| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| | * | | | | | Add a benchmark for testing our QtDBus type-validation | Thiago Macieira | 2011-04-13 | 3 | -1/+125 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | Our code is much faster than libdbus-1, even when built in debug mode. | |||||
| | * | | | | | Move the QDBusPerformance test to the tests/benchmark dir | Thiago Macieira | 2011-04-13 | 9 | -0/+431 | |
| | | |_|_|/ | |/| | | | ||||||
| * | | | | | Merge branch 'master' of scm.dev.nokia.troll.no:qt/qt into earth-master | axis | 2011-04-07 | 3 | -1/+235 | |
| |\ \ \ \ \ | |/ / / / | | | | | | | | | | | | | | | | Conflicts: src/corelib/thread/qthread_unix.cpp | |||||
| | * | | | | Merge remote branch 'qt-master/master' | Guoqing Zhang | 2011-04-05 | 12 | -22/+73541 | |
| | |\ \ \ \ | | | |_|/ | | |/| | | ||||||
| | * | | | | Merge remote branch 'qt-master/master' | Guoqing Zhang | 2011-03-16 | 7 | -7/+7 | |
| | |\ \ \ \ | | | |_|/ | | |/| | | | | | | | | | | | | Conflicts: src/corelib/thread/qthread_unix.cpp | |||||
| | * | | | | Remerging SymbianLite to team repository. | Sami Lempinen | 2011-03-09 | 4 | -5/+18 | |
| | |\ \ \ \ | ||||||
| | * \ \ \ \ | Merging SymbianLite branch to master. | Sami Lempinen | 2011-03-09 | 3 | -1/+235 | |
| | |\ \ \ \ \ | ||||||
| | | * | | | | | Using Symbian native mutex type in tst_bench_qmutex | mread | 2011-03-09 | 1 | -1/+20 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The QMutex benchmark contains benchmarks for the native mutex type. It was using pthread_mutex_t on Symbian, which isn't the true native mutex type and so isn't really representative of true native performance. Now it uses RMutex. Task-number: QTBUG-13990 Reviewed-by: Shane Kearns | |||||
| | | * | | | | | making the QWaitCondition benchmark more reliable | mread | 2011-03-09 | 1 | -8/+21 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The benchmark sometimes hung because wait conditions were sometimes signalled before being waited for. This change adds a real condition to control the interactions. Task-number: QTBUG-13990 Reviewed-by: Shane Kearns | |||||
| | | * | | | | | Starting a benchmark test for QWaitCondition | mread | 2011-03-09 | 2 | -0/+202 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This new test is added in anticipation of a native implementation of QWaitCondition for Symbian. Task-number: QTBUG-13990 Reviewed-by: Shane Kearns | |||||
| * | | | | | | | QNetworkDiskCache: clean up after merge request | Peter Hartmann | 2011-04-01 | 1 | -5/+5 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | changes include whitespace, parenting, recovering erroneously deleted code etc. Reviewed-by: Markus Goetz | |||||
| * | | | | | | | Speed benchmark for QNetworkDiskCache | Siddharth Mathur | 2011-04-01 | 3 | -1/+452 | |
| | |_|_|_|/ / |/| | | | | | | | | | | | | | | | | | | | | | | | Merge-Request: 2505 Reviewed-by: Peter Hartmann | |||||
| * | | | | | | Add a once-over unrolled fromLatin1 conversion (32 characters) | Thiago Macieira | 2011-03-27 | 1 | -0/+46 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This was suggested by Matthias Kretz on my blog. It improves on GCC (one case up to 5%), but worsens on ICC (up to 5%). | |||||
| * | | | | | | Add 16-byte loads of the Neon fromLatin1 functions | Thiago Macieira | 2011-03-27 | 1 | -0/+68 | |
| | | | | | | | ||||||
| * | | | | | | Remove the use of the QtTest baseline feature I reverted | Thiago Macieira | 2011-03-22 | 1 | -4/+4 | |
| | | | | | | | ||||||
| * | | | | | | Use the ARM version of UTF-8 detection in the Neon code | Thiago Macieira | 2011-03-22 | 1 | -16/+13 | |
