kern

log

age author description
Sat, 15 Oct 2011 08:07:09 +0300 John Tsiombikas testing of the exit syscall
Sat, 15 Oct 2011 08:06:58 +0300 John Tsiombikas fixed cleanup_vm to avoid cleaning up kernel pages
Sat, 15 Oct 2011 08:06:10 +0300 John Tsiombikas fixed a potential null dereference when deleting a bug in the redblack tree
Sat, 15 Oct 2011 07:45:56 +0300 John Tsiombikas - implemented syscalls: exit, waitpid, getppid
Thu, 13 Oct 2011 05:22:35 +0300 John Tsiombikas changed test_proc to modify memory after the fork in order to try out copy-on-write,
Wed, 12 Oct 2011 14:46:00 +0300 John Tsiombikas added a comment in clone_vm
Wed, 12 Oct 2011 14:39:40 +0300 John Tsiombikas ok I *think* i've fixed it now
Mon, 10 Oct 2011 04:16:01 +0300 John Tsiombikas - added a generic red-black tree data structure
Sun, 09 Oct 2011 20:38:35 +0300 John Tsiombikas added strcmp
Thu, 06 Oct 2011 04:28:21 +0300 John Tsiombikas ops, forgot the -m in -m elf_i386 ld arg
Wed, 05 Oct 2011 02:34:58 +0300 John Tsiombikas changed the makefile to only pass "emulation" (-m32 -melf_i386) only when not
Wed, 17 Aug 2011 05:43:47 +0300 John Tsiombikas forgot to handle the cow argument to clone_vm
Wed, 17 Aug 2011 05:42:44 +0300 John Tsiombikas added errno.h in klibc
Wed, 17 Aug 2011 05:39:39 +0300 John Tsiombikas added a comment in the tss showing the interesting bits
Wed, 17 Aug 2011 05:38:40 +0300 John Tsiombikas removed the unimplemented exit syscall
Wed, 17 Aug 2011 05:37:30 +0300 John Tsiombikas fixed the comments in test_proc.S
Wed, 17 Aug 2011 05:33:03 +0300 John Tsiombikas added a comment describing the push of just_forked address to the stack frame of the newly-forked process.
Wed, 17 Aug 2011 05:30:06 +0300 John Tsiombikas added a comment to just_forked
Tue, 16 Aug 2011 03:26:53 +0300 John Tsiombikas I think we're done for now. two processes seem to be scheduled and switched just fine, fork seems to work (NO CoW YET!)
Mon, 15 Aug 2011 06:17:58 +0300 John Tsiombikas finally, runqueues, blocking, waking up, idle loop etc, all seem to work fine
Mon, 15 Aug 2011 04:03:39 +0300 John Tsiombikas Fixed most important task switching bugs.
Sun, 14 Aug 2011 16:57:23 +0300 John Tsiombikas bringing the task switching thing into shape with proper per-process kernel stacks and shit
Mon, 08 Aug 2011 09:53:10 +0300 John Tsiombikas llalallalala
Sun, 07 Aug 2011 06:42:00 +0300 John Tsiombikas picking this up again, let's fix it
Mon, 01 Aug 2011 06:45:29 +0300 John Tsiombikas lalalala
Sat, 30 Jul 2011 07:35:53 +0300 John Tsiombikas tidyed up the intr_ret bit, made it a bit more reasonably structured, and cleaned up some debugging things
Sat, 30 Jul 2011 07:21:54 +0300 John Tsiombikas fuck yeah, now do priviledge levels and TSS
Thu, 28 Jul 2011 17:56:05 +0300 John Tsiombikas fixed the page fault
Thu, 28 Jul 2011 05:43:04 +0300 John Tsiombikas continuing with the process implementation. not done yet, panics.
Thu, 28 Jul 2011 05:33:59 +0300 John Tsiombikas apparently free_phys_page was never tested. the check for double-freeing a page
Tue, 26 Jul 2011 02:41:33 +0300 John Tsiombikas mainly additions to the VM to support processes etc. not complete
Mon, 25 Jul 2011 11:29:57 +0300 John Tsiombikas added the first test process code and system call numbers header
Mon, 25 Jul 2011 11:29:02 +0300 John Tsiombikas - implemented clone_vmem
Sun, 24 Jul 2011 18:29:24 +0300 John Tsiombikas started implementing processes
Tue, 14 Jun 2011 01:19:07 +0300 John Tsiombikas merged the timer/rtc/etc from the misc branch
Sun, 12 Jun 2011 15:45:21 +0300 John Tsiombikas - now read_rtc doesn't try to read weekday as it's probably wrong anyway, and doesn't set yearday either
Sun, 12 Jun 2011 01:46:53 +0300 John Tsiombikas - removed redundant call to day_of_year in gmtime_r
Fri, 10 Jun 2011 23:19:25 +0300 John Tsiombikas - added makefile pattern to produce assembly output from c source files.
Fri, 10 Jun 2011 19:54:47 +0300 John Tsiombikas finished with the klibc/time.c stuff
Fri, 10 Jun 2011 05:33:38 +0300 John Tsiombikas - added cmos rtc code
Thu, 09 Jun 2011 05:09:49 +0300 John Tsiombikas - made interrupt number - irq mapping macros public in intr.h
Thu, 09 Jun 2011 02:45:49 +0300 John Tsiombikas forgot to include timer.h in timer.c
Wed, 08 Jun 2011 03:02:42 +0300 John Tsiombikas 8254 timer code
Sat, 28 May 2011 08:06:47 +0300 John Tsiombikas merged the task bits, not completed
Fri, 27 May 2011 13:08:23 +0300 John Tsiombikas ops ... range->size = pgcount * PGSIZE not pg * PGSIZE
Fri, 27 May 2011 22:09:10 +0300 John Tsiombikas refactored the code a bit, and merged with the malloc implementation branch
Sat, 23 Apr 2011 00:51:18 +0300 John Tsiombikas starting on the task switching implementation
Fri, 27 May 2011 12:05:38 +0300 John Tsiombikas implemented malloc
Sat, 09 Apr 2011 07:22:27 +0300 John Tsiombikas moved enable_intr out of init_intr and after the kernel initialization is completed
Sat, 09 Apr 2011 07:14:06 +0300 John Tsiombikas fixes here and there
Wed, 06 Apr 2011 07:42:44 +0300 John Tsiombikas protected critical sections in VM and the physical memory manager by disabling interrupts
Tue, 05 Apr 2011 02:09:02 +0300 John Tsiombikas switched the vm to use recursive page tables
Mon, 04 Apr 2011 23:34:06 +0300 John Tsiombikas - fixed tragic mistake in the initial kernel image mapping
Sun, 03 Apr 2011 18:42:19 +0300 John Tsiombikas writing the vm
Sun, 03 Apr 2011 08:23:07 +0300 John Tsiombikas now ignoring parts of the memory map beyond 4gb instead of silently wrapping around and marking all holes as usable :)
Wed, 30 Mar 2011 23:14:29 +0300 John Tsiombikas added a few comments in mem.c
Wed, 30 Mar 2011 22:42:16 +0300 John Tsiombikas I think I'm done with the physical memory page allocator
Sun, 27 Mar 2011 06:28:26 +0300 John Tsiombikas fixed the stupid mistake. forgot to page-align the page directory
Sat, 26 Mar 2011 21:39:14 +0200 John Tsiombikas forgot to add a shitload of files
Sat, 26 Mar 2011 12:27:57 +0200 John Tsiombikas adding paging