Make TSO segment transmit size decisions at send time not earlier.
The basic scheme is that we try to build as large a TSO frame as
possible when pulling in the user data, but the size of the TSO frame
output to the card is determined at transmit time.
This is guided by tp->xmit_size_goal. It is always set to a multiple
of MSS and tells sendmsg/sendpage how large an SKB to try and build.
Later, tcp_write_xmit() and tcp_push_one() chop up the packet if
necessary and conditions warrant. These routines can also decide to
"defer" in order to wait for more ACKs to arrive and thus allow larger
TSO frames to be emitted.
A general observation is that TSO elongates the pipe, thus requiring a
larger congestion window and larger buffering especially at the sender
side. Therefore, it is important that applications 1) get a large
enough socket send buffer (this is accomplished by our dynamic send
buffer expansion code) 2) do large enough writes.
Signed-off-by: David S. Miller <davem@davemloft.net>
'nonagle' should be passed to the tcp_snd_test() function
as 'TCP_NAGLE_PUSH' if we are checking an SKB not at the
tail of the write_queue. This is because Nagle does not
apply to such frames since we cannot possibly tack more
data onto them.
However, while doing this __tcp_push_pending_frames() makes
all of the packets in the write_queue use this modified
'nonagle' value.
Fix the bug and simplify this function by just calling
tcp_write_xmit() directly if sk_send_head is non-NULL.
As a result, we can now make tcp_data_snd_check() just call
tcp_push_pending_frames() instead of the specialized
__tcp_data_snd_check().
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp_write_xmit() uses tcp_current_mss(), but some of it's callers,
namely __tcp_push_pending_frames(), already has this value available
already.
While we're here, fix the "cur_mss" argument to be "unsigned int"
instead of plain "unsigned".
Signed-off-by: David S. Miller <davem@davemloft.net>
The tcp_cwnd_validate() function should only be invoked
if we actually send some frames, yet __tcp_push_pending_frames()
will always invoke it. tcp_write_xmit() does the call for us,
so the call here can simply be removed.
Also, tcp_write_xmit() can be marked static.
Signed-off-by: David S. Miller <davem@davemloft.net>
It reimplements portions of tcp_snd_check(), so it
we move it to tcp_output.c we can consolidate it's
logic much easier in a later change.
Signed-off-by: David S. Miller <davem@davemloft.net>
This just moves the code into tcp_output.c, no code logic changes are
made by this patch.
Using this as a baseline, we can begin to untangle the mess of
comparisons for the Nagle test et al. We will also be able to reduce
all of the redundant computation that occurs when outputting data
packets.
Signed-off-by: David S. Miller <davem@davemloft.net>
On each packet output, we call tcp_dec_quickack_mode()
if the ACK flag is set. It drops tp->ack.quick until
it hits zero, at which time we deflate the ATO value.
When doing TSO, we are emitting multiple packets with
ACK set, so we should decrement tp->ack.quick that many
segments.
Note that, unlike this case, tcp_enter_cwr() should not
take the tcp_skb_pcount(skb) into consideration. That
function, one time, readjusts tp->snd_cwnd and moves
into TCP_CA_CWR state.
Signed-off-by: David S. Miller <davem@davemloft.net>
The ideal and most optimal layout for an SKB when doing
scatter-gather is to put all the headers at skb->data, and
all the user data in the page array.
This makes SKB splitting and combining extremely simple,
especially before a packet goes onto the wire the first
time.
So, when sk_stream_alloc_pskb() is given a zero size, make
sure there is no skb_tailroom(). This is achieved by applying
SKB_DATA_ALIGN() to the header length used here.
Next, make select_size() in TCP output segmentation use a
length of zero when NETIF_F_SG is true on the outgoing
interface.
Signed-off-by: David S. Miller <davem@davemloft.net>
I suspect "#define __ARGS(x) ()" was deprecated before I was born.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Adds qdisc_alloc() to share code between qdisc_create()
and qdisc_create_dflt(). Hides the qdisc alignment behind
macros and makes use of them.
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
This crept in with the resync-to-mainline. Nothing uses 802.11-crypt in
mainline, so we can safely comment it out for now.
