As this is also used after -exec-continue we won't see a result
before the next stop and consequently run into the watchdog
timeout regularly.
Change-Id: Ibb7e42ae1810d4d9f536cf755219fba6d8aa0f96
Reviewed-by: Orgad Shaneh <orgads@gmail.com>
For some reason, the executable is no determined correctly.
Feed it to gdb so it will have correct debugging symbols.
Fixes: QTCREATORBUG-26208
Change-Id: Id7da405e5ec9f9331399eb049fa383099981cb94
Reviewed-by: hjk <hjk@qt.io>
Reviewed-by: David Schulz <david.schulz@qt.io>
COOKIE FOR TOKEN 0 ALREADY EATEN (InferiorStopOk). TWO RESPONSES FOR ONE
COMMAND?
Change-Id: I569a9715f073f3fbba71045f14fa30a06cd434e9
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Reviewed-by: hjk <hjk@qt.io>
The type is implicitly given by the command's executble FilePath.
Also, rename the rarely used addArgs(const CommandLine &) overload
to addCommandLineAsArgs() and make it strip scheme and host from
the wrapped executable as there are no uses expected where keeping
them would be the right thing.
Change-Id: Id0b76778e7e01ac16e477f36bf30bb28d96bb177
Reviewed-by: Artem Sokolovskii <artem.sokolovskii@qt.io>
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
This patch needs to be applied together with the parent change.
There are 3 basic cases:
1. The user doesn't write anything to the write channel:
You don't need to call closeWriteChannel manually anymore.
By default the QtcProcess you create is in ProcessMode::Reader mode.
Internally it opens the process in ReadOnly mode and
closes the write channel just after starting.
2. The user writes some initial data (after being started)
and then closes the write channel:
All what is needed now it to set the write data
(QtcProcess::setWriteData) before calling start.
You also use the default ProcessMode::Reader mode.
Internally it opens the process in ReadWrite mode
and writes the writeData asynchonously when the process
already started. After writing the data it closes the
write channel automatically.
3. The user writes the data also after calling start.
All you need now is to create a process with
ProcessMode::Writer mode. In this mode the write
channel is not closed.
Internally it opens the process in ReadWrite mode
as some writers also read the data from the process.
All the code base is adapted here to the above rules.
Change-Id: Id103019d1d71a3012fd1eade226fe96b9aaa48c2
Reviewed-by: hjk <hjk@qt.io>
Mid-term plan is to concentrate on use of QtcProcess::result()
instead which is a bit more system-agnostic.
There's quite a bit of potential for downstream cleanup by
re-using QtcProcess::exitMessage() now.
Change-Id: I3806b3f5933d96e64b7cfb18cc6c52823fddcbcd
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
GDB 8.3+ can automatically save indices of DWARF symbols on disk
to speed up further loading of the same binaries.
Make this option available to the user and switch it on by default.
No harm is done for gdb builds without this feature.
Task-number: QTCREATORBUG-23207
Change-Id: Id0d467eee429a94f1d8e826a883179796732d31e
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
The Windows-only restriction is nowadays handled inside QtcProcess.
Change-Id: I77d6914831ff172026665a429b497940c60970ac
Reviewed-by: David Schulz <david.schulz@qt.io>
To make clear that this is not just any finish.
Also change FinishedError to FinishedWithError, to create
symmetry.
Also adapt enum member description to reality.
Change-Id: I13e05391eb86fdb24e2ae660f14dfddb282e1104
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
This is not recommended to use, and using Environment::systemEnvironment()
also saves a conversion.
Change-Id: I7af24a5bbf432617a75421e6950292acc4b529a3
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
The previous indirection was useful as long as they were overloads
for QProcessEnvironment and QStringList, but these are gone now.
Change-Id: I5066bd2e72fd06948a5cc7bbac6dda9006db96ed
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
Makes run() more similar to what start() looks like.
Also add some asserts to make sure run() and related functions are
only called on SyncronousProcesses, as these are currently the only
ones where this works.
Change-Id: Idee6076c3f40a484db5c17f5bb348698cc83d220
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
The result is fully stored in the object anyway. Using the extra
SynchronousProcessResponse structure only causes copies of
the data and complicates access on the user side in
a lot of cases.
The result bits are now also accessible individually.
There's obvious room for follow-up changes on the topic, e.g.
ShellCommand::runCommand's parameter list could shrink to
just a SynchronousProcess parameter.
Change-Id: I45aa7eb23832340be06905929280c012e1217263
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
Mechanical to prepare merging the actual classes.
Adapting #includes.
Change-Id: I77a2c28129287778bc870c30cb890cd26bc2e62b
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Makes the interface more similar to QtcProcess.
Change-Id: I58e57d9fdb7c37eb0d2a5c5eef8643d6be97c3cc
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Reviewed-by: Orgad Shaneh <orgads@gmail.com>
Saves some code on the user side.
Change-Id: I32cd220b6e533f5497a1865f9c34ab9db4cfda79
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Eike Ziller <eike.ziller@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>
For local run and GDB debug, with or without terminal.
Task-number: QTCREATORBUG-2831
Task-number: QTCREATORBUG-25330
Change-Id: I9b5d2156bcffea4f358474ecdbcad580a4419917
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
Previously, ptrace was used to keep a tight hold on the spawned process
across the exec(), followed by raising a SIGSTOP to allow un-racy
attaching a debugger.
This makes implementation of 'Run as root' more difficult and is
apparently not needed: instead of the ptrace use, the SIGSTOP can be
raised directly, before exec().
Change-Id: I36025ac547b2a335e2a203c728d221830e4c0a7d
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
Currently only the previous from/to pair, but can be extended
by codec state.
Task-number: QTCREATORBUG-24462
Change-Id: I3d101e74d1fef65bb75ddaab1dc2eaf77201dcde
Reviewed-by: David Schulz <david.schulz@qt.io>
These days the ISAs are adding more and more registers, which sometimes
are quite hard to follow. For instance if you're interested only in float
or only in system/CSRs registers you have to scroll down quite a lot for
monitor them.
Grouping them by thier type, it makes it much easier to follow each
register type.
Change-Id: I268626b3fe71aa8fa54be0999836caf03e8780ca
Reviewed-by: hjk <hjk@qt.io>
This avoids pending breakpoints in local attach setups now.
Change-Id: I6872a97020955180d4c3f915f6bb19c1f771f07f
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Chances are high these are triggered by the process stub. There is
a small chance for false positives here, but true SIGSTOPs by
other external sources in these locations are very unlikely.
Task-number: QTCREATORBUG-25073
Task-number: QTCREATORBUG-25082
Change-Id: I25478caf6803877f2f8cbb2edef68ae1183223a6
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Called with different timing, no real shared code currently.
Change-Id: I5a8cc32c5f38f533acfa4512518fb51932d8aab2
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
In case of 'attach' this is still a bit too early to avoid pending
breakpoints, but we are getting there.
Also, lump in the commandsAfterConnect bits into the initial breakpoint
setting, as this is the right time. The function is mis-nomed now, will
be fixed in a later patch.
Change-Id: I8c4c53d02b4eab31e8e095865fc3cb24265c5028
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
