* 3.7.3+: Bad paging request in ip_rcv_finish while running NFS traffic.
From: Ben Greear @ 2013-01-21 21:07 UTC (permalink / raw)
To: netdev, linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
I posted about this a few days ago, but this time the patches applied
are minimal and there are no out-of-tree kernel modules loaded.
I have to have the NFS patches (see below) for this test case to run
at all.
Test case is 2500 macvlans, each with a file-IO process running
on it doing writing on it's own mountpoint. This is using NFSv4,
O_DIRECT, and 10k write() calls (and each file is 10k long, so
open/close file operations as well. Smaller files and write sizes
showed the problem as well.
The bug is most easily hit during my application's restart, where interfaces
are being (re)configured and the file-io processes are being
started. This includes changes to: IP addrs, routes, ip routing rules,
mount/unmount operations, etc.
I suspect it may be some sort of race between tearing down the IP protocol
and having traffic currently being received on that
interface. I have not been able to hit this using HTTP traffic,
but perhaps that is just part of the race and not a particular
problem with NFS.
The patches applied on top of 3.7.3 are some NFS crash fixes Trond posted
last Friday and my own patches I just posted to linux-nfs:
http://www.spinics.net/lists/linux-nfs/msg34811.html
If anyone has any suggestions for how to further debug this,
please let me know.
All of the crashes are in the same place, but the value of
the bad address changes...sometimes appears as if NULL were
dereferenced...
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /home/greearb/kernel/2.6/linux-3.7.x64/vmlinux...done.
(gdb) l *(ip_rcv_finish+0x2b7)
0xffffffff8149c783 is in ip_rcv_finish (/home/greearb/git/linux-3.7.dev.y/net/ipv4/ip_input.c:373).
368 skb->len);
369 } else if (rt->rt_type == RTN_BROADCAST)
370 IP_UPD_PO_STATS_BH(dev_net(rt->dst.dev), IPSTATS_MIB_INBCAST,
371 skb->len);
372
373 return dst_input(skb);
374
375 drop:
376 kfree_skb(skb);
377 return NET_RX_DROP;
(gdb) quit
IPv6: ADDRCONF(NETDEV_CHANGE): eth2#903: link becomes ready
IPv6: ADDRCONF(NETDEV_CHANGE): eth2#923: link becomes ready
IPv6: ADDRCONF(NETDEV_CHANGE): eth2#943: link becomes ready
IPv6: ADDRCONF(NETDEV_CHANGE): eth2#963: link becomes ready
IPv6: ADDRCONF(NETDEV_CHANGE): eth2#983: link becomes ready
kernel tried to execute NX-protected page - exploit attempt? (uid: 0)
BUG: unable to handle kernel paging request at ffff88040d7d8000
IP: [<ffff88040d7d8000>] 0xffff88040d7d7fff
PGD 1a0c063 PUD df78e067 PMD 800000040d6001e3
Oops: 0011 [#1] PREEMPT SMP
Modules linked in: macvlan pktgen lockd sunrpc uinput coretemp hwmon kvm_intel kvm gpio_ich iTCO_wdt iTCO_vendor_support microcode pcspkr lpc_ich i2c_i801
e1000e ioatdma igb i7core_edac edac_core ptp pps_core dca ipv6 mgag200 i2c_algo_bit drm_kms_helper ttm drm i2c_core
CPU 6
Pid: 47, comm: ksoftirqd/6 Tainted: G C 3.7.3+ #37 Iron Systems Inc. EE2610R/X8ST3
RIP: 0010:[<ffff88040d7d8000>] [<ffff88040d7d8000>] 0xffff88040d7d7fff
RSP: 0018:ffff88040d75bc00 EFLAGS: 00010282
RAX: ffff88040435cd80 RBX: ffff8803dd83e300 RCX: ffff8803dd83e300
RDX: 0000000000000000 RSI: 0000000000000002 RDI: ffff8803dd83e300
RBP: ffff88040d75bc28 R08: ffffffff8149c4cc R09: ffff88040d75bbf0
R10: dead000000200200 R11: dead000000100100 R12: ffff8803e10c88fc
R13: ffff8803dd83e300 R14: ffff88040d38e000 R15: 0000000000000000
FS: 0000000000000000(0000) GS:ffff88041fcc0000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: ffff88040d7d8000 CR3: 0000000001a0b000 CR4: 00000000000007e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process ksoftirqd/6 (pid: 47, threadinfo ffff88040d75a000, task ffff88040d751730)
Stack:
ffffffff8149c783 ffff8803dd83e300 ffffffff8149c4cc ffff8803dd83e300
ffff88040d38e000 ffff88040d75bc58 ffffffff8149cae8 0000000080000000
0000000000000001 ffff8803dd83e300 ffff88040d38e000 ffff88040d75bc88
Call Trace:
[<ffffffff8149c783>] ? ip_rcv_finish+0x2b7/0x2cf
[<ffffffff8149c4cc>] ? inet_del_protocol+0x37/0x37
[<ffffffff8149cae8>] NF_HOOK.clone.1+0x4c/0x53
[<ffffffff8149cd73>] ip_rcv+0x237/0x268
[<ffffffff81468c46>] __netif_receive_skb+0x487/0x530
[<ffffffff81468de5>] process_backlog+0xf6/0x1d7
[<ffffffff8146b19b>] net_rx_action+0xad/0x20c
[<ffffffff8108d292>] __do_softirq+0x9c/0x161
[<ffffffff8108d37a>] run_ksoftirqd+0x23/0x42
[<ffffffff810a6f3b>] smpboot_thread_fn+0x253/0x258
[<ffffffff810a6ce8>] ? test_ti_thread_flag.clone.0+0x11/0x11
[<ffffffff8109ff60>] kthread+0xbf/0xc7
[<ffffffff8109fea1>] ? __init_kthread_worker+0x37/0x37
[<ffffffff815292bc>] ret_from_fork+0x7c/0xb0
[<ffffffff8109fea1>] ? __init_kthread_worker+0x37/0x37
Code: 36 a4 36 6a ea 6a ea 6a ea 6a ea e9 81 e9 81 e9 81 e9 81 71 48 71 48 71 48 71 48 ea 50 ea 50 ea 50 ea 50 f7 e4 f7 e4 f7 e4 f7 e4 <22> 22 00 00 ad 4e ad de
ff ff ff ff 00 00 00 00 ff ff ff ff ff
RIP [<ffff88040d7d8000>] 0xffff88040d7d7fff
RSP <ffff88040d75bc00>
CR2: ffff88040d7d8000
---[ end trace 1c131533c1de5e8b ]---
Thanks,
Ben
--
Ben Greear <greearb-my8/4N5VtI7c+919tysfdA@public.gmane.org>
Candela Technologies Inc http://www.candelatech.com
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH 14/21] NTB: Fix Sparse Warnings
From: Jon Mason @ 2013-01-21 21:13 UTC (permalink / raw)
To: Greg KH; +Cc: linux-kernel, netdev, Dave Jiang, Nicholas Bellinger
In-Reply-To: <20130120234534.GD14196@kroah.com>
On Sun, Jan 20, 2013 at 03:45:34PM -0800, Greg KH wrote:
> On Sat, Jan 19, 2013 at 02:02:28AM -0700, Jon Mason wrote:
> > Address the sparse warnings and resulting fallout
> >
> > Signed-off-by: Jon Mason <jon.mason@intel.com>
> > ---
> > drivers/ntb/ntb_hw.c | 7 ++++---
> > drivers/ntb/ntb_hw.h | 4 ++--
> > drivers/ntb/ntb_transport.c | 32 ++++++++++++++++----------------
> > 3 files changed, 22 insertions(+), 21 deletions(-)
> >
> > diff --git a/drivers/ntb/ntb_hw.c b/drivers/ntb/ntb_hw.c
> > index 18cb5dc..b792ccd 100644
> > --- a/drivers/ntb/ntb_hw.c
> > +++ b/drivers/ntb/ntb_hw.c
> > @@ -104,8 +104,9 @@ MODULE_DEVICE_TABLE(pci, ntb_pci_tbl);
> > *
> > * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
> > */
> > -int ntb_register_event_callback(struct ntb_device *ndev,
> > - void (*func)(void *handle, unsigned int event))
> > +int
> > +ntb_register_event_callback(struct ntb_device *ndev,
> > + void (*func)(void *handle, enum ntb_hw_event event))
>
> What is sparse complaining about here that moving the function name to
> the start of the line fixes? That shouldn't be necessary at all. Ah,
> the enum? If so, why change the first line at all?
All an effort to get it to fit in under 80 chars.
>
> > {
> > if (ndev->event_cb)
> > return -EINVAL;
> > @@ -344,7 +345,7 @@ int ntb_read_remote_spad(struct ntb_device *ndev, unsigned int idx, u32 *val)
> > *
> > * RETURNS: pointer to virtual address, or NULL on error.
> > */
> > -void *ntb_get_mw_vbase(struct ntb_device *ndev, unsigned int mw)
> > +void __iomem *ntb_get_mw_vbase(struct ntb_device *ndev, unsigned int mw)
>
> That makes sense to fix.
>
> > {
> > if (mw > NTB_NUM_MW)
> > return NULL;
> > diff --git a/drivers/ntb/ntb_hw.h b/drivers/ntb/ntb_hw.h
> > index 5e00951..3a3038c 100644
> > --- a/drivers/ntb/ntb_hw.h
> > +++ b/drivers/ntb/ntb_hw.h
> > @@ -165,14 +165,14 @@ int ntb_register_db_callback(struct ntb_device *ndev, unsigned int idx,
> > void ntb_unregister_db_callback(struct ntb_device *ndev, unsigned int idx);
> > int ntb_register_event_callback(struct ntb_device *ndev,
> > void (*event_cb_func) (void *handle,
> > - unsigned int event));
> > + enum ntb_hw_event event));
> > void ntb_unregister_event_callback(struct ntb_device *ndev);
> > int ntb_get_max_spads(struct ntb_device *ndev);
> > int ntb_write_local_spad(struct ntb_device *ndev, unsigned int idx, u32 val);
> > int ntb_read_local_spad(struct ntb_device *ndev, unsigned int idx, u32 *val);
> > int ntb_write_remote_spad(struct ntb_device *ndev, unsigned int idx, u32 val);
> > int ntb_read_remote_spad(struct ntb_device *ndev, unsigned int idx, u32 *val);
> > -void *ntb_get_mw_vbase(struct ntb_device *ndev, unsigned int mw);
> > +void __iomem *ntb_get_mw_vbase(struct ntb_device *ndev, unsigned int mw);
> > resource_size_t ntb_get_mw_size(struct ntb_device *ndev, unsigned int mw);
> > void ntb_ring_sdb(struct ntb_device *ndev, unsigned int idx);
> > void *ntb_find_transport(struct pci_dev *pdev);
> > diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c
> > index b3afb24..e0bdfd7 100644
> > --- a/drivers/ntb/ntb_transport.c
> > +++ b/drivers/ntb/ntb_transport.c
> > @@ -58,7 +58,7 @@
> > #include <linux/ntb.h>
> > #include "ntb_hw.h"
> >
> > -#define NTB_TRANSPORT_VERSION 1
> > +#define NTB_TRANSPORT_VERSION 2
>
> How is this a sparse fix?
I changed the format of the header below.
>
> > static unsigned int transport_mtu = 0x401E;
> > module_param(transport_mtu, uint, 0644);
> > @@ -91,14 +91,14 @@ struct ntb_transport_qp {
> > bool qp_link;
> > u8 qp_num; /* Only 64 QP's are allowed. 0-63 */
> >
> > - struct ntb_rx_info *rx_info;
> > + struct ntb_rx_info __iomem *rx_info;
> > struct ntb_rx_info *remote_rx_info;
> >
> > void (*tx_handler) (struct ntb_transport_qp *qp, void *qp_data,
> > void *data, int len);
> > struct list_head tx_free_q;
> > spinlock_t ntb_tx_free_q_lock;
> > - void *tx_mw;
> > + void __iomem *tx_mw;
> > unsigned int tx_index;
> > unsigned int tx_max_entry;
> > unsigned int tx_max_frame;
> > @@ -166,7 +166,7 @@ enum {
> > };
> >
> > struct ntb_payload_header {
> > - u64 ver;
> > + unsigned int ver;
>
> why does sparse care about this?
I am using the number of tx/rx packets transmitted as a sanity check
on the data. This is a 64bit counter. There is no iowrite64, so I
truncated this to 32bits. I can rework this part of the patch to do 2
iowrite32's and keep the value as 64bit.
Thanks,
Jon
>
> thanks,
>
> greg k-h
^ permalink raw reply
* Re: [PATCH 1/2] ipv4: Invalidate the socket cached route on pmtu events if possible
From: Julian Anastasov @ 2013-01-21 21:18 UTC (permalink / raw)
To: Steffen Klassert; +Cc: David Miller, Yurij M. Plotnikov, netdev
In-Reply-To: <20130121115911.GE30530@secunet.com>
Hello,
On Mon, 21 Jan 2013, Steffen Klassert wrote:
> The route lookup in ipv4_sk_update_pmtu() might return a route
> different from the route we cached at the socket. This is because
> standart routes are per cpu, so each cpu has it's own struct rtable.
> This means that we do not invalidate the socket cached route if the
> NET_RX_SOFTIRQ is not served by the same cpu that the sending socket
> uses. As a result, the cached route reused until we disconnect.
>
> With this patch we invalidate the socket cached route if possible.
> If the socket is owened by the user, we can't update the cached
> route directly. A followup patch will implement socket release
> callback functions for datagram sockets to handle this case.
>
> Reported-by: Yurij M. Plotnikov <Yurij.Plotnikov@oktetlabs.ru>
> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
> ---
> net/ipv4/route.c | 42 +++++++++++++++++++++++++++++++++++++++++-
> 1 file changed, 41 insertions(+), 1 deletion(-)
>
> diff --git a/net/ipv4/route.c b/net/ipv4/route.c
> index 259cbee..132737a 100644
> --- a/net/ipv4/route.c
> +++ b/net/ipv4/route.c
> @@ -965,7 +965,7 @@ void ipv4_update_pmtu(struct sk_buff *skb, struct net *net, u32 mtu,
> }
> EXPORT_SYMBOL_GPL(ipv4_update_pmtu);
>
> -void ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu)
> +static void __ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu)
> {
> const struct iphdr *iph = (const struct iphdr *) skb->data;
> struct flowi4 fl4;
> @@ -978,6 +978,46 @@ void ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu)
> ip_rt_put(rt);
> }
> }
> +
> +void ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu)
> +{
> + const struct iphdr *iph = (const struct iphdr *) skb->data;
> + struct flowi4 fl4;
> + struct rtable *rt;
> + struct dst_entry *dst;
> +
> + bh_lock_sock(sk);
> + rt = (struct rtable *) __sk_dst_get(sk);
I just saw another problem, sorry that
I missed it the first time. Here __sk_dst_get
does not get reference...