| | | | | | | | ||||||
| * | | | | | | Update the data files for the QString benchmark test | Thiago Macieira | 2011-03-22 | 3 | -204/+71503 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | Also add the Neon build flags | |||||
| * | | | | | | Fix a bug in the SSE2 UTF-8 decoder. | Thiago Macieira | 2011-03-22 | 1 | -0/+4 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the multibyte sequence crossed the end of the SSE2 area, then len would have a wrong value and the data wouldn't be read. | |||||
| * | | | | | | Add ARM Neon versions of fromLatin1 and fromUtf8 | Thiago Macieira | 2011-03-22 | 1 | -0/+192 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The fromLatin1 code is very simple, yet the handwritten assembly performs better due to the use of post-increments. The fromUtf8 code has two alternatives. Neon lacks an instruction similar to SSE2's _mm_movemask_epi8 (PMOVMSKB) which extracts one bit from each byte and stores it in a register. We used that in the UTF-8 code to detect bytes with the highest bit set. To compensate, we used two alternatives: 1) AND the comparison result with a vector containing {128, 64, ...,1 } Do 3 parallel-adds (VPADD.I8), which will make the mask propagate to the lowest component in the vector. Trick found in: http://hilbert-space.de/?p=22 (comment 16-17) 2) Extract the two words from the doubleword Neon register and do the work in ARM assembly. It looks like the latter version is performing better. | |||||
| * | | | | | | Make this compile on Atom/Core2 (no SSE4) and on ARM (no SSE) | Thiago Macieira | 2011-03-22 | 1 | -1/+18 | |
| | | | | | | | ||||||
| * | | | | | | Add baselines and zeros to the benchmarks. | Thiago Macieira | 2011-03-22 | 1 | -6/+13 | |
| | | | | | | | ||||||
| * | | | | | | Add an UTF-8 conversion on trusted data and no BOM. | Thiago Macieira | 2011-03-22 | 1 | -0/+48 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This assumes that there are no overlong sequences, no continuation characters without the leading, no missing continuations and no BOM. | |||||
| * | | | | | | Make it easier to write a UTF-8 conversion on trusted data | Thiago Macieira | 2011-03-22 | 1 | -38/+58 | |
| | | | | | | | ||||||
| * | | | | | | Add the missing tests and 4-byte UTF-8 sequences | Thiago Macieira | 2011-03-22 | 1 | -1/+23 | |
| | | | | | | | ||||||
| * | | | | | | Improve the code and avoid unnecessary stores | Thiago Macieira | 2011-03-22 | 1 | -7/+9 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If there's an UTF-8 high byte in the first 8 bytes, don't try to save the latter 8 characters | |||||
| * | | | | | | Add an UTF-8 conversion optimised for ASCII using SSE2 | Thiago Macieira | 2011-03-22 | 1 | -1/+54 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This code is 2x faster than the original UTF-8 code and within 35% of the pure Latin1 code | |||||
| * | | | | | | Add an UTF-8 conversion code that is optimised for ASCII | Thiago Macieira | 2011-03-22 | 1 | -2/+94 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | This code is ~1.5x faster than the original UTF-8 code | |||||
| * | | | | | | Add a stateless copy of the Qt 4.7 UTF-8 codec. | Thiago Macieira | 2011-03-22 | 1 | -0/+99 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | For whatever reason, this code is worse than the stateful code... | |||||
| * | | | | | | Add UTF-8 code benchmarks | Thiago Macieira | 2011-03-22 | 1 | -0/+223 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Also compare to the Latin1 functions. The UTF-8 algorithm in Qt 4.7 right now is 109% slower than the unoptimised Latin-1 algo, 120% than the Qt 4.7 SSE2 code | |||||
