* bug#60971: build failure of v3.0.9rc1 on mac os 12.6
@ 2023-01-20 19:16 lloda
[not found] ` <handler.60971.B.167424218722695.ack@debbugs.gnu.org>
0 siblings, 1 reply; 6+ messages in thread
From: lloda @ 2023-01-20 19:16 UTC (permalink / raw)
To: 60971
Hello,
v3.0.9rc1 fails on mac os 12.6.2 & gcc 12.2 or clang 14, same error in either case. gcc's error is:
------------------------------------
CC libguile_3.0_la-posix.lo
In file included from ../../../src/guile4/libguile/posix.c:82:
../../../src/guile4/libguile/posix.c:109:9: error: lvalue required as unary '&' operand
109 | verify (WEXITSTATUS (W_EXITCODE (127, 0)) == 127);
| ^~~~~~~~~~~
../../../src/guile4/lib/verify.h:213:57: note: in definition of macro '_GL_VERIFY'
213 | # define _GL_VERIFY(R, DIAGNOSTIC, ...) _Static_assert (R, DIAGNOSTIC)
| ^
../../../src/guile4/libguile/posix.c:109:1: note: in expansion of macro 'verify'
109 | verify (WEXITSTATUS (W_EXITCODE (127, 0)) == 127);
| ^~~~~~
../../../src/guile4/libguile/posix.c:109:9: error: expression in static assertion is not an integer
109 | verify (WEXITSTATUS (W_EXITCODE (127, 0)) == 127);
| ^~~~~~~~~~~
../../../src/guile4/lib/verify.h:213:57: note: in definition of macro '_GL_VERIFY'
213 | # define _GL_VERIFY(R, DIAGNOSTIC, ...) _Static_assert (R, DIAGNOSTIC)
| ^
../../../src/guile4/libguile/posix.c:109:1: note: in expansion of macro 'verify'
109 | verify (WEXITSTATUS (W_EXITCODE (127, 0)) == 127);
| ^~~~~~
------------------------------------
The problematic section in libguile/posix.c
...
#if HAVE_SYS_WAIT_H
# include <sys/wait.h>
#endif
#ifndef WEXITSTATUS
# define WEXITSTATUS(stat_val) ((unsigned)(stat_val) >> 8)
#endif
#ifndef WIFEXITED
# define WIFEXITED(stat_val) (((stat_val) & 255) == 0)
#endif
#ifndef W_EXITCODE
/* Macro for constructing a status value. Found in glibc. */
# ifdef _WIN32 /* see Gnulib's posix-w32.h */
# define W_EXITCODE(ret, sig) (ret)
# else
# define W_EXITCODE(ret, sig) ((ret) << 8 | (sig))
# endif
#endif
verify (WEXITSTATUS (W_EXITCODE (127, 0)) == 127);
...
adding -o .libs/libguile_3.0_la-posix.i -E to the build line shows line 109 as
_Static_assert ( (((*(int *)&((( 127 ) << 8 | ( 0 )))) >> 8) & 0x000000ff) == 127, "verify (" "WEXITSTATUS (W_EXITCODE (127, 0)) == 127" ")");
^
forcing redefinition of WEXITSTATUS shows the previous definition at /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/wait.h
...
