* Re: tbf/htb qdisc limitations
From: Jarek Poplawski @ 2010-10-17 20:36 UTC (permalink / raw)
To: Bill Fink; +Cc: Eric Dumazet, Rick Jones, Steven Brudenell, netdev
In-Reply-To: <20101016212434.72ae5250.billfink@mindspring.com>
On Sat, Oct 16, 2010 at 09:24:34PM -0400, Bill Fink wrote:
> On Sat, 16 Oct 2010, Jarek Poplawski wrote:
...
> > http://code.google.com/p/pspacer/wiki/HTBon10GbE
> >
> > If it doesn't help reconsider hfsc.
>
> Thanks for the link. From his results, it appears you can
> get better accuracy by keeping TSO/GSO enabled and upping
> the tc mtu parameter to 64000. I will have to try that out.
Sure, but you have to remember that scheduler doesn't know real packet
sizes and rate tables are less accurate especially for smaller packets,
so it depends on conditions.
> For the very high bandwidth cases I tend to deal with, would
> there be any advantage to further reducing the PSCHED_SHIFT
> from its current value of 6?
If you don't use low rates and/or large buffers it might be a good
idea, especially on x64 (for 32-bit longs htb needs some change for
this value below 5).
Jarek P.
^ permalink raw reply
* 2.6.36-rc8-git3: Reported regressions from 2.6.35
From: Rafael J. Wysocki @ 2010-10-17 20:15 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Maciej Rutecki, Florian Mickler, Andrew Morton, Linus Torvalds,
Kernel Testers List, Network Development, Linux ACPI,
Linux PM List, Linux SCSI List, Linux Wireless List, DRI
This message contains a list of some regressions from 2.6.35,
for which there are no fixes in the mainline known to the tracking team.
If any of them have been fixed already, please let us know.
If you know of any other unresolved regressions from 2.6.35, please let us
know either and we'll add them to the list. Also, please let us know
if any of the entries below are invalid.
Each entry from the list will be sent additionally in an automatic reply
to this message with CCs to the people involved in reporting and handling
the issue.
Listed regressions statistics:
Date Total Pending Unresolved
----------------------------------------
2010-10-17 70 27 27
2010-10-10 56 16 15
2010-10-03 52 16 14
2010-09-26 46 15 13
2010-09-20 38 15 15
2010-09-12 28 14 13
2010-08-30 21 16 15
Unresolved regressions
----------------------
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20472
Subject : [2.6.35.5 -> .7] INFO: task rpcbind:14163 blocked for more than 120 seconds.
Submitter : Pawel Sikora <pluto-PIIpFW8S9c0@public.gmane.org>
Date : 2010-10-15 15:02 (3 days old)
Message-ID : <201010151702.15675.pluto-PIIpFW8S9c0@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128715683507549&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20462
Subject : 2.6.36-rc7-git2 - panic/GPF: e1000e/vlans?
Submitter : Nikola Ciprich <extmaillist-Jp3n8lUXroTtwjQa/ONI9g@public.gmane.org>
Date : 2010-10-15 7:10 (3 days old)
Message-ID : <20101015071008.GA8714-xTMdSLfc3Wpi51D5yjT6kKVXKuFTiq87@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128712984831303&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20352
Subject : Fwd: Re: UML kernel crash of v2.6.36-rcX kernel
Submitter : Toralf Förster <toralf.foerster-Mmb7MZpHnFY@public.gmane.org>
Date : 2010-10-13 11:28 (5 days old)
First-Bad-Commit: http://git.kernel.org/linus/365b18189789bfa1acd9939e6312b8a4b4577b28
Message-ID : <201010131328.06465.toralf.foerster-Mmb7MZpHnFY@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128696930132503&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20342
Subject : [LogFS] [2.6.36-rc7] Deadlock in logfs_get_wblocks, hold and wait on same lock super->s_write_mutex
Submitter : Prasad Joshi <prasadjoshi124-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date : 2010-10-13 9:49 (5 days old)
Message-ID : <AANLkTinvsMxTxEbDEFmb5M-6fYjdRvErU==Zs7+qANkV-Rq2MYuBUFGs@public.gmane.orgail.com>
References : http://marc.info/?l=linux-kernel&m=128696335024718&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20332
Subject : [LogFS] [2.6.36-rc7] Kernel BUG at lib/btree.c:465!
Submitter : Prasad Joshi <prasadjoshi124-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date : 2010-10-12 18:56 (6 days old)
Message-ID : <AANLkTimAbCZNhLQ5nADUiAC+7JpAeJBEmjFwdxyZ-FxO-JsoAwUIsXotQFR93xxRIaA@public.gmane.orgcom>
References : http://marc.info/?l=linux-kernel&m=128690910501830&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20322
Subject : 2.6.36-rc7: inconsistent lock state: inconsistent {IN-RECLAIM_FS-R} -> {RECLAIM_FS-ON-W} usage.
Submitter : Dave Jones <davej-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Date : 2010-10-11 20:10 (7 days old)
Message-ID : <20101011201007.GA29707-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128682782828453&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20232
Subject : kworker consumes ~100% CPU on HP Elitebook 8540w running 2.6.36_rc6-git4
Submitter : Ozan Caglayan <ozan-caicS1wCkhO6A22drWdTBw@public.gmane.org>
Date : 2010-10-13 06:13 (5 days old)
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20182
Subject : 2.6.36-rc7: NULL pointer dereference in ehci_clear_tt_buffer_complete
Submitter : Stefan Richter <stefanr-MtYdepGKPcBMYopoZt5u/LNAH6kLmebB@public.gmane.org>
Date : 2010-10-11 20:56 (7 days old)
Message-ID : <tkrat.9f585ea47331a2a2-MtYdepGKPcBMYopoZt5u/LNAH6kLmebB@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128683064431749&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20172
Subject : Lenovo S12 2.6.36-rc7 lockup
Submitter : Chris Vine <chris-TF6qbakwsgc2epGFuHBODCp2UmYkHbXO@public.gmane.org>
Date : 2010-10-10 12:59 (8 days old)
Message-ID : <20101010135912.76eb61b7-PlFwYbz7hoIyinQH9hAyzg@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128671620721712&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20162
Subject : [LogFS][2.6.36.rc7+] Kernel BUG at readwrite.c:1193
Submitter : Prasad Joshi <prasadjoshi124-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date : 2010-10-10 17:44 (8 days old)
Message-ID : <AANLkTi=JkcuWBPo+X-i+9o-BJFVqjea1J3e=Mr=HvAWF@mail.gmail.com>
References : http://marc.info/?l=linux-kernel&m=128673196203340&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20032
Subject : 2.6.36-rc7 continuos kernel panics due to of cpu_idle (and cpu_intel_idle)
Submitter : Patrizio Bassi <patrizio.bassi-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date : 2010-10-08 18:23 (10 days old)
Message-ID : <AANLkTi=2irM5tM2rNC5wHVV=Nw4cs_CK59pvXDD+Y5Dz-Rq2MYuBUFGs@public.gmane.orgail.com>
References : http://marc.info/?l=linux-kernel&m=128656222919323&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=20022
Subject : "do_IRQ: 0.89 No irq handler for vector (irq -1)"
Submitter : Dave Airlie <airlied-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date : 2010-10-07 8:55 (11 days old)
Message-ID : <AANLkTikZj21CtFwC2ttvc0wJPx2qyedn7b1xXaxH659E-JsoAwUIsXotQFR93xxRIaA@public.gmane.orgcom>
References : http://marc.info/?l=linux-kernel&m=128644173420501&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=19862
Subject : [REGRESSION] no sound on T60 laptop (HDA Intel)
Submitter : Miklos Szeredi <miklos-sUDqSbJrdHQHWmgEVkV9KA@public.gmane.org>
Date : 2010-10-05 9:25 (13 days old)
Message-ID : <E1P33mT-0003qa-Fv-8f8m9JG5TPIdUIPVzhDTVZP2KDSNp7ea@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128627079921512&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=19802
Subject : [drm:init_ring_common] *ERROR* render ring head not reset to zero
Submitter : Thomas Meyer <thomas-VsYtu1Qij5c@public.gmane.org>
Date : 2010-10-04 20:31 (14 days old)
Message-ID : <201010042231.20320.thomas-VsYtu1Qij5c@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128622441331199&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=19782
Subject : 2.6.36-rc6-git2 -- BUG dentry: Poison overwritten (after resume from hibernation)
Submitter : Miles Lane <miles.lane-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date : 2010-10-03 3:54 (15 days old)
Message-ID : <AANLkTimrqtzBHAZSuDhEs_3CKA6pCbM35b1BtKH=MvC_-JsoAwUIsXov1KXRcyAk9cg@public.gmane.orgl.com>
References : http://marc.info/?l=linux-kernel&m=128607809314079&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=19632
Subject : 2.6.36-rc6: modprobe Not tainted warning
Submitter : Heinz Diehl <htd-iEI8Y0CNJBYdnm+yROfE0A@public.gmane.org>
Date : 2010-09-30 18:25 (18 days old)
Message-ID : <20100930182516.GA15089-iEI8Y0CNJBYdnm+yROfE0A@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128587114004680&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=19392
Subject : WARNING: at drivers/net/wireless/ath/ath5k/base.c:3475 ath5k_bss_info_changed+0x44/0x168 [ath5k]()
Submitter : Justin Mattock <justinmattock-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date : 2010-09-28 22:30 (20 days old)
Message-ID : <AANLkTim5WCGKPvEkOkO_YnMF9pg8mvLfQoFBNUFpfa_k-JsoAwUIsXotQFR93xxRIaA@public.gmane.orgcom>
References : http://marc.info/?l=linux-kernel&m=128571307018635&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=19372
Subject : 2.6.36-rc6: WARNING: at drivers/gpu/drm/radeon/radeon_fence.c:235 radeon_fence_wait+0x35a/0x3c0
Submitter : Alexey Dobriyan <adobriyan-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date : 2010-09-29 21:29 (19 days old)
Message-ID : <20100929212923.GA5578-y0M6fkzdUYllgR+Ck+lCww@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128579579400315&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=19142
Subject : Screen flickers when switching from the console to X
Submitter : Andrey Rahmatullin <wrar-u2l5PoMzF/Vg9hUCZPvPmw@public.gmane.org>
Date : 2010-09-27 12:05 (21 days old)
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=19072
Subject : [2.6.36-rc regression] occasional complete system hangs on sparc64 SMP
Submitter : Mikael Pettersson <mikpe-1zs4UD6AkMk@public.gmane.org>
Date : 2010-09-23 17:02 (25 days old)
Message-ID : <<19611.34846.813757.309183-tgku4HJDRZih8lFjZTKsyTAV6s6igYVG@public.gmane.org>>
References : http://marc.info/?l=linux-kernel&m=128526136531048&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=19052
Subject : 2.6.36-rc5-git1 -- [drm:i915_report_and_clear_eir] *ERROR* EIR stuck: 0x00000010, masking
Submitter : Miles Lane <miles.lane-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date : 2010-09-22 23:47 (26 days old)
Message-ID : <AANLkTikWQjUQjFJU9MO1+XbSLAEE-GARz+S+Dz2Fgu4h-JsoAwUIsXotQFR93xxRIaA@public.gmane.orgcom>
References : http://marc.info/?l=linux-kernel&m=128519926626322&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=19002
Subject : Radeon rv730 AGP/KMS/DRM kernel lockup
Submitter : Duncan <1i5t5.duncan-j9pdmedNgrk@public.gmane.org>
Date : 2010-09-23 16:48 (25 days old)
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=17361
Subject : Random kmemcheck errors and kernel freeze on 2.6.36-rc*
Submitter : Christian Casteyde <casteyde.christian-GANU6spQydw@public.gmane.org>
Date : 2010-08-29 19:59 (50 days old)
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=17121
Subject : Two blank rectangles more than 10 cm long when booting
Submitter : Eric Valette <eric.valette-GANU6spQydw@public.gmane.org>
Date : 2010-08-26 17:24 (53 days old)
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=17061
Subject : 2.6.36-rc1 on zaurus: bluetooth regression
Submitter : Pavel Machek <pavel-+ZI9xUNit7I@public.gmane.org>
Date : 2010-08-21 15:24 (58 days old)
Message-ID : <20100821152445.GA1536-+ZI9xUNit7I@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128240433828087&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=16971
Subject : qla4xxx compile failure on 32-bit PowerPC: missing readq and writeq
Submitter : Meelis Roos <mroos-Y27EyoLml9s@public.gmane.org>
Date : 2010-08-19 21:03 (60 days old)
Message-ID : <<alpine.SOC.1.00.1008192359310.19654-ptEonEWSGqKptlylMvRsHA@public.gmane.org>>
References : http://marc.info/?l=linux-kernel&m=128225184900892&w=2
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=16951
Subject : hackbench regression with 2.6.36-rc1
Submitter : Zhang, Yanmin <yanmin_zhang-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
Date : 2010-08-18 6:18 (61 days old)
Message-ID : <1282112318.21202.8.camel-sz7BYL/Y5Hu/P+R7jlPCFVaTQe2KTcn/@public.gmane.org>
References : http://marc.info/?l=linux-kernel&m=128211235904910&w=2
For details, please visit the bug entries and follow the links given in
references.
As you can see, there is a Bugzilla entry for each of the listed regressions.
There also is a Bugzilla entry used for tracking the regressions from 2.6.35,
unresolved as well as resolved, at:
http://bugzilla.kernel.org/show_bug.cgi?id=16444
Please let the tracking team know if there are any Bugzilla entries that
should be added to the list in there.
Thanks!
^ permalink raw reply
* EAPOL bridging
From: Benjamin Poirier @ 2010-10-17 18:06 UTC (permalink / raw)
To: bridge; +Cc: netdev, linux-kernel
Hello,
I have some trouble bridging EAPOL frames. I'd like to do this to allow
wired 802.1x authentication from within a kvm virtual machine. I have
the following setup:
kvm -- tap0 -- br0 -- eth1 -- 802.1x authenticator (switch) -- more network
and it doesn't work. I've added a few logging rules to ebtables. I only
see an EAPOL frame going through the INPUT chain of tap0. It seems to be
dropped by the bridge. The EAPOL frame is an ethernet link local
multicast frame with destination address 01-80-C2-00-00-03, "IEEE Std
802.1X PAE address".
