Previously the correctly gdb-identified dynamic types were overwritten
with the value's static type. This is probably because the dynamic type
doesn't include the "&" for "reference". That, however, can easily be
fixed by just appending "&". As we're only handling references there it
should be safe to do so.
Task-number: QTCREATORBUG-10888
Change-Id: I7310916ce662956e66491423ad26658c32c8776b
Reviewed-by: Christian Stenger <christian.stenger@digia.com>
Reviewed-by: hjk <hjk121@nokiamail.com>
This only affects MinGW where the feature triggered QTCREATORBUG-5200 in
the past. Right now the performance impact of c9c0f80 (fixing
QTCREATORBUG-10826) is unclear and no time to investigate. Trading the
feature for robustness seems in order. Power-users can remove the return
line in the script to get it back.
Change-Id: If5a30808b72969493e5f40dd0fcbafdcccec343e
Reviewed-by: Eike Ziller <eike.ziller@digia.com>
Reviewed-by: Orgad Shaneh <orgads@gmail.com>
We probably should completely skip the code in the cases where Qt
itself already propagates the object name to the underlying threads.
Task-number: QTCREATORBUG-10826
Change-Id: I7e9e5b7c990a270c8563e40c1fe1e900724c1e5c
Reviewed-by: Ryan May <ryan.may@eecradar.com>
Reviewed-by: hjk <hjk121@nokiamail.com>
gdbbridge.py + dumper.py:
added detection on QNX target and ARM arch
added detection of Qt 3 support
fixed detection of Qt version
qttypes.py:
fixes of different memory alignment
stdtypes.py:
support of different libstdc++ internal structures on QNX
Change-Id: I808ee048c66c73c38bf5a8403e9cf881e767442e
Reviewed-by: hjk <hjk121@nokiamail.com>
Just cosmetics. Function was oddly named, in an odd place.
Change-Id: I8bfb33f7f41f01309f4b2be8a802d4093ae25e41
Reviewed-by: hjk <hjk121@nokiamail.com>
Pretty much the same code for GDB and LLDB, no reason to duplicate.
Change-Id: I4df4ba55e3b6bdfb6422179050f9a31bd1ebe082
Reviewed-by: Eike Ziller <eike.ziller@digia.com>
Recent versions of GDB seem to require 'make install' to find their
Python bits. Given that this is not really an option, check the
usual suspects, too.
Change-Id: I5217c0184681d4a86992fe0b8989498843b26cea
Reviewed-by: David Schulz <david.schulz@digia.com>
Reviewed-by: hjk <hjk121@nokiamail.com>
gdb.selected_inferior is too new for now.
Change-Id: I4290574b6dd2803a2fd2f860700693e7b2ac6efb
Reviewed-by: Rainer Keller <rainer.keller@digia.com>
Reviewed-by: hjk <hjk121@nokiamail.com>
That also makes lldb dumper actually test for bitness instead of
assuming 64bit
Change-Id: I6207bac04817e1f4700247dc00d2dfdace1e5100
Reviewed-by: hjk <hjk121@nokiamail.com>
Make it work with LLDB, show simple values always directly.
Change-Id: I463ef81183792f85243d679dee69a41db00bed07
Reviewed-by: hjk <hjk121@nokiamail.com>
This won't work without inferior calls (e.g. when loading core files),
so the old code is needed as fallback.
Change-Id: I1c9efc06374b00bc391bf55528546ccefc60349b
Reviewed-by: hjk <hjk121@nokiamail.com>
creatortypes.py wasn't included, and some d. qualification missing.
Change-Id: I1f444358cc1489f47083af39565147576f1885b7
Reviewed-by: hjk <hjk121@nokiamail.com>
That's a recent regression after the *bridge.py split.
Change-Id: I6488052e45167586bfe455d411985324bf882a15
Reviewed-by: David Schulz <david.schulz@digia.com>
... into qttypes, stdtypes, boosttypes, and misctypes.py
Change-Id: I3bce8dfe912c528f3a0d8d6670108f8d3272b4af
Reviewed-by: David Schulz <david.schulz@digia.com>
Don't try to use the compact method if we can't determine the
underlying format as we wouldn't know how to display the data
on the Creator side.
Change-Id: I985b9354c883da3a75bce1fe170b1f6cccc1dc21
Reviewed-by: David Schulz <david.schulz@digia.com>
This was left over from the first LLDB integration.
Change-Id: I99bf43d49d4a7650e1f1b2f572e264c1f234fd36
Reviewed-by: David Schulz <david.schulz@digia.com>
