- 11 Jul, 2007 40 commits
-
-
Patrick McHardy authored
Add a nested compat attribute type that can be used to convert attributes that contain a structure to nested attributes in a backwards compatible way. The attribute looks like this: struct { [ compat contents ] struct rtattr { .rta_len = total size, .rta_type = type, } rta; struct old_structure struct; [ nested top-level attribute ] struct rtattr { .rta_len = nest size, .rta_type = type, } nest_attr; [ optional 0 .. n nested attributes ] struct rtattr { .rta_len = private attribute len, .rta_type = private attribute typ, } nested_attr; struct nested_data data; }; Since both userspace and kernel deal correctly with attributes that are larger than expected old versions will just parse the compat part and ignore the rest. Signed-off-by:Patrick McHardy <kaber@trash.net> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Patrick McHardy authored
Add a nested compat attribute type that can be used to convert attributes that contain a structure to nested attributes in a backwards compatible way. Signed-off-by:
-
Patrick McHardy authored
Currently NAT (and others) that want to modify cloned skbs copy them, even if in the vast majority of cases its not necessary because the skb is a clone made by TCP and the portion NAT wants to modify is actually writable because TCP release the header reference before cloning. The problem is that there is no clean way for NAT to find out how long the writable header area is, so this patch introduces skb->hdr_len to hold this length. When a headerless skb is cloned skb->hdr_len is set to the current headroom, for regular clones it is copied from the original. A new function skb_clone_writable(skb, len) returns whether the skb is writable up to len bytes from skb->data. To avoid enlarging the skb the mac_len field is reduced to 16 bit and the new hdr_len field is put in the remaining 16 bit. I've done a few rough benchmarks of NAT (not with this exact patch, but a very similar one). As expected it saves huge amounts of system time in case of sendfile, bringing it down to basically the same amount as without NAT, with sendmsg it only helps on loopback, probably because of the large MTU. Transmit a 1GB file using sendfile/sendmsg over eth0/lo with and without NAT: - sendfile eth0, no NAT: sys 0m0.388s - sendfile eth0, NAT: sys 0m1.835s - sendfile eth0: NAT + path: sys 0m0.370s (~ -80%) - sendfile lo, no NAT: sys 0m0.258s - sendfile lo, NAT: sys 0m2.609s - sendfile lo, NAT + patch: sys 0m0.260s (~ -90%) - sendmsg eth0, no NAT: sys 0m2.508s - sendmsg eth0, NAT: sys 0m2.539s - sendmsg eth0, NAT + patch: sys 0m2.445s (no change) - sendmsg lo, no NAT: sys 0m2.151s - sendmsg lo, NAT: sys 0m3.557s - sendmsg lo, NAT + patch: sys 0m2.159s (~ -40%) I expect other users can see a similar performance improvement, packet mangling iptables targets, ipip and ip_gre come to mind .. Signed-off-by:
-
Krishna Kumar authored
Changes : - netif_queue_stopped need not be called inside qdisc_restart as it has been called already in qdisc_run() before the first skb is sent, and in __qdisc_run() after each intermediate skb is sent (note : we are the only sender, so the queue cannot get stopped while the tx lock was got in the ~LLTX case). - BUG_ON((int) q->q.qlen < 0) was a relic from old times when -1 meant more packets are available, and __qdisc_run used to loop when qdisc_restart() returned -1. During those days, it was necessary to make sure that qlen is never less than zero, since __qdisc_run would get into an infinite loop if no packets are on the queue and this bug in qdisc was there (and worse - no more skbs could ever get queue'd as we hold the queue lock too). With Herbert's recent change to return values, this check is not required. Hopefully Herbert can validate this change. If at all this is required, it should be added to skb_dequeue (in failure case), and not to qdisc_qlen. Signed-off-by:
Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by:
-
Krishna Kumar authored
New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by:
-
Gerrit Renker authored
ccid3_hc_tx_send_packet currently returns 0 when the time difference between current time and t_nom is less than 1000 microseconds. In this case the packet is sent immediately; but, unlike other packets that can be emitted on first attempt, it will not have its window counter updated and its options set as required. This is a bug. Fix: Require the time difference to be at least 1000 microseconds. The algorithm then converges: time differences > 1000 microseconds trigger the timer in dccp_write_xmit; after timer expiry this function is tried again; when the time difference is less than 1000, the packet will have its options added and window counter updated as required. Signed-off-by:
Gerrit Renker <gerrit@erg.abdn.ac.uk> Signed-off-by:
Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
-
Gerrit Renker authored
This updates the computation of t_nom and t_last_win_count to use the newer gettimeofday interface. Committer note: used ktime_to_timeval to set the 'now' variable to t_ld in ccid3hctx_no_feedback_timer Signed-off-by: -
Arnaldo Carvalho de Melo authored
Signed-off-by: -
Gerrit Renker authored
This provides a reusable time difference function which returns the difference in microseconds, as often used in the DCCP code. Commiter note: renamed ktime_delta to ktime_us_delta and put it in ktime.h. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
