[pypy-commit] pypy default: catching up with the backend changes load_from_gc_table. stuffed the gc references before the constant pool, which makes it easy to access them using r13

Author: Richard Plangger Branch: Changeset: r83609:2fd0d166512f Date: 2016-04-12 11:08 +0200 http://bitbucket.org/pypy/pypy/changeset/2fd0d166512f/ Log:catching up with the backend changes load_from_gc_table. stuffed the gc references before the constant pool,

[pypy-commit] pypy default: wrong parameter to LG command and wrong jump target to bridge

Author: Richard Plangger Branch: Changeset: r83610:b55e5e29e2f0 Date: 2016-04-12 11:47 +0200 http://bitbucket.org/pypy/pypy/changeset/b55e5e29e2f0/ Log:wrong parameter to LG command and wrong jump target to bridge diff --git a/rpython/jit/backend/zarch/assembler.py

[pypy-commit] pypy new-jit-log: use tmpdir instead of tempfile, added a version to the jitlog

Author: Richard Plangger Branch: new-jit-log Changeset: r83614:68ab27b22964 Date: 2016-04-12 14:56 +0200 http://bitbucket.org/pypy/pypy/changeset/68ab27b22964/ Log:use tmpdir instead of tempfile, added a version to the jitlog diff --git

[pypy-commit] pypy default: (s390x) execute patch_stack_checks before materialize loop, translation issue fixed

Author: Richard Plangger Branch: Changeset: r83617:4fd5a8562c39 Date: 2016-04-12 16:23 +0200 http://bitbucket.org/pypy/pypy/changeset/4fd5a8562c39/ Log:(s390x) execute patch_stack_checks before materialize loop, translation issue fixed diff --git

[pypy-commit] pypy default: (s390x) jumping to the wrong address

Author: Richard Plangger Branch: Changeset: r83618:c893449515d6 Date: 2016-04-12 16:44 +0200 http://bitbucket.org/pypy/pypy/changeset/c893449515d6/ Log:(s390x) jumping to the wrong address diff --git a/rpython/jit/backend/zarch/assembler.py

[pypy-commit] pypy new-jit-log: writing core dump just after each resoperation, added test to check that

Author: Richard Plangger Branch: new-jit-log Changeset: r83115:24a3cc49b4ee Date: 2016-03-17 19:28 +0100 http://bitbucket.org/pypy/pypy/changeset/24a3cc49b4ee/ Log:writing core dump just after each resoperation, added test to check that diff --git

[pypy-commit] pypy fix-longevity: catchup with default

Author: Richard Plangger Branch: fix-longevity Changeset: r83087:f64a2c834041 Date: 2016-03-15 11:52 +0100 http://bitbucket.org/pypy/pypy/changeset/f64a2c834041/ Log:catchup with default diff too long, truncating to 2000 out of 17154 lines diff --git a/.hgignore

[pypy-commit] pypy gcstress-hypothesis: generating control flow using basic blocks, need conditions and loops

Author: Richard Plangger Branch: gcstress-hypothesis Changeset: r83086:65a4e92ce40f Date: 2016-03-15 11:31 +0100 http://bitbucket.org/pypy/pypy/changeset/65a4e92ce40f/ Log:generating control flow using basic blocks, need conditions and loops diff --git

[pypy-commit] buildbot default: added a new buildbot to test if this is a machine config problem

Author: Richard Plangger Branch: Changeset: r995:5b2d6b27aacf Date: 2016-03-16 17:59 +0100 http://bitbucket.org/pypy/buildbot/changeset/5b2d6b27aacf/ Log:added a new buildbot to test if this is a machine config problem diff --git a/bot2/pypybuildbot/master.py

[pypy-commit] pypy jit-leaner-frontend: removed statements that copy rd_* on guard descr. test_vector passing

Author: Richard Plangger Branch: jit-leaner-frontend Changeset: r83242:721744197b8f Date: 2016-03-22 09:59 +0100 http://bitbucket.org/pypy/pypy/changeset/721744197b8f/ Log:removed statements that copy rd_* on guard descr. test_vector passing diff --git

[pypy-commit] pypy new-jit-log: translation issue

Author: Richard Plangger Branch: new-jit-log Changeset: r83263:f4455cafdb2d Date: 2016-03-22 18:45 +0100 http://bitbucket.org/pypy/pypy/changeset/f4455cafdb2d/ Log:translation issue diff --git a/rpython/jit/backend/x86/assembler.py

[pypy-commit] pypy new-jit-log: catchup with default

Author: Richard Plangger Branch: new-jit-log Changeset: r83262:3ad35c51e7c7 Date: 2016-03-22 17:10 +0100 http://bitbucket.org/pypy/pypy/changeset/3ad35c51e7c7/ Log:catchup with default diff --git a/.hgtags b/.hgtags --- a/.hgtags +++ b/.hgtags @@ -19,3 +19,4 @@

[pypy-commit] pypy release-5.x: calculate index value (const scale, const offset) before emiting the load for ConstPtrs

Author: Richard Plangger Branch: release-5.x Changeset: r83315:30178fbdc748 Date: 2016-03-01 08:32 +0100 http://bitbucket.org/pypy/pypy/changeset/30178fbdc748/ Log:calculate index value (const scale, const offset) before emiting the load for ConstPtrs diff --git

