I thought it was maybe a problem with not enough RAM, but from responses I got here in Hardware section, it seems that is not the problem. Doubt that it’s over heating, in BIOS the temperature readouts are 32C CPU and 17C system. Using lmsensors I get readouts of 0C CPU and -10C GPU, which is obviously NOT correct.
I currently have the open source video driver, heard lots of other people saying that the fglrx driver causes the system to not boot into gui or not boot at all.
Is the whole problem due to graphics driver? Or something else?
SSD (not currently installed or in use, will take a performance hit from mobo having only SATA II ports, have to learn how to install for SSD as boot and HDD for storage… and noticing this shows as 256GB, mine is 128GB) http://www.newegg.com/Product/Product.aspx?Item=N82E16820227791
In case it’s hard to see it says
[firmware bug]: AMD-Vi: IOAPIC[0] not in IVRS table
[firmware bug]: AMD-Vi: No southbridge IOAPIC found
AMD-Vi: Disabling interrupt remapping
did you install the ati propitiatory drivers (I’m not sure if they work with build-in and gpu’s they should)
unlike intel’s build-in gpu which is supported nativly by the Linux Kernel amd’s are not see this page https://en.opensuse.org/SDB:AMD_fglrx
no I have not installed amd/ati proprietary driver
in KInfoCenter I can see for Graphical Information - OpenGL it lists 3D accelerator ‘unknown’ and Renderer as ‘Gallium 0.4 on AMD ARUBA’
OpenGL/ES version 3.0 Mesa 10.3.7
the worst seems to be with anything that is flash based (firefox won’t even display, chromium is the only browser I can really use, chrome wouldn’t even install)… while any flash content loads the screen just shows blank background and I have to minimize the browser then restore it for anything to even show up. network shows a downspeed of 1.8MB/s (I guess that is the max speed I can get on this N300 wireless adapter?), it’s almost as if the CPU is pausing momentarily every few seconds or so.
Here’s what I see for readout of ‘top’
no idea what half of that is or if it should even be there.
You can edit that with Yast2-bootloader in > Kernel Parameters > Optional Kernel Command Line Parameter
or by editing the file with superuser rights, like opening a terminal and typing:
su -
<root password>
nano /etc/default/grub
edit as required, then Ctrl+O, <Enter>, Ctrl+X.
You might have to install “nano” or use your favourite editor if you can’t do that.
Then refresh grub as written by tsu2.
OK, made the change via Yast2-bootloader
updated the grub as tsu2 said
rebooted machine (that was the longest reboot I’ve ever seen since an old Intel Pentium machine)
it still shows the same bug problem and everything still seems to be crawling like a snail… it still keeps flashing the HDD activity LED almost non stop as well.
in a console run top and see what process is eating the cpu cycles. perhaps it is the indexer it can take time to index things and can use a large amount of resources.
I had already posted before the readout of top. It hasn’t changed any. From what I see 8.357% is the highest CPU use. Should there really be such a huge gap of nothing like that between where I put in the command and it actually starts outputting info?
I think they just make hardware and software far too complicated and they should try simplifying it.
Nothing unusual in your “top” output, your A4 APU is no roaring beast but the problem seems not to be there.
BTW “top” here shows just two empty lines before useful output, maybe the gap you see is part of your problem.
tsu2 gave an useful hint pointing to IOAPIC, maybe you need to experiment with different parameters; your MoBo might have a different layout compared to those in the original Ubuntu thread. I can’t be of any further help here, never touched an A4 APU (yet), sorry.
The disk light constantly flashing and the snail-booting might point to a nearly full /root partition or other disk problems.
Posting the output of “df”, or of “btrfs filesystem df /” if you are using btrfs for /root, might help if you see something odd.
OK, here is the readout for both of those. I know this APU is not exactly the best out there, but when my Pentium 4 machine died unexpectedly (motherboard stopped POSTing but still gets power), I had to get something and this was the best machine I could afford. Sadly, about 2 weeks after getting this one I priced up components to build similar and could have just built my own with an A8-7600k, 2x8GB DDR3-1866 and a mobo that has USB3 and SATA3 all for about $100 less than I spent on this prebuilt machine. Lesson definitely learned!
Last time I did try smartctl… and unless I did something wrong with the command, it was absolutely useless. said it would take up to an hour to run, but it never even ran as far as I could tell.
linux:/home/justin # smartctl --test=long /dev/sdasmartctl 6.3 2014-07-26 r3976 [x86_64-linux-3.16.7-29-desktop] (SUSE RPM)Copyright (C) 2002-14, Bruce Allen, Christian Franke, www.smartmontools.org
=== START OF OFFLINE IMMEDIATE AND SELF-TEST SECTION ===
Sending command: "Execute SMART Extended self-test routine immediately in off-line mode".
Drive command "Execute SMART Extended self-test routine immediately in off-line mode" successful.
Testing has begun.
