* Re: [PATCH net] x86: bpf_jit: fix two bugs in eBPF JIT compiler
From: Eric Dumazet @ 2014-10-11 3:12 UTC (permalink / raw)
To: Alexei Starovoitov
Cc: David S. Miller, Darrick J. Wong, Eric Dumazet, Daniel Borkmann,
H. Peter Anvin, Thomas Gleixner, Ingo Molnar, netdev,
linux-kernel
In-Reply-To: <1412995493-16100-1-git-send-email-ast@plumgrid.com>
On Fri, 2014-10-10 at 19:44 -0700, Alexei Starovoitov wrote:
> 2.
> while staring at the code realized that 64-byte buffer may not be enough
> when 1st insn is large, so increase it to 128 to avoid buffer overflow
> (theoretical maximum size of prologue+div is 109) and add runtime check.
>
> diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
> index d56cd1f..8266896 100644
> --- a/arch/x86/net/bpf_jit_comp.c
> +++ b/arch/x86/net/bpf_jit_comp.c
> @@ -187,7 +187,8 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image,
> {
> struct bpf_insn *insn = bpf_prog->insnsi;
> int insn_cnt = bpf_prog->len;
> - u8 temp[64];
> + bool seen_ld_abs = ctx->seen_ld_abs | (oldproglen == 0);
> + u8 temp[128];
Hmmm. I would use some guard like :
#define BPF_MAX_INSN_SIZE 128
#define BPF_INSN_SAFETY 64
u8 temp[MAX_INSN_SIZE + BPF_INSN_SAFETY];
> + if (ilen >= sizeof(temp)) {
if (ilen > BPF_MAX_INSN_SIZE) {
...
> + pr_err("bpf_jit_compile fatal insn size error\n");
> + return -EFAULT;
> + }
> +
Otherwise, we might have corrupted stack and panic anyway.
^ permalink raw reply
* [PATCH net] x86: bpf_jit: fix two bugs in eBPF JIT compiler
From: Alexei Starovoitov @ 2014-10-11 2:44 UTC (permalink / raw)
To: David S. Miller
Cc: Darrick J. Wong, Eric Dumazet, Daniel Borkmann, H. Peter Anvin,
Thomas Gleixner, Ingo Molnar, netdev, linux-kernel
1.
JIT compiler using multi-pass approach to converge to final image size,
since x86 instructions are variable length. It starts with large
gaps between instructions (so some jumps may use imm32 instead of imm8)
and iterates until total program size is the same as in previous pass.
This algorithm works only if program size is strictly decreasing.
Programs that use LD_ABS insn need additional code in prologue, but it
was not emitted during 1st pass, so there was a chance that 2nd pass would
adjust imm32->imm8 jump offsets to the same number of bytes as increase in
prologue, which may cause algorithm to erroneously decide that size converged.
Fix it by always emitting largest prologue in the first pass which
is detected by oldproglen==0 check.
Also change error check condition 'proglen != oldproglen' to fail gracefully.
2.
while staring at the code realized that 64-byte buffer may not be enough
when 1st insn is large, so increase it to 128 to avoid buffer overflow
(theoretical maximum size of prologue+div is 109) and add runtime check.
Fixes: 622582786c9e ("net: filter: x86: internal BPF JIT")
Reported-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
---
note in classic BPF programs 1st insn is always short move, but native eBPF
programs may trigger buffer overflow. I couldn't force the crash with overflow,
since there are no further calls while this part of stack is used.
Both are ugly bugs regardless.
When net-next opens I will add narrowed down testcase from 'nmap' to testsuite.
arch/x86/net/bpf_jit_comp.c | 21 +++++++++++++++------
1 file changed, 15 insertions(+), 6 deletions(-)
diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index d56cd1f..8266896 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -187,7 +187,8 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image,
{
struct bpf_insn *insn = bpf_prog->insnsi;
int insn_cnt = bpf_prog->len;
- u8 temp[64];
+ bool seen_ld_abs = ctx->seen_ld_abs | (oldproglen == 0);
+ u8 temp[128];
int i;
int proglen = 0;
u8 *prog = temp;
@@ -225,7 +226,7 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image,
EMIT2(0x31, 0xc0); /* xor eax, eax */
EMIT3(0x4D, 0x31, 0xED); /* xor r13, r13 */
- if (ctx->seen_ld_abs) {
+ if (seen_ld_abs) {
/* r9d : skb->len - skb->data_len (headlen)
* r10 : skb->data
*/
@@ -685,7 +686,7 @@ xadd: if (is_imm8(insn->off))
case BPF_JMP | BPF_CALL:
func = (u8 *) __bpf_call_base + imm32;
jmp_offset = func - (image + addrs[i]);
- if (ctx->seen_ld_abs) {
+ if (seen_ld_abs) {
EMIT2(0x41, 0x52); /* push %r10 */
EMIT2(0x41, 0x51); /* push %r9 */
/* need to adjust jmp offset, since
@@ -699,7 +700,7 @@ xadd: if (is_imm8(insn->off))
return -EINVAL;
}
EMIT1_off32(0xE8, jmp_offset);
- if (ctx->seen_ld_abs) {
+ if (seen_ld_abs) {
EMIT2(0x41, 0x59); /* pop %r9 */
EMIT2(0x41, 0x5A); /* pop %r10 */
}
@@ -804,7 +805,8 @@ emit_jmp:
goto common_load;
case BPF_LD | BPF_ABS | BPF_W:
func = CHOOSE_LOAD_FUNC(imm32, sk_load_word);
-common_load: ctx->seen_ld_abs = true;
+common_load:
+ ctx->seen_ld_abs = seen_ld_abs = true;
jmp_offset = func - (image + addrs[i]);
if (!func || !is_simm32(jmp_offset)) {
pr_err("unsupported bpf func %d addr %p image %p\n",
@@ -878,6 +880,11 @@ common_load: ctx->seen_ld_abs = true;
}
ilen = prog - temp;
+ if (ilen >= sizeof(temp)) {
+ pr_err("bpf_jit_compile fatal insn size error\n");
+ return -EFAULT;
+ }
+
if (image) {
if (unlikely(proglen + ilen > oldproglen)) {
pr_err("bpf_jit_compile fatal error\n");
@@ -934,9 +941,11 @@ void bpf_int_jit_compile(struct bpf_prog *prog)
goto out;
}
if (image) {
- if (proglen != oldproglen)
+ if (proglen != oldproglen) {
pr_err("bpf_jit: proglen=%d != oldproglen=%d\n",
proglen, oldproglen);
+ goto out;
+ }
break;
}
if (proglen == oldproglen) {
--
1.7.9.5
^ permalink raw reply related
* Re: fec_main: context imbalance in 'fec_set_features'
From: Christopher Li @ 2014-10-11 2:07 UTC (permalink / raw)
To: Fabio Estevam
Cc: Frank Li, Duan Fugang-B38611, David S. Miller,
netdev@vger.kernel.org, Linux-Sparse
In-Reply-To: <CAOMZO5CJKp7+d8EPRZdr-3Bd9UxVU8fxLybFZ4uLn+cUQjPH8w@mail.gmail.com>
On Thu, Oct 2, 2014 at 8:15 AM, Fabio Estevam <festevam@gmail.com> wrote:
> Hi,
>
> sparse complains the following:
>
> drivers/net/ethernet/freescale/fec_main.c:2835:12: warning: context
> imbalance in 'fec_set_features' - different lock contexts for basic
> block
That is expected to receive warnings. Sparse can't really tell the lock
and unlock are actually on the same condition and it never change.
>
> The code looks like this:
>
> static int fec_set_features(struct net_device *netdev,
> netdev_features_t features)
> {
> struct fec_enet_private *fep = netdev_priv(netdev);
> netdev_features_t changed = features ^ netdev->features;
>
> /* Quiesce the device if necessary */
> if (netif_running(netdev) && changed & FEATURES_NEED_QUIESCE) {
e.g. netdev is running here, you take the lock.
> napi_disable(&fep->napi);
> netif_tx_lock_bh(netdev);
> fec_stop(netdev);
> }
>
> netdev->features = features;
>>
> /* Resume the device after updates */
> if (netif_running(netdev) && changed & FEATURES_NEED_QUIESCE) {
Then there is race some one shutdown the netdev before the code hits
here (is it possible? it is hard to tell from this source alone.)
Then the unlock is skipped.
> fec_restart(netdev);
> netif_tx_wake_all_queues(netdev);
Sparse is complaining, there exist a code path, from execution flow
point of view, (without considering the data flow analyze), there exists a
code path lock and unlock are not balanced.
It is possible that warning code path is not actually executable
due to data flow reason(e.g. "changed" never change during the same function).
But sparse did not have data flow analyze to know that.
If you move the code, e.g. the code inside the lock area into a function.
You can write:
if (need_lock) {
lock();
do_something_real();
unlock();
} else {
do_something_real();
}
That way, sparse don't see such a code path can trigger lock unbalanced.
Chris
^ permalink raw reply
* [PATCH net] tcp: fix ooo_okay setting vs Small Queues
From: Eric Dumazet @ 2014-10-11 1:06 UTC (permalink / raw)
To: David Miller; +Cc: netdev
From: Eric Dumazet <edumazet@google.com>
TCP Small Queues (tcp_tsq_handler()) can hold one reference on
sk->sk_wmem_alloc, preventing skb->ooo_okay being set.
We should relax test done to set skb->ooo_okay to take care
of this extra reference.
Minimal truesize of skb containing one byte of payload is
SKB_TRUESIZE(1)
Without this fix, we have more chance locking flows into the wrong
transmit queue.
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
net/ipv4/tcp_output.c | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 8d4eac793700..e2be1f80e0be 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -914,9 +914,13 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
tcp_ca_event(sk, CA_EVENT_TX_START);
/* if no packet is in qdisc/device queue, then allow XPS to select
- * another queue.
+ * another queue. We can be called from tcp_tsq_handler()
+ * which holds one reference to sk_wmem_alloc.
+ *
+ * TODO: Ideally, in-flight pure ACK packets should not matter here.
+ * One way to get this would be to set skb->truesize = 2 on them.
*/
- skb->ooo_okay = sk_wmem_alloc_get(sk) == 0;
+ skb->ooo_okay = sk_wmem_alloc_get(sk) < SKB_TRUESIZE(1);
skb_push(skb, tcp_header_size);
skb_reset_transport_header(skb);
^ permalink raw reply related
* RE: [PATCH net 1/1] hyperv: Fix a bug in netvsc_send()
From: Long Li @ 2014-10-10 23:39 UTC (permalink / raw)
To: Sitsofe Wheeler, David Miller
Cc: olaf@aepfle.de, netdev@vger.kernel.org, jasowang@redhat.com,
linux-kernel@vger.kernel.org, apw@canonical.com,
devel@linuxdriverproject.org
In-Reply-To: <20141009133200.GA17134@sucs.org>
Thanks Sitsofe. Can you provide more details on the test setup?
The kernel trace shows that skb->mac_header=0xffff (which means not yet set, it's in RCX: 000000000000ffff).
-----Original Message-----
From: devel [mailto:driverdev-devel-bounces@linuxdriverproject.org] On Behalf Of Sitsofe Wheeler
Sent: Thursday, October 09, 2014 6:32 AM
To: David Miller
Cc: olaf@aepfle.de; netdev@vger.kernel.org; jasowang@redhat.com; linux-kernel@vger.kernel.org; apw@canonical.com; devel@linuxdriverproject.org
Subject: Re: [PATCH net 1/1] hyperv: Fix a bug in netvsc_send()
On Sun, Oct 05, 2014 at 09:11:29PM -0400, David Miller wrote:
> From: "K. Y. Srinivasan" <kys@microsoft.com>
> Date: Sun, 5 Oct 2014 10:42:51 -0700
>
> > After the packet is successfully sent, we should not touch the
> > packet as it may have been freed. This patch is based on the work
> > done by Long Li <longli@microsoft.com>.
> >
> > David, please queue this up for stable.