[pypy-commit] pypy new-jit-log: test to generate a jitlog

Author: Richard Plangger Branch: new-jit-log Changeset: r83331:a3cbc9a1c5d6 Date: 2016-03-24 17:53 +0100 http://bitbucket.org/pypy/pypy/changeset/a3cbc9a1c5d6/ Log:test to generate a jitlog diff --git a/rpython/jit/backend/x86/test/test_jitlog.py

[pypy-commit] pypy jit-leaner-frontend: test_vector passing again. mostly issues with methods moved to rop from ResOperation

Author: Richard Plangger Branch: jit-leaner-frontend Changeset: r83348:472192cef9c7 Date: 2016-03-25 12:38 +0100 http://bitbucket.org/pypy/pypy/changeset/472192cef9c7/ Log:test_vector passing again. mostly issues with methods moved to rop from ResOperation diff

[pypy-commit] pypy jit-leaner-frontend: remove descr after parse_loop in assert_vectorize, this is need to successfully convert the loop to a trace obj

Author: Richard Plangger Branch: jit-leaner-frontend Changeset: r83343:2bbf906956f4 Date: 2016-03-25 11:08 +0100 http://bitbucket.org/pypy/pypy/changeset/2bbf906956f4/ Log:remove descr after parse_loop in assert_vectorize, this is need to successfully convert the

[pypy-commit] pypy jit-leaner-frontend: adapting the test interface for test_vecopt

Author: Richard Plangger Branch: jit-leaner-frontend Changeset: r83342:5e609847041f Date: 2016-03-22 17:09 +0100 http://bitbucket.org/pypy/pypy/changeset/5e609847041f/ Log:adapting the test interface for test_vecopt diff --git

[pypy-commit] pypy jit-leaner-frontend: instead of invoking _do_optimize_loop, optimize_vector is called instead, the additional other optimizations are not necessary

Author: Richard Plangger Branch: jit-leaner-frontend Changeset: r83344:e5c868861aa2 Date: 2016-03-25 12:09 +0100 http://bitbucket.org/pypy/pypy/changeset/e5c868861aa2/ Log:instead of invoking _do_optimize_loop, optimize_vector is called instead, the additional

[pypy-commit] pypy jit-leaner-frontend: missing argument, now optimizeopt tests all pass

Author: Richard Plangger Branch: jit-leaner-frontend Changeset: r83347:c8d47f78736b Date: 2016-03-25 12:21 +0100 http://bitbucket.org/pypy/pypy/changeset/c8d47f78736b/ Log:missing argument, now optimizeopt tests all pass diff --git

[pypy-commit] pypy s390x-z196: close branch

Author: Richard Plangger Branch: s390x-z196 Changeset: r83219:2c67ae9165f5 Date: 2016-03-21 09:00 +0100 http://bitbucket.org/pypy/pypy/changeset/2c67ae9165f5/ Log:close branch ___ pypy-commit mailing list

[pypy-commit] pypy new-jit-log: class (const) attributes moved to outer scope

Author: Richard Plangger Branch: new-jit-log Changeset: r83218:8022b7c95b47 Date: 2016-03-21 08:59 +0100 http://bitbucket.org/pypy/pypy/changeset/8022b7c95b47/ Log:class (const) attributes moved to outer scope diff --git a/rpython/jit/backend/x86/assembler.py

[pypy-commit] pypy new-jit-log: added patch mark (rewrite machinecode). some refactoring to parse it more easily

Author: Richard Plangger Branch: new-jit-log Changeset: r83221:0b922ecc3d86 Date: 2016-03-21 16:55 +0100 http://bitbucket.org/pypy/pypy/changeset/0b922ecc3d86/ Log:added patch mark (rewrite machinecode). some refactoring to parse it more easily diff --git

[pypy-commit] pypy new-jit-log: translation issues

Author: Richard Plangger Branch: new-jit-log Changeset: r83220:9bd952847a5f Date: 2016-03-21 09:50 +0100 http://bitbucket.org/pypy/pypy/changeset/9bd952847a5f/ Log:translation issues diff --git a/rpython/jit/metainterp/jitlog.py b/rpython/jit/metainterp/jitlog.py ---

[pypy-commit] pypy new-jit-log: encode the offset into the core dump

Author: Richard Plangger Branch: new-jit-log Changeset: r83224:d96b7e680539 Date: 2016-03-21 18:39 +0100 http://bitbucket.org/pypy/pypy/changeset/d96b7e680539/ Log:encode the offset into the core dump diff --git a/rpython/jit/metainterp/jitlog.py

[pypy-commit] pypy release-5.x: macros (e.g. WCOREDUMP) got parameter type Signed, on little endian this does not make a difference, but it does on big endian. changed to rffi.INT

Author: Richard Plangger Branch: release-5.x Changeset: r83393:78860f97739e Date: 2016-01-22 14:56 +0100 http://bitbucket.org/pypy/pypy/changeset/78860f97739e/ Log:macros (e.g. WCOREDUMP) got parameter type Signed, on little endian this does not make a

[pypy-commit] pypy release-5.x: call int should provide rffi.INT instead of rffi.SIGNED, this works on little endian, but not big