> +
> + if (sock_owned_by_user(sk) || !rt) {
> + __ipv4_sk_update_pmtu(skb, sk, mtu);
> + goto out;
> + }
> +
> + __build_flow_key(&fl4, sk, iph, 0, 0, 0, 0, 0);
> +
> + if (!__sk_dst_check(sk, 0)) {
> + rt = ip_route_output_flow(sock_net(sk), &fl4, sk);
> + if (IS_ERR(rt))
> + goto out;
but here rt->dst comes with reference.
> + }
May be here we can use 'else dst_hold(&rt->dst);' ?
It is needed for __sk_dst_set.
> +
> + __ip_rt_update_pmtu((struct rtable *) rt->dst.path, &fl4, mtu);
> +
> + dst = dst_check(&rt->dst, 0);
> + if (!dst) {
dst_release(&rt->dst);
> + rt = ip_route_output_flow(sock_net(sk), &fl4, sk);
> + if (IS_ERR(rt))
> + goto out;
> +
> + dst = &rt->dst;
Remove above line...
> + }
> +
and use here __sk_dst_set(sk, &rt->dst) instead:
> + __sk_dst_set(sk, dst);
Another variant is to remember with flag 'new_rt'
that we should call __sk_dst_set, eg. when rt comes from
ip_route_output_flow. By this way we can avoid some of
the dst_hold/dst_release calls if sk_dst_cache is not
changed. IIRC, according to sk_dst_check, dst_check can
not return different dst from the ->check method.
> +
> +out:
> + bh_unlock_sock(sk);
> +}
> EXPORT_SYMBOL_GPL(ipv4_sk_update_pmtu);
>
> void ipv4_redirect(struct sk_buff *skb, struct net *net,
> --
> 1.7.9.5
Regards
--
Julian Anastasov <ja@ssi.bg>
^ permalink raw reply
* Re: [RFC:] struct net_device_ops: Add function pointer to fill device specific ndisc information
From: Stephan Gatzka @ 2013-01-21 21:16 UTC (permalink / raw)
To: YOSHIFUJI Hideaki; +Cc: Stefan Richter, netdev, linux1394-devel, David Miller
In-Reply-To: <50FD9DB7.60802@linux-ipv6.org>
>> Full ack. And that's the reason why I feel very uncomfortable with a Yoshifujis hardware address extensions by fifo_addr, spd, and max_rec.
>>
>> This seems possible with a single netdevice for IPv4/6 only if we _always_ use the same fifo address for both IPv4 and IPv6. Do we all agree on that?
>
> I do not understand what "that" and "this" mean here.
> Do you want to have different FIFO on single net_device? If yes, for what?
>
Oh, I'm sorry for being mixed-up. I meant that your proposal for
enhancing the hardware address seems to be a feasible solution, if we
_never_ want to have distinct fifo addresses for IPv4 and IPv6. This
will be just impossible with a single netdevice.
But I have to admit that I have no good use case for different fifo
addresses...
So overall, your proposal seems to be a possible solution. Nevertheless,
I can't help that I find it a bit awkward to have speed, max_rec and
fifo address in the hardware address of the network device. For me,
that's just the GUID.
I want to have an ack from Stefan Richter before I'm going to implement
your solution.
Regards,
Stephan
^ permalink raw reply
* Re: [RFC] State resolution packet queue for IPsec
From: David Miller @ 2013-01-21 21:17 UTC (permalink / raw)
To: steffen.klassert; +Cc: herbert, netdev
In-Reply-To: <20130121094847.GC30530@secunet.com>
From: Steffen Klassert <steffen.klassert@secunet.com>
Date: Mon, 21 Jan 2013 10:48:48 +0100
> I did a implementation of a state resolution packet queue for IPsec.
> The original idea is described in git commit 14e50e57a
> ([XFRM]: Allow packet drops during larval state resolution.)
>
> Since we don't have any notifiers for xfrm state resolution, I used
> a timer that does a relookup in given intervals. This is mainly to speed
> up the start of tcp connections, so I tried to do the relookup in a higher
> rate that tcp would resend the syn packet. However, the relookup intervals,
> the queue size and the timeout value are choosen by gut instincts, so
> could be suboptimal. If anyone has some insight on good choices of these
> values, please let me know. All other comment are welcome too, of course.
This is probably the best way to solve this problem without adding
xfrm_state resolution notifications.
But really why would notifications be so bad even if they would not
be fine grained?
Any time a state is resolved, any state, you run what you have put
currently into this new timer function.
Most of the time, defer list is empty, so no work is done. Otherwise
quick notification is essential and worth running the list.
Then there are no time based heuristics to come up with at all.
^ permalink raw reply
* Re: [PATCH 14/21] NTB: Fix Sparse Warnings
From: Greg KH @ 2013-01-21 21:37 UTC (permalink / raw)
To: Jon Mason; +Cc: linux-kernel, netdev, Dave Jiang, Nicholas Bellinger
In-Reply-To: <20130121211350.GH22208@jonmason-lab>
On Mon, Jan 21, 2013 at 02:13:50PM -0700, Jon Mason wrote:
> On Sun, Jan 20, 2013 at 03:45:34PM -0800, Greg KH wrote:
> > On Sat, Jan 19, 2013 at 02:02:28AM -0700, Jon Mason wrote:
> > > Address the sparse warnings and resulting fallout
> > >
> > > Signed-off-by: Jon Mason <jon.mason@intel.com>
> > > ---
> > > drivers/ntb/ntb_hw.c | 7 ++++---
> > > drivers/ntb/ntb_hw.h | 4 ++--
> > > drivers/ntb/ntb_transport.c | 32 ++++++++++++++++----------------
> > > 3 files changed, 22 insertions(+), 21 deletions(-)
> > >
> > > diff --git a/drivers/ntb/ntb_hw.c b/drivers/ntb/ntb_hw.c
> > > index 18cb5dc..b792ccd 100644
> > > --- a/drivers/ntb/ntb_hw.c
> > > +++ b/drivers/ntb/ntb_hw.c
> > > @@ -104,8 +104,9 @@ MODULE_DEVICE_TABLE(pci, ntb_pci_tbl);
> > > *
> > > * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
> > > */
> > > -int ntb_register_event_callback(struct ntb_device *ndev,
> > > - void (*func)(void *handle, unsigned int event))
> > > +int
> > > +ntb_register_event_callback(struct ntb_device *ndev,
> > > + void (*func)(void *handle, enum ntb_hw_event event))
> >
> > What is sparse complaining about here that moving the function name to
> > the start of the line fixes? That shouldn't be necessary at all. Ah,
> > the enum? If so, why change the first line at all?
>
> All an effort to get it to fit in under 80 chars.
The original lines were less than 80 characters :)
> > > --- a/drivers/ntb/ntb_transport.c
> > > +++ b/drivers/ntb/ntb_transport.c
> > > @@ -58,7 +58,7 @@
> > > #include <linux/ntb.h>
> > > #include "ntb_hw.h"
> > >
> > > -#define NTB_TRANSPORT_VERSION 1
> > > +#define NTB_TRANSPORT_VERSION 2
> >
> > How is this a sparse fix?
>
> I changed the format of the header below.
Ah, ok.
Care to resend this without the first chunk?
thanks,
greg k-h
^ permalink raw reply
* Re: [PATCH v3] net, wireless: overwrite default_ethtool_ops
From: Luis R. Rodriguez @ 2013-01-21 21:47 UTC (permalink / raw)
To: David Miller
Cc: mcgrof-3uybbJdB1yH774rrrx3eTA, sgruszka-H+wXaHxf7aLQT0dZR+AlfA,
bhutchings-s/n/eUQHGBpZroRs9YW3xA, netdev-u79uwXL29TY76Z2rM5mHXA,
edumazet-hpIqsD4AKlfQT0dZR+AlfA, greearb-my8/4N5VtI7c+919tysfdA,
bjorn-yOkvZcmFvRU, linux-wireless-u79uwXL29TY76Z2rM5mHXA,
mirqus-Re5JQEeQqe8AvxtiuMwx3w, johannes-cdvu00un1VgdHxzADdlk8Q
In-Reply-To: <20130121.160404.1007504714461228602.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
On Mon, Jan 21, 2013 at 04:04:04PM -0500, David Miller wrote:
>
> It's queued up for -stable already as is clearly seen at:
>
> http://patchwork.ozlabs.org/user/bundle/2566/?state=*
Thanks, I was not aware of this bundle. In this case assuming
this goes into v3.7.5 this is being backported with special handling
between 3.7.0 and 3.7.5 as follows onto compat.
From: "Luis R. Rodriguez" <mcgrof-3uybbJdB1yH774rrrx3eTA@public.gmane.org>
Subject: [PATCH] compat: backport netdev_set_default_ethtool_ops()
Stanislaw found that due to commit 2c60db03 by Eric Dumazet
the wireless core was not assigning driver specific ethtool_ops.
This was fixed by Stanislaw's commit d07d7507 which added
netdev_set_default_ethtool_ops(). Since Eric's commit 2c60db03
is on v3.7-rc1 Stanislaw's fix is required down to v3.7 as well.
The d07d7507 commit is currently present on v3.8-rc4 and is on
its way to what we think may be v3.7.5. Because of this kernels
older than v3.7.5 will require the full implementation while
older kernels than v3.7.0 will require just assigning the ops
passed only if netdev has no ops already set just as we used
to have it implemented on cfg80211.
mcgrof@frijol ~/linux-stable (git::linux-3.8.y)$ git describe --contains 2c60db
v3.7-rc1~145^2~142
mcgrof@frijol ~/linux-stable (git::linux-3.8.y)$ git describe --contains d07d75
v3.8-rc4~29^2~4
ckmake results:
1 2.6.24 [ OK ]
2 2.6.25 [ OK ]
3 2.6.26 [ OK ]
4 2.6.27 [ OK ]
5 2.6.28 [ OK ]
6 2.6.29 [ OK ]
7 2.6.30 [ OK ]
8 2.6.31 [ OK ]
9 2.6.32 [ OK ]
10 2.6.33 [ OK ]
11 2.6.34 [ OK ]
12 2.6.35 [ OK ]
13 2.6.36 [ OK ]
14 2.6.37 [ OK ]
15 2.6.38 [ OK ]
16 2.6.39 [ OK ]
17 3.0.50 [ OK ]
18 3.1.10 [ OK ]
19 3.2.33 [ OK ]
20 3.3.8 [ OK ]
21 3.4.17 [ OK ]
22 3.5.7 [ OK ]
23 3.6.5 [ OK ]
24 3.7.0 [ OK ]
real 0m34.791s
user 11m38.572s
sys 3m56.927s
Signed-off-by: Luis R. Rodriguez <mcgrof-3uybbJdB1yH774rrrx3eTA@public.gmane.org>
---
compat/compat-3.8.c | 22 ++++++++++++++++++++++
include/linux/compat-3.8.h | 6 ++++++
2 files changed, 28 insertions(+)
diff --git a/compat/compat-3.8.c b/compat/compat-3.8.c
index 034dd77..1867258 100644
--- a/compat/compat-3.8.c
+++ b/compat/compat-3.8.c
@@ -16,6 +16,28 @@
#include <linux/hid.h>
#include <linux/module.h>
#include "hid-ids.h"
+#include <linux/netdevice.h>
+
+#if (LINUX_VERSION_CODE < KERNEL_VERSION(3,7,5))
+
+#if (LINUX_VERSION_CODE < KERNEL_VERSION(3,7,0))
+void netdev_set_default_ethtool_ops(struct net_device *dev,
+ const struct ethtool_ops *ops)
+{
+ if (!dev->ethtool_ops)
+ dev->ethtool_ops = ops;
+}
+#else /* kernel is between 3.7.0 and 3.7.4 */;
+void netdev_set_default_ethtool_ops(struct net_device *dev,
+ const struct ethtool_ops *ops)
+{
+ if (dev->ethtool_ops == &default_ethtool_ops)
+ dev->ethtool_ops = ops;
+}
+#endif
+
+EXPORT_SYMBOL_GPL(netdev_set_default_ethtool_ops);
+#endif /* (LINUX_VERSION_CODE < KERNEL_VERSION(3,7,5) */
/* a list of devices that shouldn't be handled by HID core at all */
static const struct hid_device_id hid_ignore_list[] = {
diff --git a/include/linux/compat-3.8.h b/include/linux/compat-3.8.h
index 052de95..942b4cb 100644
--- a/include/linux/compat-3.8.h
+++ b/include/linux/compat-3.8.h
@@ -6,6 +6,12 @@
#if (LINUX_VERSION_CODE < KERNEL_VERSION(3,8,0))
#include <linux/hid.h>
+#include <linux/netdevice.h>
+
+#if (LINUX_VERSION_CODE < KERNEL_VERSION(3,7,5))
+extern void netdev_set_default_ethtool_ops(struct net_device *dev,
+ const struct ethtool_ops *ops);
+#endif
#define HID_BUS_ANY 0xffff
#define HID_GROUP_ANY 0x0000
--
1.7.10.4
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH 0/6] ISDN patches for net-next
From: Tilman Schmidt @ 2013-01-21 21:57 UTC (permalink / raw)
To: Karsten Keil, David Miller
Cc: Hansjoerg Lipp, Karsten Keil, i4ldeveloper, netdev, linux-kernel
Karsten, David,
here's a series of six patches to the ISDN subsystem. Please consider
for application to net-next. The last one should also go into -stable
and is tagged "CC: stable" accordingly.
Thanks,
Tilman
Tilman Schmidt (6):
isdn/divert: fix readability damage
isdn/gigaset: leave DLE mode before hanging up
isdn/gigaset: beautify interface.c
isdn/gigaset: beautify common.c
isdn/gigaset: beautify ev-layer.c
isdn/gigaset: fix zero size border case in debug dump
drivers/isdn/divert/divert_init.c | 33 +-
drivers/isdn/divert/isdn_divert.c | 423 ++++++++++++++++++--------------------
drivers/isdn/divert/isdn_divert.h | 28 +-
drivers/isdn/gigaset/capi.c | 2
drivers/isdn/gigaset/common.c | 31 --
drivers/isdn/gigaset/ev-layer.c | 124 +++++------
drivers/isdn/gigaset/gigaset.h | 9
drivers/isdn/gigaset/interface.c | 46 +---
8 files changed, 332 insertions(+), 364 deletions(-)
^ permalink raw reply
* [PATCH 2/6] isdn/gigaset: leave DLE mode before hanging up
From: Tilman Schmidt @ 2013-01-21 21:57 UTC (permalink / raw)
To: Karsten Keil, David Miller
Cc: Hansjoerg Lipp, Karsten Keil, i4ldeveloper, netdev, linux-kernel
In-Reply-To: <20130121-patch-isdn-00.tilman@imap.cc>
Some firmware releases of Gigaset M105 do not accept AT+VLS=0 command
in DLE mode, so always leave DLE mode before sending the command.