| * | | | | | | Improve a little more the core loop and propagate to the other code | Thiago Macieira | 2011-03-22 | 1 | -18/+22 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently are that the "improved SSE2" version and the SSE4.1 version are yielding the best results, within 1% of each other. These results are around 20% better than the Qt 4.7 code. | |||||
| * | | | | | | Reduce the number of operations in the main loop | Thiago Macieira | 2011-03-22 | 1 | -8/+7 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The more operations there are, the more time it takes. More importantly, the more variables we touch, the more the compiler may want to use the stack instead of registers. | |||||
| * | | | | | | Add an SSE4 version using PMOVZXBW and PSRLDQ | Thiago Macieira | 2011-03-22 | 1 | -0/+22 | |
| | | | | | | | ||||||
| * | | | | | | Attempt to improve the epilog code | Thiago Macieira | 2011-03-22 | 1 | -19/+72 | |
| | | | | | | | ||||||
| * | | | | | | Add two SIMD overcommit prologs | Thiago Macieira | 2011-03-22 | 1 | -0/+21 | |
| | | | | | | | ||||||
| * | | | | | | Try to remove the non-determinism by moving the malloc() away | Thiago Macieira | 2011-03-22 | 1 | -10/+15 | |
| | | | | | | | ||||||
| * | | | | | | Optimise the prolog even further | Thiago Macieira | 2011-03-22 | 1 | -16/+33 | |
| | | | | | | | ||||||
| * | | | | | | Try to improve the prolog by doing less operations | Thiago Macieira | 2011-03-22 | 1 | -7/+7 | |
| | | | | | | | ||||||
| * | | | | | | Add an unrolled prolog | Thiago Macieira | 2011-03-22 | 1 | -1/+22 | |
| | | | | | | | ||||||
| * | | | | | | Make the prolog function more generic | Thiago Macieira | 2011-03-22 | 1 | -12/+6 | |
| | | | | | | | ||||||
| * | | | | | | Add an SSE2 alternative with prolog | Thiago Macieira | 2011-03-22 | 1 | -0/+48 | |
| | | | | | | | ||||||
| * | | | | | | Add some boundary/spill protection | Thiago Macieira | 2011-03-22 | 1 | -4/+8 | |
| | | | | | | | ||||||
| * | | | | | | Correct the code: use ushorts, not QChar | Thiago Macieira | 2011-03-22 | 1 | -4/+4 | |
| | | | | | | | ||||||
| * | | | | | | Add the SSE2 code we have in Qt 4.7 to the benchmark | Thiago Macieira | 2011-03-22 | 1 | -0/+26 | |
| | | | | | | | ||||||
| * | | | | | | Add a test for QString::fromLatin1 performace alternatives | Thiago Macieira | 2011-03-22 | 5 | -11/+414 | |
| | | | | | | | ||||||
| * | | | | | | Remove the __attribute__((optimize)) marks | Thiago Macieira | 2011-03-22 | 1 | -11/+11 | |
| | | | | | | | ||||||
| * | | | | | | Add a script that generates a string table of 8-bits | Thiago Macieira | 2011-03-22 | 1 | -0/+203 | |
| | | | | | | | ||||||
| * | | | | | | Add a benchmark for QString::fromLatin1 | Thiago Macieira | 2011-03-22 | 1 | -0/+36 | |
| | | | | | | | ||||||
| * | | | | | | Compile in C++0x mode | Thiago Macieira | 2011-03-22 | 1 | -10/+10 | |
| | | | | | | | ||||||
| * | | | | | | add license headers to qhash_string benchmark | hjk | 2011-03-21 | 2 | -1/+40 | |
| | | | | | | | ||||||
| * | | | | | | benchmarks: first shot at alternative implementation for qHash(QString) | hjk | 2011-03-16 | 5 | -0/+435 | |
| | |_|_|/ / |/| | | | | ||||||
 |
index : Qt.git | |
| Qt s a cross-platform application framework that is used for developing application software that can be run on various software and hardware platforms with little or no change in the underlying codebase, while still being a native application with native capabilities and speed. |
| summaryrefslogtreecommitdiffstats |