- cmucl-cvs - mailman3.common-lisp.net

[Git][cmucl/cmucl] Pushed new branch issue-299-enable-xoroshiro-assem-routine
by Raymond Toy (＠rtoy) 04 Apr '24

04 Apr '24

Raymond Toy pushed new branch issue-299-enable-xoroshiro-assem-routine at cmucl / cmucl -- View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/tree/issue-299-enable-xoroshir… You're receiving this email because of your account on gitlab.common-lisp.net.

1 0

[Git][cmucl/cmucl][issue-298-with-float-rounding-mode] 2 commits: Update cmucl.pot with new docstrings
by Raymond Toy (＠rtoy) 02 Apr '24

02 Apr '24

Raymond Toy pushed to branch issue-298-with-float-rounding-mode at cmucl / cmucl Commits: bf417b47 by Raymond Toy at 2024-04-02T08:33:30-07:00 Update cmucl.pot with new docstrings - - - - - e32034e0 by Raymond Toy at 2024-04-02T08:33:52-07:00 Add tests for with-float-rounding-mode Not 100% sure the expected results are right. They seem plausible, but I didn't actually analyze that the rounding is correct. - - - - - 2 changed files: - src/i18n/locale/cmucl.pot - tests/float.lisp Changes: ===================================== src/i18n/locale/cmucl.pot ===================================== @@ -4868,6 +4868,25 @@ msgid "" " their testing within, and restored on exit." msgstr "" +#: src/code/float-trap.lisp +msgid "" +"Execute BODY with the floating-point rounding mode set to\n" +" ROUNDING-MODE. ROUNDING-MODE must a one of\n" +"\n" +" :NEAREST\n" +" the default mode of round to nearest even\n" +" :ZERO\n" +" round numbers down towards zero. Positive numbers round down\n" +" and negative numbers round up.\n" +" :POSITIVE-INFINITY\n" +" round numbers up towards positive infinity\n" +" :NEGATIVE-INFINITY\n" +" round numbers down towards negative infinity\n" +"\n" +" Only the rounding mode is restored on exit; other floating-point\n" +" modes are not modified. " +msgstr "" + #: src/code/float.lisp msgid "Return true if the float X is denormalized." msgstr "" ===================================== tests/float.lisp ===================================== @@ -212,3 +212,38 @@ ;; most-positive-double-float. And a really big single-float. (assert-error 'reader-error (read-from-string "1.8d308")) (assert-error 'reader-error (read-from-string "1d999999999"))) + +(defun rounding-test (x) + (declare (double-float x) + (optimize (speed 3))) + (* x (/ 1d0 x))) + +(define-test rounding-mode.nearest + (:tag :issues) + (ext:with-float-rounding-mode (:nearest) + (assert-equal 1d0 (rounding-test 3d0)))) + +(define-test rounding-mode.zero.1 + (:tag :issues) + (ext:with-float-rounding-mode (:zero) + (assert-equal 0.9999999999999999d0 + (rounding-test 3d0)))) + +(define-test rounding-mode.zero.2 + (:tag :issues) + (ext:with-float-rounding-mode (:zero) + (assert-equal 0.9999999999999999d0 + (rounding-test -3d0)))) + +(define-test rounding-mode.positive-infinity + (:tag :issues) + (ext:with-float-rounding-mode (:positive-infinity) + (assert-equal 1.0000000000000002d0 + (rounding-test 3d0)))) + +(define-test rounding-mode.negative-infinity + (:tag :issues) + (ext:with-float-rounding-mode (:negative-infinity) + (assert-equal 0.9999999999999999d0 + (rounding-test 3d0)))) + View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/compare/7543b7202a5661892dea14… -- View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/compare/7543b7202a5661892dea14… You're receiving this email because of your account on gitlab.common-lisp.net.

1 0

[Git][cmucl/cmucl] Pushed new branch issue-298-with-float-rounding-mode
by Raymond Toy (＠rtoy) 02 Apr '24

02 Apr '24

Raymond Toy pushed new branch issue-298-with-float-rounding-mode at cmucl / cmucl -- View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/tree/issue-298-with-float-roun… You're receiving this email because of your account on gitlab.common-lisp.net.

1 0

[Git][cmucl/cmucl] Pushed new branch issue-295-docstring-for-define-assembly-routine
by Raymond Toy (＠rtoy) 02 Apr '24

02 Apr '24

Raymond Toy pushed new branch issue-295-docstring-for-define-assembly-routine at cmucl / cmucl -- View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/tree/issue-295-docstring-for-d… You're receiving this email because of your account on gitlab.common-lisp.net.

