Kevin O'Connor | 2ad3744 | 2008-05-06 19:49:01 -0400 | [diff] [blame] | 1 | Split rombios32.c up into multiple files. |
| 2 | |
Kevin O'Connor | 22dfb85 | 2008-06-07 15:24:19 -0400 | [diff] [blame] | 3 | Support parsing of linuxbios/coreboot table. |
| 4 | |
| 5 | Do a pci scan for ide controllers - don't just assume ISA ports are |
| 6 | available. |
| 7 | |
Kevin O'Connor | c781293 | 2008-06-08 23:08:12 -0400 | [diff] [blame] | 8 | The keyboard driver should detect ctrl+alt+del and reboot the |
| 9 | machine. |
| 10 | |
Kevin O'Connor | 2ad3744 | 2008-05-06 19:49:01 -0400 | [diff] [blame] | 11 | Review changes committed to coreboot, virtualbox, qemu, kvm, and bochs |
| 12 | cvs tip. |
Kevin O'Connor | 22dfb85 | 2008-06-07 15:24:19 -0400 | [diff] [blame] | 13 | * bochs cvs (1.209) -- all changes synched |
| 14 | * coreboot (r3348): 1.163 -- no noteworthy enhancements |
| 15 | * qemu (r4675): 1.207 -- supports >4Gig memory |
Kevin O'Connor | c781293 | 2008-06-08 23:08:12 -0400 | [diff] [blame] | 16 | * kvm (45a442c): 1.182 -- >4gig, e820 vmx pages, enable cache?, |
| 17 | wrmsr in smp_start and rombios32, acpi insrcovr table |
Kevin O'Connor | 22dfb85 | 2008-06-07 15:24:19 -0400 | [diff] [blame] | 18 | * virtualbox (r9404): 1.176 -- f11/f12 kbd, lots of mouse changes, |
| 19 | logo, set text mode?, int 1589, floppy data rate?, |
| 20 | dummy_isr_function, int19 calls post |
Kevin O'Connor | 2ad3744 | 2008-05-06 19:49:01 -0400 | [diff] [blame] | 21 | |
Kevin O'Connor | 5c4a8c6 | 2008-05-12 23:50:16 -0400 | [diff] [blame] | 22 | Look into ways to reduce stack usage. Alter ISRs so that they do not |
| 23 | enable irqs. Only call out to other bios functions after minimizing |
| 24 | stack usage. |
Kevin O'Connor | 2ad3744 | 2008-05-06 19:49:01 -0400 | [diff] [blame] | 25 | |
Kevin O'Connor | 7a558e4 | 2008-03-11 20:38:33 -0400 | [diff] [blame] | 26 | Audit all sti/cli calls. Audit all call16 calls to make sure flags is |
| 27 | setup properly with respect to irqs. |
Kevin O'Connor | 44c631d | 2008-03-02 11:24:36 -0500 | [diff] [blame] | 28 | |
Kevin O'Connor | ed12849 | 2008-03-11 11:14:59 -0400 | [diff] [blame] | 29 | Audit statements where a 32bit intermediary changes meaning of a 16bit |
| 30 | comparison. |
| 31 | |
Kevin O'Connor | 63dbcfb | 2008-03-01 22:17:07 -0500 | [diff] [blame] | 32 | Code assumes ebda segment is static - it doesn't read 0x40e. |
| 33 | |
Kevin O'Connor | b8aacb0 | 2008-03-01 14:56:07 -0500 | [diff] [blame] | 34 | The __call16 code does a long jump to the interrupt trampolines - this |
| 35 | is unnecessary. |
Kevin O'Connor | f076a3e | 2008-02-25 22:25:15 -0500 | [diff] [blame] | 36 | |
| 37 | Fix makefiles so that they rebuild the required files automatically. |
| 38 | |
Kevin O'Connor | f076a3e | 2008-02-25 22:25:15 -0500 | [diff] [blame] | 39 | Cleanup setting of ES on GET/SET_BDA |
| 40 | |
Kevin O'Connor | 2ad3744 | 2008-05-06 19:49:01 -0400 | [diff] [blame] | 41 | Possibly implement 32bit pcibios support. |
Kevin O'Connor | f076a3e | 2008-02-25 22:25:15 -0500 | [diff] [blame] | 42 | |
| 43 | Allow one to select adding 32 bit code to 0xf000 or in a separate |
| 44 | location. |
| 45 | |
Kevin O'Connor | 22dfb85 | 2008-06-07 15:24:19 -0400 | [diff] [blame] | 46 | Look at integrating the lgpl vgabios into tree. |
| 47 | |
Kevin O'Connor | f076a3e | 2008-02-25 22:25:15 -0500 | [diff] [blame] | 48 | Try generating bios tables at compile time. |