11:05 Alligator40: It's Estonian networks fucked only their exit nodes, but those are the most shitty abusers world has ever seen, but tcp is not a complex protocol , in the header you have offsets and polarity of operation whether read or write, it's easy to exploit it, and other protocols are alike, I know what criminals in cambodia, their biggest mistake was to humiliate me, those bandits are really done. It's bit weirdo to attack the author with his code and hope
11:05 Alligator40: things turn out ok, I know how to burn them all to hell.
11:06 Alligator40: To jam tcp is easy, cause it's hw on colliding accesses is the cult of the issue.
11:46 Alligator40: Those are all Estonian scamretards who caused worlds biggest protest and shit show fraud in the world, they caused my life to halt all the times, total retards.
11:52 Alligator40: Not understandable to human beings as to what their activity is, it's some mental illness, they go around the world and do something that is too difficult to understand, it's dramatic.
11:56 Alligator40: I do not get much , but indeed it's like what the illness is, I am worlds antihero, and those are very sick stories I hear.
12:12 Alligator40: Overall I think the military knows well how to update himars software, it's what we call a supercomputer and it's algorithm is very effective what I can see, it needs low latency emitter, to do such frequency double triple stuff on radio waves, I know that tech I think, Americans know that too, they will succeed in new sw.
12:15 Alligator40: It's ghost signal in terms of gps, it's modulation is random, man in the middle attacks can not work
13:29 MrCooper: mareko pepp: on this laptop with Picasso APU, I'm getting bogus values (all bits except the 6 LSBs are always 0) from GL_TIMESTAMP/PIPE_QUERY_TIMESTAMP queries; any ideas?
14:07 agd5f: MrCooper, maybe this patch will fix it: https://lists.freedesktop.org/archives/amd-gfx/2023-May/093006.html
15:20 MrCooper: agd5f: hmm, the synchronous GPU timestamp values aren't affected though (at least they have more significant non-0 bits), only the asynchronous ones retrieved via queries in the command stream
15:24 Venemo: agd5f: here is a log of what I mean: flip_done timed out and a backtrace of a kernel crash: https://pastebin.com/raw/eF1bTAYP but there is no ring gfx timeout
16:58 agd5f: Venemo, hmm, might be related to PSR. Does amdgpu.dcdebugmask=0x10 help (disables PSR)?
17:10 Venemo: agd5f: I will try
18:05 Venemo: agd5f: the issue is somewhat "random" but I'll let you know my experience with that flag