I've looked at http://standards.ieee.org/regauth/groupmac/tutorial.html,
which says that frames with a destination in the range 01-80-C2-00-00-00
to 01-80-C2-00-00-0F should not be forwarded by standard conformant
bridges. I've also looked at net/bridge/br_input.c and br_handle_frame()
seems quite intent on "bending" the standard when STP is disabled, but
only for 01-80-C2-00-00-00. However there are more applications that use
similar addresses, EAPOL included:
http://standards.ieee.org/regauth/groupmac/Standard_Group_MAC_Address_assignments.pdf
Given the current state of affairs, would it be acceptable to make the
code more permissive by forwarding all the range of reserved group
addresses when STP is disabled? If not, what would be the way to go
about enabling 802.1x authentication from within a virtual machine?
BTW, it seems this issue has been raised before,
https://lists.linux-foundation.org/pipermail/bridge/2007-November/005629.html
with the conclusion that
> Despite what the standards say, many users are using bridging code for invisible
> firewalls etc, and in those cases they want STP and EAPOL frames to be forwarded.
Thanks,
-Ben
^ permalink raw reply
* [PATCH v2] CAPI: Silence lockdep warning on get_capi_appl_by_nr usage
From: Jan Kiszka @ 2010-10-17 15:18 UTC (permalink / raw)
To: David S. Miller
Cc: Linux Kernel Mailing List, i4ldeveloper, Linux Netdev List,
Karsten Keil, Peter Zijlstra
In-Reply-To: <1287311135.1998.154.camel@laptop>
As long as we hold capi_controller_lock, we can safely access
capi_applications without RCU protection as no one can modify the
application list underneath us. Introduce an RCU-free
__get_capi_appl_by_nr for this purpose. This silences lockdep warnings
on suspicious rcu_dereference usage.
Signed-off-by: Jan Kiszka <jan.kiszka@web.de>
---
v2:
- use lockdep_assert_held in favor of WARN_ON_ONCE(!mutex_is_locked)
as suggested by Peter
drivers/isdn/capi/kcapi.c | 19 ++++++++++++++-----
1 files changed, 14 insertions(+), 5 deletions(-)
diff --git a/drivers/isdn/capi/kcapi.c b/drivers/isdn/capi/kcapi.c
index 2648d39..2b33b26 100644
--- a/drivers/isdn/capi/kcapi.c
+++ b/drivers/isdn/capi/kcapi.c
@@ -99,6 +99,16 @@ static inline struct capi_ctr *get_capi_ctr_by_nr(u16 contr)
return capi_controller[contr - 1];
}
+static inline struct capi20_appl *__get_capi_appl_by_nr(u16 applid)
+{
+ lockdep_assert_held(&capi_controller_lock);
+
+ if (applid - 1 >= CAPI_MAXAPPL)
+ return NULL;
+
+ return capi_applications[applid - 1];
+}
+
static inline struct capi20_appl *get_capi_appl_by_nr(u16 applid)
{
if (applid - 1 >= CAPI_MAXAPPL)
@@ -186,10 +196,9 @@ static void notify_up(u32 contr)
ctr->state = CAPI_CTR_RUNNING;
for (applid = 1; applid <= CAPI_MAXAPPL; applid++) {
- ap = get_capi_appl_by_nr(applid);
- if (!ap)
- continue;
- register_appl(ctr, applid, &ap->rparam);
+ ap = __get_capi_appl_by_nr(applid);
+ if (ap)
+ register_appl(ctr, applid, &ap->rparam);
}
wake_up_interruptible_all(&ctr->state_wait_queue);
@@ -216,7 +225,7 @@ static void ctr_down(struct capi_ctr *ctr, int new_state)
memset(ctr->serial, 0, sizeof(ctr->serial));
for (applid = 1; applid <= CAPI_MAXAPPL; applid++) {
- ap = get_capi_appl_by_nr(applid);
+ ap = __get_capi_appl_by_nr(applid);
if (ap)
capi_ctr_put(ctr);
}
--
1.7.1
^ permalink raw reply related
* Re: [PATCH V2 0/7] can: mcp251x: fix and optimize driver
From: Wolfgang Grandegger @ 2010-10-17 14:29 UTC (permalink / raw)
To: Marc Kleine-Budde
Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w,
netdev-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1287139762-23356-1-git-send-email-mkl-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
On 10/15/2010 12:49 PM, Marc Kleine-Budde wrote:
> Moin,
>
> this series of patches improves the mcp251x driver. It first fixes the
> local_softirq_pending problem. Then the amount of SPI transfers is reduced
> in order to optimise the driver.
>
> This series has been tested with a mcp2515 on i.MX35.
>
> Changes since V1:
> - Fix broken encoding in S-o-b
>
> Please review, test and consider to apply.
You can add to all patches my
Acked-by: Wolfgang Grandegger <wg-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
Wolfgang.
^ permalink raw reply
* [PATCH v2] b44: fix resume, request_irq after hw reset
From: James Hogan @ 2010-10-17 11:48 UTC (permalink / raw)
To: Gary Zambrano
Cc: David S. Miller, Jiri Pirko, FUJITA Tomonori, Hauke Mehrtens,
Larry Finger, netdev, linux-kernel, Andrew Morton
On resume, call request_irq() after resetting the hardware rather than
before. It's a shared interrupt so the handler could be called
immediately if another device on the same irq interrupts (and will be
called immediately if CONFIG_DEBUG_SHIRQ=y), but unless the hardware is
reinitialised with b44_init_hw() the read of the interrupt status
register will hang the system.
Signed-off-by: James Hogan <james@albanarts.com>
---
drivers/net/b44.c | 19 ++++++++++++++-----
1 files changed, 14 insertions(+), 5 deletions(-)
diff --git a/drivers/net/b44.c b/drivers/net/b44.c
index efeffdf..f950505 100644
--- a/drivers/net/b44.c
+++ b/drivers/net/b44.c
@@ -2296,18 +2296,27 @@ static int b44_resume(struct ssb_device *sdev)
if (!netif_running(dev))
return 0;
+ spin_lock_irq(&bp->lock);
+ b44_init_rings(bp);
+ b44_init_hw(bp, B44_FULL_RESET);
+ spin_unlock_irq(&bp->lock);
+
+ /*
+ * As a shared interrupt, the handler can be called immediately. To be
+ * able to check the interrupt status the hardware must already be
+ * powered back on (b44_init_hw).
+ */
rc = request_irq(dev->irq, b44_interrupt, IRQF_SHARED, dev->name, dev);
if (rc) {
netdev_err(dev, "request_irq failed\n");
+ spin_lock_irq(&bp->lock);
+ b44_halt(bp);
+ b44_free_rings(bp);
+ spin_unlock_irq(&bp->lock);
return rc;
}
- spin_lock_irq(&bp->lock);
-
- b44_init_rings(bp);
- b44_init_hw(bp, B44_FULL_RESET);
netif_device_attach(bp->dev);
- spin_unlock_irq(&bp->lock);
b44_enable_ints(bp);
netif_wake_queue(dev);
--
1.7.2.3
^ permalink raw reply related
* Re: [PATCH] CAPI: Silence lockdep warning on get_capi_appl_by_nr usage
From: Peter Zijlstra @ 2010-10-17 10:25 UTC (permalink / raw)
To: Jan Kiszka
Cc: David S. Miller, Linux Kernel Mailing List, i4ldeveloper,
Linux Netdev List, Karsten Keil
In-Reply-To: <4CB9891B.4060606@web.de>
On Sat, 2010-10-16 at 13:14 +0200, Jan Kiszka wrote:
> + WARN_ON_ONCE(!mutex_is_locked(&capi_controller_lock));
Note there is also lockdep_assert_held() which will actually validate
that the current task is the owner.
^ permalink raw reply
* [PATCH] net_sched: remove the unused parameter of qdisc_create_dflt()
From: Changli Gao @ 2010-10-16 23:04 UTC (permalink / raw)
To: David S. Miller
Cc: Jamal Hadi Salim, Stephen Hemminger, netdev, netem, Changli Gao
The first parameter dev isn't in use in qdisc_create_dflt().
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
---
include/net/sch_generic.h | 3 +--
net/sched/sch_atm.c | 5 ++---
net/sched/sch_cbq.c | 12 ++++++------
net/sched/sch_drr.c | 4 ++--
net/sched/sch_dsmark.c | 6 ++----
net/sched/sch_fifo.c | 3 +--
net/sched/sch_generic.c | 10 ++++------
net/sched/sch_hfsc.c | 8 +++-----
net/sched/sch_htb.c | 8 +++-----
net/sched/sch_mq.c | 2 +-
net/sched/sch_multiq.c | 3 +--
net/sched/sch_netem.c | 3 +--
net/sched/sch_prio.c | 2 +-
13 files changed, 28 insertions(+), 41 deletions(-)
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index eda8808..ea1f8a8 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -328,8 +328,7 @@ extern void qdisc_destroy(struct Qdisc *qdisc);
extern void qdisc_tree_decrease_qlen(struct Qdisc *qdisc, unsigned int n);
extern struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
struct Qdisc_ops *ops);
-extern struct Qdisc *qdisc_create_dflt(struct net_device *dev,
- struct netdev_queue *dev_queue,
+extern struct Qdisc *qdisc_create_dflt(struct netdev_queue *dev_queue,
struct Qdisc_ops *ops, u32 parentid);
extern void qdisc_calculate_pkt_len(struct sk_buff *skb,
struct qdisc_size_table *stab);
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 6318e11..2825407 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -275,8 +275,7 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent,
goto err_out;
}
flow->filter_list = NULL;
- flow->q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- &pfifo_qdisc_ops, classid);
+ flow->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid);
if (!flow->q)
flow->q = &noop_qdisc;
pr_debug("atm_tc_change: qdisc %p\n", flow->q);
@@ -543,7 +542,7 @@ static int atm_tc_init(struct Qdisc *sch, struct nlattr *opt)
INIT_LIST_HEAD(&p->flows);
INIT_LIST_HEAD(&p->link.list);
list_add(&p->link.list, &p->flows);
- p->link.q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
+ p->link.q = qdisc_create_dflt(sch->dev_queue,
&pfifo_qdisc_ops, sch->handle);
if (!p->link.q)
p->link.q = &noop_qdisc;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 28c01ef..eb76315 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1379,9 +1379,9 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt)
q->link.sibling = &q->link;
q->link.common.classid = sch->handle;
q->link.qdisc = sch;
- if (!(q->link.q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- &pfifo_qdisc_ops,
- sch->handle)))
+ q->link.q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops,
+ sch->handle);
+ if (!q->link.q)
q->link.q = &noop_qdisc;
q->link.priority = TC_CBQ_MAXPRIO-1;
@@ -1623,7 +1623,7 @@ static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
struct cbq_class *cl = (struct cbq_class*)arg;
if (new == NULL) {
- new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
+ new = qdisc_create_dflt(sch->dev_queue,
&pfifo_qdisc_ops, cl->common.classid);
if (new == NULL)
return -ENOBUFS;
@@ -1874,8 +1874,8 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
cl->R_tab = rtab;
rtab = NULL;
cl->refcnt = 1;
- if (!(cl->q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- &pfifo_qdisc_ops, classid)))
+ cl->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid);
+ if (!cl->q)
cl->q = &noop_qdisc;
cl->common.classid = classid;
cl->tparent = parent;
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index b74046a..aa8b531 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -110,7 +110,7 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
cl->refcnt = 1;
cl->common.classid = classid;
cl->quantum = quantum;
- cl->qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
+ cl->qdisc = qdisc_create_dflt(sch->dev_queue,
&pfifo_qdisc_ops, classid);
if (cl->qdisc == NULL)
cl->qdisc = &noop_qdisc;
@@ -218,7 +218,7 @@ static int drr_graft_class(struct Qdisc *sch, unsigned long arg,
struct drr_class *cl = (struct drr_class *)arg;
if (new == NULL) {
- new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
+ new = qdisc_create_dflt(sch->dev_queue,
&pfifo_qdisc_ops, cl->common.classid);
if (new == NULL)
new = &noop_qdisc;
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index 63d41f8..1d295d6 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -61,8 +61,7 @@ static int dsmark_graft(struct Qdisc *sch, unsigned long arg,
sch, p, new, old);
if (new == NULL) {
- new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- &pfifo_qdisc_ops,
+ new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops,
sch->handle);
if (new == NULL)
new = &noop_qdisc;
@@ -384,8 +383,7 @@ static int dsmark_init(struct Qdisc *sch, struct nlattr *opt)
p->default_index = default_index;
p->set_tc_index = nla_get_flag(tb[TCA_DSMARK_SET_TC_INDEX]);
- p->q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- &pfifo_qdisc_ops, sch->handle);
+ p->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, sch->handle);
if (p->q == NULL)
p->q = &noop_qdisc;
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index 5948baf..4dfecb0 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -172,8 +172,7 @@ struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops,
struct Qdisc *q;
int err = -ENOMEM;
- q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- ops, TC_H_MAKE(sch->handle, 1));
+ q = qdisc_create_dflt(sch->dev_queue, ops, TC_H_MAKE(sch->handle, 1));
if (q) {
err = fifo_set_limit(q, limit);
if (err < 0) {
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 3d57681..2a95272 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -574,10 +574,8 @@ errout:
return ERR_PTR(err);
}
-struct Qdisc * qdisc_create_dflt(struct net_device *dev,
- struct netdev_queue *dev_queue,
- struct Qdisc_ops *ops,
- unsigned int parentid)
+struct Qdisc *qdisc_create_dflt(struct netdev_queue *dev_queue,
+ struct Qdisc_ops *ops, unsigned int parentid)
{
struct Qdisc *sch;
@@ -682,7 +680,7 @@ static void attach_one_default_qdisc(struct net_device *dev,
struct Qdisc *qdisc;
if (dev->tx_queue_len) {
- qdisc = qdisc_create_dflt(dev, dev_queue,
+ qdisc = qdisc_create_dflt(dev_queue,
&pfifo_fast_ops, TC_H_ROOT);
if (!qdisc) {
printk(KERN_INFO "%s: activation failed\n", dev->name);
@@ -709,7 +707,7 @@ static void attach_default_qdiscs(struct net_device *dev)
dev->qdisc = txq->qdisc_sleeping;
atomic_inc(&dev->qdisc->refcnt);
} else {
- qdisc = qdisc_create_dflt(dev, txq, &mq_qdisc_ops, TC_H_ROOT);
+ qdisc = qdisc_create_dflt(txq, &mq_qdisc_ops, TC_H_ROOT);
if (qdisc) {
qdisc->ops->attach(qdisc);
dev->qdisc = qdisc;
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 4749609..069c62b 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1088,7 +1088,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
cl->refcnt = 1;
cl->sched = q;
cl->cl_parent = parent;
- cl->qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
+ cl->qdisc = qdisc_create_dflt(sch->dev_queue,
&pfifo_qdisc_ops, classid);
if (cl->qdisc == NULL)
cl->qdisc = &noop_qdisc;
@@ -1209,8 +1209,7 @@ hfsc_graft_class(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
if (cl->level > 0)
return -EINVAL;
if (new == NULL) {
- new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- &pfifo_qdisc_ops,
+ new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops,
cl->cl_common.classid);
if (new == NULL)
new = &noop_qdisc;
@@ -1452,8 +1451,7 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt)
q->root.cl_common.classid = sch->handle;
q->root.refcnt = 1;
q->root.sched = q;
- q->root.qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- &pfifo_qdisc_ops,
+ q->root.qdisc = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops,
sch->handle);
if (q->root.qdisc == NULL)
q->root.qdisc = &noop_qdisc;
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 87d9449..01b519d 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1121,8 +1121,7 @@ static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
if (cl->level)
return -EINVAL;
if (new == NULL &&
- (new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- &pfifo_qdisc_ops,
+ (new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops,
cl->common.classid)) == NULL)
return -ENOBUFS;
@@ -1247,8 +1246,7 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg)
return -EBUSY;
if (!cl->level && htb_parent_last_child(cl)) {
- new_q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- &pfifo_qdisc_ops,
+ new_q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops,
cl->parent->common.classid);
last_child = 1;
}
@@ -1377,7 +1375,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
/* create leaf qdisc early because it uses kmalloc(GFP_KERNEL)
so that can't be used inside of sch_tree_lock
-- thanks to Karlis Peisenieks */
- new_q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
+ new_q = qdisc_create_dflt(sch->dev_queue,
&pfifo_qdisc_ops, classid);
sch_tree_lock(sch);
if (parent && !parent->level) {
diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c
index fe91e50..ecc302f 100644
--- a/net/sched/sch_mq.c
+++ b/net/sched/sch_mq.c
@@ -56,7 +56,7 @@ static int mq_init(struct Qdisc *sch, struct nlattr *opt)
for (ntx = 0; ntx < dev->num_tx_queues; ntx++) {
dev_queue = netdev_get_tx_queue(dev, ntx);
- qdisc = qdisc_create_dflt(dev, dev_queue, &pfifo_fast_ops,
+ qdisc = qdisc_create_dflt(dev_queue, &pfifo_fast_ops,
TC_H_MAKE(TC_H_MAJ(sch->handle),
TC_H_MIN(ntx + 1)));
if (qdisc == NULL)
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index 6ae2512..32690de 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -227,8 +227,7 @@ static int multiq_tune(struct Qdisc *sch, struct nlattr *opt)
for (i = 0; i < q->bands; i++) {
if (q->queues[i] == &noop_qdisc) {
struct Qdisc *child, *old;
- child = qdisc_create_dflt(qdisc_dev(sch),
- sch->dev_queue,
+ child = qdisc_create_dflt(sch->dev_queue,
&pfifo_qdisc_ops,
TC_H_MAKE(sch->handle,
i + 1));
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 4714ff1..e5593c0 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -538,8 +538,7 @@ static int netem_init(struct Qdisc *sch, struct nlattr *opt)
qdisc_watchdog_init(&q->watchdog, sch);
- q->qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
- &tfifo_qdisc_ops,
+ q->qdisc = qdisc_create_dflt(sch->dev_queue, &tfifo_qdisc_ops,
TC_H_MAKE(sch->handle, 1));
if (!q->qdisc) {
pr_debug("netem: qdisc create failed\n");
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 0748fb1..b1c95bc 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -200,7 +200,7 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
for (i=0; i<q->bands; i++) {
if (q->queues[i] == &noop_qdisc) {
struct Qdisc *child, *old;
- child = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
+ child = qdisc_create_dflt(sch->dev_queue,
&pfifo_qdisc_ops,
TC_H_MAKE(sch->handle, i + 1));
if (child) {
^ permalink raw reply related
* [RFC PATCH net-next] drivers/net/bonding/bond_3ad: Remove struct mac_addr and typedef
From: Joe Perches @ 2010-10-17 7:38 UTC (permalink / raw)
To: Jay Vosburgh; +Cc: bonding-devel, netdev, LKML
Use more standard u8 name[ETH_ALEN];
Remove MAC_ADDRESS_COMPARE and use compare_ether_addr
Remove null_mac_addr and use is_zero_ether_addr and memset
Use memcpy for copying ether addresses
Size increases ~200 bytes so it's not great,
but it might be better for style consistency.