With 3.17.0 g782d59c (which should include this patch) I'm still seeing the following:
Oct 09 13:14:51 a network[428]: Bringing up interface eth0:
Oct 09 13:14:51 a dhclient[538]: DHCPREQUEST on eth0 to 255.255.255.255 port 67 (xid=0x1dd33078) Oct 09 13:14:51 a dhclient[538]: DHCPACK from 10.x.x.x (xid=0x1dd33078) Oct 09 13:14:55 a kernel: BUG: unable to handle kernel paging request at ffff8800ed2e72e3 Oct 09 13:14:55 a kernel: IP: [<ffffffff814ede1d>] netvsc_select_queue+0x3d/0x150 Oct 09 13:14:55 a kernel: PGD 2db5067 PUD 2075be067 PMD 207454067 PTE 80000000ed2e7060 Oct 09 13:14:55 a kernel: Oops: 0000 [#1] SMP DEBUG_PAGEALLOC Oct 09 13:14:55 a kernel: CPU: 6 PID: 566 Comm: arping Not tainted 3.17.0.x86_64-05585-g782d59c #147 Oct 09 13:14:55 a kernel: Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS 090006 05/23/2012 Oct 09 13:14:55 a kernel: task: ffff8801f978b9f0 ti: ffff8801f3b84000 task.ti: ffff8801f3
b84000 Oct 09 13:14:55 a kernel: RIP: 0010:[<ffffffff814ede1d>] [<ffffffff814ede1d>] netvsc_select_queue+0x3d/0x150 Oct 09 13:14:55 a kernel: RSP: 0018:ffff8801f3b87c60 EFLAGS: 00010202 Oc
t 09 13:14:55 a kernel: RAX: 0000000000000000 RBX: ffff8800f13e8000 RCX: 000000000000ffff Oct 09 13:14:55 a kernel: RDX: ffff8800ed2d72d8 RSI: ffff8801fabca1c0 RDI: ffff8800f13e8000 Oct 09 13:14:55 a kernel: RBP: ffff8801f3b87c88 R08: 000000000000002a R09: 0000000000000000 Oct 09 13:14:55 a kernel: R10: ffff8801f83b3f60 R11: 0000000000000008 R12: ffff8801fabca1c0 Oct 09 13:14:55 a kernel: R13: 0000000000000000 R14: ffff8800ed359bd8 R15: ffff8801fabca1c0 Oct 09 13:14:55 a kernel: FS: 00007f943a5c9740(0000) GS:ffff880206cc0000(0000) knlGS:0000000000000000 Oct 09 13:14:55 a kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 Oct 09 13:14:55 a kernel: CR2: ffff8800ed2e72e3 CR3: 00000001f3957000 CR4: 00000000000406e0 Oct 09 13:14:55 a kernel: Stack:
Oct 09 13:14:55 a kernel: ffffffff816a0221 ffff8800f13e8000 000000000000001c 0000000000000000 Oct 09 13:14:55 a kernel: ffff8800ed359bd8 ffff8801f3b87d48 ffffffff816a3fce ffff8801f3b87cb0 Oct 09 13:14:55 a kernel: ffffffff816c34a7 0000000000000001 ffff8801f3b87db8 000000000000001c Oct 09 13:14:55 a kernel: Call Trace:
Oct 09 13:14:55 a kernel: [<ffffffff816a0221>] ? packet_pick_tx_queue+0x31/0xa0 Oct 09 13:14:55 a kernel: [<ffffffff816a3fce>] packet_sendmsg+0xc6e/0xe30 Oct 09 13:14:55 a kernel: [<ffffffff816c34a7>] ? _raw_spin_unlock+0x27/0x40 Oct 09 13:14:55 a kernel: [<ffffffff81091bba>] ? prepare_creds+0x3a/0x170 Oct 09 13:14:55 a kernel: [<ffffffff815d2e08>] sock_sendmsg+0x88/0xb0 Oct 09 13:14:55 a kernel: [<ffffffff81188f83>] ? might_fault+0xa3/0xb0 Oct 09 13:14:55 a kernel: [<ffffffff81188f3a>] ? might_fault+0x5a/0xb0 Oct 09 13:14:55 a kernel: [<ffffffff815d2f3e>] SYSC_sendto+0x10e/0x150 Oct 09 13:14:55 a kernel: [<ffffffff81188f3a>] ? might_fault+0x5a/0xb0 Oct 09 13:14:55 a kernel: [<ffffffff816c41d5>] ? sysret_check+0x22/0x5d Oct 09 13:14:55 a kernel: [<ffffffff810ba3fd>] ? trace_hard
irqs_on_caller+0x17d/0x210
Oct 09 13:14:55 a kernel: [<ffffffff813a20ee>] ? trace_hardirqs_on_thunk+0x3a/0x3f Oct 09 13:14:55 a kernel: [<ffffffff815d3f1e>] SyS_sendto+0xe/0x10 Oct 09 13:14:55 a kernel: [<ffffffff816c41a9>] system_call_fastpath+0x16/0x1b Oct 09 13:14:55 a kernel: Code: 00 4d 85 d2 0f 84 1c 01 00 00 44 8b 9f 8c 03 00 00 31 c0 41 83 fb 01 0f 86 1b 01 00 00 0f b7 8e b6 00 00 00 Oct 09 13:14:55 a kernel: RIP [<ffffffff814ede1d>] netvsc_select_queue+0x3d/0x150 Oct 09 13:14:55 a kernel: RSP <ffff8801f3b87c60> Oct 09 13:14:55 a kernel: CR2: ffff8800ed2e72e3 Oct 09 13:14:55 a kernel: ---[ end trace e52f922dd7435e0d ]---
Was the above meant to have been fixed by the patch "[PATCH 1/1]
Drivers: net: hyperv: Cleanup netvsc_change_mtu ()" from
https://lkml.org/lkml/2014/8/29/369 ? If so will that patch be resent?
--
Sitsofe | http://sucs.org/~sits/
_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
^ permalink raw reply
* Re: [RFC v2 3/6] kthread: warn on kill signal if not OOM
From: Tom Gundersen @ 2014-10-10 22:45 UTC (permalink / raw)
To: Anatol Pomozov
Cc: One Thousand Gnomes, Takashi Iwai, Kay Sievers, Sreekanth Reddy,
James Bottomley, Praveen Krishnamoorthy, hare,
Nagalakshmi Nandigama, Wu Zhangjin, Tetsuo Handa,
mpt-fusionlinux.pdl, Tim Gardner, Benjamin Poirier,
Santosh Rastapur, Casey Leedom, Hariprasad S, Pierre Fersing,
Arjan van de Ven, Abhijit Mahajan, systemd Mailing List
In-Reply-To: <CAOMFOmVSwPAbPtNzPn93QGjbXh1B3QC7B2Ahd1Qxc=gWb6H7fQ@mail.gmail.com>
On Fri, Oct 10, 2014 at 11:54 PM, Anatol Pomozov
<anatol.pomozov@gmail.com> wrote:
> 1) Why not to make the timeout configurable through config file? There
> is already udev.conf you can put config option there. Thus people with
> modprobe issues can easily "fix" the problem. And then decrease
> default timeout back to 30 seconds. I agree that long module loading
> (more than 30 secs) is abnormal and should be investigated by driver
> authors.
We can already configure this either on the udev or kernel
commandline, is that not sufficient (I don't object to also adding it
to the config file, just asking)?
> 2) Could you add 'echo w > /proc/sysrq-trigger' to udev code right
> before killing the "modprobe" thread? sysrq will print information
> about stuck threads (including modprobe itself) this will make
> debugging easier. e.g. dmesg here
> https://bugs.archlinux.org/task/40454 says nothing where the threads
> were stuck.
Are the current warnings (in udev git) sufficient (should tell you
which module is taking long, but still won't tell you which kernel
thread of course)?
Cheers,
Tom
^ permalink raw reply
* Problem - Connection reset - Ralink RT3290 - HP Split 13 x2
From: Vincent Fortier @ 2014-10-10 22:39 UTC (permalink / raw)
To: netdev
My connection stop really really often. Sometimes it works for up to
an hour but usually within 10-15 minutes it just stop/reset.
Presumably networkmanager restarts it.
Tested using 3.16 & 3.17.
Device:
03:00.0 Network controller: Ralink corp. RT3290 Wireless 802.11n 1T/1R PCIe
Subsystem: Hewlett-Packard Company Ralink RT3290LE 802.11bgn 1x1
Wi-Fi and Bluetooth 4.0 Combo Adapter
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort-
<TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 0, Cache Line Size: 64 bytes
Interrupt: pin A routed to IRQ 18
Region 0: Memory at b0410000 (32-bit, non-prefetchable) [size=64K]
Capabilities: <access denied>
Kernel driver in use: rt2800pci
Getting the following messsages in syslog:
Oct 10 17:45:36 brutus kernel: [ 4.937845] ieee80211 phy0:
rt2x00_set_rt: Info - RT chipset 3290, rev 0015 detected
Oct 10 17:45:36 brutus kernel: [ 4.941563] ieee80211 phy0:
rt2x00_set_rf: Info - RF chipset 3290 detected
Oct 10 17:45:36 brutus kernel: [ 4.983868] ieee80211 phy0: Selected
rate control algorithm 'minstrel_ht'
Oct 10 17:45:37 brutus NetworkManager[877]: <info> rfkill0: found WiFi
radio killswitch (at
/sys/devices/pci0000:00/0000:00:1c.2/0000:03:00.0/ieee80211/phy0/rfkill0)
(driver rt2800pci)
Oct 10 17:45:37 brutus kernel: [ 5.610165] ieee80211 phy0:
rt2x00lib_request_firmware: Info - Loading firmware file 'rt3290.bin'
Oct 10 17:45:37 brutus kernel: [ 5.611453] ieee80211 phy0:
rt2x00lib_request_firmware: Info - Firmware detected - version: 0.37
Oct 10 17:46:01 brutus kernel: [ 29.621979] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:46:01 brutus kernel: [ 30.070125] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 0 failed to flush
Oct 10 17:46:01 brutus kernel: [ 30.230153] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:46:02 brutus kernel: [ 30.678194] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:46:02 brutus kernel: [ 31.190166] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 0 failed to flush
Oct 10 17:46:02 brutus kernel: [ 31.350260] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:46:03 brutus kernel: [ 31.798309] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:46:03 brutus kernel: [ 32.246352] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 0 failed to flush
Oct 10 17:46:03 brutus kernel: [ 32.406369] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:46:04 brutus kernel: [ 32.918438] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:49:01 brutus kernel: [ 210.316739] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:49:02 brutus kernel: [ 210.764804] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:49:02 brutus kernel: [ 211.276854] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:49:03 brutus kernel: [ 211.724905] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:50:04 brutus kernel: [ 272.647105] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:51:27 brutus kernel: [ 355.999778] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:51:28 brutus kernel: [ 357.183902] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:53:11 brutus kernel: [ 460.055960] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:57:10 brutus kernel: [ 699.528350] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 0 failed to flush
Oct 10 17:57:11 brutus kernel: [ 699.688416] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 17:57:11 brutus kernel: [ 700.136713] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 18:01:10 brutus kernel: [ 939.336908] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Oct 10 18:01:11 brutus kernel: [ 939.801190] ieee80211 phy0:
rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
Help much appreciated. Thnx in advance.
- vin
^ permalink raw reply
* [PATCHv2 1/1] ip-link: add switch to show human readable output
From: Christian Hesse @ 2014-10-10 22:27 UTC (permalink / raw)
To: netdev; +Cc: Christian Hesse
Byte and packet count can increase to really big numbers. This adds a
switch to show human readable output.
% ip -s link ls wl
4: wl: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DORMANT group default qlen 1000
link/ether 00:de:ad:be:ee:ef brd ff:ff:ff:ff:ff:ff
RX: bytes packets errors dropped overrun mcast
113570876 156975 0 0 0 0
TX: bytes packets errors dropped carrier collsns
27290790 94313 0 0 0 0
% ip -s -h link ls wl
4: wl: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DORMANT group default qlen 1000
link/ether 00:de:ad:be:ee:ef brd ff:ff:ff:ff:ff:ff
RX: bytes packets errors dropped overrun mcast
122368888 169840 0 0 0 0
116.7Mi 165.8Ki 0 0 0 0
TX: bytes packets errors dropped carrier collsns
29087507 102309 0 0 0 0
27.7Mi 99.9Ki 0 0 0 0
---
include/utils.h | 1 +
ip/ip.c | 5 ++
ip/ipaddress.c | 143 ++++++++++++++++++++++++++++++++++++++++++++++++--
man/man8/ip-link.8.in | 1 +
4 files changed, 146 insertions(+), 4 deletions(-)
diff --git a/include/utils.h b/include/utils.h
index 704dc51..7bb19e9 100644
--- a/include/utils.h
+++ b/include/utils.h
@@ -11,6 +11,7 @@
#include "rtm_map.h"
extern int preferred_family;
+extern int human_readable;
extern int show_stats;
extern int show_details;
extern int show_raw;
diff --git a/ip/ip.c b/ip/ip.c
index 739b88d..6b352c8 100644
--- a/ip/ip.c
+++ b/ip/ip.c
@@ -24,6 +24,7 @@
#include "ip_common.h"
int preferred_family = AF_UNSPEC;
+int human_readable = 0;
int show_stats = 0;
int show_details = 0;
int resolve_hosts = 0;
@@ -47,6 +48,7 @@ static void usage(void)
" tunnel | tuntap | maddr | mroute | mrule | monitor | xfrm |\n"
" netns | l2tp | tcp_metrics | token | netconf }\n"
" OPTIONS := { -V[ersion] | -s[tatistics] | -d[etails] | -r[esolve] |\n"
+" -h[uman-readable] |\n"
" -f[amily] { inet | inet6 | ipx | dnet | bridge | link } |\n"
" -4 | -6 | -I | -D | -B | -0 |\n"
" -l[oops] { maximum-addr-flush-attempts } |\n"
@@ -212,6 +214,9 @@ int main(int argc, char **argv)
preferred_family = AF_DECnet;
} else if (strcmp(opt, "-B") == 0) {
preferred_family = AF_BRIDGE;
