From: "Mattias Engdegård" <mattiase@acm.org>
To: 35770@debbugs.gnu.org
Cc: Stefan Monnier <monnier@iro.umontreal.ca>, vibhavp@gmail.com
Subject: bug#35770: [PATCH] Broken duplicate case elimination in switch byte-compilation
Date: Fri, 17 May 2019 11:33:41 +0200 [thread overview]
Message-ID: <91DFE4D0-BDB6-4365-BA51-A2A3F209177C@acm.org> (raw)
[-- Attachment #1: Type: text/plain, Size: 543 bytes --]
The byte-code compiler attempts to eliminate duplicated cases when turning a `cond' into a switch table, as in
(cond ((eq x 'a) 1)
(eq x 'b) 2)
(eq x 'a) 3)) ; remove
but it doesn't work properly, because of a confusion between expressions and their values, and a logic error that would have discarded the entire table instead of just skipped the duplicate.
This patch attempts to rectify that. I also removed a seemingly redundant condition, the (consp condition), which should always be true at that point.
[-- Attachment #2: 0001-Correctly-eliminate-duplicate-cases-in-switch-compil.patch --]
[-- Type: application/octet-stream, Size: 2503 bytes --]
From adb7adf2271e65023adc0ce5d6251054b55c6452 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mattias=20Engdeg=C3=A5rd?= <mattiase@acm.org>
Date: Fri, 17 May 2019 11:25:06 +0200
Subject: [PATCH] Correctly eliminate duplicate cases in switch compilation
Fix code mistakes that prevented the correct elimination of duplicated
cases when compiling a `cond' form to a switch bytecode, as in
(cond ((eq x 'a) 1)
((eq x 'b) 2)
((eq x 'a) 3)) ; should be elided
* lisp/emacs-lisp/bytecomp.el (byte-compile-cond-vars): Return obj2 eval'ed.
(byte-compile-cond-jump-table-info):
Discard redundant condition. Use `obj2' as evaluated.
Discard duplicated cases instead of failing the table generation.
---
lisp/emacs-lisp/bytecomp.el | 13 ++++++-------
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/lisp/emacs-lisp/bytecomp.el b/lisp/emacs-lisp/bytecomp.el
index e76baf5ed0..7f7cbe5a68 100644
--- a/lisp/emacs-lisp/bytecomp.el
+++ b/lisp/emacs-lisp/bytecomp.el
@@ -4091,8 +4091,8 @@ that suppresses all warnings during execution of BODY."
;; and the other is a constant expression whose value can be
;; compared with `eq' (with `macroexp-const-p').
(or
- (and (symbolp obj1) (macroexp-const-p obj2) (cons obj1 obj2))
- (and (symbolp obj2) (macroexp-const-p obj1) (cons obj2 obj1))))
+ (and (symbolp obj1) (macroexp-const-p obj2) (cons obj1 (eval obj2)))
+ (and (symbolp obj2) (macroexp-const-p obj1) (cons obj2 (eval obj1)))))
(defconst byte-compile--default-val (cons nil nil) "A unique object.")
@@ -4121,12 +4121,11 @@ Return a list of the form ((TEST . VAR) ((VALUE BODY) ...))"
(unless prev-test
(setq prev-test test))
(if (and obj1 (memq test '(eq eql equal))
- (consp condition)
(eq test prev-test)
- (eq obj1 prev-var)
- ;; discard duplicate clauses
- (not (assq obj2 cases)))
- (push (list (if (consp obj2) (eval obj2) obj2) body) cases)
+ (eq obj1 prev-var))
+ ;; discard duplicate clauses
+ (unless (assq obj2 cases)
+ (push (list obj2 body) cases))
(if (and (macroexp-const-p condition) condition)
(progn (push (list byte-compile--default-val
(or body `(,condition)))
--
2.20.1 (Apple Git-117)
next reply other threads:[~2019-05-17 9:33 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-17 9:33 Mattias Engdegård [this message]
2019-05-17 12:16 ` bug#35770: [PATCH] Broken duplicate case elimination in switch byte-compilation Stefan Monnier
2019-05-17 14:48 ` Mattias Engdegård
2019-05-22 10:52 ` Mattias Engdegård
2019-05-22 10:58 ` Eli Zaretskii
2019-05-22 11:11 ` Mattias Engdegård
2019-05-22 11:21 ` Noam Postavsky
2019-05-22 11:23 ` Eli Zaretskii
2019-05-22 14:19 ` Stefan Monnier
2019-05-26 17:05 ` Mattias Engdegård
2019-05-26 18:43 ` Eli Zaretskii
2019-05-26 22:06 ` Stefan Monnier
2019-05-27 11:27 ` Mattias Engdegård
2019-06-01 13:58 ` Noam Postavsky
2019-06-01 15:47 ` Eli Zaretskii
2019-06-01 21:53 ` Noam Postavsky
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=91DFE4D0-BDB6-4365-BA51-A2A3F209177C@acm.org \
--to=mattiase@acm.org \
--cc=35770@debbugs.gnu.org \
--cc=monnier@iro.umontreal.ca \
--cc=vibhavp@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this external index
https://git.savannah.gnu.org/cgit/emacs.git
https://git.savannah.gnu.org/cgit/emacs/org-mode.git
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.