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
---
drivers/isdn/gigaset/ev-layer.c | 5 +++++
1 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/drivers/isdn/gigaset/ev-layer.c b/drivers/isdn/gigaset/ev-layer.c
index 2e6963d..e2383ec 100644
--- a/drivers/isdn/gigaset/ev-layer.c
+++ b/drivers/isdn/gigaset/ev-layer.c
@@ -1693,6 +1693,11 @@ static void process_command_flags(struct cardstate *cs)
for (i = 0; i < cs->channels; ++i) {
bcs = cs->bcs + i;
if (bcs->at_state.pending_commands & PC_HUP) {
+ if (cs->dle) {
+ cs->curchannel = bcs->channel;
+ schedule_sequence(cs, &cs->at_state, SEQ_DLE0);
+ return;
+ }
bcs->at_state.pending_commands &= ~PC_HUP;
if (bcs->at_state.pending_commands & PC_CID) {
/* not yet dialing: PC_NOCID is sufficient */
--
1.7.3.4
^ permalink raw reply related
* [PATCH 3/6] isdn/gigaset: beautify interface.c
From: Tilman Schmidt @ 2013-01-21 21:57 UTC (permalink / raw)
To: Karsten Keil, David Miller
Cc: Hansjoerg Lipp, Karsten Keil, i4ldeveloper, netdev, linux-kernel
In-Reply-To: <20130121-patch-isdn-00.tilman@imap.cc>
Avoid forward declarations and remove a needless initialization.
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
---
drivers/isdn/gigaset/interface.c | 46 ++++++++++++-------------------------
1 files changed, 15 insertions(+), 31 deletions(-)
diff --git a/drivers/isdn/gigaset/interface.c b/drivers/isdn/gigaset/interface.c
index 67abf3f..20b7e7a 100644
--- a/drivers/isdn/gigaset/interface.c
+++ b/drivers/isdn/gigaset/interface.c
@@ -112,36 +112,6 @@ static int if_config(struct cardstate *cs, int *arg)
}
/*** the terminal driver ***/
-/* stolen from usbserial and some other tty drivers */
-
-static int if_open(struct tty_struct *tty, struct file *filp);
-static void if_close(struct tty_struct *tty, struct file *filp);
-static int if_ioctl(struct tty_struct *tty,
- unsigned int cmd, unsigned long arg);
-static int if_write_room(struct tty_struct *tty);
-static int if_chars_in_buffer(struct tty_struct *tty);
-static void if_throttle(struct tty_struct *tty);
-static void if_unthrottle(struct tty_struct *tty);
-static void if_set_termios(struct tty_struct *tty, struct ktermios *old);
-static int if_tiocmget(struct tty_struct *tty);
-static int if_tiocmset(struct tty_struct *tty,
- unsigned int set, unsigned int clear);
-static int if_write(struct tty_struct *tty,
- const unsigned char *buf, int count);
-
-static const struct tty_operations if_ops = {
- .open = if_open,
- .close = if_close,
- .ioctl = if_ioctl,
- .write = if_write,
- .write_room = if_write_room,
- .chars_in_buffer = if_chars_in_buffer,
- .set_termios = if_set_termios,
- .throttle = if_throttle,
- .unthrottle = if_unthrottle,
- .tiocmget = if_tiocmget,
- .tiocmset = if_tiocmset,
-};
static int if_open(struct tty_struct *tty, struct file *filp)
{
@@ -355,7 +325,7 @@ done:
static int if_write_room(struct tty_struct *tty)
{
struct cardstate *cs = tty->driver_data;
- int retval = -ENODEV;
+ int retval;
gig_dbg(DEBUG_IF, "%u: %s()", cs->minor_index, __func__);
@@ -498,6 +468,20 @@ out:
mutex_unlock(&cs->mutex);
}
+static const struct tty_operations if_ops = {
+ .open = if_open,
+ .close = if_close,
+ .ioctl = if_ioctl,
+ .write = if_write,
+ .write_room = if_write_room,
+ .chars_in_buffer = if_chars_in_buffer,
+ .set_termios = if_set_termios,
+ .throttle = if_throttle,
+ .unthrottle = if_unthrottle,
+ .tiocmget = if_tiocmget,
+ .tiocmset = if_tiocmset,
+};
+
/* wakeup tasklet for the write operation */
static void if_wake(unsigned long data)
--
1.7.3.4
^ permalink raw reply related
* [PATCH 4/6] isdn/gigaset: beautify common.c
From: Tilman Schmidt @ 2013-01-21 21:57 UTC (permalink / raw)
To: Karsten Keil, David Miller
Cc: Hansjoerg Lipp, Karsten Keil, i4ldeveloper, netdev, linux-kernel
In-Reply-To: <20130121-patch-isdn-00.tilman@imap.cc>
Rearrange the gigaset_freecs() function to make it more readable,
and adapt gigaset_initcs() accordingly.
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
---
drivers/isdn/gigaset/common.c | 31 +++++++++----------------------
1 files changed, 9 insertions(+), 22 deletions(-)
diff --git a/drivers/isdn/gigaset/common.c b/drivers/isdn/gigaset/common.c
index 6849a11..7c78144 100644
--- a/drivers/isdn/gigaset/common.c
+++ b/drivers/isdn/gigaset/common.c
@@ -467,11 +467,6 @@ void gigaset_freecs(struct cardstate *cs)
mutex_lock(&cs->mutex);
- if (!cs->bcs)
- goto f_cs;
- if (!cs->inbuf)
- goto f_bcs;
-
spin_lock_irqsave(&cs->lock, flags);
cs->running = 0;
spin_unlock_irqrestore(&cs->lock, flags); /* event handler and timer are
@@ -507,17 +502,16 @@ void gigaset_freecs(struct cardstate *cs)
gig_dbg(DEBUG_INIT, "clearing at_state");
clear_at_state(&cs->at_state);
dealloc_temp_at_states(cs);
+ clear_events(cs);
tty_port_destroy(&cs->port);
/* fall through */
case 0: /* error in basic setup */
- clear_events(cs);
gig_dbg(DEBUG_INIT, "freeing inbuf");
kfree(cs->inbuf);
+ kfree(cs->bcs);
}
-f_bcs: gig_dbg(DEBUG_INIT, "freeing bcs[]");
- kfree(cs->bcs);
-f_cs: gig_dbg(DEBUG_INIT, "freeing cs");
+
mutex_unlock(&cs->mutex);
free_cs(cs);
}
@@ -687,19 +681,6 @@ struct cardstate *gigaset_initcs(struct gigaset_driver *drv, int channels,
return NULL;
}
- gig_dbg(DEBUG_INIT, "allocating bcs[0..%d]", channels - 1);
- cs->bcs = kmalloc(channels * sizeof(struct bc_state), GFP_KERNEL);
- if (!cs->bcs) {
- pr_err("out of memory\n");
- goto error;
- }
- gig_dbg(DEBUG_INIT, "allocating inbuf");
- cs->inbuf = kmalloc(sizeof(struct inbuf_t), GFP_KERNEL);
- if (!cs->inbuf) {
- pr_err("out of memory\n");
- goto error;
- }
-
cs->cs_init = 0;
cs->channels = channels;
cs->onechannel = onechannel;
@@ -729,6 +710,12 @@ struct cardstate *gigaset_initcs(struct gigaset_driver *drv, int channels,
cs->mode = M_UNKNOWN;
cs->mstate = MS_UNINITIALIZED;
+ cs->bcs = kmalloc(channels * sizeof(struct bc_state), GFP_KERNEL);
+ cs->inbuf = kmalloc(sizeof(struct inbuf_t), GFP_KERNEL);
+ if (!cs->bcs || !cs->inbuf) {
+ pr_err("out of memory\n");
+ goto error;
+ }
++cs->cs_init;
gig_dbg(DEBUG_INIT, "setting up at_state");
--
1.7.3.4
^ permalink raw reply related
* [PATCH 6/6] isdn/gigaset: fix zero size border case in debug dump
From: Tilman Schmidt @ 2013-01-21 21:57 UTC (permalink / raw)
To: Karsten Keil, David Miller
Cc: Hansjoerg Lipp, Karsten Keil, i4ldeveloper, netdev, linux-kernel
In-Reply-To: <20130121-patch-isdn-00.tilman@imap.cc>
If subtracting 12 from l leaves zero we'd do a zero size allocation,
leading to an oops later when we try to set the NUL terminator.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
CC: stable <stable@kernel.org>
---
drivers/isdn/gigaset/capi.c | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/drivers/isdn/gigaset/capi.c b/drivers/isdn/gigaset/capi.c
index 68452b7..03a0a01 100644
--- a/drivers/isdn/gigaset/capi.c
+++ b/drivers/isdn/gigaset/capi.c
@@ -248,6 +248,8 @@ static inline void dump_rawmsg(enum debuglevel level, const char *tag,
CAPIMSG_APPID(data), CAPIMSG_MSGID(data), l,
CAPIMSG_CONTROL(data));
l -= 12;
+ if (l <= 0)
+ return;
dbgline = kmalloc(3 * l, GFP_ATOMIC);
if (!dbgline)
return;
--
1.7.3.4
^ permalink raw reply related
* [PATCH 5/6] isdn/gigaset: beautify ev-layer.c
From: Tilman Schmidt @ 2013-01-21 21:57 UTC (permalink / raw)
To: Karsten Keil, David Miller
Cc: Hansjoerg Lipp, Karsten Keil, i4ldeveloper, netdev, linux-kernel
In-Reply-To: <20130121-patch-isdn-00.tilman@imap.cc>
Cosmetic changes to drivers/isdn/gigaset/ev-layer.c and
drivers/isdn/gigaset/gigaset.h to improve readability.
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
---
drivers/isdn/gigaset/ev-layer.c | 119 +++++++++++++++++++--------------------
drivers/isdn/gigaset/gigaset.h | 9 ++-
2 files changed, 64 insertions(+), 64 deletions(-)
diff --git a/drivers/isdn/gigaset/ev-layer.c b/drivers/isdn/gigaset/ev-layer.c
index e2383ec..7459b12 100644
--- a/drivers/isdn/gigaset/ev-layer.c
+++ b/drivers/isdn/gigaset/ev-layer.c
@@ -351,10 +351,11 @@ struct reply_t gigaset_tab_cid[] =
static const struct resp_type_t {
- unsigned char *response;
- int resp_code;
- int type;
-} resp_type[] =
+ char *response;
+ int resp_code;
+ int type;
+}
+resp_type[] =
{
{"OK", RSP_OK, RT_NOTHING},
{"ERROR", RSP_ERROR, RT_NOTHING},
@@ -374,11 +375,12 @@ static const struct resp_type_t {
};
static const struct zsau_resp_t {
- unsigned char *str;
- int code;
-} zsau_resp[] =
+ char *str;
+ int code;
+}
+zsau_resp[] =
{
- {"OUTGOING_CALL_PROCEEDING", ZSAU_OUTGOING_CALL_PROCEEDING},
+ {"OUTGOING_CALL_PROCEEDING", ZSAU_PROCEEDING},
{"CALL_DELIVERED", ZSAU_CALL_DELIVERED},
{"ACTIVE", ZSAU_ACTIVE},
{"DISCONNECT_IND", ZSAU_DISCONNECT_IND},
@@ -434,7 +436,7 @@ void gigaset_handle_modem_response(struct cardstate *cs)
len = cs->cbytes;
if (!len) {
/* ignore additional LFs/CRs (M10x config mode or cx100) */
- gig_dbg(DEBUG_MCMD, "skipped EOL [%02X]", cs->respdata[len]);
+ gig_dbg(DEBUG_MCMD, "skipped EOL [%02X]", cs->respdata[0]);
return;
}
cs->respdata[len] = 0;
@@ -707,27 +709,29 @@ static void schedule_init(struct cardstate *cs, int state)
cs->commands_pending = 1;
}
-/* Add "AT" to a command, add the cid, dle encode it, send the result to the
- hardware. */
-static void send_command(struct cardstate *cs, const char *cmd, int cid,
- int dle, gfp_t kmallocflags)
+/* send an AT command
+ * adding the "AT" prefix, cid and DLE encapsulation as appropriate
+ */
+static void send_command(struct cardstate *cs, const char *cmd,
+ struct at_state_t *at_state)
{
+ int cid = at_state->cid;
struct cmdbuf_t *cb;
size_t buflen;
buflen = strlen(cmd) + 12; /* DLE ( A T 1 2 3 4 5 <cmd> DLE ) \0 */
- cb = kmalloc(sizeof(struct cmdbuf_t) + buflen, kmallocflags);
+ cb = kmalloc(sizeof(struct cmdbuf_t) + buflen, GFP_ATOMIC);
if (!cb) {
dev_err(cs->dev, "%s: out of memory\n", __func__);
return;
}
if (cid > 0 && cid <= 65535)
cb->len = snprintf(cb->buf, buflen,
- dle ? "\020(AT%d%s\020)" : "AT%d%s",
+ cs->dle ? "\020(AT%d%s\020)" : "AT%d%s",
cid, cmd);
else
cb->len = snprintf(cb->buf, buflen,
- dle ? "\020(AT%s\020)" : "AT%s",
+ cs->dle ? "\020(AT%s\020)" : "AT%s",
cmd);
cb->offset = 0;
cb->next = NULL;
@@ -886,7 +890,7 @@ static void finish_shutdown(struct cardstate *cs)
gigaset_isdn_stop(cs);
}
- /* The rest is done by cleanup_cs () in user mode. */
+ /* The rest is done by cleanup_cs() in process context. */
cs->cmd_result = -ENODEV;
cs->waiting = 0;
@@ -976,10 +980,9 @@ exit:
}
static void handle_icall(struct cardstate *cs, struct bc_state *bcs,
- struct at_state_t **p_at_state)
+ struct at_state_t *at_state)
{
int retval;
- struct at_state_t *at_state = *p_at_state;
retval = gigaset_isdn_icall(at_state);
switch (retval) {
@@ -1176,7 +1179,7 @@ static void do_action(int action, struct cardstate *cs,
spin_unlock_irqrestore(&cs->lock, flags);
break;
case ACT_ICALL:
- handle_icall(cs, bcs, p_at_state);
+ handle_icall(cs, bcs, at_state);
break;
case ACT_FAILSDOWN:
dev_warn(cs->dev, "Could not shut down the device.\n");
@@ -1264,7 +1267,7 @@ static void do_action(int action, struct cardstate *cs,
cs->commands_pending = 1;
break;
}
- /* fall through */
+ /* bad cid: fall through */
case ACT_FAILCID:
cs->cur_at_seq = SEQ_NONE;
channel = cs->curchannel;
@@ -1339,7 +1342,6 @@ static void do_action(int action, struct cardstate *cs,
*p_resp_code = RSP_ERROR;
break;
}
- /*at_state->getstring = 1;*/
cs->gotfwver = 0;
break;
case ACT_GOTVER:
@@ -1471,7 +1473,6 @@ static void process_event(struct cardstate *cs, struct event_t *ev)
int rcode;
int genresp = 0;
int resp_code = RSP_ERROR;
- int sendcid;
struct at_state_t *at_state;
int index;
int curact;
@@ -1499,7 +1500,6 @@ static void process_event(struct cardstate *cs, struct event_t *ev)
at_state->ConState, ev->type);
bcs = at_state->bcs;
- sendcid = at_state->cid;
/* Setting the pointer to the dial array */
rep = at_state->replystruct;
@@ -1510,10 +1510,12 @@ static void process_event(struct cardstate *cs, struct event_t *ev)
|| !at_state->timer_active) {
ev->type = RSP_NONE; /* old timeout */