$ size drivers/net/bonding/*.o.*
text data bss dec hex filename
25580 776 5800 32156 7d9c drivers/net/bonding/bond_3ad.o.new
25396 776 5808 31980 7cec drivers/net/bonding/bond_3ad.o.old
Signed-off-by: Joe Perches <joe@perches.com>
---
drivers/net/bonding/bond_3ad.c | 50 ++++++++++++++++++---------------------
drivers/net/bonding/bond_3ad.h | 20 ++++++---------
2 files changed, 31 insertions(+), 39 deletions(-)
diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index 079b9d1..5a38eef 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -90,10 +90,6 @@
#define AD_LINK_SPEED_BITMASK_10000MBPS 0x10
//endalloun
-// compare MAC addresses
-#define MAC_ADDRESS_COMPARE(A, B) memcmp(A, B, ETH_ALEN)
-
-static struct mac_addr null_mac_addr = {{0, 0, 0, 0, 0, 0}};
static u16 ad_ticks_per_sec;
static const int ad_delta_in_ticks = (AD_TIMER_INTERVAL * HZ) / 1000;
@@ -218,7 +214,7 @@ static inline struct aggregator *__get_next_agg(struct aggregator *aggregator)
*/
static inline int __agg_has_partner(struct aggregator *agg)
{
- return !is_zero_ether_addr(agg->partner_system.mac_addr_value);
+ return !is_zero_ether_addr(agg->partner_system);
}
/**
@@ -478,7 +474,7 @@ static void __choose_matched(struct lacpdu *lacpdu, struct port *port)
// check if all parameters are alike
if (((ntohs(lacpdu->partner_port) == port->actor_port_number) &&
(ntohs(lacpdu->partner_port_priority) == port->actor_port_priority) &&
- !MAC_ADDRESS_COMPARE(&(lacpdu->partner_system), &(port->actor_system)) &&
+ !compare_ether_addr(lacpdu->partner_system, port->actor_system) &&
(ntohs(lacpdu->partner_system_priority) == port->actor_system_priority) &&
(ntohs(lacpdu->partner_key) == port->actor_oper_port_key) &&
((lacpdu->partner_state & AD_STATE_AGGREGATION) == (port->actor_oper_port_state & AD_STATE_AGGREGATION))) ||
@@ -510,7 +506,7 @@ static void __record_pdu(struct lacpdu *lacpdu, struct port *port)
// record the new parameter values for the partner operational
partner->port_number = ntohs(lacpdu->actor_port);
partner->port_priority = ntohs(lacpdu->actor_port_priority);
- partner->system = lacpdu->actor_system;
+ memcpy(partner->system, lacpdu->actor_system, ETH_ALEN);
partner->system_priority = ntohs(lacpdu->actor_system_priority);
partner->key = ntohs(lacpdu->actor_key);
partner->port_state = lacpdu->actor_state;
@@ -568,7 +564,7 @@ static void __update_selected(struct lacpdu *lacpdu, struct port *port)
// check if any parameter is different
if (ntohs(lacpdu->actor_port) != partner->port_number ||
ntohs(lacpdu->actor_port_priority) != partner->port_priority ||
- MAC_ADDRESS_COMPARE(&lacpdu->actor_system, &partner->system) ||
+ compare_ether_addr(lacpdu->actor_system, partner->system) ||
ntohs(lacpdu->actor_system_priority) != partner->system_priority ||
ntohs(lacpdu->actor_key) != partner->key ||
(lacpdu->actor_state & AD_STATE_AGGREGATION) != (partner->port_state & AD_STATE_AGGREGATION)) {
@@ -599,7 +595,7 @@ static void __update_default_selected(struct port *port)
// check if any parameter is different
if (admin->port_number != oper->port_number ||
admin->port_priority != oper->port_priority ||
- MAC_ADDRESS_COMPARE(&admin->system, &oper->system) ||
+ compare_ether_addr(admin->system, oper->system) ||
admin->system_priority != oper->system_priority ||
admin->key != oper->key ||
(admin->port_state & AD_STATE_AGGREGATION)
@@ -629,7 +625,7 @@ static void __update_ntt(struct lacpdu *lacpdu, struct port *port)
// check if any parameter is different
if ((ntohs(lacpdu->partner_port) != port->actor_port_number) ||
(ntohs(lacpdu->partner_port_priority) != port->actor_port_priority) ||
- MAC_ADDRESS_COMPARE(&(lacpdu->partner_system), &(port->actor_system)) ||
+ compare_ether_addr(lacpdu->partner_system, port->actor_system) ||
(ntohs(lacpdu->partner_system_priority) != port->actor_system_priority) ||
(ntohs(lacpdu->partner_key) != port->actor_oper_port_key) ||
((lacpdu->partner_state & AD_STATE_LACP_ACTIVITY) != (port->actor_oper_port_state & AD_STATE_LACP_ACTIVITY)) ||
@@ -787,7 +783,7 @@ static inline void __update_lacpdu_from_port(struct port *port)
*/
lacpdu->actor_system_priority = htons(port->actor_system_priority);
- lacpdu->actor_system = port->actor_system;
+ memcpy(lacpdu->actor_system, port->actor_system, ETH_ALEN);
lacpdu->actor_key = htons(port->actor_oper_port_key);
lacpdu->actor_port_priority = htons(port->actor_port_priority);
lacpdu->actor_port = htons(port->actor_port_number);
@@ -799,7 +795,7 @@ static inline void __update_lacpdu_from_port(struct port *port)
*/
lacpdu->partner_system_priority = htons(partner->system_priority);
- lacpdu->partner_system = partner->system;
+ memcpy(lacpdu->partner_system, partner->system, ETH_ALEN);
lacpdu->partner_key = htons(partner->key);
lacpdu->partner_port_priority = htons(partner->port_priority);
lacpdu->partner_port = htons(partner->port_number);
@@ -1133,7 +1129,7 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
break;
case AD_RX_CURRENT:
// detect loopback situation
- if (!MAC_ADDRESS_COMPARE(&(lacpdu->actor_system), &(port->actor_system))) {
+ if (!compare_ether_addr(lacpdu->actor_system, port->actor_system)) {
// INFO_RECEIVED_LOOPBACK_FRAMES
pr_err("%s: An illegal loopback occurred on adapter (%s).\n"
"Check the configuration to verify that all adapters are connected to 802.3ad compliant switch ports\n",
@@ -1340,11 +1336,11 @@ static void ad_port_selection_logic(struct port *port)
}
// check if current aggregator suits us
if (((aggregator->actor_oper_aggregator_key == port->actor_oper_port_key) && // if all parameters match AND
- !MAC_ADDRESS_COMPARE(&(aggregator->partner_system), &(port->partner_oper.system)) &&
+ !compare_ether_addr(aggregator->partner_system, port->partner_oper.system) &&
(aggregator->partner_system_priority == port->partner_oper.system_priority) &&
(aggregator->partner_oper_aggregator_key == port->partner_oper.key)
) &&
- ((MAC_ADDRESS_COMPARE(&(port->partner_oper.system), &(null_mac_addr)) && // partner answers
+ ((is_zero_ether_addr(port->partner_oper.system) && // partner answers
!aggregator->is_individual) // but is not individual OR
)
) {
@@ -1382,7 +1378,7 @@ static void ad_port_selection_logic(struct port *port)
port->aggregator->actor_admin_aggregator_key = port->actor_admin_port_key;
port->aggregator->actor_oper_aggregator_key = port->actor_oper_port_key;
- port->aggregator->partner_system=port->partner_oper.system;
+ memcpy(port->aggregator->partner_system, port->partner_oper.system, ETH_ALEN);
port->aggregator->partner_system_priority = port->partner_oper.system_priority;
port->aggregator->partner_oper_aggregator_key = port->partner_oper.key;
port->aggregator->receive_state = 1;
@@ -1636,7 +1632,7 @@ static void ad_clear_agg(struct aggregator *aggregator)
aggregator->is_individual = false;
aggregator->actor_admin_aggregator_key = 0;
aggregator->actor_oper_aggregator_key = 0;
- aggregator->partner_system = null_mac_addr;
+ memset(aggregator->partner_system, 0, ETH_ALEN);
aggregator->partner_system_priority = 0;
aggregator->partner_oper_aggregator_key = 0;
aggregator->receive_state = 0;
@@ -1659,7 +1655,7 @@ static void ad_initialize_agg(struct aggregator *aggregator)
if (aggregator) {
ad_clear_agg(aggregator);
- aggregator->aggregator_mac_address = null_mac_addr;
+ memset(aggregator->aggregator_mac_address, 0, ETH_ALEN);
aggregator->aggregator_identifier = 0;
aggregator->slave = NULL;
}
@@ -1695,7 +1691,7 @@ static void ad_initialize_port(struct port *port, int lacp_fast)
if (port) {
port->actor_port_number = 1;
port->actor_port_priority = 0xff;
- port->actor_system = null_mac_addr;
+ memset(port->actor_system, 0, ETH_ALEN);
port->actor_system_priority = 0xffff;
port->actor_port_aggregator_identifier = 0;
port->ntt = false;
@@ -1754,7 +1750,7 @@ static void ad_enable_collecting_distributing(struct port *port)
*/
static void ad_disable_collecting_distributing(struct port *port)
{
- if (port->aggregator && MAC_ADDRESS_COMPARE(&(port->aggregator->partner_system), &(null_mac_addr))) {
+ if (port->aggregator && is_zero_ether_addr(port->aggregator->partner_system)) {
pr_debug("Disabling port %d(LAG %d)\n",
port->actor_port_number,
port->aggregator->aggregator_identifier);
@@ -1875,14 +1871,14 @@ static u16 aggregator_identifier;
void bond_3ad_initialize(struct bonding *bond, u16 tick_resolution, int lacp_fast)
{
// check that the bond is not initialized yet
- if (MAC_ADDRESS_COMPARE(&(BOND_AD_INFO(bond).system.sys_mac_addr),
- bond->dev->dev_addr)) {
+ if (compare_ether_addr(BOND_AD_INFO(bond).system.sys_mac_addr,
+ bond->dev->dev_addr)) {
aggregator_identifier = 0;
BOND_AD_INFO(bond).lacp_fast = lacp_fast;
BOND_AD_INFO(bond).system.sys_priority = 0xFFFF;
- BOND_AD_INFO(bond).system.sys_mac_addr = *((struct mac_addr *)bond->dev->dev_addr);
+ memcpy(BOND_AD_INFO(bond).system.sys_mac_addr, bond->dev->dev_addr, ETH_ALEN);
// initialize how many times this module is called in one second(should be about every 100ms)
ad_ticks_per_sec = tick_resolution;
@@ -1936,7 +1932,7 @@ int bond_3ad_bind_slave(struct slave *slave)
port->sm_vars &= ~AD_PORT_LACP_ENABLED;
}
// actor system is the bond's system
- port->actor_system = BOND_AD_INFO(bond).system.sys_mac_addr;
+ memcpy(port->actor_system, BOND_AD_INFO(bond).system.sys_mac_addr, ETH_ALEN);
// tx timer(to verify that no more than MAX_TX_IN_SECOND lacpdu's are sent in one second)
port->sm_tx_timer_counter = ad_ticks_per_sec/AD_MAX_TX_IN_SECOND;
port->aggregator = NULL;
@@ -1951,7 +1947,7 @@ int bond_3ad_bind_slave(struct slave *slave)
ad_initialize_agg(aggregator);
- aggregator->aggregator_mac_address = *((struct mac_addr *)bond->dev->dev_addr);
+ memcpy(aggregator->aggregator_mac_address, bond->dev->dev_addr, ETH_ALEN);
aggregator->aggregator_identifier = (++aggregator_identifier);
aggregator->slave = slave;
aggregator->is_active = 0;
@@ -2027,7 +2023,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
new_aggregator->is_individual = aggregator->is_individual;
new_aggregator->actor_admin_aggregator_key = aggregator->actor_admin_aggregator_key;
