06:11:26 <Inge-> "i dont know if this is a break through, a big deal or a mining cherry pop but I got xmrig to push 560 H/s on my Note9"
06:11:42 <Inge-> hyc ^ thought you might be amused
10:58:59 <sech1> https://www.techpowerup.com/266650/tachyum-prodigy-is-a-small-128-core-processor-with-crazy-i-o-options-64-core-sibling-enroute-production
10:59:10 <sech1> 128 core 16 TFLOPS CPU?
10:59:35 <sech1> That's more TFLOPS than RTX 2080 Ti
11:17:13 <Inge-> "For starters, the chip features two 400 Gbps PHYs integrated, along with a PCI-Express gen 5.0"
11:17:51 <Inge-> fun stuff
13:57:04 <p3rL> lulz @ https://i.imgur.com/tLTV1Wi.mp4
14:54:09 <hyc> PCIe gen5 standard is final already?
14:54:32 <hyc> Inge-: nice
15:00:44 <hyc> hm, tachyum - VLIW in-order processor, relies on compiler for OoOE
15:00:50 <hyc> sounds like Itanium
15:01:16 <hyc> not good for a multithreaded OS, but probably good for HPC
15:01:43 <hyc> with those PHYs, possibly good for a poll-driven network controller
15:01:55 <hyc> not good for an interrupt-driven one
15:58:17 <gingeropolous> total miners 38310
18:29:07 <Inge-> friend of mine had to clean out a linux print server running xmrig malware
18:29:25 <Inge-> asked him for which xmrig version. will report back..
19:27:52 <sech1> moneromooo Is incoming block PoW verification done on the main thread? Because it changes the rounding mode and can mess up other code which uses floating point
19:28:31 <sech1> Loki had a big network fuckup today supposedly because they use FP in their diff adjustment code
19:29:03 <sech1> Monero doesn't use FP there btw which is good
19:29:35 <moneromooo> IIRC, main thread if just one, in sync mode it'll be pushed to N threads.
19:29:53 <sech1> then keep in mind that rounding mode is... random then
19:30:03 <sech1> maybe we should fix it in librandomx
19:30:28 <sech1> tevador ^
19:43:17 <hyc> or just never use FP anywhere else. which we don't
20:08:59 <moneromooo> IIRC someone posted a diff adjustment patch for monero which used floating point. Maybe someone saw the patch and applied without seeing it.
21:15:57 <hyc> I assume the rounding mode is per-core? so saving it on entry to library and restoring it on return from library is sufficient?
21:16:32 <moneromooo> Is it saved/restored on context switch ?
21:16:51 <hyc> part of FPU state, it would have to be
21:16:53 <moneromooo> Must be. It'd be dangerous otherise.
22:37:07 <tevador> in the slow mode, it would be possible to reset the rounding mode after each hash without any performance impact