* [v3, 4/9] ptp_qoriq: add little enadian support
From: Yangbo Lu @ 2019-02-12 4:23 UTC (permalink / raw)
To: netdev, devicetree
Cc: David S . Miller, Richard Cochran, Rob Herring, Claudiu Manoil,
Yangbo Lu
In-Reply-To: <20190212042404.15575-1-yangbo.lu@nxp.com>
There is QorIQ 1588 timer IP block on the new ENETC Ethernet
controller. However it uses little endian mode which is different
with before. This patch is to add little endian support for the
driver by using "little-endian" dts node property.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
---
Changes for v2:
- None.
Changes for v3:
- Rebased.
---
drivers/ptp/ptp_qoriq.c | 69 ++++++++++++++++++++++-----------------
drivers/ptp/ptp_qoriq_debugfs.c | 12 +++---
include/linux/fsl/ptp_qoriq.h | 21 ++++++++---
3 files changed, 60 insertions(+), 42 deletions(-)
diff --git a/drivers/ptp/ptp_qoriq.c b/drivers/ptp/ptp_qoriq.c
index db4f929..ed4dc39 100644
--- a/drivers/ptp/ptp_qoriq.c
+++ b/drivers/ptp/ptp_qoriq.c
@@ -43,8 +43,8 @@ static u64 tmr_cnt_read(struct ptp_qoriq *ptp_qoriq)
u64 ns;
u32 lo, hi;
- lo = qoriq_read(®s->ctrl_regs->tmr_cnt_l);
- hi = qoriq_read(®s->ctrl_regs->tmr_cnt_h);
+ lo = ptp_qoriq->read(®s->ctrl_regs->tmr_cnt_l);
+ hi = ptp_qoriq->read(®s->ctrl_regs->tmr_cnt_h);
ns = ((u64) hi) << 32;
ns |= lo;
return ns;
@@ -57,8 +57,8 @@ static void tmr_cnt_write(struct ptp_qoriq *ptp_qoriq, u64 ns)
u32 hi = ns >> 32;
u32 lo = ns & 0xffffffff;
- qoriq_write(®s->ctrl_regs->tmr_cnt_l, lo);
- qoriq_write(®s->ctrl_regs->tmr_cnt_h, hi);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_cnt_l, lo);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_cnt_h, hi);
}
/* Caller must hold ptp_qoriq->lock. */
@@ -73,8 +73,8 @@ static void set_alarm(struct ptp_qoriq *ptp_qoriq)
ns -= ptp_qoriq->tclk_period;
hi = ns >> 32;
lo = ns & 0xffffffff;
- qoriq_write(®s->alarm_regs->tmr_alarm1_l, lo);
- qoriq_write(®s->alarm_regs->tmr_alarm1_h, hi);
+ ptp_qoriq->write(®s->alarm_regs->tmr_alarm1_l, lo);
+ ptp_qoriq->write(®s->alarm_regs->tmr_alarm1_h, hi);
}
/* Caller must hold ptp_qoriq->lock. */
@@ -83,8 +83,8 @@ static void set_fipers(struct ptp_qoriq *ptp_qoriq)
struct ptp_qoriq_registers *regs = &ptp_qoriq->regs;
set_alarm(ptp_qoriq);
- qoriq_write(®s->fiper_regs->tmr_fiper1, ptp_qoriq->tmr_fiper1);
- qoriq_write(®s->fiper_regs->tmr_fiper2, ptp_qoriq->tmr_fiper2);
+ ptp_qoriq->write(®s->fiper_regs->tmr_fiper1, ptp_qoriq->tmr_fiper1);
+ ptp_qoriq->write(®s->fiper_regs->tmr_fiper2, ptp_qoriq->tmr_fiper2);
}
static int extts_clean_up(struct ptp_qoriq *ptp_qoriq, int index,
@@ -115,8 +115,8 @@ static int extts_clean_up(struct ptp_qoriq *ptp_qoriq, int index,
event.index = index;
do {
- lo = qoriq_read(reg_etts_l);
- hi = qoriq_read(reg_etts_h);
+ lo = ptp_qoriq->read(reg_etts_l);
+ hi = ptp_qoriq->read(reg_etts_h);
if (update_event) {
event.timestamp = ((u64) hi) << 32;
@@ -124,7 +124,7 @@ static int extts_clean_up(struct ptp_qoriq *ptp_qoriq, int index,
ptp_clock_event(ptp_qoriq->clock, &event);
}
- stat = qoriq_read(®s->ctrl_regs->tmr_stat);
+ stat = ptp_qoriq->read(®s->ctrl_regs->tmr_stat);
} while (ptp_qoriq->extts_fifo_support && (stat & valid));
return 0;
@@ -144,8 +144,8 @@ irqreturn_t ptp_qoriq_isr(int irq, void *priv)
spin_lock(&ptp_qoriq->lock);
- val = qoriq_read(®s->ctrl_regs->tmr_tevent);
- mask = qoriq_read(®s->ctrl_regs->tmr_temask);
+ val = ptp_qoriq->read(®s->ctrl_regs->tmr_tevent);
+ mask = ptp_qoriq->read(®s->ctrl_regs->tmr_temask);
spin_unlock(&ptp_qoriq->lock);
@@ -173,14 +173,14 @@ irqreturn_t ptp_qoriq_isr(int irq, void *priv)
ns = ptp_qoriq->alarm_value + ptp_qoriq->alarm_interval;
hi = ns >> 32;
lo = ns & 0xffffffff;
- qoriq_write(®s->alarm_regs->tmr_alarm2_l, lo);
- qoriq_write(®s->alarm_regs->tmr_alarm2_h, hi);
+ ptp_qoriq->write(®s->alarm_regs->tmr_alarm2_l, lo);
+ ptp_qoriq->write(®s->alarm_regs->tmr_alarm2_h, hi);
ptp_qoriq->alarm_value = ns;
} else {
spin_lock(&ptp_qoriq->lock);
- mask = qoriq_read(®s->ctrl_regs->tmr_temask);
+ mask = ptp_qoriq->read(®s->ctrl_regs->tmr_temask);
mask &= ~ALM2EN;
- qoriq_write(®s->ctrl_regs->tmr_temask, mask);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_temask, mask);
spin_unlock(&ptp_qoriq->lock);
ptp_qoriq->alarm_value = 0;
ptp_qoriq->alarm_interval = 0;
@@ -194,7 +194,7 @@ irqreturn_t ptp_qoriq_isr(int irq, void *priv)
}
if (ack) {
- qoriq_write(®s->ctrl_regs->tmr_tevent, ack);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_tevent, ack);
return IRQ_HANDLED;
} else
return IRQ_NONE;
@@ -229,7 +229,7 @@ int ptp_qoriq_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
tmr_add = neg_adj ? tmr_add - diff : tmr_add + diff;
- qoriq_write(®s->ctrl_regs->tmr_add, tmr_add);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_add, tmr_add);
return 0;
}
@@ -326,15 +326,15 @@ int ptp_qoriq_enable(struct ptp_clock_info *ptp,
spin_lock_irqsave(&ptp_qoriq->lock, flags);
- mask = qoriq_read(®s->ctrl_regs->tmr_temask);
+ mask = ptp_qoriq->read(®s->ctrl_regs->tmr_temask);
if (on) {
mask |= bit;
- qoriq_write(®s->ctrl_regs->tmr_tevent, bit);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_tevent, bit);
} else {
mask &= ~bit;
}
- qoriq_write(®s->ctrl_regs->tmr_temask, mask);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_temask, mask);
spin_unlock_irqrestore(&ptp_qoriq->lock, flags);
return 0;
@@ -496,6 +496,14 @@ int ptp_qoriq_init(struct ptp_qoriq *ptp_qoriq, void __iomem *base,
return -ENODEV;
}
+ if (of_property_read_bool(node, "little-endian")) {
+ ptp_qoriq->read = qoriq_read_le;
+ ptp_qoriq->write = qoriq_write_le;
+ } else {
+ ptp_qoriq->read = qoriq_read_be;
+ ptp_qoriq->write = qoriq_write_be;
+ }
+
if (of_device_is_compatible(node, "fsl,fman-ptp-timer")) {
ptp_qoriq->regs.ctrl_regs = base + FMAN_CTRL_REGS_OFFSET;
ptp_qoriq->regs.alarm_regs = base + FMAN_ALARM_REGS_OFFSET;
@@ -519,13 +527,14 @@ int ptp_qoriq_init(struct ptp_qoriq *ptp_qoriq, void __iomem *base,
spin_lock_irqsave(&ptp_qoriq->lock, flags);
regs = &ptp_qoriq->regs;
- qoriq_write(®s->ctrl_regs->tmr_ctrl, tmr_ctrl);
- qoriq_write(®s->ctrl_regs->tmr_add, ptp_qoriq->tmr_add);
- qoriq_write(®s->ctrl_regs->tmr_prsc, ptp_qoriq->tmr_prsc);
- qoriq_write(®s->fiper_regs->tmr_fiper1, ptp_qoriq->tmr_fiper1);