Author: Richard Plangger Branch: release-5.x Changeset: r83400:403d1e816438 Date: 2016-02-05 20:46 +0100 http://bitbucket.org/pypy/pypy/changeset/403d1e816438/ Log:call int should provide rffi.INT instead of rffi.SIGNED, this works on little endian, but not big

[pypy-commit] pypy release-5.x: fixed callsite of clibffi with the same big endian issues as found yesterday evening

Author: Richard Plangger Branch: release-5.x Changeset: r83399:c8ccc2e362ae Date: 2016-01-22 12:51 +0100 http://bitbucket.org/pypy/pypy/changeset/c8ccc2e362ae/ Log:fixed callsite of clibffi with the same big endian issues as found yesterday evening (grafted from

[pypy-commit] pypy release-5.x: ffi call fixed in deprecated api that is still used (fix before that was not sufficient), fixed legacy tests test_libffi

Author: Richard Plangger Branch: release-5.x Changeset: r83401:d1470cfc1db9 Date: 2016-02-05 22:37 +0100 http://bitbucket.org/pypy/pypy/changeset/d1470cfc1db9/ Log:ffi call fixed in deprecated api that is still used (fix before that was not sufficient), fixed

[pypy-commit] pypy release-5.x: advancing the pointer to correctly return the value written from libffi, the reason why this does not happen on e.g. ppc bigendian 64 bit is: libffi ppc casts the retur

Author: Richard Plangger Branch: release-5.x Changeset: r83396:9d2afadd0b72 Date: 2016-01-21 20:37 +0100 http://bitbucket.org/pypy/pypy/changeset/9d2afadd0b72/ Log:advancing the pointer to correctly return the value written from libffi, the reason why this does

[pypy-commit] pypy release-5.x: half way through the ndarray tests (endian issues)

Author: Richard Plangger Branch: release-5.x Changeset: r83397:a2399bd7c38d Date: 2016-01-22 10:56 +0100 http://bitbucket.org/pypy/pypy/changeset/a2399bd7c38d/ Log:half way through the ndarray tests (endian issues) (grafted from

[pypy-commit] pypy release-5.x: and the other part of the bigendian issues (micronumpy tests)

Author: Richard Plangger Branch: release-5.x Changeset: r83398:d48bab9286fe Date: 2016-01-22 12:00 +0100 http://bitbucket.org/pypy/pypy/changeset/d48bab9286fe/ Log:and the other part of the bigendian issues (micronumpy tests) (grafted from

[pypy-commit] pypy release-5.x: _rawffi bigendian issue in callbacks, callback writes narrow integer directly to MSB of 64 bit value on s390x (wrong when value is passed along)

Author: Richard Plangger Branch: release-5.x Changeset: r83402:3624c8c2c3be Date: 2016-02-10 18:07 +0100 http://bitbucket.org/pypy/pypy/changeset/3624c8c2c3be/ Log:_rawffi bigendian issue in callbacks, callback writes narrow integer directly to MSB of 64 bit

[pypy-commit] pypy new-jit-log: added jitlog to the existing call sites of the old logger

Author: Richard Plangger Branch: new-jit-log Changeset: r83118:df3f2f49084a Date: 2016-03-18 08:18 +0100 http://bitbucket.org/pypy/pypy/changeset/df3f2f49084a/ Log:added jitlog to the existing call sites of the old logger diff --git

[pypy-commit] pypy s390x-enhance-speed: partly enabling the literal pool. it is now simpler and does not allocate 32bit values

Author: Richard Plangger Branch: s390x-enhance-speed Changeset: r82843:9c1e430cbed5 Date: 2016-03-07 12:00 +0100 http://bitbucket.org/pypy/pypy/changeset/9c1e430cbed5/ Log:partly enabling the literal pool. it is now simpler and does not allocate 32bit values

[pypy-commit] pypy s390x-backend: updated s390x docu

Author: Richard Plangger Branch: s390x-backend Changeset: r82844:a144c706eae1 Date: 2016-03-07 12:18 +0100 http://bitbucket.org/pypy/pypy/changeset/a144c706eae1/ Log:updated s390x docu diff --git a/rpython/doc/s390x.rst b/rpython/doc/s390x.rst ---

[pypy-commit] pypy s390x-backend: jump location was off, shift by parameter of sradi is weird (but it works now)

Author: Richard Plangger Branch: s390x-backend Changeset: r82874:78f8d539bef6 Date: 2016-03-08 09:25 +0100 http://bitbucket.org/pypy/pypy/changeset/78f8d539bef6/ Log:jump location was off, shift by parameter of sradi is weird (but it works now) diff --git

[pypy-commit] pypy gcstress-hypothesis: block of bytecodes is now supported, it seems that some cases are generated that trash the constraint max_size of the initial stack is 0

Author: Richard Plangger Branch: gcstress-hypothesis Changeset: r82770:4dbf54a3736b Date: 2016-03-04 15:58 +0100 http://bitbucket.org/pypy/pypy/changeset/4dbf54a3736b/ Log:block of bytecodes is now supported, it seems that some cases are generated that trash the

[pypy-commit] pypy gcstress-hypothesis: added some tests to ensure bytecode generation considers the top elements on the stack

