aboutsummaryrefslogtreecommitdiff
Commit message (Expand)AuthorAgeFilesLines
* Backed out changeset f1c2880a6b49Carl Friedrich Bolz-Tereick2021-02-153-141/+5
* fix type-unsafe testCarl Friedrich Bolz-Tereick2020-04-281-5/+6
* better reasoning about upper bounds of or and xor, and about lower bounds of orCarl Friedrich Bolz-Tereick2020-04-293-5/+141
* make the JIT reason about int_invert and int_negCarl Friedrich Bolz-Tereick2020-04-265-0/+163
* kill unnecessary importsCarl Friedrich Bolz-Tereick2021-02-042-4/+2
* more of sameCarl Friedrich Bolz-Tereick2021-02-043-19/+37
* call more appropriate methods a bit everywhereCarl Friedrich Bolz-Tereick2021-02-043-58/+45
* use low-level interface where applicable, add at least a minimal hypothesis t...Carl Friedrich Bolz-Tereick2021-02-042-13/+21
* remove unreachable code(?)Carl Friedrich Bolz-Tereick2021-02-031-4/+0
* add a comment about something that I have to re-learn every time I read this ...Carl Friedrich Bolz-Tereick2020-04-291-0/+6
* remove copy-pasted commentCarl Friedrich Bolz-Tereick2020-04-291-2/+0
* express the make_le/ge/gt/lt methods in terms of their _const variantsCarl Friedrich Bolz-Tereick2020-04-291-8/+32
* optimize chains of longlong2float(float2longlong(x)) and vice versaCarl Friedrich Bolz-Tereick2021-01-312-0/+30
* fix a bug in the heapcache around nonstandard virtualizables (it doesn't occurCarl Friedrich Bolz-Tereick2021-01-183-11/+21
* merge default into branchMatti Picus2020-10-059-34/+717
|\
| * get rid of mentions of bitbucketMatti Picus2020-09-241-1/+1
| * Issue #3297Armin Rigo2020-09-104-27/+207
| * Disable the work done on optimize_UINT_LT & friends, with explanationArmin Rigo2020-09-032-8/+30
| * Some intbound checking around the 'uint' comparisonsArmin Rigo2020-09-022-0/+438
| * add jit_hooks.stats_memmgr_release_all()Armin Rigo2020-08-303-6/+49
* | fixes for metainterp/test/test_fficall.pyArmin Rigo2020-08-111-3/+4
* | fix type in rpython\jit\metainterp\test\test_virtual.pynulano2020-07-301-1/+1
|/
* Older versions of CPython 2.7.x (like 2.7.6) don't support this new syntaxArmin Rigo2020-07-171-6/+6
* Fixes for 560856d4f545: need to add an ARRAYMOVE operation in the jit, similarArmin Rigo2020-06-256-27/+122
* Use py3-style names for function attributesRonan Lamy2020-04-226-9/+9
* Modernize exec syntax in rpython/Ronan Lamy2020-04-186-44/+44
* Fix commentsArmin Rigo2020-04-111-2/+3
* fix test: one of the ptr_eq is no longer necessary, due to the improvedCarl Friedrich Bolz-Tereick2020-02-271-1/+1
* merge warmup-improvements-various:Carl Friedrich Bolz-Tereick2020-02-268-19/+267
|\
| * merge defaultCarl Friedrich Bolz-Tereick2020-02-223-0/+41
| |\
| * | when tracing, we know that newly allocated frames must be different from the ...Carl Friedrich Bolz-Tereick2020-02-173-2/+48
| * | more robust caching of quasiimmut_field when applied to *constants* (which isCarl Friedrich Bolz-Tereick2020-02-173-5/+85
| * | cache result of call_loopinvariant in the heapcacheCarl Friedrich Bolz-Tereick2020-02-164-4/+64
| * | fix testCarl Friedrich Bolz-Tereick2020-02-161-1/+1
| * | redo 8fcba04d307f: don't generate tons and tons of guard_not_invalidated whileCarl Friedrich Bolz-Tereick2020-02-165-7/+69
* | | merge defaultCarl Friedrich Bolz-Tereick2020-02-211-1/+1
|\ \ \ | | |/ | |/|
| * | fix translationCarl Friedrich Bolz-Tereick2020-02-211-1/+1
* | | merge defaultCarl Friedrich Bolz-Tereick2020-02-214-7/+50
|\| |
| * | count the number of ops that aren't even executed, because they are caught byCarl Friedrich Bolz-Tereick2020-02-213-0/+41
| |/
| * Backed out changeset 8fcba04d307fCarl Friedrich Bolz-Tereick2020-02-153-32/+2
| * don't generate tons and tons of guard_not_invalidated while tracingCarl Friedrich Bolz-Tereick2020-02-123-2/+32
| * use a linked list instead of a listCarl Friedrich Bolz-Tereick2020-02-121-7/+9
* | typoCarl Friedrich Bolz-Tereick2020-02-191-2/+2
|/
* Issue #3128Armin Rigo2019-12-192-1/+17
* in-progress: remove asmgcc completely (modern versions of gcc show testArmin Rigo2019-12-111-2/+1
* Revert 18443d3a74d5: the branch int-test-is-zero is maybe a nice ideaArmin Rigo2019-11-177-60/+0
* Make the usage of INT_TEST_xxx conditional on the backendArmin Rigo2019-11-161-9/+10
* Generate more compact code for bit tests using the TEST instruction:Armin Rigo2019-11-167-0/+59
* fix translation, tweak outputCarl Friedrich Bolz-Tereick2019-10-111-5/+7
* debug-print quasi-immutable invalidationsCarl Friedrich Bolz-Tereick2019-10-101-3/+9