- 03 Jan, 2011 9 commits
-
-
Eric Dumazet authored
Le lundi 03 janvier 2011 à 11:40 -0800, David Miller a écrit : > From: Jarek Poplawski <jarkao2@gmail.com> > Date: Mon, 3 Jan 2011 20:37:03 +0100 > > > On Sun, Jan 02, 2011 at 09:24:36PM +0100, Eric Dumazet wrote: > >> Le mercredi 29 décembre 2010 ?? 00:07 +0100, Jarek Poplawski a écrit : > >> > >> > Ingress is before vlans handler so these features and the > >> > NETIF_F_HW_VLAN_TX flag seem useful for ifb considering > >> > dev_hard_start_xmit() checks. > >> > >> OK, here is v2 of the patch then, thanks everybody. > >> > >> > >> [PATCH v2 net-next-2.6] ifb: add performance flags > >> > >> IFB can use the full set of features flags (NETIF_F_SG | > >> NETIF_F_FRAGLIST | NETIF_F_TSO | NETIF_F_NO_CSUM | NETIF_F_HIGHDMA) to > >> avoid unnecessary split of some packets (GRO for example) > >> > >> Changli suggested to also set vlan_features, > > > > He also suggested more GSO flags of which especially NETIF_F_TSO6 > > seems interesting (wrt GRO)? > > I think at least TSO6 would very much be appropriate here. Yes, why not, I am only wondering why loopback / dummy (and others ?) only set NETIF_F_TSO :) Since I want to play with ECN, I might also add NETIF_F_TSO_ECN ;) For other flags, I really doubt it can matter on ifb ? [PATCH v3 net-next-2.6] ifb: add performance flags IFB can use the full set of features flags (NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_TSO | NETIF_F_NO_CSUM | NETIF_F_HIGHDMA) to avoid unnecessary split of some packets (GRO for example) Changli suggested to also set vlan_features, NETIF_F_TSO6, NETIF_F_TSO_ECN. Jarek suggested to add NETIF_F_HW_VLAN_TX as well. Signed-off-by:
Eric Dumazet <eric.dumazet@gmail.com> Cc: Changli Gao <xiaosuo@gmail.com> Cc: Jarek Poplawski <jarkao2@gmail.com> Cc: Pawel Staszewski <pstaszewski@itcare.pl> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Eric Dumazet authored
Provide child qdisc backlog (byte count) information so that "tc -s qdisc" can report it to user. packet count is already correctly provided. qdisc red 11: parent 1:11 limit 60Kb min 15Kb max 45Kb ecn Sent 3116427684 bytes 1415782 pkt (dropped 8, overlimits 7866 requeues 0) rate 242385Kbit 13630pps backlog 13560b 8p requeues 0 marked 7865 early 1 pdrop 7 other 0 Signed-off-by:
-
Shmulik Ravid authored
More accurate return values for the following (new) dcbnl routines: dcbnl_getdcbx() dcbnl_setdcbx() dcbnl_getfeatcfg() dcbnl_setfeatcfg() Signed-off-by:
Shmulik Ravid <shmulikr@broadcom.com> Signed-off-by:
-
Michal Simek authored
Replace __init by __devinit. Warning message: WARNING: vmlinux.o(.data+0xbc14): Section mismatch in reference from the variable temac_of_driver to the function .init.text:temac_of_probe() The variable temac_of_driver references the function __init temac_of_probe() If the reference is valid then annotate the variable with __init* or __refdata (see linux/init.h) or name the variable: *_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console, Signed-off-by:
Michal Simek <monstr@monstr.eu> Acked-by:
Grant Likely <grant.likely@secretlab.ca> Signed-off-by:
-
Michal Simek authored
Signed-off-by:
-
amit salecha authored
Signed-off-by:
Amit Kumar Salecha <amit.salecha@qlogic.com> Signed-off-by:
-
Sucheta Chakraborty authored
o Enable/disable LRO in device based on NETIF_F_LRO flag, instead of using driver private flag. o Disable LRO, if rx csum offloading is off. David Miller, You should use netdev_info() instead of dev_info(). Signed-off-by:
Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by:
-
Paul Gortmaker authored
With the original 8 bit ISA ne1000 card being over 20 years old, it only makes sense to consider ne.c and all the other toplevel 8390 based driver files as legacy for obsolete hardware. The most recent thing made in large quantities that was 8390 based were those crazy PCI ne2k clones - and even they are now 10+ years old. Also remove myself as maintainer, since the only changes to these drivers going forward will be the generic API type changes that touch all drivers. Signed-off-by:
Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by:
-
Paul Gortmaker authored
The drivers associated with the prototypes in this commit have been deleted some time ago, but the externs escaped detection. Using a simple "git grep" shows that these references are historical artefacts, only mentioned by the deleted lines. Signed-off-by:
-
- 01 Jan, 2011 23 commits
-
-
Paul Gortmaker authored
The tipc/dbg.h file was recently renamed to tipc/log.h, but the re-include define was not updated accordingly. Signed-off-by:
-
stephen hemminger authored