Author: Richard Plangger Branch: gcstress-hypothesis Changeset: r82769:13234c967ffc Date: 2016-03-04 10:40 +0100 http://bitbucket.org/pypy/pypy/changeset/13234c967ffc/ Log:added some tests to ensure bytecode generation considers the top elements on the stack

[pypy-commit] pypy gcstress-hypothesis: creating a list of byte codes using hypothesis. the state along each instruction is passed using shared

Author: Richard Plangger Branch: gcstress-hypothesis Changeset: r82768:5ae35a1cd368 Date: 2016-03-03 10:52 +0100 http://bitbucket.org/pypy/pypy/changeset/5ae35a1cd368/ Log:creating a list of byte codes using hypothesis. the state along each instruction is passed

[pypy-commit] pypy s390x-backend: catchup with default

Author: Richard Plangger Branch: s390x-backend Changeset: r82807:9ff14f05a92d Date: 2016-03-06 10:24 +0100 http://bitbucket.org/pypy/pypy/changeset/9ff14f05a92d/ Log:catchup with default diff --git a/pypy/doc/contributor.rst b/pypy/doc/contributor.rst ---

[pypy-commit] pypy s390x-backend: fixed some tests

Author: Richard Plangger Branch: s390x-backend Changeset: r82806:f563f1ea7c20 Date: 2016-03-06 10:21 +0100 http://bitbucket.org/pypy/pypy/changeset/f563f1ea7c20/ Log:fixed some tests diff --git a/rpython/jit/backend/llsupport/test/test_gc_integration.py

[pypy-commit] pypy s390x-backend: added whats new entry for memop-simplify3

Author: Richard Plangger Branch: s390x-backend Changeset: r82810:71943846762a Date: 2016-03-06 12:01 +0100 http://bitbucket.org/pypy/pypy/changeset/71943846762a/ Log:added whats new entry for memop-simplify3 diff --git a/pypy/doc/whatsnew-head.rst

[pypy-commit] pypy memop-simplify3: close branch

Author: Richard Plangger Branch: memop-simplify3 Changeset: r82809:1cd36e0809b5 Date: 2016-03-06 11:58 +0100 http://bitbucket.org/pypy/pypy/changeset/1cd36e0809b5/ Log:close branch ___ pypy-commit mailing list

[pypy-commit] pypy s390x-backend: reverted x86 assembler (malloc_cond_varsize), related to the issue with bytesize and length

Author: Richard Plangger Branch: s390x-backend Changeset: r82808:572470f1b45a Date: 2016-03-06 11:57 +0100 http://bitbucket.org/pypy/pypy/changeset/572470f1b45a/ Log:reverted x86 assembler (malloc_cond_varsize), related to the issue with bytesize and length diff

[pypy-commit] pypy s390x-backend: fixed messed up whatsnew entry by merge...

Author: Richard Plangger Branch: s390x-backend Changeset: r82811:5e9210fd34f7 Date: 2016-03-06 12:10 +0100 http://bitbucket.org/pypy/pypy/changeset/5e9210fd34f7/ Log:fixed messed up whatsnew entry by merge... diff --git a/pypy/doc/whatsnew-5.0.0.rst

[pypy-commit] pypy s390x-backend: catchup with default

Author: Richard Plangger Branch: s390x-backend Changeset: r82832:b31ac68c3821 Date: 2016-03-06 21:35 +0100 http://bitbucket.org/pypy/pypy/changeset/b31ac68c3821/ Log:catchup with default diff --git a/pypy/doc/release-5.0.0.rst b/pypy/doc/release-5.0.0.rst ---

[pypy-commit] pypy s390x-backend: kill unused methods (_apply_scale, _multiply_by_constant)

Author: Richard Plangger Branch: s390x-backend Changeset: r82819:539acda35771 Date: 2016-03-06 18:35 +0100 http://bitbucket.org/pypy/pypy/changeset/539acda35771/ Log:kill unused methods (_apply_scale, _multiply_by_constant) diff --git

[pypy-commit] pypy s390x-backend: typo it is curRpos

Author: Richard Plangger Branch: s390x-backend Changeset: r82820:974fa6a90ecd Date: 2016-03-06 18:37 +0100 http://bitbucket.org/pypy/pypy/changeset/974fa6a90ecd/ Log:typo it is curRpos diff --git a/rpython/jit/backend/ppc/opassembler.py

[pypy-commit] pypy s390x-backend: _multiply_by_constant is still used by malloc_cond_varsize

Author: Richard Plangger Branch: s390x-backend Changeset: r82821:64b4e6d07aed Date: 2016-03-06 18:38 +0100 http://bitbucket.org/pypy/pypy/changeset/64b4e6d07aed/ Log:_multiply_by_constant is still used by malloc_cond_varsize diff --git

[pypy-commit] pypy s390x-backend: fixed resop comment + param count

Author: Richard Plangger Branch: s390x-backend Changeset: r82815:372d00e95882 Date: 2016-03-06 17:02 +0100 http://bitbucket.org/pypy/pypy/changeset/372d00e95882/ Log:fixed resop comment + param count diff --git a/rpython/jit/backend/llsupport/rewrite.py

[pypy-commit] pypy s390x-backend: (untested) fix zero array for ppc

Author: Richard Plangger Branch: s390x-backend Changeset: r82818:8ab56b6e79d4 Date: 2016-03-06 18:28 +0100 http://bitbucket.org/pypy/pypy/changeset/8ab56b6e79d4/ Log:(untested) fix zero array for ppc diff --git a/rpython/jit/backend/ppc/opassembler.py

