| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | merge 8.5; fix notifier mask bug and Tcl_Read performance regression | dgp | 2014-08-22 | 2 | -21/+12 |
| |\ | |||||
| | * | Correct performance regression in a series of short binary reads from a | dgp | 2014-08-22 | 1 | -19/+7 |
| | |\ | | | | | | | socket. Many thanks to Eric Boudaillier for the report and testing support. | ||||
| | | * | Same results; simpler logic. | dgp | 2014-08-22 | 1 | -21/+7 |
| | | | | |||||
| | | * | Test fix for likely cause of reported I/O slowdown. | dgp | 2014-08-21 | 1 | -0/+2 |
| | |/ | | | | | | | | | | | | | | | | | In a DoRead() revision, it came to favor making every effort to fill buffers, in preference to a more sensible goal of favoring avoiding calls out to the driver if there's already enough data in the buffers to satisfy the read operation. Result is many more calls out to recv() than are a good idea. Ought to show up most glaringly when many Tcl_Read() calls asking for small numbers of bytes (compared to buffer size) each, and that matches the reported case. | ||||
| | * | Fix reviewed and accepted. | andreask | 2014-08-20 | 1 | -1/+1 |
| | |\ | |||||
| | | * | Docs for Tcl_CreateChannelHandler() state that the registered handler proc | dgp | 2014-08-20 | 1 | -1/+1 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | will be called back with a mask value. "Mask is an integer mask indicating which of the requested conditions actually exists for the channel; it will contain ***a subset of the bits from the mask argument*** to Tcl_CreateChannelHandler when the handler was created." (emhpasis added). Tcl_NotifyChannel is not honoring this. It passes a mask value that may contain bits not in common with the mask argument to T_CCH(). This commit is a one-liner patch adding in the masking step to make things behave as documented. Thanks to apn for digging this out. (In combination with other questionable code, this led to a hang in test http-4.6 on Windows) Tcl_NotifyChannel() has had this error in all of recorded Tcl history. It's hard to imagine any code dependent on it though. If any exists, it can be revised to pass the mask value it truly needs to T_CCH() and end up with code suitable both before and after this change. If you concur, please merge to core-8-5-branch, and I'll take it from there. | ||||
| | * | | Make test io-36.1.1 more portable. | dgp | 2014-08-20 | 1 | -1/+4 |
| | |/ | |||||
| * | | [74e073599e5e7c5eb535e2f60f1b06feb468eaae|74e073599e]: tclsh is using old ↵ | jan.nijtmans | 2014-08-20 | 1 | -0/+18 |
| |\ \ | |/ | | | | | style dialogs when Tk is loaded on Windows | ||||
| | * | [74e073599e5e7c5eb535e2f60f1b06feb468eaae|74e073599e]: tclsh is using old ↵ | jan.nijtmans | 2014-08-20 | 1 | -0/+18 |
| | | | | | | | | | style dialogs when Tk is loaded on Windows | ||||
| * | | [7d52e1101b] oo::object should know that oo::class is a subclass. | dkf | 2014-08-17 | 2 | -0/+18 |
| | | | |||||
| * | | Put back SystemV timezone files, which were removed in previous commit. ↵ | jan.nijtmans | 2014-08-13 | 13 | -0/+65 |
| |\ \ | |/ | | | | | Don't know if that was by accident, but seems unwise in a Tcl patch release. | ||||
| | * | Put back SystemV timezone files, which were removed in previous commit. ↵ | jan.nijtmans | 2014-08-13 | 13 | -0/+65 |
| | | | | | | | | | Don't know if that was by accident, but seems unwise in a Tcl patch release. | ||||
| * | | merge tzdata2014f from ietf.org | Kevin B Kenny | 2014-08-13 | 91 | -2874/+3044 |
| |\ \ | |/ | |||||
| | * | tzdata2014f from ietf.org | Kevin B Kenny | 2014-08-13 | 91 | -2874/+3044 |
| | | | |||||
| * | | merge-mark. Do not accommodate udp bugs in Tcl 8.6.2. | dgp | 2014-08-12 | 0 | -0/+0 |
| |\ \ | |/ | |||||
| | * | Accommodate the "udp" package, and any other Tcl package that expects a | dgp | 2014-08-12 | 1 | -1/+1 |
| | | | | | | | | | | | [read] on a channel @ EOF to attempt another pass through the channel drivers instead of immediately returning an empty string. Correcting this misbehavior appears too disruptive in a patch release to a long stable branch. | ||||
| * | | Fixed type conversion warnings which break a Win32 debug build (MSVC). | andreask | 2014-08-06 | 1 | -2/+10 |
| | | | |||||
| * | | Add tclsh.exe.manifest.in to the distribution | dgp | 2014-08-04 | 1 | -0/+1 |
| |\ \ | |/ | |||||
| | * | Add tclsh.exe.manifest.in to the distribution. | dgp | 2014-08-04 | 1 | -0/+1 |
| | | | |||||
| * | | Another Preserve/Release on channels is needed during IO finalization. | dgp | 2014-08-02 | 1 | -0/+2 |
| |\ \ | |/ | |||||
| | * | Another Preserve/Release on channels is needed during IO finalization. | dgp | 2014-08-02 | 1 | -0/+3 |
| | | | |||||
| * | | Addition of a cast in tclWinFile.c to match types in a comparison, and | andreask | 2014-08-01 | 2 | -2/+2 |
