summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2012-03-24Sync from internal Google development.Joshua Haberman
Many improvements, too many to mention. One significant perf regression warrants investigation: omitfp.parsetoproto2_googlemessage1.upb_jit: 343 -> 252 (-26.53) plain.parsetoproto2_googlemessage1.upb_jit: 334 -> 251 (-24.85) 25% regression for this benchmark is bad, but since I don't think there's any fundamental design issue that caused it I'm going to go ahead with the commit anyway. Can investigate and fix later. Other benchmarks were neutral or showed slight improvement.
2011-12-23Fixes to un-break "make descriptorgen"Joshua Haberman
2011-12-22Small Makefile fix.Joshua Haberman
2011-12-22Fixed the open-source build.Joshua Haberman
2011-12-22Sync with internal Google development.Joshua Haberman
This breaks the open-source build, will follow up with a change to fix it.
2011-11-23Refinement of upb_bytesrc interface.Joshua Haberman
Added a upb_byteregion that tracks a region of the input buffer; decoders use this instead of using a upb_bytesrc directly. upb_byteregion is also used as the way of passing a string to a upb_handlers callback. This symmetry makes decoders compose better; if you want to take a parsed string and decode it as something else, you can take the string directly from the callback and feed it as input to another parser. A commented-out version of a pinning interface is present; I decline to actually implement it (and accept its extra complexity) until/unless it is clear that it is actually a win. But it is included as a proof-of-concept, to show that it fits well with the existing interface.
2011-11-19Changes to get upb compiling inside Google.Joshua Haberman
2011-09-21Fix GDB unwinding through JIT-compiled frames.Joshua Haberman
2011-09-20Fixes to get upb to compile inside Google.Joshua Haberman
2011-09-20x86 JIT: now works without -NDNDEBUG (also small perf increase).Joshua Haberman
2011-09-20Small bugfix for x86->x64 rename.Joshua Haberman
2011-09-20Fix perf-tests.sh to skip building non-upb tests.Joshua Haberman
2011-09-19Rename x86 -> x64, since JIT is x64-only.Joshua Haberman
2011-09-15x86 JIT: avoid storing pointer for every field.Joshua Haberman
2011-09-14Small typo fix.Joshua Haberman
2011-09-14x86 JIT: add easy switch between replicated/common dispatch.Joshua Haberman
2011-09-12Added an example, constified some more methods.Joshua Haberman
2011-09-12Fleshed out fielddef default functionality.Joshua Haberman
Fixes unit test submitted by Hunter Morris (thanks!).
2011-09-09Add failing unit testHunter Morris
This might actually just bring to light my misuse of the upb_fielddef functions. The test assertions are fine, but an assertion in upb/upb.h fails: ./upb/upb.h:181: upb_value_getptr: Assertion `val.type == 33' failed.
2011-09-05Merge pull request #2 from winksaville/fix-typo-handler.hJoshua Haberman
Fix typo in handler.h
2011-09-05Fix typo in handler.hWink Saville
Signed-off-by: Wink Saville <wink@saville.com>
2011-09-04Header tweaking.Joshua Haberman
2011-09-04Fleshed out C++ def wrappers some.Joshua Haberman
2011-09-04Const invasion: large parts of upb made const-correct.Joshua Haberman
2011-09-03Work on C++ wrappers, added C++ test.Joshua Haberman
2011-08-31Header cleanup, clarify/correct comments for interfaces.Joshua Haberman
2011-08-31Make "benchmark" build "benchmarks", add bug note.Joshua Haberman
2011-08-31Revised upb_bytesink, refactored upb_textprinter (untested).Joshua Haberman
2011-08-27Refined upb_status.Joshua Haberman
2011-08-27Add packed field support (untested).Joshua Haberman
2011-08-24Some updates to the experimental C++ wrapper.Joshua Haberman
2011-08-22Add comment clarifying that the proto2 benchmark is ugly and temporary.Joshua Haberman
2011-08-21Some progress on upb_stdio.Josh Haberman
It's still not correct, but it's closer.
2011-08-20Fold gen-deps.sh into Makefile.Joshua Haberman
2011-08-20Let the JIT emit hasbit-setting code in addition to calling a callback.Joshua Haberman
This leads to a major (20-40%) improvement in the parsetoproto2 benchmark with small messages. We now are faster than proto2 in all apples-to-apples comparisons, at least given the (admittedly limited) set of benchmarks in this source tree.
2011-08-20Fix stack alignment on OS X.Joshua Haberman
2011-08-20Prime the decoder buf for modest perf improvement on small messages.Joshua Haberman
2011-08-20Some Makefile cleanup.Joshua Haberman
2011-08-19More cleanup.Joshua Haberman
2011-08-19Some source cleanup/commenting.Joshua Haberman
2011-08-19Renamed lang_ext -> bindings, README updates.Joshua Haberman
2011-08-19Avoid longjmp() in successful case.Josh Haberman
Speeds up short messages by 15-25%.
2011-08-19Fix proto2 benchmark for newer proto2.Josh Haberman
2011-08-19Benchmark to parse into proto2 messages.Joshua Haberman
2011-08-17x86 JIT: add callback specializations for a 10% speedup when parsing to struct.Joshua Haberman
2011-08-17Benchmark JIT vs no JIT without forcing a whole-project recompile.Joshua Haberman
2011-08-12Python: fleshed out accessors.Joshua Haberman
2011-08-12Python: basic SymbolTable support and empty accessors.Joshua Haberman
2011-08-11Python: fixed object cache and fleshed out MessageDef a bit.Joshua Haberman
2011-08-10Fixed overzealous assert().Joshua Haberman
generated by cgit on debian on lair
contact matthew@masot.net with questions or feedback