| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
option to disable.
|
|
|
|
|
|
|
| |
performance.
* tools/tsdPerf.tcl A simplistic tool that uses the thread
extension and tsdPerf.so to get some performance metrics by,
simulating, simple TSD contention.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* generic/tclInt.h: Change around some function names and
add some new per-platform declarations for thread-specific data
functions.
* generic/tclThread.c: Make use of of the new function names
that no longer have a Tclp prefix.
* generic/tclThreadStorage.c: Replace the core thread-specific data
(TSD) mechanism with an array offset solution that eliminates the
hash tables, and only uses one slot of native TSD.
Many thanks to Kevin B. Kenny for his help with this.
* unix/tclUnixThrd.c: Add platform-specific TSD functions for use
by tclThreadStorage.c.
* win/tclWinThrd.c: Add platform-specific TSD functions for use
by tclThreadStorage.c.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Restored compilability of 'stitcher.exe' on vc2k5.
* cameraparams.h:
* compressor.cpp:
* imageDirectory.h:
* stitchEngine.h:
* stitcher.cpp:
Added computation of Brenner gradient and median of
several images to allow for computation of white balance
on the fly.
Replaced a lot of 'unsigned' quantities with 'int' to avoid
surprises on overflow.
Turned the images to run in X-major order again. This
requires a 64-bit machine for the larger images, but is
considerably more cache-friendly, and reduces stitch times
of the smaller images by nearly half.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
to delete the async handler (common trap). As, only the one
that created the handler is allowed to delete it.
|
|
|
|
|
| |
to locate handler token fails. Happens when some other
thread attempts to delete somebody else's token.
|
|
|
|
|
|
|
|
|
|
| |
implementation. Added test cases about how it handles if the rug
is pulled out from under a channel (= killing threads,
interpreters containing the tcl command for a channel, and channel
sitting in a different interpreter/thread.)
* generic/tclIORChan.c: Fixed the bugs exposed by the new
testcases, redone most of the cleanup and exit handling.
|
|
|
|
|
|
|
|
|
| |
* generic/tclTest.c: USE_OBSOLETE_FS_HOOKS, completing
* tests/ioCmd.test: the deprecation path for these
* tests/ioUtil.test (removed): obsolete interfaces. (Code was active
in Tcl 8.4, present but enabled only by customized compile switch in
Tcl 8.5, and now completely gone for Tcl 8.6). Also removed all tests
relevant only to the removed interfaces.
|
|
|
|
| |
Thanks to hat0 for spotting this.
|
|
|
|
|
|
|
|
| |
* generic/tclStubInit.c: export only module-scope pointers to
* generic/tclStubLib.c: the main stubs tables (for package
* tools/genStubs.tcl: initialization). [Patch 1938497]
* generic/tclBasic.c (Tcl_CreateInterp):
* generic/tclTomMathInterface.c (TclTommath_Init):
|
|
|
|
|
|
|
|
| |
* generic/tclNotify.c: module-scope hooks table instead of
* generic/tclStubInit.c: runtime stubs-table modification;
* macosx/tclMacOSXNotify.c: ensure all hookable notifier functions
* win/tclWinNotify.c: check for hooks; remove hook checks in
* unix/tclUnixNotfy.c: notifier API callers. [Patch 1938497]
|
|
|
|
|
|
|
| |
* io.test (io-53.8a): Ferrieux <ferrieux@users.sourceforge.net>,
* chanio.test (chan-io-53.8a): to shift EOF handling to the async
part of the command if a callback is specified, should the channel
be at EOF already when fcopy is called. Testcase by myself.
|
|
|
|
| |
of tclStubLib.o from TCL_OBJS. [Bug 1942795]
|
|
|
|
|
| |
Added comments to the test that it can fail on a heavily loaded
system.
|
|
|
|
| |
'struct timezone' in the call to 'gettimeofday'. [Bug 1942197].
|
|
|
|
|
|
|
|
|
|
|
|
| |
values, changed to not be an error, but behave like the special
value -1 (copy all, default).
* tests/iocmd.test (iocmd-15.{12,13}): Removed.
* tests/io.test (io-52.5{,a,b}): Reverted last change, added
* tests/chanio.test (chan-io-52.5{,a,b}): comment regarding the
meaning of -1, added two more testcases for other negative values,
and input wrapped to negative.
|
| |
|
| |
|
|
|
|
|
| |
* tests/io.test (io-52.5): does not seem to have any bearing, and
was an illegal value.
|
|
|
|
|
|
|
|
| |
* tests/ioCmd.test (iocmd-15.{13,14}): value to reject negative
values, and values overflowing 32-bit signed. [Bug 1557855]. Basic
patch by Alexandre Ferrieux <ferrieux@users.sourceforge.net>, with
modifications from me to separate overflow from true negative
value. Extended testsuite.
|
|
|
|
| |
* tests/io.test (io-53.8,53.9,53.10): spaces in builddir path
|
|
|
|
| |
used in GrowEvaluationStack() and friends.
|
|
|
|
| |
*StubsPtr as EXTERN instead of extern.
|
|
|
|
|
|
|
|
| |
* tests/chanio.test:
* generic/tclIO.c: Additional changes to data structures for fcopy
* generic/tclIO.h: and channels to perform proper cleanup in case
of a channel having two background copy operations running as is
now possible.
|
| |
|
|
|
|
|
|
| |
TclCopyChannel): New macro, and the places using it. This change
allows for bi-directional fcopy on channels. Thanks to Alexandre
Ferrieux <ferrieux@users.sourceforge.net> for the patch.
|
|
|
|
|
|
| |
Fix [format {% d}] so that it behaves the same way as in 8.4 and
as C's printf().