gig_dbg(DEBUG_EVENT, "old timeout");
- } else if (!at_state->waiting)
- gig_dbg(DEBUG_EVENT, "timeout occurred");
- else
- gig_dbg(DEBUG_EVENT, "stopped waiting");
+ } else {
+ if (at_state->waiting)
+ gig_dbg(DEBUG_EVENT, "stopped waiting");
+ else
+ gig_dbg(DEBUG_EVENT, "timeout occurred");
+ }
}
spin_unlock_irqrestore(&cs->lock, flags);
@@ -1561,45 +1563,40 @@ static void process_event(struct cardstate *cs, struct event_t *ev)
do_action(rep->action[curact], cs, bcs, &at_state, &p_command,
&genresp, &resp_code, ev);
if (!at_state)
- break; /* may be freed after disconnect */
+ /* at_state destroyed by disconnect */
+ return;
}
- if (at_state) {
- /* Jump to the next con-state regarding the array */
- if (rep->new_ConState >= 0)
- at_state->ConState = rep->new_ConState;
+ /* Jump to the next con-state regarding the array */
+ if (rep->new_ConState >= 0)
+ at_state->ConState = rep->new_ConState;
- if (genresp) {
- spin_lock_irqsave(&cs->lock, flags);
+ if (genresp) {
+ spin_lock_irqsave(&cs->lock, flags);
+ at_state->timer_expires = 0;
+ at_state->timer_active = 0;
+ spin_unlock_irqrestore(&cs->lock, flags);
+ gigaset_add_event(cs, at_state, resp_code, NULL, 0, NULL);
+ } else {
+ /* Send command to modem if not NULL... */
+ if (p_command) {
+ if (cs->connected)
+ send_command(cs, p_command, at_state);
+ else
+ gigaset_add_event(cs, at_state, RSP_NODEV,
+ NULL, 0, NULL);
+ }
+
+ spin_lock_irqsave(&cs->lock, flags);
+ if (!rep->timeout) {
at_state->timer_expires = 0;
at_state->timer_active = 0;
- spin_unlock_irqrestore(&cs->lock, flags);
- gigaset_add_event(cs, at_state, resp_code,
- NULL, 0, NULL);
- } else {
- /* Send command to modem if not NULL... */
- if (p_command) {
- if (cs->connected)
- send_command(cs, p_command,
- sendcid, cs->dle,
- GFP_ATOMIC);
- else
- gigaset_add_event(cs, at_state,
- RSP_NODEV,
- NULL, 0, NULL);
- }
-
- spin_lock_irqsave(&cs->lock, flags);
- if (!rep->timeout) {
- at_state->timer_expires = 0;
- at_state->timer_active = 0;
- } else if (rep->timeout > 0) { /* new timeout */
- at_state->timer_expires = rep->timeout * 10;
- at_state->timer_active = 1;
- ++at_state->timer_index;
- }
- spin_unlock_irqrestore(&cs->lock, flags);
+ } else if (rep->timeout > 0) { /* new timeout */
+ at_state->timer_expires = rep->timeout * 10;
+ at_state->timer_active = 1;
+ ++at_state->timer_index;
}
+ spin_unlock_irqrestore(&cs->lock, flags);
}
}
diff --git a/drivers/isdn/gigaset/gigaset.h b/drivers/isdn/gigaset/gigaset.h
index 8e2fc8f..eb63a0f 100644
--- a/drivers/isdn/gigaset/gigaset.h
+++ b/drivers/isdn/gigaset/gigaset.h
@@ -111,11 +111,10 @@ void gigaset_dbg_buffer(enum debuglevel level, const unsigned char *msg,
/* connection state */
#define ZSAU_NONE 0
-#define ZSAU_DISCONNECT_IND 4
-#define ZSAU_OUTGOING_CALL_PROCEEDING 1
#define ZSAU_PROCEEDING 1
#define ZSAU_CALL_DELIVERED 2
#define ZSAU_ACTIVE 3
+#define ZSAU_DISCONNECT_IND 4
#define ZSAU_NULL 5
#define ZSAU_DISCONNECT_REQ 6
#define ZSAU_UNKNOWN -1
@@ -183,18 +182,22 @@ void gigaset_dbg_buffer(enum debuglevel level, const unsigned char *msg,
#define AT_NUM 7
/* variables in struct at_state_t */
+/* - numeric */
#define VAR_ZSAU 0
#define VAR_ZDLE 1
#define VAR_ZCTP 2
+/* total number */
#define VAR_NUM 3
-
+/* - string */
#define STR_NMBR 0
#define STR_ZCPN 1
#define STR_ZCON 2
#define STR_ZBC 3
#define STR_ZHLC 4
+/* total number */
#define STR_NUM 5
+/* event types */
#define EV_TIMEOUT -105
#define EV_IF_VER -106
#define EV_PROC_CIDMODE -107
--
1.7.3.4
^ permalink raw reply related
* Re: [PATCH net 2/3] net: cdc_mbim: send ZLP after max sized NTBs
From: Bjørn Mork @ 2013-01-21 22:01 UTC (permalink / raw)
To: Yauheni Kaliuta
Cc: netdev-u79uwXL29TY76Z2rM5mHXA, linux-usb-u79uwXL29TY76Z2rM5mHXA,
Greg Suarez, Alexey Orishko, Oliver Neukum
In-Reply-To: <87wqv64gr9.fsf-jZyLUQyd5ymekGDzARDqUg@public.gmane.org>
Yauheni Kaliuta <y.kaliuta-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:
>>>>>> "BM" == Bjørn Mork writes:
>
> > We normally avoid sending ZLPs by padding NTBs with a zero byte
> > if the NTB is shorter than dwNtbOutMaxSize, resulting in a short
> > USB packet instead of a ZLP. But in the case where the NTB length
> > is exactly dwNtbOutMaxSize and this is an exact multiplum of
> > wMaxPacketSize, then we must send a ZLP.
>
> The idea of NCM was to avoid extra ZLPs. If your transfer is exactly
> dwNtbOutMaxSize, it's known, you can submit such request on the receiver
> side and you do not need any EOT indicatation, so the frametime can be
> used for useful data.
Yes, that makes sense. And I understand that both you and Alexey are of
this opinion.
But this idea is by no means made clear (to me) in the spec. I do not
think the current wording is precise enough to expect every implementor
to understand any such intent. The only place I find either "short
packet" or ZLP mentioned in the NCM spec is in tables 3-1 and 3-2,
describing the 16bit and 32bit NTH formats, in the description of the
(d)wBlockLength fields. And even there it is only mentioned in the
context of the special (d)wBlockLength = 0x0000 handling.
If the intent was to prevent ZLPs, then it would have been wise to write
that in the NCM standard. As it stands you have to use a lot of
imagination to read that intent into the current spec.
> I didn't check MBIM specs, but I guess, it wasn't changed. But better get
> Alexey's answer for sure.
No, there are no changes to this area in the MBIM spec AFAIK, so
whatever is correct for NCM is also correct for MBIM. The question is
what is correct for NCM.
> > This fixes an issue seen on a Sierra Wireless MC7710 device
> > where the transmission would fail whenever we ended up padding
> > the NTBs to max size.
>
> Is it buggy?
That is not unlikely. However, if so then it is buggy in a way we just
have to deal with because Windows does...
But before claiming this to be a bug I would like to understand how we
have come to this conclusion that NCM and MBIM devices should not need
ZLPs. I agree that it would have made sense to have such a requirement,
but I cannot find it in the spec.
I would also like to know how Windows works around this issue, because I
am pretty confident that this device works with Windows 8. If it didn't
then the firmware wouldn't have been flagged for release, and it is. It
is even distributed by 3rd party integrators (aka laptop vendors).
Bjørn
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH net 0/3] cdc_ncm and cdc_mbim fixes for 3.8
From: Bjørn Mork @ 2013-01-21 22:12 UTC (permalink / raw)
To: Alexey Orishko
Cc: Alexey ORISHKO, David Miller, netdev@vger.kernel.org,
linux-usb@vger.kernel.org, gsuarez@smithmicro.com,
oneukum@suse.de
In-Reply-To: <CAL_Kpj2QsUqLLCnME3zNHxJpcrJZufN1ZMT=vRsr_gzs70gvpQ@mail.gmail.com>
Alexey Orishko <alexey.orishko@gmail.com> writes:
> On Mon, Jan 21, 2013 at 9:16 PM, Bjørn Mork <bjorn@mork.no> wrote:
>> Alexey ORISHKO <alexey.orishko@stericsson.com> wrote:
>>
>>
>> There are still a couple of questions I am wondering about
>> though. The most important one: what is Windows 8 doing different
>> from us? Does anyone know? I do not have a Windows 8 system myself
>> and I wouldn't know how to snoop on it either.
>>
>
> Are you talking about bulk data? They have their own bugs :-) looks
> like can't handle divisor/remainder in NTB right.
Yes, I am talking about bulk data. Why don't they hit the ZLP issue?
Don't they zero-pad the NTB like we do?
> If you want, I can send you Ellisys log.
Yes, please. That would be very nice to have.
Bjørn
^ permalink raw reply
* [PATCH 1/6] isdn/divert: fix readability damage
From: Tilman Schmidt @ 2013-01-21 21:57 UTC (permalink / raw)
To: Karsten Keil, David Miller
Cc: Hansjoerg Lipp, Karsten Keil, i4ldeveloper, netdev, linux-kernel
In-Reply-To: <20130121-patch-isdn-00.tilman@imap.cc>
Fix up some of the readibility deterioration caused by last year's
ISDN whitespace coding style cleanup.
Note that the checkpatch complaints all apply to the state of the
source before this patch as well, and in many cases even more so.
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
---
drivers/isdn/divert/divert_init.c | 33 ++--
drivers/isdn/divert/isdn_divert.c | 423 ++++++++++++++++++-------------------
drivers/isdn/divert/isdn_divert.h | 28 ++--
3 files changed, 237 insertions(+), 247 deletions(-)
diff --git a/drivers/isdn/divert/divert_init.c b/drivers/isdn/divert/divert_init.c
index 5374c25..267dede 100644
--- a/drivers/isdn/divert/divert_init.c
+++ b/drivers/isdn/divert/divert_init.c
@@ -22,13 +22,13 @@ MODULE_LICENSE("GPL");
/****************************************/
/* structure containing interface to hl */
/****************************************/
-isdn_divert_if divert_if =
-{ DIVERT_IF_MAGIC, /* magic value */
- DIVERT_CMD_REG, /* register cmd */
- ll_callback, /* callback routine from ll */
- NULL, /* command still not specified */
- NULL, /* drv_to_name */
- NULL, /* name_to_drv */
+isdn_divert_if divert_if = {
+ DIVERT_IF_MAGIC, /* magic value */
+ DIVERT_CMD_REG, /* register cmd */
+ ll_callback, /* callback routine from ll */
+ NULL, /* command still not specified */
+ NULL, /* drv_to_name */
+ NULL, /* name_to_drv */
};
/*************************/
@@ -36,14 +36,15 @@ isdn_divert_if divert_if =
/* no cmd line parms */
/*************************/
static int __init divert_init(void)
-{ int i;
+{
+ int i;
- if (divert_dev_init())
- { printk(KERN_WARNING "dss1_divert: cannot install device, not loaded\n");
+ if (divert_dev_init()) {
+ printk(KERN_WARNING "dss1_divert: cannot install device, not loaded\n");
return (-EIO);
}
- if ((i = DIVERT_REG_NAME(&divert_if)) != DIVERT_NO_ERR)
- { divert_dev_deinit();
+ if ((i = DIVERT_REG_NAME(&divert_if)) != DIVERT_NO_ERR) {
+ divert_dev_deinit();
printk(KERN_WARNING "dss1_divert: error %d registering module, not loaded\n", i);
return (-EIO);
}
@@ -61,13 +62,13 @@ static void __exit divert_exit(void)
spin_lock_irqsave(&divert_lock, flags);
divert_if.cmd = DIVERT_CMD_REL; /* release */
- if ((i = DIVERT_REG_NAME(&divert_if)) != DIVERT_NO_ERR)
- { printk(KERN_WARNING "dss1_divert: error %d releasing module\n", i);
+ if ((i = DIVERT_REG_NAME(&divert_if)) != DIVERT_NO_ERR) {
+ printk(KERN_WARNING "dss1_divert: error %d releasing module\n", i);
spin_unlock_irqrestore(&divert_lock, flags);
return;
}
- if (divert_dev_deinit())
- { printk(KERN_WARNING "dss1_divert: device busy, remove cancelled\n");
+ if (divert_dev_deinit()) {
+ printk(KERN_WARNING "dss1_divert: device busy, remove cancelled\n");
spin_unlock_irqrestore(&divert_lock, flags);
return;
}
diff --git a/drivers/isdn/divert/isdn_divert.c b/drivers/isdn/divert/isdn_divert.c
index e61e55f..db432e6 100644
--- a/drivers/isdn/divert/isdn_divert.c
+++ b/drivers/isdn/divert/isdn_divert.c
@@ -19,8 +19,8 @@
/**********************************/
/* structure keeping calling info */
/**********************************/
-struct call_struc
-{ isdn_ctrl ics; /* delivered setup + driver parameters */
+struct call_struc {
+ isdn_ctrl ics; /* delivered setup + driver parameters */
ulong divert_id; /* Id delivered to user */
unsigned char akt_state; /* actual state */
char deflect_dest[35]; /* deflection destination */
@@ -34,8 +34,8 @@ struct call_struc
/********************************************/
/* structure keeping deflection table entry */
/********************************************/
-struct deflect_struc
-{ struct deflect_struc *next, *prev;
+struct deflect_struc {
+ struct deflect_struc *next, *prev;
divert_rule rule; /* used rule */
};
@@ -64,16 +64,16 @@ static void deflect_timer_expire(ulong arg)
del_timer(&cs->timer); /* delete active timer */
spin_unlock_irqrestore(&divert_lock, flags);
- switch (cs->akt_state)
- { case DEFLECT_PROCEED:
- cs->ics.command = ISDN_CMD_HANGUP; /* cancel action */
- divert_if.ll_cmd(&cs->ics);
- spin_lock_irqsave(&divert_lock, flags);
- cs->akt_state = DEFLECT_AUTODEL; /* delete after timeout */