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove two more unused IPV6_AUTHHDR option things,
which I failed to remove them last time,
plus, mark IPV6_AUTHHDR obsolete.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Another rollup of patches which give various symbols static scope
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Allow using setsockopt to set TCP congestion control to use on a per
socket basis.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow TCP to have multiple pluggable congestion control algorithms.
Algorithms are defined by a set of operations and can be built in
or modules. The legacy "new RENO" algorithm is used as a starting
point and fallback.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch is a follow up to patch 1 regarding "Selective Sub Address
matching with call user data". It allows use of the Fast-Select-Acceptance
optional user facility for X.25.
This patch just implements fast select with no restriction on response
(NRR). What this means (according to ITU-T Recomendation 10/96 section
6.16) is that if in an incoming call packet, the relevant facility bits are
set for fast-select-NRR, then the called DTE can issue a direct response to
the incoming packet using a call-accepted packet that contains
call-user-data. This patch allows such a response.
The called DTE can also respond with a clear-request packet that contains
call-user-data. However, this feature is currently not implemented by the
patch.
How is Fast Select Acceptance used?
By default, the system does not allow fast select acceptance (as before).
To enable a response to fast select acceptance,
After a listen socket in created and bound as follows
socket(AF_X25, SOCK_SEQPACKET, 0);
bind(call_soc, (struct sockaddr *)&locl_addr, sizeof(locl_addr));
but before a listen system call is made, the following ioctl should be used.
ioctl(call_soc,SIOCX25CALLACCPTAPPRV);
Now the listen system call can be made
listen(call_soc, 4);
After this, an incoming-call packet will be accepted, but no call-accepted
packet will be sent back until the following system call is made on the socket
that accepts the call
ioctl(vc_soc,SIOCX25SENDCALLACCPT);
The network (or cisco xot router used for testing here) will allow the
application server's call-user-data in the call-accepted packet,
provided the call-request was made with Fast-select NRR.
Signed-off-by: Shaun Pereira <spereira@tusc.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
From: Shaun Pereira <spereira@tusc.com.au>
This is the first (independent of the second) patch of two that I am
working on with x25 on linux (tested with xot on a cisco router). Details
are as follows.
Current state of module:
A server using the current implementation (2.6.11.7) of the x25 module will
accept a call request/ incoming call packet at the listening x.25 address,
from all callers to that address, as long as NO call user data is present
in the packet header.
If the server needs to choose to accept a particular call request/ incoming
call packet arriving at its listening x25 address, then the kernel has to
allow a match of call user data present in the call request packet with its
own. This is required when multiple servers listen at the same x25 address
and device interface. The kernel currently matches ALL call user data, if
present.
Current Changes:
This patch is a follow up to the patch submitted previously by Andrew
Hendry, and allows the user to selectively control the number of octets of
call user data in the call request packet, that the kernel will match. By
default no call user data is matched, even if call user data is present.
To allow call user data matching, a cudmatchlength > 0 has to be passed
into the kernel after which the passed number of octets will be matched.
Otherwise the kernel behavior is exactly as the original implementation.
This patch also ensures that as is normally the case, no call user data
will be present in the Call accepted / call connected packet sent back to
the caller
Future Changes on next patch:
There are cases however when call user data may be present in the call
accepted packet. According to the X.25 recommendation (ITU-T 10/96)
section 5.2.3.2 call user data may be present in the call accepted packet
provided the fast select facility is used. My next patch will include this
fast select utility and the ability to send up to 128 octets call user data
in the call accepted packet provided the fast select facility is used. I
am currently testing this, again with xot on linux and cisco.
Signed-off-by: Shaun Pereira <spereira@tusc.com.au>
(With a fix from Alexey Dobriyan <adobriyan@gmail.com>)
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch implements a number of smp_processor_id() cleanup ideas that
Arjan van de Ven and I came up with.
The previous __smp_processor_id/_smp_processor_id/smp_processor_id API
spaghetti was hard to follow both on the implementational and on the
usage side.
Some of the complexity arose from picking wrong names, some of the
complexity comes from the fact that not all architectures defined
__smp_processor_id.
In the new code, there are two externally visible symbols:
- smp_processor_id(): debug variant.
- raw_smp_processor_id(): nondebug variant. Replaces all existing
uses of _smp_processor_id() and __smp_processor_id(). Defined
by every SMP architecture in include/asm-*/smp.h.
