12:39:44 hello 12:40:04 any news about compiling debug version for android 12:40:09 malinero? 12:48:22 Guest123422: it worked fine 12:48:26 you have to do clean build 12:49:38 one more details is that dockerfile create apk which i unzip copy monerod and run 12:49:58 maybe this is wrong? 12:50:16 but with version Release work without issues 12:51:19 he said it builds fine 12:51:33 I don't know if he tested the apk afterwards 12:51:46 can you share the create apk code? 12:52:35 https://github.com/monero-project/monero-gui/blob/master/Dockerfile.android 13:01:10 don't see any reason for it 13:02:19 DCMAKE_BUILD_TYPE=Release DCMAKE_BUILD_TYPE=Debug 13:02:23 the only change! 13:08:09 what do you need debug build for? 13:08:56 to c why cant mine on huawei smartphone that i have 13:09:26 if i mine either apk or monerod alone crash 13:10:25 i read about some fixes for arm64 but still crash 13:10:46 then build daemon with debug 13:10:50 and start the daemon separately 13:10:57 without GUI involved 13:11:14 im doing this 13:11:21 no gui involved 13:11:31 i strip daemon from apk 13:11:34 extract 13:11:48 that's not going to work if the shared libraries are inside the .apk 13:12:28 why run without libraries the release version 13:12:39 for release version i do the same 13:12:39 do you find a libdaemonizer.so inside the.apk? 13:12:51 got it i search right now 13:47:39 there is no libdaemonizer.so anywhere after unzippping apk 13:47:48 apk has no libdaemonizer.so 13:48:44 and there is not libdaemonizer.so on release version too 15:36:35 hello 16:52:50 hmm, do you know if randomx allocates >= 1MB on the stack (not the heap) ? 17:08:48 out of curiosity, can gdb on x86 reads a binary and a core dump for another arch? 17:09:32 s/for/from/ 17:52:24 hello 17:59:18 Hi 18:01:04 smth u told about randomx 18:01:30 "randomx allocates >= 1MB on the stack " 18:01:41 this is about gdb or monerod 18:02:20 https://github.com/monero-project/monero/blob/master/src/crypto/CMakeLists.txt#L108 18:04:08 if randomx allocates more than 1MB, it won't work when used through the GUI but should work when used in standalone. If it allocates more than 4 MB, it won't work with all android builds 18:04:43 (i think it should still work with rooted devices and arm builds) 18:05:46 interesting 18:06:30 how MB randomx take on laptop,pc 18:06:37 around 18:07:12 i forgot one important detail: allocates *on the stack*, not the heap 18:08:05 so it doesn't matter what the OS says 18:08:19 randomx code is too complex for me :) 18:08:40 on pc works without issues 18:08:46 solo mining works 18:09:29 yep, it's an android specific limitation 18:10:06 if we give more RAM to monerod? 18:10:44 the pb is not how much RAM in total 18:10:59 it's the "stack" size 18:11:18 google "heap VS stack" if you want to know more about it 18:11:39 arm registers are fewer than x86? 18:12:26 iiuc, it's an android limitation, not arm 18:13:19 i remember that before 1year or more mining was working on smartphones 18:13:29 smth changed 18:13:31 dont know 18:13:42 dont remember when exactly 18:13:52 introduction of randomx ? :) 18:14:03 maybe........ 18:14:46 https://github.com/tevador/RandomX/pull/202 18:14:54 this one was not fixed 18:15:00 definetly not 18:18:43 Are you sure you have included that commit ? 18:19:13 that commits are inserted on randomx code now 18:19:19 but do not work 18:23:26 they are inserted in randomx, but doesn't mean you have them locally 18:24:02 i have i checked 18:25:49 have you tried the arm build ? 18:26:59 make depends target=aarch64-linux-android 18:27:00 this? 18:28:55 nope, target=aarch64-linux-gnu 18:29:37 no but can this run on android? 18:29:54 the getmonero.org android version runs on android 18:30:08 yep 18:30:44 this will make linux binary target=aarch64-linux-gnu 18:30:51 how to run on android? 18:31:03 maybe im wrong 18:31:07 just asking 18:32:07 start with termux 18:32:10 I guess? 18:32:57 https://monero.stackexchange.com/questions/2702/is-it-possible-to-run-a-full-monerod-node-on-android 18:33:08 i think you'll need gnurrot 18:33:14 gnuroot 18:34:04 i use termux for running android binary but ok i test linux aarch64 binary too 18:46:38 i mine on mediatek p22t 8core and no crash 18:46:58 is aarch64 android 18:47:03 any suggestion? 18:53:41 maybe it has some low level CPU instructions that your phone doesn't phone. I would open an issue in randomx repo and give details about the hardware specs 18:57:56 jit?? 19:05:10 no idea (but i'd guess that jit is a high level feature that requires some specific ASM instructions to make it fast enough) 19:16:50 ArticMine: are you free to approve the matrix relay to this channel? 19:17:28 he did write yes last time, or does he have to do something more? 19:18:07 last time no one was opped I believe 19:18:22 now that someone is opped I can have the bridge DM them for approval, then it will be relayed 19:19:54 * malinero doesn't understand the existence of so many channels while #monero has become a faucet channel ... 19:20:28 #monero was supposed to be bridged, I don't know the status of it... 19:21:07 afaik bridging large channels with over 200 users or so require extra permission by matrix and it's stuck there 19:21:36 ah ok 19:21:37 actually 100, and yeah looks like that channel isn't bridged yet either 19:24:27 19:52 <@ArticMine> Yes 19:24:35 that was from a couple days ago (as OP) 19:25:09 He will have to say "yes" to a DM that Matrix sends him. I may have missed that message a few days ago 19:25:17 ok 20:14:08 out of curiosity, have "TheLounge" (https://thelounge.chat/) ever been considered as an alternative or an addition to matrix ?