- cs->timer.expires = jiffies + (HZ * AUTODEL_TIME);
- add_timer(&cs->timer);
- spin_unlock_irqrestore(&divert_lock, flags);
- break;
+ switch (cs->akt_state) {
+ case DEFLECT_PROCEED:
+ cs->ics.command = ISDN_CMD_HANGUP; /* cancel action */
+ divert_if.ll_cmd(&cs->ics);
+ spin_lock_irqsave(&divert_lock, flags);
+ cs->akt_state = DEFLECT_AUTODEL; /* delete after timeout */
+ cs->timer.expires = jiffies + (HZ * AUTODEL_TIME);
+ add_timer(&cs->timer);
+ spin_unlock_irqrestore(&divert_lock, flags);
+ break;
case DEFLECT_ALERT:
cs->ics.command = ISDN_CMD_REDIR; /* protocol */
@@ -111,7 +111,8 @@ static void deflect_timer_expire(ulong arg)
int cf_command(int drvid, int mode,
u_char proc, char *msn,
u_char service, char *fwd_nr, ulong *procid)
-{ unsigned long flags;
+{
+ unsigned long flags;
int retval, msnlen;
int fwd_len;
char *p, *ielenp, tmp[60];
@@ -130,8 +131,8 @@ int cf_command(int drvid, int mode,
*p++ = 1; /* length */
*p++ = service; /* service to handle */
- if (mode == 1)
- { if (!*fwd_nr) return (-EINVAL); /* destination missing */
+ if (mode == 1) {
+ if (!*fwd_nr) return (-EINVAL); /* destination missing */
if (strchr(fwd_nr, '.')) return (-EINVAL); /* subaddress not allowed */
fwd_len = strlen(fwd_nr);
*p++ = 0x30; /* number enumeration */
@@ -144,12 +145,12 @@ int cf_command(int drvid, int mode,
msnlen = strlen(msn);
*p++ = 0x80; /* msn number */
- if (msnlen > 1)
- { *p++ = msnlen; /* length */
+ if (msnlen > 1) {
+ *p++ = msnlen; /* length */
strcpy(p, msn);
p += msnlen;
- }
- else *p++ = 0;
+ } else
+ *p++ = 0;
*ielenp = p - ielenp - 1; /* set total IE length */
@@ -186,14 +187,13 @@ int cf_command(int drvid, int mode,
retval = divert_if.ll_cmd(&cs->ics); /* execute command */
- if (!retval)
- { cs->prev = NULL;
+ if (!retval) {
+ cs->prev = NULL;
spin_lock_irqsave(&divert_lock, flags);
cs->next = divert_head;
divert_head = cs;
spin_unlock_irqrestore(&divert_lock, flags);
- }
- else
+ } else
kfree(cs);
return (retval);
} /* cf_command */
@@ -203,15 +203,16 @@ int cf_command(int drvid, int mode,
/* handle a external deflection command */
/****************************************/
int deflect_extern_action(u_char cmd, ulong callid, char *to_nr)
-{ struct call_struc *cs;
+{
+ struct call_struc *cs;
isdn_ctrl ic;
unsigned long flags;
int i;
if ((cmd & 0x7F) > 2) return (-EINVAL); /* invalid command */
cs = divert_head; /* start of parameter list */
- while (cs)
- { if (cs->divert_id == callid) break; /* found */
+ while (cs) {
+ if (cs->divert_id == callid) break; /* found */
cs = cs->next;
} /* search entry */
if (!cs) return (-EINVAL); /* invalid callid */
@@ -220,32 +221,30 @@ int deflect_extern_action(u_char cmd, ulong callid, char *to_nr)
ic.arg = cs->ics.arg;
i = -EINVAL;
if (cs->akt_state == DEFLECT_AUTODEL) return (i); /* no valid call */
- switch (cmd & 0x7F)
- { case 0: /* hangup */
- del_timer(&cs->timer);
- ic.command = ISDN_CMD_HANGUP;
- i = divert_if.ll_cmd(&ic);
- spin_lock_irqsave(&divert_lock, flags);
- cs->akt_state = DEFLECT_AUTODEL; /* delete after timeout */
- cs->timer.expires = jiffies + (HZ * AUTODEL_TIME);
- add_timer(&cs->timer);
- spin_unlock_irqrestore(&divert_lock, flags);
- break;
+ switch (cmd & 0x7F) {
+ case 0: /* hangup */
+ del_timer(&cs->timer);
+ ic.command = ISDN_CMD_HANGUP;
+ i = divert_if.ll_cmd(&ic);
+ spin_lock_irqsave(&divert_lock, flags);
+ cs->akt_state = DEFLECT_AUTODEL; /* delete after timeout */
+ cs->timer.expires = jiffies + (HZ * AUTODEL_TIME);
+ add_timer(&cs->timer);
+ spin_unlock_irqrestore(&divert_lock, flags);
+ break;
case 1: /* alert */
if (cs->akt_state == DEFLECT_ALERT) return (0);
cmd &= 0x7F; /* never wait */
del_timer(&cs->timer);
ic.command = ISDN_CMD_ALERT;
- if ((i = divert_if.ll_cmd(&ic)))
- {
+ if ((i = divert_if.ll_cmd(&ic))) {
spin_lock_irqsave(&divert_lock, flags);
cs->akt_state = DEFLECT_AUTODEL; /* delete after timeout */
cs->timer.expires = jiffies + (HZ * AUTODEL_TIME);
add_timer(&cs->timer);
spin_unlock_irqrestore(&divert_lock, flags);
- }
- else
+ } else
cs->akt_state = DEFLECT_ALERT;
break;
@@ -254,15 +253,13 @@ int deflect_extern_action(u_char cmd, ulong callid, char *to_nr)
strlcpy(cs->ics.parm.setup.phone, to_nr, sizeof(cs->ics.parm.setup.phone));
strcpy(cs->ics.parm.setup.eazmsn, "Testtext manual");
ic.command = ISDN_CMD_REDIR;
- if ((i = divert_if.ll_cmd(&ic)))
- {
+ if ((i = divert_if.ll_cmd(&ic))) {
spin_lock_irqsave(&divert_lock, flags);
cs->akt_state = DEFLECT_AUTODEL; /* delete after timeout */
cs->timer.expires = jiffies + (HZ * AUTODEL_TIME);
add_timer(&cs->timer);
spin_unlock_irqrestore(&divert_lock, flags);
- }
- else
+ } else
cs->akt_state = DEFLECT_ALERT;
break;
@@ -274,19 +271,19 @@ int deflect_extern_action(u_char cmd, ulong callid, char *to_nr)
/* insert a new rule before idx */
/********************************/
int insertrule(int idx, divert_rule *newrule)
-{ struct deflect_struc *ds, *ds1 = NULL;
+{
+ struct deflect_struc *ds, *ds1 = NULL;
unsigned long flags;
- if (!(ds = kmalloc(sizeof(struct deflect_struc),
- GFP_KERNEL)))
+ if (!(ds = kmalloc(sizeof(struct deflect_struc), GFP_KERNEL)))
return (-ENOMEM); /* no memory */
ds->rule = *newrule; /* set rule */
spin_lock_irqsave(&divert_lock, flags);
- if (idx >= 0)
- { ds1 = table_head;
+ if (idx >= 0) {
+ ds1 = table_head;
while ((ds1) && (idx > 0))
{ idx--;
ds1 = ds1->next;
@@ -294,17 +291,16 @@ int insertrule(int idx, divert_rule *newrule)
if (!ds1) idx = -1;
}
- if (idx < 0)
- { ds->prev = table_tail; /* previous entry */
+ if (idx < 0) {
+ ds->prev = table_tail; /* previous entry */
ds->next = NULL; /* end of chain */
if (ds->prev)
ds->prev->next = ds; /* last forward */
else
table_head = ds; /* is first entry */
table_tail = ds; /* end of queue */
- }
- else
- { ds->next = ds1; /* next entry */
+ } else {
+ ds->next = ds1; /* next entry */
ds->prev = ds1->prev; /* prev entry */
ds1->prev = ds; /* backward chain old element */
if (!ds->prev)
@@ -319,17 +315,18 @@ int insertrule(int idx, divert_rule *newrule)
/* delete the rule at position idx */
/***********************************/
int deleterule(int idx)
-{ struct deflect_struc *ds, *ds1;
+{
+ struct deflect_struc *ds, *ds1;
unsigned long flags;
- if (idx < 0)
- { spin_lock_irqsave(&divert_lock, flags);
+ if (idx < 0) {
+ spin_lock_irqsave(&divert_lock, flags);
ds = table_head;
table_head = NULL;
table_tail = NULL;
spin_unlock_irqrestore(&divert_lock, flags);
- while (ds)
- { ds1 = ds;
+ while (ds) {
+ ds1 = ds;
ds = ds->next;
kfree(ds1);
}
@@ -339,13 +336,12 @@ int deleterule(int idx)
spin_lock_irqsave(&divert_lock, flags);
ds = table_head;
- while ((ds) && (idx > 0))
- { idx--;
+ while ((ds) && (idx > 0)) {
+ idx--;
ds = ds->next;
}
- if (!ds)
- {
+ if (!ds) {
spin_unlock_irqrestore(&divert_lock, flags);
return (-EINVAL);
}
@@ -369,12 +365,13 @@ int deleterule(int idx)
/* get a pointer to a specific rule number */
/*******************************************/
divert_rule *getruleptr(int idx)
-{ struct deflect_struc *ds = table_head;
+{
+ struct deflect_struc *ds = table_head;
if (idx < 0) return (NULL);
- while ((ds) && (idx >= 0))
- { if (!(idx--))
- { return (&ds->rule);
+ while ((ds) && (idx >= 0)) {
+ if (!(idx--)) {
+ return (&ds->rule);
break;
}
ds = ds->next;
@@ -386,7 +383,8 @@ divert_rule *getruleptr(int idx)
/* called from common module on an incoming call */
/*************************************************/
static int isdn_divert_icall(isdn_ctrl *ic)
-{ int retval = 0;
+{
+ int retval = 0;
unsigned long flags;
struct call_struc *cs = NULL;
struct deflect_struc *dv;
@@ -394,8 +392,8 @@ static int isdn_divert_icall(isdn_ctrl *ic)
u_char accept;
/* first check the internal deflection table */
- for (dv = table_head; dv; dv = dv->next)
- { /* scan table */
+ for (dv = table_head; dv; dv = dv->next) {
+ /* scan table */
if (((dv->rule.callopt == 1) && (ic->command == ISDN_STAT_ICALLW)) ||
((dv->rule.callopt == 2) && (ic->command == ISDN_STAT_ICALL)))
continue; /* call option check */
@@ -409,10 +407,10 @@ static int isdn_divert_icall(isdn_ctrl *ic)
p = dv->rule.my_msn;
p1 = ic->parm.setup.eazmsn;
accept = 0;
- while (*p)
- { /* complete compare */
- if (*p == '-')
- { accept = 1; /* call accepted */
+ while (*p) {
+ /* complete compare */
+ if (*p == '-') {
+ accept = 1; /* call accepted */
break;
}
if (*p++ != *p1++)
@@ -422,14 +420,15 @@ static int isdn_divert_icall(isdn_ctrl *ic)
} /* complete compare */
if (!accept) continue; /* not accepted */
- if ((strcmp(dv->rule.caller, "0")) || (ic->parm.setup.phone[0]))
- { p = dv->rule.caller;
+ if ((strcmp(dv->rule.caller, "0")) ||
+ (ic->parm.setup.phone[0])) {
+ p = dv->rule.caller;
p1 = ic->parm.setup.phone;
accept = 0;
- while (*p)
- { /* complete compare */
- if (*p == '-')
- { accept = 1; /* call accepted */
+ while (*p) {
+ /* complete compare */
+ if (*p == '-') {
+ accept = 1; /* call accepted */
break;
}
if (*p++ != *p1++)
@@ -440,10 +439,10 @@ static int isdn_divert_icall(isdn_ctrl *ic)
if (!accept) continue; /* not accepted */
}
- switch (dv->rule.action)
- { case DEFLECT_IGNORE:
- return (0);
- break;
+ switch (dv->rule.action) {
+ case DEFLECT_IGNORE:
+ return (0);
+ break;
case DEFLECT_ALERT:
case DEFLECT_PROCEED:
@@ -465,31 +464,29 @@ static int isdn_divert_icall(isdn_ctrl *ic)
cs->ics.parm.setup.screen = dv->rule.screen;
if (dv->rule.waittime)
cs->timer.expires = jiffies + (HZ * dv->rule.waittime);
+ else if (dv->rule.action == DEFLECT_PROCEED)
+ cs->timer.expires = jiffies + (HZ * extern_wait_max);
else
- if (dv->rule.action == DEFLECT_PROCEED)
- cs->timer.expires = jiffies + (HZ * extern_wait_max);
- else
- cs->timer.expires = 0;
+ cs->timer.expires = 0;
cs->akt_state = dv->rule.action;
spin_lock_irqsave(&divert_lock, flags);
cs->divert_id = next_id++; /* new sequence number */
spin_unlock_irqrestore(&divert_lock, flags);
cs->prev = NULL;
- if (cs->akt_state == DEFLECT_ALERT)
- { strcpy(cs->deflect_dest, dv->rule.to_nr);
- if (!cs->timer.expires)
- { strcpy(ic->parm.setup.eazmsn, "Testtext direct");
+ if (cs->akt_state == DEFLECT_ALERT) {
+ strcpy(cs->deflect_dest, dv->rule.to_nr);
+ if (!cs->timer.expires) {
+ strcpy(ic->parm.setup.eazmsn,
+ "Testtext direct");
ic->parm.setup.screen = dv->rule.screen;
strlcpy(ic->parm.setup.phone, dv->rule.to_nr, sizeof(ic->parm.setup.phone));
cs->akt_state = DEFLECT_AUTODEL; /* delete after timeout */
cs->timer.expires = jiffies + (HZ * AUTODEL_TIME);
retval = 5;
- }
- else
+ } else
retval = 1; /* alerting */
- }
- else
- { cs->deflect_dest[0] = '\0';
+ } else {
+ cs->deflect_dest[0] = '\0';
retval = 4; /* only proceed */
}
sprintf(cs->info, "%d 0x%lx %s %s %s %s 0x%x 0x%x %d %d %s\n",
@@ -505,8 +502,8 @@ static int isdn_divert_icall(isdn_ctrl *ic)
dv->rule.waittime,
cs->deflect_dest);
if ((dv->rule.action == DEFLECT_REPORT) ||
- (dv->rule.action == DEFLECT_REJECT))
- { put_info_buffer(cs->info);
+ (dv->rule.action == DEFLECT_REJECT)) {
+ put_info_buffer(cs->info);
kfree(cs); /* remove */
return ((dv->rule.action == DEFLECT_REPORT) ? 0 : 2); /* nothing to do */
}
@@ -519,8 +516,8 @@ static int isdn_divert_icall(isdn_ctrl *ic)
break;
} /* scan_table */
- if (cs)
- { cs->prev = NULL;
+ if (cs) {
+ cs->prev = NULL;
spin_lock_irqsave(&divert_lock, flags);
cs->next = divert_head;
divert_head = cs;
@@ -529,21 +526,21 @@ static int isdn_divert_icall(isdn_ctrl *ic)
put_info_buffer(cs->info);
return (retval);
- }
- else
+ } else
return (0);
} /* isdn_divert_icall */
void deleteprocs(void)
-{ struct call_struc *cs, *cs1;
+{
+ struct call_struc *cs, *cs1;
unsigned long flags;
spin_lock_irqsave(&divert_lock, flags);
cs = divert_head;
divert_head = NULL;
- while (cs)
- { del_timer(&cs->timer);
+ while (cs) {
+ del_timer(&cs->timer);
cs1 = cs;
cs = cs->next;
kfree(cs1);
@@ -555,12 +552,13 @@ void deleteprocs(void)
/* put a address including address type into buffer */
/****************************************************/
static int put_address(char *st, u_char *p, int len)
-{ u_char retval = 0;
+{
