TODO revision 1.43 
1/* $NetBSD: TODO,v 1.43 2021/06/25 19:13:12 palle Exp $ */
2
3Things to be done:
4
5common:
6- make %g6 point to curcpu
7- make %g7 point to curlwp
8- change run-time checks for cpu type to function pointers
9
10sun4u:
11- GENERIC.UP kernel hangs on v445 (missing interrupt?)
12
13sun4v:
14 - current status
15     T5 ldom with 2 VCPU and 4GB:
16       - kernel boots from miniroot.fs via ldom fisk (vdsk)
17	   - ldom virtual network interface (vnet) is working
18	     (verified by exiting sysinst and issuing a ping command)
19	   - the sysinst tool starts, disk setup is working,
20	     but the process crashes when selecting network installation method.
21		 The %i,%l and %o registers are corrupted, after a call to the libcurses
22		 wrefresh() function.
23		 If the sysinst insallation source is local directory, the sets are properly
24		 unpacked, but the system hangs afterwards when running the makedev script.
25	 T2000 ldom with 8 VCPU and 4GB:
26	   - crashes in /sbin/init doing an access() call where %o0 is corrupted (zero)
27	 S7 ldom with 8 VCPU and 16GB (primary ldom is Solaris 11.4 SRU30):
28	   - during autoconfiguration of vnet devices the call to hv_vintr_setenabled()
29	     returns error code 7 (EBADTRAP?)
30- 64-bit kernel support
31- 32-bit kernel support
32- libkvm
33- ofwboot: tlb_init_sun4v() hardcodes number of slots to 64
34- locore.s: sun4v_datatrap missing implementation for trap level 1
35- check build without SUN4V defined
36- replace relevant references to %ver with GET_MAXCWP
37- pmap_mp_init(): sun4v missing handling
38- replace constructs like "wrpr %g0, PSTATE_KERN, %pstate" with NORMAL_GLOBALS
39- replace constructs line "wrpr %g0, PSTATE_INTR, %pstate" with ALTERNATE_GOBALS
40- sun4v tsb no need to lock... per cpu... anyway...
41- ci_tsb_desc->td_ctxidx: -1 or 1?
42- MP support - currently bypassed in pmap_bootstrap() for sun4v
43- vpci.c/vpcivar.h: cleanup FIXMEs
44- interrupts not handled properly (com at ebus only...)
45- mpt(4) complains: mpt0: Phy 0: Link Status Unknown
46- man pages for drivers imported from OpenBSD lke vpci, vbus, cbus, vdsk, ldc etc.
47- vdsk and ldc drivers: code maked with OPENBSD_BUSDMA - make the bus_dma stuff work properly
48- vbus.c: handle prom_getprop() memory leaks
49- locore.s: rft_user (sun4v specific manaul fill) - seems to work, but is it good enough (compared to openbsds rft_user?
50- openfirmware interface
51    bounce handlng could need rework
52      - only #if __arch64__ independent for SUN4V, as OF for all ultrasparcs is
53        plain 32bit code too (and this #if only because the point below becomes
54        always false otherwise)
55      - only bounce if the address is >= 1UL<<32
56      - perhaps apply the lock to OF calls in general, not only the bounce'd ones
57        (in other words: let's protect the OF context with the lock, not
58        only the bounce buffer contents).
59	  - Are there any other bounce-worthy OF calls? If so, apply handling 
60	    to all of them now
61      - consider method used by Solaris/OpenSolaris/Illumos (look for PROM_32BIT_ADDRS in
62	    the Illumos codebase)
63 - platforms tested so far:
64 -- qemu sun4v/niagara - reaches the point where the kernel needs to mount a root fs
65 -- T2000 - mmu fault somewhere
66 -- T5 - stickcmpr is not properly initialized ???
67