+ } else if (matches(opt, "-human") == 0 ||
+ matches(opt, "-human-readable") == 0) {
+ ++human_readable;
} else if (matches(opt, "-stats") == 0 ||
matches(opt, "-statistics") == 0) {
++show_stats;
diff --git a/ip/ipaddress.c b/ip/ipaddress.c
index 45729d8..d625434 100644
--- a/ip/ipaddress.c
+++ b/ip/ipaddress.c
@@ -319,6 +319,54 @@ static void print_vfinfo(FILE *fp, struct rtattr *vfinfo)
}
}
+static void print_human64(FILE *fp, int length, uint64_t count)
+{
+ int written;
+
+ if (count > 1125899906842624) /* 2**50 */
+ written = fprintf(fp, "%"PRIu64".%"PRIu64"Pi",
+ count / 1125899906842624,
+ count * 10 / 1125899906842624 % 10);
+ else if (count > 1099511627776) /* 2**40 */
+ written = fprintf(fp, "%"PRIu64".%"PRIu64"Ti",
+ count / 1099511627776,
+ count * 10 / 1099511627776 % 10);
+ else if (count > 1073741824) /* 2**30 */
+ written = fprintf(fp, "%"PRIu64".%"PRIu64"Gi",
+ count / 1073741824, count * 10 / 1073741824 % 10);
+ else if (count > 1048576) /* 2**20 */
+ written = fprintf(fp, "%"PRIu64".%"PRIu64"Mi",
+ count / 1048576, count * 10 / 1048576 % 10);
+ else if (count > 1024) /* 2**10 */
+ written = fprintf(fp, "%"PRIu64".%"PRIu64"Ki",
+ count / 1024, count * 10 / 1024 % 10);
+ else
+ written = fprintf(fp, "%"PRIu64, count);
+
+ while(written++ <= length)
+ fputc(' ', fp);
+}
+
+static void print_human32(FILE *fp, int length, uint32_t count)
+{
+ int written;
+
+ if (count > 1073741824) /* 2**30 */
+ written = fprintf(fp, "%u.%uGi",
+ count / 1073741824, count * 10 / 1073741824 % 10);
+ else if (count > 1048576) /* 2**20 */
+ written = fprintf(fp, "%u.%uMi",
+ count / 1048576, count * 10 / 1048576 % 10);
+ else if (count > 1024) /* 2**10 */
+ written = fprintf(fp, "%u.%uKi",
+ count / 1024, count * 10 / 1024 % 10);
+ else
+ written = fprintf(fp, "%u", count);
+
+ while(written++ <= length)
+ fputc(' ', fp);
+}
+
static void print_link_stats64(FILE *fp, const struct rtnl_link_stats64 *s,
const struct rtattr *carrier_changes)
{
@@ -334,15 +382,36 @@ static void print_link_stats64(FILE *fp, const struct rtnl_link_stats64 *s,
if (s->rx_compressed)
fprintf(fp, " %-7"PRIu64"",
(uint64_t)s->rx_compressed);
+ if (human_readable) {
+ fprintf(fp, "%s", _SL_);
+ fprintf(fp, " ");
+ print_human64(fp, 10, (uint64_t)s->rx_bytes);
+ print_human64(fp, 8, (uint64_t)s->rx_packets);
+ print_human64(fp, 7, (uint64_t)s->rx_errors);
+ print_human64(fp, 7, (uint64_t)s->rx_dropped);
+ print_human64(fp, 7, (uint64_t)s->rx_over_errors);
+ print_human64(fp, 7, (uint64_t)s->multicast);
+ if (s->rx_compressed)
+ print_human64(fp, 7, (uint64_t)s->rx_compressed);
+ }
if (show_stats > 1) {
fprintf(fp, "%s", _SL_);
fprintf(fp, " RX errors: length crc frame fifo missed%s", _SL_);
- fprintf(fp, " %-7"PRIu64" %-7"PRIu64" %-7"PRIu64" %-7"PRIu64" %-7"PRIu64"",
+ fprintf(fp, " %-8"PRIu64" %-7"PRIu64" %-7"PRIu64" %-7"PRIu64" %-7"PRIu64"",
(uint64_t)s->rx_length_errors,
(uint64_t)s->rx_crc_errors,
(uint64_t)s->rx_frame_errors,
(uint64_t)s->rx_fifo_errors,
(uint64_t)s->rx_missed_errors);
+ if (human_readable) {
+ fprintf(fp, "%s", _SL_);
+ fprintf(fp, " ");
+ print_human64(fp, 8, (uint64_t)s->rx_length_errors);
+ print_human64(fp, 7, (uint64_t)s->rx_crc_errors);
+ print_human64(fp, 7, (uint64_t)s->rx_frame_errors);
+ print_human64(fp, 7, (uint64_t)s->rx_fifo_errors);
+ print_human64(fp, 7, (uint64_t)s->rx_missed_errors);
+ }
}
fprintf(fp, "%s", _SL_);
fprintf(fp, " TX: bytes packets errors dropped carrier collsns %s%s",
@@ -357,13 +426,25 @@ static void print_link_stats64(FILE *fp, const struct rtnl_link_stats64 *s,
if (s->tx_compressed)
fprintf(fp, " %-7"PRIu64"",
(uint64_t)s->tx_compressed);
+ if (human_readable) {
+ fprintf(fp, "%s", _SL_);
+ fprintf(fp, " ");
+ print_human64(fp, 10, (uint64_t)s->tx_bytes);
+ print_human64(fp, 8, (uint64_t)s->tx_packets);
+ print_human64(fp, 7, (uint64_t)s->tx_errors);
+ print_human64(fp, 7, (uint64_t)s->tx_dropped);
+ print_human64(fp, 7, (uint64_t)s->tx_carrier_errors);
+ print_human64(fp, 7, (uint64_t)s->collisions);
+ if (s->tx_compressed)
+ print_human64(fp, 7, (uint64_t)s->tx_compressed);
+ }
if (show_stats > 1) {
fprintf(fp, "%s", _SL_);
fprintf(fp, " TX errors: aborted fifo window heartbeat");
if (carrier_changes)
fprintf(fp, " transns");
fprintf(fp, "%s", _SL_);
- fprintf(fp, " %-7"PRIu64" %-7"PRIu64" %-7"PRIu64" %-8"PRIu64"",
+ fprintf(fp, " %-8"PRIu64" %-7"PRIu64" %-7"PRIu64" %-8"PRIu64"",
(uint64_t)s->tx_aborted_errors,
(uint64_t)s->tx_fifo_errors,
(uint64_t)s->tx_window_errors,
@@ -371,6 +452,17 @@ static void print_link_stats64(FILE *fp, const struct rtnl_link_stats64 *s,
if (carrier_changes)
fprintf(fp, " %-7u",
*(uint32_t*)RTA_DATA(carrier_changes));
+ if (human_readable) {
+ fprintf(fp, "%s", _SL_);
+ fprintf(fp, " ");
+ print_human64(fp, 8, (uint64_t)s->tx_aborted_errors);
+ print_human64(fp, 7, (uint64_t)s->tx_fifo_errors);
+ print_human64(fp, 7, (uint64_t)s->tx_window_errors);
+ print_human64(fp, 7, (uint64_t)s->tx_heartbeat_errors);
+ if (carrier_changes)
+ print_human64(fp, 7, (uint64_t)*(uint32_t*)RTA_DATA(carrier_changes));
+ }
+
}
}
@@ -386,16 +478,37 @@ static void print_link_stats32(FILE *fp, const struct rtnl_link_stats *s,
);
if (s->rx_compressed)
fprintf(fp, " %-7u", s->rx_compressed);
+ if (human_readable) {
+ fprintf(fp, "%s", _SL_);
+ fprintf(fp, " ");
+ print_human32(fp, 10, s->rx_bytes);
+ print_human32(fp, 8, s->rx_packets);
+ print_human32(fp, 7, s->rx_errors);
+ print_human32(fp, 7, s->rx_dropped);
+ print_human32(fp, 7, s->rx_over_errors);
+ print_human32(fp, 7, s->multicast);
+ if (s->rx_compressed)
+ print_human32(fp, 7, s->rx_compressed);
+ }
if (show_stats > 1) {
fprintf(fp, "%s", _SL_);
fprintf(fp, " RX errors: length crc frame fifo missed%s", _SL_);
- fprintf(fp, " %-7u %-7u %-7u %-7u %-7u",
+ fprintf(fp, " %-8u %-7u %-7u %-7u %-7u",
s->rx_length_errors,
s->rx_crc_errors,
s->rx_frame_errors,
s->rx_fifo_errors,
s->rx_missed_errors
);
+ if (human_readable) {
+ fprintf(fp, "%s", _SL_);
+ fprintf(fp, " ");
+ print_human32(fp, 8, s->rx_length_errors);
+ print_human32(fp, 7, s->rx_crc_errors);
+ print_human32(fp, 7, s->rx_frame_errors);
+ print_human32(fp, 7, s->rx_fifo_errors);
+ print_human32(fp, 7, s->rx_missed_errors);
+ }
}
fprintf(fp, "%s", _SL_);
fprintf(fp, " TX: bytes packets errors dropped carrier collsns %s%s",
@@ -405,13 +518,25 @@ static void print_link_stats32(FILE *fp, const struct rtnl_link_stats *s,
s->tx_dropped, s->tx_carrier_errors, s->collisions);
if (s->tx_compressed)
fprintf(fp, " %-7u", s->tx_compressed);
+ if (human_readable) {
+ fprintf(fp, "%s", _SL_);
+ fprintf(fp, " ");
+ print_human32(fp, 10, s->tx_bytes);
+ print_human32(fp, 8, s->tx_packets);
+ print_human32(fp, 7, s->tx_errors);
+ print_human32(fp, 7, s->tx_dropped);
+ print_human32(fp, 7, s->tx_carrier_errors);
+ print_human32(fp, 7, s->collisions);
+ if (s->tx_compressed)
+ print_human32(fp, 7, s->tx_compressed);
+ }
if (show_stats > 1) {
fprintf(fp, "%s", _SL_);
fprintf(fp, " TX errors: aborted fifo window heartbeat");
if (carrier_changes)
fprintf(fp, " transns");
fprintf(fp, "%s", _SL_);
- fprintf(fp, " %-7u %-7u %-7u %-8u",
+ fprintf(fp, " %-8u %-7u %-7u %-8u",
s->tx_aborted_errors,
s->tx_fifo_errors,
s->tx_window_errors,
@@ -420,6 +545,16 @@ static void print_link_stats32(FILE *fp, const struct rtnl_link_stats *s,
if (carrier_changes)
fprintf(fp, " %-7u",
*(uint32_t*)RTA_DATA(carrier_changes));
+ if (human_readable) {
+ fprintf(fp, "%s", _SL_);
+ fprintf(fp, " ");
+ print_human32(fp, 8, s->tx_aborted_errors);
+ print_human32(fp, 7, s->tx_fifo_errors);
+ print_human32(fp, 7, s->tx_window_errors);
+ print_human32(fp, 7, s->tx_heartbeat_errors);
+ if (carrier_changes)
+ print_human32(fp, 7, *(uint32_t*)RTA_DATA(carrier_changes));
+ }
}
}
diff --git a/man/man8/ip-link.8.in b/man/man8/ip-link.8.in
index 383917a..9c20dd0 100644
--- a/man/man8/ip-link.8.in
+++ b/man/man8/ip-link.8.in
@@ -16,6 +16,7 @@ ip-link \- network device configuration
.ti -8
.IR OPTIONS " := { "
\fB\-V\fR[\fIersion\fR] |
+\fB\-h\fR[\fIuman-readable\fR] |
\fB\-s\fR[\fItatistics\fR] |
\fB\-r\fR[\fIesolve\fR] |
\fB\-f\fR[\fIamily\fR] {
--
2.1.2
^ permalink raw reply related
* Re: [PATCH net 1/3] net: sctp: fix skb_over_panic when receiving malformed ASCONF chunks
From: Daniel Borkmann @ 2014-10-10 22:06 UTC (permalink / raw)
To: Joshua Kinard; +Cc: davem, linux-sctp, netdev, Vlad Yasevich
In-Reply-To: <5437AF27.2030506@gentoo.org>
On 10/10/2014 12:04 PM, Joshua Kinard wrote:
...
> If I am reading correctly, this crash can only be triggered by actually getting
> through the SCTP handshake, then sending this specially-crafted ASCONF chunk?
> Meaning a blind nmap scan using this tactic against a random netblock wouldn't
> just randomly knock servers offline? This would seem to reduce the attack
> surface a quite bit by requiring the remote endpoint to actually respond.
Sorry, have been on travel almost whole day ... yes, handshake has to be
completed before that. So a scan/probe would need to establish a connection
first and ASCONF would need to be supported.
> Is there a CVE # for this?
CVE-2014-3673
^ permalink raw reply
* Re: [PATCH net 2/3] net: sctp: fix panic on duplicate ASCONF chunks
From: Daniel Borkmann @ 2014-10-10 22:02 UTC (permalink / raw)
To: Neil Horman; +Cc: davem, linux-sctp, netdev, Vlad Yasevich
In-Reply-To: <20141010153954.GE19499@hmsreliant.think-freely.org>
On 10/10/2014 05:39 PM, Neil Horman wrote:
...
> Is it worth adding a WARN_ON, to indicate that two ASCONF chunks have been
> received with duplicate serials?
Don't think so, as this would be triggerable from outside.
Thanks,
Daniel
^ permalink raw reply
* Re: [PATCH v1 2/3] drivers: net: xgene: Add SGMII based 1GbE support
From: Francois Romieu @ 2014-10-10 22:01 UTC (permalink / raw)
To: Iyappan Subramanian
Cc: davem, netdev, devicetree, linux-arm-kernel, patches, kchudgar
In-Reply-To: <1412972316-16344-3-git-send-email-isubramanian@apm.com>
Iyappan Subramanian <isubramanian@apm.com> :
[...]
> diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
> index c8f3824..63ea194 100644
> --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
> +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
> @@ -410,7 +410,6 @@ static void xgene_gmac_set_mac_addr(struct xgene_enet_pdata *pdata)
> addr0 = (dev_addr[3] << 24) | (dev_addr[2] << 16) |
> (dev_addr[1] << 8) | dev_addr[0];
> addr1 = (dev_addr[5] << 24) | (dev_addr[4] << 16);
> - addr1 |= pdata->phy_addr & 0xFFFF;
phy_addr removal is harmless (all zeroes from netdev priv data) but it's
unrelated to $SUBJECT.
You may split this patch as:
1. prettyfication / cruft removal
2. add link_state in xgene_mac_ops / pdata->rm shuffle
3. SGMII based 1GbE support
Mostly stylistic review below.
[...]
> diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
> index 15ec426..dc024c1 100644
> --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
> +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
[...]
> @@ -179,7 +180,6 @@ enum xgene_enet_rm {
> #define TUND_ADDR 0x4a
>
> #define TSO_IPPROTO_TCP 1
> -#define FULL_DUPLEX 2
See above.
>
> #define USERINFO_POS 0
> #define USERINFO_LEN 32
[...]
> diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c
> new file mode 100644
> index 0000000..6038596
> --- /dev/null
> +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c
[...]
> +static void xgene_enet_wr_csr(struct xgene_enet_pdata *pdata,
> + u32 offset, u32 val)
> +{
> + void __iomem *addr = pdata->eth_csr_addr + offset;
> +
> + iowrite32(val, addr);
> +}
Replace 'pdata' with one of 'xp', 'pd', 'p' ?
You should be able to pack a lot.
static void xgene_enet_wr_csr(struct xgene_enet_pdata *p, u32 offset, u32 val)
{
iowrite32(val, p->eth_csr_addr + offset);
}
There are several of those.
[...]
