mirror of
https://github.com/Zenithsiz/ftmemsim-valgrind.git
synced 2026-02-03 10:05:29 +00:00
55c12b3a18
-ffunction-sections -fdata-sections and the linker option
-Wl,--gc-sections, --read-var-info=yes gives the following:
valgrind: m_debuginfo/d3basics.c:973 (vgModuleLocal_evaluate_GX): Assertion 'aMax == ~(Addr)0' failed.
host stacktrace:
==18521== at 0x38057C54: show_sched_status_wrk (m_libcassert.c:308)
==18521== by 0x38057F50: report_and_quit (m_libcassert.c:367)
==18521== by 0x38058151: vgPlain_assert_fail (m_libcassert.c:432)
==18521== by 0x3813F084: vgModuleLocal_evaluate_GX (d3basics.c:973)
==18521== by 0x38098300: data_address_is_in_var (debuginfo.c:2769)
==18521== by 0x38099E26: vgPlain_get_data_description (debuginfo.c:3298)
...
The problem is that -Wl,--gc-sections eliminates the unused functions
but keeps some debug info for the functions or their compilation units.
The dwarf entry has low and high pc, but both are equal to 0.
The dwarf reader of Valgrind is confused by this, as the varstack becomes
empty, while it should not. This then causes local (eliminated) variables
to be put in the global scope, leading afterwards to evaluation errors
when describing any other variables.
The fix is to also push something on the varstack when
a CU that has low and high pc given but with 0 value.
This is similar to the varstack_push done for a CU that has
no low pc, no high pc and no range.
Despite considerable effort to make a small reproducer, the problem
could only be produced with a big executable.
After the fix, everything was working properly.
The wrong behaviour for dwarf entries produce the following trace:
<2><2ff291a>: Abbrev Number: 23 (DW_TAG_formal_parameter)
DW_AT_name : AET
DW_AT_decl_file : 1
DW_AT_decl_line : 243
DW_AT_type : <2ff2811>
DW_AT_location : 18288554
Recording this variable, with 1 PC range(s)
....
<2ff291a> addVar: level 0: AET :: EdgeTableEntry*
Loc=GX(final){[0x0,0x8]=50,[0x9,0x1d]=53,[0x1e,0x26]=51,[0x27,0x29]=53,[0x2a,0x2f]=51,[0x44,0x4a]=53,[0x4d,0x5e]=51,[0x5f,0x62]=53}
FrB=none
declared at: gdkpolyreg-generic.c:243
ACQUIRE for range(s) [0x0,0xffffffff]
The AET is a formal parameter of a function, but is wrongly added
at level 0, with a PC range covering the full space. It has a Loc GX
which uses non biased program counters (e.g. 0x0,0x8).
This dwarf entry will require a FrB (and registers when evaluating)
but no such things are available (or given) when evaluating a variable
in the global scope.
The fix is to handle compilation units with lo and hi pc == 0x0
similarly to a CU that has no lo and hi pc.
With this fix, valgrind --read-var-info=yes could properly
handle a big application with plenty of eliminated functions.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@13941
On 4 Apr 06, the debuginfo reader (m_debuginfo) was majorly cleaned up
and restructured. It has been a bit of a tangle for a while. The new
structure looks like this:
debuginfo.c
readelf.c
readdwarf.c readstabs.c
storage.c
Each .c can only call those below it on the page.
storage.c contains the SegInfo structure and stuff for
maintaining/searching arrays of symbols, line-numbers, and Dwarf CF
info records.
readdwarf.c and readstabs.c parse the relevant kind of info and
call storage.c to store the results.
readelf.c reads ELF format, hands syms directly to storage.c,
then delegates to readdwarf.c/readstabs.c for debug info. All
straightforward.
debuginfo.c is the top-level file, and is quite small.
There are 3 goals to this:
(1) Generally tidy up something which needs tidying up
(2) Introduce more modularity, so as to make it easier to add
readers for other formats, if needed
(3) Simplify the stabs reader.
Rationale for (1) and (2) are obvious.
Re (3), the stabs reader has for a good year contained a sophisticated
and impressive parser for stabs strings, with the aim of recording in
detail the types of variables (I think) (Jeremy's work). Unfortunately
that has caused various segfaults reading stabs info in the past few months
(#77869, #117936, #119914, #120345 and another to do with deeply nested
template types).
The worst thing is that it is the stabs type reader that is crashing,
not the stabs line-number reader, but the type info is only used by
Helgrind, which is looking pretty dead at the moment. So I have lifed
out the type-reader code and put it in UNUSED_STABS.txt for safe
storage, just leaving the line-number reader in place.
If Helgrind ever does come back to life we will need to reinstate the
type storage/reader stuff but with DWARF as its primary target.
Placing the existing stabs type-reader in hibernation improves
stability whilst retaining the development effort/expertise that went
into it for possible future reinstatement.