- The CPU usage is 10% according to CPUinfo
- I start OpenSFV to process files on my HD simultaneously,
- The CPU usage is now about 13-15% as long as both processes are running,
- When the copying is finished, CPU usage increases to 80%, and OpenSFV gets the resources it need.
This means the system prevents other tasks using resources while reading from a DVD-rom.

d) When copying files from a DVD to HD, the max transfer rate seems to be 2.5 MB/sec, even though the DVD rotates in full speed.

e) IDETool. The DVD's spins always up to MAX rotating speed, and even though I set the reading speed to 1x in IDETool, the rotating speed is still max (causes unnecessary noise, heat and wear).


7) I'm also experiencing "lockups" in some cases when writing to HD. Normally, writing to HD is fast, but in some cases, this happens:
- the copying process stops for a while,
- CPU usage runs at idle.
- While this happens, the filesystem is "locked", so multitasking is limited. Example: Video playback stops (possibly due to limited disk access), and I cannot get access to directories and files via either the WB or Directory Opus.
PS: I don't know exactly why the writing process stops (in my test case it was a large file, so maybe the file system had problems finding fragments large enough? Just speculations..), but it isn't this I think is the real issue.
The issue is the "locking" of the (file)system, which prevents me from doing other tasks while I wait for the file copy to finish.

8) "Waving effect" of the displayed image in f.ex. Multiview when scrolling a shown image horizontal or vertical.



Details of my system:
- Hardware:
o ACube SAM440EP 667MHz (mini-ITX) with 512 MB RAM and onboard Radeon M9 gfx chip.
o SATA controller is also onboard (Silicon Image SII3114, I think).
o As storage, I use a 30GB OCZ SSD ("Solid series") and a 120GB OCZ SSD ("Vertex 2").
o Internal SATA DVD writer is a Plextor PX-820SA.
o External USB DVD writer is a Samsung WriteMaster SE-S224.
o Mice: "Razer Copperhead" and "AEon Amiga" (with boing ball logo). I am using only one at once.
o Keyboard: Logitech.
o USB hub (4 ports).
- Patches for the hardware:
o Acubes setup tool for Sam440ep (increases VIDEO BUS read speed from 14 to 39 MB/sec).
- OS:
o Amiga OS4.1 update 3.
- Filesystem:
o SFS2 on the disks in my tests above.


Feel free to ask if you need more details to easier solve the issues.
I'll be more than happy to help.



Below are output from GFXBench2D-results run on my system with OS4.1 update 3 and 2 accordingly.

GFXBench2D on update 3 shows these numbers (What "copy from RAM to VIDEO RAM" is concerned, the copy*-commands are slower than on update 2, except "WritePixelArray" which is much faster):
-------------------------------------------
System Information:
OS: AmigaOS Release 4.1 - Update 3
Motherboard: Sam440EP
CPU: AMCC PPC440EP, 1.3 @ 666.667 MHz
L1 Cache Size: 32768, L2 Cache Size: 0, L3 Cache Size: 0
Total RAM: 512 MiB
External Bus (FSB) Speed: 133.333 MHz
Exec.library version: 53.22
Graphics.library version: 53.4
Intuition.library version: 53.32
Picasso96API.library version: 2.356
Rtg.library version: 41.4355


Opening screen: P96-0:Radeon M9:1280x1024

Board name: Radeon M9
Product ID: 0x4c66 Vendor ID: 0x1002 SubProduct ID: 0x4c66 SubVendor ID: 0x1002
Board driver: Unknown (0, 0)
Chip driver: Unknown (0, 0)
VRAM: 63.9844 MiB
Display mode: 1280x1024@60 (32 bpp)
Copy from RAM to VRAM:
Transfer size: 15452160 bytes
Src: 0x59325000, Dest: 0xa8da6d00
copy32: 52.407 MiB/s (took 0.281190 seconds)
copy64: 53.357 MiB/s (took 0.276186 seconds)
copy64f: 53.417 MiB/s (took 0.275871 seconds)
copy64x2: 52.402 MiB/s (took 0.281218 seconds)
copy64fx2: 53.448 MiB/s (took 0.275711 seconds)
copy64fx2PF: 53.447 MiB/s (took 0.275720 seconds)
copy64fx4PF: 53.445 MiB/s (took 0.275731 seconds)
useMemcpy: 26.487 MiB/s (took 0.556353 seconds)
useExecCopyMem: 53.333 MiB/s (took 0.276309 seconds)
copyToVRAM: 52.488 MiB/s (took 0.280756 seconds)
WritePixelArray: 90.989 MiB/s (took 0.161958 seconds).