#if defined(_POSIX_C_SOURCE) && !defined(_DARWIN_C_SOURCE)
#define _W_INT(i) (i)
#else
#define _W_INT(w) (*(int *)&(w)) /* convert union wait to int */ <------------- problem
#define WCOREFLAG 0200
#endif /* (_POSIX_C_SOURCE && !_DARWIN_C_SOURCE) */
/* These macros are permited, as they are in the implementation namespace */
#define _WSTATUS(x) (_W_INT(x) & 0177)
#define _WSTOPPED 0177 /* _WSTATUS if process is stopped */
/*
* [XSI] The <sys/wait.h> header shall define the following macros for
* analysis of process status values
*/
#if __DARWIN_UNIX03
#define WEXITSTATUS(x) ((_W_INT(x) >> 8) & 0x000000ff)
#else /* !__DARWIN_UNIX03 */
#define WEXITSTATUS(x) (_W_INT(x) >> 8)
#endif /* !__DARWIN_UNIX03 */
/* 0x13 == SIGCONT */
#define WSTOPSIG(x) (_W_INT(x) >> 8)
#define WIFCONTINUED(x) (_WSTATUS(x) == _WSTOPPED && WSTOPSIG(x) == 0x13)
#define WIFSTOPPED(x) (_WSTATUS(x) == _WSTOPPED && WSTOPSIG(x) != 0x13)
#define WIFEXITED(x) (_WSTATUS(x) == 0)
#define WIFSIGNALED(x) (_WSTATUS(x) != _WSTOPPED && _WSTATUS(x) != 0)
#define WTERMSIG(x) (_WSTATUS(x))
#if (!defined(_POSIX_C_SOURCE) || defined(_DARWIN_C_SOURCE))
#define WCOREDUMP(x) (_W_INT(x) & WCOREFLAG)
#define W_EXITCODE(ret, sig) ((ret) << 8 | (sig))
#define W_STOPCODE(sig) ((sig) << 8 | _WSTOPPED)
#endif /* (!defined(_POSIX_C_SOURCE) || defined(_DARWIN_C_SOURCE)) */
...
so that's the problem, but I'm not sure what the solution is. The way it's
written, W_EXITCODE() is always going to give up a number...
Thanks
Daniel
^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#61660: [feature request] optimization of case-lambda
@ 2023-02-20 17:56 lloda
0 siblings, 0 replies; 6+ messages in thread
From: lloda @ 2023-02-20 17:56 UTC (permalink / raw)
To: 61660
On 3.0.9
> ,optimize ((case-lambda (() 0)))
= 0
but
> ,optimize ((case-lambda (() 0) ((a) 1)))
= ((case-lambda (() 0) ((a) 1)))
The problem with this is that when the output of a macro contains case-lambda, recursive application results in geometrical increase of code size. It seems that it should be possible to resolve the application on the spot; at last to reduce the case-lambda to a simple lambda when the arity is available (which now only happens when the case-lambda contains a single clause).
Thanks
Daniel
^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#61660: [feature request] optimization of case-lambda
[not found] ` <handler.60971.B.167424218722695.ack@debbugs.gnu.org>
@ 2023-02-23 19:00 ` lloda
2023-02-24 17:26 ` lloda
0 siblings, 1 reply; 6+ messages in thread
From: lloda @ 2023-02-23 19:00 UTC (permalink / raw)
To: 61660
[-- Attachment #1: Type: text/plain, Size: 39 bytes --]
Patch attached, tests tbd. Thoughts?
[-- Attachment #2: 0001-peval-reduces-some-inlined-case-lambda-calls.patch --]
[-- Type: application/octet-stream, Size: 1952 bytes --]
From a1f605598a7a49b2971072c8db635b788e209e3a Mon Sep 17 00:00:00 2001
From: Daniel Llorens <lloda@sarc.name>
Date: Thu, 23 Feb 2023 17:38:10 +0100
Subject: [PATCH] peval reduces some inlined case-lambda calls
* module/language/tree-il/peval.scm (peval): Reduce multiple case lambda
in <call> trees according to the number of arguments. Lambda cases
with more than required args (either optional, keyword, or rest
arguments) are ignored.
---
module/language/tree-il/peval.scm | 15 +++++++++++++++
1 file changed, 15 insertions(+)
diff --git a/module/language/tree-il/peval.scm b/module/language/tree-il/peval.scm
index 7945fd9b9..5c9799edb 100644
--- a/module/language/tree-il/peval.scm
+++ b/module/language/tree-il/peval.scm
@@ -1668,6 +1668,21 @@ top-level bindings from ENV and return the resulting expression."
(log 'inline-end result exp)
result)))))
+ (($ <lambda> src-proc meta body)
+ ;; If there are multiple cases and one matches nargs, omit all the others.