net/dccp/ccids/lib/loss_interval.c is the only place where this struct is used. Signed-off-by:Arnaldo Carvalho de Melo <acme@redhat.com>
-
Arnaldo Carvalho de Melo authored
It had just a slab cache, so, for the sake of simplicity just make dccp_trfc_lib module init routine create the slab cache, no need for users of the lib to create a private loss_interval object. Signed-off-by: -
Arnaldo Carvalho de Melo authored
Not used outside the loss_interval code anymore. Signed-off-by: -
Arnaldo Carvalho de Melo authored
Now its only used inside the loss_interval code. Signed-off-by: -
Arnaldo Carvalho de Melo authored
Renaming it to dccp_li_update_li. Also based on previous work by Ian McDonald. Signed-off-by: -
Arnaldo Carvalho de Melo authored
Now ccid3_hc_rx_update_li is ready to be moved to net/dccp/ccids/lib/loss_interval, it uses the same interface as the other functions there. Signed-off-by: -
Arnaldo Carvalho de Melo authored
This is a preparatory patch for moving these loss interval functions from net/dccp/ccids/ccid3.c to net/dccp/ccids/lib/loss_interval.c. Based on a patch by Ian McDonald. Signed-off-by: -
Ian McDonald authored
When compiling with EXTRA_CFLAGS=-W noticed that tstamp is not initialised correctly in dccp_li_calc_first_li. Signed-off-by:
Ian McDonald <ian.mcdonald@jandi.co.nz>
-
Ian McDonald authored
When compiling with EXTRA_CFLAGS=-W notice that we have signed/unsigned issue in dccp.h. Signed-off-by:
-
Ian McDonald authored
Signed-off-by:
-
Patrick McHardy authored
Signed-off-by:
-
Patrick McHardy authored
Signed-off-by:
-
Patrick McHardy authored
Keep track of the number of configured ingress/egress QoS mappings to avoid iteration while calculating the netlink attribute size. Signed-off-by:
-
Patrick McHardy authored
skb->priority has only 32 bits and even VLAN uses 32 bit values in its API. Signed-off-by:
-
Patrick McHardy authored
The returned device is unused, return proper error codes instead and avoid having the ioctl handler guess the error. Signed-off-by:
-
Patrick McHardy authored
Move device registration and configuration of the underlying device to a seperate function. Signed-off-by:
-
Patrick McHardy authored
Move the checks of the underlying device to a seperate function. Signed-off-by:
-
Patrick McHardy authored
Move group allocation to a seperate function to clean up the code a bit and allocate groups before registering the device. Device registration is globally visible and causes netlink events, so we shouldn't fail afterwards. Signed-off-by:
-
Patrick McHardy authored
Move some device initialization code to new dev->init callback to make it shareable with netlink. Additionally this fixes a minor bug, dev->iflink is set after registration, which causes an incorrect value in the initial netlink message. Signed-off-by:
-
Patrick McHardy authored
Move the device lookup and checks to the ioctl handler under the RTNL and change all name-based interfaces to take a struct net_device * instead. This allows to use them from a netlink interface, which identifies devices based on ifindex not name. It also avoids races between the ioctl interface and the (upcoming) netlink interface since now all changes happen under the RTNL. As a nice side effect this greatly simplifies error handling in the helper functions and fixes a number of incorrect error codes like -EINVAL for device not found. Signed-off-by:
-
Patrick McHardy authored
Signed-off-by:
-
Patrick McHardy authored
Use a list instead of an array to allow creating new devices. Signed-off-by:
-
Patrick McHardy authored
Signed-off-by:
-
Patrick McHardy authored
Use a list instead of an array to allow creating new devices. Signed-off-by:
-
Patrick McHardy authored
Use dev->stats instead of netdev_priv(). Signed-off-by:
-
Patrick McHardy authored
Add rtnetlink API for creating, changing and deleting software devices. Signed-off-by:
-
Patrick McHardy authored
Split up rtnl_setlink into a function performing validation and a function performing the actual changes. This allows to share the modifcation logic with rtnl_newlink, which is introduced by the next patch. Signed-off-by:
-
Patrick McHardy authored
Signed-off-by:
-
Larry Finger authored
At present, transmission rate information for mac80211 is available only if verbose debugging is turned on, and then only in the logs. This patch implements the SIOCGIWRATE ioctl, which adds the current transmission rate to the output of iwconfig. Signed-off-by:
Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
John W. Linville <linville@tuxdriver.com> Signed-off-by:
-
David S. Miller authored
It hasn't "summed" anything in over 7 years, and it's just a straight mempcy ala skb_copy_to_linear_data() so just get rid of it. Signed-off-by: -
Herbert Xu authored
Currently the code for /proc/net/tcp disable BH while iterating over the entire established hash table. Even though we call cond_resched_softirq for each entry, we still won't process softirq's as regularly as we would otherwise do which results in poor performance when the system is loaded near capacity. This anomaly comes from the 2.4 code where this was all in a single function and the local_bh_disable might have made sense as a small optimisation. The cost of each local_bh_disable is so small when compared against the increased latency in keeping it disabled over a large but mostly empty TCP established hash table that we should just move it to the individual read_lock/read_unlock calls as we do in inet_diag. Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by:
-