new_aggregator->actor_oper_aggregator_key = aggregator->actor_oper_aggregator_key;
- new_aggregator->partner_system = aggregator->partner_system;
+ memcpy(new_aggregator->partner_system, aggregator->partner_system, ETH_ALEN);
new_aggregator->partner_system_priority = aggregator->partner_system_priority;
new_aggregator->partner_oper_aggregator_key = aggregator->partner_oper_aggregator_key;
new_aggregator->receive_state = aggregator->receive_state;
@@ -2371,7 +2367,7 @@ int bond_3ad_get_active_agg_info(struct bonding *bond, struct ad_info *ad_info)
ad_info->ports = aggregator->num_of_ports;
ad_info->actor_key = aggregator->actor_oper_aggregator_key;
ad_info->partner_key = aggregator->partner_oper_aggregator_key;
- memcpy(ad_info->partner_system, aggregator->partner_system.mac_addr_value, ETH_ALEN);
+ memcpy(ad_info->partner_system, aggregator->partner_system, ETH_ALEN);
return 0;
}
diff --git a/drivers/net/bonding/bond_3ad.h b/drivers/net/bonding/bond_3ad.h
index 2c46a154..648792c 100644
--- a/drivers/net/bonding/bond_3ad.h
+++ b/drivers/net/bonding/bond_3ad.h
@@ -37,10 +37,6 @@
#define AD_LACP_SLOW 0
#define AD_LACP_FAST 1
-typedef struct mac_addr {
- u8 mac_addr_value[ETH_ALEN];
-} mac_addr_t;
-
enum {
BOND_AD_STABLE = 0,
BOND_AD_BANDWIDTH = 1,
@@ -112,7 +108,7 @@ typedef struct lacpdu {
u8 tlv_type_actor_info; // = actor information(type/length/value)
u8 actor_information_length; // = 20
__be16 actor_system_priority;
- struct mac_addr actor_system;
+ u8 actor_system[ETH_ALEN];
__be16 actor_key;
__be16 actor_port_priority;
__be16 actor_port;
@@ -121,7 +117,7 @@ typedef struct lacpdu {
u8 tlv_type_partner_info; // = partner information
u8 partner_information_length; // = 20
__be16 partner_system_priority;
- struct mac_addr partner_system;
+ u8 partner_system[ETH_ALEN];
__be16 partner_key;
__be16 partner_port_priority;
__be16 partner_port;
@@ -149,7 +145,7 @@ typedef struct bond_marker {
// = 0x02 (marker response information)
u8 marker_length; // = 0x16
u16 requester_port; // The number assigned to the port by the requester
- struct mac_addr requester_system; // The requester's system id
+ u8 requester_system[ETH_ALEN]; // The requester's system id
u32 requester_transaction_id; // The transaction id allocated by the requester,
u16 pad; // = 0
u8 tlv_type_terminator; // = 0x00
@@ -175,12 +171,12 @@ struct port;
// aggregator structure(43.4.5 in the 802.3ad standard)
typedef struct aggregator {
- struct mac_addr aggregator_mac_address;
+ u8 aggregator_mac_address[ETH_ALEN];
u16 aggregator_identifier;
bool is_individual;
u16 actor_admin_aggregator_key;
u16 actor_oper_aggregator_key;
- struct mac_addr partner_system;
+ u8 partner_system[ETH_ALEN];
u16 partner_system_priority;
u16 partner_oper_aggregator_key;
u16 receive_state; // BOOLEAN
@@ -193,7 +189,7 @@ typedef struct aggregator {
} aggregator_t;
struct port_params {
- struct mac_addr system;
+ u8 system[ETH_ALEN];
u16 system_priority;
u16 key;
u16 port_number;
@@ -205,7 +201,7 @@ struct port_params {
typedef struct port {
u16 actor_port_number;
u16 actor_port_priority;
- struct mac_addr actor_system; // This parameter is added here although it is not specified in the standard, just for simplification
+ u8 actor_system[ETH_ALEN]; // This parameter is added here although it is not specified in the standard, just for simplification
u16 actor_system_priority; // This parameter is added here although it is not specified in the standard, just for simplification
u16 actor_port_aggregator_identifier;
bool ntt;
@@ -239,7 +235,7 @@ typedef struct port {
// system structure
struct ad_system {
u16 sys_priority;
- struct mac_addr sys_mac_addr;
+ u8 sys_mac_addr[ETH_ALEN];
};
#ifdef __ia64__
^ permalink raw reply related
* Re: [RFC PATCH 0/9] ipvs network name space (netns) aware
From: Simon Horman @ 2010-10-17 6:47 UTC (permalink / raw)
To: Hans Schillstrom
Cc: lvs-devel, netdev, netfilter-devel, ja, wensong, daniel.lezcano
In-Reply-To: <201010081316.38317.hans.schillstrom@ericsson.com>
On Fri, Oct 08, 2010 at 01:16:36PM +0200, Hans Schillstrom wrote:
> This patch series adds network name space (netns) support to the LVS.
>
> REVISION
>
> This is version 1
>
> OVERVIEW
>
> The patch doesn't remove or add any functionality except for netns.
> For users that don't use network name space (netns) this patch is
> completely transparent.
>
> No it's possible to run LVS in a Linux container (see lxc-tools)
> i.e. a light weight virtualization. For example it's possible to run
> one or several lvs on a real server in their own network name spaces.
> >From the LVS point of view it looks like it runs on it's own machine.
>
> IMPLEMENTATION
> Basic requirements for netns awareness
> - Global variables has to be moved to dyn. allocated memory.
>
> Most global variables now resides in a struct ipvs { } in netns/ip_vs.h.
> What is moved and what is not ?
>
> Some cache aligned locks are still in global, module init params and some debug_level.
>
> Algorithm files they are untouched.
>
> QUESTIONS
> Drop rate in ip_vs_ctl per netns or grand total ?
My gut-feeling is that per netns makes more sense.
> Should more lock variables be moved (or less) ?
I'm unsure what you are asking here but I will make a general statement
about locking in IPVS: it needs work.
>
> PATCH SET
> This patch set is based upon net-next-2.6 (2.6.36-rc3) from 4 oct 2010
> and [patch v4] ipvs: IPv6 tunnel mode
>
> Note: ip_vs_xmit.c will not work without "[patch v4] ipvs: IPv6 tunnel mode"
Unfortunately the patches don't apply with the persistence engine
patches which were recently merged into nf-next-2.6 (although
"[patch v4.1 ]ipvs: IPv6 tunnel mode" is still unmerged).
I'm happy to work with you to make the required changes there.
(I realise those patches weren't merged when you made your post.
But regardless, either your or me will need to update the patches).
Another issue is that your patches seem to be split in a way
where the build breaks along the way. E.g. after applying
patch 1, the build breaks. Could you please split things up
in a manner such that this doesn't happen. The reason being
that it breaks bisection.
Lastly, could you provide a unique subject for each patch.
I know its a bit tedious, but it does make a difference when
browsing the changelog.
^ permalink raw reply
* [PATCH 2/2] bonding: cleanup: remove braces from single block statements
From: Bandan Das @ 2010-10-17 6:19 UTC (permalink / raw)
To: David Miller; +Cc: LKML, NetDev, joe, Eric Dumazet, Bandan Das
In-Reply-To: <1287296399-25142-1-git-send-email-bandan.das@stratus.com>
checkpatch.pl cleanup : Remove braces from single statement
blocks.
Signed-off-by: Bandan Das <bandan.das@stratus.com>
---
drivers/net/bonding/bond_3ad.c | 163 ++++++++++++++++++----------------------
1 files changed, 72 insertions(+), 91 deletions(-)
diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index 6b06b00..881914b 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -129,9 +129,8 @@ static void ad_marker_response_received(struct bond_marker *marker, struct port
*/
static inline struct bonding *__get_bond_by_port(struct port *port)
{
- if (port->slave == NULL) {
+ if (port->slave == NULL)
return NULL;
- }
return bond_get_bond_by_slave(port->slave);
}
@@ -144,9 +143,8 @@ static inline struct bonding *__get_bond_by_port(struct port *port)
*/
static inline struct port *__get_first_port(struct bonding *bond)
{
- if (bond->slave_cnt == 0) {
+ if (bond->slave_cnt == 0)
return NULL;
- }
return &(SLAVE_AD_INFO(bond->first_slave).port);
}
@@ -164,9 +162,8 @@ static inline struct port *__get_next_port(struct port *port)
struct slave *slave = port->slave;
// If there's no bond for this port, or this is the last slave
- if ((bond == NULL) || (slave->next == bond->first_slave)) {
+ if ((bond == NULL) || (slave->next == bond->first_slave))
return NULL;
- }
return &(SLAVE_AD_INFO(slave->next).port);
}
@@ -183,9 +180,8 @@ static inline struct aggregator *__get_first_agg(struct port *port)
struct bonding *bond = __get_bond_by_port(port);
// If there's no bond for this port, or bond has no slaves
- if ((bond == NULL) || (bond->slave_cnt == 0)) {
+ if ((bond == NULL) || (bond->slave_cnt == 0))
return NULL;
- }
return &(SLAVE_AD_INFO(bond->first_slave).aggregator);
}
@@ -203,9 +199,8 @@ static inline struct aggregator *__get_next_agg(struct aggregator *aggregator)
struct bonding *bond = bond_get_bond_by_slave(slave);
// If there's no bond for this aggregator, or this is the last slave
- if ((bond == NULL) || (slave->next == bond->first_slave)) {
+ if ((bond == NULL) || (slave->next == bond->first_slave))
return NULL;
- }
return &(SLAVE_AD_INFO(slave->next).aggregator);
}
@@ -240,9 +235,8 @@ static inline void __enable_port(struct port *port)
{
struct slave *slave = port->slave;
- if ((slave->link == BOND_LINK_UP) && IS_UP(slave->dev)) {
+ if ((slave->link == BOND_LINK_UP) && IS_UP(slave->dev))
bond_set_slave_active_flags(slave);
- }
}
/**
@@ -265,9 +259,8 @@ static inline u32 __get_agg_selection_mode(struct port *port)
{
struct bonding *bond = __get_bond_by_port(port);
- if (bond == NULL) {
+ if (bond == NULL)
return BOND_AD_STABLE;
- }
return BOND_AD_INFO(bond).agg_select_mode;
}
@@ -281,9 +274,8 @@ static inline int __check_agg_selection_timer(struct port *port)
{
struct bonding *bond = __get_bond_by_port(port);
- if (bond == NULL) {
+ if (bond == NULL)
return 0;
- }
return BOND_AD_INFO(bond).agg_select_timer ? 1 : 0;
}
@@ -328,9 +320,9 @@ static u16 __get_link_speed(struct port *port)
* link down, it sets the speed to 0.
* This is done in spite of the fact that the e100 driver reports 0 to be
* compatible with MVT in the future.*/
- if (slave->link != BOND_LINK_UP) {
+ if (slave->link != BOND_LINK_UP)
speed = 0;
- } else {
+ else {
switch (slave->speed) {
case SPEED_10:
speed = AD_LINK_SPEED_BITMASK_10MBPS;
@@ -375,9 +367,9 @@ static u8 __get_duplex(struct port *port)
// handling a special case: when the configuration starts with
// link down, it sets the duplex to 0.
