tclquadcode

Timeline
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

50 ancestors of [90e908dae3]

2019-01-14
03:46
Further development of varargs. Note that the invocation sequence is much, much simpler than it used to be, so 'invoke.tcl' is no more. check-in: 90e908dae3 user: kbk tags: notworking, kbk-refactor-callframe
2019-01-13
15:43
Clean out dead 'exists.tcl' source check-in: f283d28ebd user: kbk tags: notworking, kbk-refactor-callframe
15:38
Merge the (not-working) vararg reform branch. It appears that both these tasks need to be attacked at the same time because the changes are tightly interwoven. check-in: 05c93c9cc5 user: kbk tags: notworking, kbk-refactor-callframe
2019-01-09
01:52
Add support for compiled procs that alter the callframe, then fail. check-in: 30d6a54fdb user: kbk tags: notworking, kbk-refactor-callframe
2019-01-08
04:17
Fix more trivial errors in 'translate' and missing cases in code gen check-in: fba0ec7241 user: kbk tags: notworking, kbk-refactor-callframe
2019-01-05
22:48
Generate only one moveFromCallFrame per bytecode instruction check-in: 9dbaca7922 user: kbk tags: notworking, kbk-refactor-callframe
2019-01-02
02:20
Oops, one extra callframe ref in 'callframe-error-quads' check-in: cdc4d10e86 user: kbk tags: notworking, kbk-refactor-callframe
2019-01-01
23:06
Update quadcode generation and documentation so that 'direct' operations refer to the callframe (which is necessary in the event that they are emitted to deal with double-dereference). CODE ISSUER IS NOT YET UPDATED TO COPE WITH THE CHANGE. check-in: 104debc734 user: kbk tags: notworking, kbk-refactor-callframe
21:57
Move callframe moves to the actual points where callframe values are moved. They can then be optimized by avaiability and liveness analysis. check-in: dbd9571178 user: kbk tags: notworking, kbk-refactor-callframe
2018-12-29
21:00
Keep dictionary iterations out of the callframe - they can never be addressed by upvar check-in: ea7b788fb4 user: kbk tags: kbk-refactor-callframe
20:26
Open a branch for some experiments on refactoring callframe operations check-in: 7454a5228c user: kbk tags: kbk-refactor-callframe
2018-12-27
19:47
Add 'buiilder.tcl' inadvertently omitted from previous commit Closed-Leaf check-in: 2e3b3ea76f user: kbk tags: notworking, kbk-refactor-varargs
19:44
Very rough beginning of a refactored 'varargs' pass check-in: 7c9d400e5e user: kbk tags: notworking, kbk-refactor-varargs
04:12
Eliminate the 'isBoolean' instruction in favour of using the type checking machinery Leaf check-in: 0718166269 user: kbk tags: trunk
04:09
Eliminate the 'isBoolean' opcode in favour of 'instanceOf', to allow for type inferemce based on tryCvtToBoolean Closed-Leaf check-in: bd009801ca user: kbk tags: kbk-isBoolean
03:53
Open a branch for rework of 'isBoolean' so that it informs type inference check-in: 2a34883328 user: kbk tags: kbk-isBoolean
2018-12-18
15:12
OOPS: remove 'source' of unused file check-in: 3bf74c48dc user: kbk tags: trunk
02:32
Add a long-forgotten 'specializer.md' discussing what the specializer does. check-in: 7bd4d23ac9 user: kbk tags: trunk
2018-12-17
23:19
Integrate kbk-jumpthread: replace the node-by-node splitting with a single pass that identifies many threading opportunities and also reduces the number of splits. Eliminate the old nodesplit pass, and the renameTemps pass, which is no longer required. check-in: a934a75e1f user: kbk tags: trunk
23:13
result, returnCode, returnOptions must be split into FAIL and non-FAIL paths because the backend isn't prepared to deal with all combinations of FAIL + someOtherType. Closed-Leaf check-in: 94358b53ea user: kbk tags: kbk-jumpthread
22:08