+ u_char retval = 0;
u_char adr_typ = 0; /* network standard */
if (len < 2) return (retval);
- if (*p == 0xA1)
- { retval = *(++p) + 2; /* total length */
+ if (*p == 0xA1) {
+ retval = *(++p) + 2; /* total length */
if (retval > len) return (0); /* too short */
len = retval - 2; /* remaining length */
if (len < 3) return (0);
@@ -572,16 +570,13 @@ static int put_address(char *st, u_char *p, int len)
if (*p++ != 0x12) return (0);
if (*p > len) return (0); /* check number length */
len = *p++;
- }
- else
- if (*p == 0x80)
- { retval = *(++p) + 2; /* total length */
- if (retval > len) return (0);
- len = retval - 2;
- p++;
- }
- else
- return (0); /* invalid address information */
+ } else if (*p == 0x80) {
+ retval = *(++p) + 2; /* total length */
+ if (retval > len) return (0);
+ len = retval - 2;
+ p++;
+ } else
+ return (0); /* invalid address information */
sprintf(st, "%d ", adr_typ);
st += strlen(st);
@@ -598,7 +593,8 @@ static int put_address(char *st, u_char *p, int len)
/* report a successful interrogation */
/*************************************/
static int interrogate_success(isdn_ctrl *ic, struct call_struc *cs)
-{ char *src = ic->parm.dss1_io.data;
+{
+ char *src = ic->parm.dss1_io.data;
int restlen = ic->parm.dss1_io.datalen;
int cnt = 1;
u_char n, n1;
@@ -608,50 +604,44 @@ static int interrogate_success(isdn_ctrl *ic, struct call_struc *cs)
if (*src++ != 0x30) return (-101);
if ((n = *src++) > 0x81) return (-102); /* invalid length field */
restlen -= 2; /* remaining bytes */
- if (n == 0x80)
- { if (restlen < 2) return (-103);
+ if (n == 0x80) {
+ if (restlen < 2) return (-103);
if ((*(src + restlen - 1)) || (*(src + restlen - 2))) return (-104);
restlen -= 2;
- }
+ } else if (n == 0x81) {
+ n = *src++;
+ restlen--;
+ if (n > restlen) return (-105);
+ restlen = n;
+ } else if (n > restlen)
+ return (-106);
else
- if (n == 0x81)
- { n = *src++;
- restlen--;
- if (n > restlen) return (-105);
- restlen = n;
- }
- else
- if (n > restlen) return (-106);
- else
- restlen = n; /* standard format */
+ restlen = n; /* standard format */
if (restlen < 3) return (-107); /* no procedure */
if ((*src++ != 2) || (*src++ != 1) || (*src++ != 0x0B)) return (-108);
restlen -= 3;
if (restlen < 2) return (-109); /* list missing */
- if (*src == 0x31)
- { src++;
+ if (*src == 0x31) {
+ src++;
if ((n = *src++) > 0x81) return (-110); /* invalid length field */
restlen -= 2; /* remaining bytes */
- if (n == 0x80)
- { if (restlen < 2) return (-111);
+ if (n == 0x80) {
+ if (restlen < 2) return (-111);
if ((*(src + restlen - 1)) || (*(src + restlen - 2))) return (-112);
restlen -= 2;
- }
+ } else if (n == 0x81) {
+ n = *src++;
+ restlen--;
+ if (n > restlen) return (-113);
+ restlen = n;
+ } else if (n > restlen)
+ return (-114);
else
- if (n == 0x81)
- { n = *src++;
- restlen--;
- if (n > restlen) return (-113);
- restlen = n;
- }
- else
- if (n > restlen) return (-114);
- else
- restlen = n; /* standard format */
+ restlen = n; /* standard format */
} /* result list header */
- while (restlen >= 2)
- { stp = st;
+ while (restlen >= 2) {
+ stp = st;
sprintf(stp, "%d 0x%lx %d %s ", DIVERT_REPORT, ic->parm.dss1_io.ll_id,
cnt++, divert_if.drv_to_name(ic->driver));
stp += strlen(stp);
@@ -674,8 +664,8 @@ static int interrogate_success(isdn_ctrl *ic, struct call_struc *cs)
sprintf(stp, "%d ", (*p++) & 0xFF);
stp += strlen(stp);
n -= 6;
- if (n > 2)
- { if (*p++ != 0x30) continue;
+ if (n > 2) {
+ if (*p++ != 0x30) continue;
if (*p > (n - 2)) continue;
n = *p++;
if (!(n1 = put_address(stp, p, n & 0xFF))) continue;
@@ -692,58 +682,58 @@ static int interrogate_success(isdn_ctrl *ic, struct call_struc *cs)
/* callback for protocol specific extensions */
/*********************************************/
static int prot_stat_callback(isdn_ctrl *ic)
-{ struct call_struc *cs, *cs1;
+{
+ struct call_struc *cs, *cs1;
int i;
unsigned long flags;
cs = divert_head; /* start of list */
cs1 = NULL;
- while (cs)
- { if (ic->driver == cs->ics.driver)
- { switch (cs->ics.arg)
- { case DSS1_CMD_INVOKE:
- if ((cs->ics.parm.dss1_io.ll_id == ic->parm.dss1_io.ll_id) &&
- (cs->ics.parm.dss1_io.hl_id == ic->parm.dss1_io.hl_id))
- { switch (ic->arg)
- { case DSS1_STAT_INVOKE_ERR:
- sprintf(cs->info, "128 0x%lx 0x%x\n",
- ic->parm.dss1_io.ll_id,
- ic->parm.dss1_io.timeout);
- put_info_buffer(cs->info);
- break;
-
- case DSS1_STAT_INVOKE_RES:
- switch (cs->ics.parm.dss1_io.proc)
- { case 7:
- case 8:
- put_info_buffer(cs->info);
- break;
-
- case 11:
- i = interrogate_success(ic, cs);
- if (i)
- sprintf(cs->info, "%d 0x%lx %d\n", DIVERT_REPORT,
- ic->parm.dss1_io.ll_id, i);
- put_info_buffer(cs->info);
- break;
-
- default:
- printk(KERN_WARNING "dss1_divert: unknown proc %d\n", cs->ics.parm.dss1_io.proc);
- break;
- }
-
+ while (cs) {
+ if (ic->driver == cs->ics.driver) {
+ switch (cs->ics.arg) {
+ case DSS1_CMD_INVOKE:
+ if ((cs->ics.parm.dss1_io.ll_id == ic->parm.dss1_io.ll_id) &&
+ (cs->ics.parm.dss1_io.hl_id == ic->parm.dss1_io.hl_id)) {
+ switch (ic->arg) {
+ case DSS1_STAT_INVOKE_ERR:
+ sprintf(cs->info, "128 0x%lx 0x%x\n",
+ ic->parm.dss1_io.ll_id,
+ ic->parm.dss1_io.timeout);
+ put_info_buffer(cs->info);
+ break;
+
+ case DSS1_STAT_INVOKE_RES:
+ switch (cs->ics.parm.dss1_io.proc) {
+ case 7:
+ case 8:
+ put_info_buffer(cs->info);
+ break;
+ case 11:
+ i = interrogate_success(ic, cs);
+ if (i)
+ sprintf(cs->info, "%d 0x%lx %d\n", DIVERT_REPORT,
+ ic->parm.dss1_io.ll_id, i);
+ put_info_buffer(cs->info);
break;
default:
- printk(KERN_WARNING "dss1_divert unknown invoke answer %lx\n", ic->arg);
+ printk(KERN_WARNING "dss1_divert: unknown proc %d\n", cs->ics.parm.dss1_io.proc);
break;
}
- cs1 = cs; /* remember structure */
- cs = NULL;
- continue; /* abort search */
- } /* id found */
- break;
+
+ break;
+
+ default:
+ printk(KERN_WARNING "dss1_divert unknown invoke answer %lx\n", ic->arg);
+ break;
+ }
+ cs1 = cs; /* remember structure */
+ cs = NULL;
+ continue; /* abort search */
+ } /* id found */
+ break;
case DSS1_CMD_INVOKE_ABORT:
printk(KERN_WARNING "dss1_divert unhandled invoke abort\n");
@@ -757,13 +747,12 @@ static int prot_stat_callback(isdn_ctrl *ic)
} /* driver ok */
}
- if (!cs1)
- { printk(KERN_WARNING "dss1_divert unhandled process\n");
+ if (!cs1) {
+ printk(KERN_WARNING "dss1_divert unhandled process\n");
return (0);
}
- if (cs1->ics.driver == -1)
- {
+ if (cs1->ics.driver == -1) {
spin_lock_irqsave(&divert_lock, flags);
del_timer(&cs1->timer);
if (cs1->prev)
@@ -784,20 +773,22 @@ static int prot_stat_callback(isdn_ctrl *ic)
/* status callback from HL */
/***************************/
static int isdn_divert_stat_callback(isdn_ctrl *ic)
-{ struct call_struc *cs, *cs1;
+{
+ struct call_struc *cs, *cs1;
unsigned long flags;
int retval;
retval = -1;
cs = divert_head; /* start of list */
- while (cs)
- { if ((ic->driver == cs->ics.driver) && (ic->arg == cs->ics.arg))
- { switch (ic->command)
- { case ISDN_STAT_DHUP:
- sprintf(cs->info, "129 0x%lx\n", cs->divert_id);
- del_timer(&cs->timer);
- cs->ics.driver = -1;
- break;
+ while (cs) {
+ if ((ic->driver == cs->ics.driver) &&
+ (ic->arg == cs->ics.arg)) {
+ switch (ic->command) {
+ case ISDN_STAT_DHUP:
+ sprintf(cs->info, "129 0x%lx\n", cs->divert_id);
+ del_timer(&cs->timer);
+ cs->ics.driver = -1;
+ break;
case ISDN_STAT_CAUSE:
sprintf(cs->info, "130 0x%lx %s\n", cs->divert_id, ic->parm.num);
@@ -818,8 +809,7 @@ static int isdn_divert_stat_callback(isdn_ctrl *ic)
}
cs1 = cs;
cs = cs->next;
- if (cs1->ics.driver == -1)
- {
+ if (cs1->ics.driver == -1) {
spin_lock_irqsave(&divert_lock, flags);
if (cs1->prev)
cs1->prev->next = cs1->next; /* forward link */
@@ -840,20 +830,19 @@ static int isdn_divert_stat_callback(isdn_ctrl *ic)
/********************/
int ll_callback(isdn_ctrl *ic)
{
- switch (ic->command)
- { case ISDN_STAT_ICALL:
+ switch (ic->command) {
+ case ISDN_STAT_ICALL:
case ISDN_STAT_ICALLW:
return (isdn_divert_icall(ic));
break;
case ISDN_STAT_PROT:
- if ((ic->arg & 0xFF) == ISDN_PTYPE_EURO)
- { if (ic->arg != DSS1_STAT_INVOKE_BRD)
+ if ((ic->arg & 0xFF) == ISDN_PTYPE_EURO) {
+ if (ic->arg != DSS1_STAT_INVOKE_BRD)
return (prot_stat_callback(ic));
else
return (0); /* DSS1 invoke broadcast */
- }
- else
+ } else
return (-1); /* protocol not euro */
default:
diff --git a/drivers/isdn/divert/isdn_divert.h b/drivers/isdn/divert/isdn_divert.h
index 42f2893..55033dd 100644
--- a/drivers/isdn/divert/isdn_divert.h
+++ b/drivers/isdn/divert/isdn_divert.h
@@ -43,8 +43,8 @@
#define DEFLECT_ALL_IDS 0xFFFFFFFF /* all drivers selected */
-typedef struct
-{ ulong drvid; /* driver ids, bit mapped */
+typedef struct {
+ ulong drvid; /* driver ids, bit mapped */
char my_msn[35]; /* desired msn, subaddr allowed */
char caller[35]; /* caller id, partial string with * + subaddr allowed */
char to_nr[35]; /* deflected to number incl. subaddress */
@@ -65,18 +65,18 @@ typedef struct
u_char waittime; /* maximum wait time for proceeding */
} divert_rule;
-typedef union
-{ int drv_version; /* return of driver version */
- struct
- { int drvid; /* id of driver */
+typedef union {
+ int drv_version; /* return of driver version */
+ struct {
+ int drvid; /* id of driver */
char drvnam[30]; /* name of driver */
} getid;
- struct
- { int ruleidx; /* index of rule */
+ struct {
+ int ruleidx; /* index of rule */
divert_rule rule; /* rule parms */
} getsetrule;
- struct
- { u_char subcmd; /* 0 = hangup/reject,
+ struct {
+ u_char subcmd; /* 0 = hangup/reject,
1 = alert,
2 = deflect */
ulong callid; /* id of call delivered by ascii output */
@@ -84,8 +84,8 @@ typedef union
else uus1 string (maxlen 31),
data from rule used if empty */
} fwd_ctrl;
- struct
- { int drvid; /* id of driver */
+ struct {
+ int drvid; /* id of driver */
u_char cfproc; /* cfu = 0, cfb = 1, cfnr = 2 */
ulong procid; /* process id returned when no error */
u_char service; /* basically coded service, 0 = all */
@@ -104,8 +104,8 @@ typedef union
/**************************************************/
/* structure keeping ascii info for device output */
/**************************************************/
-struct divert_info
-{ struct divert_info *next;
+struct divert_info {
+ struct divert_info *next;
ulong usage_cnt; /* number of files still to work */
char info_start[2]; /* info string start */
};
--
1.7.3.4
^ permalink raw reply related
* Re: [PATCH 14/21] NTB: Fix Sparse Warnings
From: Jon Mason @ 2013-01-21 22:28 UTC (permalink / raw)
To: Greg KH; +Cc: linux-kernel, netdev, Dave Jiang, Nicholas Bellinger
In-Reply-To: <20130121213719.GA20506@kroah.com>
On Mon, Jan 21, 2013 at 01:37:19PM -0800, Greg KH wrote:
> On Mon, Jan 21, 2013 at 02:13:50PM -0700, Jon Mason wrote:
> > On Sun, Jan 20, 2013 at 03:45:34PM -0800, Greg KH wrote:
> > > On Sat, Jan 19, 2013 at 02:02:28AM -0700, Jon Mason wrote:
> > > > Address the sparse warnings and resulting fallout
> > > >
> > > > Signed-off-by: Jon Mason <jon.mason@intel.com>
> > > > ---
> > > > drivers/ntb/ntb_hw.c | 7 ++++---
> > > > drivers/ntb/ntb_hw.h | 4 ++--
> > > > drivers/ntb/ntb_transport.c | 32 ++++++++++++++++----------------
> > > > 3 files changed, 22 insertions(+), 21 deletions(-)
> > > >
> > > > diff --git a/drivers/ntb/ntb_hw.c b/drivers/ntb/ntb_hw.c
> > > > index 18cb5dc..b792ccd 100644
> > > > --- a/drivers/ntb/ntb_hw.c
> > > > +++ b/drivers/ntb/ntb_hw.c
> > > > @@ -104,8 +104,9 @@ MODULE_DEVICE_TABLE(pci, ntb_pci_tbl);
> > > > *
> > > > * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
> > > > */
> > > > -int ntb_register_event_callback(struct ntb_device *ndev,
> > > > - void (*func)(void *handle, unsigned int event))
> > > > +int
> > > > +ntb_register_event_callback(struct ntb_device *ndev,
> > > > + void (*func)(void *handle, enum ntb_hw_event event))
> > >
> > > What is sparse complaining about here that moving the function name to
> > > the start of the line fixes? That shouldn't be necessary at all. Ah,
> > > the enum? If so, why change the first line at all?
