Conversation
05d36a7 to
5db0df5
Compare
DrDeano
left a comment
There was a problem hiding this comment.
Sorry for the brutal commenting
5db0df5 to
ed88c5e
Compare
DrDeano
left a comment
There was a problem hiding this comment.
I was wondering, is it possible to have a run time test for calling a fake syscall, could just have something hard coded for now, but then could extend this out to a proper runtime testing that can call test functions. So could pass a boolean into the build of the kernel which it can check
Yeah I could add a test that always runs when the kernel boots, and later move it to a proper runtime test when we figure out how to pass options to the build. |
I've actually found out how to pass options to the build so I will make a PR for that and add a syscall runtime test as part of #60 |
|
I actually don't think this is quite ready to be merged yet since I need to do some more investigation on how syscall arguments should be handled. |
|
Ok I did some experimentation and made the syscall triggering much better and able to pass arguments. Ready for a review! |
DrDeano
left a comment
There was a problem hiding this comment.
Also, wheres those runtime tests :P
Waiting for the build options to be merged :p I actually need to improve that patch a little. Will do so when I can. |
|
No problem, shall I'll wait until the build options PR is done |
Alright that one should be ready now. I added some replies here too |
c0a8813 to
9b9c82a
Compare
DrDeano
left a comment
There was a problem hiding this comment.
I keep finding things soz. Have a small change, I have a lot of comments lol
093a1c6 to
f7526b2
Compare
This patch adds syscalls for x86. They are triggered by putting the syscall number in
eaxthen trriggering interrupt128(The same as linux). It adds support for calling syscalls with a maximum of 5 arguments. Closes #36