- if (slave->link != BOND_LINK_UP) {
+ if (slave->link != BOND_LINK_UP)
retval = 0x0;
- } else {
+ else {
switch (slave->duplex) {
case DUPLEX_FULL:
retval = 0x1;
@@ -423,11 +415,10 @@ static u16 __ad_timer_to_ticks(u16 timer_type, u16 par)
switch (timer_type) {
case AD_CURRENT_WHILE_TIMER: // for rx machine usage
- if (par) { // for short or long timeout
+ if (par)
retval = (AD_SHORT_TIMEOUT_TIME*ad_ticks_per_sec); // short timeout
- } else {
+ else
retval = (AD_LONG_TIMEOUT_TIME*ad_ticks_per_sec); // long timeout
- }
break;
case AD_ACTOR_CHURN_TIMER: // for local churn machine
retval = (AD_CHURN_DETECTION_TIME*ad_ticks_per_sec);
@@ -519,11 +510,11 @@ static void __record_pdu(struct lacpdu *lacpdu, struct port *port)
port->actor_oper_port_state &= ~AD_STATE_DEFAULTED;
// set the partner sync. to on if the partner is sync. and the port is matched
- if ((port->sm_vars & AD_PORT_MATCHED) && (lacpdu->actor_state & AD_STATE_SYNCHRONIZATION)) {
+ if ((port->sm_vars & AD_PORT_MATCHED)
+ && (lacpdu->actor_state & AD_STATE_SYNCHRONIZATION))
partner->port_state |= AD_STATE_SYNCHRONIZATION;
- } else {
+ else
partner->port_state &= ~AD_STATE_SYNCHRONIZATION;
- }
}
}
@@ -710,11 +701,10 @@ static void __set_agg_ports_ready(struct aggregator *aggregator, int val)
for (port = aggregator->lag_ports; port;
port = port->next_port_in_aggregator) {
- if (val) {
+ if (val)
port->sm_vars |= AD_PORT_READY;
- } else {
+ else
port->sm_vars &= ~AD_PORT_READY;
- }
}
}
@@ -838,9 +828,8 @@ static int ad_lacpdu_send(struct port *port)
int length = sizeof(struct lacpdu_header);
skb = dev_alloc_skb(length);
- if (!skb) {
+ if (!skb)
return -ENOMEM;
- }
skb->dev = slave->dev;
skb_reset_mac_header(skb);
@@ -879,9 +868,8 @@ static int ad_marker_send(struct port *port, struct bond_marker *marker)
int length = sizeof(struct bond_marker_header);
skb = dev_alloc_skb(length + 16);
- if (!skb) {
+ if (!skb)
return -ENOMEM;
- }
skb_reserve(skb, 16);
@@ -922,9 +910,10 @@ static void ad_mux_machine(struct port *port)
} else {
switch (port->sm_mux_state) {
case AD_MUX_DETACHED:
- if ((port->sm_vars & AD_PORT_SELECTED) || (port->sm_vars & AD_PORT_STANDBY)) { // if SELECTED or STANDBY
+ if ((port->sm_vars & AD_PORT_SELECTED)
+ || (port->sm_vars & AD_PORT_STANDBY))
+ /* if SELECTED or STANDBY */
port->sm_mux_state = AD_MUX_WAITING; // next state
- }
break;
case AD_MUX_WAITING:
// if SELECTED == FALSE return to DETACH state
@@ -938,18 +927,18 @@ static void ad_mux_machine(struct port *port)
}
// check if the wait_while_timer expired
- if (port->sm_mux_timer_counter && !(--port->sm_mux_timer_counter)) {
+ if (port->sm_mux_timer_counter
+ && !(--port->sm_mux_timer_counter))
port->sm_vars |= AD_PORT_READY_N;
- }
// in order to withhold the selection logic to check all ports READY_N value
// every callback cycle to update ready variable, we check READY_N and update READY here
__set_agg_ports_ready(port->aggregator, __agg_ports_are_ready(port->aggregator));
// if the wait_while_timer expired, and the port is in READY state, move to ATTACHED state
- if ((port->sm_vars & AD_PORT_READY) && !port->sm_mux_timer_counter) {
+ if ((port->sm_vars & AD_PORT_READY)
+ && !port->sm_mux_timer_counter)
port->sm_mux_state = AD_MUX_ATTACHED; // next state
- }
break;
case AD_MUX_ATTACHED:
// check also if agg_select_timer expired(so the edable port will take place only after this timer)
@@ -1044,13 +1033,14 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
// check if state machine should change state
// first, check if port was reinitialized
- if (port->sm_vars & AD_PORT_BEGIN) {
- port->sm_rx_state = AD_RX_INITIALIZE; // next state
- }
+ if (port->sm_vars & AD_PORT_BEGIN)
+ /* next state */
+ port->sm_rx_state = AD_RX_INITIALIZE;
// check if port is not enabled
- else if (!(port->sm_vars & AD_PORT_BEGIN) && !port->is_enabled && !(port->sm_vars & AD_PORT_MOVED)) {
- port->sm_rx_state = AD_RX_PORT_DISABLED; // next state
- }
+ else if (!(port->sm_vars & AD_PORT_BEGIN)
+ && !port->is_enabled && !(port->sm_vars & AD_PORT_MOVED))
+ /* next state */
+ port->sm_rx_state = AD_RX_PORT_DISABLED;
// check if new lacpdu arrived
else if (lacpdu && ((port->sm_rx_state == AD_RX_EXPIRED) || (port->sm_rx_state == AD_RX_DEFAULTED) || (port->sm_rx_state == AD_RX_CURRENT))) {
port->sm_rx_timer_counter = 0; // zero timer
@@ -1072,13 +1062,16 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
// if no lacpdu arrived and no timer is on
switch (port->sm_rx_state) {
case AD_RX_PORT_DISABLED:
- if (port->sm_vars & AD_PORT_MOVED) {
+ if (port->sm_vars & AD_PORT_MOVED)
port->sm_rx_state = AD_RX_INITIALIZE; // next state
- } else if (port->is_enabled && (port->sm_vars & AD_PORT_LACP_ENABLED)) {
+ else if (port->is_enabled
+ && (port->sm_vars
+ & AD_PORT_LACP_ENABLED))
port->sm_rx_state = AD_RX_EXPIRED; // next state
- } else if (port->is_enabled && ((port->sm_vars & AD_PORT_LACP_ENABLED) == 0)) {
+ else if (port->is_enabled
+ && ((port->sm_vars
+ & AD_PORT_LACP_ENABLED) == 0))
port->sm_rx_state = AD_RX_LACP_DISABLED; // next state
- }
break;
default: //to silence the compiler
break;
@@ -1094,11 +1087,10 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
port->sm_rx_state);
switch (port->sm_rx_state) {
case AD_RX_INITIALIZE:
- if (!(port->actor_oper_port_key & AD_DUPLEX_KEY_BITS)) {
+ if (!(port->actor_oper_port_key & AD_DUPLEX_KEY_BITS))
port->sm_vars &= ~AD_PORT_LACP_ENABLED;
- } else {
+ else
port->sm_vars |= AD_PORT_LACP_ENABLED;
- }
port->sm_vars &= ~AD_PORT_SELECTED;
__record_default(port);
port->actor_oper_port_state &= ~AD_STATE_EXPIRED;
@@ -1152,9 +1144,10 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
// verify that if the aggregator is enabled, the port is enabled too.
//(because if the link goes down for a short time, the 802.3ad will not
// catch it, and the port will continue to be disabled)
- if (port->aggregator && port->aggregator->is_active && !__port_is_enabled(port)) {
+ if (port->aggregator
+ && port->aggregator->is_active
+ && !__port_is_enabled(port))
__enable_port(port);
- }
break;
default: //to silence the compiler
break;
@@ -1220,9 +1213,9 @@ static void ad_periodic_machine(struct port *port)
// If not expired, check if there is some new timeout parameter from the partner state
switch (port->sm_periodic_state) {
case AD_FAST_PERIODIC:
- if (!(port->partner_oper.port_state & AD_STATE_LACP_TIMEOUT)) {
+ if (!(port->partner_oper.port_state
+ & AD_STATE_LACP_TIMEOUT))
port->sm_periodic_state = AD_SLOW_PERIODIC; // next state
- }
break;
case AD_SLOW_PERIODIC:
if ((port->partner_oper.port_state & AD_STATE_LACP_TIMEOUT)) {
@@ -1241,11 +1234,11 @@ static void ad_periodic_machine(struct port *port)
port->sm_periodic_state = AD_FAST_PERIODIC; // next state
break;
case AD_PERIODIC_TX:
- if (!(port->partner_oper.port_state & AD_STATE_LACP_TIMEOUT)) {
+ if (!(port->partner_oper.port_state
+ & AD_STATE_LACP_TIMEOUT))
port->sm_periodic_state = AD_SLOW_PERIODIC; // next state
- } else {
+ else
port->sm_periodic_state = AD_FAST_PERIODIC; // next state
- }
break;
default: //to silence the compiler
break;
@@ -1291,9 +1284,8 @@ static void ad_port_selection_logic(struct port *port)
int found = 0;
// if the port is already Selected, do nothing
- if (port->sm_vars & AD_PORT_SELECTED) {
+ if (port->sm_vars & AD_PORT_SELECTED)
return;
- }
// if the port is connected to other aggregator, detach it
if (port->aggregator) {
@@ -1321,9 +1313,8 @@ static void ad_port_selection_logic(struct port *port)
port->actor_port_number,
temp_aggregator->aggregator_identifier);
// if the aggregator is empty, clear its parameters, and set it ready to be attached
- if (!temp_aggregator->lag_ports) {
+ if (!temp_aggregator->lag_ports)
ad_clear_agg(temp_aggregator);
- }
break;
}
}
@@ -1341,9 +1332,8 @@ static void ad_port_selection_logic(struct port *port)
// keep a free aggregator for later use(if needed)
if (!aggregator->lag_ports) {
- if (!free_aggregator) {
+ if (!free_aggregator)
free_aggregator = aggregator;
- }
continue;
}
// check if current aggregator suits us
@@ -1384,11 +1374,11 @@ static void ad_port_selection_logic(struct port *port)
// update the new aggregator's parameters
// if port was responsed from the end-user
- if (port->actor_oper_port_key & AD_DUPLEX_KEY_BITS) {// if port is full duplex
+ if (port->actor_oper_port_key & AD_DUPLEX_KEY_BITS)
+ /* if port is full duplex */
port->aggregator->is_individual = false;
- } else {
+ else
port->aggregator->is_individual = true;
- }
port->aggregator->actor_admin_aggregator_key = port->actor_admin_port_key;
port->aggregator->actor_oper_aggregator_key = port->actor_oper_port_key;
@@ -1716,9 +1706,8 @@ static void ad_initialize_port(struct port *port, int lacp_fast)
port->actor_admin_port_state = AD_STATE_AGGREGATION | AD_STATE_LACP_ACTIVITY;
port->actor_oper_port_state = AD_STATE_AGGREGATION | AD_STATE_LACP_ACTIVITY;
- if (lacp_fast) {
+ if (lacp_fast)
port->actor_oper_port_state |= AD_STATE_LACP_TIMEOUT;
- }
memcpy(&port->partner_admin, &tmpl, sizeof(tmpl));
memcpy(&port->partner_oper, &tmpl, sizeof(tmpl));
@@ -1947,9 +1936,8 @@ int bond_3ad_bind_slave(struct slave *slave)
port->actor_admin_port_key |= (__get_link_speed(port) << 1);
port->actor_oper_port_key = port->actor_admin_port_key;
// if the port is not full duplex, then the port should be not lacp Enabled
- if (!(port->actor_oper_port_key & AD_DUPLEX_KEY_BITS)) {
+ if (!(port->actor_oper_port_key & AD_DUPLEX_KEY_BITS))
port->sm_vars &= ~AD_PORT_LACP_ENABLED;
- }
// actor system is the bond's system
port->actor_system = BOND_AD_INFO(bond).system.sys_mac_addr;
// tx timer(to verify that no more than MAX_TX_IN_SECOND lacpdu's are sent in one second)
@@ -2021,9 +2009,10 @@ void bond_3ad_unbind_slave(struct slave *slave)
new_aggregator = __get_first_agg(port);
for (; new_aggregator; new_aggregator = __get_next_agg(new_aggregator)) {
// if the new aggregator is empty, or it is connected to our port only
- if (!new_aggregator->lag_ports || ((new_aggregator->lag_ports == port) && !new_aggregator->lag_ports->next_port_in_aggregator)) {
+ if (!new_aggregator->lag_ports
+ || ((new_aggregator->lag_ports == port)
+ && !new_aggregator->lag_ports->next_port_in_aggregator))
break;
- }
}
// if new aggregator found, copy the aggregator's parameters
// and connect the related lag_ports to the new aggregator
@@ -2061,9 +2050,8 @@ void bond_3ad_unbind_slave(struct slave *slave)
// clear the aggregator
ad_clear_agg(aggregator);
- if (select_new_active_agg) {
+ if (select_new_active_agg)
ad_agg_selection_logic(__get_first_agg(port));
- }
} else {
pr_warning("%s: Warning: unbinding aggregator, and could not find a new aggregator for its ports\n",
slave->dev->master->name);
@@ -2091,11 +2079,10 @@ void bond_3ad_unbind_slave(struct slave *slave)
prev_port = temp_port,
temp_port = temp_port->next_port_in_aggregator) {
if (temp_port == port) { // the aggregator found - detach the port from this aggregator
- if (prev_port) {
+ if (prev_port)
prev_port->next_port_in_aggregator = temp_port->next_port_in_aggregator;
- } else {
+ else
temp_aggregator->lag_ports = temp_port->next_port_in_aggregator;
- }
temp_aggregator->num_of_ports--;
if (temp_aggregator->num_of_ports == 0) {
select_new_active_agg = temp_aggregator->is_active;
@@ -2137,14 +2124,12 @@ void bond_3ad_state_machine_handler(struct work_struct *work)
read_lock(&bond->lock);
- if (bond->kill_timers) {
+ if (bond->kill_timers)
goto out;
- }
//check if there are any slaves
- if (bond->slave_cnt == 0) {
+ if (bond->slave_cnt == 0)
goto re_arm;
- }
// check if agg_select_timer timer after initialize is timed out
if (BOND_AD_INFO(bond).agg_select_timer && !(--BOND_AD_INFO(bond).agg_select_timer)) {
@@ -2177,9 +2162,8 @@ void bond_3ad_state_machine_handler(struct work_struct *work)
ad_tx_machine(port);
// turn off the BEGIN bit, since we already handled it
- if (port->sm_vars & AD_PORT_BEGIN) {
+ if (port->sm_vars & AD_PORT_BEGIN)
port->sm_vars &= ~AD_PORT_BEGIN;
- }
}
re_arm:
@@ -2417,9 +2401,8 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
*/
read_lock(&bond->lock);
- if (!BOND_IS_OK(bond)) {
+ if (!BOND_IS_OK(bond))
goto out;
- }
if (bond_3ad_get_active_agg_info(bond, &ad_info)) {
pr_debug("%s: Error: bond_3ad_get_active_agg_info failed\n",
@@ -2443,9 +2426,8 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
if (agg && (agg->aggregator_identifier == agg_id)) {
slave_agg_no--;
- if (slave_agg_no < 0) {
+ if (slave_agg_no < 0)
break;
- }
}
}
@@ -2461,9 +2443,8 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
int slave_agg_id = 0;
struct aggregator *agg = SLAVE_AD_INFO(slave).port.aggregator;
- if (agg) {
+ if (agg)
slave_agg_id = agg->aggregator_identifier;
- }
if (SLAVE_IS_OK(slave) && agg && (slave_agg_id == agg_id)) {
res = bond_dev_queue_xmit(bond, skb, slave->dev);
--
1.7.2.3
^ permalink raw reply related
* [PATCH 1/2] bonding: cleanup : add space around operators
From: Bandan Das @ 2010-10-17 6:19 UTC (permalink / raw)
To: David Miller; +Cc: LKML, NetDev, joe, Eric Dumazet, Bandan Das
In-Reply-To: <1287296399-25142-1-git-send-email-bandan.das@stratus.com>
checkpatch.pl cleanup: Added spaces around operators at various places.
Also fixed some c99 style comments that I came across.