There is one new internal symbol, dependent on DEBUG_PREEMPT:
- debug_smp_processor_id(): internal debug variant, mapped to
smp_processor_id().
Also, i moved debug_smp_processor_id() from lib/kernel_lock.c into a new
lib/smp_processor_id.c file. All related comments got updated and/or
clarified.
I have build/boot tested the following 8 .config combinations on x86:
{SMP,UP} x {PREEMPT,!PREEMPT} x {DEBUG_PREEMPT,!DEBUG_PREEMPT}
I have also build/boot tested x64 on UP/PREEMPT/DEBUG_PREEMPT. (Other
architectures are untested, but should work just fine.)
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Essentially netlink at the moment always reports a pid and sequence of 0
always for v6 route activities.
To understand the repurcassions of this look at:
http://lists.quagga.net/pipermail/quagga-dev/2005-June/003507.html
While fixing this, i took the liberty to resolve the outstanding issue
of IPV6 routes inserted via ioctls to have the correct pids as well.
This patch tries to behave as close as possible to the v4 routes i.e
maintains whatever PID the socket issuing the command owns as opposed to
the process. That made the patch a little bulky.
I have tested against both netlink derived utility to add/del routes as
well as ioctl derived one. The Quagga folks have tested against quagga.
This fixes the problem and so far hasnt been detected to introduce any
new issues.
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Below is a more generic patch to do fib_lookup via netlink. For others
we should say that we discussed this as a way to verify route selection.
It's also possible there are others uses for this.
In short the fist half of struct fib_result_nl is filled in by caller
and netlink call fills in the other half and returns it.
In case anyone is interested there is a corresponding user app to compare
the full routing table this was used to test implementation of the LC-trie.
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the xfrm_state_afinfo->init_flags hook which allows
each address family to perform any common initialisation that does
not require a corresponding destructor call.
It will be used subsequently to set the XFRM_STATE_NOPMTUDISC flag
in IPv4.
It also fixes up the error codes returned by xfrm_init_state.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: James Morris <jmorris@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds xfrm_init_state which is simply a wrapper that calls
xfrm_get_type and subsequently x->type->init_state. It also gets rid
of the unused args argument.
Abstracting it out allows us to add common initialisation code, e.g.,
to set family-specific flags.
The add_time setting in xfrm_user.c was deleted because it's already
set by xfrm_state_alloc.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: James Morris <jmorris@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implements sctp_connectx() as defined in the SCTP sockets API draft by
tunneling the request through a setsockopt().
Signed-off-by: Frank Filz <ffilzlnx@us.ibm.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implements an interface to be used by leaf qdiscs maintaining an internal
skb queue. The interface maintains a backlog in bytes additionaly
to the skb_queue_len() maintained by the queue itself. Relevant statistics
get incremented automatically. Every function comes in two variants, one
assuming Qdisc->q is used as queue and the second taking a sk_buff_head
as argument. Be aware that, if you use multiple queues, you still have to
maintain the Qdisc->q.qlen counter yourself.
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
To retrieve the neighbour tables send RTM_GETNEIGHTBL with the
NLM_F_DUMP flag set. Every neighbour table configuration is
spread over multiple messages to avoid running into message
size limits on systems with many interfaces. The first message
in the sequence transports all not device specific data such as
statistics, configuration, and the default parameter set.
This message is followed by 0..n messages carrying device
specific parameter sets.
Although the ordering should be sufficient, NDTA_NAME can be
used to identify sequences. The initial message can be identified
by checking for NDTA_CONFIG. The device specific messages do
not contain this TLV but have NDTPA_IFINDEX set to the
corresponding interface index.
To change neighbour table attributes, send RTM_SETNEIGHTBL
with NDTA_NAME set. Changeable attribute include NDTA_THRESH[1-3],
NDTA_GC_INTERVAL, and all TLVs in NDTA_PARMS unless marked
otherwise. Device specific parameter sets can be changed by
setting NDTPA_IFINDEX to the interface index of the corresponding
device.
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
This chunks out the accept_queue and tcp_listen_opt code and moves
them to net/core/request_sock.c and include/net/request_sock.h, to
make it useful for other transport protocols, DCCP being the first one
to use it.
