netfilter-devel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Duncan Roe <duncan_roe@optusnet.com.au>
To: pablo@netfilter.org
Cc: netfilter-devel@vger.kernel.org
Subject: [PATCH libnetfilter_queue 1/2] src: doc: Major re-work of user packet buffer documentation
Date: Mon, 18 Nov 2019 14:36:37 +1100	[thread overview]
Message-ID: <20191118033638.26472-1-duncan_roe@optusnet.com.au> (raw)

Divide functions into a hierarchy:
 top-level: Functions all programs that modify data will use
 2nd-level: Rarely-used functions
 3rd-level: Functions not to use (should have been declared static)
Only the top-level functions appear on the "User-space network packet buffer"
page, which looks a lot less daunting than it used to.

Parameter descriptions all match prototypes

All non-void functions have a "Returns" paragraph

Code change:

 pktb_alloc: set errno to EPROTONOSUPPORT before doing error return because
             protocol is not supported

Detailed other updates (top-level)

 pktb_alloc: - Add "Errors" para
             - Add "See also" para

 pktb_data, pktb_len: Add "appropriate use" line

 pktb_mangle: Add warning to use a different function unless mangling MAC hddr

 pktb_mangled: Add usage hint line

Detailed other updates (2nd-level)

 pktb_mac_header: Point out only for AF_BRIDGE

 pktb_tailroom: Point out no dynamic expansion

 pktb_transport_header: Add note that programmer must code to set this

Signed-off-by: Duncan Roe <duncan_roe@optusnet.com.au>
---
 src/extra/pktbuff.c | 154 +++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 117 insertions(+), 37 deletions(-)

diff --git a/src/extra/pktbuff.c b/src/extra/pktbuff.c
index f1f8323..26d7ca8 100644
--- a/src/extra/pktbuff.c
+++ b/src/extra/pktbuff.c
@@ -9,6 +9,7 @@
  * This code has been sponsored by Vyatta Inc. <http://www.vyatta.com>
  */
 
+#include <errno.h>
 #include <stdlib.h>
 #include <string.h> /* for memcpy */
 #include <stdbool.h>
@@ -30,7 +31,8 @@
 
 /**
  * pktb_alloc - allocate a new packet buffer
- * \param family Indicate what family, eg. AF_BRIDGE, AF_INET, AF_INET6, ...
+ * \param family Indicate what family. Currently supported families are
+ * AF_BRIDGE, AF_INET & AF_INET6.
  * \param data Pointer to packet data
  * \param len Packet length
  * \param extra Extra memory in the tail to be allocated (for mangling)
@@ -38,7 +40,13 @@
  * This function returns a packet buffer that contains the packet data and
  * some extra memory room in the tail (if requested).
  *
- * \return a pointer to a new queue handle or NULL on failure.
+ * \return Pointer to a new userspace packet buffer or NULL on failure.
+ * \par Errors
+ * __ENOMEM__ From __calloc__()
+ * \n
+ * __EPROTONOSUPPORT__ _family_ was __AF_BRIDGE__ and this is not an IP packet
+ * (v4 or v6)
+ * \sa __calloc__(3)
  */
 EXPORT_SYMBOL
 struct pkt_buff *pktb_alloc(int family, void *data, size_t len, size_t extra)
@@ -78,6 +86,7 @@ struct pkt_buff *pktb_alloc(int family, void *data, size_t len, size_t extra)
 			break;
 		default:
 			/* This protocol is unsupported. */
+			errno = EPROTONOSUPPORT;
 			free(pktb);
 			return NULL;
 		}
@@ -88,8 +97,12 @@ struct pkt_buff *pktb_alloc(int family, void *data, size_t len, size_t extra)
 }
 
 /**
- * pktb_data - return pointer to the beginning of the packet buffer
- * \param pktb Pointer to packet buffer
+ * pktb_data - get pointer to network packet
+ * \param pktb Pointer to userspace packet buffer
+ * \return Pointer to start of network packet data within __pktb__
+ * \par
+ * It is appropriate to use _pktb_data_ as the second argument of
+ * nfq_nlmsg_verdict_put_pkt()
  */
 EXPORT_SYMBOL
 uint8_t *pktb_data(struct pkt_buff *pktb)
@@ -99,7 +112,11 @@ uint8_t *pktb_data(struct pkt_buff *pktb)
 
 /**
  * pktb_len - return length of the packet buffer
- * \param pktb Pointer to packet buffer
+ * \param pktb Pointer to userspace packet buffer
+ * \return Length of packet contained within __pktb__
+ * \par
+ * It is appropriate to use _pktb_len_ as the third argument of
+ * nfq_nlmsg_verdict_put_pkt()
  */
 EXPORT_SYMBOL
 uint32_t pktb_len(struct pkt_buff *pktb)
@@ -109,7 +126,7 @@ uint32_t pktb_len(struct pkt_buff *pktb)
 
 /**
  * pktb_free - release packet buffer
- * \param pktb Pointer to packet buffer
+ * \param pktb Pointer to userspace packet buffer
  */
 EXPORT_SYMBOL
 void pktb_free(struct pkt_buff *pktb)
@@ -118,8 +135,35 @@ void pktb_free(struct pkt_buff *pktb)
 }
 