- qoriq_write(®s->fiper_regs->tmr_fiper2, ptp_qoriq->tmr_fiper2);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_ctrl, tmr_ctrl);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_add, ptp_qoriq->tmr_add);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_prsc, ptp_qoriq->tmr_prsc);
+ ptp_qoriq->write(®s->fiper_regs->tmr_fiper1, ptp_qoriq->tmr_fiper1);
+ ptp_qoriq->write(®s->fiper_regs->tmr_fiper2, ptp_qoriq->tmr_fiper2);
set_alarm(ptp_qoriq);
- qoriq_write(®s->ctrl_regs->tmr_ctrl, tmr_ctrl|FIPERST|RTPE|TE|FRD);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_ctrl,
+ tmr_ctrl|FIPERST|RTPE|TE|FRD);
spin_unlock_irqrestore(&ptp_qoriq->lock, flags);
@@ -543,8 +552,8 @@ void ptp_qoriq_free(struct ptp_qoriq *ptp_qoriq)
{
struct ptp_qoriq_registers *regs = &ptp_qoriq->regs;
- qoriq_write(®s->ctrl_regs->tmr_temask, 0);
- qoriq_write(®s->ctrl_regs->tmr_ctrl, 0);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_temask, 0);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_ctrl, 0);
ptp_qoriq_remove_debugfs(ptp_qoriq);
ptp_clock_unregister(ptp_qoriq->clock);
diff --git a/drivers/ptp/ptp_qoriq_debugfs.c b/drivers/ptp/ptp_qoriq_debugfs.c
index 3a70daf..e8dddce 100644
--- a/drivers/ptp/ptp_qoriq_debugfs.c
+++ b/drivers/ptp/ptp_qoriq_debugfs.c
@@ -11,7 +11,7 @@ static int ptp_qoriq_fiper1_lpbk_get(void *data, u64 *val)
struct ptp_qoriq_registers *regs = &ptp_qoriq->regs;
u32 ctrl;
- ctrl = qoriq_read(®s->ctrl_regs->tmr_ctrl);
+ ctrl = ptp_qoriq->read(®s->ctrl_regs->tmr_ctrl);
*val = ctrl & PP1L ? 1 : 0;
return 0;
@@ -23,13 +23,13 @@ static int ptp_qoriq_fiper1_lpbk_set(void *data, u64 val)
struct ptp_qoriq_registers *regs = &ptp_qoriq->regs;
u32 ctrl;
- ctrl = qoriq_read(®s->ctrl_regs->tmr_ctrl);
+ ctrl = ptp_qoriq->read(®s->ctrl_regs->tmr_ctrl);
if (val == 0)
ctrl &= ~PP1L;
else
ctrl |= PP1L;
- qoriq_write(®s->ctrl_regs->tmr_ctrl, ctrl);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_ctrl, ctrl);
return 0;
}
@@ -42,7 +42,7 @@ static int ptp_qoriq_fiper2_lpbk_get(void *data, u64 *val)
struct ptp_qoriq_registers *regs = &ptp_qoriq->regs;
u32 ctrl;
- ctrl = qoriq_read(®s->ctrl_regs->tmr_ctrl);
+ ctrl = ptp_qoriq->read(®s->ctrl_regs->tmr_ctrl);
*val = ctrl & PP2L ? 1 : 0;
return 0;
@@ -54,13 +54,13 @@ static int ptp_qoriq_fiper2_lpbk_set(void *data, u64 val)
struct ptp_qoriq_registers *regs = &ptp_qoriq->regs;
u32 ctrl;
- ctrl = qoriq_read(®s->ctrl_regs->tmr_ctrl);
+ ctrl = ptp_qoriq->read(®s->ctrl_regs->tmr_ctrl);
if (val == 0)
ctrl &= ~PP2L;
else
ctrl |= PP2L;
- qoriq_write(®s->ctrl_regs->tmr_ctrl, ctrl);
+ ptp_qoriq->write(®s->ctrl_regs->tmr_ctrl, ctrl);
return 0;
}
diff --git a/include/linux/fsl/ptp_qoriq.h b/include/linux/fsl/ptp_qoriq.h
index 757aec3..1f8bb6a 100644
--- a/include/linux/fsl/ptp_qoriq.h
+++ b/include/linux/fsl/ptp_qoriq.h
@@ -157,21 +157,30 @@ struct ptp_qoriq {
u32 cksel;
u32 tmr_fiper1;
u32 tmr_fiper2;
+ u32 (*read)(unsigned __iomem *addr);
+ void (*write)(unsigned __iomem *addr, u32 val);
};
-static inline u32 qoriq_read(unsigned __iomem *addr)
+static inline u32 qoriq_read_be(unsigned __iomem *addr)
{
- u32 val;
-
- val = ioread32be(addr);
- return val;
+ return ioread32be(addr);
}
-static inline void qoriq_write(unsigned __iomem *addr, u32 val)
+static inline void qoriq_write_be(unsigned __iomem *addr, u32 val)
{
iowrite32be(val, addr);
}
+static inline u32 qoriq_read_le(unsigned __iomem *addr)
+{
+ return ioread32(addr);
+}
+
+static inline void qoriq_write_le(unsigned __iomem *addr, u32 val)
+{
+ iowrite32(val, addr);
+}
+
irqreturn_t ptp_qoriq_isr(int irq, void *priv);
int ptp_qoriq_init(struct ptp_qoriq *ptp_qoriq, void __iomem *base,
const struct ptp_clock_info caps);
--
1.7.1
^ permalink raw reply related
* [v3, 7/9] ptp: add QorIQ PTP support for ENETC
From: Yangbo Lu @ 2019-02-12 4:24 UTC (permalink / raw)
To: netdev, devicetree
Cc: David S . Miller, Richard Cochran, Rob Herring, Claudiu Manoil,
Yangbo Lu
In-Reply-To: <20190212042404.15575-1-yangbo.lu@nxp.com>
This patch is to add QorIQ PTP support for ENETC.
ENETC PTP driver which is a PCI driver for same
1588 timer IP block will reuse QorIQ PTP driver.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
---
Chanbges for v2:
- None.
Chanbges for v3:
- None.
---
drivers/ptp/Kconfig | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig
index aeb4a8b..7fe1863 100644
--- a/drivers/ptp/Kconfig
+++ b/drivers/ptp/Kconfig
@@ -43,7 +43,7 @@ config PTP_1588_CLOCK_DTE
config PTP_1588_CLOCK_QORIQ
tristate "Freescale QorIQ 1588 timer as PTP clock"
- depends on GIANFAR || FSL_DPAA_ETH
+ depends on GIANFAR || FSL_DPAA_ETH || FSL_ENETC || FSL_ENETC_VF
depends on PTP_1588_CLOCK
default y
help
--
1.7.1
^ permalink raw reply related
* Re: [PATCH bpf-next v2 0/4] selftests: bpf: improve bpf object file rules
From: Alexei Starovoitov @ 2019-02-12 4:34 UTC (permalink / raw)
To: Jiong Wang; +Cc: daniel, netdev, oss-drivers
In-Reply-To: <1549886481-25848-1-git-send-email-jiong.wang@netronome.com>
On Mon, Feb 11, 2019 at 12:01:17PM +0000, Jiong Wang wrote:
> This set improves bpf object file related rules in selftests Makefile.
> - tell git to ignore the build dir "alu32".
> - extend sub-register mode compilation to all bpf object files to give
> LLVM compiler bpf back-end more exercise.
> - auto-generate bpf kernel object file list.
> - relax sub-register mode compilation criteria.
>
> v1 -> v2:
> - rename "kern_progs" to "progs". (Alexei)
> - spin a new patch to remove build server kernel requirement for
> sub-register mode compilation (Alexei)
> - rebase on top of KaFai’s latest "test_sock_fields" patch set.
Looks great. Both test_progs and test_progs_32 are passing for me.
Applied to bpf-next. Thanks!
^ permalink raw reply
* Re: [PATCH v2] net: fix IPv6 prefix route residue
From: David Miller @ 2019-02-12 4:38 UTC (permalink / raw)
To: liuzhiqiang26
Cc: kuznet, yoshfuji, 0xeffeff, edumazet, netdev, mingfangsen,
zhangwenhao8, wangxiaogang3, zhoukang7, dsahern, thaller,
maowenan
In-Reply-To: <131bd64c-1484-39c3-d813-dde47ff4198c@huawei.com>
From: Zhiqiang Liu <liuzhiqiang26@huawei.com>
Date: Mon, 11 Feb 2019 10:57:46 +0800
> From: Zhiqiang Liu <liuzhiqiang26@huawei.com>
>
> Follow those steps:
> # ip addr add 2001:123::1/32 dev eth0
> # ip addr add 2001:123:456::2/64 dev eth0
> # ip addr del 2001:123::1/32 dev eth0
> # ip addr del 2001:123:456::2/64 dev eth0
> and then prefix route of 2001:123::1/32 will still exist.