1 0

[Git][cmucl/cmucl][issue-297-pprint-assemble] 2 commits: Clarify release notes to say new-assem:assemble
by Raymond Toy (＠rtoy) 02 Apr '24

02 Apr '24

Raymond Toy pushed to branch issue-297-pprint-assemble at cmucl / cmucl Commits: a4eb4787 by Raymond Toy at 2024-04-01T18:41:35-07:00 Clarify release notes to say new-assem:assemble - - - - - 22f0bec1 by Raymond Toy at 2024-04-01T19:26:37-07:00 Clean up pprint new-assem:assemble test In the source, use the VM package for the symbols instead of defaulting to the `PPRINT-TESTS` package. This means the expected results need to be updated too. - - - - - 2 changed files: - src/general-info/release-21f.md - tests/pprint.lisp Changes: ===================================== src/general-info/release-21f.md ===================================== @@ -65,7 +65,7 @@ public domain. * ~~#288~~ Re-enable `deftransform` for random integers. * ~~#290~~ Pprint `with-float-traps-masked` better * ~~#291~~ Pprint `handler-case` neatly. - * ~~#297~~ Pprint `assemble` with less indentation. + * ~~#297~~ Pprint `new-assem:assemble` with less indentation. * Other changes: * Improvements to the PCL implementation of CLOS: * Changes to building procedure: ===================================== tests/pprint.lisp ===================================== @@ -126,11 +126,12 @@ (:tag :issues) (assert-equal " -(NEW-ASSEM:ASSEMBLE (*CODE-SEGMENT* 'XOROSHIRO-UPDATE) - XOROSHIRO-UPDATE - (PUSH (CONS 'XOROSHIRO-UPDATE XOROSHIRO-UPDATE) C::*ASSEMBLER-ROUTINES*))" +(NEW-ASSEM:ASSEMBLE (C:*CODE-SEGMENT* 'X86::XOROSHIRO-UPDATE) + X86::XOROSHIRO-UPDATE + (PUSH (CONS 'X86::XOROSHIRO-UPDATE X86::XOROSHIRO-UPDATE) + C::*ASSEMBLER-ROUTINES*))" (with-output-to-string (s) - (pprint '(new-assem:assemble (*code-segment* 'xoroshiro-update) - xoroshiro-update - (push (cons 'xoroshiro-update xoroshiro-update) c::*assembler-routines*)) + (pprint '(new-assem:assemble (c::*code-segment* 'vm::xoroshiro-update) + vm::xoroshiro-update + (push (cons 'vm::xoroshiro-update vm::xoroshiro-update) c::*assembler-routines*)) s)))) View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/compare/e9eed900d911463276db32… -- View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/compare/e9eed900d911463276db32… You're receiving this email because of your account on gitlab.common-lisp.net.

1 0

[Git][cmucl/cmucl][issue-297-pprint-assemble] 2 commits: Add a test for the pprinter for new-assem:assemble
by Raymond Toy (＠rtoy) 02 Apr '24

02 Apr '24

Raymond Toy pushed to branch issue-297-pprint-assemble at cmucl / cmucl Commits: a79e94ac by Raymond Toy at 2024-04-01T18:40:03-07:00 Add a test for the pprinter for new-assem:assemble - - - - - e9eed900 by Raymond Toy at 2024-04-01T18:40:20-07:00 Update release notes for issue #297 - - - - - 2 changed files: - src/general-info/release-21f.md - tests/pprint.lisp Changes: ===================================== src/general-info/release-21f.md ===================================== @@ -65,6 +65,7 @@ public domain. * ~~#288~~ Re-enable `deftransform` for random integers. * ~~#290~~ Pprint `with-float-traps-masked` better * ~~#291~~ Pprint `handler-case` neatly. + * ~~#297~~ Pprint `assemble` with less indentation. * Other changes: * Improvements to the PCL implementation of CLOS: * Changes to building procedure: ===================================== tests/pprint.lisp ===================================== @@ -121,3 +121,16 @@ (:no-error () (format nil "Nothing bad happened."))) s)))) + +(define-test pprint.assemble + (:tag :issues) + (assert-equal + " +(NEW-ASSEM:ASSEMBLE (*CODE-SEGMENT* 'XOROSHIRO-UPDATE) + XOROSHIRO-UPDATE + (PUSH (CONS 'XOROSHIRO-UPDATE XOROSHIRO-UPDATE) C::*ASSEMBLER-ROUTINES*))" + (with-output-to-string (s) + (pprint '(new-assem:assemble (*code-segment* 'xoroshiro-update) + xoroshiro-update + (push (cons 'xoroshiro-update xoroshiro-update) c::*assembler-routines*)) + s)))) View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/compare/cf6803a34ed3ec008fb0c7… -- View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/compare/cf6803a34ed3ec008fb0c7… You're receiving this email because of your account on gitlab.common-lisp.net.