+ (or (and (lambda-case-alternate body)
+ (let ((nargs (length orig-args)))
+ (let loop ((body body))
+ (match body
+ (#f #f)
+ (($ <lambda-case> src-body req opt rest kw inits gensyms body alt)
+ (if (and (not opt) (not rest) (not kw) (= nargs (length req)))
+ (revisit-proc
+ (make-lambda
+ src-proc meta
+ (make-lambda-case src-body req opt rest kw inits gensyms body #f)))
+ (loop alt)))))))
+ (make-call src (for-call orig-proc) (map for-value orig-args))))
(($ <let> _ _ _ vals _)
;; Attempt to inline `let' in the operator position.
;;
--
2.30.2
^ permalink raw reply related [flat|nested] 6+ messages in thread
* bug#61660: [feature request] optimization of case-lambda
2023-02-23 19:00 ` bug#61660: [feature request] optimization of case-lambda lloda
@ 2023-02-24 17:26 ` lloda
2023-02-27 10:11 ` Ludovic Courtès
0 siblings, 1 reply; 6+ messages in thread
From: lloda @ 2023-02-24 17:26 UTC (permalink / raw)
To: 61660
[-- Attachment #1: Type: text/plain, Size: 59 bytes --]
Fixed patch handling rest & #:optional, with test cases.
[-- Attachment #2: 0001-peval-reduces-some-inlined-case-lambda-calls.patch --]
[-- Type: application/octet-stream, Size: 4757 bytes --]
From 61ed612fb36108e395bdee4b1bbb46b49ef017b3 Mon Sep 17 00:00:00 2001
From: Daniel Llorens <lloda@sarc.name>
Date: Thu, 23 Feb 2023 17:38:10 +0100
Subject: [PATCH] peval reduces some inlined case-lambda calls
* module/language/tree-il/peval.scm (peval): Reduce multiple case lambda
in <call> trees according to the number of arguments. Do not try to
reduce case-lambda using keyword arguments.
* test-suite/tests/peval.test: Tests.
---
module/language/tree-il/peval.scm | 23 ++++++++++
test-suite/tests/peval.test | 72 +++++++++++++++++++++++++++++++
2 files changed, 95 insertions(+)
diff --git a/module/language/tree-il/peval.scm b/module/language/tree-il/peval.scm
index 7945fd9b9..7c05e9a2e 100644
--- a/module/language/tree-il/peval.scm
+++ b/module/language/tree-il/peval.scm
@@ -1668,6 +1668,29 @@ top-level bindings from ENV and return the resulting expression."
(log 'inline-end result exp)
result)))))
+ (($ <lambda> src-proc meta orig-body)
+ ;; If there are multiple cases and one matches nargs, omit all the others.
+ (or (and
+ (lambda-case-alternate orig-body)
+ (let ((nargs (length orig-args)))
+ (let loop ((body orig-body))
+ (match body
+ (#f #f) ;; No matching case; an error.
+ (($ <lambda-case> src-case req opt rest kw inits gensyms case-body alt)
+ (cond (kw
+ ;; FIXME: Not handling keyword cases.
+ #f)
+ ((let ((nreq (length req)))
+ (if rest
+ (<= nreq nargs)
+ (<= nreq nargs (+ nreq (if opt (length opt) 0)))))
+ ;; Keep only this case.
+ (revisit-proc
+ (make-lambda
+ src-proc meta
+ (make-lambda-case src-case req opt rest kw inits gensyms case-body #f))))
+ (else (loop alt))))))))
+ (make-call src (for-call orig-proc) (map for-value orig-args))))
(($ <let> _ _ _ vals _)
;; Attempt to inline `let' in the operator position.
;;
diff --git a/test-suite/tests/peval.test b/test-suite/tests/peval.test
index a2e4975d9..8a8f0124a 100644
--- a/test-suite/tests/peval.test
+++ b/test-suite/tests/peval.test
@@ -1456,6 +1456,78 @@
;; <https://bugs.gnu.org/60522>.
