Since I only have a keyboard and mouse plugged into the USB ports, I don't think it's because I'm pulling too much power on that circuit. Why did I get another? I didn't see it happen the first time. The PC had shutdown and I wasn't sure what happened. The second time I watched the LED go dark on my mouse, never to light up again on that port.

What gives?

Secondly, due to an unrelated issue of an unrecognized IDE by the BIOS, I had to reset the CMOS. After 5 or 6 resets to get the IDE drive working, the Dr DOS USB error that I had been seeing was also gone. The formerly non-working USB ports now worked and the boot time was back to 2 seconds.

For about a week. Then it was back to broken and the CMOS reset trick doesn't seem to be working this time. Is there some undocumented factory reset I just happened to stumble on before?

I've just purchased mine, and after one week I have the exact same problem as you earlier mentioned. One way to fix the loooong POST is to disable the USB Legacy. But, if you do that, you can no longer go into BIOS (mouse and keyboard doesn't work). I've also checked *ALL* the possible combinations into USB bios settings but the Intel USB 3.0 ports are NOT working at all.

I've also sent an e-mail to ASRock waiting for a reply. This is ... lame.

PS: I've updated to the latest firmware. Nothing happened.

I use Linux for my OS and I have this problem every time the bootloader updates. Every. Single. Time. This is my last ASRock board. I already have a replacement from a different brand. I need a stable, reliable machine and this is not cutting it.

Before you fly off the handle and start accusing moderators of being fanbois try contacting tech support directly with your issue and see if they are able to help you. Even though linux is not a supported OS there may well be something that can be done, be it reverting to a previous BIOS version or some workaround. 

[edit] The OS effects the UEFI so yes, it can be OS related. Unlike the old simple BIOS system the UEFI is designed to interact with the OS. Now it may not be the OS but the only way to rule it out is to install Windows and check if the problem persists. Please cease the inflammatory comments, the people posting here have been trying to help you and do not deserve to be insulted for taking the time to reply to you.

There is no way a bootloader update should cause my USB ports to stop working without a fault of some kind. This is a defect and if done by design, a design flaw. I'm sorry if you don't like to hear that.

[edit] Just saw your edit. The only people who tried to help me were the noobs. You guys all blew me off.

In your situation I would try and rule out hardware being the issue by doing as I stated, install windows and see if your ports fail there too. If that is the case then your particular board may have a defect or BIOS issue in which case an BIOS update or RMA would be warranted. Given the issue arises only when you perform an update it is fairly clear that Linux is making a change to the UEFI that is causing this problem. It may be enabling or disabling legacy USB support for example or any number of inaccessible, hidden controls within the UEFI that can only be set by the OS.  I think the misunderstanding here is stemming from the fact that you were not aware of how much the OS interacts with the UEFI. We understand what you are saying and are trying to provide info as to what we think may be the issue.

It's not so much as Linux as GRUB which gets updated. But here's the thing, even if GRUB is doing something wrong, which it may be, why can't I fix it even inside the BIOS config? And why does it only affect the Intel USB hub and not he ASMedia one? Or more generally, why is only USB effected? This is the cause of the slow boot BTW. The ports are there but they never respond.

As I said in my last post, there are more functions available in the UEFI than are present in the UI. Many of these features are only accessible by the OS/bootloader and are not alterable by the user. All kinds of things are determined by the UEFI and many of these settings could damage the system if the user were able to access them and alter them. The issue in your case seems to be stemming from the intel USB controller that is built into the chipset. As it is chipset based it is far more intricately connected to the UEFI than the Asmedia controller which has its own separate IC. It has been a long time since I delved into the inner workings of the BIOS but if I had to guess, GRUB may be trying to alter a setting related to bootable USB devices. This change is obviously not working correctly with ASRock's UEFI, possibly due to outdated coding in the GRUB system or because compatibility for your board is not properly implemented. 

Once you clear CMOS after an update the system functions as it should until another update is required. This is a good thing as it means the issue is likely external, stemming from something GRUB is changing. It should be easy to test if it is GRUB causing the issue by setting a hidden UEFI flag incorrectly. Next time you perform an update and the issue occurs (it may not if the update includes a fix for the issue) try resetting to factory defaults within your UEFI. If this works then the problem is likely a setting you have missed as it will reset all user alterable values. If it does not work but clear CMOS does then you know for certain the GRUB is altering something you cannot access. Clear CMOS resets all data including hidden registers and variables. This is the reason both exist and the further battery removal method will additionally reset the Real Time Clock or RTC as it is usually referred to.

If this is the case then your best bet is posting your issue on your distro's user forums and seeing if anyone there can help you. You can try Tech Support here directly and they may be able to help you but bare in mind they are not obligated to do so given the issue stems from an unsupported OS. Generally though ASRock's tech support is very helpful and eager to resolve user issues whenever they can.

I hope this helps clarify how intricately connected the OS, bootloaders and UEFI are. 

@Xalter -- that's the clearest explanation yet. I don't need to reset the BIOS to factory defaults because I never change anything there but I tried that anyway . It requires a CMOS reset to get the Intel ports working again. So it's not the hardware itself. At this point it *seems* to be a combination of GRUB setting something it shouldn't and ASRock UEFI letting it do something it shouldn't. Thank you. You've been very helpful.

I have everything working for now and I really don't want to intentionally cause a fault. My desktop is mounted inside a wooden cabinet so it's a real PITA to disassemble it and reset it. Besides I already know that if I run the GRUB installer (which happens every time there is a kernel update) this problem recurs. In fact, the last break was due to uninstalling an old kernel. So, it was the same exact kernel and just running  GRUB to remove an older unused one caused the fault. It's very repeatable but I don't want to repeat it again unless I have to. Only once of the five times this has happened was I able to *just* clear the CMOS and have everything work. The other times I've had to disconnect all the drives and remove my video card, clear it and then re-assemble it. Clearing the CMOS seems to be like driving a tack with a sledgehammer. This may explain some of my obvious frustration in earlier posts. If I have offended anyone, I am sorry.

I still don't know if GRUB is the culprit or merely the catalyst and I don't want to point fingers anyway. I'm just trying to get as much information out as I can. This is a real issue. Who's issue it belongs to is still open and to me irrelevant. I'm not a GRUB developer or a firmware engineer so who get's the workaround doesn't matter to me.

Dude if you think wood cages influence RFI you need to talk to Mr Faraday. Maybe if it was ironwood ...

I think you mean well but you're tilting at the wrong windmill here. I will try your experiment when it breaks again because why not, it's already broken.

Often this is what I do to resolve it:  shutdown the computer and switch off the power supply.  Pull out all USB peripherals, and then press and hold the power button to completely discharge the M/B capacitors (should turn on for like a half a sec).  I then turn it all back on without the keyboard/mouse.  Sometimes this fixes it immediately or after a restart or two - no need to reset the CMOS.  Not sure if being able to boot into a non-linux OS helps my cause or not.  Hopefully this will help some of you recover quicker from this USB port issue.

There was a kernel update today and also a grub&efi update but I have not rebooted yet. I'll wait till the weekend since I also use this machine for work. I can't afford to risk a several hour delay if I have to pull it apart again and reassemble it.

DIY OS X eh? Sounds interesting! My last laptop was OS X with Linux and Windows VMs but I never tried native dual/triple boot with it. Topic for another thread perhaps.

Thanks to everyone that has helped!