From: Sebastian Huber <sebastian.huber@embedded-brains.de>
To: newlib@sourceware.org
Cc: devel@rtems.org
Subject: [PATCH v2 4/4] sys/tree.h: Red child with black sibling rotations
Date: Tue, 5 Oct 2021 17:33:44 +0200 [thread overview]
Message-ID: <20211005153344.91266-6-sebastian.huber@embedded-brains.de> (raw)
In-Reply-To: <20211005153344.91266-1-sebastian.huber@embedded-brains.de>
Add specialized rotations RB_RED_ROTATE_LEFT() and RB_RED_ROTATE_RIGHT() which
may be used if we rotate a red child which has a black sibling. Such a red
node must have at least two child nodes so that the following red-black tree
invariant is fulfilled:
Every path from a given node to any of its descendant NULL nodes goes through
the same number of black nodes.
PARENT
/ \
BLACK RED
/ \
BLACK BLACK
---
newlib/libc/include/sys/tree.h | 43 ++++++++++++++++++++++++++++++++--
1 file changed, 41 insertions(+), 2 deletions(-)
diff --git a/newlib/libc/include/sys/tree.h b/newlib/libc/include/sys/tree.h
index f6190f7f4..7a3b55bdf 100644
--- a/newlib/libc/include/sys/tree.h
+++ b/newlib/libc/include/sys/tree.h
@@ -412,6 +412,45 @@ struct { \
RB_AUGMENT(right); \
} while (/*CONSTCOND*/ 0)
+/*
+ * The RB_RED_ROTATE_LEFT() and RB_RED_ROTATE_RIGHT() rotations are specialized
+ * versions of RB_ROTATE_LEFT() and RB_ROTATE_RIGHT() which may be used if we
+ * rotate a red child element which has a black sibling. Such a red node must
+ * have at least two child nodes so that the following red-black tree invariant
+ * is fulfilled:
+ *
+ * Every path from a given node to any of its descendant NULL nodes goes
+ * through the same number of black nodes.
+ *
+ * parent
+ * / \
+ * BLACK RED (elm)
+ * / \
+ * BLACK BLACK
+ */
+
+#define RB_RED_ROTATE_LEFT(head, elm, tmp, field) do { \
+ (tmp) = RB_RIGHT(elm, field); \
+ RB_RIGHT(elm, field) = RB_LEFT(tmp, field); \
+ RB_SET_PARENT(RB_RIGHT(elm, field), elm, field); \
+ RB_SET_PARENT(tmp, RB_PARENT(elm, field), field); \
+ RB_SWAP_CHILD(head, elm, tmp, field); \
+ RB_LEFT(tmp, field) = (elm); \
+ RB_SET_PARENT(elm, tmp, field); \
+ RB_AUGMENT(elm); \
+} while (/*CONSTCOND*/ 0)
+
+#define RB_RED_ROTATE_RIGHT(head, elm, tmp, field) do { \
+ (tmp) = RB_LEFT(elm, field); \
+ RB_LEFT(elm, field) = RB_RIGHT(tmp, field); \
+ RB_SET_PARENT(RB_LEFT(elm, field), elm, field); \
+ RB_SET_PARENT(tmp, RB_PARENT(elm, field), field); \
+ RB_SWAP_CHILD(head, elm, tmp, field); \
+ RB_RIGHT(tmp, field) = (elm); \
+ RB_SET_PARENT(elm, tmp, field); \
+ RB_AUGMENT(elm); \
+} while (/*CONSTCOND*/ 0)
+
/* Generates prototypes and inline functions */
#define RB_PROTOTYPE(name, type, field, cmp) \
RB_PROTOTYPE_INTERNAL(name, type, field, cmp,)
@@ -526,7 +565,7 @@ name##_RB_REMOVE_COLOR(struct name *head, struct type *parent) \
tmp = RB_RIGHT(parent, field); \
if (RB_COLOR(tmp, field) == RB_RED) { \
RB_SET_BLACKRED(tmp, parent, field); \
- RB_ROTATE_LEFT(head, parent, tmp, field);\
+ RB_RED_ROTATE_LEFT(head, parent, tmp, field); \
tmp = RB_RIGHT(parent, field); \
} \
if (RB_ISRED(RB_RIGHT(tmp, field), field)) \
@@ -552,7 +591,7 @@ name##_RB_REMOVE_COLOR(struct name *head, struct type *parent) \
tmp = RB_LEFT(parent, field); \
if (RB_COLOR(tmp, field) == RB_RED) { \
RB_SET_BLACKRED(tmp, parent, field); \
- RB_ROTATE_RIGHT(head, parent, tmp, field);\
+ RB_RED_ROTATE_RIGHT(head, parent, tmp, field); \
tmp = RB_LEFT(parent, field); \
} \
if (RB_ISRED(RB_LEFT(tmp, field), field)) \
--
2.26.2
next prev parent reply other threads:[~2021-10-05 15:33 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-05 15:33 [PATCH v2 0/4] Optimize red-black tree insert/extract Sebastian Huber
2021-10-05 15:33 ` [PATCH v2] Align *utime*() with POSIX/glibc Sebastian Huber
2021-10-05 15:35 ` Sebastian Huber
2021-10-05 15:33 ` [PATCH v2 1/4] sys/tree.h: Simplify loop condition Sebastian Huber
2021-10-05 15:33 ` [PATCH v2 2/4] sys/tree.h: Simplify chain of conditions Sebastian Huber
2021-10-05 15:33 ` [PATCH v2 3/4] sys/tree.h: Add parent rotations Sebastian Huber
2021-10-05 15:33 ` Sebastian Huber [this message]
2021-10-06 18:10 ` [PATCH v2 0/4] Optimize red-black tree insert/extract Jeff Johnston
2021-10-07 12:10 ` Sebastian Huber
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=20211005153344.91266-6-sebastian.huber@embedded-brains.de \
--to=sebastian.huber@embedded-brains.de \
--cc=devel@rtems.org \
--cc=newlib@sourceware.org \
/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.
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).