Signed-off-by: Bandan Das <bandan.das@stratus.com>
---
drivers/net/bonding/bond_3ad.c | 111 ++++++++++++++++++++++++----------------
1 files changed, 67 insertions(+), 44 deletions(-)
diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index 079b9d1..6b06b00 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -93,7 +93,7 @@
// compare MAC addresses
#define MAC_ADDRESS_COMPARE(A, B) memcmp(A, B, ETH_ALEN)
-static struct mac_addr null_mac_addr = {{0, 0, 0, 0, 0, 0}};
+static struct mac_addr null_mac_addr = { { 0, 0, 0, 0, 0, 0 } };
static u16 ad_ticks_per_sec;
static const int ad_delta_in_ticks = (AD_TIMER_INTERVAL * HZ) / 1000;
@@ -329,7 +329,7 @@ static u16 __get_link_speed(struct port *port)
* This is done in spite of the fact that the e100 driver reports 0 to be
* compatible with MVT in the future.*/
if (slave->link != BOND_LINK_UP) {
- speed=0;
+ speed = 0;
} else {
switch (slave->speed) {
case SPEED_10:
@@ -376,17 +376,17 @@ static u8 __get_duplex(struct port *port)
// handling a special case: when the configuration starts with
// link down, it sets the duplex to 0.
if (slave->link != BOND_LINK_UP) {
- retval=0x0;
+ retval = 0x0;
} else {
switch (slave->duplex) {
case DUPLEX_FULL:
- retval=0x1;
+ retval = 0x1;
pr_debug("Port %d Received status full duplex update from adapter\n",
port->actor_port_number);
break;
case DUPLEX_HALF:
default:
- retval=0x0;
+ retval = 0x0;
pr_debug("Port %d Received status NOT full duplex update from adapter\n",
port->actor_port_number);
break;
@@ -419,7 +419,7 @@ static inline void __initialize_port_locks(struct port *port)
*/
static u16 __ad_timer_to_ticks(u16 timer_type, u16 par)
{
- u16 retval=0; //to silence the compiler
+ u16 retval = 0; /* to silence the compiler */
switch (timer_type) {
case AD_CURRENT_WHILE_TIMER: // for rx machine usage
@@ -653,7 +653,7 @@ static void __update_ntt(struct lacpdu *lacpdu, struct port *port)
*/
static void __attach_bond_to_agg(struct port *port)
{
- port=NULL; // just to satisfy the compiler
+ port = NULL; /* just to satisfy the compiler */
// This function does nothing since the parser/multiplexer of the receive
// and the parser/multiplexer of the aggregator are already combined
}
@@ -668,7 +668,7 @@ static void __attach_bond_to_agg(struct port *port)
*/
static void __detach_bond_from_agg(struct port *port)
{
- port=NULL; // just to satisfy the compiler
+ port = NULL; /* just to satisfy the compiler */
// This function does nothing sience the parser/multiplexer of the receive
// and the parser/multiplexer of the aggregator are already combined
}
@@ -685,7 +685,9 @@ static int __agg_ports_are_ready(struct aggregator *aggregator)
if (aggregator) {
// scan all ports in this aggregator to verfy if they are all ready
- for (port=aggregator->lag_ports; port; port=port->next_port_in_aggregator) {
+ for (port = aggregator->lag_ports;
+ port;
+ port = port->next_port_in_aggregator) {
if (!(port->sm_vars & AD_PORT_READY_N)) {
retval = 0;
break;
@@ -706,7 +708,8 @@ static void __set_agg_ports_ready(struct aggregator *aggregator, int val)
{
struct port *port;
- for (port=aggregator->lag_ports; port; port=port->next_port_in_aggregator) {
+ for (port = aggregator->lag_ports; port;
+ port = port->next_port_in_aggregator) {
if (val) {
port->sm_vars |= AD_PORT_READY;
} else {
@@ -722,7 +725,7 @@ static void __set_agg_ports_ready(struct aggregator *aggregator, int val)
*/
static u32 __get_agg_bandwidth(struct aggregator *aggregator)
{
- u32 bandwidth=0;
+ u32 bandwidth = 0;
u32 basic_speed;
if (aggregator->num_of_ports) {
@@ -744,7 +747,7 @@ static u32 __get_agg_bandwidth(struct aggregator *aggregator)
bandwidth = aggregator->num_of_ports * 10000;
break;
default:
- bandwidth=0; // to silent the compilor ....
+ bandwidth = 0; /*to silence the compiler ....*/
}
}
return bandwidth;
@@ -1183,7 +1186,8 @@ static void ad_tx_machine(struct port *port)
}
}
// restart tx timer(to verify that we will not exceed AD_MAX_TX_IN_SECOND
- port->sm_tx_timer_counter=ad_ticks_per_sec/AD_MAX_TX_IN_SECOND;
+ port->sm_tx_timer_counter =
+ ad_ticks_per_sec/AD_MAX_TX_IN_SECOND;
}
}
@@ -1294,20 +1298,24 @@ static void ad_port_selection_logic(struct port *port)
// if the port is connected to other aggregator, detach it
if (port->aggregator) {
// detach the port from its former aggregator
- temp_aggregator=port->aggregator;
- for (curr_port=temp_aggregator->lag_ports; curr_port; last_port=curr_port, curr_port=curr_port->next_port_in_aggregator) {
+ temp_aggregator = port->aggregator;
+ for (curr_port = temp_aggregator->lag_ports; curr_port;
+ last_port = curr_port,
+ curr_port = curr_port->next_port_in_aggregator) {
if (curr_port == port) {
temp_aggregator->num_of_ports--;
if (!last_port) {// if it is the first port attached to the aggregator
- temp_aggregator->lag_ports=port->next_port_in_aggregator;
+ temp_aggregator->lag_ports =
+ port->next_port_in_aggregator;
} else {// not the first port attached to the aggregator
- last_port->next_port_in_aggregator=port->next_port_in_aggregator;
+ last_port->next_port_in_aggregator =
+ port->next_port_in_aggregator;
}
// clear the port's relations to this aggregator
port->aggregator = NULL;
- port->next_port_in_aggregator=NULL;
- port->actor_port_aggregator_identifier=0;
+ port->next_port_in_aggregator = NULL;
+ port->actor_port_aggregator_identifier = 0;
pr_debug("Port %d left LAG %d\n",
port->actor_port_number,
@@ -1334,7 +1342,7 @@ static void ad_port_selection_logic(struct port *port)
// keep a free aggregator for later use(if needed)
if (!aggregator->lag_ports) {
if (!free_aggregator) {
- free_aggregator=aggregator;
+ free_aggregator = aggregator;
}
continue;
}
@@ -1350,10 +1358,11 @@ static void ad_port_selection_logic(struct port *port)
) {
// attach to the founded aggregator
port->aggregator = aggregator;
- port->actor_port_aggregator_identifier=port->aggregator->aggregator_identifier;
- port->next_port_in_aggregator=aggregator->lag_ports;
+ port->actor_port_aggregator_identifier =
+ port->aggregator->aggregator_identifier;
+ port->next_port_in_aggregator = aggregator->lag_ports;
port->aggregator->num_of_ports++;
- aggregator->lag_ports=port;
+ aggregator->lag_ports = port;
pr_debug("Port %d joined LAG %d(existing LAG)\n",
port->actor_port_number,
port->aggregator->aggregator_identifier);
@@ -1370,7 +1379,8 @@ static void ad_port_selection_logic(struct port *port)
if (free_aggregator) {
// assign port a new aggregator
port->aggregator = free_aggregator;
- port->actor_port_aggregator_identifier=port->aggregator->aggregator_identifier;
+ port->actor_port_aggregator_identifier =
+ port->aggregator->aggregator_identifier;
// update the new aggregator's parameters
// if port was responsed from the end-user
@@ -1382,8 +1392,10 @@ static void ad_port_selection_logic(struct port *port)
port->aggregator->actor_admin_aggregator_key = port->actor_admin_port_key;
port->aggregator->actor_oper_aggregator_key = port->actor_oper_port_key;
- port->aggregator->partner_system=port->partner_oper.system;
- port->aggregator->partner_system_priority = port->partner_oper.system_priority;
+ port->aggregator->partner_system =
+ port->partner_oper.system;
+ port->aggregator->partner_system_priority =
+ port->partner_oper.system_priority;
port->aggregator->partner_oper_aggregator_key = port->partner_oper.key;
port->aggregator->receive_state = 1;
port->aggregator->transmit_state = 1;
@@ -1785,13 +1797,16 @@ static void ad_marker_info_send(struct port *port)
marker.requester_port = (((port->actor_port_number & 0xFF) << 8) |((u16)(port->actor_port_number & 0xFF00) >> 8));
marker.requester_system = port->actor_system;
// convert requester_port(u32) to Big Endian
- marker.requester_transaction_id = (((++port->transaction_id & 0xFF) << 24) |((port->transaction_id & 0xFF00) << 8) |((port->transaction_id & 0xFF0000) >> 8) |((port->transaction_id & 0xFF000000) >> 24));
+ marker.requester_transaction_id =
+ (((++port->transaction_id & 0xFF) << 24)
+ | ((port->transaction_id & 0xFF00) << 8)
+ | ((port->transaction_id & 0xFF0000) >> 8)
+ | ((port->transaction_id & 0xFF000000) >> 24));
marker.pad = 0;
marker.tlv_type_terminator = 0x00;
marker.terminator_length = 0x00;
- for (index=0; index<90; index++) {
- marker.reserved_90[index]=0;
- }
+ for (index = 0; index < 90; index++)
+ marker.reserved_90[index] = 0;
// send the marker information
if (ad_marker_send(port, &marker) >= 0) {
@@ -1816,7 +1831,7 @@ static void ad_marker_info_received(struct bond_marker *marker_info,
//marker = *marker_info;
memcpy(&marker, marker_info, sizeof(struct bond_marker));
// change the marker subtype to marker response
- marker.tlv_type=AD_MARKER_RESPONSE_SUBTYPE;
+ marker.tlv_type = AD_MARKER_RESPONSE_SUBTYPE;
// send the marker response
if (ad_marker_send(port, &marker) >= 0) {
@@ -1837,8 +1852,8 @@ static void ad_marker_info_received(struct bond_marker *marker_info,
static void ad_marker_response_received(struct bond_marker *marker,
struct port *port)
{
- marker=NULL; // just to satisfy the compiler
- port=NULL; // just to satisfy the compiler
+ marker = NULL; /* just to satisfy the compiler */
+ port = NULL; /* just to satisfy the compiler */
// DO NOTHING, SINCE WE DECIDED NOT TO IMPLEMENT THIS FEATURE FOR NOW
}
@@ -2037,8 +2052,9 @@ void bond_3ad_unbind_slave(struct slave *slave)
new_aggregator->num_of_ports = aggregator->num_of_ports;
// update the information that is written on the ports about the aggregator
- for (temp_port=aggregator->lag_ports; temp_port; temp_port=temp_port->next_port_in_aggregator) {
- temp_port->aggregator=new_aggregator;
+ for (temp_port = aggregator->lag_ports; temp_port;
+ temp_port = temp_port->next_port_in_aggregator) {
+ temp_port->aggregator = new_aggregator;
temp_port->actor_port_aggregator_identifier = new_aggregator->aggregator_identifier;
}
@@ -2071,7 +2087,9 @@ void bond_3ad_unbind_slave(struct slave *slave)
for (; temp_aggregator; temp_aggregator = __get_next_agg(temp_aggregator)) {
prev_port = NULL;
// search the port in the aggregator's related ports
- for (temp_port=temp_aggregator->lag_ports; temp_port; prev_port=temp_port, temp_port=temp_port->next_port_in_aggregator) {
+ for (temp_port = temp_aggregator->lag_ports; temp_port;
+ prev_port = temp_port,
+ temp_port = temp_port->next_port_in_aggregator) {
if (temp_port == port) { // the aggregator found - detach the port from this aggregator
if (prev_port) {
prev_port->next_port_in_aggregator = temp_port->next_port_in_aggregator;
@@ -2079,7 +2097,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
temp_aggregator->lag_ports = temp_port->next_port_in_aggregator;
}
temp_aggregator->num_of_ports--;
- if (temp_aggregator->num_of_ports==0) {
+ if (temp_aggregator->num_of_ports == 0) {
select_new_active_agg = temp_aggregator->is_active;
// clear the aggregator
ad_clear_agg(temp_aggregator);
@@ -2094,7 +2112,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
}
}
}
- port->slave=NULL;
+ port->slave = NULL;
}
/**
@@ -2245,7 +2263,8 @@ void bond_3ad_adapter_speed_changed(struct slave *slave)
}
port->actor_admin_port_key &= ~AD_SPEED_KEY_BITS;
- port->actor_oper_port_key=port->actor_admin_port_key |= (__get_link_speed(port) << 1);
+ port->actor_oper_port_key = port->actor_admin_port_key |=
+ (__get_link_speed(port) << 1);
pr_debug("Port %d changed speed\n", port->actor_port_number);
// there is no need to reselect a new aggregator, just signal the
// state machines to reinitialize
@@ -2262,7 +2281,7 @@ void bond_3ad_adapter_duplex_changed(struct slave *slave)
{
struct port *port;
- port=&(SLAVE_AD_INFO(slave).port);
+ port = &(SLAVE_AD_INFO(slave).port);
// if slave is null, the whole port is not initialized
if (!port->slave) {
@@ -2272,7 +2291,8 @@ void bond_3ad_adapter_duplex_changed(struct slave *slave)
}
port->actor_admin_port_key &= ~AD_DUPLEX_KEY_BITS;
- port->actor_oper_port_key=port->actor_admin_port_key |= __get_duplex(port);
+ port->actor_oper_port_key = port->actor_admin_port_key |=
+ __get_duplex(port);
pr_debug("Port %d changed duplex\n", port->actor_port_number);
// there is no need to reselect a new aggregator, just signal the
// state machines to reinitialize
@@ -2304,14 +2324,17 @@ void bond_3ad_handle_link_change(struct slave *slave, char link)
if (link == BOND_LINK_UP) {
port->is_enabled = true;
port->actor_admin_port_key &= ~AD_DUPLEX_KEY_BITS;
- port->actor_oper_port_key=port->actor_admin_port_key |= __get_duplex(port);
+ port->actor_oper_port_key = port->actor_admin_port_key |=
+ __get_duplex(port);
port->actor_admin_port_key &= ~AD_SPEED_KEY_BITS;
- port->actor_oper_port_key=port->actor_admin_port_key |= (__get_link_speed(port) << 1);
+ port->actor_oper_port_key = port->actor_admin_port_key |=
+ (__get_link_speed(port) << 1);
} else {
/* link has failed */
port->is_enabled = false;
port->actor_admin_port_key &= ~AD_DUPLEX_KEY_BITS;
- port->actor_oper_port_key= (port->actor_admin_port_key &= ~AD_SPEED_KEY_BITS);
+ port->actor_oper_port_key = (port->actor_admin_port_key &=
+ ~AD_SPEED_KEY_BITS);
}
//BOND_PRINT_DBG(("Port %d changed link status to %s", port->actor_port_number, ((link == BOND_LINK_UP)?"UP":"DOWN")));
// there is no need to reselect a new aggregator, just signal the
--
1.7.2.3
^ permalink raw reply related
* [PATCH 0/2] bonding: checkpatch.pl cleanups in bond_3ad.c
From: Bandan Das @ 2010-10-17 6:19 UTC (permalink / raw)
To: David Miller; +Cc: LKML, NetDev, joe, Eric Dumazet, Bandan Das
Just some cleanups based on errors/warnings reported by checkpatch.pl.