[pypy-commit] pypy s390x-backend: (untested) length is imm, we still need to check if enough space can be written. eliminated first stdux, doing addr calc. before entering the loops

Author: Richard Plangger Branch: s390x-backend Changeset: r82829:1afedf2dd2d2 Date: 2016-03-06 20:49 +0100 http://bitbucket.org/pypy/pypy/changeset/1afedf2dd2d2/ Log:(untested) length is imm, we still need to check if enough space can be written. eliminated first

[pypy-commit] pypy s390x-backend: (untested) fixed two issues in the ppc zero_array

Author: Richard Plangger Branch: s390x-backend Changeset: r82824:4cca8fad1c1f Date: 2016-03-06 19:17 +0100 http://bitbucket.org/pypy/pypy/changeset/4cca8fad1c1f/ Log:(untested) fixed two issues in the ppc zero_array diff --git a/rpython/jit/backend/ppc/opassembler.py

[pypy-commit] pypy s390x-backend: catchup default

Author: Richard Plangger Branch: s390x-backend Changeset: r82665:90730c0a4880 Date: 2016-03-02 21:57 +0100 http://bitbucket.org/pypy/pypy/changeset/90730c0a4880/ Log:catchup default diff too long, truncating to 2000 out of 12259 lines diff --git a/.hgignore b/.hgignore

[pypy-commit] pypy s390x-backend: fix in assembly. 1 func for addr generation of gc_load/gc_store(_indexed) instead of 4

Author: Richard Plangger Branch: s390x-backend Changeset: r82664:7ec50a9c18a6 Date: 2016-03-02 21:44 +0100 http://bitbucket.org/pypy/pypy/changeset/7ec50a9c18a6/ Log:fix in assembly. 1 func for addr generation of gc_load/gc_store(_indexed) instead of 4 diff

[pypy-commit] pypy gcstress-hypothesis: moved out test test into new branch

Author: Richard Plangger Branch: gcstress-hypothesis Changeset: r82666:f2615fd00d55 Date: 2016-03-02 22:11 +0100 http://bitbucket.org/pypy/pypy/changeset/f2615fd00d55/ Log:moved out test test into new branch diff --git a/rpython/jit/backend/llsupport/tl/__init__.py

[pypy-commit] pypy s390x-backend: undo some of the changes of memop-simplify3 and pass length to the length parameter instead of byte size

Author: Richard Plangger Branch: s390x-backend Changeset: r82630:2ea190038b88 Date: 2016-03-01 12:43 +0100 http://bitbucket.org/pypy/pypy/changeset/2ea190038b88/ Log:undo some of the changes of memop-simplify3 and pass length to the length parameter instead of

[pypy-commit] pypy s390x-backend: created a test that tries to execute a byte code block (stack is not prepared)

Author: Richard Plangger Branch: s390x-backend Changeset: r82655:2c9ed008895a Date: 2016-03-01 18:48 +0100 http://bitbucket.org/pypy/pypy/changeset/2c9ed008895a/ Log:created a test that tries to execute a byte code block (stack is not prepared) diff --git

[pypy-commit] pypy s390x-backend: removed pool and loading constant using immediate values, test_runner passes already

Author: Richard Plangger Branch: s390x-backend Changeset: r82656:62b48d6dd7ca Date: 2016-03-02 14:41 +0100 http://bitbucket.org/pypy/pypy/changeset/62b48d6dd7ca/ Log:removed pool and loading constant using immediate values, test_runner passes already diff --git

[pypy-commit] pypy s390x-backend: ignore register 13 in pair allocation

Author: Richard Plangger Branch: s390x-backend Changeset: r82657:3e0b424d69d1 Date: 2016-03-02 16:02 +0100 http://bitbucket.org/pypy/pypy/changeset/3e0b424d69d1/ Log:ignore register 13 in pair allocation diff --git

[pypy-commit] pypy s390x-enhance-speed: missing module import, wrong attr access

Author: Richard Plangger Branch: s390x-enhance-speed Changeset: r82848:adcad73cca0f Date: 2016-03-07 16:28 +0100 http://bitbucket.org/pypy/pypy/changeset/adcad73cca0f/ Log:missing module import, wrong attr access diff --git a/rpython/jit/backend/zarch/pool.py

[pypy-commit] pypy s390x-enhance-speed: added two more pointer to the literal pool, exchanged some registers for better pipeline flow

Author: Richard Plangger Branch: s390x-enhance-speed Changeset: r82847:cc0fd9f1b25f Date: 2016-03-07 15:38 +0100 http://bitbucket.org/pypy/pypy/changeset/cc0fd9f1b25f/ Log:added two more pointer to the literal pool, exchanged some registers for better pipeline

[pypy-commit] pypy s390x-enhance-speed: using load_imm using pool instead of load_imm_plus for gcrootmap (root stack top addr). there is not gain for doing so on s390x.

Author: Richard Plangger Branch: s390x-enhance-speed Changeset: r82845:62241a24deb5 Date: 2016-03-07 15:10 +0100 http://bitbucket.org/pypy/pypy/changeset/62241a24deb5/ Log:using load_imm using pool instead of load_imm_plus for gcrootmap (root stack top addr).