Copy from VRAM to RAM:
Transfer size: 15452160 bytes
Src: 0xa8da6d00, Dest: 0x59325000
copy32: 26.601 MiB/s (took 0.553985 seconds)
copy64: 28.472 MiB/s (took 0.517564 seconds)
copy64f: 30.294 MiB/s (took 0.486438 seconds)
useMemcpy: 10.245 MiB/s (took 1.438438 seconds)
useExecCopyMem: 26.596 MiB/s (took 0.554079 seconds)
copyFromVRAM: 29.993 MiB/s (took 0.491327 seconds)
ReadPixelArray: 33.809 MiB/s (took 0.435872 seconds).





...and these on update 2:
----------------------------------------
System Information:
OS: AmigaOS Release 4.1 - Update 2
Motherboard: Sam440EP
CPU: AMCC PPC440EP, 1.3 @ 666.667 MHz
L1 Cache Size: 32768, L2 Cache Size: 0, L3 Cache Size: 0
Total RAM: 512 MiB
External Bus (FSB) Speed: 133.333 MHz
Exec.library version: 53.12
Graphics.library version: 53.4
Intuition.library version: 53.30
Picasso96API.library version: 2.356
Rtg.library version: 41.4340


Opening screen: P96-0:Radeon M9:1280x1024

Board name: Radeon M9
Product ID: 0x4c66 Vendor ID: 0x1002 SubProduct ID: 0x4c66 SubVendor ID: 0x1002
Board driver: Unknown (0, 0)
Chip driver: Unknown (0, 0)
VRAM: 63.9844 MiB
Display mode: 1280x1024@60 (32 bpp)
Copy from RAM to VRAM:
Transfer size: 15452160 bytes
Src: 0x484ee000, Dest: 0xa8d322e0
copy32: 63.009 MiB/s (took 0.233878 seconds)
copy64: 55.838 MiB/s (took 0.263911 seconds)
copy64f: 63.218 MiB/s (took 0.233103 seconds)
copy64x2: 62.984 MiB/s (took 0.233971 seconds)
copy64fx2: 63.219 MiB/s (took 0.233100 seconds)
copy64fx2PF: 56.008 MiB/s (took 0.263109 seconds)
copy64fx4PF: 63.206 MiB/s (took 0.233149 seconds)
useMemcpy: 31.548 MiB/s (took 0.467104 seconds)
useExecCopyMem: 55.851 MiB/s (took 0.263853 seconds)
copyToVRAM: 63.204 MiB/s (took 0.233155 seconds)
WritePixelArray: 62.760 MiB/s (took 0.234803 seconds).

Copy from VRAM to RAM:
Transfer size: 15452160 bytes
Src: 0xa8d322e0, Dest: 0x484ee000
copy32: 20.961 MiB/s (took 0.703020 seconds)
copy64: 22.791 MiB/s (took 0.646581 seconds)
copy64f: 22.857 MiB/s (took 0.644705 seconds)
useMemcpy: 11.023 MiB/s (took 1.336833 seconds)
useExecCopyMem: 20.958 MiB/s (took 0.703147 seconds)
copyFromVRAM: 22.860 MiB/s (took 0.644627 seconds)
ReadPixelArray: 21.882 MiB/s (took 0.673447 seconds).
User avatar
jostein_aarbakk
Beta Tester
Beta Tester
 
Posts: 37
Joined: Fri Aug 05, 2011 12:08 am
Location: Norway

Re: Amiga OS4.1 update 3 issues

Postby Gebrochen » Wed Nov 02, 2011 2:19 pm

Your issue 7 interests me

I experience worse than this, simply put, any file larger than 3mb seems to stay on 0% copied, and the only way to fix this issue is to use the shell command to copy 700 mb or more files across to portable HDD or USB keys.

