Although already tried to lookup a native type with an
lldb internal mechanism we still tried to do it "manually"
again.
This secondary approach was needed at some point when lldb
had no integrated way to lookup a type inside all modules.
Lookups done manually will not provide a better result than
the lldb internal one.
The error output generated with this blocks debugging which
makes QC unusable at this state.
So, remove the secondary lookup completely.
Fixes: QTCREATORBUG-25185
Fixes: QTCREATORBUG-25217
Change-Id: Ibd8a125a89633c611bf750e0f1759c639717e1d2
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Reviewed-by: hjk <hjk@qt.io>
Previously, the starting inferior was sent a SIGSTOP to avoid
progress before the debugger could attach.
However, these signals are then also visible in the debugger and
need to be ignored as part of the startup handling in Creator.
The waiting effect can be achieved less intrusively by waiting
on a pipe read between fork() and exec().
Task-number: QTCREATORBUG-25073
Task-number: QTCREATORBUG-25082
Task-number: QTCREATORBUG-25227
Change-Id: Ie70b9eb5ea865f85411c26b0dbf377a019fec8d5
Reviewed-by: Eike Ziller <eike.ziller@qt.io>
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Looking up function signatures ends up in massive
stderr output which blocks the debugger for ages, so omit
them as long we are not sure where they originate.
Task-number: QTCREATORBUG-25185
Task-number: QTCREATORBUG-25217
Change-Id: I9b022d2194a6cb61651ee0648be526fc94a02da9
Reviewed-by: hjk <hjk@qt.io>
Case like 'template<typename T> using TVector = std::vector<T>'
end up only as 'TVector' in the debug info. This is unsuitable
as an id in our type info caching as this would only allow one
type layout for all specializations, which is not the case for
e.g. std::vector<bool>.
The solution is to mangle the target type into the id, as already
done for GDB.
This makes the Typedef2 dumper test pass.
Change-Id: I11538bbf6431f61a11c18366a2a2b4911cdc2e0e
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
This fixes the Internal2, EnumFlags and EnumInClass tests on Linux.
The values lose now the previously hand-crafted Class:: prefixes,
but the context is clear from the type column, and it's what LLDB
developers (and potentially users) consider normal.
Change-Id: I09e41f7b4fb4f078ef3f535fe650d06e7c2a0331
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Reviewed-by: hjk <hjk@qt.io>
In the newest version of LLDB, they remove anonymous and inline
namespaces from their display name, which breaks all debugger helper
code on Macs with the newest version of Xcode, since Qt Creator logic
determines libc++ vs. libstdc++ by the presence of the "__1" inline
namespace.
Change-Id: I139d0654ffbc9dec2e42b40eaad92ea2e3c067c0
Reviewed-by: hjk <hjk@qt.io>
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
... to something that works with LLDB 12.
Task-number: QTCREATORBUG-24596
Change-Id: Ib2f8255d45a02fa3d2c737864994397f587e9b15
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Reviewed-by: Eike Ziller <eike.ziller@qt.io>
Sample broken output:
>result={token="9",lldbid="1",valid="1",hitcount="0",threadid="0",oneshot="0",
condition="",enabled="1",valid="1",ignorecount="0",locations=[
{locid="1",function="foo()",enabled="1",resolved="0",valid="1",ignorecount="0",
file="F:\Projects\test\main.cpp",line="5",addr="4199979"},],
file="F:\Projects\test\main.cpp",line="5"}@
When parsed as GdbMi value, the strings are expected to be escaped.
Change-Id: Idb923516c1cf6e25b970ad08fae977bdf1045b4b
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Partially revert 7f958700a0 for 'remote-ios'.
The original commit changed the way attaching to a remote server or process
works, attempting to make it work with lldb-server on a remote linux device.
That breaks connecting to the debugging server on iOS devices.
Fixes: QTCREATORBUG-23995
Change-Id: I7a793fa73a564a4ef19cf82e13c2ad50d4247ee3
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
... and remove some of the now implicit putNumChild() calls.
Also, adapt docs.
Change-Id: I1ab1dafada95a1703cd6a9ba5b9e7e4166b48cd9
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
This could be extended to the other bridges, not done in this patch.
Change-Id: I620290049b7c95f8e3fb7584d4ca99a42fd343d4
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
This is apparently the right thing to do anyway and also helps to
select the right emulator or device when multiple ones are connected.
In that situation otherwise an error
"Expected a single connected device, got instead 2 -
try setting 'ANDROID_SERIAL'"
would occur.
Change-Id: I650a221d1a321d4dd9035411f85c7a68244c20e2
Reviewed-by: BogDan Vatra <bogdan@kdab.com>
This can appear as (non-existing) preferred frame index in backtraces
consisting only of frames without debug info. Fall back to top most
frame in that case.
Change-Id: Id7bb941d5fc3080aff3853cebb7b92afdd0ba4ea
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
No need to start the debugger event loop if the engine setup fails
Change-Id: I6bdeda0d2609c79aa5fafff7bc4b373a0bf14fbd
Reviewed-by: David Schulz <david.schulz@qt.io>
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Breakpoints inserting is async, so they did end up when the process
was running already.
This adds an extra roundtrip. Better, but more intrusive solution
might be to set the initial breakpoints synchronously, but that
would touch all engines.
Change-Id: Ia728a9c5ae8f1d6d29d3cc02b9e2d04952091fe9
Reviewed-by: David Schulz <david.schulz@qt.io>
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
The device on remote side *has* to configure (/usr/bin/)lldb-server as
"GDB server executable" in the device settings.
A real gdbserver does not work.
Change-Id: I045ffb60a824e06ee683d8bdfffeb480a580af5f
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
generated with autopep8 and the introduced setup.cfg
Change-Id: I5b55a7937f2cde81be59d2b434e94ada1a4af5ab
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Using imports like
from foo import *
is considered as bad habit and it reduces
static code analysis usability.
Change-Id: I56a175f4c7b231e2b8e486bd9d1c65543720f56a
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Tweak handling for LLDB provided by Xcode 11 slightly,
as the state handling of the process went astray when
using the debugger Python 3 based.
Task-number: QTCREATORBUG-22955
Change-Id: Ie070049d210a570993507ed940d141237aab89ce
Reviewed-by: hjk <hjk@qt.io>
Reviewed-by: Eike Ziller <eike.ziller@qt.io>
If the system or the running inferior produces output while
fetching data the output did interfere with the state handling
of the debugger engine which in turn lead to a complete
inconsistent state of the debugger where the states of the
debugger bridge and the debugger engine (QC internal) assumed
complete different states.
Circumvent by explicitly ignoring the state for inferior output
on stdout or stderr.
Change-Id: I81aa0bacd41de4266a9278c66fed908dd72c9d44
Reviewed-by: hjk <hjk@qt.io>
Reviewed-by: Eike Ziller <eike.ziller@qt.io>
As per suggestion from Pyls, this changes
if not needle in haystack:
to
if needle not in haystack:
Change-Id: I4a482604e13e61ecee9e02935479632419710ff7
Reviewed-by: hjk <hjk@qt.io>
Reviewed-by: Robert Loehning <robert.loehning@qt.io>
Reviewed-by: Alessandro Portale <alessandro.portale@qt.io>