> +static bool xgene_enet_wr_indirect(void __iomem *addr, void __iomem *wr,
> + void __iomem *cmd, void __iomem *cmd_done,
> + u32 wr_addr, u32 wr_data)
> +{
> + u32 done;
> + u8 wait = 10;
> +
> + iowrite32(wr_addr, addr);
> + iowrite32(wr_data, wr);
> + iowrite32(XGENE_ENET_WR_CMD, cmd);
> +
> + /* wait for write command to complete */
> + while (!(done = ioread32(cmd_done)) && wait--)
> + udelay(1);
> +
> + if (!done)
> + return false;
int i;
for (i = 0; i < 10; i++) {
if (ioread32(cmd_done)) {
iowrite32(0, cmd);
return true;
}
udelay(1);
}
return false;
> +
> + iowrite32(0, cmd);
> +
> + return true;
> +}
> +
> +static void xgene_enet_wr_mac(struct xgene_enet_pdata *pdata,
> + u32 wr_addr, u32 wr_data)
> +{
> + void __iomem *addr, *wr, *cmd, *cmd_done;
> +
> + addr = pdata->mcx_mac_addr + MAC_ADDR_REG_OFFSET;
> + wr = pdata->mcx_mac_addr + MAC_WRITE_REG_OFFSET;
> + cmd = pdata->mcx_mac_addr + MAC_COMMAND_REG_OFFSET;
> + cmd_done = pdata->mcx_mac_addr + MAC_COMMAND_DONE_REG_OFFSET;
struct xgene_indirect_ctl {
void __iomem *addr;
void __iomem *ctl;
void __iomem *cmd;
void __iomem *cmd_done;
};
static void xgene_enet_wr_mac(struct xgene_enet_pdata *p, u32 addr, u32 data)
{
struct xgene_indirect_ctl ctl = {
.addr = p->mcx_mac_addr + MAC_ADDR_REG_OFFSET;
.ctl = p->mcx_mac_addr + MAC_WRITE_REG_OFFSET;
.cmd = p->mcx_mac_addr + MAC_COMMAND_REG_OFFSET;
.cmd_done = p->mcx_mac_addr + MAC_COMMAND_DONE_REG_OFFSET;
};
if (!xgene_enet_wr_indirect(&ctl, wr_addr, wr_data)) {
...
It's syntaxic sugar that avoids (an excess of) 'void *' parameters.
You could reuse it for xgene_enet_rd_mac.
> +
> + if (!xgene_enet_wr_indirect(addr, wr, cmd, cmd_done, wr_addr, wr_data))
> + netdev_err(pdata->ndev, "MCX mac write failed, addr: %04x\n",
> + wr_addr);
> +}
> +
> +static void xgene_enet_rd_csr(struct xgene_enet_pdata *pdata,
> + u32 offset, u32 *val)
> +{
> + void __iomem *addr = pdata->eth_csr_addr + offset;
> +
> + *val = ioread32(addr);
> +}
static u32 xgene_enet_rd_csr(struct xgene_enet_pdata *pdata, u32 offset)
{
return ioread32(pdata->eth_csr_addr + offset);
}
> +
> +static void xgene_enet_rd_diag_csr(struct xgene_enet_pdata *pdata,
> + u32 offset, u32 *val)
> +{
> + void __iomem *addr = pdata->eth_diag_csr_addr + offset;
> +
> + *val = ioread32(addr);
> +}
> +
> +static bool xgene_enet_rd_indirect(void __iomem *addr, void __iomem *rd,
> + void __iomem *cmd, void __iomem *cmd_done,
> + u32 rd_addr, u32 *rd_data)
> +{
> + u32 done;
> + u8 wait = 10;
> +
> + iowrite32(rd_addr, addr);
> + iowrite32(XGENE_ENET_RD_CMD, cmd);
> +
> + /* wait for read command to complete */
> + while (!(done = ioread32(cmd_done)) && wait--)
> + udelay(1);
> +
> + if (!done)
> + return false;
See above.
[...]
> +static void xgene_sgmac_rx_enable(struct xgene_enet_pdata *pdata)
> +{
> + u32 data;
> +
> + xgene_enet_rd_mac(pdata, MAC_CONFIG_1_ADDR, &data);
> + xgene_enet_wr_mac(pdata, MAC_CONFIG_1_ADDR, data | RX_EN);
> +}
> +
> +static void xgene_sgmac_tx_enable(struct xgene_enet_pdata *pdata)
> +{
> + u32 data;
> +
> + xgene_enet_rd_mac(pdata, MAC_CONFIG_1_ADDR, &data);
> + xgene_enet_wr_mac(pdata, MAC_CONFIG_1_ADDR, data | TX_EN);
> +}
static void _xgene_sgmac_rxtx(struct xgene_enet_pdata *p, bool set, u32 bits)
{
u32 data;
xgene_enet_rd_mac(pdata, MAC_CONFIG_1_ADDR, &data);
if (set)
data |= bits;
else
data &= ~bits;
xgene_enet_wr_mac(pdata, MAC_CONFIG_1_ADDR, data);
}
(or _xgene_sgmac_rxtx(struct xgene_enet_pdata *p, u32 set, u32 clear))
static void xgene_sgmac_rxtx_set(struct xgene_enet_pdata *p, u32 bits)
{
_xgene_sgmac_rxtx(p, true, bits);
}
static void xgene_sgmac_rx_enable(struct xgene_enet_pdata *p)
{
xgene_sgmac_rxtx_set(p, RX_EN);
}
static void xgene_sgmac_tx_enable(struct xgene_enet_pdata *p)
{
xgene_sgmac_rxtx_set(p, TX_EN);
}
static void xgene_sgmac_rxtx_clear(struct xgene_enet_pdata *p, u32 bits)
{
_xgene_sgmac_rxtx(p, false, bits);
}
etc.
[...]
> +struct xgene_mac_ops xgene_sgmac_ops = {
> + .init = xgene_sgmac_init,
> + .reset = xgene_sgmac_reset,
> + .rx_enable = xgene_sgmac_rx_enable,
> + .tx_enable = xgene_sgmac_tx_enable,
> + .rx_disable = xgene_sgmac_rx_disable,
> + .tx_disable = xgene_sgmac_tx_disable,
> + .set_mac_addr = xgene_sgmac_set_mac_addr,
> + .link_state = xgene_enet_link_state
Please use tabs before '='.
> +};
> +
> +struct xgene_port_ops xgene_sgport_ops = {
> + .reset = xgene_enet_reset,
> + .cle_bypass = xgene_enet_cle_bypass,
> + .shutdown = xgene_enet_shutdown,
See above.
[...]
> diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h
> new file mode 100644
> index 0000000..de43246
> --- /dev/null
> +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h
[...]
> +#define PHY_ADDR(src) (((src)<<8) & GENMASK(12, 8))
#define PHY_ADDR(src) (((src) << 8) & GENMASK(12, 8))
--
Ueimor
^ permalink raw reply
* Re: [RFC v2 3/6] kthread: warn on kill signal if not OOM
From: Anatol Pomozov @ 2014-10-10 21:54 UTC (permalink / raw)
To: Luis R. Rodriguez
Cc: One Thousand Gnomes, Takashi Iwai, Kay Sievers, Sreekanth Reddy,
Praveen Krishnamoorthy, hare, Nagalakshmi Nandigama, Wu Zhangjin,
Tetsuo Handa, mpt-fusionlinux.pdl, Tim Gardner, Benjamin Poirier,
Santosh Rastapur, Casey Leedom, Hariprasad S, Pierre Fersing,
Arjan van de Ven, Andrew Morton, Abhijit Mahajan,
systemd Mailing List, Lin
In-Reply-To: <CAB=NE6XS9pWmfO7xjTpLFG1OdOJ=UktbOK5pjJCK8hzfXagfXQ@mail.gmail.com>
Hi
On Fri, Sep 12, 2014 at 1:09 PM, Luis R. Rodriguez
<mcgrof@do-not-panic.com> wrote:
> On Thu, Sep 11, 2014 at 10:48 PM, Tom Gundersen <teg@jklm.no> wrote:
>> On Fri, Sep 12, 2014 at 12:26 AM, Luis R. Rodriguez
>> <mcgrof@do-not-panic.com> wrote:
>>> On Thu, Sep 11, 2014 at 2:43 PM, Tom Gundersen <teg@jklm.no> wrote:
>>>> How about simply introducing a new flag to finit_module() to indicate
>>>> that the caller does not care about asynchronicity. We could then pass
>>>> this from udev, but existing scripts calling modprobe/insmod will not
>>>> be affected.
>>>
>>> Do you mean that you *do want asynchronicity*?
>>
>> Precisely, udev would opt-in, but existing scripts etc would not.
>
> Sure that's the other alternative that Tejun was mentioning.
>
>>>> But isn't finit_module() taking a long time a serious problem given
>>>> that it means no other module can be loaded in parallel?
>>>
>>> Indeed but having a desire to make the init() complete fast is
>>> different than the desire to have the combination of both init and
>>> probe fast synchronously.
>>
>> I guess no one is arguing that probe should somehow be required to be
>> fast, but rather:
>>
>>> If userspace wants init to be fast and let
>>> probe be async then userspace has no option but to deal with the fact
>>> that async probe will be async, and it should then use other methods
>>> to match any dependencies if its doing that itself.
>>
>> Correct. And this therefore likely needs to be opt-in behaviour per
>> finit_module() invocation to avoid breaking old assumptions.
>
> Sure.
>
>>> For example
>>> networking should not kick off after a network driver is loaded but
>>> rather one the device creeps up on udev. We should be good with
>>> networking dealing with this correctly today but not sure about other
>>> subsystems. depmod should be able to load the required modules in
>>> order and if bus drivers work right then probe of the remnant devices
>>> should happen asynchronously. The one case I can think of that is a
>>> bit different is modules-load.d things but those *do not rely on the
>>> timeout*, but are loaded prior to a service requirement. Note though
>>> that if those modules had probe and they then run async'd then systemd
>>> service would probably need to consider that the requirements may not
>>> be there until later. If this is not carefully considered that could
>>> introduce regression to users of modules-load.d when async probe is
>>> fully deployed. The same applies to systemd making assumptions of kmod
>>> loading a module and a dependency being complete as probe would have
>>> run it before.
>>
>> Yeah, these all needs to be considered when deciding whether or not to
>> enable async in each specific case.
>
> Yes and come to think of it I'd recommend opting out of async
> functionality for modules-load.d given that it does *not* hooked with
> the timeout and there is a good chances its users likely do want to
> wait for probe to run at this point.
>
> Given this I also am inclined now for the per module request to be
> async or not (default) from userspace. The above would be a good
> example starting use case.
>
>>> I believe one concern here lies in on whether or not userspace
>>> is properly equipped to deal with the requirements on module loading
>>> doing async probing and that possibly failing. Perhaps systemd might
>>> think all userspace is ready for that but are we sure that's the case?
>>
>> There almost certainly are custom things out there relying on the
>> synchronous behaviour, but if we make it opt-in we should not have a
>> problem.
We recently discussed this "timeout module loading" issue in Arch IRC
and here are few more ideas:
1) Why not to make the timeout configurable through config file? There
is already udev.conf you can put config option there. Thus people with
modprobe issues can easily "fix" the problem. And then decrease
default timeout back to 30 seconds. I agree that long module loading
(more than 30 secs) is abnormal and should be investigated by driver
authors.
2) Could you add 'echo w > /proc/sysrq-trigger' to udev code right
before killing the "modprobe" thread? sysrq will print information
about stuck threads (including modprobe itself) this will make
debugging easier. e.g. dmesg here
https://bugs.archlinux.org/task/40454 says nothing where the threads
were stuck.
^ permalink raw reply
* [PATCH v2] fm10k: Add skb->xmit_more support
From: alexander.duyck @ 2014-10-10 21:30 UTC (permalink / raw)
To: e1000-devel, netdev, davem, jeffrey.t.kirsher; +Cc: matthew.vick
From: Alexander Duyck <alexander.h.duyck@redhat.com>
This change adds support for skb->xmit_more based on the changes that were
made to igb to support the feature. The main changes are moving up the
check for maybe_stop_tx so that we can check netif_xmit_stopped to determine
if we must write the tail because we can add no further buffers.
Acked-by: Matthew Vick <matthew.vick@intel.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com>
---
v2: Minor fix to patch description for typo.