[pypy-commit] pypy s390x-backend: correct shift for ppc zero array, ofs_loc can be a register

Author: Richard Plangger Branch: s390x-backend Changeset: r82869:014ea5f68140 Date: 2016-03-07 19:47 +0100 http://bitbucket.org/pypy/pypy/changeset/014ea5f68140/ Log:correct shift for ppc zero array, ofs_loc can be a register diff --git

[pypy-commit] pypy s390x-backend: correct positioning of the ptr to write

Author: Richard Plangger Branch: s390x-backend Changeset: r82871:89a10fc86c4b Date: 2016-03-07 20:47 +0100 http://bitbucket.org/pypy/pypy/changeset/89a10fc86c4b/ Log:correct positioning of the ptr to write diff --git a/rpython/jit/backend/ppc/opassembler.py

[pypy-commit] pypy s390x-backend: r.length_loc does not exist! removed "r."

Author: Richard Plangger Branch: s390x-backend Changeset: r82836:04dc68d8235f Date: 2016-03-07 08:20 +0100 http://bitbucket.org/pypy/pypy/changeset/04dc68d8235f/ Log:r.length_loc does not exist! removed "r." diff --git a/rpython/jit/backend/ppc/opassembler.py

[pypy-commit] pypy s390x-backend: translating the interpreter and feeding it with hypothesis, it compiles but does not correctly enter the dispatch loop

Author: Richard Plangger Branch: s390x-backend Changeset: r82638:ae5c221a741c Date: 2016-03-01 17:25 +0100 http://bitbucket.org/pypy/pypy/changeset/ae5c221a741c/ Log:translating the interpreter and feeding it with hypothesis, it compiles but does not correctly

[pypy-commit] pypy s390x-backend: removed the function field _arguments_, a new function rebuilds the information needed for the auto encoding test

Author: Richard Plangger Branch: s390x-backend Changeset: r82636:a5ef58cc Date: 2016-03-01 15:02 +0100 http://bitbucket.org/pypy/pypy/changeset/a5ef58cc/ Log:removed the function field _arguments_, a new function rebuilds the information needed for the

[pypy-commit] pypy s390x-backend: trying to translate the current interpreter in a test and later feed it with hypothesis.

Author: Richard Plangger Branch: s390x-backend Changeset: r82637:9eab49cb6677 Date: 2016-03-01 15:45 +0100 http://bitbucket.org/pypy/pypy/changeset/9eab49cb6677/ Log:trying to translate the current interpreter in a test and later feed it with hypothesis. in

[pypy-commit] pypy s390x-backend: bytecode and constants are correctly passed, need to modify hypothesis to generate correct programs

Author: Richard Plangger Branch: s390x-backend Changeset: r82640:b9cd7126874f Date: 2016-03-01 17:57 +0100 http://bitbucket.org/pypy/pypy/changeset/b9cd7126874f/ Log:bytecode and constants are correctly passed, need to modify hypothesis to generate correct

[pypy-commit] pypy s390x-backend: merged the speed improvements from s390x-enhance-speed

Author: Richard Plangger Branch: s390x-backend Changeset: r82899:c63346ce0b33 Date: 2016-03-09 09:46 +0100 http://bitbucket.org/pypy/pypy/changeset/c63346ce0b33/ Log:merged the speed improvements from s390x-enhance-speed diff --git

[pypy-commit] pypy s390x-backend: merged default

Author: Richard Plangger Branch: s390x-backend Changeset: r82900:acc1954c74fa Date: 2016-03-09 09:52 +0100 http://bitbucket.org/pypy/pypy/changeset/acc1954c74fa/ Log:merged default diff --git a/LICENSE b/LICENSE --- a/LICENSE +++ b/LICENSE @@ -240,6 +240,7 @@

[pypy-commit] pypy s390x-backend: removed skip from signal test

Author: Richard Plangger Branch: s390x-backend Changeset: r82901:2d40f0c2257e Date: 2016-03-09 10:39 +0100 http://bitbucket.org/pypy/pypy/changeset/2d40f0c2257e/ Log:removed skip from signal test diff --git a/pypy/module/__pypy__/test/test_signal.py

[pypy-commit] buildbot default: replaced python vm for building pypy with pypy! builds for s390x now spawn NOT at the same time, but shifted by 2-3h

Author: Richard Plangger Branch: Changeset: r991:2041ec4f50e8 Date: 2016-03-09 10:57 +0100 http://bitbucket.org/pypy/buildbot/changeset/2041ec4f50e8/ Log:replaced python vm for building pypy with pypy! builds for s390x now spawn NOT at the same time, but shifted

[pypy-commit] pypy new-jit-log: exposed enable_jitlog in module _vmprof, JITLOG env variable now controls the jitlog output file (bugfix), setup_once is now called while initializing the metainterp_sd

Author: Richard Plangger Branch: new-jit-log Changeset: r83499:f42dc412ebd9 Date: 2016-04-01 17:19 +0200 http://bitbucket.org/pypy/pypy/changeset/f42dc412ebd9/ Log:exposed enable_jitlog in module _vmprof, JITLOG env variable now controls the jitlog output file