>
> This is because ipv6_prefix_equal in check_cleanup_prefix_route
> func does not check whether two IPv6 addresses have the same
> prefix length. If the prefix of one address starts with another
> shorter address prefix, even though their prefix lengths are
> different, the return value of ipv6_prefix_equal is true.
>
> Here I add a check of whether two addresses have the same prefix
> to decide whether their prefixes are equal.
>
> Fixes: 5b84efecb7d9 ("ipv6 addrconf: don't cleanup prefix route
> for IFA_F_NOPREFIXROUTE")
> Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
> Reported-by: Wenhao Zhang <zhangwenhao8@huawei.com>
Applied and queued up for -stable.
Please do not split up long Fixes: tag lines, keep the entire tag on
one line only.
I fixed it up for you this time.
Thanks.
^ permalink raw reply
* Re: [PATCH bpf-next v9 7/7] selftests: bpf: add test_lwt_ip_encap selftest
From: Alexei Starovoitov @ 2019-02-12 4:39 UTC (permalink / raw)
To: Peter Oskolkov
Cc: Alexei Starovoitov, Daniel Borkmann, netdev, Peter Oskolkov,
David Ahern, Willem de Bruijn
In-Reply-To: <20190212004249.219268-8-posk@google.com>
On Mon, Feb 11, 2019 at 04:42:49PM -0800, Peter Oskolkov wrote:
> This patch adds a bpf self-test to cover BPF_LWT_ENCAP_IP mode
> in bpf_lwt_push_encap.
>
> Covered:
> - encapping in LWT_IN and LWT_XMIT
> - IPv4 and IPv6
>
> A follow-up patch will add GSO and VRF-enabled tests.
>
> Signed-off-by: Peter Oskolkov <posk@google.com>
> ---
> tools/testing/selftests/bpf/Makefile | 5 +-
> .../testing/selftests/bpf/test_lwt_ip_encap.c | 85 +++++
> .../selftests/bpf/test_lwt_ip_encap.sh | 311 ++++++++++++++++++
> 3 files changed, 399 insertions(+), 2 deletions(-)
> create mode 100644 tools/testing/selftests/bpf/test_lwt_ip_encap.c
> create mode 100755 tools/testing/selftests/bpf/test_lwt_ip_encap.sh
>
> diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
> index c7e1e3255448..3ebd41a0c253 100644
> --- a/tools/testing/selftests/bpf/Makefile
> +++ b/tools/testing/selftests/bpf/Makefile
> @@ -36,7 +36,7 @@ BPF_OBJ_FILES = \
> get_cgroup_id_kern.o socket_cookie_prog.o test_select_reuseport_kern.o \
> test_skb_cgroup_id_kern.o bpf_flow.o netcnt_prog.o test_xdp_vlan.o \
> xdp_dummy.o test_map_in_map.o test_spin_lock.o test_map_lock.o \
> - test_sock_fields_kern.o
> + test_sock_fields_kern.o test_lwt_ip_encap.o
This needs one more rebase on top of Jiong's makefile changes.
Other than this it looks great.
Thanks!
^ permalink raw reply
* Re: [PATCH bpf-next v9 5/7] bpf: add handling of BPF_LWT_REROUTE to lwt_bpf.c
From: David Ahern @ 2019-02-12 4:39 UTC (permalink / raw)
To: Peter Oskolkov, Alexei Starovoitov, Daniel Borkmann, netdev
Cc: Peter Oskolkov, Willem de Bruijn
In-Reply-To: <20190212004249.219268-6-posk@google.com>
On 2/11/19 5:42 PM, Peter Oskolkov wrote:
> @@ -88,6 +90,35 @@ static int run_lwt_bpf(struct sk_buff *skb, struct bpf_lwt_prog *lwt,
> return ret;
> }
>
> +static int bpf_lwt_input_reroute(struct sk_buff *skb)
> +{
> + int err = -EINVAL;
> +
> + if (skb->protocol == htons(ETH_P_IP)) {
> + struct iphdr *iph = ip_hdr(skb);
> +
> + err = ip_route_input_noref(skb, iph->daddr, iph->saddr,
> + iph->tos, skb_dst(skb)->dev);
> + } else if (skb->protocol == htons(ETH_P_IPV6)) {
> +#if IS_ENABLED(CONFIG_IPV6)
> + err = ipv6_stub->ipv6_route_input(skb);
> +#else
> + pr_warn_once("BPF_LWT_REROUTE input: IPV6 not available\n");
> +#endif
The stub defines ipv6_route_input when IPv6 is not enabled.
addrconf_core.o is compiled in if NET is enabled irregardless of
CONFIG_IPV6. ie., you don't need the IS_ENABLED check here.
If a bpf program pushes a v6 header the stub returns -EAFNOSUPPORT based
on patch 4.
> + } else {
> + pr_warn_once("BPF_LWT_REROUTE input: unsupported proto %d\n",
> + skb->protocol);
You don't need a warn on here; just return -EAFNOSUPPORT.
> + }
> +
> + if (err)
> + goto err;
> + return dst_input(skb);
> +
> +err:
> + kfree_skb(skb);
> + return err;
> +}
> +
> static int bpf_input(struct sk_buff *skb)
> {
> struct dst_entry *dst = skb_dst(skb);
> @@ -99,6 +130,8 @@ static int bpf_input(struct sk_buff *skb)
> ret = run_lwt_bpf(skb, &bpf->in, dst, NO_REDIRECT);
> if (ret < 0)
> return ret;
> + if (ret == BPF_LWT_REROUTE)
> + return bpf_lwt_input_reroute(skb);
> }
>
> if (unlikely(!dst->lwtstate->orig_input)) {
> @@ -148,6 +181,95 @@ static int xmit_check_hhlen(struct sk_buff *skb)
> return 0;
> }
>
> +static int bpf_lwt_xmit_reroute(struct sk_buff *skb)
> +{
> + struct net_device *l3mdev = l3mdev_master_dev_rcu(skb_dst(skb)->dev);
> + int oif = l3mdev ? l3mdev->ifindex : 0;
> + struct dst_entry *dst = NULL;
> + struct sock *sk;
> + struct net *net;
> + bool ipv4;
> + int err;
> +
> + if (skb->protocol == htons(ETH_P_IP)) {
> + ipv4 = true;
> + } else if (skb->protocol == htons(ETH_P_IPV6)) {
> + ipv4 = false;
> + } else {
> + pr_warn_once("BPF_LWT_REROUTE xmit: unsupported proto %d\n",
> + skb->protocol);
same here - no warn on.
> + return -EINVAL;
> + }
> +
> + sk = sk_to_full_sk(skb->sk);
> + if (sk) {
> + if (sk->sk_bound_dev_if)
> + oif = sk->sk_bound_dev_if;
> + net = sock_net(sk);
> + } else {
> + net = dev_net(skb_dst(skb)->dev);
> + }
> +
> + if (ipv4) {
> + struct iphdr *iph = ip_hdr(skb);
> + struct flowi4 fl4 = {};
> + struct rtable *rt;
> +
> + fl4.flowi4_oif = oif;
> + fl4.flowi4_mark = skb->mark;
> + fl4.flowi4_uid = sock_net_uid(net, sk);
> + fl4.flowi4_tos = RT_TOS(iph->tos);
> + fl4.flowi4_flags = FLOWI_FLAG_ANYSRC;
> + fl4.flowi4_proto = iph->protocol;
> + fl4.daddr = iph->daddr;
> + fl4.saddr = iph->saddr;
> +
> + rt = ip_route_output_key(net, &fl4);
> + if (IS_ERR(rt) || rt->dst.error)
> + return -EINVAL;
> + dst = &rt->dst;
> + } else {
> +#if IS_ENABLED(CONFIG_IPV6)
> + struct ipv6hdr *iph6 = ipv6_hdr(skb);
> + struct flowi6 fl6 = {};
> +
> + fl6.flowi6_oif = oif;
> + fl6.flowi6_mark = skb->mark;
> + fl6.flowi6_uid = sock_net_uid(net, sk);
> + fl6.flowlabel = ip6_flowinfo(iph6);
> + fl6.flowi6_proto = iph6->nexthdr;
> + fl6.daddr = iph6->daddr;
> + fl6.saddr = iph6->saddr;
> +
> + err = ipv6_stub->ipv6_dst_lookup(net, skb->sk, &dst, &fl6);
> + if (err || IS_ERR(dst) || dst->error)
> + return -EINVAL;
> +#else
> + pr_warn_once("BPF_LWT_REROUTE xmit: IPV6 not available\n");
> + return -EINVAL;
> +#endif
No #if .. #endif needed. The stub handles it.
