All of lore.kernel.org
 help / color / mirror / Atom feed
From: Thomas Haller <thaller@redhat.com>
To: NetFilter <netfilter-devel@vger.kernel.org>
Cc: Thomas Haller <thaller@redhat.com>
Subject: [PATCH nft 2/5] datatype: don't clone static name/desc strings for datatype
Date: Wed, 27 Sep 2023 21:57:25 +0200	[thread overview]
Message-ID: <20230927200143.3798124-3-thaller@redhat.com> (raw)
In-Reply-To: <20230927200143.3798124-1-thaller@redhat.com>

Avoid cloning static strings for "struct datatype". With
concat_type_alloc(), the name/desc are generated dynamically and need to
be allocated. However, datatype_clone() only reuses the original
name/desc strings. If those strings are static already, we don't need to
clone them.

Note that there are no other places that also want to change or set the
name/desc. If there were, they would need to handle the new fact that
the strings may or may not be dynamically allocated.

Signed-off-by: Thomas Haller <thaller@redhat.com>
---
 include/datatype.h |  2 ++
 src/datatype.c     | 15 ++++++++++-----
 2 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/include/datatype.h b/include/datatype.h
index b53a739e1e6c..465ade290652 100644
--- a/include/datatype.h
+++ b/include/datatype.h
@@ -132,6 +132,7 @@ struct parse_ctx;
  * @f_prefix:	preferred representation for ranges is a prefix
  * @f_alloc:	whether the instance is dynamically allocated. If not, datatype_get() and
  *		datatype_free() are NOPs.
+ * @f_allocated_strings: whether @name and @desc are heap allocated or static.
  * @name:	type name
  * @desc:	type description
  * @basetype:	basetype for subtypes, determines type compatibility
@@ -153,6 +154,7 @@ struct datatype {
 	enum byteorder			byteorder;
 	bool				f_prefix:1;
 	bool				f_alloc:1;
+	bool				f_allocated_strings:1;
 
 	const char			*name;
 	const char			*desc;
diff --git a/src/datatype.c b/src/datatype.c
index 464eb49171c6..1c557a06c751 100644
--- a/src/datatype.c
+++ b/src/datatype.c
@@ -1244,8 +1244,10 @@ struct datatype *datatype_clone(const struct datatype *orig_dtype)
 
 	dtype = xzalloc(sizeof(*dtype));
 	*dtype = *orig_dtype;
-	dtype->name = xstrdup(orig_dtype->name);
-	dtype->desc = xstrdup(orig_dtype->desc);
+	if (orig_dtype->f_allocated_strings) {
+		dtype->name = xstrdup(orig_dtype->name);
+		dtype->desc = xstrdup(orig_dtype->desc);
+	}
 	dtype->f_alloc = true;
 	dtype->refcnt = 1;
 
@@ -1265,8 +1267,10 @@ void datatype_free(const struct datatype *ptr)
 	if (--dtype->refcnt > 0)
 		return;
 
-	xfree(dtype->name);
-	xfree(dtype->desc);
+	if (dtype->f_allocated_strings) {
+		xfree(dtype->name);
+		xfree(dtype->desc);
+	}
 	xfree(dtype);
 }
 
@@ -1299,7 +1303,8 @@ const struct datatype *concat_type_alloc(uint32_t type)
 	dtype		= datatype_alloc();
 	dtype->type	= type;
 	dtype->size	= size;
-	dtype->subtypes = subtypes;
+	dtype->subtypes	= subtypes;
+	dtype->f_allocated_strings = true;
 	dtype->name	= xstrdup(name);
 	dtype->desc	= xstrdup(desc);
 	dtype->parse	= concat_type_parse;
-- 
2.41.0


  parent reply	other threads:[~2023-09-27 20:02 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-09-27 19:57 [PATCH nft 0/5] more various cleanups related to struct datatype Thomas Haller
2023-09-27 19:57 ` [PATCH nft 1/5] datatype: make "flags" field of datatype struct simple booleans Thomas Haller
2024-08-19 22:41   ` Pablo Neira Ayuso
2023-09-27 19:57 ` Thomas Haller [this message]
2023-09-27 19:57 ` [PATCH nft 3/5] datatype: don't clone datatype in set_datatype_alloc() if byteorder already matches Thomas Haller
2023-09-27 19:57 ` [PATCH nft 4/5] datatype: extend set_datatype_alloc() to change size Thomas Haller
2023-09-27 19:57 ` [PATCH nft 5/5] datatype: use xmalloc() for allocating datatype in datatype_clone() Thomas Haller
2023-09-28 19:11   ` Pablo Neira Ayuso

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=20230927200143.3798124-3-thaller@redhat.com \
    --to=thaller@redhat.com \
    --cc=netfilter-devel@vger.kernel.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 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.