| d8286ed3496bcd69cd4796f57a5b5c04859d2378 |
|
04-Nov-2013 |
Petr Machata <pmachata@redhat.com> |
Show "nil" also for void* values
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| cb943f1d54fb8d45879e9545f97f236f833bf7ea |
|
26-Sep-2013 |
Peter Wu <lekensteyn@gmail.com> |
Add tests for %p format specifier crash and zero(EXPR) leaks
Notes for these regression tests:
- The %p test should run when ltrace is built with -fsanitize=address or
some other form of protection that makes the program abort when trying
to free() static memory.
- The zero(EXPR) test needs valgrind to test for the previous memleak
bug, but as it stands now, it can also test whether the zero
functionality works or not.
Signed-off-by: Peter Wu <lekensteyn@gmail.com>
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| 1ad42523638fd49c1200cbf81ba6a39f864c91cb |
|
22-Mar-2013 |
Petr Machata <pmachata@redhat.com> |
Fix a problem in lookup of typedef X_* when X is an intrinsic type
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| e8701b4cc4333dbc5e3c683468fa8fb077977ba6 |
|
06-Dec-2012 |
Petr Machata <pmachata@redhat.com> |
Move ltraceParamTest from parameters2.exp to ltrace.exp, name ltraceLibTest
Also, change the semantics: ltraceLibTest doesn't do matching anymore,
but is a wrapper around several ltraceSource, ltraceCompile and
ltraceRun. Matching is done by the callee. This allows us to choose
the exact matcher to use.
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| bb04f57417d132a2f4aeb738d7f47acbdbbbf457 |
|
24-Nov-2012 |
Petr Machata <pmachata@redhat.com> |
Add test for "addr" type
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| 0965420dfd1f4167609c4b3bbee5dc0277423897 |
|
05-Dec-2012 |
Petr Machata <pmachata@redhat.com> |
Don't mangle static symbol names internally
Instead, write the @soname bit separately. I'm not sure why this was done
like that in the first place. I may have cut'n'pasted this from RHEL
patches for dlopen support, where these sorts of hacks were necessary.
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| e5a8cef0bf0e96480d6ca0355bb64d3121a88858 |
|
29-Nov-2012 |
Petr Machata <pmachata@redhat.com> |
Don't use system-wide 'true', compile own binary instead
On biarch systems, true may come from the wrong arch, and ltrace may be
unable to trace it.
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| f197727e6247be1ee08d2a667931aee20512ae18 |
|
29-Nov-2012 |
Petr Machata <pmachata@redhat.com> |
Allow formatting floats, doubles in hex
Include a test suite case.
Describe the extension in ltrace.conf.5.
Mention it in NEWS.
There's no similar agreed-upon convention for oct, and displaying which
bits are set in a floating number like what bitvect does is certainly not
helpful. Hence, remove the corresponding TODO item. Add a different
(unrelated) one instead.
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| ec4ab256e5c7fe1596dd90a3c8e8d4b052ca1207 |
|
20-Nov-2012 |
Petr Machata <pmachata@redhat.com> |
Add lens "bitvec" for displaying objects as bit vectors
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| 3219c86c7fa6d5464fe4b14ce038002851d0d5cc |
|
19-Nov-2012 |
Petr Machata <pmachata@redhat.com> |
Allow typedefs of lenses
This worked for keywords such as enum that are handled specially, because
those behave as types for backward compatibility. It was however not
possible to typedef e.g. hex.
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| 6248a0ac394b2aa3b2267eaa1220a90b609b6f86 |
|
19-Nov-2012 |
Petr Machata <pmachata@redhat.com> |
Guard number of expanded structures of the same type
The logic behind that is that if we limit array output, we should similarly
limit linked list output. Number of expansions of the same-typed structure
is a reasonable metric for that. Double recursive structures will expand
a bit too much, but on the other hand we want the stuff that is typed out
in the configure file to display fully, and not be considered for trimming.
Test suite included.
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| 307b90b158b79ba3aae61d5c6612b4769b10be5f |
|
19-Nov-2012 |
Petr Machata <pmachata@redhat.com> |
Allow forward declarations of structs
Now ltrace can display singly-linked lists. Recursion detection was added
to format_pointer. It relies on the fact that ltrace is not multi-threaded
and doesn't need per-process or per-thread cache of already-displayed
values, because we display one value at a time anyway.
Several test cases added.
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| b17863f2dc7a0f2edbf6a2482028c829b91c2c09 |
|
18-Nov-2012 |
Petr Machata <pmachata@redhat.com> |
Guard against redefinition of typedef
Test case included
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|
| 1a399cde69dca38591be563e062f60380bbd41e4 |
|
18-Nov-2012 |
Petr Machata <pmachata@redhat.com> |
Parse correctly typedefs with common prefix
More exactly, if there is a typedef X and another XY, and a reference is
made to X, don't instead match XY. Test case included.
/external/ltrace/testsuite/ltrace.main/parameters2.exp
|