^ permalink raw reply
* Re: [PATCH net-next v2 0/5] devlink: minor tweaks to reported device info
From: David Miller @ 2019-02-12 4:40 UTC (permalink / raw)
To: jakub.kicinski; +Cc: jiri, netdev, oss-drivers
In-Reply-To: <20190211033531.12928-1-jakub.kicinski@netronome.com>
From: Jakub Kicinski <jakub.kicinski@netronome.com>
Date: Sun, 10 Feb 2019 19:35:26 -0800
> This series contains two minor touch ups for devlink code. First
> || is corrected to && in the ethtool compat code. Next patch
> decreases the stack allocation size.
>
> On the nfp side after further discussions with the manufacturing
> team we decided to realign the serial number contents slightly and
> rename one of the other fields from "vendor" to "mfr", short for
> "manufacture".
>
> v2: - add patch 3 - move board maker as a generic attribute.
Series applied, thanks Jakub.
^ permalink raw reply
* Re: [PATCH bpf-next v9 7/7] selftests: bpf: add test_lwt_ip_encap selftest
From: David Ahern @ 2019-02-12 4:40 UTC (permalink / raw)
To: Alexei Starovoitov, Peter Oskolkov
Cc: Alexei Starovoitov, Daniel Borkmann, netdev, Peter Oskolkov,
Willem de Bruijn
In-Reply-To: <20190212043938.xln5tkoerpovvkcl@ast-mbp>
On 2/11/19 9:39 PM, Alexei Starovoitov wrote:
> This needs one more rebase on top of Jiong's makefile changes.
> Other than this it looks great.
> Thanks!
needs more than that. just sent comments about patch 5.
^ permalink raw reply
* Re: [PATCH bpf-next v9 7/7] selftests: bpf: add test_lwt_ip_encap selftest
From: Alexei Starovoitov @ 2019-02-12 4:41 UTC (permalink / raw)
To: David Ahern
Cc: Peter Oskolkov, Alexei Starovoitov, Daniel Borkmann,
Network Development, Peter Oskolkov, Willem de Bruijn
In-Reply-To: <da9a2e62-a83f-f192-5c29-03d73317f2e5@gmail.com>
On Mon, Feb 11, 2019 at 8:40 PM David Ahern <dsahern@gmail.com> wrote:
>
> On 2/11/19 9:39 PM, Alexei Starovoitov wrote:
> > This needs one more rebase on top of Jiong's makefile changes.
> > Other than this it looks great.
> > Thanks!
>
> needs more than that. just sent comments about patch 5.
yep. just saw your comment. agree.
^ permalink raw reply
* Re: [PATCH bpf-next 4/4] selftests/bpf: Test static data relocation
From: Alexei Starovoitov @ 2019-02-12 5:01 UTC (permalink / raw)
To: Joe Stringer; +Cc: bpf, netdev, daniel, ast
In-Reply-To: <20190212004729.535-5-joe@wand.net.nz>
On Mon, Feb 11, 2019 at 04:47:29PM -0800, Joe Stringer wrote:
> Add tests for libbpf relocation of static variable references into the
> .data and .bss sections of the ELF.
>
> Signed-off-by: Joe Stringer <joe@wand.net.nz>
...
> +#define __fetch(x) (__u32)(&(x))
> +
> +static __u32 static_bss = 0; /* Reloc reference to .bss section */
> +static __u32 static_data = 42; /* Reloc reference to .data section */
> +
> +/**
> + * Load a u32 value from a static variable into a map, for the userland test
> + * program to validate.
> + */
> +SEC("static_data_load")
> +int load_static_data(struct __sk_buff *skb)
> +{
> + __u32 key, value;
> +
> + key = 0;
> + value = __fetch(static_bss);
If we proceed with this approach we will not be able to add support
for normal 'value = static_bss;' C code in the future.
Let's figure out the way to do it right from the start.
Support for global and static variables is must have feature to add asap,
but let's not cut the corner like this.
We did such hacks in the past and every time it came back to bite us.
^ permalink raw reply
* Re: [net] tipc: fix link session and re-establish issues
From: David Miller @ 2019-02-12 5:26 UTC (permalink / raw)
To: tuong.t.lien; +Cc: jon.maloy, ying.xue, netdev, tipc-discussion
In-Reply-To: <20190211062943.4864-1-tuong.t.lien@dektech.com.au>
From: Tuong Lien <tuong.t.lien@dektech.com.au>
Date: Mon, 11 Feb 2019 13:29:43 +0700
> When a link endpoint is re-created (e.g. after a node reboot or
> interface reset), the link session number is varied by random, the peer
> endpoint will be synced with this new session number before the link is
> re-established.
>
> However, there is a shortcoming in this mechanism that can lead to the
> link never re-established or faced with a failure then. It happens when
> the peer endpoint is ready in ESTABLISHING state, the 'peer_session' as
> well as the 'in_session' flag have been set, but suddenly this link
> endpoint leaves. When it comes back with a random session number, there
> are two situations possible:
>
> 1/ If the random session number is larger than (or equal to) the
> previous one, the peer endpoint will be updated with this new session
> upon receipt of a RESET_MSG from this endpoint, and the link can be re-
> established as normal. Otherwise, all the RESET_MSGs from this endpoint
> will be rejected by the peer. In turn, when this link endpoint receives
> one ACTIVATE_MSG from the peer, it will move to ESTABLISHED and start
> to send STATE_MSGs, but again these messages will be dropped by the
> peer due to wrong session.
> The peer link endpoint can still become ESTABLISHED after receiving a
> traffic message from this endpoint (e.g. a BCAST_PROTOCOL or
> NAME_DISTRIBUTOR), but since all the STATE_MSGs are invalid, the link
> will be forced down sooner or later!
>
> Even in case the random session number is larger than the previous one,
> it can be that the ACTIVATE_MSG from the peer arrives first, and this
> link endpoint moves quickly to ESTABLISHED without sending out any
> RESET_MSG yet. Consequently, the peer link will not be updated with the
> new session number, and the same link failure scenario as above will
> happen.
>
> 2/ Another situation can be that, the peer link endpoint was reset due
> to any reasons in the meantime, its link state was set to RESET from
> ESTABLISHING but still in session, i.e. the 'in_session' flag is not
> reset...
> Now, if the random session number from this endpoint is less than the
> previous one, all the RESET_MSGs from this endpoint will be rejected by
> the peer. In the other direction, when this link endpoint receives a
> RESET_MSG from the peer, it moves to ESTABLISHING and starts to send
> ACTIVATE_MSGs, but all these messages will be rejected by the peer too.
> As a result, the link cannot be re-established but gets stuck with this
> link endpoint in state ESTABLISHING and the peer in RESET!
>
> Solution:
> ===========
>
> This link endpoint should not go directly to ESTABLISHED when getting
> ACTIVATE_MSG from the peer which may belong to the old session if the
> link was re-created. To ensure the session to be correct before the
> link is re-established, the peer endpoint in ESTABLISHING state will
> send back the last session number in ACTIVATE_MSG for a verification at
> this endpoint. Then, if needed, a new and more appropriate session
> number will be regenerated to force a re-synch first.
>
> In addition, when a link in ESTABLISHING state is reset, its state will
> move to RESET according to the link FSM, along with resetting the
> 'in_session' flag (and the other data) as a normal link reset, it will
> also be deleted if requested.
>
> The solution is backward compatible.
>
> Acked-by: Jon Maloy <jon.maloy@ericsson.com>
> Acked-by: Ying Xue <ying.xue@windriver.com>
> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Applied.
^ permalink raw reply
* Re: [PATCH] Documentation: fix some freescale dpio-driver.rst warnings
From: David Miller @ 2019-02-12 5:29 UTC (permalink / raw)
To: rdunlap
Cc: netdev, stuyoder, laurentiu.tudor, ruxandra.radulescu,
madalin.bucur, linux-doc
In-Reply-To: <99c9792a-7218-76d9-9b53-8477afa69f79@infradead.org>
From: Randy Dunlap <rdunlap@infradead.org>
Date: Sun, 10 Feb 2019 22:32:42 -0800
> From: Randy Dunlap <rdunlap@infradead.org>
>
> Fix markup warnings for one list by using correct list syntax.
> Fix markup warnings for another list by using blank lines before the
> list.
>
> Documentation/networking/device_drivers/freescale/dpaa2/dpio-driver.rst:30: WARNING: Unexpected indentation.
> Documentation/networking/device_drivers/freescale/dpaa2/dpio-driver.rst:143: WARNING: Unexpected indentation.
>
> Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Applied, thanks.
^ permalink raw reply
* Re: [PATCH net-next 1/1] flow_offload: Fix flow action infrastructure
From: David Miller @ 2019-02-12 5:31 UTC (permalink / raw)
To: elibr; +Cc: netdev, roid, pablo, jiri, saeedm
In-Reply-To: <20190211075259.39058-1-elibr@mellanox.com>
From: Eli Britstein <elibr@mellanox.com>
Date: Mon, 11 Feb 2019 09:52:59 +0200
> Implementation of macro "flow_action_for_each" introduced in
> commit e3ab786b42535 ("flow_offload: add flow action infrastructure")
> and used in commit 738678817573c ("drivers: net: use flow action
> infrastructure") iterated the first item twice and did not reach the
> last one. Fix it.