 /**
- * pktb_push - update pointer to the beginning of the packet buffer
- * \param pktb Pointer to packet buffer
+ * \defgroup otherfns Other functions
+ *
+ * The library provides a number of other functions which many user-space
+ * programs will never need. These divide into 2 groups:
+ * \n
+ * 1. Functions to get values of members of opaque __struct pktbuff__, described
+ * below
+ * \n
+ * 2. Internal functions, described in Module __Internal functions__
+ *
+ * @{
+ */
+
+/**
+ * \defgroup uselessfns Internal functions
+ *
+ * \warning Do not use these functions. Instead, always use the mangle
+ * function appropriate to the level at which you are working.
+ * \n
+ * pktb_mangle() uses all the below functions except _pktb_pull_, which is not
+ * used by anything.
+ *
+ * @{
+ */
+
+/**
+ * pktb_push - decrement pointer to packet buffer
+ * \param pktb Pointer to userspace packet buffer
+ * \param len Number of bytes to subtract from packet start address
  */
 EXPORT_SYMBOL
 void pktb_push(struct pkt_buff *pktb, unsigned int len)
@@ -129,8 +173,9 @@ void pktb_push(struct pkt_buff *pktb, unsigned int len)
 }
 
 /**
- * pktb_pull - update pointer to the beginning of the packet buffer
- * \param pktb Pointer to packet buffer
+ * pktb_pull - increment pointer to packet buffer
+ * \param pktb Pointer to userspace packet buffer
+ * \param len Number of bytes to add to packet start address
  */
 EXPORT_SYMBOL
 void pktb_pull(struct pkt_buff *pktb, unsigned int len)
@@ -141,7 +186,8 @@ void pktb_pull(struct pkt_buff *pktb, unsigned int len)
 
 /**
  * pktb_put - add extra bytes to the tail of the packet buffer
- * \param pktb Pointer to packet buffer
+ * \param pktb Pointer to userspace packet buffer
+ * \param len Number of bytes to add to packet tail (and length)
  */
 EXPORT_SYMBOL
 void pktb_put(struct pkt_buff *pktb, unsigned int len)
@@ -152,7 +198,8 @@ void pktb_put(struct pkt_buff *pktb, unsigned int len)
 
 /**
  * pktb_trim - set new length for this packet buffer
- * \param pktb Pointer to packet buffer
+ * \param pktb Pointer to userspace packet buffer
+ * \param len New packet length (tail is adjusted to reflect this)
  */
 EXPORT_SYMBOL
 void pktb_trim(struct pkt_buff *pktb, unsigned int len)
@@ -162,8 +209,17 @@ void pktb_trim(struct pkt_buff *pktb, unsigned int len)
 }
 
 /**
- * pktb_tailroom - get room in bytes in the tail of the packet buffer
- * \param pktb Pointer to packet buffer
+ * @}
+ */
+
+/**
+ * pktb_tailroom - get room available for packet expansion
+ * \param pktb Pointer to userspace packet buffer
+ * \return room in bytes after the tail of the packet buffer
+ * \n
+ * This starts off as the __extra__ argument to pktb_alloc().
+ * Programmers should ensure this __extra__ argument is sufficient for any
+ * packet mangle, as packet buffers cannot be expanded dynamically.
  */
 EXPORT_SYMBOL
 unsigned int pktb_tailroom(struct pkt_buff *pktb)
@@ -172,8 +228,11 @@ unsigned int pktb_tailroom(struct pkt_buff *pktb)
 }
 
 /**
- * pktb_mac_header - return pointer to layer 2 header (if any)
- * \param pktb Pointer to packet buffer
+ * pktb_mac_header - get address of layer 2 header (if any)
+ * \param pktb Pointer to userspace packet buffer
+ * \return Pointer to MAC header or NULL if no such header present.
+ * \n
+ * Only packet buffers in family __AF_BRIDGE__ have a non-NULL MAC header.
  */
 EXPORT_SYMBOL
 uint8_t *pktb_mac_header(struct pkt_buff *pktb)
@@ -182,8 +241,10 @@ uint8_t *pktb_mac_header(struct pkt_buff *pktb)
 }
 
 /**
- * pktb_network_header - return pointer to layer 3 header
- * \param pktb Pointer to packet buffer
+ * pktb_network_header - get address of layer 3 header
+ * \param pktb Pointer to userspace packet buffer
+ * \return Pointer to layer 3 header or NULL if the packet buffer was created
+ * with an unsupported family
  */
 EXPORT_SYMBOL
 uint8_t *pktb_network_header(struct pkt_buff *pktb)
@@ -192,8 +253,13 @@ uint8_t *pktb_network_header(struct pkt_buff *pktb)
 }
 
