I have had USB thumb drives that just wouldn’t boot on some computers. Try a different USB drive maybe.

https://lwn.net/Articles/335415/

The evince PDF reader ran into this issue back in 2005. It is now rare to find a distributor shipping a version of evince which implements copy restrictions. Xpdf implements copy restrictions unconditionally, but Debian patched that code out in 2002, and that patch has spread to other distributors as well. In general, as one would expect, free PDF readers tend not to implement this behavior. Okular is about the only exception that your editor can find; it’s interesting to note that the version of Okular shipped with Fedora Rawhide also implements copy restrictions by default. Perhaps this behavior is result of the relative newness of this application; as it accumulates more users, the pressure for more user-friendly behavior is likely to grow.

Another downside of flatpak is that I don’t trust upstream devs to have my best interests at heart, but I trust Debian developers far more. I’ve seen upstream do some annoying or stupid shit and the Debian maintainers not budging.

I think it was poppler or evince that decided they were going to enforce the no-copy-and-paste bit you can set on pdfs. Debian patched it out. I’ve seen Mozilla decide they were going to enforce their trademarks. They carved out special exceptions for various distros but that still would have meant you would have to rename Firefox if you were to fork Debian. Debian had none of it. There were many dodgy copyright and licensing problems upstream devs gave no shit about. Debian not including these often eventually put pressure on them to fix this shit or for some replacement to get developed.

Oh yeah that’s where I was getting at, but I didn’t have time to write that out earlier. I agree that OP probably pulled out the usb stick before buffers were flushed. I imagine that direct I/O would mitigate this problem a lot because presumably whatever buffers still exist (there would some hardware buffers and I think Linux kernel I/O buffers) will be minimal compared to the potentially large amount of dirty pages one might accumulate using normal cached writes. So I imagine those buffers would be empty very shortly (less than one second maybe?) after dd finishes, whereas I’ve seen regular dd finish tens of seconds before my usb stick stopped blinking it’s LED. Still if you wait for that long the result will be the same.

I don’t think oflags=direct has any influence on the result. Apparently that’s about disabling the page cache in the kernel, which can avoid a situation in which the system slows down due to buildup yet-to-write pages.