Hi To enable EEH suuport for this pci-express adapter, we added 3 EEH callback functions in the driver and also msi/pcie state needs to be saved and restored for the adapter. Thanks, Wendy Signed-off-by: Wendy Xiong <wendyx@us.ibm.com> diff -Nuarp linux-2.6.25-rc9.orig/drivers/net/bnx2.c linux-2.6.25-rc9.eeh/drivers/net/bnx2.c --- linux-2.6.25-rc9.orig/drivers/net/bnx2.c 2008-04-18 22:12:53.000000000 -0500 +++ linux-2.6.25-rc9.eeh/drivers/net/bnx2.c 2008-04-18 22:28:42.00000000...

[dropped lkml from the Cc] correct, that fix is only for very specific adapters which are based on a totally basically it's inserted into a new motherboard? what was the old motherboard? can you check the BIOS and disable things like "PCI Write combining" or "Writeback" or any option looking similar to that? It appears you hit an issue that is exposed by these adapters on some AMD/NVIDIA chipset-based motherboards. This issue is known and we are investigating this and have been for a l...

yes, that's what the reports are. it appears to be related to a bridge chip which is common on both older and newer TCP Segmentation offload - the hardware will split up the payload into MTU-size fragments itself instead of doing it in the kernel. ethtool -K ethX tso off Auke --

David; I was about to put this in the trivial tree, but I think it should probably be merged via the net tree instead. The change looks sane to me. /Jesper Juhl ---------- Forwarded message ---------- Date: Mon, 03 Mar 2008 22:38:26 +0100 From: Arnd Hannemann <hannemann@nets.rwth-aachen.de> To: trivial@kernel.org Subject: [PATCH] make use of before macro in tcp_input.c Make use of tcp before macro. Signed-off-by: Arnd Hannemann <hannemann@nets.rwth-aachen.de> diff -...

These cli()/sti() calls are made in start_timer() and are therefor redundant since the register_lock is now used to protect register io from within scc_isr() and write_scc() (where all calls to start_timer() originate). Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com> --- drivers/net/hamradio/dmascc.c | 3 --- 1 files changed, 0 insertions(+), 3 deletions(-) diff --git a/drivers/net/hamradio/dmascc.c b/drivers/net/hamradio/dmascc.c index e04bf99..0b94833 100644 --- a/dr...

I'm not sure what the right number is here. Say worst case is header which goes over a page boundary then MAX_SKB_FRAGS in the skb, but for some reason that already has a +2: /* To allow 64K frame to be packed as single skb without frag_list */ #define MAX_SKB_FRAGS (65536/PAGE_SIZE + 2) Unless someone explains, I'll change the xmit sg to 2+MAX_SKB_FRAGS as well. Thanks, Rusty. --

Delete the non-napi code from the driver and Kconfig. Tested x86_64. Apply at next open opportunity. Signed-off-by: Don Fry <pcnet32@verizon.net> --- linux-2.6.25/drivers/net/orig.pcnet32.c 2008-04-16 19:49:44.000000000 -0700 +++ linux-2.6.25/drivers/net/pcnet32.c 2008-04-21 08:24:55.000000000 -0700 @@ -22,12 +22,8 @@ *************************************************************************/ #define DRV_NAME "pcnet32" -#ifdef CONFIG_PCNET32_NAPI -#define DRV_VERSION "1.34-NAPI" -#...

The netns start-stop engine can happily live with any of init or exit callbacks set to NULL. Isn't it too late to merge this one? Signed-off-by: Pavel Emelyanov <xemul@openvz.org> --- diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 8a0fd40..e591e09 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -4338,12 +4338,6 @@ int unregister_inet6addr_notifier(struct notifier_block *nb) EXPORT_SYMBOL(unregister_inet6addr_notifier); - -static int addrconf_net_i...

The seq_open_net() call should be accompanied with seq_release_net() one. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> --- diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index f52f7f8..11b22ab 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -787,7 +787,7 @@ static const struct file_operations xt_table_ops = { .open = xt_table_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release, + .release = seq_release_net, }...