drivers/net/ethernet/intel/fm10k/fm10k_main.c | 65 +++++++++++++------------
1 file changed, 34 insertions(+), 31 deletions(-)
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
index 6c800a3..8ad7ff4 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
@@ -930,6 +930,30 @@ static bool fm10k_tx_desc_push(struct fm10k_ring *tx_ring,
return i == tx_ring->count;
}
+static int __fm10k_maybe_stop_tx(struct fm10k_ring *tx_ring, u16 size)
+{
+ netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index);
+
+ smp_mb();
+
+ /* We need to check again in a case another CPU has just
+ * made room available. */
+ if (likely(fm10k_desc_unused(tx_ring) < size))
+ return -EBUSY;
+
+ /* A reprieve! - use start_queue because it doesn't call schedule */
+ netif_start_subqueue(tx_ring->netdev, tx_ring->queue_index);
+ ++tx_ring->tx_stats.restart_queue;
+ return 0;
+}
+
+static inline int fm10k_maybe_stop_tx(struct fm10k_ring *tx_ring, u16 size)
+{
+ if (likely(fm10k_desc_unused(tx_ring) >= size))
+ return 0;
+ return __fm10k_maybe_stop_tx(tx_ring, size);
+}
+
static void fm10k_tx_map(struct fm10k_ring *tx_ring,
struct fm10k_tx_buffer *first)
{
@@ -1023,13 +1047,18 @@ static void fm10k_tx_map(struct fm10k_ring *tx_ring,
tx_ring->next_to_use = i;
+ /* Make sure there is space in the ring for the next send. */
+ fm10k_maybe_stop_tx(tx_ring, DESC_NEEDED);
+
/* notify HW of packet */
- writel(i, tx_ring->tail);
+ if (netif_xmit_stopped(txring_txq(tx_ring)) || !skb->xmit_more) {
+ writel(i, tx_ring->tail);
- /* we need this if more than one processor can write to our tail
- * at a time, it synchronizes IO on IA64/Altix systems
- */
- mmiowb();
+ /* we need this if more than one processor can write to our tail
+ * at a time, it synchronizes IO on IA64/Altix systems
+ */
+ mmiowb();
+ }
return;
dma_error:
@@ -1049,30 +1078,6 @@ dma_error:
tx_ring->next_to_use = i;
}
-static int __fm10k_maybe_stop_tx(struct fm10k_ring *tx_ring, u16 size)
-{
- netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index);
-
- smp_mb();
-
- /* We need to check again in a case another CPU has just
- * made room available. */
- if (likely(fm10k_desc_unused(tx_ring) < size))
- return -EBUSY;
-
- /* A reprieve! - use start_queue because it doesn't call schedule */
- netif_start_subqueue(tx_ring->netdev, tx_ring->queue_index);
- ++tx_ring->tx_stats.restart_queue;
- return 0;
-}
-
-static inline int fm10k_maybe_stop_tx(struct fm10k_ring *tx_ring, u16 size)
-{
- if (likely(fm10k_desc_unused(tx_ring) >= size))
- return 0;
- return __fm10k_maybe_stop_tx(tx_ring, size);
-}
-
netdev_tx_t fm10k_xmit_frame_ring(struct sk_buff *skb,
struct fm10k_ring *tx_ring)
{
@@ -1117,8 +1122,6 @@ netdev_tx_t fm10k_xmit_frame_ring(struct sk_buff *skb,
fm10k_tx_map(tx_ring, first);
- fm10k_maybe_stop_tx(tx_ring, DESC_NEEDED);
-
return NETDEV_TX_OK;
out_drop:
^ permalink raw reply related
* Re: kernel crash in bpf_jit on x86_64 when running nmap
From: Alexei Starovoitov @ 2014-10-10 21:17 UTC (permalink / raw)
To: Darrick J. Wong
Cc: David S. Miller, linux-kernel, Daniel Borkmann, H. Peter Anvin,
Thomas Gleixner, Ingo Molnar, Network Development
In-Reply-To: <20141010204403.GA30018@birch.djwong.org>
On Fri, Oct 10, 2014 at 1:44 PM, Darrick J. Wong
<darrick.wong@oracle.com> wrote:
> Hi everyone,
>
> I was running nmap on a x86_64 qemu guest and experienced the following crash:
>
> # nmap -sS -O -vvv 192.168.122.1
> Starting Nmap 6.40 ( http://nmap.org ) at 2014-10-10 13:14 PDT
> Initiating ARP Ping Scan at 13:14
> Scanning 192.168.122.1 [1 port]
> <kaboom>
>
> dmesg output is as follows (I set net.core.bpf_jit_enable=2 the second time):
>
> [ 32.376291] flen=3 proglen=82 pass=0 image=ffffffffc01ac65b
> [ 32.377595] JIT code: 00000000: 55 48 89 e5 48 81 ec 28 02 00 00 48 89 9d d8 fd
> [ 32.379243] JIT code: 00000010: ff ff 4c 89 ad e0 fd ff ff 4c 89 b5 e8 fd ff ff
> [ 32.380984] JIT code: 00000020: 4c 89 bd f0 fd ff ff 31 c0 4d 31 ed 48 89 fb b8
> [ 32.382606] JIT code: 00000030: 00 00 00 00 48 8b 9d d8 fd ff ff 4c 8b ad e0 fd
> [ 32.384280] JIT code: 00000040: ff ff 4c 8b b5 e8 fd ff ff 4c 8b bd f0 fd ff ff
> [ 32.385911] JIT code: 00000050: c9 c3
> [ 32.386841] bpf_jit: proglen=265 != oldproglen=269
thanks for the report. The line above indicates that JIT tried to emit
4 byte longer program than during pre-final pass.
Will look at it asap.
^ permalink raw reply
* Re: [PATCH] fm10k: Add skb->xmit_more support
From: Vick, Matthew @ 2014-10-10 21:12 UTC (permalink / raw)
To: alexander.duyck@gmail.com, e1000-devel@lists.sourceforge.net,
netdev@vger.kernel.org, Kirsher, Jeffrey T
In-Reply-To: <20141010201714.20593.67813.stgit@ahduyck-workstation.home>
On 10/10/14, 1:17 PM, "alexander.duyck@gmail.com"
<alexander.duyck@gmail.com> wrote:
>From: Alexander Duyck <alexander.h.duyck@redhat.com>
>
>This change adds suport for skb->xmit_more based on the changes that were
>made to igb to support the feature. The main changes are moving up the
>check for maybe_stop_tx so that we can check netif_xmit_stopped to
>determine
>if we must write the tail because we can add no further buffers.
>
>Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com>
s/suport/support in the first line of the patch description, but other
than that this looks good to me. Thanks, Alex!
Acked-by: Matthew Vick <matthew.vick@intel.com>
^ permalink raw reply
* [PATCH net] skbuff: fix ftrace handling in skb_unshare
From: Alexander Aring @ 2014-10-10 21:10 UTC (permalink / raw)
To: netdev; +Cc: kernel, Alexander Aring
If the skb is not dropped afterwards we should run consume_skb instead
kfree_skb. Inside of function skb_unshare we do always a kfree_skb,
doesn't depend if skb_copy failed or was successful.
This patch switch this behaviour like skb_share_check, if allocation of
sk_buff failed we use kfree_skb otherwise consume_skb.
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
---
include/linux/skbuff.h | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index abde271..d150734 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1116,7 +1116,12 @@ static inline struct sk_buff *skb_unshare(struct sk_buff *skb,
might_sleep_if(pri & __GFP_WAIT);
if (skb_cloned(skb)) {
struct sk_buff *nskb = skb_copy(skb, pri);
- kfree_skb(skb); /* Free our shared copy */
+
+ /* Free our shared copy */
+ if (likely(nskb))
+ consume_skb(skb);
+ else
+ kfree_skb(skb);
skb = nskb;
}
return skb;
--
2.1.2
^ permalink raw reply related
* Re: [Nios2-dev] [PATCH net-next 0/2] Altera TSE with no PHY
From: Matthew Gerlach @ 2014-10-10 20:46 UTC (permalink / raw)
To: Walter Lozano, netdev@vger.kernel.org
Cc: f.fainelli@gmail.com, tobias.klauser@gmail.com,
vbridgers2013@gmail.com, nios2-dev@lists.rocketboards.org,
davem@davemloft.net, guido@vanguardiasur.com.ar
In-Reply-To: <1412359741-8423-1-git-send-email-walter@vanguardiasur.com.ar>
Hi Walter,
I've examined your patch, and it makes sense to me. I tested your patch again the TSE example design, and it did not break anything.
Thanks,
Matthew Gerlach
________________________________________
From: nios2-dev-bounces@lists.rocketboards.org <nios2-dev-bounces@lists.rocketboards.org> on behalf of Walter Lozano <walter@vanguardiasur.com.ar>
Sent: Friday, October 3, 2014 11:08 AM
To: netdev@vger.kernel.org
Cc: f.fainelli@gmail.com; tobias.klauser@gmail.com; vbridgers2013@gmail.com; nios2-dev@lists.rocketboards.org; davem@davemloft.net; guido@vanguardiasur.com.ar
Subject: [Nios2-dev] [PATCH net-next 0/2] Altera TSE with no PHY
In some scenarios there is no PHY chip present, for example in optical links.
This serie of patches moves PHY get addr and MDIO create to a new function and
avoids PHY and MDIO probing in these cases.
Walter Lozano (2):
Altera TSE: Move PHY get addr and MDIO create
Altera TSE: Add support for no PHY
drivers/net/ethernet/altera/altera_tse_main.c | 65 +++++++++++++++++--------
1 file changed, 44 insertions(+), 21 deletions(-)
--
1.7.10.4
_______________________________________________
Nios2-dev mailing list
Nios2-dev@lists.rocketboards.org
http://lists.rocketboards.org/cgi-bin/mailman/listinfo/nios2-dev
^ permalink raw reply
* Re: [PATCH iproute2] Changed action labels to [action ] format
From: Vadim Kochan @ 2014-10-10 20:59 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: netdev@vger.kernel.org
In-Reply-To: <20141010155648.1f1b51bf@uryu.home.lan>
But the scripts can be easy rewritten ?
On Fri, Oct 10, 2014 at 11:56 PM, Stephen Hemminger
<stephen@networkplumber.org> wrote:
> On Fri, 10 Oct 2014 19:51:56 +0300
> Vadim Kochan <vadim4j@gmail.com> wrote:
>
>> The reasons for this change were:
>> - different ip utils uses different labels for action: Deleted/delete
>> - make an action labels to be easy found in logs
>
> Fine in concept. But people do stupid things like parse the output of commands.
> I am loathe to make cosmetic change.
>
^ permalink raw reply
* Re: [PATCH iproute2] Changed action labels to [action ] format
From: Stephen Hemminger @ 2014-10-10 20:56 UTC (permalink / raw)
To: Vadim Kochan; +Cc: netdev
In-Reply-To: <1412959916-32169-1-git-send-email-vadim4j@gmail.com>
On Fri, 10 Oct 2014 19:51:56 +0300
Vadim Kochan <vadim4j@gmail.com> wrote:
> The reasons for this change were:
> - different ip utils uses different labels for action: Deleted/delete
> - make an action labels to be easy found in logs
Fine in concept. But people do stupid things like parse the output of commands.
I am loathe to make cosmetic change.
^ permalink raw reply
* kernel crash in bpf_jit on x86_64 when running nmap
From: Darrick J. Wong @ 2014-10-10 20:44 UTC (permalink / raw)
To: Alexei Starovoitov, David S. Miller
Cc: linux-kernel, Daniel Borkmann, H. Peter Anvin, Thomas Gleixner,
Ingo Molnar, netdev
Hi everyone,
I was running nmap on a x86_64 qemu guest and experienced the following crash:
# nmap -sS -O -vvv 192.168.122.1
Starting Nmap 6.40 ( http://nmap.org ) at 2014-10-10 13:14 PDT
Initiating ARP Ping Scan at 13:14
Scanning 192.168.122.1 [1 port]
<kaboom>
dmesg output is as follows (I set net.core.bpf_jit_enable=2 the second time):
[ 32.376291] flen=3 proglen=82 pass=0 image=ffffffffc01ac65b
[ 32.377595] JIT code: 00000000: 55 48 89 e5 48 81 ec 28 02 00 00 48 89 9d d8 fd
[ 32.379243] JIT code: 00000010: ff ff 4c 89 ad e0 fd ff ff 4c 89 b5 e8 fd ff ff
[ 32.380984] JIT code: 00000020: 4c 89 bd f0 fd ff ff 31 c0 4d 31 ed 48 89 fb b8
[ 32.382606] JIT code: 00000030: 00 00 00 00 48 8b 9d d8 fd ff ff 4c 8b ad e0 fd
[ 32.384280] JIT code: 00000040: ff ff 4c 8b b5 e8 fd ff ff 4c 8b bd f0 fd ff ff
[ 32.385911] JIT code: 00000050: c9 c3
[ 32.386841] bpf_jit: proglen=265 != oldproglen=269
[ 32.387936] flen=33 proglen=265 pass=0 image=ffffffffc01ae3a2
