From a630d7fb9666bbacdd714ddc12d3c42ef1466ba7 Mon Sep 17 00:00:00 2001 From: Kartik Agaram Date: Wed, 13 Feb 2019 00:01:14 -0800 Subject: 4959 It's always seemed ugly to explain the rules for segment names. Let's just always require a fixed name for the code and data segments. --- subx/034compute_segment_address.cc | 39 ++++++++++++++++++++++++++++---------- subx/Readme.md | 3 +-- 2 files changed, 30 insertions(+), 12 deletions(-) (limited to 'subx') diff --git a/subx/034compute_segment_address.cc b/subx/034compute_segment_address.cc index b6c191e4..d37060a4 100644 --- a/subx/034compute_segment_address.cc +++ b/subx/034compute_segment_address.cc @@ -34,12 +34,15 @@ if (!starts_with(segment_title, "0x")) { Currently_parsing_named_segment = true; if (!contains_key(Segment_index, segment_title)) { trace(99, "parse") << "new segment '" << segment_title << "'" << end(); - if (segment_title == "code") - put(Segment_index, segment_title, 0); - else if (segment_title == "data") - put(Segment_index, segment_title, 1); - else - put(Segment_index, segment_title, max(2, SIZE(out.segments))); + if (out.segments.empty() && segment_title != "code") { + raise << "first segment must be 'code' but is '" << segment_title << "'\n" << end(); + return; + } + if (SIZE(out.segments) == 1 && segment_title != "data") { + raise << "second segment must be 'data' but is '" << segment_title << "'\n" << end(); + return; + } + put(Segment_index, segment_title, SIZE(out.segments)); out.segments.push_back(segment()); } else { @@ -50,10 +53,7 @@ if (!starts_with(segment_title, "0x")) { :(before "End flush(p, lines) Special-cases") if (Currently_parsing_named_segment) { - if (p.segments.empty() || Currently_parsing_segment_index < 0) { - raise << "input does not start with a '==' section header\n" << end(); - return; - } + assert(!p.segments.empty()); trace(99, "parse") << "flushing to segment" << end(); vector& curr_segment_data = p.segments.at(Currently_parsing_segment_index).lines; curr_segment_data.insert(curr_segment_data.begin(), lines.begin(), lines.end()); @@ -81,6 +81,25 @@ if (Currently_parsing_named_segment) { +load: 0x0900005c -> 0c +load: 0x0900005d -> 0d +:(scenario error_on_missing_segment_header) +% Hide_errors = true; +05/add-to-EAX 0/imm32 ++error: input does not start with a '==' section header + +:(scenario error_on_first_segment_not_code) +% Hide_errors = true; +== data +05 00 00 00 00 ++error: first segment must be 'code' but is 'data' + +:(scenario error_on_second_segment_not_data) +% Hide_errors = true; +== code +05/add-to-EAX 0/imm32 +== bss +05 00 00 00 00 ++error: second segment must be 'data' but is 'bss' + //: compute segment address :(before "End Level-2 Transforms") diff --git a/subx/Readme.md b/subx/Readme.md index de7a0fa7..b03176f0 100644 --- a/subx/Readme.md +++ b/subx/Readme.md @@ -274,8 +274,7 @@ SubX programs map to the same ELF binaries that a conventional Linux system uses. Linux ELF binaries consist of a series of _segments_. In particular, they distinguish between code and data. Correspondingly, SubX programs consist of a series of segments, each starting with a header line: `==` followed by a name. -The first segment is assumed to be for code, and the second for data. By -convention, I name them `code` and `data`. +The first segment must be named `code`; the second must be named `data`. Execution always begins at the start of the `code` segment. -- cgit 1.4.1-2-gfad0 mit.php?id=5b5afb12176437e977f210ee0bd1ed188f390611'>^ 
f91f1db ^

ae8cc2f ^




f37aa32 ^





ae8cc2f ^




f37aa32 ^


ae8cc2f ^

f37aa32 ^

ae8cc2f ^





ae8cc2f ^




1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116

117
118

119

120
121
122
123

124
125
126
127
128

129
130
131
132

133
134

135

136

137
138
139
140
141

142
143
144
145