There are still a ton of lines greater than 80 chars and I tried to fix
a few that I came across.
Bandan Das (2):
bonding: cleanup : add space around operators
bonding: cleanup: remove braces from single block statements
drivers/net/bonding/bond_3ad.c | 274 ++++++++++++++++++++--------------------
1 files changed, 139 insertions(+), 135 deletions(-)
--
Bandan
^ permalink raw reply
* Re: Documenting UNIX domain autobind
From: Tetsuo Handa @ 2010-10-17 5:28 UTC (permalink / raw)
To: mtk.manpages; +Cc: netdev, mzxreary
In-Reply-To: <AANLkTimtTazMmDvQFgASsVGa=AYvPZqUaUmyg5p7VVY1@mail.gmail.com>
Michael Kerrisk wrote:
> I think this text correctly documents the technical details (but let
> me know if you see errors). What is lacking is an explanation of why
> this feature exists. Is someone able to explain where this feature is
> used and why?
What we can see is that unix_autobind() was added in Linux 2.1.15
http://lxr.linux.no/linux-old+v2.1.15/net/unix/af_unix.c#L464
addr->len = sprintf(addr->name->sun_path+1, "%08x", ordernum) + 1 + sizeof(short);
49 * Differences from 2.0.0-11-... (ANK)
50 * Bug fixes and improvements.
51 * - client shutdown killed server socket.
52 * - removed all useless cli/sti pairs.
53 * - (suspicious!) not allow connect/send to connected not to us
54 * socket, return EPERM.
55 *
56 * Semantic changes/extensions.
57 * - generic control message passing.
58 * - SCM_CREDENTIALS control message.
59 * - "Abstract" (not FS based) socket bindings.
60 * Abstract names are sequences of bytes (not zero terminated)
61 * started by 0, so that this name space does not intersect
62 * with BSD names.
and was changed to allow from 2^32 names to 2^20 names in Linux 2.3.15.
http://lxr.linux.no/linux-old+v2.3.15/net/unix/af_unix.c#L514
addr->len = sprintf(addr->name->sun_path+1, "%05x", ordernum) + 1 + sizeof(short);
I don't know the reason.
Regards.
^ permalink raw reply
* Re: [PATCH 2/3] cxgb4: function namespace cleanup (v2)
From: Steve Wise @ 2010-10-17 3:30 UTC (permalink / raw)
To: Dimitris Michailidis
Cc: Stephen Hemminger, Divy Le Ray, David S. Miller, Casey Leedom,
netdev
In-Reply-To: <4CB9434D.1060702@chelsio.com>
On 10/16/2010 1:16 AM, Dimitris Michailidis wrote:
> Stephen Hemminger wrote:
>> On Fri, 15 Oct 2010 18:11:42 -0700
>> Dimitris Michailidis <dm@chelsio.com> wrote:
>>
>>> Stephen Hemminger wrote:
>>>> Make functions only used in one file local.
>>>> Remove lots of dead code, relating to unsupported functions
>>>> in mainline driver like RSS, IPv6, and TCP offload.
>>> Thanks, this looks OK. One exception, cxgb4_get_tcp_stats was
>>> intended to be used by the rdma driver. I see that driver doesn't
>>> call it presently but if you don't mind can we give Steve a few
>>> hours to tell us if he has any imminent plans to use it. If he
>>> doesn't offer to do something to use it for .37 it goes.
>>
>> The kernel source tree is not your development place holder tree.
>> At least #ifdef the code out for now.
>
> I am trying to protect Stephen Rothwell's time by checking that the IB
> folks don't plan to add a call to this in their tree while we remove
> the function in net-next. There's supposed to be a call in the IB
> driver. I don't know why there isn't one or whether they are planning
> to fix it for .37. I see the potential for a linux-next conflict and
> I am trying to avoid it. #ifdef doesn't help, if it's not needed we
> can remove it for good.
I'll add a patch this week to utilize the tcp stats.
Steve.
^ permalink raw reply
* [PATCH resend] netfilter: fix kconfig unmet dependency warning
From: Randy Dunlap @ 2010-10-17 2:11 UTC (permalink / raw)
To: netdev; +Cc: davem, netfilter-devel, Patrick McHardy
From: Randy Dunlap <randy.dunlap@oracle.com>
Fix netfilter kconfig unmet dependencies warning & spell out
"compatible" while there.
warning: (IP_NF_TARGET_TTL && NET && INET && NETFILTER && IP_NF_IPTABLES && NETFILTER_ADVANCED || IP6_NF_TARGET_HL && NET && INET && IPV6 && NETFILTER && IP6_NF_IPTABLES && NETFILTER_ADVANCED) selects NETFILTER_XT_TARGET_HL which has unmet direct dependencies ((IP_NF_MANGLE || IP6_NF_MANGLE) && NETFILTER_ADVANCED)
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
---
net/ipv4/netfilter/Kconfig | 4 ++--
net/ipv6/netfilter/Kconfig | 4 ++--
2 files changed, 4 insertions(+), 4 deletions(-)
[orig. sent 2010-SEP-23]
--- linux-next-20100921.orig/net/ipv4/netfilter/Kconfig
+++ linux-next-20100921/net/ipv4/netfilter/Kconfig
@@ -324,10 +324,10 @@ config IP_NF_TARGET_ECN
config IP_NF_TARGET_TTL
tristate '"TTL" target support'
- depends on NETFILTER_ADVANCED
+ depends on NETFILTER_ADVANCED && IP_NF_MANGLE
select NETFILTER_XT_TARGET_HL
---help---
- This is a backwards-compat option for the user's convenience
+ This is a backwards-compatible option for the user's convenience
(e.g. when running oldconfig). It selects
CONFIG_NETFILTER_XT_TARGET_HL.
--- linux-next-20100921.orig/net/ipv6/netfilter/Kconfig
+++ linux-next-20100921/net/ipv6/netfilter/Kconfig
@@ -132,10 +132,10 @@ config IP6_NF_MATCH_RT
# The targets
config IP6_NF_TARGET_HL
tristate '"HL" hoplimit target support'
- depends on NETFILTER_ADVANCED
+ depends on NETFILTER_ADVANCED && IP6_NF_MANGLE
select NETFILTER_XT_TARGET_HL
---help---
- This is a backwards-compat option for the user's convenience
+ This is a backwards-compatible option for the user's convenience
(e.g. when running oldconfig). It selects
CONFIG_NETFILTER_XT_TARGET_HL.
--
^ permalink raw reply
* Re: tbf/htb qdisc limitations
From: Bill Fink @ 2010-10-17 1:24 UTC (permalink / raw)
To: Jarek Poplawski; +Cc: Eric Dumazet, Rick Jones, Steven Brudenell, netdev
In-Reply-To: <20101016205824.GA2113@del.dom.local>
On Sat, 16 Oct 2010, Jarek Poplawski wrote:
> On Sat, Oct 16, 2010 at 12:51:06AM -0400, Bill Fink wrote:
> > On Sat, 16 Oct 2010, Jarek Poplawski wrote:
> >
> > > On Fri, Oct 15, 2010 at 05:37:46PM -0400, Bill Fink wrote:
> > > ...
> > > > i7test7% tc -s -d qdisc show dev eth2
> > > > qdisc prio 1: root refcnt 33 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
> > > > Sent 11028687119 bytes 1223828 pkt (dropped 293, overlimits 0 requeues 0)
> > > > backlog 0b 0p requeues 0
> > > > qdisc tbf 10: parent 1:1 rate 8900Mbit burst 1112500b/64 mpu 0b lat 4295.0s
> > > > Sent 11028687077 bytes 1223827 pkt (dropped 293, overlimits 593 requeues 0)
> > > > backlog 0b 0p requeues 0
> > > >
> > > > I'm not sure how you can have so many dropped but not have
> > > > any TCP retransmissions (or not show up as requeues). But
> > > > there's probably something basic I just don't understand
> > > > about how all this stuff works.
> > >
> > > Me either, but it seems higher "limit" might help with these drops.
> >
> > You were of course correct about the higher limit helping.
> > I finally upgraded the field system to 2.6.35, and did some
> > testing on the real data path of interest, which has an RTT
> > of about 29 ms. I set up a rate limit of 8 Gbps using the
> > following commands:
> >
> > tc qdisc add dev eth2 root handle 1: prio
> > tc qdisc add dev eth2 parent 1:1 handle 10: tbf rate 8000mbit limit 35000000 burst 20000 mtu 9000
> > tc filter add dev eth2 protocol ip parent 1: prio 1 u32 match ip protocol 6 0xff match ip dst 192.168.1.23 flowid 10:1
> >
> > hecn-i7sl1% nuttcp -T10 -i1 -w50m 192.168.1.23
> > 676.3750 MB / 1.00 sec = 5673.4646 Mbps 0 retrans
> > 948.5625 MB / 1.00 sec = 7957.1508 Mbps 0 retrans
> > 948.8125 MB / 1.00 sec = 7959.5902 Mbps 0 retrans
> > 948.3750 MB / 1.00 sec = 7955.5382 Mbps 0 retrans
> > 949.0000 MB / 1.00 sec = 7960.6696 Mbps 0 retrans
> > 948.7500 MB / 1.00 sec = 7958.7873 Mbps 0 retrans
> > 948.6875 MB / 1.00 sec = 7958.0959 Mbps 0 retrans
> > 948.6250 MB / 1.00 sec = 7957.4205 Mbps 0 retrans
> > 948.7500 MB / 1.00 sec = 7958.7237 Mbps 0 retrans
> > 948.4375 MB / 1.00 sec = 7956.3648 Mbps 0 retrans
> >
> > 9270.5625 MB / 10.09 sec = 7707.7457 Mbps 24 %TX 36 %RX 0 retrans 29.38 msRTT
> >
> > hecn-i7sl1% tc -s -d qdisc show dev eth2
> > qdisc prio 1: root refcnt 33 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
> > Sent 9779476756 bytes 1084943 pkt (dropped 0, overlimits 0 requeues 0)
> > backlog 0b 0p requeues 0
> > qdisc tbf 10: parent 1:1 rate 8000Mbit burst 19000b/64 mpu 0b lat 35.0ms
> > Sent 9779476756 bytes 1084943 pkt (dropped 0, overlimits 1831360 requeues 0)
> > backlog 0b 0p requeues 0
> >
> > No drops!
> >
> > BTW the effective rate limit seems to be a very coarse adjustment
> > at these speeds. I was seeing some data path issues at 8.9 Gbps
> > so I tried setting slightly lower rates such as 8.8 Gbps, 8.7 Gbps,
> > etc, but they still gave me an effective rate limit of about 8.9 Gbps.
> > It wasn't until I got down to a setting of 8 Gbps that I actually
> > got an effective rate limit of 8 Gbps.
> >
> > Also the man page for tbf seems to be wrong/misleading about
> > the burst parameter. It states:
> >
> > "If your buffer is too small, packets may be dropped because more
> > tokens arrive per timer tick than fit in your bucket. The minimum
> > buffer size can be calculated by dividing the rate by HZ.
> >
> > According to that, with a rate of 8 Gbps and HZ=1000, the minimum
> > burst should be 1000000 bytes. But my testing shows that a burst
> > of just 20000 works just fine. That's only 2 9000-byte packets
> > or about 20 usec of traffic at the 8 Gbps rate. Using too large
> > a value for burst can actually be harmful as it allows the traffic
> > to temporarily exceed the desired rate limit.
>
> As I mentioned before, it could work, but your config is really on
> the edge. Anyway, if lower than minimum buffer size is needed
> something else is definitely wrong. (Btw, this size can matter less
> with high resolution timers.) You could try if my iproute patch:
> "tc_core: Use double in tc_core_time2tick()" (not merged) can help
> here. While googling for this patch I found this page, which might be
> interesting to you (besides the link to the thread with the patch at
> the end, take 1 or 2, shouldn't matter):
>
> http://code.google.com/p/pspacer/wiki/HTBon10GbE
>
> If it doesn't help reconsider hfsc.
Thanks for the link. From his results, it appears you can
get better accuracy by keeping TSO/GSO enabled and upping
the tc mtu parameter to 64000. I will have to try that out.
For the very high bandwidth cases I tend to deal with, would
there be any advantage to further reducing the PSCHED_SHIFT
from its current value of 6?
-Bill
^ permalink raw reply
* Re: "src" attribute ignored for IPv6 (preferred source address selection)
From: Daniel Roesen @ 2010-10-17 1:12 UTC (permalink / raw)
To: netdev
In-Reply-To: <4C8215C8.8000908@httrack.com>
On Sat, Sep 04, 2010 at 11:47:52AM +0200, Xavier Roche wrote:
> In an attempt to change the source address in various conditions (network
> prefix, outgoing protocol, user owner id, ..), it appeared that the src
> attribute of an IPv6 route command (ip -6 route add .. src ..) was simply
> ignored.
>
> The outgoing interface can be selected, but not the preferred source
> address (see (1)), which remains always the same within an interface.
>
> If this a bug ? Or a known limitation of the kernel ?
Well, at least it was a known limitation six years ago, when I
discovered the same:
http://lkml.indiana.edu/hypermail/linux/kernel/0409.0/1768.html
Unfortunately I don't have the time (and most probably not the necessary
kernel knowhow) to hack this up myself, but I'm still interested. :-)
Best regards,
Daniel
--
CLUE-RIPE -- Jabber: dr@cluenet.de -- dr@IRCnet -- PGP: 0xA85C8AA0
^ permalink raw reply
* r8169: high CPU load (softirq) bottlenecking throughput
From: Lasse Makholm @ 2010-10-16 21:54 UTC (permalink / raw)
To: netdev
Hello netdev people,
I'm seeing what seems to be an unreasonably high CPU utilization from
the r8169 driver on an Intel Atom board.