[ 32.389288] JIT code: 00000000: 55 48 89 e5 48 81 ec 28 02 00 00 48 89 9d d8 fd
[ 32.390916] JIT code: 00000010: ff ff 4c 89 ad e0 fd ff ff 4c 89 b5 e8 fd ff ff
[ 32.393150] JIT code: 00000020: 4c 89 bd f0 fd ff ff 31 c0 4d 31 ed 44 8b 4f 68
[ 32.394820] JIT code: 00000030: 44 2b 4f 6c 4c 8b 97 c8 00 00 00 48 89 fb be 0c
[ 32.397303] JIT code: 00000040: 00 00 00 e8 07 85 ec c0 48 81 f8 06 08 00 00 0f
[ 32.399712] JIT code: 00000050: 85 95 00 00 00 be 0c 00 00 00 e8 f0 84 ec c0 48
[ 32.402163] JIT code: 00000060: 81 f8 06 08 00 00 75 7e b8 12 00 00 00 89 45 c0
[ 32.404627] JIT code: 00000070: 44 8b 6d c0 4c 89 ee 83 c6 0e e8 ab 84 ec c0 89
[ 32.407027] JIT code: 00000080: 45 c4 b8 39 00 54 52 89 45 c8 44 8b 6d c8 8b 45
[ 32.409498] JIT code: 00000090: c4 44 29 e8 48 83 f8 00 75 4c be 0c 00 00 00 e8
[ 32.411916] JIT code: 000000a0: a7 84 ec c0 48 81 f8 06 08 00 00 75 39 b8 16 00
[ 32.414479] JIT code: 000000b0: 00 00 89 45 c8 44 8b 6d c8 4c 89 ee 83 c6 0e e8
[ 32.417109] JIT code: 000000c0: 7f 84 ec c0 89 45 cc b8 36 15 00 00 89 45 d0 44
[ 32.419497] JIT code: 000000d0: 8b 6d d0 8b 45 cc 44 29 e8 48 83 f8 00 75 07 b8
[ 32.426032] JIT code: 000000e0: 00 01 00 00 eb 05 b8 00 00 00 00 48 8b 9d d8 fd
[ 32.428445] JIT code: 000000f0: ff ff 4c 8b ad e0 fd ff ff 4c 8b b5 e8 fd ff ff
[ 32.430839] JIT code: 00000100: 4c 8b bd f0 fd ff ff c9 c3
[ 32.432562] BUG: unable to handle kernel NULL pointer dereference at 0000000000000012
[ 32.435275] IP: [<ffffffff810768cc>] efi_call+0xfc/0x100
[ 32.436464] PGD 1f6b59067 PUD 1f79b6067 PMD 0
[ 32.436464] Oops: 0002 [#1] PREEMPT SMP
[ 32.436464] Modules linked in: sch_fq_codel lpc_ich mfd_core fuse nfsd auth_rpcgss exportfs virtio_scsi af_packet
[ 32.436464] CPU: 0 PID: 3577 Comm: nmap Tainted: G W 3.17.0-mcsum #1
[ 32.436464] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS Bochs 01/01/2011
[ 32.436464] task: ffff8801f6c69840 ti: ffff8801f6dd4000 task.ti: ffff8801f6dd4000
[ 32.436464] RIP: 0010:[<ffffffff810768cc>] [<ffffffff810768cc>] efi_call+0xfc/0x100
[ 32.436464] RSP: 0018:ffff8801f6dd7860 EFLAGS: 00010212
[ 32.436464] RAX: 0000000000000012 RBX: ffff8801f76e2f00 RCX: ffffffff81a94d40
[ 32.436464] RDX: ffff88007b1aa080 RSI: 0000000000000020 RDI: ffff8801f76e2f00
[ 32.436464] RBP: ffff8801f6dd7a90 R08: 00000000000000cc R09: 000000000000002a
[ 32.436464] R10: ffff8801f796e420 R11: 000000000000002a R12: ffff8801f796e420
[ 32.436464] R13: 0000000000000012 R14: ffff8801f6981000 R15: 000000000000002a
[ 32.436464] FS: 00007fd30509c780(0000) GS:ffff8801ff600000(0000) knlGS:0000000000000000
[ 32.436464] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 32.436464] CR2: 0000000000000012 CR3: 00000001f6a43000 CR4: 00000000000006f0
[ 32.436464] Stack:
[ 32.436464] ffffffffc01ae421 ffff8801f76e2f00 ffff8801f6f5c000 ffff8801f6981000
[ 32.436464] 000000000000002a 00000000002284d0 0000000000000010 00000000ffffffff
[ 32.436464] 0000000000000000 0000010000000003 ffffffff81aa64c0 ffffffff81aa7248
[ 32.436464] Call Trace:
[ 32.436464] [<ffffffff8114bf1a>] ? __alloc_pages_nodemask+0x14a/0xa40
[ 32.436464] [<ffffffff8129a662>] ? put_dec+0x72/0x90
[ 32.436464] [<ffffffff8129b593>] ? number.isra.2+0x323/0x360
[ 32.436464] [<ffffffff810a8538>] ? __enqueue_entity+0x78/0x80
[ 32.436464] [<ffffffff810a9bc5>] ? set_next_entity+0x95/0xb0
[ 32.436464] [<ffffffff810af0d2>] ? pick_next_task_fair+0x722/0x880
[ 32.436464] [<ffffffff81001625>] ? __switch_to+0x165/0x5a0
[ 32.436464] [<ffffffff8103fa28>] ? lookup_address+0x28/0x30
[ 32.436464] [<ffffffff8103facb>] ? _lookup_address_cpa.isra.11+0x3b/0x40
[ 32.436464] [<ffffffffc013140f>] tpacket_rcv+0xdf/0x88d [af_packet]
[ 32.436464] [<ffffffff815a3b78>] ? __schedule+0x348/0x800
[ 32.436464] [<ffffffff81039e29>] ? kvm_clock_get_cycles+0x9/0x10
[ 32.436464] [<ffffffff8149e4a7>] dev_queue_xmit_nit+0x1b7/0x230
[ 32.436464] [<ffffffff814a07dc>] dev_hard_start_xmit+0x2fc/0x650
[ 32.436464] [<ffffffff814be62e>] sch_direct_xmit+0xee/0x1c0
[ 32.436464] [<ffffffff814a0d15>] __dev_queue_xmit+0x1e5/0x4f0
[ 32.436464] [<ffffffff814a1030>] dev_queue_xmit+0x10/0x20
[ 32.436464] [<ffffffffc0130b93>] packet_sendmsg+0xd33/0x1070 [af_packet]
[ 32.436464] [<ffffffff81485c1e>] sock_sendmsg+0x6e/0x90
[ 32.436464] [<ffffffff81485db1>] SYSC_sendto+0x121/0x1d0
[ 32.436464] [<ffffffff81039e07>] ? kvm_clock_read+0x27/0x40
[ 32.436464] [<ffffffff81039e29>] ? kvm_clock_get_cycles+0x9/0x10
[ 32.436464] [<ffffffff810df857>] ? __getnstimeofday64+0x37/0xd0
[ 32.436464] [<ffffffff810df8fe>] ? getnstimeofday64+0xe/0x30
[ 32.436464] [<ffffffff810df93a>] ? do_gettimeofday+0x1a/0x50
[ 32.436464] [<ffffffff8148684e>] SyS_sendto+0xe/0x10
[ 32.436464] [<ffffffff815a882d>] system_call_fastpath+0x1a/0x1f
[ 32.436464] Code: 0f 28 4c 24 50 0f 28 54 24 40 0f 28 5c 24 30 0f 28 64 24 20 0f 28 6c 24 10 48 8b 74 24 08 0f 22 c6 48 8b 24 24 c3 66 0f 1f 84 00 <00> 00 00 00 85 f6 0f 88 cc 00 00 00 44 89 c8 29 f0 83 f8 03 7e
[ 32.436464] RIP [<ffffffff810768cc>] efi_call+0xfc/0x100
[ 32.436464] RSP <ffff8801f6dd7860>
[ 32.436464] CR2: 0000000000000012
[ 32.436464] ---[ end trace c2167eb2b612f788 ]---
[ 32.436464] Kernel panic - not syncing: Fatal exception in interrupt
[ 32.436464] Kernel Offset: 0x0 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
[ 32.436464] ---[ end Kernel panic - not syncing: Fatal exception in interrupt
git bisect traced it back to commit 622582786c9e041d0bd52bde201787adeab249f8
("net: filter: x86: internal BPF JIT") in 3.15-rc4. Reverting it or setting
bpf_jit_enable=0 makes the crash go away and nmap runs to completion.
I'm not sure why %rip is efi_call, since the QEMU guest doesn't support UEFI.
I'm guessing the CPU is off in the weeds, possibly as some side effect of the
proglen mismatch. I can help debug, but I'm not a bpf_jit expert by any means.
:)
Thanks,
--Darrick
^ permalink raw reply
* [PATCH] net: can: esd_usb2: fix memory leak on disconnect
From: Alexey Khoroshilov @ 2014-10-10 20:31 UTC (permalink / raw)
To: Matthias Fuchs, Wolfgang Grandegger, Marc Kleine-Budde
Cc: Alexey Khoroshilov, linux-can, netdev, linux-kernel, ldv-project
It seems struct esd_usb2 dev is not deallocated on disconnect.
The patch adds the deallocation.
Found by Linux Driver Verification project (linuxtesting.org).
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
---
drivers/net/can/usb/esd_usb2.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/can/usb/esd_usb2.c b/drivers/net/can/usb/esd_usb2.c
index b7c9e8b11460..7a90075529c3 100644
--- a/drivers/net/can/usb/esd_usb2.c
+++ b/drivers/net/can/usb/esd_usb2.c
@@ -1143,6 +1143,7 @@ static void esd_usb2_disconnect(struct usb_interface *intf)
}
}
unlink_all_urbs(dev);
+ kfree(dev);
}
}
--
1.9.1
^ permalink raw reply related
* [PATCH v1 2/3] drivers: net: xgene: Add SGMII based 1GbE support
From: Iyappan Subramanian @ 2014-10-10 20:18 UTC (permalink / raw)
To: davem-fT/PcQaiUtIeIZ0/mPfg9Q, netdev-u79uwXL29TY76Z2rM5mHXA,
devicetree-u79uwXL29TY76Z2rM5mHXA
Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
patches-qTEPVZfXA3Y, kchudgar-qTEPVZfXA3Y, Iyappan Subramanian
In-Reply-To: <1412972316-16344-1-git-send-email-isubramanian-qTEPVZfXA3Y@public.gmane.org>
Signed-off-by: Iyappan Subramanian <isubramanian-qTEPVZfXA3Y@public.gmane.org>
Signed-off-by: Keyur Chudgar <kchudgar-qTEPVZfXA3Y@public.gmane.org>
---
drivers/net/ethernet/apm/xgene/Makefile | 2 +-
drivers/net/ethernet/apm/xgene/xgene_enet_hw.c | 1 -
drivers/net/ethernet/apm/xgene/xgene_enet_hw.h | 2 +-
drivers/net/ethernet/apm/xgene/xgene_enet_main.c | 24 +-
drivers/net/ethernet/apm/xgene/xgene_enet_main.h | 5 +-
drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c | 407 ++++++++++++++++++++++
drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h | 41 +++
drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c | 3 +-
drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h | 4 -
9 files changed, 473 insertions(+), 16 deletions(-)
create mode 100644 drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c
create mode 100644 drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h
diff --git a/drivers/net/ethernet/apm/xgene/Makefile b/drivers/net/ethernet/apm/xgene/Makefile
index 589b352..68be5655 100644
--- a/drivers/net/ethernet/apm/xgene/Makefile
+++ b/drivers/net/ethernet/apm/xgene/Makefile
@@ -2,6 +2,6 @@
# Makefile for APM X-Gene Ethernet Driver.
#
-xgene-enet-objs := xgene_enet_hw.o xgene_enet_xgmac.o \
+xgene-enet-objs := xgene_enet_hw.o xgene_enet_sgmac.o xgene_enet_xgmac.o \
xgene_enet_main.o xgene_enet_ethtool.o
obj-$(CONFIG_NET_XGENE) += xgene-enet.o
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
index c8f3824..63ea194 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
@@ -410,7 +410,6 @@ static void xgene_gmac_set_mac_addr(struct xgene_enet_pdata *pdata)
addr0 = (dev_addr[3] << 24) | (dev_addr[2] << 16) |
(dev_addr[1] << 8) | dev_addr[0];
addr1 = (dev_addr[5] << 24) | (dev_addr[4] << 16);
- addr1 |= pdata->phy_addr & 0xFFFF;
xgene_enet_wr_mcx_mac(pdata, STATION_ADDR0_ADDR, addr0);
xgene_enet_wr_mcx_mac(pdata, STATION_ADDR1_ADDR, addr1);
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
index 15ec426..dc024c1 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
@@ -44,6 +44,7 @@ static inline u32 xgene_get_bits(u32 val, u32 start, u32 end)
enum xgene_enet_rm {
RM0,
+ RM1,
RM3 = 3
};
@@ -179,7 +180,6 @@ enum xgene_enet_rm {
#define TUND_ADDR 0x4a
#define TSO_IPPROTO_TCP 1
-#define FULL_DUPLEX 2
#define USERINFO_POS 0
#define USERINFO_LEN 32
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
index 9b85239..743eccf 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
@@ -21,6 +21,7 @@
#include "xgene_enet_main.h"
#include "xgene_enet_hw.h"
+#include "xgene_enet_sgmac.h"
#include "xgene_enet_xgmac.h"
static void xgene_enet_init_bufpool(struct xgene_enet_desc_ring *buf_pool)
@@ -813,6 +814,7 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata)
return pdata->phy_mode;
}
if (pdata->phy_mode != PHY_INTERFACE_MODE_RGMII &&
+ pdata->phy_mode != PHY_INTERFACE_MODE_SGMII &&
pdata->phy_mode != PHY_INTERFACE_MODE_XGMII) {
dev_err(dev, "Incorrect phy-connection-type specified\n");
return -ENODEV;
@@ -830,14 +832,12 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata)
pdata->eth_csr_addr = base_addr + BLOCK_ETH_CSR_OFFSET;
pdata->eth_ring_if_addr = base_addr + BLOCK_ETH_RING_IF_OFFSET;
pdata->eth_diag_csr_addr = base_addr + BLOCK_ETH_DIAG_CSR_OFFSET;
- if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) {
- pdata->mcx_mac_addr = base_addr + BLOCK_ETH_MAC_OFFSET;
- pdata->mcx_mac_csr_addr = base_addr + BLOCK_ETH_MAC_CSR_OFFSET;
- pdata->rm = RM3;
- } else {
+ if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) {
pdata->mcx_mac_addr = base_addr + BLOCK_AXG_MAC_OFFSET;