[pypy-commit] pypy default: _patch_jump_for_descr cannot be used to patch any jump (stitch bridge would not work if the offset was too big. thx arigato)

Author: Richard Plangger Branch: Changeset: r83463:c5ec7f300d0a Date: 2016-03-31 15:58 +0200 http://bitbucket.org/pypy/pypy/changeset/c5ec7f300d0a/ Log:_patch_jump_for_descr cannot be used to patch any jump (stitch bridge would not work if the offset was too

[pypy-commit] pypy new-jit-log: refactoring the jitlog, inlines encode methods, finish_once of cpu now prints enter count of bridges, entries and loops

Author: Richard Plangger Branch: new-jit-log Changeset: r83484:992293e7c5b7 Date: 2016-04-01 10:56 +0200 http://bitbucket.org/pypy/pypy/changeset/992293e7c5b7/ Log:refactoring the jitlog, inlines encode methods, finish_once of cpu now prints enter count of

[pypy-commit] pypy new-jit-log: catchup default

Author: Richard Plangger Branch: new-jit-log Changeset: r83485:6f53abf601b1 Date: 2016-04-01 10:57 +0200 http://bitbucket.org/pypy/pypy/changeset/6f53abf601b1/ Log:catchup default diff --git a/pypy/module/__pypy__/interp_magic.py b/pypy/module/__pypy__/interp_magic.py

[pypy-commit] pypy new-jit-log: merged default, translation issue

Author: Richard Plangger Branch: new-jit-log Changeset: r83482:132688540925 Date: 2016-04-01 09:53 +0200 http://bitbucket.org/pypy/pypy/changeset/132688540925/ Log:merged default, translation issue diff --git a/pypy/doc/__pypy__-module.rst b/pypy/doc/__pypy__-module.rst

[pypy-commit] pypy new-jit-log: try init from env variable rewritten, there is currently no filtering mechanism. filtering does not make sense anymore

Author: Richard Plangger Branch: new-jit-log Changeset: r83490:b8dc684ce886 Date: 2016-04-01 12:50 +0200 http://bitbucket.org/pypy/pypy/changeset/b8dc684ce886/ Log:try init from env variable rewritten, there is currently no filtering mechanism. filtering does not

[pypy-commit] pypy new-jit-log: do not release gil for two logger functions. they are called in @rgc.no_release_gil

Author: Richard Plangger Branch: new-jit-log Changeset: r83488:4a28028053f5 Date: 2016-04-01 11:54 +0200 http://bitbucket.org/pypy/pypy/changeset/4a28028053f5/ Log:do not release gil for two logger functions. they are called in @rgc.no_release_gil diff --git

[pypy-commit] pypy new-jit-log: another releasegil=False

Author: Richard Plangger Branch: new-jit-log Changeset: r83489:0edef2d07438 Date: 2016-04-01 12:42 +0200 http://bitbucket.org/pypy/pypy/changeset/0edef2d07438/ Log:another releasegil=False diff --git a/rpython/rlib/rvmprof/cintf.py b/rpython/rlib/rvmprof/cintf.py ---

[pypy-commit] pypy new-jit-log: minor changed in the emitted format

Author: Richard Plangger Branch: new-jit-log Changeset: r83457:26c9229ea302 Date: 2016-03-31 14:30 +0200 http://bitbucket.org/pypy/pypy/changeset/26c9229ea302/ Log:minor changed in the emitted format diff --git a/rpython/jit/backend/x86/assembler.py

[pypy-commit] pypy default: is_jit_debug method was moved. fixes crashes on s390x

Author: Richard Plangger Branch: Changeset: r83458:f1a599da9f67 Date: 2016-03-31 14:37 +0200 http://bitbucket.org/pypy/pypy/changeset/f1a599da9f67/ Log:is_jit_debug method was moved. fixes crashes on s390x diff --git a/rpython/jit/backend/zarch/pool.py

[pypy-commit] pypy new-jit-log: catchup with default

Author: Richard Plangger Branch: new-jit-log Changeset: r83459:83bff96fb266 Date: 2016-03-31 14:43 +0200 http://bitbucket.org/pypy/pypy/changeset/83bff96fb266/ Log:catchup with default diff too long, truncating to 2000 out of 9743 lines diff --git

[pypy-commit] pypy default: is_jit_debug takes the integer opnum as argument

Author: Richard Plangger Branch: Changeset: r83474:7e0e8c6d8985 Date: 2016-04-01 08:38 +0200 http://bitbucket.org/pypy/pypy/changeset/7e0e8c6d8985/ Log:is_jit_debug takes the integer opnum as argument diff --git a/rpython/jit/backend/zarch/pool.py

[pypy-commit] pypy new-jit-log: removed old files, adapted main pypy jit driver

Author: Richard Plangger Branch: new-jit-log Changeset: r83868:dcee45542016 Date: 2016-04-25 15:54 +0200 http://bitbucket.org/pypy/pypy/changeset/dcee45542016/ Log:removed old files, adapted main pypy jit driver diff --git a/pypy/module/pypyjit/interp_jit.py

[pypy-commit] pypy new-jit-log: creating jit logger for base optimization tests

Author: Richard Plangger Branch: new-jit-log Changeset: r83867:c186763aab26 Date: 2016-04-25 15:30 +0200 http://bitbucket.org/pypy/pypy/changeset/c186763aab26/ Log:creating jit logger for base optimization tests diff --git