>
> Fixes: e3ab786b42535 ("flow_offload: add flow action infrastructure")
> Fixes: 738678817573c ("drivers: net: use flow action infrastructure")
> Signed-off-by: Eli Britstein <elibr@mellanox.com>
> Reviewed-by: Roi Dayan <roid@mellanox.com>
Applied, thanks.
^ permalink raw reply
* [Patch net] team: avoid complex list operations in team_nl_cmd_options_set()
From: Cong Wang @ 2019-02-12 5:59 UTC (permalink / raw)
To: netdev
Cc: Cong Wang, syzbot+4d4af685432dc0e56c91,
syzbot+68ee510075cf64260cc4, Jiri Pirko, Paolo Abeni
The current opt_inst_list operations inside team_nl_cmd_options_set()
is too complex to track:
LIST_HEAD(opt_inst_list);
nla_for_each_nested(...) {
list_for_each_entry(opt_inst, &team->option_inst_list, list) {
if (__team_option_inst_tmp_find(&opt_inst_list, opt_inst))
continue;
list_add(&opt_inst->tmp_list, &opt_inst_list);
}
}
team_nl_send_event_options_get(team, &opt_inst_list);
as while we retrieve 'opt_inst' from team->option_inst_list, it could
be added to the local 'opt_inst_list' for multiple times. The
__team_option_inst_tmp_find() doesn't work, as the setter
team_mode_option_set() still calls team->ops.exit() which uses
->tmp_list too in __team_options_change_check().
Simplify the list operations by moving the 'opt_inst_list' and
team_nl_send_event_options_get() into the nla_for_each_nested() loop so
that it can be guranteed that we won't insert a same list entry for
multiple times. Therefore, __team_option_inst_tmp_find() can be removed
too.
Fixes: 4fb0534fb7bb ("team: avoid adding twice the same option to the event list")
Fixes: 2fcdb2c9e659 ("team: allow to send multiple set events in one message")
Reported-by: syzbot+4d4af685432dc0e56c91@syzkaller.appspotmail.com
Reported-by: syzbot+68ee510075cf64260cc4@syzkaller.appspotmail.com
Cc: Jiri Pirko <jiri@resnulli.us>
Cc: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
---
drivers/net/team/team.c | 27 +++++----------------------
1 file changed, 5 insertions(+), 22 deletions(-)
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index afd9d25d1992..958f1cf67282 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -256,17 +256,6 @@ static void __team_option_inst_mark_removed_port(struct team *team,
}
}
-static bool __team_option_inst_tmp_find(const struct list_head *opts,
- const struct team_option_inst *needle)
-{
- struct team_option_inst *opt_inst;
-
- list_for_each_entry(opt_inst, opts, tmp_list)
- if (opt_inst == needle)
- return true;
- return false;
-}
-
static int __team_options_register(struct team *team,
const struct team_option *option,
size_t option_count)
@@ -2460,7 +2449,6 @@ static int team_nl_cmd_options_set(struct sk_buff *skb, struct genl_info *info)
int err = 0;
int i;
struct nlattr *nl_option;
- LIST_HEAD(opt_inst_list);
rtnl_lock();
@@ -2480,6 +2468,7 @@ static int team_nl_cmd_options_set(struct sk_buff *skb, struct genl_info *info)
struct nlattr *opt_attrs[TEAM_ATTR_OPTION_MAX + 1];
struct nlattr *attr;
struct nlattr *attr_data;
+ LIST_HEAD(opt_inst_list);
enum team_option_type opt_type;
int opt_port_ifindex = 0; /* != 0 for per-port options */
u32 opt_array_index = 0;
@@ -2584,23 +2573,17 @@ static int team_nl_cmd_options_set(struct sk_buff *skb, struct genl_info *info)
if (err)
goto team_put;
opt_inst->changed = true;
-
- /* dumb/evil user-space can send us duplicate opt,
- * keep only the last one
- */
- if (__team_option_inst_tmp_find(&opt_inst_list,
- opt_inst))
- continue;
-
list_add(&opt_inst->tmp_list, &opt_inst_list);
}
if (!opt_found) {
err = -ENOENT;
goto team_put;
}
- }
- err = team_nl_send_event_options_get(team, &opt_inst_list);
+ err = team_nl_send_event_options_get(team, &opt_inst_list);
+ if (err)
+ break;
+ }
team_put:
team_nl_team_put(team);
--
2.20.1
^ permalink raw reply related
* linux-next: manual merge of the akpm tree with the bpf tree
From: Stephen Rothwell @ 2019-02-12 6:14 UTC (permalink / raw)
To: Andrew Morton, Daniel Borkmann, Alexei Starovoitov, Networking
Cc: Linux Next Mailing List, Linux Kernel Mailing List,
Davidlohr Bueso
[-- Attachment #1: Type: text/plain, Size: 712 bytes --]
Hi all,
Today's linux-next merge of the akpm tree got a conflict in:
net/xdp/xdp_umem.c
between commit:
e451eb510684 ("xsk: share the mmap_sem for page pinning")
from the bpf tree and patch:
"net/xdp/xdp_umem.c: do not use mmap_sem"
from the akpm tree.
I fixed it up (I dropped the akpm tree patch) and can carry the fix as
necessary. This is now fixed as far as linux-next is concerned, but any
non trivial conflicts should be mentioned to your upstream maintainer
when your tree is submitted for merging. You may also want to consider
cooperating with the maintainer of the conflicting tree to minimise any
particularly complex conflicts.
--
Cheers,
Stephen Rothwell
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
^ permalink raw reply
* Re: [PATCH net] dsa: mv88e6xxx: Ensure all pending interrupts are handled prior to exit
From: Heiner Kallweit @ 2019-02-12 6:51 UTC (permalink / raw)
To: Andrew Lunn, John David Anglin
Cc: Russell King, Vivien Didelot, Florian Fainelli, netdev
In-Reply-To: <20190212035806.GE19023@lunn.ch>
On 12.02.2019 04:58, Andrew Lunn wrote:
>>> Hi David
>>>
>>> I just tested this on one of my boards. It loops endlessly:
>>>
>>> [ 47.173396] mv88e6xxx_g1_irq_thread_work: c881 a8 80
>>> [ 47.182108] mv88e6xxx_g1_irq_thread_work: c881 a8 80
>>> [ 47.190820] mv88e6xxx_g1_irq_thread_work: c881 a8 80
>>> [ 47.199535] mv88e6xxx_g1_irq_thread_work: c881 a8 80
>>> [ 47.208254] mv88e6xxx_g1_irq_thread_work: c881 a8 80
>>>
>>> These are reg, ctl1, reg & ctl1.
>>>
>>> So there is an unhandled device interrupt.
>
> Hi Heiner
>
> Your patch Fixes: 2b3e88ea6528 ("net: phy: improve phy state
> checking") is causing me problems with interrupts for the Marvell
> switches.
>
Hi Andrew,
what kernel version is it?
And the PHY driver in use is "Marvell 88E6390" ?
> That change means we don't check the PHY device if it caused an
> interrupt when its state is less than UP.
>
> What i'm seeing is that the PHY is interrupting pretty early on after
> a reboot when the previous boot had the interface up.
>
So this means that when going down for reboot the interrupts are not
properly masked / disabled? Because (at least for net-next) we enable
interrupts in phy_start() only.
> [ 10.125702] Marvell 88E6390 mv88e6xxx-0:02: phy_start_interrupts
> [ 10.162798] Marvell 88E6390 mv88e6xxx-0:02: phy_enable_interrupts
> [ 10.168931] Marvell 88E6390 mv88e6xxx-0:02: marvell_ack_interrupt
> [ 10.180164] Marvell 88E6390 mv88e6xxx-0:02: marvell_config_intr 1
>
> a little later it interrupts:
>
> [ 12.999717] mv88e6xxx_g1_irq_thread_fn
> [ 13.007253] mv88e6xxx_g2_irq_thread_fn: 4 811c 4
> [ 13.012015] libphy: __phy_is_started: phydev->state 1 PHY_UP 3
> [ 13.017941] Marvell 88E6390 mv88e6xxx-0:02: phy_interrupt: phy_is_started(phydev) 0
>
> The current code just causes it to be ignored. So the interrupts fires
> again, and again...
>
I would have more expected the opposite. If the interrupt is ignored
(IRQ_NONE returned), then it doesn't get acked. And if it's not acked
new interrupts should be blocked. Or is it different with this chip?