 /**
- * pktb_transport_header - return pointer to layer 4 header (if any)
- * \param pktb Pointer to packet buffer
+ * pktb_transport_header - get address of layer 4 header (if known)
+ * \param pktb Pointer to userspace packet buffer
+ * \return Pointer to layer 4 header or NULL if not (yet) set
+ * \note
+ * Unlike the lower-level headers, it is the programmer's responsibility to
+ * create the level 4 (transport) header pointer by caling e.g.
+ * nfq_ip_set_transport_header()
  */
 EXPORT_SYMBOL
 uint8_t *pktb_transport_header(struct pkt_buff *pktb)
@@ -201,6 +267,10 @@ uint8_t *pktb_transport_header(struct pkt_buff *pktb)
 	return pktb->transport_header;
 }
 
+/**
+ * @}
+ */
+
 static int pktb_expand_tail(struct pkt_buff *pkt, int extra)
 {
 	/* No room in packet, cannot mangle it. We don't support dynamic
@@ -228,19 +298,24 @@ static int enlarge_pkt(struct pkt_buff *pkt, unsigned int extra)
 
 /**
  * pktb_mangle - adjust contents of a packet
- * \param pkt Pointer to packet buffer
+ * \param pktb Pointer to userspace packet buffer
  * \param dataoff Offset to layer 4 header. Specify zero to access layer 3 (IP)
- * header
+ * header (layer 2 for family \b AF_BRIDGE)
  * \param match_offset Further offset to content that you want to mangle
  * \param match_len Length of the existing content you want to mangle
  * \param rep_buffer Pointer to data you want to use to replace current content
  * \param rep_len Length of data you want to use to replace current content
  * \returns 1 for success and 0 for failure. Failure will occur if the \b extra
- * argument to the pktb_alloc() call that created \b pkt is less than the excess
- * of \b rep_len over \b match_len
+ * argument to the pktb_alloc() call that created \b pktb is less than the
+ * excess of \b rep_len over \b match_len
+ \warning pktb_mangle does not update any checksums. Developers should use the
+ appropriate mangler for the protocol level: nfq_ip_mangle(),
+ nfq_tcp_mangle_ipv4() or nfq_udp_mangle_ipv4(). IPv6 versions are planned.
+ \n
+ It is appropriate to use pktb_mangle to change the MAC header.
  */
 EXPORT_SYMBOL
-int pktb_mangle(struct pkt_buff *pkt,
+int pktb_mangle(struct pkt_buff *pktb,
 		unsigned int dataoff,
 		unsigned int match_offset,
 		unsigned int match_len,
@@ -250,39 +325,44 @@ int pktb_mangle(struct pkt_buff *pkt,
 	unsigned char *data;
 
 	if (rep_len > match_len &&
-	    rep_len - match_len > pktb_tailroom(pkt) &&
-	    !enlarge_pkt(pkt, rep_len - match_len))
+	    rep_len - match_len > pktb_tailroom(pktb) &&
+	    !enlarge_pkt(pktb, rep_len - match_len))
 		return 0;
 
-	data = pkt->network_header + dataoff;
+	data = pktb->network_header + dataoff;
 
 	/* move post-replacement */
 	memmove(data + match_offset + rep_len,
 		data + match_offset + match_len,
-		pkt->tail - (pkt->network_header + dataoff +
+		pktb->tail - (pktb->network_header + dataoff +
 			     match_offset + match_len));
 
 	/* insert data from buffer */
 	memcpy(data + match_offset, rep_buffer, rep_len);
 
-	/* update pkt info */
+	/* update packet info */
 	if (rep_len > match_len)
-		pktb_put(pkt, rep_len - match_len);
+		pktb_put(pktb, rep_len - match_len);
 	else
-		pktb_trim(pkt, pkt->len + rep_len - match_len);
+		pktb_trim(pktb, pktb->len + rep_len - match_len);
 
-	pkt->mangled = true;
+	pktb->mangled = true;
 	return 1;
 }
 
 /**
- * pktb_mangled - return true if packet has been mangled
- * \param pktb Pointer to packet buffer
+ * pktb_mangled - test whether packet has been mangled
+ * \param pktb Pointer to userspace packet buffer
+ * \return __true__ if packet has been mangled (modified), else __false__
+ * \par
+ * When assembling a verdict, it is not necessary to return the contents of
+ * un-modified packets. Use _pktb_mangled_ to decide whether packet contents
+ * need to be returned.
  */
 EXPORT_SYMBOL
-bool pktb_mangled(const struct pkt_buff *pkt)
+bool pktb_mangled(const struct pkt_buff *pktb)
 {
-	return pkt->mangled;
+	return pktb->mangled;
 }
 
 /**
-- 
2.14.5


             reply	other threads:[~2019-11-18  3:36 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-18  3:36 Duncan Roe [this message]
2019-11-18  3:36 ` [PATCH libnetfilter_queue 2/2] Minor tweak to pktb_len function description Duncan Roe
2019-12-07 17:58   ` Pablo Neira Ayuso
2019-12-07  0:32 ` [PATCH libnetfilter_queue 1/2] src: doc: Major re-work of user packet buffer documentation Duncan Roe
2019-12-07 17:58 ` 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=20191118033638.26472-1-duncan_roe@optusnet.com.au \
    --to=duncan_roe@optusnet.com.au \
    --cc=netfilter-devel@vger.kernel.org \
    --cc=pablo@netfilter.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 NNTP newsgroup(s).