|
|
1.1 root 1: the m68k exception vector table lists the spaces used for the different vectors
2:
3: clean up posix-serial callouts to be like 3c400
4:
5: is the (brief) extension word handling correct for PC-relative
6: EAs when there are multiple brief extension words? (i.e., is
7: first_ea_extword_offset updated (does it need to be?))
8:
9: in type 8 frames, we currently store the address as it appeared
10: on the bus - i.e., only 24 bits, not the full 32. maybe this is
11: incorrect?
12:
13: should tme_m68k_rte_start have more RESTARTING checks?
14:
15: are the data input buffer and data output buffers being kept in the
16: right endianness in both the trap frames and in the ic buffers?
17:
18: watch out for (raw[0] << 4) possibly staying 8-bit and then overflowing
19: in tme_m68k_insn_buffer_xfer
20:
21: optimization: have the fast executor only execute an instruction if
22: there are TME_M68K_INSN_WORDS_MAX * sizeof(tme_uint16_t) bytes
23: available. note that this does *not* eliminate the need for the
24: group0 hook entirely, since while we guarantee that we won't fault on
25: an instruction fetch, we still might fault on data in
26: mid-instruction-fetch. this optimization would however remove a
27: test and goto in the fast FETCH macros.
28:
29: probably the way ENA_NONBOOT works on the sun2 is that all FC_SP
30: references go to the PROM
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.