I am thinking of starting a seperate thread for this issue,

as I also tested my rear ports, while placing my kybd and mouse to the front ports, still no go, it seems the only way for me to copy anything is to use shell.

Hyperion better fix this for the new users, who are definately going to only use GUI and no command line interfaces.

(Side NOTE:)
BTW, while im at it, thank you Hyperion for continuously supporting and developing Amiga Operating System and expanding the new gen hardware it will and can run on.

Cheers
Gebrochen
 
Posts: 5
Joined: Wed Oct 19, 2011 12:46 pm

Re: Amiga OS4.1 update 3 issues

Postby xenic » Wed Nov 02, 2011 4:00 pm

@Gebrochen
I just copied an 18MB raw photograph file from USB stick to RAM: and then copied the file back to the USB stick with WorkBench drag'n drop. No problem here. If nobody else can reproduce the problem then you probably need to tell us a little more about your system. I'm using a SAM Flex 800 with only printer, mouse, keyboard plugged into USB with no external hubs. I tested while online with one ethernet port connected to an ethernet wireless adapter. The USB stick is FAT formatted and I am using the internal sound and the graphics card that was supplied with my system.
AmigaOne X1000 with 2GB memory - OS4.1 FE
xenic
 
Posts: 1160
Joined: Sun Jun 19, 2011 1:06 am

Re: Amiga OS4.1 update 3 issues

Postby xenic » Wed Nov 02, 2011 4:57 pm

@jostein_aarbakk
When you cover so much territory in one post you are less likely to get a response because readers are overwhelmed. Here are my comments on your numbered issues:
1. This is a similar problem I have when plugging a mouse in through at least one of my USB hubs. If the mouse/hub is plugged in before a reset it will be recognized, but will not be recognized after a warm reboot. There have been a number of reported problems with USB hub, mouse & keyboard recognition so I think that issue may get some work before the next update.

2. I've had a similar problem with my keyboard since Update1. Occasionally (rarely) it becomes deactivated for no apparent reason. I added an icon to AmiDock that runs "UsbCtrl Restart" from a script which reactivates my keyboard.

3 & 4. I don't play many videos or perform speed tests so I can't comment on these.

5. USB read/write speeds are significantly slower to FAT formatted media that to SFS formatted media. If you have spare USB sticks or memory cards, you might want to try formatting one with SFS and compare the speed. Personally, I have formatted most of my USB sticks with SFS because it's faster and less error-prone. Others have speculated in other forums that there are some problems with CrossDos.

6. The only thing I can suggest regarding your DVD writer is to check the transfer setting with NVGetvar to see if your DVD writer is using PIO mode or a DMA mode for transfers.

7. The only comment I can make here is a personal opinion that multitasking doesn't seem to works as well as in some previous versions of OS4.xx but it could be a third part software issue. For example, if I move the mouse around while Loview or Picshow are loading large JPG images the pointer will temporarily freeze right before the image is displayed. If I load the same image with WarpView there is no pointer freeze. I could be wrong but it seems like some programmers are using methods to speed up their program performance at the expense of smooth multitasking (An opinion; not a fact or accusation).

8. Can't reproduce with Multiview.

Sorry I can't be more helpful.
AmigaOne X1000 with 2GB memory - OS4.1 FE
xenic
 
Posts: 1160
Joined: Sun Jun 19, 2011 1:06 am

Re: Amiga OS4.1 update 3 issues

Postby mechanic » Wed Nov 02, 2011 6:37 pm

jostein_aarbakk wrote:Details of my system:
- Hardware:
o ACube SAM440EP 667MHz (mini-ITX) with 512 MB RAM and onboard Radeon M9 gfx chip.
o SATA controller is also onboard (Silicon Image SII3114, I think).
o As storage, I use a 30GB OCZ SSD ("Solid series") and a 120GB OCZ SSD ("Vertex 2").
o Internal SATA DVD writer is a Plextor PX-820SA.
o External USB DVD writer is a Samsung WriteMaster SE-S224.
o Mice: "Razer Copperhead" and "AEon Amiga" (with boing ball logo). I am using only one at once.
o Keyboard: Logitech.
o USB hub (4 ports).
- Patches for the hardware:
o Acubes setup tool for Sam440ep (increases VIDEO BUS read speed from 14 to 39 MB/sec).
- OS:
o Amiga OS4.1 update 3.
- Filesystem:
o SFS2 on the disks in my tests above.