This implements 64 bit statistics support and fixes races when reading counter values. The PHY counters can only be accessed 16 bits at a time, so they are subject to carry races. NB: * TX/RX counters are maintained in software because the the hardware packet count is only a 32 bit value. * Error counters are really only 32 bit. * Old 32 bit counter fields in dev->stats still used for some software counters Signed-off-by:Stephen Hemminger <shemminger@vyatta.com> Signed-off-by:
-
Allan Stephens authored
Cleans up TIPC's source code to eliminate the presence of unnecessary use of {} around single statements. These changes are purely cosmetic and do not alter the operation of TIPC in any way. Signed-off-by:Allan Stephens <Allan.Stephens@windriver.com> Signed-off-by:
-
Allan Stephens authored
Cleans up TIPC's source code to eliminate the needless initialization of static variables to zero. These changes are purely cosmetic and do not alter the operation of TIPC in any way. Signed-off-by:
-
Allan Stephens authored
Cleans up TIPC's source code to eliminate assigning values to variables within conditional expressions, improving code readability and reducing warnings from various code checker tools. These changes are purely cosmetic and do not alter the operation of TIPC in any way. Signed-off-by:
-
Allan Stephens authored
Cleans up TIPC's source code to eliminate deviations from generally accepted coding conventions relating to leading/trailing white space and white space around commas, braces, cases, and sizeof. These changes are purely cosmetic and do not alter the operation of TIPC in any way. Signed-off-by:
-
Paul Gortmaker authored
The existing code for the copy to user and error handling at the end of getsockopt isn't easy to follow, due to the excessive use of if/else. By simply using return where appropriate, it can be made smaller and easier to follow at the same time. Signed-off-by:
-
Allan Stephens authored
It is acceptable to call kfree() with NULL, so these checks are not serving any useful purpose. Signed-off-by:
-
Allan Stephens authored
Eliminates a number of #include statements that no longer serve any useful purpose. Signed-off-by:
-
Allan Stephens authored
Completes the simplification of TIPC's debugging capabilities. By default TIPC includes no debugging code, and any debugging code added by developers that calls the dbg() and dbg_macros() is compiled out. If debugging support is enabled, TIPC prints out some additional data about its internal state when certain abnormal conditions occur, and any developer-added calls to the TIPC debug macros are compiled in. Signed-off-by:
-
Allan Stephens authored
Eliminates most link-specific debugging code in TIPC, which is now largely unnecessary. All calls to the link-specific debugging macros have been removed, as are the macros themselves; in addition, the optional allocation of print buffers to hold debugging information for each link endpoint has been removed. The ability for TIPC to print out helpful diagnostic information when link retransmit failures occur has been retained for the time being, as an aid in tracking down the cause of such failures. Signed-off-by:
-
Allan Stephens authored
Eliminates calls to two debugging macros that are being completely obsoleted, as well as any associated debugging routines that are no longer required. Signed-off-by:
-
Allan Stephens authored
Eliminates obsolete calls to two of TIPC's main debugging macros, as well as a pair of associated debugging routines that are no longer required. Signed-off-by:
-
Allan Stephens authored
As the first step in removing obsolete debugging code from TIPC the files that implement TIPC's non-debug-related log buffer subsystem are renamed to better reflect their true nature. Signed-off-by:
-
Allan Stephens authored
Eliminates a sorted list TIPC uses to keep track of the neighboring nodes it has links to, since this duplicates information already present in the internal array of node object pointers. Signed-off-by:
-
Allan Stephens authored
Eliminates routines, data structures, and files that make up TIPC's user registry. The user registry is no longer needed since the native API routines that utilized it no longer exist and there are no longer any internal TIPC services that use it. Signed-off-by:
-
Allan Stephens authored
Simplifies TIPC's network topology service so that it no longer registers its ports with the user registry, since the service doesn't take advantage of any of the registry's capabilities. Signed-off-by:
-
Allan Stephens authored
Simplifies TIPC's configuration service so that it no longer registers its port with the user registry, since the service doesn't take advantage of any of the registry's capabilities. Signed-off-by:
-
Allan Stephens authored