Next patches will rename tcp_listen_opt to accept_sock and remove the
inline tcp functions that just call a reqsk_queue_ function.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ok, this one just renames some stuff to have a better namespace and to
dissassociate it from TCP:
struct open_request -> struct request_sock
tcp_openreq_alloc -> reqsk_alloc
tcp_openreq_free -> reqsk_free
tcp_openreq_fastfree -> __reqsk_free
With this most of the infrastructure closely resembles a struct
sock methods subset.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Kept this first changeset minimal, without changing existing names to
ease peer review.
Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn
has two new members:
->slab, that replaces tcp_openreq_cachep
->obj_size, to inform the size of the openreq descendant for
a specific protocol
The protocol specific fields in struct open_request were moved to a
class hierarchy, with the things that are common to all connection
oriented PF_INET protocols in struct inet_request_sock, the TCP ones
in tcp_request_sock, that is an inet_request_sock, that is an
open_request.
I.e. this uses the same approach used for the struct sock class
hierarchy, with sk_prot indicating if the protocol wants to use the
open_request infrastructure by filling in sk_prot->rsk_prot with an
or_calltable.
Results? Performance is improved and TCP v4 now uses only 64 bytes per
open request minisock, down from 96 without this patch :-)
Next changeset will rename some of the structs, fields and functions
mentioned above, struct or_calltable is way unclear, better name it
struct request_sock_ops, s/struct open_request/struct request_sock/g,
etc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch removes XFRM_SAP_* and converts them over to XFRM_MSG_*.
The netlink interface is meant to map directly onto the underlying
xfrm subsystem. Therefore rather than using a new independent
representation for the events we can simply use the existing ones
from xfrm_user.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This patch ensures that the hard state/policy expire notifications are
only sent when the state/policy is successfully removed from their
respective tables.
As it is, it's possible for a state/policy to both expire through
reaching a hard limit, as well as being deleted by the user.
Note that this behaviour isn't actually forbidden by RFC 2367.
However, it is a quality of implementation issue.
As an added bonus, the restructuring in this patch will help
eventually in moving the expire notifications from softirq
context into process context, thus improving their reliability.
One important side-effect from this change is that SAs reaching
their hard byte/packet limits are now deleted immediately, just
like SAs that have reached their hard time limits.
Previously they were announced immediately but only deleted after
30 seconds.
This is bad because it prevents the system from issuing an ACQUIRE
command until the existing state was deleted by the user or expires
after the time is up.
In the scenario where the expire notification was lost this introduces
a 30 second delay into the system for no good reason.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Heres the final patch.
What this patch provides
- netlink xfrm events
- ability to have events generated by netlink propagated to pfkey
and vice versa.
- fixes the acquire lets-be-happy-with-one-success issue
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
It cannot work properly, so just ignore it in drr
and rr multipath algorithms just like the random
multipath algorithm does.
Suggested by Herbert Xu.
Signed-off by: Pravin B. Shelar <pravins@calsoftinc.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
[XFRM] Call dst_check() with appropriate cookie
This fixes infinite loop issue with IPv6 tunnel mode.
Signed-off-by: Kazunori Miyazawa <kazunori@miyazawa.org>
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ross moved. Remove the bad email address so people will find the correct
one in ./CREDITS.
Signed-off-by: Jesper Juhl <juhl-lkml@dif.dk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This causes sk->sk_prot to change, which makes the socket
release free the sock into the wrong SLAB cache. Fix this
by introducing sk_prot_creator so that we always remember
where the sock came from.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
I found a bug that stopped IPsec/IPv6 from working. About
a month ago IPv6 started using rt6i_idev->dev on the cached socket dst
entries. If the cached socket dst entry is IPsec, then rt6i_idev will
be NULL.
Since we want to look at the rt6i_idev of the original route in this
case, the easiest fix is to store rt6i_idev in the IPsec dst entry just
as we do for a number of other IPv6 route attributes. Unfortunately
this means that we need some new code to handle the references to
rt6i_idev. That's why this patch is bigger than it would otherwise be.
I've also done the same thing for IPv4 since it is conceivable that
once these idev attributes start getting used for accounting, we
probably need to dereference them for IPv4 IPsec entries too.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller