| 00:23.06 | *** join/#card10badge DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
| 01:42.07 | *** join/#card10badge ebi__ (~quassel@dynamic-077-002-036-049.77.2.pool.telefonica.de) |
| 01:43.07 | *** join/#card10badge mkie_ (~quassel@dynamic-077-002-036-049.77.2.pool.telefonica.de) |
| 02:26.48 | *** join/#card10badge Topy44 (XFkgaptW3Z@bellatrix.uberspace.de) |
| 03:41.14 | *** join/#card10badge jrt (~jrt@dynamic-089-012-124-137.89.12.pool.telefonica.de) |
| 05:08.19 | rahix[m] | I think the problem is that the compiler is technically allowed to assume that a pointer to type T is pointing to an address with valid alignment for T |
| 05:47.33 | *** join/#card10badge DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
| 06:00.17 | *** join/#card10badge jobroe (~manjaro-u@p579eb41c.dip0.t-ipconnect.de) |
| 07:51.26 | rahix[m] | Looks like it: "A pointer to an object type may be converted to a pointer to a different object type. If the resulting pointer is not correctly aligned for the referenced type, the behavior is undefined." |
| 07:52.04 | rahix[m] | http://port70.net/~nsz/c/c11/n1570.html#6.3.2.3p7 |
| 09:53.25 | *** join/#card10badge theAntimon (~matze@p5b313092.dip0.t-ipconnect.de) |
| 10:07.03 | *** join/#card10badge msvb-lab (~msvb-lab@monero/hardware/michael) |
| 10:52.28 | *** join/#card10badge theAntimon (~matze@p5b313092.dip0.t-ipconnect.de) |
| 11:34.20 | *** part/#card10badge theAntimon (~matze@p5b313092.dip0.t-ipconnect.de) |
| 12:17.52 | *** join/#card10badge mmu_man (~revol@vaf26-2-82-244-111-82.fbx.proxad.net) |
| 12:29.17 | *** join/#card10badge erlehmann (~erlehmann@dynamic-046-114-032-032.46.114.pool.telefonica.de) |
| 12:31.56 | *** join/#card10badge taziden (~taziden@flexiden.org) |
| 13:10.05 | *** join/#card10badge theAntimon (~matze@141.76.180.9) |
| 13:11.12 | *** join/#card10badge theAntimon1 (~matze@141.76.180.9) |
| 13:11.37 | *** part/#card10badge theAntimon (~matze@141.76.180.9) |
| 13:28.39 | *** join/#card10badge theAntimon (~matze@141.76.180.9) |
| 13:29.15 | *** part/#card10badge theAntimon (~matze@141.76.180.9) |
| 13:40.16 | *** join/#card10badge DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
| 14:40.15 | schneider | i've got exception printing to the display working: https://git.card10.badge.events.ccc.de/-/snippets/48 |
| 14:40.24 | schneider | it is a hack though |
| 14:41.08 | schneider | rahix[m]: would you merge something like that? i can lobby upstream to add a hook but until then MP does not provide a reasonable interface |
| 15:15.04 | *** join/#card10badge theAntimon (~matze@141.76.180.9) |
| 16:15.05 | *** join/#card10badge oida (~oida@gateway/tor-sasl/oida) |
| 16:15.07 | *** join/#card10badge theAntimon (~matze@141.76.180.9) |
| 16:16.07 | *** part/#card10badge theAntimon (~matze@141.76.180.9) |
| 16:18.44 | *** join/#card10badge msvb-lab (~msvb-lab@monero/hardware/michael) |
| 17:11.25 | *** join/#card10badge theAntimon (~matze@141.76.180.9) |
| 17:11.30 | *** part/#card10badge theAntimon (~matze@141.76.180.9) |
| 18:02.40 | *** join/#card10badge revbx0[m] (revbx0matr@gateway/shell/matrix.org/x-rsvrsgikcttecspd) |
| 19:29.42 | Rahix | heh, that really is a hack |
| 19:58.34 | pixtxa[m] | Already seen an "exceptions" on screen today. It said kernel panic and displayed a bomb or something. |
| 20:09.31 | *** join/#card10badge theAntimon (~matze@p5b313092.dip0.t-ipconnect.de) |
| 20:17.56 | *** part/#card10badge theAntimon (~matze@p5b313092.dip0.t-ipconnect.de) |
| 20:32.28 | pixtxa[m] | <schneider "Rahix: https://git.card10.badge."> Alpha-channel would be nice to put circles/letters/⦠to the display. 1 bit would be ok for me. |
| 21:04.10 | pixtxa[m] | <schneider "same for https://git.card10.badg"> Well, it's based on PixelFlut and only the needed pixels are drawn. Because of the offset, the symbols can directly be read from the filesystem and card10 don't have to manipulate them. And it's also possible to add strings onto each other to draw multiple symbols. I Kind of like it. |
| 21:08.43 | schneider | pixtxa[m]: the bomb is a failed assertion in epicardium. my snippet above deals with python exceptions |
| 21:09.14 | schneider | pixtxa[m]: a bug report would be appreciated. the serial output should also contain the information |
| 21:10.23 | schneider | pixtxa[m]: alpha is not supported at all yet as the underlying "gfx" layer does not allow that yet |
| 21:10.31 | schneider | i think it will come "soon" |
| 21:16.15 | pixtxa[m] | Well, 1 bit alpha is simply "draw pixel" or "don't draw pixel", that should be no problem. |
| 21:24.29 | pixtxa[m] | If I can give only a rectangular matrix to the api, it would be nice to have cutouts, where the display just keeps the old state. |
| 21:50.20 | pixtxa[m] | The serial output was red, then the connection closed. I use the linux screen command, haven't found a nicer way jet. |
| 21:50.31 | pixtxa[m] | * The serial output was red, then the connection closed. I use the linux screen command, haven't found a nicer way yet. |
| 22:20.33 | schneider | hm, the usual problem |
| 22:20.53 | schneider | (screen terminating because the device goes away) |
| 22:21.17 | schneider | Rahix: maybe we want to put a loop which waits for a button press after the assert |
| 22:21.32 | schneider | a mean panic() |