* tests/format.test: Add a test for '% d' and '%+d'.
|
|
|
|
|
|
|
|
|
|
|
| |
* tests/io.test (io-53.9): Made test cleanup robust against
the possibility of slow process shutdown on Windows.
* win/tcl.m4: Added -D_CRT_SECURE_NO_DEPRECATE and
-DCRT_NONSTDC_NO_DEPRECATE to the MSVC compilation flags
so that the compilation doesn't barf on perfectly reasonable
Posix system calls.
* win/configure: Manually patched (don't have the right autoconf
to hand).
|
|
|
|
|
|
|
|
|
|
|
| |
Tcl was creating an NTFS junction point (IO_REPARSE_TAG_MOUNT_POINT)
but filling in the union member for a Vista symbolic link.
We had gotten away with this error because the union member
(SymbolicLinkReparseBuffer) was misdefined in this file
and in the 'winnt.h' in early versions of MinGW. MinGW 3.4.2
has the correct definition of SymbolicLinkReparseBuffer, exposing
the mismatch, and making tests cmdAH-19.4.1, fCmd-28.*, and
filename-11.* fail.
|
|
|
|
|
|
| |
* tests/chanio.test: on Alexandre's test script. Also fixed
problem with timer in preceding test, was not canceled properly
in the ok case.
|
|
|
|
|
|
|
|
| |
write when input was zero-length anyway. Otherwise keept it an
error, and separate the message from 'written too much'.
* tests/ioCmd.test (iocmd-24.6): Testcase updated for changed
message.
|
|
|
|
|
|
|
|
|
| |
the now closed channel from the reflection map. Before we could
crash the system by invoking 'chan postevent' on a closed
reflected channel, dereferencing the dangling pointer in the
map.
* tests/ioCmd.test (iocmd-31.8): Testcase for the above.
|
|
|
|
|
|
| |
* tests/io.test: prevent fcopy from calling -command synchronously
* tests/chanio.test: the first time. Thanks to Alexandre Ferrieux
<ferrieux@users.sourceforge.net> for report and patch.
|
|
|
|
|
|
| |
* tools/genStubs.tcl: avoid exporting them from libraries that link
with -ltclstub; constify tcl*StubsPtr and stub
table hook pointers. [Bug 1819422]
|
|
|
|
|
|
|
| |
[Bug 780533], with many thanks to Alexandre Ferrieux
<ferrieux@users.sourceforge.net> for tracking it down and
providing a solution. Still have to convert his test script into
a proper test case.
|
|
|
|
|
|
|
|
| |
* tools/genStubs.tcl: creation so that "tclStubsPtr" is not present
* unix/Makefile.in: in libtcl.so, but is present only in
* win/Makefile.in: libtclstub.a. This tightens up the rules for
* win/makefile.bc: users of the stubs interfaces. [Bug 1819422]
* win/makefile.vc:
|
|
|
|
|
|
|
| |
* tools/genStubs.tcl: creation so that "tclStubsPtr" is not present
* unix/Makefile.in: in libtcl.so, but is present only in
* win/Makefile.in: libtclstub.a. This tightens up the rules for
users of the stubs interfaces. [Bug 1819422]
|
| |
|
|
|
|
|
| |
* unix/tcl.m4: rounding setup on solaris x86, native cc), provided
by Michael Schlenker.
|
|
|
|
| |
* generic/tclPkg.c (Tcl_PkgInitStubsCheck):
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* generic/tcl.h:
* library/init.tcl:
* macosx/Tcl-Common.xcconfig:
* macosx/Tcl.pbproj/default.pbxuser:
* macosx/Tcl.pbproj/project.pbxproj:
* tools/tcl.wse.in:
* unix/configure.in:
* unix/tcl.spec:
* win/README:
* win/configure.in:
* win/makefile.bc:
* win/tcl.m4:
* unix/configure: autoconf-2.59
* win/configure:
* generic/tclBasic.c: Revised stubs-generation tool and interp
* tools/genStubs.tcl: creation so that "tclStubsPtr" is not present
* unix/Makefile.in: in libtcl.so, but is present only in
libtclstub.a. This tightens up the rules for users of the stubs
interfaces. [Bug 1819422]
* generic/tclDecls.h: make genstubs
* generic/tclIntDecls.h:
* generic/tclIntPlatDecls.h:
* generic/tclPlatDecls.h:
* generic/tclTomMathDecls.h:
|
|
|
|
| |
locate isnan() after the above change.
|
|
|
|
|
|
|
|
|
| |
* unix/configure.in: Added code to the configurator to check for
a standard isnan() macro and use it if one
is found. This change avoids bugs where
the test of ((d) != (d)) is optimized away
by an overaggressive compiler. [Bug 1783544]
* unix/configure: autoconf-2.61
|
|
|
|
|
| |
to deal with (slightly buggy) math libraries in which pow()
returns an incorrectly rounded result. [Bug 1808174]
|
|
|
|
|
|
|
|
|
|
|
| |
* library/init.tcl:
* tools/tcl.wse.in:
* unix/configure.in:
* unix/tcl.spec:
* win/configure.in:
* unix/configure: autoconf-2.59
* win/configure:
|