1 0

[Git][cmucl/cmucl] Pushed new branch issue-297-pprint-assemble
by Raymond Toy (＠rtoy) 02 Apr '24

02 Apr '24

Raymond Toy pushed new branch issue-297-pprint-assemble at cmucl / cmucl -- View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/tree/issue-297-pprint-assemble You're receiving this email because of your account on gitlab.common-lisp.net.

1 0

[Git][cmucl/cmucl][issue-294-xoroshiro-lisp-assem-routine] Change the registers used for state, r1, and r0
by Raymond Toy (＠rtoy) 02 Apr '24

02 Apr '24

Raymond Toy pushed to branch issue-294-xoroshiro-lisp-assem-routine at cmucl / cmucl Commits: 1ec83248 by Raymond Toy at 2024-04-01T17:28:35-07:00 Change the registers used for state, r1, and r0 These registers show that in `xoroshiro-gen`, all the registers are in the right place so we don't need to do any additional moves. This is a nice micro-optimization. - - - - - 1 changed file: - src/assembly/x86/arith.lisp Changes: ===================================== src/assembly/x86/arith.lisp ===================================== @@ -447,8 +447,8 @@ (:policy :fast-safe) (:arg-types simple-array-double-float) (:result-types unsigned-num unsigned-num)) - ((:arg state descriptor-reg eax-offset) - (:res r1 unsigned-reg edx-offset) + ((:arg state descriptor-reg edx-offset) + (:res r1 unsigned-reg ecx-offset) (:res r0 unsigned-reg ebx-offset) (:temp s0 double-reg xmm0-offset) (:temp s1 double-reg xmm1-offset) View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/commit/1ec832483118f863620fd90… -- View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/commit/1ec832483118f863620fd90… You're receiving this email because of your account on gitlab.common-lisp.net.

1 0

[Git][cmucl/cmucl][issue-294-xoroshiro-lisp-assem-routine] 4 commits: generate-return-sequence generated incorrect :raw return
by Raymond Toy (＠rtoy) 31 Mar '24