Eliminates routines, data structures, and files that were intended to allow TIPC to support a network containing multiple clusters. Currently, TIPC supports only networks consisting of a single cluster within a single zone, so this code is unnecessary. Signed-off-by:
-
Allan Stephens authored
Eliminates routines and data structures that were intended to allow TIPC to route messages to other clusters. Currently, TIPC supports only networks consisting of a single cluster within a single zone, so this code is unnecessary. Signed-off-by:
-
Allan Stephens authored
Simplifies routines and data structures that were intended to allow TIPC to support slave nodes (i.e. nodes that did not have links to all of the other nodes in its cluster, forcing TIPC to route messages that it could not deliver directly through a non-slave node). Currently, TIPC supports only networks containing non-slave nodes, so this code is unnecessary. Note: The latest edition of the TIPC 2.0 Specification has eliminated the concept of slave nodes entirely. Signed-off-by:
-
Allan Stephens authored
Eliminates routines, data structures, and files that were intended to allows TIPC to support a network containing multiple zones. Currently, TIPC supports only networks consisting of a single cluster within a single zone, so this code is unnecessary. Signed-off-by:
-
Eric Dumazet authored
In case CONFIG_PM_SLEEP is disabled, we dont need tg3_suspend() and tg3_resume(). drivers/net/tg3.c:15056: warning: ‘tg3_suspend’ defined but not used drivers/net/tg3.c:15110: warning: ‘tg3_resume’ defined but not used Signed-off-by:
-
- 31 Dec, 2010 8 commits
-
-
stephen hemminger authored
The skge driver used the legacy PCI power management, and did its own PCI callbacks. Use the same code model as Rafael's changes to sky2. Let the PCI subsystem take care of all the PCI-specific aspects of device handling during system power transitions. Compile tested only (so far). Signed-off-by:
Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by:
-
Eric Dumazet authored
slot_dequeue_head() should make sure slot skb chain is correct in both ways, or we can crash if all possible flows are in use. Jarek pointed out slot_queue_init() can now be done in sfq_init() once, instead each time a flow is setup. Signed-off-by:
-
Eric Dumazet authored
SFQ is currently 'limited' to small packets, because it uses a 15bit allotment number per flow. Introduce a scale by 8, so that we can handle full size TSO/GRO packets. Use appropriate handling to make sure allot is positive before a new packet is dequeued, so that fairness is respected. Signed-off-by:
Jarek Poplawski <jarkao2@gmail.com> Cc: Patrick McHardy <kaber@trash.net> Signed-off-by:
-
Rafael J. Wysocki authored
The sky2 driver uses the legacy PCI power management, so it has to do some PCI-specific things in its ->suspend() and ->resume() callbacks, which isn't necessary and should better be done by the PCI sybsystem-level power management code. Moreover, it uses device_set_wakeup_enable() incorrectly (that function should be used when the WoL setting is changed rather than during suspend). Convert sky2 to the new PCI power management framework and make it let the PCI subsystem take care of all the PCI-specific aspects of device handling during system power transitions. Tested on a desktop machine with a Marvell 88E8056 PCI-E adapter. Signed-off-by:
-
Rafael J. Wysocki authored
The tg3 driver uses the legacy PCI power management, so it has to do some PCI-specific things in its ->suspend() and ->resume() callbacks, which isn't necessary and should better be done by the PCI sybsystem-level power management code. Convert tg3 to the new PCI power management framework and make it let the PCI subsystem take care of all the PCI-specific aspects of device handling during system power transitions. Tested on HP nx6325 with a NetXtreme BCM5788 adapter. Signed-off-by:
-
Dan Carpenter authored
We already verified that "dev->udev->actconfig->extralen" was non-zero so "len" is non-zero here as well. Signed-off-by:
Dan Carpenter <error27@gmail.com> Acked-by:
Oliver Neukum <oneukum@suse.de> Signed-off-by:
-
Michael Chan authored
Update MIPS firmware to 6.2.1, with improved small packet performance in RSS mode, and iSCSI CID allocation bug fix on 5708. Update driver version to 2.0.21. Signed-off-by:
Michael Chan <mchan@broadcom.com> Signed-off-by:
-
Michael Chan authored
When changing ring size, we free all memory including status block memory. If we're in INTA mode and sharing IRQ, the IRQ handler can be called and it will reference the NULL status block pointer. Because of the lockless design of the IRQ handler, there is no simple way to synchronize and prevent this. So we avoid this problem by freeing the IRQ handler before freeing the status block memory. Signed-off-by:
-