Just some notes. Same SAM440ep with up3.

First thing I would recommend is to deactivate Acube setup tool.
When I changed to a wireless mouse/keyboard (during up2) my related problems disappeared. They are on front port.
I have a 4 port hub. The hub is Full-Speed (ohci) not Hi-Speed (ehci). Not active during boot of course.

I just finished moving a 1.4GB AVI file around. DVD>Internal HD(sfs). DVD>External PATA HD(sfs) on USB. USB HD> Internal HD.
Transfer rates go between 2.5MB/S to 6.5MB/S with no problems.
I can watch and listen to a commercial DVD with DVplayer while playing a AVI vid with Mplayer (sound off) without serious frame
skip or lockups.

Could you plug in only the mouse & keyboard. Then post your usb.log file before plugging in anything else please.
A-Eon A1X1000 ATI HD6850, Creative SB1570 PCIe, RTL8139 net PCI.
User avatar
mechanic
 
Posts: 503
Joined: Sat Jun 25, 2011 10:22 pm

Re: Amiga OS4.1 update 3 issues

Postby jostein_aarbakk » Sat Nov 12, 2011 11:21 pm

@Xenic:
Thanks for your reply. I really hope Hyperion will fix these issues in a future update of the Amiga OS.

xenic wrote:@jostein_aarbakk
1) .. so I think that issue may get some work before the next update.

Great! Also, I have done a few more tests yesterday, and the problem seem to occur when attaching the mouse to my external 4 port USB HUB. This was not a problem with update 2.
Because of this, I have stopped using an external USB HUB until the bug is fixed.

xenic wrote:@jostein_aarbakk
2) .. I added an icon to AmiDock that runs "UsbCtrl Restart" from a script which reactivates my keyboard.

Sounds like a good idea. Will become handy if/when I encounter problems with my keyboard (which for now is attached to the USB 1.1 port of my SAM without an external HUB).
I forgot to mention that the trick I used when the mouse got deactivated, was to unplug/plug the mouse. This made it work a few minutes until the next deactivation (but quite burdensome..).

xenic wrote:@jostein_aarbakk
5) USB read/write speeds are significantly slower to FAT formatted media that to SFS formatted media. Others have speculated in other forums that there are some problems with CrossDos.

You are right regarding the write speed of FAT vs SFS formatted media, but my system seems to have a MAX reading speed of about 6 MB/sec (se below). This is strange.
Here are the results from my recent tests:
- READ speed of a CompactFlash card via USB is the same for FAT-formatted media as SFS formatted media (5.9 MB/sec).
- WRITE speed for FAT-formatted meda is about half (2.4 MB/sec) of a SFS formatted media (4.9 MB/sec).
- READ speed of a DVD-R read by an external USB DVD-writer is about the same as reading from a CompactFlash card (6.2 MB/sec vs 5.9 MB/sec).

The flash card used, is a 1GB CompactFlash SanDisk ExtremeIII. This card supports READ/WRITE speeds of at least 20 MegaBytes/sec.
The external USB DVD-writer has max. reading speed of 16x DVD.
Tests were done by copying the same file containing 142 MB. and done via "Drag&Drop" in Workbench to/from RAM.


xenic wrote:@jostein_aarbakk
6) The only thing I can suggest regarding your DVD writer is to check the transfer setting with NVGetvar to see if your DVD writer is using PIO mode or a DMA mode for transfers.

