the_sea_peoples has quit [Ping timeout: 480 seconds]
the_sea_peoples has joined #haiku
<phschafft>
morning.
<Begasus[m]>
noon
<phschafft>
all good?
<Begasus[m]>
yeah, can't complain, you?
<phschafft>
oh, all good.
<phschafft>
reading a bit of gordonjcp's code this fine morning.
<Begasus[m]>
interesting things for your project?
<Begasus[m]>
update script as good as ready, want to finetune it a bit so it matches upstream categories (for easier tracking)
<phschafft>
Begasus[m]: I hope so!
<phschafft>
and I must say maybe a prime case where the virtual handle concept will really shine.
<Begasus[m]>
+1 :)
<phschafft>
I was very skeptical about the virtual handles at first, but it really performs more than well. I wonder why I have never seen it before. Hope I didn't actually invent some fundamental new thing ;)
<Begasus[m]>
heh
<Begasus[m]>
any way progress :)
<phschafft>
:)
dodo75 has joined #haiku
dodo75 has quit []
dodo75 has joined #haiku
Chai-T-Rex has joined #haiku
ChaiTRex has quit [Ping timeout: 480 seconds]
zdykstra has quit [Ping timeout: 480 seconds]
andreaa has joined #haiku
zdykstra has joined #haiku
<andreaa>
I apologize for all the challenges today and in the past... I never challenge, in case I withdraw all the challenges and have mercy on me... peace and good
<erysdren>
i hope you find peace
<andreaa>
thanks
mmu_man has joined #haiku
andreaa has left #haiku [Konversation terminated!]
<phschafft>
erysdren: I was wondering about two additional header parameters for vmv0. one that tells the end of the image in terms of what is required to load to run the image. so there can be e.g. more metadata after that mark, but nothing that is required to run it.
<phschafft>
and the second is a mark to tell the location after which no executable code is found.
<phschafft>
there might be any kind of data section after that. just no code.
<phschafft>
the first is to provide an optional way to optimise loading as the system can figure out that some part is doesn't need to be loaded nor parsed. and the second is a type of simple protection against executing data.
<phschafft>
the section markers may also provide this info, but they require more parsing and stuff.
<phschafft>
so the idea would be that those markers serve as an easy, simple, and early absolute maximum value. and that e.g. the section markers may later provide more detailed data IF the implementation wants to parse them.
AndrevS has joined #haiku
HaikuUser has joined #haiku
HaikuUser has quit []
<phschafft>
gordonjcp: what is your status generally speaking today?
<gordonjcp>
phschafft: hi
<gordonjcp>
phschafft: somewhat about, pretty tired having been up since 6 dropping my sister-in-law off at the airport :-)
<phschafft>
ok, thanks for the info.
<gordonjcp>
phschafft: what did you want to ask?
<gordonjcp>
phschafft: I took a look at my PDP8 stuff, and suspect that the df32 bits might not actually work at all
<phschafft>
generally finding out more about file formats for storage devices.
<gordonjcp>
ah
<phschafft>
figuring out what is original, what is 'just' formats invented by emulators, how to parse/work with that etc.
<gordonjcp>
I'm pretty sure I treat them throughout as just raw images
<phschafft>
I think your code does that, but I would also compare your code to that of simh.
<phschafft>
we'll have dinner in a few minutes. and there is more for me to figure out. so if you don't feel good today, surely we can push that to tomorrow.
<phschafft>
but at least you have some basic context now. :)
<gordonjcp>
I don't understand simh, I'm nowhere near clever enough
<gordonjcp>
but sure, feel free to ask any questions even if I'm not about, fire them into a pm if you want to save busying up the channel
zardshard has left #haiku [Error from remote client]
zardshard has joined #haiku
<phschafft>
I feel like combining the information from all the places is the best way for me. and simh just works^TM over here. so surely it is another valid reference. even if only to see if it does the same as I expect. ;)
<scantysnax>
i'm on 32-bit haiku trying to build a 32-bit executable with all otherwise portable code.
<scantysnax>
everyhting compiles fine, but can't link
<phschafft>
that looks like the library search paths are wrong and/or something important is missing.
rakka has quit [Remote host closed the connection]
<phschafft>
crti.o is part of the compiler.
<phschafft>
if it can't find that something is strangely wrong.
<phschafft>
hence my question.
rakka has joined #haiku
<scantysnax>
i agree with you.
<scantysnax>
i'm on a nightly i think
<scantysnax>
and i have zlib installed, so i dont' know why it's not finding it.
<phschafft>
my second guess with be that something overrides the paths somewhere in the build process (not just adds some, but unsets the default ones as well).
mmu_man has joined #haiku
<jmairboeck>
scantysnax: it looks like it is using the wrong version of the linker. Did you use "setarch x86"?
epok128 has joined #haiku
<scantysnax>
yes
<jmairboeck>
ah, I read the path wrong, this is the modern version of ld
epok128 has quit []
<jmairboeck>
maybe gcc_x86_syslibs_devel is missing?
<scantysnax>
i dont see that in haikudepot
<jmairboeck>
about the missing zlib: you need zlib_x86_devel (or devel:libz_x86, which is the same)
<jmairboeck>
I think there is a menu item to show or hide devel packages
<scantysnax>
i have zlib_devel installed
mmu_man is now known as Guest23052
mmu_man has joined #haiku
<jmairboeck>
for linking with gcc 13, you need the _x86 version
<scantysnax>
okay, heading home. will work on this more tomorrow
scantysnax has quit [Quit: Vision[]: i've been blurred!]
mmu_man has quit [Ping timeout: 480 seconds]
mmu_man has joined #haiku
HaikuUser has joined #haiku
HaikuUser has quit []
<Anarchos>
how can i change the Change-ID of a PR in gerrit ? It seems i can't modifiy the change-id by editing the commit msg.
<Habbie>
i thought that would just work
<Habbie>
well, because it would make it a separate unrelated change
<Habbie>
why do you want to change it?
<Anarchos>
Habbie because three commits belong to the same change, but i sent them without change-id so gerrit did 3 pull requests with three different change-id, one for each commit !
<Habbie>
right. I think: squash, give new id, push, close the other 3?
<Anarchos>
Habbie what is 'squash' ?
<Habbie>
to combine the 3 commits into 1 commit
<Anarchos>
ok
jmairboeck has quit [Quit: Konversation terminated!]
<Anarchos>
Habbie thanks it worked !
<Habbie>
yay!
Anarchos has quit [Quit: sleep]
Aedil has quit [Quit: GN8!]
vezhlys has joined #haiku
vezhlys has quit [Remote host closed the connection]
vezhlys has joined #haiku
AndrevS has quit [Quit: umount /dev/irc]
B2IA has quit [Quit: Vision[]: i've been blurred!]