31 Mar '24

Raymond Toy pushed to branch issue-294-xoroshiro-lisp-assem-routine at cmucl / cmucl Commits: 91a947c4 by Raymond Toy at 2024-03-31T07:52:16-07:00 generate-return-sequence generated incorrect :raw return When the return style is `:raw`, we generated the wrong sequence of instructions because `generate-return-sequence` is supposed to return a list of the instructions. We were returning the "instructions" `inst` and `ret` instead of `(inst ret)`. - - - - - 06445c08 by Raymond Toy at 2024-03-31T07:54:58-07:00 Define a vop for the xoroshiro-update assembly routine. The generated VOP basically does what we want and is the same as the hand-written one in compiler/x86/arith.lisp. To make this work, just needed to add `:arg-types` and `:result-types` options. - - - - - c42702ce by Raymond Toy at 2024-03-31T08:03:45-07:00 Some cleanups and comments Rename `result[01]` back to `r[01]`. Add some comments and move the description of the algorithm from the body of the assembly routine to the comment before the definition of the assembly routine. - - - - - 4910c4f0 by Raymond Toy at 2024-03-31T08:05:54-07:00 Remove the hand-written update VOP This VOP is now generated by the assembly routine and is basically identical to what we wrote here. Hence, we don't need this, which was commented out anyway. - - - - - 3 changed files: - src/assembly/x86/arith.lisp - src/assembly/x86/support.lisp - src/compiler/x86/arith.lisp Changes: ===================================== src/assembly/x86/arith.lisp ===================================== @@ -412,15 +412,41 @@ (inst pop k) (inst ret)) - -#+(and random-xoroshiro assembler) +;;; Support for the xoroshiro128** generator. See +;;; https://prng.di.unimi.it/xoroshiro128starstar.c for the official +;;; code. +;;; +;;; This is what we're implementing, where s[] is our state vector. +;;; +;;; static uint64_t s[2]; +;;; static inline uint64_t rotl(const uint64_t x, int k) { +;;; return (x << k) | (x >> (64 - k)); +;;; } +;;; +;;; uint64_t next(void) { +;;; const uint64_t s0 = s[0]; +;;; uint64_t s1 = s[1]; +;;; const uint64_t result = rotl(s0 * 5, 7) * 9; +;;; +;;; s1 ^= s0; +;;; s[0] = rotl(s0, 24) ^ s1 ^ (s1 << 16); // a, b +;;; s[1] = rotl(s1, 37); // c +;;; +;;; return result; +;;; } +;;; +;;; A VOP is also generated to call this assembly routine. This +;;; routine computes a new 64-bit random number and also updates the +;;; state, which is (simple-array (double-float) (2)). +#+random-xoroshiro (define-assembly-routine (xoroshiro-update - (:translate kernel::xoroshiro-update) + (:translate kernel::random-xoroshiro-update) (:return-style :raw) (:cost 30) - (:policy :safe) - (:save-p t)) + (:policy :fast-safe) + (:arg-types simple-array-double-float) + (:result-types unsigned-num unsigned-num)) ((:arg state descriptor-reg eax-offset) (:res r1 unsigned-reg edx-offset) (:res r0 unsigned-reg ebx-offset) @@ -428,26 +454,6 @@ (:temp s1 double-reg xmm1-offset) (:temp t0 double-reg xmm2-offset) (:temp t1 double-reg xmm3-offset)) - ;; See https://prng.di.unimi.it/xoroshiro128starstar.c for the official code. - ;; - ;; This is what we're implementing, where s[] is our state vector. - ;; - ;; static uint64_t s[2]; - ;; static inline uint64_t rotl(const uint64_t x, int k) { - ;; return (x << k) | (x >> (64 - k)); - ;; } - ;; - ;; uint64_t next(void) { - ;; const uint64_t s0 = s[0]; - ;; uint64_t s1 = s[1]; - ;; const uint64_t result = rotl(s0 * 5, 7) * 9; - ;; - ;; s1 ^= s0; - ;; s[0] = rotl(s0, 24) ^ s1 ^ (s1 << 16); // a, b - ;; s[1] = rotl(s1, 37); // c - ;; - ;; return result; - ;; } ;; s0 = state[0] (inst movsd s0 (make-ea :dword :base state @@ -522,5 +528,4 @@ vm:word-bytes) (* 8 1)) vm:other-pointer-type)) - s1) - (inst ret)) + s1)) ===================================== src/assembly/x86/support.lisp ===================================== @@ -39,7 +39,7 @@ (def-vm-support-routine generate-return-sequence (style) (ecase style (:raw - `(inst ret)) + `((inst ret))) (:full-call `( (inst pop eax-tn) ===================================== src/compiler/x86/arith.lisp ===================================== @@ -1812,34 +1812,6 @@ ) #+random-xoroshiro -(progn (defknown kernel::random-xoroshiro-update ((simple-array double-float (2))) (values (unsigned-byte 32) (unsigned-byte 32)) (movable)) - - -(define-vop (random-xoroshiro-update) - (:policy :fast-safe) - (:translate kernel::random-xoroshiro-update) - (:args (state :scs (descriptor-reg) :target state-arg)) - (:arg-types simple-array-double-float) - (:results (hi :scs (unsigned-reg)) - (lo :scs (unsigned-reg))) - (:result-types unsigned-num unsigned-num) - (:temporary (:sc descriptor-reg :offset eax-offset) state-arg) - (:temporary (:sc double-reg :offset xmm0-offset) s0) - (:temporary (:sc double-reg :offset xmm1-offset) s1) - (:temporary (:sc double-reg :offset xmm2-offset) t0) - (:temporary (:sc double-reg :offset xmm3-offset) t1) - (:temporary (:sc unsigned-reg :offset edx-offset :target hi) r1) - (:temporary (:sc unsigned-reg :offset ebx-offset :target lo) r0) - (:generator 50 - (move state-arg state) - (move s0 s0) - (move s1 s1) - (move t0 t0) - (move t1 t1) - (inst call (make-fixup 'vm::xoroshiro-update :assembly-routine)) - (move hi r1) - (move lo r0))) -) View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/compare/204c9c34731d07e2b274c2… -- View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/compare/204c9c34731d07e2b274c2… You're receiving this email because of your account on gitlab.common-lisp.net.

1 0

[Git][cmucl/cmucl] Pushed new branch issue-294-xoroshiro-lisp-assem-routine
by Raymond Toy (＠rtoy) 31 Mar '24

31 Mar '24

Raymond Toy pushed new branch issue-294-xoroshiro-lisp-assem-routine at cmucl / cmucl -- View it on GitLab: https://gitlab.common-lisp.net/cmucl/cmucl/-/tree/issue-294-xoroshiro-lisp-… You're receiving this email because of your account on gitlab.common-lisp.net.

1 0