(primcall make-vector)))
+(with-test-prefix "case-lambda"
+ ;; one case
+ (pass-if-peval
+ ((case-lambda (() 0)))
+ (const 0))
+
+ ;; middle
+ (pass-if-peval
+ ((case-lambda (() 0) ((a b) 1) ((a) 2)) 1 2)
+ (const 1))
+
+ ;; last
+ (pass-if-peval
+ ((case-lambda ((a b) 0) ((a) 1) (() 2)))
+ (const 2))
+
+ ;; first
+ (pass-if-peval
+ ((case-lambda ((a) 0) (() 1) ((a b) 2)) 1)
+ (const 0))
+
+ ;; rest arg
+ (pass-if-peval
+ ((case-lambda (args 0) ((a b) 1) ((a) 2)) 1 2)
+ (const 0))
+
+ ;; req before rest I
+ (pass-if-peval
+ ((case-lambda ((a b) 0) (args 1) ((a) 1)) 1 2)
+ (const 0))
+
+ ;; req before rest II
+ (pass-if-peval
+ ((case-lambda ((a) 0) (args 1) ((a b) 2)) 1 2)
+ (const 1))
+
+ ;; optional
+ (pass-if-peval
+ ((case-lambda* ((a #:optional x) 0) (args 1) ((a) 2)) 1 2)
+ (const 0))
+
+ ;; optional and rest, no match I
+ (pass-if-peval
+ ((case-lambda* ((a #:optional x . rest) 0) (args 1) ((a) 2)))
+ (const 1))
+
+ ;; optional and rest, match I
+ (pass-if-peval
+ ((case-lambda* (() 0) ((a #:optional x . rest) 1) ((a) 2)) 1)
+ (const 1))
+
+ ;; optional and rest, match II
+ (pass-if-peval
+ ((case-lambda* ((a #:optional x . rest) 0) (args 1) ((a) 2)) 1)
+ (const 0))
+
+ ;; optional and rest, match III
+ (pass-if-peval
+ ((case-lambda* ((a #:optional x . rest) 0) (args 1) ((a) 2)) 1 2)
+ (const 0))
+
+ ;; optional and rest, match IV
+ (pass-if-peval
+ ((case-lambda* ((a #:optional x . rest) 0) (args 1) ((a) 2)) 1 2 3)
+ (const 0))
+
+ ;; keyword cases survive
+ (pass-if (= 1 ((case-lambda* ((a b) 0) ((a #:key x) 1)) 0 #:x 1)))
+ (pass-if (= 0 ((case-lambda* ((a b c) 0) ((a #:key x) 1)) 0 #:x 1)))
+ (pass-if (= 0 ((case-lambda* ((a #:key x) 0) ((a b) 0)) 0 #:x 1)))
+ (pass-if (= 1 ((case-lambda* ((a #:key x) 0) ((a b c) 1)) 0 1 2))))
+
(with-test-prefix "eqv?"
(pass-if-peval (eqv? x #f)
(primcall eq? (toplevel x) (const #f)))
--
2.30.2
^ permalink raw reply related [flat|nested] 6+ messages in thread
* bug#61660: [feature request] optimization of case-lambda
2023-02-24 17:26 ` lloda
@ 2023-02-27 10:11 ` Ludovic Courtès
2023-02-27 17:19 ` lloda
0 siblings, 1 reply; 6+ messages in thread
From: Ludovic Courtès @ 2023-02-27 10:11 UTC (permalink / raw)
To: lloda; +Cc: 61660
Hi Daniel,
lloda <lloda@sarc.name> skribis:
> From 61ed612fb36108e395bdee4b1bbb46b49ef017b3 Mon Sep 17 00:00:00 2001
> From: Daniel Llorens <lloda@sarc.name>
> Date: Thu, 23 Feb 2023 17:38:10 +0100
> Subject: [PATCH] peval reduces some inlined case-lambda calls
>
> * module/language/tree-il/peval.scm (peval): Reduce multiple case lambda
> in <call> trees according to the number of arguments. Do not try to
> reduce case-lambda using keyword arguments.