> If i change to code to call into the PHY driver and let it handle the
> interrupts, things keep running. A little bit later the interface is
> configured up:
>
> [ 15.921326] mv88e6085 gpio-0:00 red: configuring for phy/gmii link mode
> [ 15.928693] libphy: __phy_is_started: phydev->state 3 PHY_UP 3
> [ 15.929442] IPv6: ADDRCONF(NETDEV_UP): red: link is not ready
> [ 15.935596] Marvell 88E6390 mv88e6xxx-0:02: m88e6390_config_aneg
> [ 15.935608] Marvell 88E6390 mv88e6xxx-0:02: m88e6390_errata
>
> [ 16.071364] Marvell 88E6390 mv88e6xxx-0:02: m88e1510_config_aneg
> [ 16.112362] Marvell 88E6390 mv88e6xxx-0:02: m88e1318_config_aneg
> [ 16.151245] Marvell 88E6390 mv88e6xxx-0:02: m88e1121_config_aneg
> [ 16.368206] Marvell 88E6390 mv88e6xxx-0:02: PHY state change UP -> NOLINK
>
> and after another interrupt the link goes up.
>
> [ 19.519840] mv88e6xxx_g1_irq_thread_fn
> [ 19.528546] mv88e6xxx_g2_irq_thread_fn: 4 811c 4
> [ 19.534152] libphy: __phy_is_started: phydev->state 5 PHY_UP 3
> [ 19.540030] Marvell 88E6390 mv88e6xxx-0:02: phy_interrupt: phy_is_started(phydev) 1
> [ 19.547721] Marvell 88E6390 mv88e6xxx-0:02: m88e1121_did_interrupt
> [ 19.559829] Marvell 88E6390 mv88e6xxx-0:02: marvell_ack_interrupt
> [ 19.590753] Marvell 88E6390 mv88e6xxx-0:02: marvell_read_status
> [ 19.596712] Marvell 88E6390 mv88e6xxx-0:02: marvell_update_link
> [ 19.628387] Marvell 88E6390 mv88e6xxx-0:02: PHY state change NOLINK -> RUNNING
> [ 19.628453] mv88e6085 gpio-0:00 red: Link is Up - 1Gbps/Full - flow control off
> [ 19.635920] IPv6: ADDRCONF(NETDEV_CHANGE): red: link becomes ready
>
> I don't yet know why the first interrupt happens, before we configure
> auto-neg, etc. But it is not too unreasonable. We have configured
> interrupts, so it could be reporting link down etc.
>
> So i think we might need to revert part of this change, call into the
> driver so long as the PHY is not in state PHY_HALTED.
>
> What do you think?
>
I will take a closer look later.
> Andrew
>
Heiner
^ permalink raw reply
* [PATCH net-next 1/2] devlink: Return right error code in case of errors for region read
From: Parav Pandit @ 2019-02-12 7:09 UTC (permalink / raw)
To: jiri, davem, netdev; +Cc: parav
devlink_nl_cmd_region_read_dumpit() misses to return right error code on
most error conditions.
Return the right error code on such errors.
Fixes: 4e54795a27f5 ("devlink: Add support for region snapshot read command")
Signed-off-by: Parav Pandit <parav@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
---
net/core/devlink.c | 22 ++++++++++++++++------
1 file changed, 16 insertions(+), 6 deletions(-)
diff --git a/net/core/devlink.c b/net/core/devlink.c
index 5e2ef5a..8a198ba 100644
--- a/net/core/devlink.c
+++ b/net/core/devlink.c
@@ -3643,26 +3643,34 @@ static int devlink_nl_cmd_region_read_dumpit(struct sk_buff *skb,
goto out;
devlink = devlink_get_from_attrs(sock_net(cb->skb->sk), attrs);
- if (IS_ERR(devlink))
+ if (IS_ERR(devlink)) {
+ err = -ENODEV;
goto out;
+ }
mutex_lock(&devlink_mutex);
mutex_lock(&devlink->lock);
if (!attrs[DEVLINK_ATTR_REGION_NAME] ||
- !attrs[DEVLINK_ATTR_REGION_SNAPSHOT_ID])
+ !attrs[DEVLINK_ATTR_REGION_SNAPSHOT_ID]) {
+ err = -EINVAL;
goto out_unlock;
+ }
region_name = nla_data(attrs[DEVLINK_ATTR_REGION_NAME]);
region = devlink_region_get_by_name(devlink, region_name);
- if (!region)
+ if (!region) {
+ err = -EINVAL;
goto out_unlock;
+ }
hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
&devlink_nl_family, NLM_F_ACK | NLM_F_MULTI,
DEVLINK_CMD_REGION_READ);
- if (!hdr)
+ if (!hdr) {
+ err = -EMSGSIZE;
goto out_unlock;
+ }
err = devlink_nl_put_handle(skb, devlink);
if (err)
@@ -3673,8 +3681,10 @@ static int devlink_nl_cmd_region_read_dumpit(struct sk_buff *skb,
goto nla_put_failure;
chunks_attr = nla_nest_start(skb, DEVLINK_ATTR_REGION_CHUNKS);
- if (!chunks_attr)
+ if (!chunks_attr) {
+ err = -EMSGSIZE;
goto nla_put_failure;
+ }
if (attrs[DEVLINK_ATTR_REGION_CHUNK_ADDR] &&
attrs[DEVLINK_ATTR_REGION_CHUNK_LEN]) {
@@ -3715,7 +3725,7 @@ static int devlink_nl_cmd_region_read_dumpit(struct sk_buff *skb,
mutex_unlock(&devlink->lock);
mutex_unlock(&devlink_mutex);
out:
- return 0;
+ return err;
}
struct devlink_info_req {
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next 2/2] devlink: Fix list access without lock while reading region
From: Parav Pandit @ 2019-02-12 7:09 UTC (permalink / raw)
To: jiri, davem, netdev; +Cc: parav
While finding the devlink device during region reading,
devlink device list is accessed and devlink device is
returned without holding a lock. This could lead to user-after-free
accesses.
While at it, add lockdep assert to ensure that all future callers hold
the lock when calling devlink_get_from_attrs().
Fixes: 4e54795a27f5 ("devlink: Add support for region snapshot read command")
Signed-off-by: Parav Pandit <parav@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
---
net/core/devlink.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/net/core/devlink.c b/net/core/devlink.c
index 8a198ba..8cd773b 100644
--- a/net/core/devlink.c
+++ b/net/core/devlink.c
@@ -116,6 +116,8 @@ static struct devlink *devlink_get_from_attrs(struct net *net,
busname = nla_data(attrs[DEVLINK_ATTR_BUS_NAME]);
devname = nla_data(attrs[DEVLINK_ATTR_DEV_NAME]);
+ lockdep_assert_held(&devlink_mutex);
+
list_for_each_entry(devlink, &devlink_list, list) {
if (strcmp(devlink->dev->bus->name, busname) == 0 &&
strcmp(dev_name(devlink->dev), devname) == 0 &&
@@ -3642,13 +3644,13 @@ static int devlink_nl_cmd_region_read_dumpit(struct sk_buff *skb,
if (err)
goto out;
+ mutex_lock(&devlink_mutex);
devlink = devlink_get_from_attrs(sock_net(cb->skb->sk), attrs);
if (IS_ERR(devlink)) {
err = -ENODEV;
- goto out;
+ goto out_dev;
}
- mutex_lock(&devlink_mutex);
mutex_lock(&devlink->lock);
if (!attrs[DEVLINK_ATTR_REGION_NAME] ||
@@ -3723,6 +3725,7 @@ static int devlink_nl_cmd_region_read_dumpit(struct sk_buff *skb,
genlmsg_cancel(skb, hdr);
out_unlock:
mutex_unlock(&devlink->lock);
+out_dev:
mutex_unlock(&devlink_mutex);
out:
return err;
--
1.8.3.1
^ permalink raw reply related
* Re: [PATCH net-next 3/3] staging: fsl-dpaa2: ethsw: Remove getting PORT_BRIDGE_FLAGS
From: Greg KH @ 2019-02-12 7:28 UTC (permalink / raw)
To: Florian Fainelli
Cc: netdev, open list:STAGING SUBSYSTEM, andrew,
moderated list:ETHERNET BRIDGE, open list, vivien.didelot,
Ido Schimmel, jiri, David S. Miller
In-Reply-To: <20190211211749.19847-4-f.fainelli@gmail.com>
On Mon, Feb 11, 2019 at 01:17:49PM -0800, Florian Fainelli wrote:
> There is no code that tries to get the attribute
> SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS, remove support for doing that.