It's using DMA mode. NVGetvar says "sii3114ide_xfer=GGGG".
A few days ago, I changed "sii3114ide_xfer" to "GG0G" (after reading another post on this forum), but the results were exactly the same.
Below are some of my current values returned by NVGetVar:
s4sii_maxbus=1
ide_doreset=on
ide_reset_timeout=10
ide_cd_timeout=20
sii3114ide_xfer=GGGG

If you experience problems reconstructing the issues mentioned in 6a)-e), feel free to ask, and I'll help the best I can.


xenic wrote:@jostein_aarbakk
7) The only comment I can make here is a personal opinion that multitasking doesn't seem to works as well as in some previous versions of OS4.xx but it could be a third part software issue.

It could, but most likely not in my case.
The reason is that in my case, I used "Drag&Drop" file copy, which is standard functionality in the OS.
My "best guess" is that the source of this bug has something to do with either the filesystem itself or some related internal software components.


xenic wrote:@jostein_aarbakk
8) Can't reproduce with Multiview.

Actually, I have seen this "effect" not only in Multiview, but also OWB (the port from MorphOS).
The bug is present both with and without Acubes "Sam440ep setup".
Here is how you'll find it:
- Start MultiView with an image larger in the vertical direction than the visible screen.
- Move the vertical scrollbar slowly downwards, and look at the center of the screen. You will se a kind of thin "stripe" moving downwards as you scroll down.
- Move the vertical scrollbar slowly upwards. You will see the effect here, also.
User avatar
jostein_aarbakk
Beta Tester
Beta Tester
 
Posts: 37
Joined: Fri Aug 05, 2011 12:08 am
Location: Norway

Re: Amiga OS4.1 update 3 issues

Postby jostein_aarbakk » Sun Nov 13, 2011 1:16 am

@Mechanic:
Thanks for your reply.

mechanic wrote:First thing I would recommend is to deactivate Acube setup tool.

I tried this, but none of my issues disappeared.
The only sideeffect was that the VIDEO BUS READ speed got reduced from 39 to 14 MB/sec (according to RageMem), and that's not a good thing.


mechanic wrote:When I changed to a wireless mouse/keyboard (during up2) my related problems disappeared. They are on front port.
I have a 4 port hub. The hub is Full-Speed (ohci) not Hi-Speed (ehci). Not active during boot of course.

Until a few days ago, the HUB was always connected, and the mouse connected to it.
Now, I have stopped using the external HUB completely, and the mouse problems disappeared.
(But if I connect the mouse to the HUB again, even after bootup, the problems appear at once).

I have been using all 4 "onboard" USB ports on Sam:
- keyboard, mouse->upper, lower USB port on the back of Sam accordingly
- "USB FlashMemory card reader" and "USB external DVD writer" -> the other 2 ports (in front of my computer case)


mechanic wrote:I just finished moving a 1.4GB AVI file around. DVD>Internal HD(sfs). DVD>External PATA HD(sfs) on USB. USB HD> Internal HD.
Transfer rates go between 2.5MB/S to 6.5MB/S with no problems.

I think that the transfer rate to/from an USB device should not be limited to 6 MB/sec, but either up to the limit of USB 2.0 or the device itself.
The 6 MB/sec limit is too low.
The same goes for the 2.5 MB/sec limit when reading from an internal SATA DVD-rom drive.
This is strange.

Here are the results from my recent tests:
- READ speed of a CompactFlash card via USB is the same for FAT-formatted media as SFS formatted media (5.9 MB/sec).
- WRITE speed for FAT-formatted meda is about half (2.4 MB/sec) of a SFS formatted media (4.9 MB/sec).
- READ speed of a DVD-R read by an external USB DVD-writer is about the same as reading from a CompactFlash card (6.2 MB/sec vs 5.9 MB/sec).
- READ speed of a DVD-R read by an internal SATA DVD-writer is less than half as fast as an external USB writer (2.5 MB/sec vs 6.2 MB/sec).


mechanic wrote:I can watch and listen to a commercial DVD with DVplayer while playing a AVI vid with Mplayer (sound off) without serious frame
skip or lockups.

