- 21 Jan, 2009 40 commits
-
-
Evgeniy Polyakov authored
With simple extension to the binding mechanism, which allows to bind more than 64k sockets (or smaller amount, depending on sysctl parameters), we have to traverse the whole bind hash table to find out empty bucket. And while it is not a problem for example for 32k connections, bind() completion time grows exponentially (since after each successful binding we have to traverse one bucket more to find empty one) even if we start each time from random offset inside the hash table. So, when hash table is full, and we want to add another socket, we have to traverse the whole table no matter what, so effectivelly this will be the worst case performance and it will be constant. Attached picture shows bind() time depending on number of already bound sockets. Green area corresponds to the usual binding to zero port process, which turns on kernel port selection as described above. Red area is the bind process, when number of reuse-bound sockets is not limited by 64k (or sysctl parameters). The same exponential growth (hidden by the green area) before number of ports reaches sysctl limit. At this time bind hash table has exactly one reuse-enbaled socket in a bucket, but it is possible that they have different addresses. Actually kernel selects the first port to try randomly, so at the beginning bind will take roughly constant time, but with time number of port to check after random start will increase. And that will have exponential growth, but because of above random selection, not every next port selection will necessary take longer time than previous. So we have to consider the area below in the graph (if you could zoom it, you could find, that there are many different times placed there), so area can hide another. Blue area corresponds to the port selection optimization. This is rather simple design approach: hashtable now maintains (unprecise and racely updated) number of currently bound sockets, and when number of such sockets becomes greater than predefined value (I use maximum port range defined by sysctls), we stop traversing the whole bind hash table and just stop at first matching bucket after random start. Above limit roughly corresponds to the case, when bind hash table is full and we turned on mechanism of allowing to bind more reuse-enabled sockets, so it does not change behaviour of other sockets. Signed-off-by:
Evgeniy Polyakov <zbr@ioremap.net> Tested-by:
Denys Fedoryschenko <denys@visp.net.lb> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Herbert Xu authored
This patch makes igb invoke the GRO hooks instead of LRO. As GRO has a compatible external interface to LRO this is a very straightforward replacement. Three things of note: 1) I've kept the LRO Kconfig option until we decide to enable GRO across the board at which point it can also be killed. 2) The poll_controller stuff is broken in igb as it tries to do the same work as the normal poll routine. Since poll_controller can be called in the middle of a poll, this can't be good. I noticed this because poll_controller can invoke the GRO hooks without flushing held GRO packets. However, this should be harmless (assuming the poll_controller bug above doesn't kill you first :) since the next ->poll will clear the backlog. The only time when we'll have a problem is if we're already executing the GRO code on the same ring, but that's no worse than what happens now. 3) I kept the ip_summed check before calling GRO so that we're on par with previous behaviour. Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by:
-
Harvey Harrison authored
The base versions handle constant folding just fine, use them directly. Signed-off-by:
Harvey Harrison <harvey.harrison@gmail.com> Acked-by:
David Dillow <dave@thedillows.org> Signed-off-by:
-
Herbert Xu authored
This patch makes sfc invoke the GRO hooks instead of LRO. As GRO has a compatible external interface to LRO this is a very straightforward replacement. Everything should appear identical to the user except that the offload is now controlled by the GRO ethtool option instead of LRO. I've kept the lro module parameter as is since that's for compatibility only. I have eliminated efx_rx_mk_skb as the GRO layer can take care of all packets regardless of whether GRO is enabled or not. So the only case where we don't call GRO is if the packet checksum is absent. This is to keep the behaviour changes of the patch to a minimum. Signed-off-by:
-
Herbert Xu authored
This patch makes ixgbe invoke the GRO hooks instead of LRO. As GRO has a compatible external interface to LRO this is a very straightforward replacement. As GRO uses the napi structure to track the held packets, I've modified the code paths involved to pass that along. Signed-off-by:
-
Gerrit Renker authored
Since all feature-negotiation processing now takes place in feat.c, functions for producing verbose debugging output are concentrated there. New functions to print out values, entry records, and options are provided, and also a macro is defined to not always have the function name in the output line. Thanks a lot to Wei Yongjun and Giuseppe Galeota for help and discussion with an earlier revision of this patch. Signed-off-by:
Gerrit Renker <gerrit@erg.abdn.ac.uk> Acked-by:
Ian McDonald <ian.mcdonald@jandi.co.nz> Signed-off-by:
-
Gerrit Renker authored
This patch takes care of initialising and type-checking sysctls related to feature negotiation. Type checking is important since some of the sysctls now directly impact the feature-negotiation process. The sysctls are initialised with the known default values for each feature. For the type-checking the value constraints from RFC 4340 are used: * Sequence Window uses the specified Wmin=32, the maximum is ulong (4 bytes), tested and confirmed that it works up to 4294967295 - for Gbps speed; * Ack Ratio is between 0 .. 0xffff (2-byte unsigned integer); * CCIDs are between 0 .. 255; * request_retries, retries1, retries2 also between 0..255 for good measure; * tx_qlen is checked to be non-negative; * sync_ratelimit remains as before. Notes: ------ 1. Die s@sysctl_dccp_feat@sysctl_dccp@g since the sysctls are now in feat.c. 2. As pointed out by Arnaldo, the pattern of type-checking repeats itself in other places, sometimes with exactly the same kind of definitions (e.g. "static int zero;"). It may be a good idea (kernel janitors?) to consolidate type checking. For the sake of keeping the changeset small and in order not to affect other subsystems, I have not strived to generalise here. Signed-off-by: -
Gerrit Renker authored
This adds full support for local/remote Sequence Window feature, from which the * sequence-number-validity (W) and * acknowledgment-number-validity (W') windows derive as specified in RFC 4340, 7.5.3. Specifically, the following is contained in this patch: * integrated new socket fields into dccp_sk; * updated the update_gsr/gss routines with regard to these fields; * updated handler code: the Sequence Window feature is located at the TX side, so the local feature is meant if the handler-rx flag is false; * the initialisation of `rcv_wnd' in reqsk is removed, since - rcv_wnd is not used by the code anywhere; - sequence number checks are not done in the LISTEN state (cf. 7.5.3); - dccp_check_req checks the Ack number validity more rigorously; * the `struct dccp_minisock' became empty and is now removed. Signed-off-by: -
Gerrit Renker authored
This initialises feature negotiation from two tables, which are in turn are initialised from sysctls. As a novel feature, specifics of the implementation (e.g. that short seqnos and ECN are not yet available) are advertised for robustness. Signed-off-by:
-
Divy Le Ray authored
The driver currently drops to line interrupt mode if it did not get all the msi-x vectors it requested. Allow msi-x settings when a minimal amount of vectors is provided. Signed-off-by:
Divy Le Ray <divy@chelsio.com> Signed-off-by:
-
Herbert Xu authored
Now that VLAN has GRO support as well, we can call its GRO handler as well. Signed-off-by:
-
Ben Hutchings authored
Following the removal of the unused struct net_device * parameter from the NAPI functions named *netif_rx_* in commit 908a7a16, they are exactly equivalent to the corresponding *napi_* functions and are therefore redundant. Signed-off-by:
Ben Hutchings <bhutchings@solarflare.com> Acked-by:
Neil Horman <nhorman@tuxdriver.com> Signed-off-by:
-
Cesar Eduardo Barros authored
Signed-off-by:
Cesar Eduardo Barros <cesarb@cesarb.net> Signed-off-by:
-
Cesar Eduardo Barros authored
Instead of making up a name for the device ids, put them directly in the device id table. Also move the vendor id to pci_ids.h. Signed-off-by:
-
Cesar Eduardo Barros authored
SC92031_DESCRIPTION is only used in one place. Signed-off-by:
-
Cesar Eduardo Barros authored
The version string makes no sense anymore, since this driver is only maintained within the kernel. Signed-off-by:
-
Cesar Eduardo Barros authored
The banner currently printed when loading the module is mostly useless. Replace it with a more informative one, printed after probing the device. Output format copied from 8139cp/8139too. Signed-off-by:
-
Jan Engelhardt authored
Signed-off-by:
Jan Engelhardt <jengelh@medozas.de> Signed-off-by:
-
Krzysztof Hałasa authored
Also remove unneeded last_rx update from Synclink drivers. Synclink part mostly by Stephen Hemminger. Signed-off-by:
Krzysztof Hałasa <khc@pm.waw.pl> Signed-off-by:
-
Krzysztof Hałasa authored
Use the internal get_stats() by default. Fixes LMC and wanXL drivers. Signed-off-by:
-
Krzysztof Hałasa authored
Signed-off-by:
-
Inaky Perez-Gonzalez authored
Signed-off-by:
Inaky Perez-Gonzalez <inaky@linux.intel.com> Signed-off-by:
-
Frank Blaschka authored
claw convert to net_device_ops. Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com Signed-off-by: -
Frank Blaschka authored
netiucv convert to net_device_ops. Signed-off-by:
Frank Blaschka <frank.blaschka@de.ibm.com> Signed-off-by:
-
Frank Blaschka authored
ctcm convert to net_device_ops. Signed-off-by:
-
Frank Blaschka authored
lcs convert to net_device_ops. Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
Stephen Hemminger <shemminger@vyatta.com> Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
Thomas Sailer <t.sailer@alumni.ethz.ch> Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
Stephen Hemminger authored
Signed-off-by:
-