| | | | | | | | | | | | fix of a TRACE string literal in tclExecute.c with a bogus escape. Both would otherwise bug a Windows debug build (where warnings are errors). | ||||
| * | | Updated package "platform" to version 1.0.13. Separated the | andreask | 2014-08-01 | 5 | -26/+29 |
| |\ \ | |/ | | | | | | | identifiers "macosx*-i386-x86_64" and "macosx-universal" as incompatible with each other. | ||||
| | * | Updated package "platform" to version 1.0.13. Separated the | andreask | 2014-08-01 | 6 | -27/+30 |
| | | | | | | | | | | | | | | | | | identifiers "macosx*-i386-x86_64" and "macosx-universal" as incompatible with each other. Plus addition of a cast in tclIO.c to match types in a comparison which otherwise bugs a Windows debug build (warnings are errors). | ||||
| * | | merge-mark | dgp | 2014-08-01 | 0 | -0/+0 |
| |\ \ | |/ | |||||
| | * | Backport the cheaper Preserve/Release on channels. | dgp | 2014-08-01 | 5 | -51/+77 |
| | |\ | |||||
| * | | | [e75fabac06] Teardown line information after failed ensemble subcommand | dgp | 2014-08-01 | 1 | -0/+18 |
| | | | | | | | | | | | | | compile attempts. Copy/paste from CompileCmdCompileProc(). May be refactor opportunity here. | ||||
| * | | | Correction and commentary in the details of ensemble compiling. | dgp | 2014-08-01 | 1 | -1/+17 |
| | | | | |||||
| * | | | Fix part of [d19aef1487] (broken --disable-load on MacOSX), broken by ↵ | jan.nijtmans | 2014-07-31 | 1 | -1/+3 |
| | | | | | | | | | | | | | [4d417791c9] | ||||
| * | | | [a84a720966] Protect against double free in TclOO classChainCache teardown. | dgp | 2014-07-31 | 1 | -1/+1 |
| | | | | |||||
| * | | | [3757cdf808] More clock refactoring with spooky impact on [string match] | dgp | 2014-07-30 | 1 | -10/+22 |
| | |/ |/| | | | performance. | ||||
| * | | Complete transformation off Tcl_Preserve() legacy onto ref counting. | dgp | 2014-07-30 | 1 | -15/+7 |
| | | | |||||
| * | | Reduce the cost of Preserve/Release on channels | dgp | 2014-07-30 | 5 | -43/+74 |
| | | | |||||
| * | | [3757cdf808] Refactoring seems to reduce (maybe somewhat kinda sorta) a | dgp | 2014-07-29 | 1 | -22/+32 |
| | | | | | | | bizarre performance regression. Still not what it was. Damn Heisenbugs. | ||||
| * | | An alternative implementation of [chan copy] that operates in precisely | dgp | 2014-07-28 | 1 | -7/+254 |
| |\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | those circumstances where the incumbent implementation is known to perform a byte-for-byte copy from input channel to output channel. In that case, the new implementation ceases the activity of copying every byte from buffer to buffer to buffer to buffer to effect the copy. Instead it arranges for the very same buffer the input channel read driver read into to be the buffer the output channel write driver writes out from. No data copying (or even examining) at all. When the proper conditions apply, this ought to be noticeably faster. | ||||
| | * \ | merge trunk | dgp | 2014-07-28 | 0 | -0/+0 |
| | |\ \ | |/ / |/| | | |||||
| * | | | merge-mark | jan.nijtmans | 2014-07-27 | 0 | -0/+0 |
| |\ \ \ | | |/ | |/| | |||||
| | * | | correct EOLFIX macro, broken by [a803608ed5] | jan.nijtmans | 2014-07-27 | 1 | -2/+2 |
| | | | | |||||
| | | * | Revise MBRead() to be used in both sync and async modes. | dgp | 2014-07-28 | 1 | -44/+34 |
| | | | | |||||
| | | * | Push MBWrite() differences out to callers. | dgp | 2014-07-27 | 1 | -27/+22 |
| | | | | |||||
| | | * | Rework MBWrite() so it can be used in both sync and async modes. | dgp | 2014-07-27 | 1 | -68/+41 |
| | | | | | | | | | | Reduce code duplication. | ||||
| | | * | Use common MBError() routine for sync and async operations. | dgp | 2014-07-26 | 1 | -25/+24 |
| | | | | |||||
| | | * | Extend the "move buffer" implementation to cover the async case. | dgp | 2014-07-26 | 1 | -11/+175 |
| | | | | |||||
| | | * | merge trunk | dgp | 2014-07-24 | 1 | -2/+1 |
| | | |\ | |_|/ |/| | | |||||
| * | | | [9d19af7adb] Make sure all timer events get canceled. | dgp | 2014-07-24 | 1 | -2/+1 |
| | | | | |||||
| | | * | simplify moveBytes selection logic | dgp | 2014-07-24 | 1 | -3/+1 |
| | | | | |||||
| | | * | Make sure MoveBytes error reporting reproduces what CopyData does. Bugward ↵ | dgp | 2014-07-24 | 1 | -0/+8 |
| | | | | | | | | | | | | | compatibility! | ||||
| | | * | merge trunk | dgp | 2014-07-24 | 1 | -0/+30 |
| | | |\ | |_|/ |/| | | |||||
| * | | | Test format of write error messages during binary [chan copy] | dgp | 2014-07-24 | 1 | -0/+30 |
| | | | | |||||
| | | * | Make sure MoveBytes records read error messages. | dgp | 2014-07-24 | 1 | -1/+5 |
| | | | | |||||
 |
index : tcl.git | |
| Tcl is a high-level, general-purpose, interpreted, dynamic programming language. It was designed with the goal of being very simple but powerful. |
| summaryrefslogtreecommitdiffstats |