I have run some more tests just to be sure, and here are the results:
- If you start f.ex. DVPlayer, and then start reading a DVD media from your internal SATA DVD-rom, DVPlayer stops a few seconds (until the rotating speed has reached maximum?), but after the "freeze", video playback is actually shown as normal (don't know if this always is the case, or if the file size plays a role etc.)
- If you start a file copy (via WB's "drag&drop", OpenSFV or something else that involves heavy disk access), and then start reading a DVD media from your internal SATA DVD-rom, the CPU-usage drops, and the disk-accessing task uses very long time to finish. When the DVD-reading is finished, the CPU-usage gets normal again.
- If you do the same job as above, but from an external USB DVD-rom instead of the internal SATA DVD-rom, the "lockup"-problems of other disk-related tasks are not present.

To test the 2nd. bullet above, you can do the following:

Before you start, please make sure you have CPUInfo installed:
http://os4depot.net/share/utility/docky ... _docky.lha

1st. test: Do this in the following sequence while measuring the elapsed time:
1) Start copying a file from your "internal SATA HD" to another folder on the same HD via Workbench. The file must be big in size (1 GB is suitable).
2) Watch the CPU-usage CPUInfo is showing. Mine shows 46-49%.
3) Start copying another file via Workbench, but now from your "internal SATA DVD-drive".
The file must be smaller than the above, try one with about 80 MB.
The point is that this copy process must finish before the 1st. one.
4) Watch CPUInfo again. Mine shows 15-17%.
5) Watch the progress indicator of the "DVD to HD" copy process, and wait until it's finished (or just cancel).
6) Watch CPUInfo again while the copy in step 1) is still running. Mine shows normal CPU usage again (46-49%).

2nd. test: Do this while measuring the elapsed time:
1) Step 1) above.
2) Watch the CPU-usage CPUInfo is showing. Mine shows 46-49%.

What I have noticed, is that the elapsed time in the first test is way longer than in the 2nd., which proves that the DVD copy process is preventing other disk-related tasks to run at full speed.


mechanic wrote:Could you plug in only the mouse & keyboard. Then post your usb.log file before plugging in anything else please.


Below is the usb.log file immediatly after startup.
No USB HUB is used.
Only the keyboard and mouse are connected.