pdata->mcx_mac_csr_addr = base_addr + BLOCK_AXG_MAC_CSR_OFFSET;
- pdata->rm = RM0;
+ } else {
+ pdata->mcx_mac_addr = base_addr + BLOCK_ETH_MAC_OFFSET;
+ pdata->mcx_mac_csr_addr = base_addr + BLOCK_ETH_MAC_CSR_OFFSET;
}
pdata->rx_buff_cnt = NUM_PKT_BUF;
@@ -881,10 +881,17 @@ static void xgene_enet_setup_ops(struct xgene_enet_pdata *pdata)
case PHY_INTERFACE_MODE_RGMII:
pdata->mac_ops = &xgene_gmac_ops;
pdata->port_ops = &xgene_gport_ops;
+ pdata->rm = RM3;
+ break;
+ case PHY_INTERFACE_MODE_SGMII:
+ pdata->mac_ops = &xgene_sgmac_ops;
+ pdata->port_ops = &xgene_sgport_ops;
+ pdata->rm = RM1;
break;
default:
pdata->mac_ops = &xgene_xgmac_ops;
pdata->port_ops = &xgene_xgport_ops;
+ pdata->rm = RM0;
break;
}
}
@@ -895,6 +902,7 @@ static int xgene_enet_probe(struct platform_device *pdev)
struct xgene_enet_pdata *pdata;
struct device *dev = &pdev->dev;
struct napi_struct *napi;
+ struct xgene_mac_ops *mac_ops;
int ret;
ndev = alloc_etherdev(sizeof(struct xgene_enet_pdata));
@@ -937,10 +945,12 @@ static int xgene_enet_probe(struct platform_device *pdev)
napi = &pdata->rx_ring->napi;
netif_napi_add(ndev, napi, xgene_enet_napi, NAPI_POLL_WEIGHT);
+
+ mac_ops = pdata->mac_ops;
if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII)
ret = xgene_enet_mdio_config(pdata);
else
- INIT_DELAYED_WORK(&pdata->link_work, xgene_enet_link_state);
+ INIT_DELAYED_WORK(&pdata->link_work, mac_ops->link_state);
return ret;
err:
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
index 86cf68b..1057181 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
@@ -39,6 +39,9 @@
#define NUM_PKT_BUF 64
#define NUM_BUFPOOL 32
+#define PHY_POLL_LINK_ON (10 * HZ)
+#define PHY_POLL_LINK_OFF (PHY_POLL_LINK_ON / 5)
+
/* software context of a descriptor ring */
struct xgene_enet_desc_ring {
struct net_device *ndev;
@@ -76,6 +79,7 @@ struct xgene_mac_ops {
void (*tx_disable)(struct xgene_enet_pdata *pdata);
void (*rx_disable)(struct xgene_enet_pdata *pdata);
void (*set_mac_addr)(struct xgene_enet_pdata *pdata);
+ void (*link_state)(struct work_struct *work);
};
struct xgene_port_ops {
@@ -109,7 +113,6 @@ struct xgene_enet_pdata {
void __iomem *base_addr;
void __iomem *ring_csr_addr;
void __iomem *ring_cmd_addr;
- u32 phy_addr;
int phy_mode;
enum xgene_enet_rm rm;
struct rtnl_link_stats64 stats;
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c
new file mode 100644
index 0000000..6038596
--- /dev/null
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c
@@ -0,0 +1,407 @@
+/* Applied Micro X-Gene SoC Ethernet Driver
+ *
+ * Copyright (c) 2014, Applied Micro Circuits Corporation
+ * Authors: Iyappan Subramanian <isubramanian-qTEPVZfXA3Y@public.gmane.org>
+ * Keyur Chudgar <kchudgar-qTEPVZfXA3Y@public.gmane.org>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "xgene_enet_main.h"
+#include "xgene_enet_hw.h"
+#include "xgene_enet_sgmac.h"
+
+static void xgene_enet_wr_csr(struct xgene_enet_pdata *pdata,
+ u32 offset, u32 val)
+{
+ void __iomem *addr = pdata->eth_csr_addr + offset;
+
+ iowrite32(val, addr);
+}
+
+static void xgene_enet_wr_ring_if(struct xgene_enet_pdata *pdata,
+ u32 offset, u32 val)
+{
+ void __iomem *addr = pdata->eth_ring_if_addr + offset;
+
+ iowrite32(val, addr);
+}
+
+static void xgene_enet_wr_diag_csr(struct xgene_enet_pdata *pdata,
+ u32 offset, u32 val)
+{
+ void __iomem *addr = pdata->eth_diag_csr_addr + offset;
+
+ iowrite32(val, addr);
+}
+
+static bool xgene_enet_wr_indirect(void __iomem *addr, void __iomem *wr,
+ void __iomem *cmd, void __iomem *cmd_done,
+ u32 wr_addr, u32 wr_data)
+{
+ u32 done;
+ u8 wait = 10;
+
+ iowrite32(wr_addr, addr);
+ iowrite32(wr_data, wr);
+ iowrite32(XGENE_ENET_WR_CMD, cmd);
+
+ /* wait for write command to complete */
+ while (!(done = ioread32(cmd_done)) && wait--)
+ udelay(1);
+
+ if (!done)
+ return false;
+
+ iowrite32(0, cmd);
+
+ return true;
+}
+
+static void xgene_enet_wr_mac(struct xgene_enet_pdata *pdata,
+ u32 wr_addr, u32 wr_data)
+{
+ void __iomem *addr, *wr, *cmd, *cmd_done;
+
+ addr = pdata->mcx_mac_addr + MAC_ADDR_REG_OFFSET;
+ wr = pdata->mcx_mac_addr + MAC_WRITE_REG_OFFSET;
+ cmd = pdata->mcx_mac_addr + MAC_COMMAND_REG_OFFSET;
+ cmd_done = pdata->mcx_mac_addr + MAC_COMMAND_DONE_REG_OFFSET;
+
+ if (!xgene_enet_wr_indirect(addr, wr, cmd, cmd_done, wr_addr, wr_data))
+ netdev_err(pdata->ndev, "MCX mac write failed, addr: %04x\n",
+ wr_addr);
+}
+
+static void xgene_enet_rd_csr(struct xgene_enet_pdata *pdata,
+ u32 offset, u32 *val)
+{
+ void __iomem *addr = pdata->eth_csr_addr + offset;
+
+ *val = ioread32(addr);
+}
+
+static void xgene_enet_rd_diag_csr(struct xgene_enet_pdata *pdata,
+ u32 offset, u32 *val)
+{
+ void __iomem *addr = pdata->eth_diag_csr_addr + offset;
+
+ *val = ioread32(addr);
+}
+
+static bool xgene_enet_rd_indirect(void __iomem *addr, void __iomem *rd,
+ void __iomem *cmd, void __iomem *cmd_done,
+ u32 rd_addr, u32 *rd_data)
+{
+ u32 done;
+ u8 wait = 10;
+
+ iowrite32(rd_addr, addr);
+ iowrite32(XGENE_ENET_RD_CMD, cmd);
+
+ /* wait for read command to complete */
+ while (!(done = ioread32(cmd_done)) && wait--)
+ udelay(1);
+
+ if (!done)
+ return false;
+
+ *rd_data = ioread32(rd);
+ iowrite32(0, cmd);
+
+ return true;
+}
+
+static void xgene_enet_rd_mac(struct xgene_enet_pdata *pdata,
+ u32 rd_addr, u32 *rd_data)
+{
+ void __iomem *addr, *rd, *cmd, *cmd_done;
+
+ addr = pdata->mcx_mac_addr + MAC_ADDR_REG_OFFSET;
+ rd = pdata->mcx_mac_addr + MAC_READ_REG_OFFSET;
+ cmd = pdata->mcx_mac_addr + MAC_COMMAND_REG_OFFSET;
+ cmd_done = pdata->mcx_mac_addr + MAC_COMMAND_DONE_REG_OFFSET;
+
+ if (!xgene_enet_rd_indirect(addr, rd, cmd, cmd_done, rd_addr, rd_data))
+ netdev_err(pdata->ndev, "MCX mac read failed, addr: %04x\n",
+ rd_addr);
+}
+
+static int xgene_enet_ecc_init(struct xgene_enet_pdata *pdata)
+{
+ struct net_device *ndev = pdata->ndev;
+ u32 data;
+ u8 wait = 10;
+
+ xgene_enet_wr_diag_csr(pdata, ENET_CFG_MEM_RAM_SHUTDOWN_ADDR, 0x0);
+ do {
+ usleep_range(100, 110);
+ xgene_enet_rd_diag_csr(pdata, ENET_BLOCK_MEM_RDY_ADDR, &data);
+ } while ((data != 0xffffffff) && wait--);
+
+ if (data != 0xffffffff) {
+ netdev_err(ndev, "Failed to release memory from shutdown\n");
+ return -ENODEV;
+ }
+
+ return 0;
+}
+
+static void xgene_enet_config_ring_if_assoc(struct xgene_enet_pdata *pdata)
+{
+ xgene_enet_wr_ring_if(pdata, ENET_CFGSSQMIWQASSOC_ADDR, 0xffffffff);
+ xgene_enet_wr_ring_if(pdata, ENET_CFGSSQMIFPQASSOC_ADDR, 0xffffffff);
+}
+
+static void xgene_mii_phy_write(struct xgene_enet_pdata *pdata, u8 phy_id,
+ u32 reg, u16 data)
+{
+ u32 addr, wr_data, done;
+ u8 wait = 10;
+
+ addr = PHY_ADDR(phy_id) | REG_ADDR(reg);
+ xgene_enet_wr_mac(pdata, MII_MGMT_ADDRESS_ADDR, addr);
+
+ wr_data = PHY_CONTROL(data);
+ xgene_enet_wr_mac(pdata, MII_MGMT_CONTROL_ADDR, wr_data);
+
+ while (wait--) {
+ xgene_enet_rd_mac(pdata, MII_MGMT_INDICATORS_ADDR, &done);
+ if (!(done & BUSY_MASK))
+ break;
+ usleep_range(10, 20);
+ }
+ if (done & BUSY_MASK)
+ netdev_err(pdata->ndev, "MII_MGMT write failed\n");
+}
+
+static void xgene_mii_phy_read(struct xgene_enet_pdata *pdata, u8 phy_id,
+ u32 reg, u32 *data)
+{
+ u32 addr, done;
+ u8 wait = 10;
+
+ addr = PHY_ADDR(phy_id) | REG_ADDR(reg);
+ xgene_enet_wr_mac(pdata, MII_MGMT_ADDRESS_ADDR, addr);
+ xgene_enet_wr_mac(pdata, MII_MGMT_COMMAND_ADDR, READ_CYCLE_MASK);
+
+ while (wait--) {
+ xgene_enet_rd_mac(pdata, MII_MGMT_INDICATORS_ADDR, &done);
+ if (!(done & BUSY_MASK))
+ break;
+ usleep_range(10, 20);
+ }
+ if (done & BUSY_MASK)
+ netdev_err(pdata->ndev, "MII_MGMT read failed\n");
+
+ xgene_enet_rd_mac(pdata, MII_MGMT_STATUS_ADDR, data);
+ xgene_enet_wr_mac(pdata, MII_MGMT_COMMAND_ADDR, 0);
+}
+
+static void xgene_sgmac_reset(struct xgene_enet_pdata *pdata)
+{
+ xgene_enet_wr_mac(pdata, MAC_CONFIG_1_ADDR, SOFT_RESET1);
+ xgene_enet_wr_mac(pdata, MAC_CONFIG_1_ADDR, 0);
+}
+
+static void xgene_sgmac_set_mac_addr(struct xgene_enet_pdata *pdata)
+{
+ u32 addr0, addr1;
+ u8 *dev_addr = pdata->ndev->dev_addr;
+
+ addr0 = (dev_addr[3] << 24) | (dev_addr[2] << 16) |
+ (dev_addr[1] << 8) | dev_addr[0];
+ xgene_enet_wr_mac(pdata, STATION_ADDR0_ADDR, addr0);
+
+ xgene_enet_rd_mac(pdata, STATION_ADDR1_ADDR, &addr1);
+ addr1 |= (dev_addr[5] << 24) | (dev_addr[4] << 16);
+ xgene_enet_wr_mac(pdata, STATION_ADDR1_ADDR, addr1);
+}
+
+static u32 xgene_enet_link_status(struct xgene_enet_pdata *pdata)
+{
+ u32 data;
+
+ xgene_mii_phy_read(pdata, INT_PHY_ADDR,
+ SGMII_BASE_PAGE_ABILITY_ADDR >> 2, &data);
+
+ return data & LINK_UP;
+}
+
+static void xgene_sgmac_init(struct xgene_enet_pdata *pdata)
+{
+ u32 data, loop = 10;
+
+ xgene_sgmac_reset(pdata);
+
+ /* Enable auto-negotiation */
+ xgene_mii_phy_write(pdata, INT_PHY_ADDR,
+ SGMII_CONTROL_ADDR >> 2, 0x1000);
+ xgene_mii_phy_write(pdata, INT_PHY_ADDR,
+ SGMII_TBI_CONTROL_ADDR >> 2, 0);
+
+ while (loop--) {
+ xgene_mii_phy_read(pdata, INT_PHY_ADDR,
+ SGMII_STATUS_ADDR >> 2, &data);
+ if ((data & AUTO_NEG_COMPLETE) && (data & LINK_STATUS))
+ break;
+ usleep_range(10, 20);
+ }
+ if (!(data & AUTO_NEG_COMPLETE) || !(data & LINK_STATUS))
+ netdev_err(pdata->ndev, "Auto-negotiation failed\n");
+
+ xgene_enet_rd_mac(pdata, MAC_CONFIG_2_ADDR, &data);
+ ENET_INTERFACE_MODE2_SET(&data, 2);
+ xgene_enet_wr_mac(pdata, MAC_CONFIG_2_ADDR, data | FULL_DUPLEX2);
+ xgene_enet_wr_mac(pdata, INTERFACE_CONTROL_ADDR, ENET_GHD_MODE);
+
+ xgene_enet_rd_csr(pdata, ENET_SPARE_CFG_REG_ADDR, &data);
+ data |= MPA_IDLE_WITH_QMI_EMPTY;
+ xgene_enet_wr_csr(pdata, ENET_SPARE_CFG_REG_ADDR, data);
+
+ xgene_sgmac_set_mac_addr(pdata);
+
+ xgene_enet_rd_csr(pdata, DEBUG_REG_ADDR, &data);
+ data |= CFG_BYPASS_UNISEC_TX | CFG_BYPASS_UNISEC_RX;
+ xgene_enet_wr_csr(pdata, DEBUG_REG_ADDR, data);
+
+ /* Adjust MDC clock frequency */
+ xgene_enet_rd_mac(pdata, MII_MGMT_CONFIG_ADDR, &data);
+ MGMT_CLOCK_SEL_SET(&data, 7);
+ xgene_enet_wr_mac(pdata, MII_MGMT_CONFIG_ADDR, data);
+
+ /* Enable drop if bufpool not available */
+ xgene_enet_rd_csr(pdata, RSIF_CONFIG_REG_ADDR, &data);
+ data |= CFG_RSIF_FPBUFF_TIMEOUT_EN;
+ xgene_enet_wr_csr(pdata, RSIF_CONFIG_REG_ADDR, data);
+
+ /* Rtype should be copied from FP */
+ xgene_enet_wr_csr(pdata, RSIF_RAM_DBG_REG0_ADDR, 0);
+
+ /* Bypass traffic gating */
+ xgene_enet_wr_csr(pdata, CFG_LINK_AGGR_RESUME_0_ADDR, TX_PORT0);
+ xgene_enet_wr_csr(pdata, CFG_BYPASS_ADDR, RESUME_TX);
+ xgene_enet_wr_csr(pdata, SG_RX_DV_GATE_REG_0_ADDR, RESUME_RX0);
+}
+
+static void xgene_sgmac_rx_enable(struct xgene_enet_pdata *pdata)
+{
+ u32 data;
+
+ xgene_enet_rd_mac(pdata, MAC_CONFIG_1_ADDR, &data);
+ xgene_enet_wr_mac(pdata, MAC_CONFIG_1_ADDR, data | RX_EN);
+}
+
+static void xgene_sgmac_tx_enable(struct xgene_enet_pdata *pdata)
+{
+ u32 data;
+
+ xgene_enet_rd_mac(pdata, MAC_CONFIG_1_ADDR, &data);