Please wait 60 minutes for test to complete.
Test will complete after Wed Feb 24 02:49:24 2016
Use smartctl -X to abort test. linux:/home/justin #
Beside for that, I did run a test outside of OS from Ultimate Boot CD. The program I ran (only one that would do anything with this drive) was Vivard 0.4
First test run showed 8 errors within the first 2% of scanning, but there was also a loose power cable that was causing a voltage drop.
Second run, with the power cable plugged right and proper voltage, showed no errors at all.
I do not know what to do, I do not understand one bit of any of this. One thing tells me the drive is IDE bus, another thing says it is SCSI. I look at the drive and it is SATA.
I’m about to just take a very big hammer to this whole machine. Never had this kind of problem on any older system that has the legacy BIOS… only once they screwed things up with that **** UEFI ****. Windows is garbage just like the idiot who made it! And capitalism can bend over and get what’s coming.
Don’t have any clue how it figures a bad overclock since I have NEVER overclocked any computer before and wouldn’t even have any clue how to do so.
I pressed F1, it went into BIOS, I simply clicked saved and reboot. Then it booted normally… still with the same firmware bugs listed and all.
So looks like I’m stuck dealing with the problem as is since there is no where to get help for free. Learning should not cost anything and especially not be forbidden to those of us who are **** poor!!
You can overclock but the settings that works in Windows may not work in Linux because of differences in timings.
On the other hand it may be a hardware problem???
A full smartctl can take a long time since it scans the entire disk. try the -a option it gives the history which should point to any trouble with the device
resetting the BIOS may erase any boot info so the BIOS may not know how to boot the installed OS(s). SO you may have to point it to the right OS to recreate the entry in its memory
Unfortunately you apparently stumbled on a HW / FW / Kernel problem that affects many, according to Mr. Google, but has no standard fix it seems…
**Let’s recap:
your CPU seems OK, no odd task eating up your meal
your disk partitions seem OK
it is normal for a SATA disk to be seen as SCSI (the SW interface) or be connected to the IDE bus (HW interface)
your disk HW might or might not be always OK (is the PWR connector fixed once and for all?)
an AMD-vi / IVRS / IOMMU problem plagues many over the net across distributions, hints here and there, no standard fix according to my searches.
**
I should say that we don’t know yet if the latter might be causing the snail-speed of your setup, from what we read so far.
The errors reported only say that the kernel is trying to workaround what it sees as a faulty BIOS table; but those might affect Input/Output performance and/or prompt servicing of interrupt requests from peripherals (disks, among others).
If you are not using virtualization (VirtualBox, VMWare, KVM etc.) and unless tsu2 comes up with a brilliant solution
you might try one or more of the following:
enter the BIOS and disable everything called Virtualization, AMD-vi, AMD-V, and /or labeled IOMMU
add “iommu=soft” to your boot command line (fromthis thread); you might just hit the “E” key at the boot screen and edit the “Linux” line without rebuilding grub, then hit F10 to boot.
If your system is still able to boot with any of these and the snail is gone, you might at least have a workable system for everyday work.
No, I have never overclocked any computer I’ve ever owned. Wouldn’t know how. Besides, if I understand it right, AMD APUs can NOT be overclocked unless they are the ones signified with a k (ex. A6-6400k is unlocked, my A4-6300 is locked)… but I suppose that I could be misunderstanding how that works too.
I will try those last few suggestions. If those don’t help either then I give up with this machine, I don’t have the patience or energy to deal with it any longer. I’ll just do what I can to scrape up enough extra funding to build a new machine and give this one away to whom ever wishes to deal with it.
This time I’ll make sure to build it myself and that it has USB3 and SATA3 6.0Gbps ports on the motherboard. Then I can use my SSD for booting and it will run full performance. Can’t fathom why on earth they are still making motherboards with only SATA2 at best and no USB3 when those are now so readily available, backwards compatible and the new standard.
edit
In the mean time, I will go back to using my old laptop. It’s a Dell XPS M1330 with Core 2 Duo T9300 (2.5GHz/6M/800MHz), 4GB (2x 2GB) DDR2-800, nVidia GeForce 8600M and a 500GB Seagate Momentus (SATA 3.0Gb/s, 16MB buffer/cache).
So just out of curiosity, is that to say this is only a problem for Linux users? Or would this also affect a Windows 8/8.1/10 install?
Mainly wondering on account of I had the same problem with it constantly slowing to a crawl when it had Windows 8.1 (pre-installed). Thought I’d go ahead and check out Win 8.1 while it was there. Hated it due to the constant slow downs. So maybe it wasn’t even OS related at all to begin with?
Now I don’t even see an edit button for my post(s) here…
Checked BIOS and there are no such options in any section. I did notice that there’s an option for SATA controller that allows for IDE, AHCI or RAID and it is set to AHCI.
Think I’m definitely going to just build a new system and use a good Gigabyte board. These MSI ones are just junk.