I: [19:50:06] USB stack | | ==========
I: [19:50:06] USB stack | | USB System Software Task started
I: [19:50:06] USB stack | | Starting HCDs...
E: [19:50:06] USB stack | New Fkt | fkt: 0x5FF8ADC0 = Level -1
E: [19:50:06] USB stack | New Fkt | fkt: 0x5FF8AE60 = Level 0
E: [19:50:07] USB stack | New Fkt | fkt: 0x5FF8AF00 = Level -1
E: [19:50:07] USB stack | New Fkt | fkt: 0x5FE15010 = Level 0
E: [19:50:07] USB stack | New Fkt | fkt: 0x5FE150B0 = Level -1
E: [19:50:07] USB stack | New Fkt | fkt: 0x5FE15150 = Level 0
I: [19:50:07] USB stack | | ...HCDs started.
I: [19:50:07] USB Fkt Init | Init Fkt | Fkt 0x5FF8AE60 is {Vendor: 0x0000, Product: 0x0000, Class: 09.00}
I: [19:50:07] USB Fkt Init | Init Fkt | Fkt 0x5FE15010 is {Vendor: 0x0000, Product: 0x0000, Class: 09.00}
I: [19:50:07] USB Fkt Init | Init Fkt | Fkt 0x5FE15150 is {Vendor: 0x0000, Product: 0x0000, Class: 09.00}
I: [19:50:07] USB Fkt Init | Init Fkt | Fkt 0x5FF8AE60 initialized
I: [19:50:07] USB Fkt Init | Init Fkt | Fkt 0x5FE15010 initialized
I: [19:50:07] USB Fkt Init | Init Fkt | Fkt 0x5FE15150 initialized
I: [19:50:07] hub.usbfd | HUB | Hub "Rear Roothub (ohci.usbhcd/0)" has TT thinktime of 8 full speed bittimes and protocol 0
I: [19:50:07] hub.usbfd | HUB | Hub "Rear Roothub (ehci.usbhcd/0)" has TT thinktime of 8 full speed bittimes and protocol 0
I: [19:50:07] hub.usbfd | HUB | Hub "Front Roothub (ohci.usbhcd/1)" has TT thinktime of 8 full speed bittimes and protocol 0
I: [19:50:07] hub.usbfd | HUB | All ports are empty
I: [19:50:07] EHCI Controller Task Unit 0 | EHCI | Device at port 0 is a lowspeed device. Giving up ownership.
I: [19:50:07] hub.usbfd | HUB | All ports are empty
I: [19:50:07] hub.usbfd | HUB | All ports are empty
I: [19:50:07] hub.usbfd | Sys_EndInitialAttachmentPhase | Initial USB Attachment Phase terminated
I: [19:50:07] hub.usbfd | HUB | Device found on port 1 of hub "Rear Roothub (ohci.usbhcd/0)".
I: [19:50:07] hub.usbfd | Hub port | Device at port 1 of hub "Rear Roothub (ohci.usbhcd/0)" is low-speed.
E: [19:50:07] USB stack | New Fkt | fkt: 0x5FE153D0 = Level 1
I: [19:50:07] hub.usbfd | HUB | Device on hub port 1 successfully added.
I: [19:50:07] USB Fkt Init | Init Fkt | Fkt 0x5FE153D0 is {Vendor: 0x046D, Product: 0xC312, Class: 00.00}
I: [19:50:07] USB Fkt Init | Init Fkt | Fkt 0x5FE153D0 initialized
I: [19:50:07] EHCI Controller Task Unit 0 | EHCI | Device at port 1 is a lowspeed device. Giving up ownership.
I: [19:50:08] hub.usbfd | HUB | Device found on port 1 of hub "Front Roothub (ohci.usbhcd/1)".
I: [19:50:08] hub.usbfd | Hub port | Device at port 1 of hub "Front Roothub (ohci.usbhcd/1)" is low-speed.
E: [19:50:08] USB stack | New Fkt | fkt: 0x5FE15470 = Level 1
I: [19:50:08] hub.usbfd | HUB | Device on hub port 1 successfully added.
I: [19:50:08] USB Fkt Init | Init Fkt | Fkt 0x5FE15470 is {Vendor: 0x046D, Product: 0xC050, Class: 00.00}
I: [19:50:08] USB Fkt Init | Init Fkt | Fkt 0x5FE15470 initialized
I: [19:50:12] USB stack Process | | USB System Software Process started.
I: [19:50:12] USB stack Process | | Reading FD descriptors...
I: [19:50:12] USB stack Process | | ...FD infos read
I: [19:50:12] USB stack Process | Rebind FD | Weak FD "bootmouse.usbfd" for Ifc 0x5FE51E10 is being preempted.
I: [19:50:12] USB stack Process | Rebind FD | Weak FD "bootkeyboard.usbfd" for Ifc 0x5FE51B90 is being preempted.
I: [19:50:12] USB stack Process | | Starting HCDs...
I: [19:50:12] USB stack Process | | ...HCDs started.
I: [19:50:12] USB stack Process | Sys_EndInitialAttachmentPhase | Initial USB Attachment Phase terminated
I: [19:50:12] USB stack Process | | USB stack fullbooted