The ip6_route_net_init() performs some unneeded actions. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> --- diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 210a079..7f82e80 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2614,7 +2614,7 @@ struct ctl_table *ipv6_route_sysctl_init(struct net *net) static int ip6_route_net_init(struct net *net) { - int ret = 0; + int ret; ret = -ENOMEM; net->ipv6.ip6_dst_ops = kmemdup(&ip6_dst_ops_template, --

The del_timer() function doesn't guarantee, that the timer callback is not active by the time it exits. Thus, the fib6_net_exit() may kfree() all the data, that is required by the fib6_run_gc(). The race window is tiny, but slab poisoning can trigger this bug. Using del_timer_sync() will cure this. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> --- diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 50f3f8f..1ee4fa1 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib....

Consider an interface with two (or more) IP addresses, connected to a LAN segment with two (or more) networks. In such a case the source IP is not unique, it should be chosen depending on the destionation IP. Under some circumstances this choice is incorrect. Consider the example (an interface in two networks, trying to reach Becasue of the preferred src, the actual source IP is chosen right. Now let's flush all the routes, and then add them manually. (Certainly such a usage is a corner case,...

From: "Satoru SATOH" <satoru.satoh@gmail.com> I've applied your patch, thank you. But I had to do so by hand because your email client corrupted the tab characters into the space by changing them into spaces. Please correct this before you submit future patches. Thank you. --

From: Adrian Bunk <bunk@kernel.org> Also applied, thanks Adrian. BTW, I noticed that some of your patches get posted like with the commit ID newline missing, so that the diff specification shares the same line as the commit ID. Oddly some of your patch postings do not have this missing newline. Any idea what causes this? Just curious :-) --

After the bcm43xx removal ieee80211_wx_{get,set}_auth() were no longer used. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- include/net/ieee80211.h | 8 --- net/ieee80211/ieee80211_wx.c | 89 ----------------------------------- 2 files changed, 97 deletions(-) ec783a7fb0fdacebefd93c7bf40493467a825c70 diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h index facd0ff..b31399e 100644 --- a/include/net/ieee80211.h +++ b/include/net/ieee80211.h @@ -1309,14 +1309,...

The only other two users of ieee80211 are ipw2100 and ipw2200, and they both appear to implement the SIOCxIWAUTH calls privately. So I think these can go to the function slaughterhouse. If somebody wants to unify them later it would be better to start fresh anyway. Dan --

Hmm. It wasn't added as part of softmac so must have been around longer, not sure. James? johannes

skb_{under,over}_panic() can now become static. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- include/linux/skbuff.h | 4 ---- net/core/skbuff.c | 6 ++---- 2 files changed, 2 insertions(+), 8 deletions(-) 7bb1567021c093a87aeabdabf129ba7a4d70779f diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 11fd9f2..7bf66ba 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -385,10 +385,6 @@ extern int skb_cow_data(struct sk_buff *skb, int ...

From: Adrian Bunk <bunk@kernel.org> Patch applied, thanks Adrian. --

__ipv6_isatap_ifid() can become static. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- include/net/addrconf.h | 2 -- net/ipv6/addrconf.c | 3 +-- 2 files changed, 1 insertion(+), 4 deletions(-) \2064125a01db3d6716f93f15309fa98f75b74dd8 diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 0a2f037..8e021c1 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -224,8 +224,6 @@ static inline int ipv6_addr_is_ll_all_routers(const struct in6_addr *a...

Good plan, thanks. johannes

Well, 'git log [$file]' is even more scalable and precise, if committer (as well as author) info and patch flow info is what a tester or bug reporter seeks. But it sounds like you are making an assumption about development We're all for more bug reports and testing, but the link you are trying to draw is more than a little overblown. I'm not convinced LKML's lack of davem pull request visibility is a problem source limiting the growth of Linux today... That's a bit much. :) AFAICT y...

What do_gettimeofday() does is to call getnstimeofday() and to convert the result from timespec{} to timeval{}. We do not always need timeval{} and we can convert timespec{} when we really need (to print). Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org> --- diff --git a/net/dccp/probe.c b/net/dccp/probe.c index 7053bb8..6e1df62 100644 --- a/net/dccp/probe.c +++ b/net/dccp/probe.c @@ -46,29 +46,24 @@ struct { struct kfifo *fifo; spinlock_t lock; wait_queue_head_t wai...

This mirrors a patch which is currently in the test tree and does the update to timespec, apart from a few other changes. It is a good idea, since that in turn mirrors tcp_probe.c Hope Arnado will in some future have time to look at some of the test tree patches, in any case this patch does not harm. The University of Aberdeen is a charity registered in Scotland, No SC013683. --

From: Arnaldo Carvalho de Melo <acme@redhat.com> Thanks for reviewing, patch applied. --