 Sven/Sarah Barth
|
40f90b60eb
* DWARF line info: only use the cached module if the base address also is the same; this can happen if the initial exception occurred at a random address outside the module, but the next stack address can be found inside the correct module
|
il y a 10 mois |
 Rika Ichinose
|
2536529bc0
Reuse DWARF buffer on seeks.
|
il y a 1 an |
|
Rika Ichinose
|
8cfc87ffdf
Sink lnfodwrf.pp global variables down to their actual scopes.
|
il y a 1 an |
 Michaël Van Canneyt
|
ab3b7f8f69
* Do not use namespace prefix for lnfodwarf
|
il y a 1 an |
 Michael VAN CANNEYT
|
ccfa38c68e
* Dotted RTL compiles
|
il y a 2 ans |
|
Michael VAN CANNEYT
|
6084bb6d09
* Some forgotten string -> shortstring instances
|
il y a 2 ans |
|
Michael VAN CANNEYT
|
d2d3fe6bc3
* Char -> AnsiChar
|
il y a 2 ans |
 Jonas Maebe
|
f13b47289e
* handle case where the first instruction with line information is looked up
|
il y a 4 ans |
|
Jonas Maebe
|
0eb9dd3879
* Darwin support for printing line info for backtraces when using Dwarf,
|
il y a 4 ans |
 florian
|
45b3d8e0bf
* check for cache entry before even trying to open the file containing the debug info
|
il y a 4 ans |
|
florian
|
4dfcdeaef8
+ use a hash table to speed up line info retrieval
|
il y a 4 ans |
 nickysn
|
b18602485a
* fixed base segment calculation on i8086 (code moved after call to OpenDwarf,
|
il y a 7 ans |
|
nickysn
|
45bcbca71b
* fixed reading of DW_FORM_ref_addr on i8086
|
il y a 7 ans |
|
nickysn
|
2a5fbd4598
* continue reading the segment when walking the dwarf arange tuples on i8086
|
il y a 7 ans |
|
nickysn
|
d1cdd8be7c
* do not emit padding bytes after the dwarf arange header on i8086 for
|
il y a 7 ans |
|
nickysn
|
6fedf7e930
+ debug dump the address size from the header in ParseCompilationUnitForFunctionName
|
il y a 7 ans |
|
nickysn
|
1d3ad2b078
+ also take the segment portion of the address in lnfodwrf.ParseCompilationUnit
|
il y a 7 ans |
|
nickysn
|
4ce5db5188
+ also check the arange segment in ParseCompilationUnitForDebugInfoOffset
|
il y a 7 ans |
|
nickysn
|
81d5483ab1
* subtract the process load segment from the code segment on i8086
|
il y a 7 ans |
|
nickysn
|
875330d84f
+ pass a segment (0 on non-i8086) to all ParseCompilationUnit* functions in unit
|
il y a 7 ans |
|
nickysn
|
d7a3f755e4
+ also read arange segments on i8086
|
il y a 7 ans |
|
nickysn
|
3989e57fed
* take into account the address size in lnfodwrf.ReadAddress on i8086; support
|
il y a 7 ans |
|
nickysn
|
a7f5a69000
+ added an address size parameter to lnfodrwf.ReadAddress(); pass the expected
|
il y a 7 ans |
|
nickysn
|
dc8b78abe6
+ debug print the arange address and segment sizes
|
il y a 7 ans |
|
nickysn
|
1f76f52996
* use hexstr(addr) in the dwarf backtrace function, so it displays always a good
|
il y a 7 ans |
|
nickysn
|
96b53c168c
* use hexstr(baseaddr) to print out the pointer value, regardless of its x86 type (near/far, etc.)
|
il y a 7 ans |
|
nickysn
|
abe61d0daa
* the 'addr' and 'baseaddr' parameters of exeinfo.GetModuleByAddr changed to
|
il y a 7 ans |
|
nickysn
|
0ceb1244bf
* changed the type of the first parameter ('addr') of lnfodwrf.GetLineInfo to
|
il y a 7 ans |
|
nickysn
|
2411cb3f24
* fix save/restore of base/index/pos on i8086
|
il y a 7 ans |
|
nickysn
|
0ca05f042b
+ parse the openwatcom dwarf extension DW_LNE_SET_SEGMENT and also track segment
|
il y a 7 ans |