Serving a 1GB blob repeatedly through Apache basically maxes out one
CPU core completely...
Here is a typical snapshot of the CPU load:
11:17:23 PM CPU %usr %nice %sys %iowait %irq %soft
%steal %guest %idle
11:17:24 PM all 0.00 0.00 6.54 0.00 0.00 17.68
0.00 0.00 75.79
11:17:24 PM 0 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
11:17:24 PM 1 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
11:17:24 PM 2 0.00 0.00 27.00 0.00 0.00 73.00
0.00 0.00 0.00
11:17:24 PM 3 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
There is no disk IO while this is going on (the 1GB blob is cached),
and the machine is otherwise idle.
Throughput tops out at around 50MB/s but the receiving end and the
switch is easily able to get 100MB/s from another box.
The board is a D945GCLF2D with an Atom 330 (2 x 1.6 GHz), so
presumably it should be able to pump out 100MB/s easily...
For SSH-tunneled traffic, this is a disaster because both encryption
and I/O land on the same CPU... I'm currently maxing out at 10 - 20
MB/s with SSH-tunneled traffic
The kernel is currently 2.6.36-020636rc8-generic (Ubuntu daily
snapshot) but I've tried a handful of others between here and Ubuntu's
2.6.32-24....
lspci knows the following about the NIC:
makholm@korovyov:~$ lspci -vv -s 01:00.0
01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.
RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 02)
Subsystem: Intel Corporation Device 0001
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B- DisINTx+
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort-
<TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 0, Cache Line Size: 64 bytes
Interrupt: pin A routed to IRQ 40
Region 0: I/O ports at 1000 [size=256]
Region 2: Memory at 88100000 (64-bit, non-prefetchable) [size=4K]
Region 4: Memory at 88000000 (64-bit, prefetchable) [size=64K]
Expansion ROM at 88020000 [disabled] [size=128K]
Capabilities: <access denied>
Kernel driver in use: r8169
Kernel modules: r8169
makholm@korovyov:~$
I have a D-Link DGE-528T (also uses the r8169 driver) that shows
similar behavior in the Atom box. Trying that cards in an Athlon II
box also show a noticable CPU load by the r8169 driver, though the
Athlon obviously copes much better...
Any tips on how to improve the situation? I'd be happy to provide more
info and/or test some patches etc. if needed...
Thanks
/Lasse
^ permalink raw reply
* How to change delayed ACKs?
From: Fred . @ 2010-10-16 21:57 UTC (permalink / raw)
To: linux-net
I play online games.
Someone told me about Leatrix Latency Fix.
http://www.wowinterface.com/downloads/info13581-LeatrixLatencyFix.html
ACKs are delayed. The latency fix works by changing setting to
decrease that delay.
On Windows it modifies 'TCPAckFrequency' in the registry.
On FreeBSD and Mac OS X;
sudo sysctl -w net.inet.tcp.delayed_ack=0
This key doesn't seem to exist on Linux.
error: "net.inet.tcp.delayed_ack" is an unknown key
Is there any way to reduce the delay of delayed ACKs in order to
increase performance in online gaming?
P.S im not subscribed to mailinglist
^ permalink raw reply
* Re: tbf/htb qdisc limitations
From: Jarek Poplawski @ 2010-10-16 20:58 UTC (permalink / raw)
To: Bill Fink; +Cc: Eric Dumazet, Rick Jones, Steven Brudenell, netdev
In-Reply-To: <20101016005106.35e4cc8d.billfink@mindspring.com>
On Sat, Oct 16, 2010 at 12:51:06AM -0400, Bill Fink wrote:
> On Sat, 16 Oct 2010, Jarek Poplawski wrote:
>
> > On Fri, Oct 15, 2010 at 05:37:46PM -0400, Bill Fink wrote:
> > ...
> > > i7test7% tc -s -d qdisc show dev eth2
> > > qdisc prio 1: root refcnt 33 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
> > > Sent 11028687119 bytes 1223828 pkt (dropped 293, overlimits 0 requeues 0)
> > > backlog 0b 0p requeues 0
> > > qdisc tbf 10: parent 1:1 rate 8900Mbit burst 1112500b/64 mpu 0b lat 4295.0s
> > > Sent 11028687077 bytes 1223827 pkt (dropped 293, overlimits 593 requeues 0)
> > > backlog 0b 0p requeues 0
> > >
> > > I'm not sure how you can have so many dropped but not have
> > > any TCP retransmissions (or not show up as requeues). But
> > > there's probably something basic I just don't understand
> > > about how all this stuff works.
> >
> > Me either, but it seems higher "limit" might help with these drops.
>
> You were of course correct about the higher limit helping.
> I finally upgraded the field system to 2.6.35, and did some
> testing on the real data path of interest, which has an RTT
> of about 29 ms. I set up a rate limit of 8 Gbps using the
> following commands:
>
> tc qdisc add dev eth2 root handle 1: prio
> tc qdisc add dev eth2 parent 1:1 handle 10: tbf rate 8000mbit limit 35000000 burst 20000 mtu 9000
> tc filter add dev eth2 protocol ip parent 1: prio 1 u32 match ip protocol 6 0xff match ip dst 192.168.1.23 flowid 10:1
>
> hecn-i7sl1% nuttcp -T10 -i1 -w50m 192.168.1.23
> 676.3750 MB / 1.00 sec = 5673.4646 Mbps 0 retrans
> 948.5625 MB / 1.00 sec = 7957.1508 Mbps 0 retrans
> 948.8125 MB / 1.00 sec = 7959.5902 Mbps 0 retrans
> 948.3750 MB / 1.00 sec = 7955.5382 Mbps 0 retrans
> 949.0000 MB / 1.00 sec = 7960.6696 Mbps 0 retrans
> 948.7500 MB / 1.00 sec = 7958.7873 Mbps 0 retrans
> 948.6875 MB / 1.00 sec = 7958.0959 Mbps 0 retrans
> 948.6250 MB / 1.00 sec = 7957.4205 Mbps 0 retrans
> 948.7500 MB / 1.00 sec = 7958.7237 Mbps 0 retrans
> 948.4375 MB / 1.00 sec = 7956.3648 Mbps 0 retrans
>
> 9270.5625 MB / 10.09 sec = 7707.7457 Mbps 24 %TX 36 %RX 0 retrans 29.38 msRTT
>
> hecn-i7sl1% tc -s -d qdisc show dev eth2
> qdisc prio 1: root refcnt 33 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
> Sent 9779476756 bytes 1084943 pkt (dropped 0, overlimits 0 requeues 0)
> backlog 0b 0p requeues 0
> qdisc tbf 10: parent 1:1 rate 8000Mbit burst 19000b/64 mpu 0b lat 35.0ms
> Sent 9779476756 bytes 1084943 pkt (dropped 0, overlimits 1831360 requeues 0)
> backlog 0b 0p requeues 0
>
> No drops!
>
> BTW the effective rate limit seems to be a very coarse adjustment
> at these speeds. I was seeing some data path issues at 8.9 Gbps
> so I tried setting slightly lower rates such as 8.8 Gbps, 8.7 Gbps,
> etc, but they still gave me an effective rate limit of about 8.9 Gbps.
> It wasn't until I got down to a setting of 8 Gbps that I actually
> got an effective rate limit of 8 Gbps.
>
> Also the man page for tbf seems to be wrong/misleading about
> the burst parameter. It states:
>
> "If your buffer is too small, packets may be dropped because more
> tokens arrive per timer tick than fit in your bucket. The minimum
> buffer size can be calculated by dividing the rate by HZ.
>
> According to that, with a rate of 8 Gbps and HZ=1000, the minimum
> burst should be 1000000 bytes. But my testing shows that a burst
> of just 20000 works just fine. That's only 2 9000-byte packets
> or about 20 usec of traffic at the 8 Gbps rate. Using too large
> a value for burst can actually be harmful as it allows the traffic
> to temporarily exceed the desired rate limit.
As I mentioned before, it could work, but your config is really on
the edge. Anyway, if lower than minimum buffer size is needed
something else is definitely wrong. (Btw, this size can matter less
with high resolution timers.) You could try if my iproute patch:
"tc_core: Use double in tc_core_time2tick()" (not merged) can help
here. While googling for this patch I found this page, which might be
interesting to you (besides the link to the thread with the patch at
the end, take 1 or 2, shouldn't matter):
http://code.google.com/p/pspacer/wiki/HTBon10GbE
If it doesn't help reconsider hfsc.
Thanks,
Jarek P.
^ permalink raw reply
* Re: [Bugme-new] [Bug 19692] New: linux-2.6.36-rc5 crash with gianfar ethernet at full line rate traffic
From: Jarek Poplawski @ 2010-10-16 19:48 UTC (permalink / raw)
To: emin ak
Cc: Andrew Morton, netdev, bugzilla-daemon, bugme-daemon,
Anton Vorontsov
In-Reply-To: <AANLkTinAj=9-RqNA6xKc_5ONz2hPMRGM8uJ2Xy7XJVdr@mail.gmail.com>
On Sat, Oct 16, 2010 at 02:14:01AM +0300, emin ak wrote:
> Hi Jarek.
> Sorry for delayed answer. As I promised, I have started the tests on
> Monday (in spite of dealing with The Monday Syndrome:) I had applied
> your patch and after two billions packet (and approximatly four hours)
> passed kernel crashed with skb_over_panic error similar with first
> type. To ensure the patch is failed or not, I rerun the same test
> again. That time, surprisingly, it did'nt crashed again for two days
> with same kernel. But this situation had occured before, I think
> sometimes because of randomness of the applied ethernet traffic and
> mostly because I cant apply all full line rate random traffic to my
> target device because of wrong test setup (switch / hardware packet
> generator settings etc..), it takes three or more day to crash the
> kernel and sometimes it never crashes. My device only and only crashes
> when I can apply full line rate random traffic. Before informing you
> and the list with (official) test results, I want to be sure with the
> truth of them. So that please let me apply more test to the target
> device for a fews day more. After that I wish I'll came with good
> news!
Hi Emin,
Sorry for being impatient. Actually, waiting is no problem for me. I
simply didn't know how much this bug is reproducible, and was a bit
mislead by your forecast of one day result (and terrified btw seeing
words Monday and immediately side by side ;-) So, a few days or weeks,
no problem, it's all up to you.
On the other hand, it looks like there might be something else/more,
so I'd suggest to add this last skb_over_panic to bugzilla too.
Thanks,
Jarek P.
^ permalink raw reply
* Re: openvswitch/flow WAS ( Re: [rfc] Merging the Open vSwitch datapath
From: Jesse Gross @ 2010-10-16 19:33 UTC (permalink / raw)
To: hadi; +Cc: Ben Pfaff, netdev, ovs-team
In-Reply-To: <1287228959.3664.72.camel@bigi>
On Sat, Oct 16, 2010 at 4:35 AM, jamal <hadi@cyberus.ca> wrote:
> On Fri, 2010-10-15 at 14:35 -0700, Jesse Gross wrote:
>
>>
>> You're right, at a high level, it appears that there is a bit of an
>> overlap between bridging, tc, and Open vSwitch.
>
> It looks like openvswitch rides on top of openflow, correct?
> earlier i was looking at openflow/datapath but gleaning
> openvswitch/datapath it still looks conceptually the same
> at the lower level.
Yes, Open vSwitch supports the OpenFlow protocol. However, the Open
vSwitch kernel portion is completely different from the OpenFlow
reference implementation datapath and in fact does not speak OpenFlow
at the kernel level. You brought up the point of keeping the kernel
simple and making policy decisions in userspace. I completely agree
and, in fact, that is the reason why Open vSwitch is designed the way
it is.
I think it might be helpful if I gave a high level overview of packet
processing:
When a packet is received it, the relevant fields from the packet are
extracted and matched against a hash table. The most interesting part
is actually what happens when the packets don't match a hash entry:
they get sent up to userspace. It is userspace that makes a policy
decision about the traffic and then pushes down a flow entry for
future packets to match. Some of the things that those decisions can
be based on include: OpenFlow rules, wildcarded entries, normal L2
learning, etc. From then on, packets in that flow can be processed on
the fast path in the kernel with minimal overhead, while still getting
the benefit of the knowledge of userspace.
So I think that we are actually in agreement on quite a number of
points: the kernel should be kept as simple as possible, the control
plane should be abstracted out and handled in userspace, and it should
be possible to map the control rules (from OpenFlow or anywhere
really) onto a simpler set of primitives for handling packets.
So with those goals in mind, here's what is needed:
1. Packet field extraction and classification. Realistically speaking
a new, specialized classifier would probably be needed, as you
mention.
2. A mechanism to send/receive packets to/from userspace. This is an
important component that Open vSwitch adds to the pipeline. This will
probably expand in the future to suit different applications, like the
security processing that I talked about.
3. Output actions. A few exist today, at least some new ones will
need to be added.
So in reality, all of major components of Open vSwitch are actually
not present in the kernel today. I know the argument could be made
that certains parts can be replicated in different ways but that's
back to the simplicity point that I was making earlier. The u32
classifier isn't well suited for these types of rules and neither is
pedit. If we're going to add the needed components either way, let's
not make everyone's lives more complicated by mixing everything
together.
^ permalink raw reply
* Re: [PATCH -next] net: move MII outside of NET_ETHERNET, fix kconfig warning
From: David Miller @ 2010-10-16 18:58 UTC (permalink / raw)
To: randy.dunlap; +Cc: linux-kernel, netdev, akpm, jgarzik
In-Reply-To: <20101013181859.fc3a96fd.randy.dunlap@oracle.com>
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Wed, 13 Oct 2010 18:18:59 -0700
> From: Randy Dunlap <randy.dunlap@oracle.com>
>
> We have USB, PCMCIA, and gigabit ethernet drivers that select
> MII even though NET_ETHERNET is not enabled, so make MII not
> be dependent on NET_ETHERNET. It is still dependent on NET
> and NETDEVICES.
>
> Fixes kconfig unmet dependency warning (shortened, was very long string):
...
> Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
> Acked-by: Jeff Garzik <jgarzik@pobox.com> [2006-NOV-30]
Applied, and I added your infiniband patch too.
Thanks Randy.
^ permalink raw reply
* Re: [PATCH net-next 2/2] stmmac: make function tables const
From: David Miller @ 2010-10-16 18:57 UTC (permalink / raw)
To: shemminger; +Cc: peppe.cavallaro, deepak.sikri, netdev
In-Reply-To: <20101013175125.6f6a64db@nehalam>
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Wed, 13 Oct 2010 17:51:25 -0700
> These tables only contain function pointers.
>
> Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Applied.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox