summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2021-09-20add an offset parameter to read() and write()dzwdz
2021-09-20remove some unused variablesdzwdz
2021-09-20use a single struct for all fs_wait return valuesdzwdz
2021-09-20add some missing includes to `kernel/vfs/request.h`dzwdz
2021-09-19slightly tidy up some syscallsdzwdz
yay for pointless changes
2021-09-19show the malloc balance on halt; remove a bad kfree() calldzwdz
the kfree() call was freeing a buffer which was inside a process struct
2021-09-19remove the unused static_strcmp macrodzwdz
2021-09-19call sysenter_setup before kmain_earlydzwdz
makes the code slightly simpler, imo
2021-09-19mark run_init as _Noreturndzwdz
2021-09-18_syscall_mount: actually use the just computed `backend` vardzwdz
whoops
2021-09-18remove `struct process.stack_top` - it was unuseddzwdz
2021-09-18init: fix off-by-one error in tar_sizedzwdz
2021-09-18implement NULL mountsdzwdz
2021-09-18`init`: implement a working tar driverdzwdz
2021-09-18create `init/stdlib`dzwdz
2021-09-18put the `handle_t` typedef in `shared/types.h`dzwdz
2021-09-18merge `kernel/types.h` and `init/types.h`dzwdz
2021-09-18`init`: start work on the tar driverdzwdz
2021-09-16scaffolding for the initrd: init can read files appended to itdzwdz
currently this is just a text file, but it will be a tar archive later on
2021-09-16implement output from vfs callsdzwdz
2021-09-16fs_read stub, basic implementation in userlanddzwdz
2021-09-15fs_wait: pass the file ID toodzwdz
2021-09-14merge the `syscall_handler` and `_syscall` declarationsdzwdz
i think that making the call a bit more "transparent" makes the code nicer
2021-09-14add support for 4-parameter syscallsdzwdz
2021-09-12fix typo; use the right union elementdzwdz
the values are the same, it only caused a warning in sparse
2021-09-12simplify `fs_respond`'s signaturedzwdz
2021-09-12first working access function!dzwdz
2021-09-12reorganise init, againdzwdz
2021-09-12move `enum vs_operation` to shared/flags.hdzwdz
2021-09-12allow vfs_request_finish to returndzwdz
thanks to this, the fs provider can continue executing until the next fs_wait() call. that should speed things up a bit
2021-09-12implement part of `fs_respond`dzwdz
2021-09-12force init's main to be in .text.startupdzwdz
for some reason it's only there by default if you compile with `-O2`. because of this, you couldn't compile camellia with `-O1`
2021-09-12log cr2 on page faultsdzwdz
2021-09-12make the inline functions staticdzwdz
otherwise this doesn't compile with -O0
2021-09-12init: allocate bss at runtimedzwdz
2021-09-12barebones `memflag()` implementation - letting the user allocate pagesdzwdz
2021-09-12remove `shared/vfs.h`, it was unuseddzwdz
2021-09-12remove some useless code from initdzwdz
2021-09-12implement most of fs_waitdzwdz
awaited_req is a garbage name but i couldn't come up with a better one. i also have no idea how to handle all the failure states
2021-09-12vfs_request refactor pt2dzwdz
2021-09-12vfs request refactor pt1dzwdz
2021-09-12remove some dead codedzwdz
2021-09-12remove obsolete commentdzwdz
2021-09-11add some helper functions for copying between virtual and physical memorydzwdz
this was already possible, but now it's slightly safer
2021-09-11replace `user_ptr` with a linux-style `__user` annotationdzwdz
2021-09-11fix the `sparse` warningsdzwdz
2021-09-09basic _syscall_fs_wait() impl, doesn't pass the req yetdzwdz
2021-09-08style: simplify the `kmalloc(sizeof(...` callsdzwdz
2021-09-08save the vfs_op_request in the process structdzwdz
2021-09-08copy _syscall_fs_open's argument to a new buffer instead of a shared onedzwdz
this will be needed for storing vfs_op's properly