Finish jump threading - actually do the block duplication and redirection of jumps. Add the logic for SSA deconstruction (required by jump threading) and make SSA construction work with the deconstructed result. check-in: a41b93130e user: kbk tags: notworking, kbk-jumpthread
2018-12-16
05:18
Add constant folding for unary +. It probably ought to be folded out altogether, since it serves only to trigger type checking.    Add an outline of the actual process of jump threading once the threads have been determined. check-in: c388c8737b user: kbk tags: notworking, kbk-jumpthread
04:55
Add a 'cos2' test case to illustrate the cost of non-numeric ordering comparisons check-in: 86167d6917 user: kbk tags: notworking, kbk-jumpthread
04:43
Finish forward analysis in the jump threading pass. check-in: dc90afa15d user: kbk tags: notworking, kbk-jumpthread
2018-12-10
04:51
Remove vestiges of the old node splitter from 'inline.tcl'. Make console dribble in 'jumpthread.tcl' contingent on debug-jumpthread. check-in: fd2ea3e6f1 user: kbk tags: notworking, kbk-jumpthread
04:09
OOPS! Didn't add the new 'jumpthread.tcl' module! Remove the old 'nodesplit.tcl', and remove the 'renameTemps' pass since 'nodesplit' was the only thing that ever depended on it. check-in: 67b989853b user: kbk tags: notworking, kbk-jumpthread
01:34
Calculation of anticipable tests for jump threading check-in: f138663832 user: kbk tags: notworking, kbk-jumpthread
2018-12-09
21:52
Start with the most basic framework for invoking the jump threading pass. check-in: 6bdc33061b user: kbk tags: kbk-jumpthread
20:55
Open a branch for experiments with more accurate and faster jump threading. check-in: 67902a50a2 user: kbk tags: kbk-jumpthread
2018-12-08
21:46
Add a micropass to optimize away conditional jumps that are identical to a conditional jump in a dominator. (Partial redundancy elimination appears to create these.) check-in: 344567b919 user: kbk tags: trunk
21:45
Add a micropass to optimize away conditional jumps that are identical to a conditional jump in a dominator. (Partial redundancy elimination appears to create these.) Closed-Leaf check-in: 1c9b4510d1 user: kbk tags: kbk-deadcond
2018-12-07
02:43
Fixes that make poly1305 compilable. check-in: 602b3659c7 user: kbk tags: trunk
02:23
Increase recursion limit to allow compilation of longer procedures. Correct a bug in expandInlines that left dangling references to procedure args. Correct a bug in analyzeTypes where the INT/ENTIER break was set to int32 rather than int64. Add shift operations to constfold. Improve auditing. Merge trunk. Leaf check-in: 023d0828f0 user: kbk tags: poly1305
02:19
Increase recursion limit to allow compilation of longer procedures. Correct a bug in expandInlines that left dangling references to procedure args. Correct a bug in analyzeTypes where the INT/ENTIER break was set to int32 rather than int64. Add shift operations to constfold. Improve auditing. Merge trunk. Closed-Leaf check-in: 1984824c19 user: kbk tags: mistake
2018-12-06
21:17
merge trunk check-in: c2a6505ae9 user: dkf tags: poly1305
03:15
Merge kbk-pre - add the optimizations of loop inversion (enables loop-invariant code motion) and partial redundancy elimination, and fix multiple bugs exposed by these optimizations. check-in: 0e06123e97 user: kbk tags: trunk
03:13
Array initialization cannot be hoisted because every array needs its own initArray. Code insertion must ignore operations that it has already inserted, or infinite loops can result. All tests pass! Closed-Leaf check-in: cf87d02677 user: kbk tags: kbk-pre
01:42
Remove speculative phis if they turned out not to be useful check-in: bd6294fade user: kbk tags: notworking, kbk-pre
2018-12-05
05:37
Rewrite pre_insert and pre_phi_translate to NOT use a persistent cache. The missing piece in VanD04 is that pre_insert must call pre_phi_translate for all the anticipated expressions first, and then use the dictionary that results for the translated values. There are still further bugs, but we're over this hump at least. check-in: 0bbe5a5dd3 user: kbk tags: notworking, kbk-pre
2018-12-03
05:08
Stop constant folding from leaving dead code behind, Add a test for simple nested iterations, using [lmap]. Temporarily patch 'foreach' operations from being hoistable - I don't think this will be necessary, but it's tickling other bugs. Make translation of values across a phi work if one of the inputs to the phi is a literal. Put 'bbidom' and 'bblevel' directly after dead code elimination, because virtually everything depends on having dominators, which deadcode destroys. check-in: 4f50ed77b2 user: kbk tags: notworking, kbk-pre
01:06
Fix crashing bug in creating variable name for a phi operation. Fix bug where partial redundancy elimination got stale dominators, causing much weirdness. Add additional tracing to available expression calculation. check-in: ce1088b7a7 user: kbk tags: notworking, kbk-pre
2018-12-02
23:27
Turn on partial redundancy elimination. Correct bug in availability analysis that led to infinite loop in 'msrange2' check-in: 68da56cfee user: kbk tags: notworking, kbk-pre
2018-12-01
17:37
Restructure pass management in 'tidy'. Add constant folding of a few more operations. check-in: e2750cb57e user: kbk tags: kbk-pre
2018-11-29
00:56
merge trunk check-in: 815387c202 user: kbk tags: notworking, kbk-pre
2018-11-27
03:07
Fix bug in finding variable names among the args to 'regexp' check-in: 6311170b65 user: kbk tags: trunk
2018-11-26
04:57
Added a couple of tests to trigger loop-invariant code motion.    Corrected du-chain mismanagement in callframe.tcl (deleting moveToCallFrame), copyprop.tcl (deleting any quads), uselessphis (deleting anything), and narrow.tcl (cleanupNarrow, deleting anything). NOT WORKING: During node splitting, at least 'flightawarebench', 'msrange', 'bug7c59', and 'cse-caller' are still coming out with mismanaged chains.    Improved auditing in the 'tidy' pass of the transformer, and made it less aggressive about rechecking everything. NOT WORKING: in the 'wordcounter2' test case, it's missing at least one invocation of 'initArrayIfNotExists' because we're falling into the code gen in 'compile.tcl' for initArrayIfNotExists(ARRAY).    Advanced partial redundancy elimination to 'code complete' status. The stuff surrounding it is in bad enough shape at the moment that it is far from tested, and known not to be working for everything, but is patched out at the moment. check-in: 02ab4301aa user: kbk tags: notworking, kbk-pre
2018-11-23
18:42
Back out from quite a long false path - switch over from Simpson's modified SSAPRE to VanDrunen's GVNPRE. Implemented the two phases of BUILDSETS - results look correct for a couple of simple tests. check-in: e69362f376 user: kbk tags: kbk-pre
2018-11-13
02:57
Code complete for PRE - except that I don't know how to introduce a phi if PRE has made an expression available on two afferent flow graph edges rather than in a dominator. check-in: fa9f7e0eb3 user: kbk tags: notworking, kbk-pre
2018-11-12
16:38
Tidy up commentary and calculation of INSERT and DELETE. Build the place to start code rewriting, and add some methods to start on renaming variables wholesale. check-in: e901ec10df user: kbk tags: kbk-pre
01:38
Change signatures in preparation for code rewriting: include enough information in signatures so that INSERT can reconstruct an instruction check-in: c098311242 user: kbk tags: kbk-pre
2018-11-11
22:34
Add bitwise Boolean operations to constant folding. Modify partial redundancy elimination to use bit vectors for dataflow calculations and debug the dataflow equations. check-in: 71a4e793a1 user: kbk tags: kbk-pre