Age | Commit message (Collapse) | Author | |
---|---|---|---|
2022-03-28 | Release 0.8.0.v0.8 | Bartosz Taudul | |
2022-03-24 | Merge pull request #347 from bjacob/set-numcpus | Bartosz Taudul | |
set s_numCpus before reading it | |||
2022-03-24 | set s_numCpus before reading it | Benoit Jacob | |
2022-03-23 | Merge pull request #346 from mcourteaux/master | Bartosz Taudul | |
Warn the user about capstone deprecated branch in the manual. | |||
2022-03-23 | Fix typos. | Bartosz Taudul | |
2022-03-23 | Warn the user about capstone deprecated branch in the manual. | Martijn Courteaux | |
2022-03-19 | Merge pull request #344 from thedmd/callstack_table-clipboard_full_location | Bartosz Taudul | |
In callstack table view, copy full location with line to clipboard | |||
2022-03-19 | In callstack table view, copy location with line to clipboard instead just ↵ | thedmd | |
location. | |||
2022-03-18 | Backport SPSCQueue commit 3086fa9. | Bartosz Taudul | |
2022-03-18 | Apply SPSCQueue commit 62cdc1f3. | Bartosz Taudul | |
2022-03-18 | Update robin_hood to 3.11.5. | Bartosz Taudul | |
2022-03-18 | Bump zstd to 1.5.2. | Bartosz Taudul | |
2022-03-18 | Set focus on input field when find zone window appears. | Bartosz Taudul | |
2022-03-18 | Sync libbacktrace with 2446c6. | Bartosz Taudul | |
2022-03-17 | Merge pull request #339 from voysys/fix-server-assertion-crash | Bartosz Taudul | |
Ignore uninitialized GPU contexts when rendering GPU events | |||
2022-03-15 | Update manual. | Bartosz Taudul | |
2022-03-15 | Update NEWS. | Bartosz Taudul | |
2022-03-15 | Indicate currently selected annotation. | Bartosz Taudul | |
2022-03-15 | Allow adding full-view annotations. | Bartosz Taudul | |
2022-03-15 | Make the annotations list always accessible. | Bartosz Taudul | |
2022-03-15 | Move adding annotations to a separate function. | Bartosz Taudul | |
2022-03-15 | Add id to annotation description input box. | Bartosz Taudul | |
2022-03-11 | Define ___tracy_demangle on Android. | Bartosz Taudul | |
2022-03-10 | Merge pull request #341 from teajay-fr/bugfix/access_violation_on_error | Bartosz Taudul | |
Avoid accessing address 0 when reporting a failure. | |||
2022-03-10 | Avoid accessing address 0 when reporting a failure. | Thomas Bernard | |
2022-03-10 | Fix compilation on MSVC 2015. | Bartosz Taudul | |
2022-03-09 | Ignore uninitialized GPU contexts when rendering GPU events | Niclas Olmenius | |
This fixes an assertion crash when viewing GPU zones when no zones have been recorded on a GPU context if the client is compiled with TRACY_ON_DEMAND | |||
2022-03-07 | Merge pull request #337 from simonvanbernem/master | Bartosz Taudul | |
Make install_vcpkg_dendencies.bat work from any working directory | |||
2022-03-07 | Made install_vcpkg_dendencies.bat work from any working directory | Simon van Bernem | |
pushd %~dp0 will push the directory that contains the file as the working directory and we also pop it at the end to be a good citizen and restore the previous working directory | |||
2022-03-03 | Update manual. | Bartosz Taudul | |
2022-03-03 | Implement going to parent zone in find zone window. | Bartosz Taudul | |
2022-02-28 | Merge pull request #334 from daverigby/master | Bartosz Taudul | |
Fix macOS build breaks when _GNU_SOURCE defined | |||
2022-02-28 | program_invocation_short_name is Linux-specific | Dave Rigby | |
'program_invocation_short_name' is Linux-specific; other OSs such as macOS do not support it. Fixes build break on macOS 12.2 with _GNU_SOURCE defined. | |||
2022-02-28 | pthread_setname_np takes 1 arg on macOS | Dave Rigby | |
pthread_setname_np() can only set the name of the current thread on macOS, so only pass a single name argument. Fixes build break on macOS 12.2 with _GNU_SOURCE defined. | |||
2022-02-28 | Merge pull request #333 from sideeffects/opencl_transient | Bartosz Taudul | |
Add transient OpenCL Zones. | |||
2022-02-28 | Update TracyOpenCL.hpp | johner | |
Add transient OpenCL Zones. | |||
2022-02-24 | Merge pull request #332 from Lectem/cmake_options | Bartosz Taudul | |
CMake: Add missing options | |||
2022-02-23 | CMake: New description for TRACY_CALLSTACK | Clément Grégoire | |
2022-02-23 | Fix IPV4 -> IPv4 typo | Clément Grégoire | |
2022-02-23 | CMake: Add the new TRACY_NO_CALLSTACK option | Clément Grégoire | |
2022-02-23 | CMake: Add missing options | Clément Grégoire | |
2022-02-22 | Update manual. | Bartosz Taudul | |
2022-02-22 | Update NEWS. | Bartosz Taudul | |
2022-02-22 | Allow disabling callstack handling. | Bartosz Taudul | |
2022-02-21 | Merge pull request #325 from BrychDaneel/automatic_timer_fallback | Bartosz Taudul | |
Automatic timer fallback | |||
2022-02-19 | Codestyle fixes | Daniil Brychikov | |
2022-02-16 | Automatic timer fallback | Daniil Brychikov | |
2022-02-15 | prev_state is long. | Bartosz Taudul | |
2022-02-14 | Check if memory can be allocated in a thread. | Bartosz Taudul | |
Consider running the following code with operator new and delete overloaded to track allocations with call stacks: std::thread( []({ thread_local std::string str; }); Each call stack requires a memory allocation to be performed by the profiler, to make the stack available at a later time. When the thread is created, the TLS block is initialized and the std::string buffer can be allocated. To track this allocation, rpmalloc has to be initialized. This initialization also happens within the TLS block. Now, when the thread exits, the heap managed by rpmalloc may be released first during the TLS block destruction (and if the destruction is performed in reverse creation order, then it *will* be destroyed first, as rpmalloc was initialized only after the std::string initialization, to track the allocation performed within). The next thing to happen is destruction of std::string and release of the memory block it contains. The release is tracked by the profiler, and as mentioned earlier, to save the call stack for later use, a memory allocation is needed. But the allocator is no longer available in this thread, because rpmalloc was released just before! As a solution to this issue, profiler will detect whether the allocator is still available and will ignore the call stack, if it's not. The other solution is to disable the rpmalloc thread cleanup, which may potentially cause leak-like behavior, in case a large number of threads is spawned and destroyed. Note that this is not a water-tight solution. Other functions will still want to allocate memory for call stacks, but it is rather unlikely that such calls would be performed during TLS block destruction. It is also possible that the event queue will run out of allocated space for events at this very moment, and in such a case the allocator will also fail. | |||
2022-02-14 | Add rpmalloc thread state accessor. | Bartosz Taudul | |