> >
> > All an effort to get it to fit in under 80 chars.
>
> The original lines were less than 80 characters :)
>
> > > > --- a/drivers/ntb/ntb_transport.c
> > > > +++ b/drivers/ntb/ntb_transport.c
> > > > @@ -58,7 +58,7 @@
> > > > #include <linux/ntb.h>
> > > > #include "ntb_hw.h"
> > > >
> > > > -#define NTB_TRANSPORT_VERSION 1
> > > > +#define NTB_TRANSPORT_VERSION 2
> > >
> > > How is this a sparse fix?
> >
> > I changed the format of the header below.
>
> Ah, ok.
>
> Care to resend this without the first chunk?
Works for me :)
Resending with the minor edit now.
Thanks,
Jon
>
> thanks,
>
> greg k-h
^ permalink raw reply
* [PATCH 1/2] NTB: disable x86_32 support
From: Jon Mason @ 2013-01-21 22:28 UTC (permalink / raw)
To: Greg KH; +Cc: linux-kernel, netdev, Dave Jiang, Nicholas Bellinger
Atomic readq and writeq do not exist by default on some 32bit
architectures, thus causing compile errors due to non-existent symbols.
Since NTB has not been tested 32bit, disable x86_32 support until such
time as this and any other issues can be properly discovered.
Signed-off-by: Jon Mason <jon.mason@intel.com>
---
drivers/ntb/Kconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/ntb/Kconfig b/drivers/ntb/Kconfig
index f69df793..37ee649 100644
--- a/drivers/ntb/Kconfig
+++ b/drivers/ntb/Kconfig
@@ -1,7 +1,7 @@
config NTB
tristate "Intel Non-Transparent Bridge support"
depends on PCI
- depends on X86
+ depends on X86_64
help
The PCI-E Non-transparent bridge hardware is a point-to-point PCI-E bus
connecting 2 systems. When configured, writes to the device's PCI
--
1.7.9.5
^ permalink raw reply related
* [PATCH 2/2] NTB: Fix Sparse Warnings
From: Jon Mason @ 2013-01-21 22:28 UTC (permalink / raw)
To: Greg KH; +Cc: linux-kernel, netdev, Dave Jiang, Nicholas Bellinger
In-Reply-To: <1358807332-22782-1-git-send-email-jon.mason@intel.com>
Address the sparse warnings and resulting fallout
Signed-off-by: Jon Mason <jon.mason@intel.com>
---
drivers/ntb/ntb_hw.c | 4 ++--
drivers/ntb/ntb_hw.h | 4 ++--
drivers/ntb/ntb_transport.c | 32 ++++++++++++++++----------------
3 files changed, 20 insertions(+), 20 deletions(-)
diff --git a/drivers/ntb/ntb_hw.c b/drivers/ntb/ntb_hw.c
index 6d8e937..f802e7c 100644
--- a/drivers/ntb/ntb_hw.c
+++ b/drivers/ntb/ntb_hw.c
@@ -104,7 +104,7 @@ MODULE_DEVICE_TABLE(pci, ntb_pci_tbl);
* RETURNS: An appropriate -ERRNO error value on error, or zero for success.
*/
int ntb_register_event_callback(struct ntb_device *ndev,
- void (*func)(void *handle, unsigned int event))
+ void (*func)(void *handle, enum ntb_hw_event event))
{
if (ndev->event_cb)
return -EINVAL;
@@ -343,7 +343,7 @@ int ntb_read_remote_spad(struct ntb_device *ndev, unsigned int idx, u32 *val)
*
* RETURNS: pointer to virtual address, or NULL on error.
*/
-void *ntb_get_mw_vbase(struct ntb_device *ndev, unsigned int mw)
+void __iomem *ntb_get_mw_vbase(struct ntb_device *ndev, unsigned int mw)
{
if (mw > NTB_NUM_MW)
return NULL;
diff --git a/drivers/ntb/ntb_hw.h b/drivers/ntb/ntb_hw.h
index 5e00951..3a3038c 100644
--- a/drivers/ntb/ntb_hw.h
+++ b/drivers/ntb/ntb_hw.h
@@ -165,14 +165,14 @@ int ntb_register_db_callback(struct ntb_device *ndev, unsigned int idx,
void ntb_unregister_db_callback(struct ntb_device *ndev, unsigned int idx);
int ntb_register_event_callback(struct ntb_device *ndev,
void (*event_cb_func) (void *handle,
- unsigned int event));
+ enum ntb_hw_event event));
void ntb_unregister_event_callback(struct ntb_device *ndev);
int ntb_get_max_spads(struct ntb_device *ndev);
int ntb_write_local_spad(struct ntb_device *ndev, unsigned int idx, u32 val);
int ntb_read_local_spad(struct ntb_device *ndev, unsigned int idx, u32 *val);
int ntb_write_remote_spad(struct ntb_device *ndev, unsigned int idx, u32 val);
int ntb_read_remote_spad(struct ntb_device *ndev, unsigned int idx, u32 *val);
-void *ntb_get_mw_vbase(struct ntb_device *ndev, unsigned int mw);
+void __iomem *ntb_get_mw_vbase(struct ntb_device *ndev, unsigned int mw);
resource_size_t ntb_get_mw_size(struct ntb_device *ndev, unsigned int mw);
void ntb_ring_sdb(struct ntb_device *ndev, unsigned int idx);
void *ntb_find_transport(struct pci_dev *pdev);
diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c
index b3afb24..e0bdfd7 100644
--- a/drivers/ntb/ntb_transport.c
+++ b/drivers/ntb/ntb_transport.c
@@ -58,7 +58,7 @@
#include <linux/ntb.h>
#include "ntb_hw.h"
-#define NTB_TRANSPORT_VERSION 1
+#define NTB_TRANSPORT_VERSION 2
static unsigned int transport_mtu = 0x401E;
module_param(transport_mtu, uint, 0644);
@@ -91,14 +91,14 @@ struct ntb_transport_qp {
bool qp_link;
u8 qp_num; /* Only 64 QP's are allowed. 0-63 */
- struct ntb_rx_info *rx_info;
+ struct ntb_rx_info __iomem *rx_info;
struct ntb_rx_info *remote_rx_info;
void (*tx_handler) (struct ntb_transport_qp *qp, void *qp_data,
void *data, int len);
struct list_head tx_free_q;
spinlock_t ntb_tx_free_q_lock;
- void *tx_mw;
+ void __iomem *tx_mw;
unsigned int tx_index;
unsigned int tx_max_entry;
unsigned int tx_max_frame;
@@ -166,7 +166,7 @@ enum {
};
struct ntb_payload_header {
- u64 ver;
+ unsigned int ver;
unsigned int len;
unsigned int flags;
};
@@ -474,7 +474,7 @@ static void ntb_transport_setup_qp_mw(struct ntb_transport *nt,
u8 mw_num = QP_TO_MW(qp_num);
unsigned int i;
- WARN_ON(nt->mw[mw_num].virt_addr == 0);
+ WARN_ON(nt->mw[mw_num].virt_addr == NULL);
if (nt->max_qps % NTB_NUM_MW && mw_num < nt->max_qps % NTB_NUM_MW)
num_qps_mw = nt->max_qps / NTB_NUM_MW + 1;
@@ -933,7 +933,7 @@ static int ntb_process_rxc(struct ntb_transport_qp *qp)
entry = ntb_list_rm(&qp->ntb_rx_pend_q_lock, &qp->rx_pend_q);
if (!entry) {
dev_dbg(&ntb_query_pdev(qp->ndev)->dev,
- "no buffer - HDR ver %llu, len %d, flags %x\n",
+ "no buffer - HDR ver %u, len %d, flags %x\n",
hdr->ver, hdr->len, hdr->flags);
qp->rx_err_no_buf++;
return -ENOMEM;
@@ -946,9 +946,9 @@ static int ntb_process_rxc(struct ntb_transport_qp *qp)
return -EAGAIN;
}
- if (hdr->ver != qp->rx_pkts) {
+ if (hdr->ver != (u32) qp->rx_pkts) {
dev_dbg(&ntb_query_pdev(qp->ndev)->dev,
- "qp %d: version mismatch, expected %llu - got %llu\n",
+ "qp %d: version mismatch, expected %llu - got %u\n",
qp->qp_num, qp->rx_pkts, hdr->ver);
ntb_list_add(&qp->ntb_rx_pend_q_lock, &entry->entry,
&qp->rx_pend_q);
@@ -965,7 +965,7 @@ static int ntb_process_rxc(struct ntb_transport_qp *qp)
}
dev_dbg(&ntb_query_pdev(qp->ndev)->dev,
- "rx offset %u, ver %llu - %d payload received, buf size %d\n",
+ "rx offset %u, ver %u - %d payload received, buf size %d\n",
qp->rx_index, hdr->ver, hdr->len, entry->len);
if (hdr->len <= entry->len) {
@@ -988,7 +988,7 @@ out:
/* Ensure that the data is fully copied out before clearing the flag */
wmb();
hdr->flags = 0;
- qp->rx_info->entry = qp->rx_index;
+ iowrite32(qp->rx_index, &qp->rx_info->entry);
qp->rx_index++;
qp->rx_index %= qp->rx_max_entry;
@@ -1018,19 +1018,19 @@ static void ntb_transport_rxc_db(void *data, int db_num)
static void ntb_tx_copy_task(struct ntb_transport_qp *qp,
struct ntb_queue_entry *entry,
- void *offset)
+ void __iomem *offset)
{
- struct ntb_payload_header *hdr;
+ struct ntb_payload_header __iomem *hdr;
memcpy_toio(offset, entry->buf, entry->len);
hdr = offset + qp->tx_max_frame - sizeof(struct ntb_payload_header);
- hdr->len = entry->len;
- hdr->ver = qp->tx_pkts;
+ iowrite32(entry->len, &hdr->len);
+ iowrite32((u32) qp->tx_pkts, &hdr->ver);
/* Ensure that the data is fully copied out before setting the flag */
wmb();
- hdr->flags = entry->flags | DESC_DONE_FLAG;
+ iowrite32(entry->flags | DESC_DONE_FLAG, &hdr->flags);
ntb_ring_sdb(qp->ndev, qp->qp_num);
@@ -1052,7 +1052,7 @@ static void ntb_tx_copy_task(struct ntb_transport_qp *qp,
static int ntb_process_tx(struct ntb_transport_qp *qp,
struct ntb_queue_entry *entry)
{
- void *offset;
+ void __iomem *offset;
offset = qp->tx_mw + qp->tx_max_frame * qp->tx_index;
--
1.7.9.5
^ permalink raw reply related
* Re: [PATCH net 2/3] net: cdc_mbim: send ZLP after max sized NTBs
From: Bjørn Mork @ 2013-01-21 22:36 UTC (permalink / raw)
To: Alexey ORISHKO
Cc: netdev@vger.kernel.org, linux-usb@vger.kernel.org, Greg Suarez,
Oliver Neukum, Alexey Orishko
In-Reply-To: <2AC7D4AD8BA1C640B4C60C61C8E520154A8EE17D57@EXDCVYMBSTM006.EQ1STM.local>
Alexey ORISHKO <alexey.orishko@stericsson.com> writes:
>> -----Original Message-----
>> From: Bjørn Mork [mailto:bjorn@mork.no]
>>
>> We normally avoid sending ZLPs by padding NTBs with a zero byte if the
>> NTB is shorter than dwNtbOutMaxSize, resulting in a short USB packet
>> instead of a ZLP. But in the case where the NTB length is exactly
>> dwNtbOutMaxSize and this is an exact multiplum of wMaxPacketSize, then
>> we must send a ZLP.
>>
>> This fixes an issue seen on a Sierra Wireless MC7710 device where the
>> transmission would fail whenever we ended up padding the NTBs to max
>> size.
>
> If you add ZLP for NTBs of dwNtbOutMaxSize, you are heavily affecting CPU
> load, increasing interrupt load by factor of 2 in high load traffic
> scenario and possibly decreasing throughput for all other devices
> which behaves correctly.
>
> Why should others pay for Sierra Wireless fault?
No, they shouldn't.
> At least, do check VID/PID in usbnet and send ZLP only for malfunctioning
> devices.
Yes, this would be easy to implement by simply creating a vendor/device
specific entry in cdc_mbim.
But I do worry that this isn't Sierra Wireless fault, but Qualcomms
fault. SW are using Qualcomm firmware to do the low level stuff, and I
know the Qualcomm firmware implement MBIM functionality so it may very
well be the one to blame (I don't know this for sure, and I have no idea
how to verify it). If this is correct than we can expect hundreds of
devices needing a device specific entry in a year or two. That's not
what I was looking forward to with a class driver like cdc_mbim.
And I still believe that there is reason to ask if this is really a
fault at all.
