pcw.sys 0x0000007E BSOD

[acoul]

Greetings,

I am getting a BSOD on an Asrock D1800M board after installing UsbBoot-Installer. the system is a win-7 pro with all the latest recommended updates. on another system, a NUC6CAYH, the system boots just fine from the usb3.x ports. on another system, a GIGABYTE A320M-DS2, booting through usb fails while a regular boot doesn't produce any BSOD.

what would be the best way to a clean uninstall of UsbBoot-Installer ?

Edit: it also boots fine on my old ASRock K10N78FullHD-hSLI usb2.

a great & unique tool indeed.

[acoul]

Today I checked UsbBoot-Installer on an ASRock J3355M with the same results to those of Asrock D1800M. Luckily, I was able to recover from the BSODs by reverting to the Last Known Good Configuration which removed the UsbBoot-Installer changes. An uninstaller option though would be quite neat.

So far, I only had luck with UsbBoot-Installer on my Intel NUC6CAYH box. thereof, my bug report is as follows:

1). A320M-DS2 gives BSOD while booting on USB 3.0 port. On USB 3.1, it just hangs forever. AHCI boot works just fine.

2). Early USB-3.0 Intel platforms seem to give BSOD either booting from USB or SATA/AHCI. I suspect that this is a non usb-boot related issue, that if resolved, usb boot will be just fine.

on Device Manager with show hidden devices checked, I see many new USB-related devices with the boot flag checked on. I tried to disable a few in order to see if the BSOD would disappear, but the BSODs also appeared on my working ahci A320M-DS2.

feedback as to how further troubleshoot this case, would be quite welcomed.

[acoul]

I am using a modified win7pro installation. a friend of mine injected for me the most recent (as of Jan/2020) USB3 drivers on the boot.win image. So this image that I am using to perform these tests is kind of a blackbox for me. this image was created with the WinToUSB handy tool. I also have a regular win7pro image made with WinToUSB that is able to usb boot only on older usb2 boards. installing UsbBoot-Installer to it, didn't successfully booted on usb3 boards. The other problem I have with this "special" usb3 injected image, is that for some reason, I am unable to enable memory dumps upon BSOD no matter how hard I tried (probably not hard enough).

So just to share my adventures & findings, I renamed pcw.sys and was able to boot on safe-mode on the Asrock D1800M on both, usb2 & usb3 ports. speed was 40Mb/s on usb2 & 200Mb/s on usb3. normal boot produces a 7B BSOD that I am unable to further debug for the reasons explained above.

with the pcw.sys renamed I was able to normally boot on the A320M-DS2 on the usb3.1. speeds was 32Mb/s read & 112Mb/s write. usb3.0 boot was not possible due to a missing or bad driver I guess.

finally on the NUC6CAYH box, speeds were 65Mb/s read & 160Mb/s write

[acoul]

using Snappy Driver Installer Origin to update every single driver on the platforms I am testing this system, along with fine-tuning the start-up drivers with Autoruns, finally made the day.

All platforms are booting just fine with usb3. the A320M-DS2 still can't boot from usb3.0 and some times usb3.1 freezes, but I am suspecting a bug on the amd usb driver.

so I am closing this issue, since UsbBoot-Installer works just fine.

[acoul]

admxhc31 is quite unstable. I had to set BootFlags to 3 for both admxhc31 & admhub31 in HKEY_LOCAL_MACHINE/CurrentControlSet/services in order to get stable boots & rebooots

still, on DiskMark the write testing stage hangs all the time. Using a usb-2 cable on that same port, results on a stable system with the usb2 speeds. another odd thing is that when booting from usb3.1 with a usb2 cable, the speed on the usb3.0 with a usb3 cable also gets a performance hit that sometimes approaches usb2 speed.

admxhc30 is rock stable with speeds near 180 (the same disk score 250 on ahci)

the wonders of the M$ (and in this case AMD) universe