Here is what's being logged next after I connect my USB-hub (but without connecting anything to the HUB itself):
I: [19:55:55] hub.usbfd | HUB | Device found on port 4 of hub "Rear Roothub (ehci.usbhcd/0)".
I: [19:55:56] EHCI Controller Task Unit 0 | EHCI | Device at port 3 is a fullspeed device. Giving up ownership.
I: [19:55:56] hub.usbfd | HUB | Port 4 of hub "Rear Roothub (ehci.usbhcd/0)" reset timed out
W: [19:55:56] hub.usbfd | HUB | Resetting hub port 4 and trying again
I: [19:55:56] hub.usbfd | HUB | Reset request for port 4 failed (error -23)
E: [19:55:56] hub.usbfd | HUB | Resetting port 4 of hub "Rear Roothub (ehci.usbhcd/0)" failed. Device adding aborted
I: [19:55:58] hub.usbfd | HUB | Device found on port 2 of hub "Front Roothub (ohci.usbhcd/1)".
I: [19:55:58] hub.usbfd | Hub port | Device at port 2 of hub "Front Roothub (ohci.usbhcd/1)" is full-speed.
E: [19:55:58] USB stack | New Fkt | fkt: 0x5a347710 = Level 1
I: [19:55:58] hub.usbfd | HUB | Device on hub port 2 successfully added.
I: [19:55:58] USB Fkt Init | Init Fkt | Fkt 0x5A347710 is {Vendor: 0x1A40, Product: 0x0101, Class: 09.00}
I: [19:55:58] USB Fkt Init | Init Fkt | Fkt 0x5A347710 initialized
I: [19:55:58] hub.usbfd | HUB | Hub "USB 2.0 Hub [Safe" has TT thinktime of 8 full speed bittimes and protocol 0
I: [19:55:58] hub.usbfd | HUB | All ports are empty
User avatar
jostein_aarbakk
Beta Tester
Beta Tester
 
Posts: 37
Joined: Fri Aug 05, 2011 12:08 am
Location: Norway

Re: Amiga OS4.1 update 3 issues

Postby mechanic » Mon Nov 14, 2011 11:51 pm

[quote="jostein_aarbakk"][/quote]
I tried your tests with somewhat similar results.

Your USB.log looks just fine.

I do not think usb speed is your problem. Keep in mind the max speed in any kind
of data/file transfer is the write speed, not the max transfer speed nor the read
speed. Usually writing takes at least 50% longer than read and as much as 75% longer.

If a format conversion is involved than even longer.

You can see this by taking a 50 +- MB file and using Workbench copy it to RAMDISK,
Then from ram to a usb stick and note the time. Then Rename the file on the usb
and copy it back to RAM. You should see a noticeable difference.

My 440ep reads from a usb stick almost 2.5 times faster than writing to it.

I'm not saying you don't have some problem, and no idea what it is yet, just that
I do not think it is usb transfer speed.
A-Eon A1X1000 ATI HD6850, Creative SB1570 PCIe, RTL8139 net PCI.
User avatar
mechanic
 
Posts: 503
Joined: Sat Jun 25, 2011 10:22 pm

Re: Amiga OS4.1 update 3 issues

Postby JosDuchIt » Tue Nov 15, 2011 10:36 am

@xenic
2. I've had a similar problem with my keyboard since Update1. Occasionally (rarely) it becomes deactivated for no apparent reason. I added an icon to AmiDock that runs "UsbCtrl Restart" from a script which reactivates my keyboard.


where is UsbCtrlRestart to be found?
JosDuchIt
 
Posts: 291
Joined: Sun Jun 26, 2011 6:47 pm

Re: Amiga OS4.1 update 3 issues

Postby Gebrochen » Tue Nov 15, 2011 3:28 pm

@ helpers,

sorry for late reply, busy, busy, busy,

Yes, I can copy using shell,

Yes I can also copy, and thus far more stable, using Directory Opus......................

I will continue the saga hopefully this weekend.

It seems only my workbench with the issue.

Soon I am to have another sam system, assuming all goes well, hence, two sam systems, 440 and by the end of the week a sam flex..............the more people supporting the better, haha.

(please do not get offended with the writings below, we are in an AMIGA forum after all, so I think I can spit my thoughts)

OK, so I will test this USB copy thing on practically a fresh system, if no issues like other users posting here, then clearly it is possible to muck up OS4.x operating system itself, almost just like a windows system, then needing to re install the OS (unheard of in older AOS's)???? :-O Yes I just mentioned the enemy, I apologise, dont worry, do not use it at home..........only at work and not by choice (get payed to use shit, haha).

Cheers

P.S. haha, cant wait to test this issue on the flex, hopefully not the same..........that would indeed pose something serious to fix for GUI only users
Gebrochen
 
Posts: 5
Joined: Wed Oct 19, 2011 12:46 pm

Next

Return to Platform: AmigaOne 500 and Sam series

Who is online

Users browsing this forum: Google [Bot] and 2 guests