I don't know if this is particularily useful, but here is an usbmon
extract from the driver without the extra ZLP fix:
ffff88022aef3d40 3883111197 S Bo:8:005:1 -115 4096 = 4e434d48 0c000600 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff880213f2d2c0 3883111233 S Bi:8:005:2 -115 16384 <
ffff8802274eab40 3883111255 S Bi:8:005:2 -115 16384 <
ffff88022fd65980 3883111276 S Bi:8:005:2 -115 16384 <
ffff880204351440 3883111297 S Bi:8:005:2 -115 16384 <
ffff880213f2d8c0 3883111317 S Bi:8:005:2 -115 16384 <
ffff88022aef3d40 3883111460 C Bo:8:005:1 0 4096 >
ffff880204351e00 3883965516 S Bo:8:005:1 -115 4096 = 4e434d48 0c000700 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff880204351e00 3883965849 C Bo:8:005:1 0 4096 >
ffff880204351e00 3884973428 S Bo:8:005:1 -115 4096 = 4e434d48 0c000800 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff880204351e00 3884973752 C Bo:8:005:1 0 4096 >
ffff880204351d40 3885981392 S Bo:8:005:1 -115 4096 = 4e434d48 0c000900 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff880204351d40 3885981638 C Bo:8:005:1 0 4096 >
ffff880231b128c0 3886989391 S Bo:8:005:1 -115 4096 = 4e434d48 0c000a00 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff880231b128c0 3886989622 C Bo:8:005:1 0 4096 >
ffff8802041d8a40 3887997401 S Bo:8:005:1 -115 4096 = 4e434d48 0c000b00 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff8802041d8a40 3887997675 C Bo:8:005:1 0 4096 >
ffff8802041d8a40 3889005720 S Bo:8:005:1 -115 4096 = 4e434d48 0c000c00 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff8802041d8a40 3889006019 C Bo:8:005:1 0 4096 >
ffff8802041d8c80 3890013428 S Bo:8:005:1 -115 4096 = 4e434d48 0c000d00 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff8802041d8c80 3890013784 C Bo:8:005:1 0 4096 >
ffff880213f2d2c0 3892433962 C Bi:8:005:2 -104 0
ffff8802274eab40 3892434032 C Bi:8:005:2 -104 0
ffff88022fd65980 3892434078 C Bi:8:005:2 -104 0
ffff880204351440 3892434126 C Bi:8:005:2 -104 0
ffff880213f2d8c0 3892434172 C Bi:8:005:2 -104 0
same driver, with smaller packets keeping the NTB size below the magic
limit:
ffff88022b1de200 3896794916 S Bi:8:005:2 -115 16384 <
ffff88022767d780 3896794928 C Bo:8:005:1 0 262 >
ffff88022b1de140 3896794948 S Bi:8:005:2 -115 16384 <
ffff88022767d780 3896794971 S Bi:8:005:2 -115 16384 <
ffff88022b1deb00 3896794995 S Bi:8:005:2 -115 16384 <
ffff88022767d900 3896864077 C Bi:8:005:2 0 416 = 4e434d48 0c003500 a0010c00 49505300 10000000 1c007a01 00000000 4500017a
ffff88022767d900 3896864129 S Bi:8:005:2 -115 16384 <
ffff88022d639e40 3897644729 S Bo:8:005:1 -115 262 = 4e434d48 0c000f00 06010c00 49505300 10000000 b8004e00 00000000 00000000
ffff88022d639e40 3897644965 C Bo:8:005:1 0 262 >
ffff88022b1de380 3897711770 C Bi:8:005:2 0 112 = 4e434d48 0c003600 70000c00 49505300 10000000 1c004e00 00000000 4500004e
ffff88022b1de380 3897711812 S Bi:8:005:2 -115 16384 <
ffff88022b07c600 3898646616 S Bo:8:005:1 -115 262 = 4e434d48 0c001000 06010c00 49505300 10000000 b8004e00 00000000 00000000
ffff88022b07c600 3898646848 C Bo:8:005:1 0 262 >
ffff88022b1de200 3898697788 C Bi:8:005:2 0 112 = 4e434d48 0c003700 70000c00 49505300 10000000 1c004e00 00000000 4500004e
ffff88022b1de200 3898697829 S Bi:8:005:2 -115 16384 <
ffff88022b1de140 3901425927 C Bi:8:005:2 -104 0
driver with ZLP:
ffff8802041d82c0 3651255294 C Bo:8:005:1 0 4096 >
ffff88021a3ef380 3652103000 S Bo:8:005:1 -115 4096 = 4e434d48 0c003200 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff88021a3ef380 3652103315 C Bo:8:005:1 0 4096 >
ffff8802041af540 3652439129 C Bi:8:005:2 0 416 = 4e434d48 0c002c00 a0010c00 49505300 10000000 1c007a01 00000000 4500017a
ffff8802041af540 3652439183 S Bi:8:005:2 -115 16384 <
ffff8802041af240 3653099148 C Bi:8:005:2 0 416 = 4e434d48 0c002d00 a0010c00 49505300 10000000 1c007a01 00000000 4500017a
ffff8802041af240 3653099183 S Bi:8:005:2 -115 16384 <
ffff88021a3efb00 3653101946 S Bo:8:005:1 -115 4096 = 4e434d48 0c003300 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff88021a3efb00 3653102201 C Bo:8:005:1 0 4096 >
ffff8802041af9c0 3654039362 C Bi:8:005:2 0 416 = 4e434d48 0c002e00 a0010c00 49505300 10000000 1c007a01 00000000 4500017a
ffff8802041af9c0 3654039414 S Bi:8:005:2 -115 16384 <
ffff88021a3ef980 3654101353 S Bo:8:005:1 -115 4096 = 4e434d48 0c003400 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff88021a3ef980 3654101622 C Bo:8:005:1 0 4096 >
ffff88021a3efc80 3654999361 C Bi:8:005:2 0 416 = 4e434d48 0c002f00 a0010c00 49505300 10000000 1c007a01 00000000 4500017a
ffff88021a3efc80 3654999412 S Bi:8:005:2 -115 16384 <
ffff88021a3ef200 3655102396 S Bo:8:005:1 -115 4096 = 4e434d48 0c003500 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff88021a3ef200 3655102737 C Bo:8:005:1 0 4096 >
ffff88021a3efe00 3655949445 C Bi:8:005:2 0 416 = 4e434d48 0c003000 a0010c00 49505300 10000000 1c007a01 00000000 4500017a
ffff88021a3efe00 3655949491 S Bi:8:005:2 -115 16384 <
ffff88021a3ef500 3656103498 S Bo:8:005:1 -115 4096 = 4e434d48 0c003600 00100c00 49505300 10000000 b8007a01 00000000 00000000
ffff88021a3ef500 3656103779 C Bo:8:005:1 0 4096 >
ffff8802041af540 3657229210 C Bi:8:005:2 0 416 = 4e434d48 0c003100 a0010c00 49505300 10000000 1c007a01 00000000 4500017a
ffff8802041af540 3657229251 S Bi:8:005:2 -115 16384 <
ffff8802041af240 3659438084 C Bi:8:005:2 -104 0
ffff8802041af9c0 3659438159 C Bi:8:005:2 -104 0
ffff88021a3efc80 3659438206 C Bi:8:005:2 -104 0
Bjørn
^ permalink raw reply
* Re: [PATCH 0/6] ISDN patches for net-next
From: David Miller @ 2013-01-21 22:38 UTC (permalink / raw)
To: tilman; +Cc: isdn, hjlipp, keil, i4ldeveloper, netdev, linux-kernel
In-Reply-To: <20130121-patch-isdn-00.tilman@imap.cc>
From: Tilman Schmidt <tilman@imap.cc>
Date: Mon, 21 Jan 2013 22:57:20 +0100 (CET)
> here's a series of six patches to the ISDN subsystem. Please consider
> for application to net-next. The last one should also go into -stable
> and is tagged "CC: stable" accordingly.
Patches 1-5 applied to net-next, patch 6 applied to 'net' and queued
up for -stable.
We don't have changes bypass 'net' and go into stable from net-next
like you intended. We also don't use CC: stable for networking
patches, you ask me to queue them up into my -stable patch queue
for networking instead.
Thanks.
^ permalink raw reply
* Re: [RFC PATCH linux-next] et131x: Promote staging et131x driver to drivers/net
From: Mark Einon @ 2013-01-21 23:44 UTC (permalink / raw)
To: Dan Carpenter; +Cc: Greg KH, devel, sfr, davem, linux-kernel, netdev
In-Reply-To: <20130119110328.GO4584@mwanda>
On 19 January 2013 11:03, Dan Carpenter <dan.carpenter@oracle.com> wrote:
<snip>
>
> et131x_get_regs() has endian bugs calling et131x_mii_read().
>
</snip>
Hi Dan,
Could you be a bit more descriptive about the issues you think there
are with these calls?
cheers,
Mark
^ permalink raw reply
* Re: r8169 doing more work than napi weight
From: Francois Romieu @ 2013-01-21 23:42 UTC (permalink / raw)
To: Timo Teras; +Cc: netdev
In-Reply-To: <20130121171241.2af52e12@vostro>
Timo Teras <timo.teras@iki.fi> :
[...]
> Or just delete the whole block as obsolete. 'git log' says the problem
> should have gone away by always using hardware Rx VLAN. See commit
> 05af214 "r8169: fix Ethernet Hangup for RTL8110SC rev d".
Yes, this stuff looks really strange.
Turning the whole "(desc->opts2 & cpu_to_le32(0xfffe000)) && ... " test
as a WARN_ON_ONCE in the RxRES branch seems a good trade-off from a
coward's viewpoint.
--
Ueimor
^ permalink raw reply
* Re: 3.7.3+: Bad paging request in ip_rcv_finish while running NFS traffic.
From: Ben Greear @ 2013-01-22 0:32 UTC (permalink / raw)
To: netdev, linux-nfs@vger.kernel.org
In-Reply-To: <50FDADF4.3060601@candelatech.com>
On 01/21/2013 01:07 PM, Ben Greear wrote:
> I posted about this a few days ago, but this time the patches applied
> are minimal and there are no out-of-tree kernel modules loaded.
Here's another crash, this time with SLUB memory debugging turned on.
Seems much harder to hit this way...I've only managed this one. I
believe the RCX register might be interesting...that 6b is probably
freed memory poisioning. Maybe skb or skb_dest() is already freed?
I have added a 'verify_mem_not_deleted(skb)' before the
dst_input line in ip_rcv_finish..but so far, it hasn't
hit the problem...
general protection fault: 0000 [#1] PREEMPT SMP
Modules linked in: 8021q garp stp llc nfsv4 auth_rpcgss nfs fscache macvlan pktgen lockd sunrpc iTCO_wdt iTCO_vendor_support gpio_ich coretemp hwmon kvm_intel
kvm microcode pcspkr i2c_i801 lpc_ich igb e1000e ioatdma ptp i7core_edac pps_core edac_core dca uinput ipv6 mgag200 i2c_algo_bit drm_kms_helper ttm drm i2c_core
[last unloaded: macvlan]
CPU 2
Pid: 22, comm: rcuc/2 Tainted: G C O 3.7.3+ #38 Iron Systems Inc. EE2610R/X8ST3
RIP: 0010:[<ffffffff814c196d>] [<ffffffff814c196d>] ip_rcv_finish+0x334/0x34f
RSP: 0018:ffff88041fc43d98 EFLAGS: 00010282
RAX: ffff880369f67180 RBX: ffff8803e4144fc0 RCX: 000000006b6b6b6b
RDX: ffff8803dd632f64 RSI: ffffffff81a2a580 RDI: ffff8803e4144fc0
RBP: ffff88041fc43db8 R08: ffffffff814c1639 R09: ffffffff814c1639
R10: ffff8803dd632f64 R11: ffff8803dd632f64 R12: ffff8803dd632f64
R13: ffff88041fc58420 R14: ffff88040cfa8000 R15: 0000000000000000
FS: 0000000000000000(0000) GS:ffff88041fc40000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 000000000079e008 CR3: 00000003d5759000 CR4: 00000000000007e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process rcuc/2 (pid: 22, threadinfo ffff88040d7d4000, task ffff88040d7d8000)
Stack:
ffff8803e4144fc0 ffffffff814c1639 ffff8803e4144fc0 ffff88040cfa8000
ffff88041fc43de8 ffffffff814c1e61 ffff880480000000 0000000000000246
ffff8803e4144fc0 ffff88040cfa8000 ffff88041fc43e18 ffffffff814c20ec
Call Trace:
<IRQ>
[<ffffffff814c1639>] ? skb_dst+0x41/0x41
[<ffffffff814c1e61>] NF_HOOK.clone.1+0x4c/0x53
[<ffffffff814c20ec>] ip_rcv+0x237/0x267
[<ffffffff814887f0>] __netif_receive_skb+0x537/0x5e0
[<ffffffff8148895a>] process_backlog+0xc1/0x1b7
[<ffffffff8148b082>] ? net_rx_action+0x1c4/0x1e2
[<ffffffff8148af70>] net_rx_action+0xb2/0x1e2
[<ffffffff8108eeb8>] __do_softirq+0xab/0x17b
[<ffffffff81557cfc>] call_softirq+0x1c/0x30
<EOI>
[<ffffffff8100bd46>] do_softirq+0x46/0x9e
[<ffffffff810fb709>] ? rcu_cpu_kthread+0xf6/0x12f
[<ffffffff8108f08d>] _local_bh_enable_ip+0xb3/0xe7
[<ffffffff8108f0d9>] local_bh_enable+0xd/0x10
[<ffffffff810fb709>] rcu_cpu_kthread+0xf6/0x12f
[<ffffffff810aacde>] smpboot_thread_fn+0x253/0x259
[<ffffffff810aaa8b>] ? test_ti_thread_flag.clone.0+0x11/0x11
[<ffffffff810a3439>] kthread+0xc2/0xca
[<ffffffff810a3377>] ? __init_kthread_worker+0x56/0x56
[<ffffffff815569bc>] ret_from_fork+0x7c/0xb0
[<ffffffff810a3377>] ? __init_kthread_worker+0x56/0x56
Code: 8b 80 58 04 00 00 48 8b 80 48 01 00 00 65 48 ff 80 c8 00 00 00 8b 53 68 65 48 01 90 e8 00 00 00 48 89 df e8
8e fc ff ff 48 89 df <ff> 50 50 eb 0d 48 89 df e8 2c eb fb ff b8 01 00 00 00 5b 41 5c
RIP [<ffffffff814c196d>] ip_rcv_finish+0x334/0x34f
RSP <ffff88041fc43d98>
[drm:mgag200_bo_reserve] *ERROR* reserve failed ffff8803feba9fb0
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [PATCH V6 1/3] virtio-net: fix the set affinity bug when CPU IDs are not consecutive
From: Rusty Russell @ 2013-01-22 1:12 UTC (permalink / raw)
To: Wanlong Gao, linux-kernel
Cc: Michael S. Tsirkin, Jason Wang, Eric Dumazet, virtualization,
netdev, Wanlong Gao
In-Reply-To: <1358767524-17934-1-git-send-email-gaowanlong@cn.fujitsu.com>
Wanlong Gao <gaowanlong@cn.fujitsu.com> writes:
> As Michael mentioned, set affinity and select queue will not work very
> well when CPU IDs are not consecutive, this can happen with hot unplug.
> Fix this bug by traversal the online CPUs, and create a per cpu variable
> to find the mapping from CPU to the preferable virtual-queue.
This series looks fairly sane at a glance, to me, but MST is the Ack you
need.
Thanks,
Rusty.
^ 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