> * test-suite/tests/peval.test: Tests.
[...]
> +++ b/module/language/tree-il/peval.scm
> @@ -1668,6 +1668,29 @@ top-level bindings from ENV and return the resulting expression."
>
> (log 'inline-end result exp)
> result)))))
> + (($ <lambda> src-proc meta orig-body)
> + ;; If there are multiple cases and one matches nargs, omit all the others.
> + (or (and
> + (lambda-case-alternate orig-body)
> + (let ((nargs (length orig-args)))
> + (let loop ((body orig-body))
> + (match body
> + (#f #f) ;; No matching case; an error.
> + (($ <lambda-case> src-case req opt rest kw inits gensyms case-body alt)
> + (cond (kw
> + ;; FIXME: Not handling keyword cases.
> + #f)
Maybe s/FIXME/XXX/ since it’s at most a limitation, certainly not a bug.
It LGTM and Andy already approved it on IRC, so go ahead!
Ludo’.
^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#61660: [feature request] optimization of case-lambda
2023-02-27 10:11 ` Ludovic Courtès
@ 2023-02-27 17:19 ` lloda
0 siblings, 0 replies; 6+ messages in thread
From: lloda @ 2023-02-27 17:19 UTC (permalink / raw)
To: Ludovic Courtès; +Cc: 61660-done
> On 27 Feb 2023, at 11:11, Ludovic Courtès <ludo@gnu.org> wrote:
>
> Hi Daniel,
>
> lloda <lloda@sarc.name> skribis:
>
>> From 61ed612fb36108e395bdee4b1bbb46b49ef017b3 Mon Sep 17 00:00:00 2001
>> From: Daniel Llorens <lloda@sarc.name>
>> Date: Thu, 23 Feb 2023 17:38:10 +0100
>> Subject: [PATCH] peval reduces some inlined case-lambda calls
>>
>> * module/language/tree-il/peval.scm (peval): Reduce multiple case lambda
>> in <call> trees according to the number of arguments. Do not try to
>> reduce case-lambda using keyword arguments.
>> * test-suite/tests/peval.test: Tests.
>
> [...]
>
>> +++ b/module/language/tree-il/peval.scm
>> @@ -1668,6 +1668,29 @@ top-level bindings from ENV and return the resulting expression."
>>
>> (log 'inline-end result exp)
>> result)))))
>> + (($ <lambda> src-proc meta orig-body)
>> + ;; If there are multiple cases and one matches nargs, omit all the others.
>> + (or (and
>> + (lambda-case-alternate orig-body)
>> + (let ((nargs (length orig-args)))
>> + (let loop ((body orig-body))
>> + (match body
>> + (#f #f) ;; No matching case; an error.
>> + (($ <lambda-case> src-case req opt rest kw inits gensyms case-body alt)
>> + (cond (kw
>> + ;; FIXME: Not handling keyword cases.
>> + #f)
>
> Maybe s/FIXME/XXX/ since it’s at most a limitation, certainly not a bug.
>
> It LGTM and Andy already approved it on IRC, so go ahead!
>
> Ludo’.
Apologies for not seeing this earlier. Pushed to 3b47f87618047ebb8812788c64a44877a4f2e0dd. Thanks!
- Daniel
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2023-02-27 17:19 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-01-20 19:16 bug#60971: build failure of v3.0.9rc1 on mac os 12.6 lloda
[not found] ` <handler.60971.B.167424218722695.ack@debbugs.gnu.org>
2023-02-23 19:00 ` bug#61660: [feature request] optimization of case-lambda lloda
2023-02-24 17:26 ` lloda
2023-02-27 10:11 ` Ludovic Courtès
2023-02-27 17:19 ` lloda
-- strict thread matches above, loose matches on Subject: below --
2023-02-20 17:56 lloda
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).