>
> Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
> ---
> drivers/staging/fsl-dpaa2/ethsw/ethsw.c | 5 -----
> 1 file changed, 5 deletions(-)
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
^ permalink raw reply
* Re: [oss-drivers] Re: [RFC 1/3] devlink: add flash update command
From: Jiri Pirko @ 2019-02-12 7:26 UTC (permalink / raw)
To: Jakub Kicinski; +Cc: davem, netdev, oss-drivers, mkubecek, andrew
In-Reply-To: <20190212082530.48cbb382@cakuba.netronome.com>
Mon, Feb 11, 2019 at 08:25:30PM CET, jakub.kicinski@netronome.com wrote:
>On Mon, 11 Feb 2019 17:45:13 +0100, Jiri Pirko wrote:
>> Mon, Feb 11, 2019 at 07:59:20AM CET, jakub.kicinski@netronome.com wrote:
>> >Add devlink flash update command. Advanced NICs have firmware
>> >stored in flash and often cryptographically secured. Updating
>> >that flash is handled by management firmware. Ethtool has a
>> >flash update command which served us well, however, it has two
>> >shortcomings:
>> > - it takes rtnl_lock unnecessarily - really flash update has
>> > nothing to do with networking, so using a networking device
>> > as a handle is suboptimal, which leads us to the second one:
>> > - it requires a functioning netdev - in case device enters an
>> > error state and can't spawn a netdev (e.g. communication
>> > with the device fails) there is no netdev to use as a handle
>> > for flashing.
>> >
>> >Devlink already has the ability to report the firmware versions,
>> >now with the ability to update the firmware/flash we will be
>> >able to recover devices in bad state.
>> >
>> >To enable easy interoperability with ethtool add the target
>> >partition ID. We may or may not add a different method of
>> >identification, but there is no such immediate need.
>> >
>> >Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
>> >---
>> > include/net/devlink.h | 2 ++
>> > include/uapi/linux/devlink.h | 6 ++++++
>> > net/core/devlink.c | 30 ++++++++++++++++++++++++++++++
>> > 3 files changed, 38 insertions(+)
>> >
>> >diff --git a/include/net/devlink.h b/include/net/devlink.h
>> >index 07660fe4c0e3..55b3478b1291 100644
>> >--- a/include/net/devlink.h
>> >+++ b/include/net/devlink.h
>> >@@ -529,6 +529,8 @@ struct devlink_ops {
>> > struct netlink_ext_ack *extack);
>> > int (*info_get)(struct devlink *devlink, struct devlink_info_req *req,
>> > struct netlink_ext_ack *extack);
>> >+ int (*flash_update)(struct devlink *devlink, const char *path,
>> >+ u32 target, struct netlink_ext_ack *extack);
>> > };
>> >
>> > static inline void *devlink_priv(struct devlink *devlink)
>> >diff --git a/include/uapi/linux/devlink.h b/include/uapi/linux/devlink.h
>> >index 72d9f7c89190..f4417283fd1b 100644
>> >--- a/include/uapi/linux/devlink.h
>> >+++ b/include/uapi/linux/devlink.h
>> >@@ -103,6 +103,8 @@ enum devlink_command {
>> > DEVLINK_CMD_HEALTH_REPORTER_DUMP_GET,
>> > DEVLINK_CMD_HEALTH_REPORTER_DUMP_CLEAR,
>> >
>> >+ DEVLINK_CMD_FLASH_UPDATE,
>> >+
>> > /* add new commands above here */
>> > __DEVLINK_CMD_MAX,
>> > DEVLINK_CMD_MAX = __DEVLINK_CMD_MAX - 1
>> >@@ -326,6 +328,10 @@ enum devlink_attr {
>> > DEVLINK_ATTR_HEALTH_REPORTER_DUMP_TS, /* u64 */
>> > DEVLINK_ATTR_HEALTH_REPORTER_GRACEFUL_PERIOD, /* u64 */
>> > DEVLINK_ATTR_HEALTH_REPORTER_AUTO_RECOVER, /* u8 */
>> >+
>> >+ DEVLINK_ATTR_FLASH_UPDATE_FILE_NAME, /* string */
>> >+ DEVLINK_ATTR_FLASH_UPDATE_TARGET_ID, /* u32 */
>>
>> Do we need to carry this on? I mean, the ef->region is only checked in 4
>> drivers against ETHTOOL_FLASH_ALL_REGIONS, which is default.
>> There is this bnxt driver which is the only one working with this value.
>> I think that for compat, there should be some id-region mapping
>> provided by driver which the compat layer would use to translate.
>>
>> I also think that this should be in sync with what is returned in
>> DEVLINK_ATTR_INFO_VERSION_NAME.
>>
>> For example:
>> $ devlink dev info pci/0000:82:00.0
>> pci/0000:82:00.0:
>> driver nfp
>> serial_number 16240145
>> versions:
>> fixed:
>> board.id AMDA0081-0001
>> board.rev 15
>> board.vendor SMA
>> board.model hydrogen
>> running:
>> fw.mgmt 010181.010181.0101d4
>> fw.cpld 0x1030000
>> fw.app abm-d372b6
>> fw.undi 0.0.2
>> chip.init AMDA-0081-0001 20160318164536
>> stored:
>> fw.mgmt 010181.010181.0101d4
>> fw.app abm-d372b6
>> fw.undi 0.0.2
>> chip.init AMDA-0081-0001 20160318164536
>>
>> Now user should be able to use one of the identifiers to flash relevant
>> fw, like:
>>
>> devlink dev flash pci/0000:82:00.0 XXX fw.mgmt file flash-boot.bin
>>
>> I'm not sure about the name of "xxx" attribute. Maybe "id":
>>
>> devlink dev flash pci/0000:82:00.0 id fw.mgmt file flash-boot.bin
>> devlink dev flash pci/0000:82:00.0 id fw.cpld file some-other.bin
>
>Agreed, that looks good! TBH in case of Netronome the binary
>image contains an identifier so it will update the correct component
Okay. So in case the "component" attr is omitted, there would be some
flag passed to the driver so it would know that the file contains more
component binaries and has to do parsing itself/in-fw.
>automatically. That's why I say "no immediate need" :) (How about
>"component" instead of "id", BTW?)
Component is fine by me.
>
>I will drop the target ID, I just added it for full backward compat
>with ethtool, but it may be confusing, given it would be mostly unused.
Ok.
>I'll drop it in non-RFC, do you want me to add the id/component or leave
>it out for now?
I think it would be good to add it and have the api complete.
^ permalink raw reply
* Re: [PATCH v2] net: fix IPv6 prefix route residue
From: Zhiqiang Liu @ 2019-02-12 7:43 UTC (permalink / raw)
To: David Miller
Cc: kuznet, yoshfuji, 0xeffeff, edumazet, netdev, mingfangsen,
zhangwenhao8, wangxiaogang3, zhoukang7, dsahern, thaller,
maowenan
In-Reply-To: <20190211.203801.1570457113402893162.davem@davemloft.net>
> From: Zhiqiang Liu <liuzhiqiang26@huawei.com>
> Date: Mon, 11 Feb 2019 10:57:46 +0800
>
>> From: Zhiqiang Liu <liuzhiqiang26@huawei.com>
>>
>> Follow those steps:
>> # ip addr add 2001:123::1/32 dev eth0
>> # ip addr add 2001:123:456::2/64 dev eth0
>> # ip addr del 2001:123::1/32 dev eth0
>> # ip addr del 2001:123:456::2/64 dev eth0
>> and then prefix route of 2001:123::1/32 will still exist.
>>
>> This is because ipv6_prefix_equal in check_cleanup_prefix_route
>> func does not check whether two IPv6 addresses have the same
>> prefix length. If the prefix of one address starts with another
>> shorter address prefix, even though their prefix lengths are
>> different, the return value of ipv6_prefix_equal is true.
>>
>> Here I add a check of whether two addresses have the same prefix
>> to decide whether their prefixes are equal.
>>
>> Fixes: 5b84efecb7d9 ("ipv6 addrconf: don't cleanup prefix route
>> for IFA_F_NOPREFIXROUTE")
>> Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
>> Reported-by: Wenhao Zhang <zhangwenhao8@huawei.com>
>
> Applied and queued up for -stable.Thank you for applying the patch.
>
> Please do not split up long Fixes: tag lines, keep the entire tag on
> one line only.
>
> I fixed it up for you this time.
>
> Thanks.
This is the first patch in my life. Thank you and David Ahern again.
^ permalink raw reply
* Re: [PATCH 01/18] MIPS: lantiq: pass struct device to DMA API functions
From: Christoph Hellwig @ 2019-02-12 7:45 UTC (permalink / raw)
To: Paul Burton
Cc: Christoph Hellwig, John Crispin, Vinod Koul, Dmitry Tarnyagin,
Nicolas Ferre, Sudip Mukherjee, Felipe Balbi,
linux-mips@vger.kernel.org, linux-kernel@vger.kernel.org,
dmaengine@vger.kernel.org, netdev@vger.kernel.org,
linux-usb@vger.kernel.org, linux-fbdev@vger.kernel.org,
alsa-devel@alsa-project.org, iommu@lists.linux-foundation.org
In-Reply-To: <20190207232912.wfgejc5c6d6lk5so@pburton-laptop>
On Thu, Feb 07, 2019 at 11:29:14PM +0000, Paul Burton wrote:
> Would you like this to go through the MIPS tree or elsewhere? If the
> latter:
>
> Acked-by: Paul Burton <paul.burton@mips.com>
Please pick it up through the mips tree!