+ xgene_enet_wr_mac(pdata, MAC_CONFIG_1_ADDR, data | TX_EN);
+}
+
+static void xgene_sgmac_rx_disable(struct xgene_enet_pdata *pdata)
+{
+ u32 data;
+
+ xgene_enet_rd_mac(pdata, MAC_CONFIG_1_ADDR, &data);
+ xgene_enet_wr_mac(pdata, MAC_CONFIG_1_ADDR, data & ~RX_EN);
+}
+
+static void xgene_sgmac_tx_disable(struct xgene_enet_pdata *pdata)
+{
+ u32 data;
+
+ xgene_enet_rd_mac(pdata, MAC_CONFIG_1_ADDR, &data);
+ xgene_enet_wr_mac(pdata, MAC_CONFIG_1_ADDR, data & ~TX_EN);
+}
+
+static void xgene_enet_reset(struct xgene_enet_pdata *pdata)
+{
+ clk_prepare_enable(pdata->clk);
+ clk_disable_unprepare(pdata->clk);
+ clk_prepare_enable(pdata->clk);
+
+ xgene_enet_ecc_init(pdata);
+ xgene_enet_config_ring_if_assoc(pdata);
+}
+
+static void xgene_enet_cle_bypass(struct xgene_enet_pdata *pdata,
+ u32 dst_ring_num, u16 bufpool_id)
+{
+ u32 cb, fpsel;
+
+ xgene_enet_rd_csr(pdata, CLE_BYPASS_REG0_0_ADDR, &cb);
+ cb |= CFG_CLE_BYPASS_EN0;
+ CFG_CLE_IP_PROTOCOL0_SET(&cb, 3);
+ xgene_enet_wr_csr(pdata, CLE_BYPASS_REG0_0_ADDR, cb);
+
+ fpsel = xgene_enet_ring_bufnum(bufpool_id) - 0x20;
+ xgene_enet_rd_csr(pdata, CLE_BYPASS_REG1_0_ADDR, &cb);
+ CFG_CLE_DSTQID0_SET(&cb, dst_ring_num);
+ CFG_CLE_FPSEL0_SET(&cb, fpsel);
+ xgene_enet_wr_csr(pdata, CLE_BYPASS_REG1_0_ADDR, cb);
+}
+
+static void xgene_enet_shutdown(struct xgene_enet_pdata *pdata)
+{
+ clk_disable_unprepare(pdata->clk);
+}
+
+static void xgene_enet_link_state(struct work_struct *work)
+{
+ struct xgene_enet_pdata *pdata = container_of(to_delayed_work(work),
+ struct xgene_enet_pdata, link_work);
+ struct net_device *ndev = pdata->ndev;
+ u32 link_status, poll_interval;
+
+ link_status = xgene_enet_link_status(pdata);
+ if (link_status) {
+ if (!netif_carrier_ok(ndev)) {
+ netif_carrier_on(ndev);
+ xgene_sgmac_init(pdata);
+ xgene_sgmac_rx_enable(pdata);
+ xgene_sgmac_tx_enable(pdata);
+ netdev_info(ndev, "Link is Up - 1Gbps\n");
+ }
+ poll_interval = PHY_POLL_LINK_ON;
+ } else {
+ if (netif_carrier_ok(ndev)) {
+ xgene_sgmac_rx_disable(pdata);
+ xgene_sgmac_tx_disable(pdata);
+ netif_carrier_off(ndev);
+ netdev_info(ndev, "Link is Down\n");
+ }
+ poll_interval = PHY_POLL_LINK_OFF;
+ }
+
+ schedule_delayed_work(&pdata->link_work, poll_interval);
+}
+
+struct xgene_mac_ops xgene_sgmac_ops = {
+ .init = xgene_sgmac_init,
+ .reset = xgene_sgmac_reset,
+ .rx_enable = xgene_sgmac_rx_enable,
+ .tx_enable = xgene_sgmac_tx_enable,
+ .rx_disable = xgene_sgmac_rx_disable,
+ .tx_disable = xgene_sgmac_tx_disable,
+ .set_mac_addr = xgene_sgmac_set_mac_addr,
+ .link_state = xgene_enet_link_state
+};
+
+struct xgene_port_ops xgene_sgport_ops = {
+ .reset = xgene_enet_reset,
+ .cle_bypass = xgene_enet_cle_bypass,
+ .shutdown = xgene_enet_shutdown,
+};
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h
new file mode 100644
index 0000000..de43246
--- /dev/null
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h
@@ -0,0 +1,41 @@
+/* Applied Micro X-Gene SoC Ethernet Driver
+ *
+ * Copyright (c) 2014, Applied Micro Circuits Corporation
+ * Authors: Iyappan Subramanian <isubramanian-qTEPVZfXA3Y@public.gmane.org>
+ * Keyur Chudgar <kchudgar-qTEPVZfXA3Y@public.gmane.org>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __XGENE_ENET_SGMAC_H__
+#define __XGENE_ENET_SGMAC_H__
+
+#define PHY_ADDR(src) (((src)<<8) & GENMASK(12, 8))
+#define REG_ADDR(src) ((src) & GENMASK(4, 0))
+#define PHY_CONTROL(src) ((src) & GENMASK(15, 0))
+#define INT_PHY_ADDR 0x1e
+#define SGMII_TBI_CONTROL_ADDR 0x44
+#define SGMII_CONTROL_ADDR 0x00
+#define SGMII_STATUS_ADDR 0x04
+#define SGMII_BASE_PAGE_ABILITY_ADDR 0x14
+#define AUTO_NEG_COMPLETE BIT(5)
+#define LINK_STATUS BIT(2)
+#define LINK_UP BIT(15)
+#define MPA_IDLE_WITH_QMI_EMPTY BIT(12)
+#define SG_RX_DV_GATE_REG_0_ADDR 0x0dfc
+
+extern struct xgene_mac_ops xgene_sgmac_ops;
+extern struct xgene_port_ops xgene_sgport_ops;
+
+#endif /* __XGENE_ENET_SGMAC_H__ */
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c
index cd64b9f..67d0720 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c
@@ -284,7 +284,7 @@ static void xgene_enet_shutdown(struct xgene_enet_pdata *pdata)
clk_disable_unprepare(pdata->clk);
}
-void xgene_enet_link_state(struct work_struct *work)
+static void xgene_enet_link_state(struct work_struct *work)
{
struct xgene_enet_pdata *pdata = container_of(to_delayed_work(work),
struct xgene_enet_pdata, link_work);
@@ -322,6 +322,7 @@ struct xgene_mac_ops xgene_xgmac_ops = {
.rx_disable = xgene_xgmac_rx_disable,
.tx_disable = xgene_xgmac_tx_disable,
.set_mac_addr = xgene_xgmac_set_mac_addr,
+ .link_state = xgene_enet_link_state
};
struct xgene_port_ops xgene_xgport_ops = {
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h
index d2d59e7..5a5296a 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h
@@ -47,10 +47,6 @@
#define XG_ENET_SPARE_CFG_REG_1_ADDR 0x0410
#define XGENET_RX_DV_GATE_REG_0_ADDR 0x0804
-#define PHY_POLL_LINK_ON (10 * HZ)
-#define PHY_POLL_LINK_OFF (PHY_POLL_LINK_ON / 5)
-
-void xgene_enet_link_state(struct work_struct *work);
extern struct xgene_mac_ops xgene_xgmac_ops;
extern struct xgene_port_ops xgene_xgport_ops;
--
1.9.1
--
To unsubscribe from this list: send the line "unsubscribe devicetree" 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 v1 3/3] drivers: net: xgene: Add SGMII based 1GbE ethtool support
From: Iyappan Subramanian @ 2014-10-10 20:18 UTC (permalink / raw)
To: davem, netdev, devicetree
Cc: linux-arm-kernel, patches, kchudgar, Iyappan Subramanian
In-Reply-To: <1412972316-16344-1-git-send-email-isubramanian@apm.com>
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Keyur Chudgar <kchudgar@apm.com>
---
.../net/ethernet/apm/xgene/xgene_enet_ethtool.c | 25 +++++++++++++++-------
1 file changed, 17 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c b/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c
index c1c997b..416d6eb 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c
@@ -64,16 +64,25 @@ static int xgene_get_settings(struct net_device *ndev, struct ethtool_cmd *cmd)
return -ENODEV;
return phy_ethtool_gset(phydev, cmd);
+ } else if (pdata->phy_mode == PHY_INTERFACE_MODE_SGMII) {
+ cmd->supported = SUPPORTED_1000baseT_Full |
+ SUPPORTED_Autoneg | SUPPORTED_MII;
+ cmd->advertising = cmd->supported;
+ ethtool_cmd_speed_set(cmd, SPEED_1000);
+ cmd->duplex = DUPLEX_FULL;
+ cmd->port = PORT_MII;
+ cmd->transceiver = XCVR_INTERNAL;
+ cmd->autoneg = AUTONEG_ENABLE;
+ } else {
+ cmd->supported = SUPPORTED_10000baseT_Full | SUPPORTED_FIBRE;
+ cmd->advertising = cmd->supported;
+ ethtool_cmd_speed_set(cmd, SPEED_10000);
+ cmd->duplex = DUPLEX_FULL;
+ cmd->port = PORT_FIBRE;
+ cmd->transceiver = XCVR_INTERNAL;
+ cmd->autoneg = AUTONEG_DISABLE;
}
- cmd->supported = SUPPORTED_10000baseT_Full | SUPPORTED_FIBRE;
- cmd->advertising = cmd->supported;
- ethtool_cmd_speed_set(cmd, SPEED_10000);
- cmd->duplex = DUPLEX_FULL;
- cmd->port = PORT_FIBRE;
- cmd->transceiver = XCVR_EXTERNAL;
- cmd->autoneg = AUTONEG_DISABLE;
-
return 0;
}
--
1.9.1
^ permalink raw reply related
* [PATCH v1 0/3] Add SGMII based 1GbE support to APM X-Gene SoC ethernet driver
From: Iyappan Subramanian @ 2014-10-10 20:18 UTC (permalink / raw)
To: davem-fT/PcQaiUtIeIZ0/mPfg9Q, netdev-u79uwXL29TY76Z2rM5mHXA,
devicetree-u79uwXL29TY76Z2rM5mHXA
Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
patches-qTEPVZfXA3Y, kchudgar-qTEPVZfXA3Y, Iyappan Subramanian
Adding SGMII based 1GbE basic support to APM X-Gene SoC ethernet driver.
v1:
* Initial version
---
Iyappan Subramanian (3):
dtb: Add SGMII based 1GbE node to APM X-Gene SoC device tree
drivers: net: xgene: Add SGMII based 1GbE support
drivers: net: xgene: Add SGMII based 1GbE ethtool support
arch/arm64/boot/dts/apm-mustang.dts | 4 +
arch/arm64/boot/dts/apm-storm.dtsi | 24 ++
drivers/net/ethernet/apm/xgene/Makefile | 2 +-
.../net/ethernet/apm/xgene/xgene_enet_ethtool.c | 25 +-
drivers/net/ethernet/apm/xgene/xgene_enet_hw.c | 1 -
drivers/net/ethernet/apm/xgene/xgene_enet_hw.h | 2 +-
drivers/net/ethernet/apm/xgene/xgene_enet_main.c | 24 +-
drivers/net/ethernet/apm/xgene/xgene_enet_main.h | 5 +-
drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c | 407 +++++++++++++++++++++
drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h | 41 +++
drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c | 3 +-
drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h | 4 -
12 files changed, 518 insertions(+), 24 deletions(-)
create mode 100644 drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c
create mode 100644 drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h
--
1.9.1
--
To unsubscribe from this list: send the line "unsubscribe devicetree" 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
* [PATCH v1 1/3] dtb: Add SGMII based 1GbE node to APM X-Gene SoC device tree
From: Iyappan Subramanian @ 2014-10-10 20:18 UTC (permalink / raw)
To: davem, netdev, devicetree
Cc: linux-arm-kernel, patches, kchudgar, Iyappan Subramanian
In-Reply-To: <1412972316-16344-1-git-send-email-isubramanian@apm.com>
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Keyur Chudgar <kchudgar@apm.com>
---
arch/arm64/boot/dts/apm-mustang.dts | 4 ++++
arch/arm64/boot/dts/apm-storm.dtsi | 24 ++++++++++++++++++++++++
2 files changed, 28 insertions(+)
diff --git a/arch/arm64/boot/dts/apm-mustang.dts b/arch/arm64/boot/dts/apm-mustang.dts
index 2ae782b..71a1489 100644
--- a/arch/arm64/boot/dts/apm-mustang.dts
+++ b/arch/arm64/boot/dts/apm-mustang.dts
@@ -33,6 +33,10 @@
status = "ok";
};
+&sgenet0 {
+ status = "ok";
+};
+
&xgenet {
status = "ok";
};
diff --git a/arch/arm64/boot/dts/apm-storm.dtsi b/arch/arm64/boot/dts/apm-storm.dtsi
index d16cc03..f45bbfe 100644
--- a/arch/arm64/boot/dts/apm-storm.dtsi
+++ b/arch/arm64/boot/dts/apm-storm.dtsi
@@ -176,6 +176,16 @@
clock-output-names = "menetclk";
};
+ sge0clk: sge0clk@1f21c000 {
+ compatible = "apm,xgene-device-clock";
+ #clock-cells = <1>;
+ clocks = <&socplldiv2 0>;
+ reg = <0x0 0x1f21c000 0x0 0x1000>;
+ reg-names = "csr-reg";
+ csr-mask = <0x3>;
+ clock-output-names = "sge0clk";
+ };
+
xge0clk: xge0clk@1f61c000 {
compatible = "apm,xgene-device-clock";
#clock-cells = <1>;
@@ -446,6 +456,20 @@
};
};
+ sgenet0: ethernet@1f210000 {
+ compatible = "apm,xgene-enet";
+ status = "disabled";
+ reg = <0x0 0x1f210000 0x0 0x10000>,
+ <0x0 0x1f200000 0x0 0X10000>,
+ <0x0 0x1B000000 0x0 0X20000>;
+ reg-names = "enet_csr", "ring_csr", "ring_cmd";
+ interrupts = <0x0 0xA0 0x4>;
+ dma-coherent;
+ clocks = <&sge0clk 0>;
+ local-mac-address = [00 00 00 00 00 00];
+ phy-connection-type = "sgmii";
+ };
+
xgenet: ethernet@1f610000 {
compatible = "apm,xgene-enet";
status = "disabled";
--
1.9.1
^ permalink raw reply related
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