summaryrefslogtreecommitdiff
path: root/src/systemc/tests/working.filt
blob: 49ab3aaab3b4f88a7b53dc5c56f9fbe5e47349b0 (plain)
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
not any([

# The sc_elab_and_sim mechanism is purposefully not supported.
name == "sc_elab_and_sim",

# The wif trace format is deprecated, and probably would take a non-trivial
# amount of work to support.
path.startswith("systemc/tracing/wif_trace"),
path.startswith("systemc/misc/stars/wif_trace"),

# Phase callbacks are an experimental feature. Also calling all the interested
# sc_objects whenever time advances would add a non-trivial amount of
# complexity and is probably not worth implementing in general.
path.startswith("systemc/kernel/phase_callbacks/"),


path in (
    # Uses sc_elab_and_sim.
    "systemc/kernel/sc_main_main",

    # Tests the deprecated sc_string type which we aren't supporting.
    "systemc/bugs/sc_string_bracket_assign",
    "systemc/misc/stars/star124010",

    # This test declares a constructor it never defines, so it's not clear how
    # it would ever work?
    "systemc/misc/stars/star104726",

    # Another test which calls an undefined function.
    "systemc/misc/gnats/pr-480",

    # Two more tests with undefined functions.
    "systemc/misc/unit/extern",
    "systemc/misc/unit/extern2",

    # These tests rely on implicit FXVAL constructors based on primitive types,
    # but enabling that makes other operator overloads ambiguous so things
    # don't compile. It also seems dangerous to change what the headers look
    # like between the backing implementation and the consuming application.
    "systemc/datatypes/fx/fast_constructors",
    "systemc/datatypes/fx/constructors",

    # These tests uses sensitive_neg and sensitive_pos and overloads which are
    # deprecated and not worth supporting. The non-deprecated version,
    # sensitive, is exercised elsewhere.
    "systemc/kernel/sc_sensitive/test02",
    "systemc/kernel/sc_sensitive/test03",

    # This test relies on being able to set the time resolution even after it's
    # been fixed by telling systemc to ignore the error message. gem5 has no
    # facility to ignore the error it raises for the same reason, and it
    # wouldn't make sense to add one.
    "systemc/kernel/sc_time/test19",

    # This test is for a mechanism to change to a non-standard signal write
    # checking behavior using an environment variable which we don't intent to
    # support at this time.
    "systemc/communication/sc_signal/check_writer/test16",
),

])