^ permalink raw reply
* [PATCH bpf] xsk: do not remove umem from netdevice on fall-back to copy-mode
From: Björn Töpel @ 2019-02-12 7:51 UTC (permalink / raw)
To: ast, daniel, netdev
Cc: Björn Töpel, magnus.karlsson, magnus.karlsson,
jan.sokolowski
From: Björn Töpel <bjorn.topel@intel.com>
Commit c9b47cc1fabc ("xsk: fix bug when trying to use both copy and
zero-copy on one queue id") stores the umem into the netdev._rx
struct. However, the patch incorrectly removed the umem from the
netdev._rx struct when user-space passed "best-effort" mode
(i.e. select the fastest possible option available), and zero-copy
mode was not available. This commit fixes that.
Fixes: c9b47cc1fabc ("xsk: fix bug when trying to use both copy and zero-copy on one queue id")
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
---
net/xdp/xdp_umem.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c
index 597866e7c441..37e1fe180769 100644
--- a/net/xdp/xdp_umem.c
+++ b/net/xdp/xdp_umem.c
@@ -125,9 +125,10 @@ int xdp_umem_assign_dev(struct xdp_umem *umem, struct net_device *dev,
return 0;
err_unreg_umem:
- xdp_clear_umem_at_qid(dev, queue_id);
if (!force_zc)
err = 0; /* fallback to copy mode */
+ if (err)
+ xdp_clear_umem_at_qid(dev, queue_id);
out_rtnl_unlock:
rtnl_unlock();
return err;
--
2.19.1
^ permalink raw reply related
* [PATCH net-next] nfp: flower: remove double new line
From: Jakub Kicinski @ 2019-02-12 8:18 UTC (permalink / raw)
To: davem; +Cc: netdev, oss-drivers, Jakub Kicinski
Recent cls_flower offload rewrite added a double new line.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
---
drivers/net/ethernet/netronome/nfp/flower/offload.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c
index fe1469d201af..450d7296fd57 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/offload.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c
@@ -187,7 +187,6 @@ nfp_flower_calculate_key_layers(struct nfp_app *app,
if (ipv4_addrs.mask->dst != cpu_to_be32(~0))
return -EOPNOTSUPP;
-
flow_rule_match_enc_ports(rule, &enc_ports);
if (enc_ports.mask->dst != cpu_to_be16(~0))
return -EOPNOTSUPP;
--
2.19.2
^ permalink raw reply related
* [PATCH bpf-next] bpf: offload: add priv field for drivers
From: Jakub Kicinski @ 2019-02-12 8:20 UTC (permalink / raw)
To: alexei.starovoitov, daniel; +Cc: oss-drivers, netdev, Jakub Kicinski
Currently bpf_offload_dev does not have any priv pointer, forcing
the drivers to work backwards from the netdev in program metadata.
This is not great given programs are conceptually associated with
the offload device, and it means one or two unnecessary deferences.
Add a priv pointer to bpf_offload_dev.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
---
drivers/net/ethernet/netronome/nfp/bpf/main.c | 2 +-
drivers/net/ethernet/netronome/nfp/bpf/offload.c | 4 +---
drivers/net/netdevsim/bpf.c | 5 +++--
include/linux/bpf.h | 3 ++-
kernel/bpf/offload.c | 10 +++++++++-
5 files changed, 16 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/bpf/main.c b/drivers/net/ethernet/netronome/nfp/bpf/main.c
index dccae0319204..275de9f4c61c 100644
--- a/drivers/net/ethernet/netronome/nfp/bpf/main.c
+++ b/drivers/net/ethernet/netronome/nfp/bpf/main.c
@@ -465,7 +465,7 @@ static int nfp_bpf_init(struct nfp_app *app)
app->ctrl_mtu = nfp_bpf_ctrl_cmsg_mtu(bpf);
}
- bpf->bpf_dev = bpf_offload_dev_create(&nfp_bpf_dev_ops);
+ bpf->bpf_dev = bpf_offload_dev_create(&nfp_bpf_dev_ops, bpf);
err = PTR_ERR_OR_ZERO(bpf->bpf_dev);
if (err)
goto err_free_neutral_maps;
diff --git a/drivers/net/ethernet/netronome/nfp/bpf/offload.c b/drivers/net/ethernet/netronome/nfp/bpf/offload.c
index 55c7dbf8b421..15dce97650a5 100644
--- a/drivers/net/ethernet/netronome/nfp/bpf/offload.c
+++ b/drivers/net/ethernet/netronome/nfp/bpf/offload.c
@@ -185,8 +185,6 @@ static void nfp_prog_free(struct nfp_prog *nfp_prog)
static int nfp_bpf_verifier_prep(struct bpf_prog *prog)
{
- struct nfp_net *nn = netdev_priv(prog->aux->offload->netdev);
- struct nfp_app *app = nn->app;
struct nfp_prog *nfp_prog;
int ret;
@@ -197,7 +195,7 @@ static int nfp_bpf_verifier_prep(struct bpf_prog *prog)
INIT_LIST_HEAD(&nfp_prog->insns);
nfp_prog->type = prog->type;
- nfp_prog->bpf = app->priv;
+ nfp_prog->bpf = bpf_offload_dev_priv(prog->aux->offload->offdev);
ret = nfp_prog_prepare(nfp_prog, prog->insnsi, prog->len);
if (ret)
diff --git a/drivers/net/netdevsim/bpf.c b/drivers/net/netdevsim/bpf.c
index 172b271c8bd2..f92c43453ec6 100644
--- a/drivers/net/netdevsim/bpf.c
+++ b/drivers/net/netdevsim/bpf.c
@@ -248,7 +248,7 @@ static int nsim_bpf_create_prog(struct netdevsim *ns, struct bpf_prog *prog)
static int nsim_bpf_verifier_prep(struct bpf_prog *prog)
{
- struct netdevsim *ns = netdev_priv(prog->aux->offload->netdev);
+ struct netdevsim *ns = bpf_offload_dev_priv(prog->aux->offload->offdev);
if (!ns->bpf_bind_accept)
return -EOPNOTSUPP;
@@ -589,7 +589,8 @@ int nsim_bpf_init(struct netdevsim *ns)
if (IS_ERR_OR_NULL(ns->sdev->ddir_bpf_bound_progs))
return -ENOMEM;
- ns->sdev->bpf_dev = bpf_offload_dev_create(&nsim_bpf_dev_ops);
+ ns->sdev->bpf_dev = bpf_offload_dev_create(&nsim_bpf_dev_ops,
+ ns);
err = PTR_ERR_OR_ZERO(ns->sdev->bpf_dev);
if (err)
return err;
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index bd169a7bcc93..d3126ff4994a 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -767,8 +767,9 @@ int bpf_map_offload_get_next_key(struct bpf_map *map,
bool bpf_offload_prog_map_match(struct bpf_prog *prog, struct bpf_map *map);
struct bpf_offload_dev *
-bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops);
+bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops, void *priv);
void bpf_offload_dev_destroy(struct bpf_offload_dev *offdev);
+void *bpf_offload_dev_priv(struct bpf_offload_dev *offdev);
int bpf_offload_dev_netdev_register(struct bpf_offload_dev *offdev,
struct net_device *netdev);
void bpf_offload_dev_netdev_unregister(struct bpf_offload_dev *offdev,
diff --git a/kernel/bpf/offload.c b/kernel/bpf/offload.c
index 39dba8c90331..ba635209ae9a 100644
--- a/kernel/bpf/offload.c
+++ b/kernel/bpf/offload.c
@@ -35,6 +35,7 @@ static DECLARE_RWSEM(bpf_devs_lock);
struct bpf_offload_dev {
const struct bpf_prog_offload_ops *ops;
struct list_head netdevs;
+ void *priv;
};
struct bpf_offload_netdev {
@@ -669,7 +670,7 @@ void bpf_offload_dev_netdev_unregister(struct bpf_offload_dev *offdev,
EXPORT_SYMBOL_GPL(bpf_offload_dev_netdev_unregister);
struct bpf_offload_dev *
-bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops)
+bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops, void *priv)
{
struct bpf_offload_dev *offdev;
int err;
@@ -688,6 +689,7 @@ bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops)
return ERR_PTR(-ENOMEM);
offdev->ops = ops;
+ offdev->priv = priv;
INIT_LIST_HEAD(&offdev->netdevs);
return offdev;
@@ -700,3 +702,9 @@ void bpf_offload_dev_destroy(struct bpf_offload_dev *offdev)
kfree(offdev);
}
EXPORT_SYMBOL_GPL(bpf_offload_dev_destroy);
+
+void *bpf_offload_dev_priv(struct bpf_offload_dev *offdev)
+{
+ return offdev->priv;
+}
+EXPORT_SYMBOL_GPL(bpf_offload_dev_priv);
--
2.19.2
^ 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