[pypy-commit] pypy.org extradoc: updated the architectures on features.html and regenerated them

Author: Richard Plangger Branch: extradoc Changeset: r737:7da22d5ebd79 Date: 2016-04-29 08:37 +0200 http://bitbucket.org/pypy/pypy.org/changeset/7da22d5ebd79/ Log:updated the architectures on features.html and regenerated them diff --git a/compat.html b/compat.html ---

[pypy-commit] pypy.org extradoc: x86 on one line, regenerated source

Author: Richard Plangger Branch: extradoc Changeset: r739:f5505f8ce414 Date: 2016-04-29 09:05 +0200 http://bitbucket.org/pypy/pypy.org/changeset/f5505f8ce414/ Log:x86 on one line, regenerated source diff --git a/features.html b/features.html --- a/features.html +++

[pypy-commit] pypy.org extradoc: some clarification and added operating systems on which they run (regenerated the page as well)

Author: Richard Plangger Branch: extradoc Changeset: r738:53b0a434ba24 Date: 2016-04-29 08:48 +0200 http://bitbucket.org/pypy/pypy.org/changeset/53b0a434ba24/ Log:some clarification and added operating systems on which they run (regenerated the page as well)

[pypy-commit] pypy default: updated the FAQ entry (cpu architectures supported)

Author: Richard Plangger Branch: Changeset: r84020:9806eaad6bd9 Date: 2016-04-29 08:57 +0200 http://bitbucket.org/pypy/pypy/changeset/9806eaad6bd9/ Log:updated the FAQ entry (cpu architectures supported) diff --git a/pypy/doc/faq.rst b/pypy/doc/faq.rst ---

[pypy-commit] pypy new-jit-log: setting the metainterp_sd earlier to be able to resolve constptr even for the trace that has not been optimized

Author: Richard Plangger Branch: new-jit-log Changeset: r83901:ebe7bd49cf45 Date: 2016-04-26 10:54 +0200 http://bitbucket.org/pypy/pypy/changeset/ebe7bd49cf45/ Log:setting the metainterp_sd earlier to be able to resolve constptr even for the trace that has not

[pypy-commit] pypy new-jit-log: start_new_trace was called too late, it must be called before the optimizer is invoked. added an ABORT_TRACE marker and encoding it into the jitlog

Author: Richard Plangger Branch: new-jit-log Changeset: r83900:1ba36757f66e Date: 2016-04-26 09:17 +0200 http://bitbucket.org/pypy/pypy/changeset/1ba36757f66e/ Log:start_new_trace was called too late, it must be called before the optimizer is invoked. added an

[pypy-commit] pypy new-jit-log: init merge point needs a counter before encoding the types,

Author: Richard Plangger Branch: new-jit-log Changeset: r83913:ef963d113f77 Date: 2016-04-26 17:20 +0200 http://bitbucket.org/pypy/pypy/changeset/ef963d113f77/ Log:init merge point needs a counter before encoding the types, removed one wrong start_new_trace which

[pypy-commit] pypy new-jit-log: added jitlog_disable to the _vmprof module

Author: Richard Plangger Branch: new-jit-log Changeset: r83969:5e6017560e6b Date: 2016-04-27 11:15 +0200 http://bitbucket.org/pypy/pypy/changeset/5e6017560e6b/ Log:added jitlog_disable to the _vmprof module diff --git a/pypy/module/_vmprof/__init__.py

[pypy-commit] pypy new-jit-log: stitching was not written to the log, but it is now

Author: Richard Plangger Branch: new-jit-log Changeset: r83971:33c6e3af459c Date: 2016-04-27 14:43 +0200 http://bitbucket.org/pypy/pypy/changeset/33c6e3af459c/ Log:stitching was not written to the log, but it is now diff --git a/rpython/jit/backend/x86/assembler.py

[pypy-commit] pypy new-jit-log: fixed tests and added missing method in rvmprof/__init__.py (for disable_jitlog)

Author: Richard Plangger Branch: new-jit-log Changeset: r83981:638ea201b5b8 Date: 2016-04-27 15:04 +0200 http://bitbucket.org/pypy/pypy/changeset/638ea201b5b8/ Log:fixed tests and added missing method in rvmprof/__init__.py (for disable_jitlog) diff --git

[pypy-commit] pypy new-jit-log: copy machine code byte by byte instead of converting it to HEX

Author: Richard Plangger Branch: new-jit-log Changeset: r83980:c6b6bb67de34 Date: 2016-04-27 14:58 +0200 http://bitbucket.org/pypy/pypy/changeset/c6b6bb67de34/ Log:copy machine code byte by byte instead of converting it to HEX diff --git a/rpython/rlib/jitlog.py

[pypy-commit] pypy new-jit-log: hurrayy! the new get_location function API is now extensible and pypy translates (this took me quite a while)

Author: Richard Plangger Branch: new-jit-log Changeset: r83847:a02168e4834f Date: 2016-04-25 08:38 +0200 http://bitbucket.org/pypy/pypy/changeset/a02168e4834f/ Log:hurrayy! the new get_location function API is now extensible and pypy translates (this took me

<    5   6   7   8   9   10   11   12   13   14   >