Hi,
This patch set is addressing several issues in the DPAA Ethernet
driver suite:
- module unload crash caused by wrong reference to device being left
in the cleanup code after the DSA related changes
- scheduling wile atomic bug in QMan code revealed during dpaa_eth
module unload
- a couple of error counter fixes, a duplicated init in dpaa_eth.
Madalin
Camelia Groza (3):
dpaa_eth: remove duplicate initialization
dpaa_eth: increment the RX dropped counter when needed
dpaa_eth: remove duplicate increment of the tx_errors counter
Madalin Bucur (2):
soc/fsl/qbman: fix issue in qman_delete_cgr_safe()
dpaa_eth: fix error in dpaa_remove()
drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 8 ++++----
drivers/soc/fsl/qbman/qman.c | 28 +++++---------------------
2 files changed, 9 insertions(+), 27 deletions(-)
--
2.1.0
The recent changes that make the driver probing compatible with DSA
were not propagated in the dpa_remove() function, breaking the
module unload function. Using the proper device to address the issue.
Signed-off-by: Madalin Bucur <[email protected]>
---
drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 159dc2d..128e0a4 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -2871,7 +2871,7 @@ static int dpaa_remove(struct platform_device *pdev)
struct device *dev;
int err;
- dev = &pdev->dev;
+ dev = pdev->dev.parent;
net_dev = dev_get_drvdata(dev);
priv = netdev_priv(net_dev);
--
2.1.0
From: Camelia Groza <[email protected]>
The tx_errors counter is incremented by the dpaa_xmit caller.
Signed-off-by: Camelia Groza <[email protected]>
Signed-off-by: Madalin Bucur <[email protected]>
---
drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 76b3c9e..fd43f98 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -2021,7 +2021,6 @@ static inline int dpaa_xmit(struct dpaa_priv *priv,
}
if (unlikely(err < 0)) {
- percpu_stats->tx_errors++;
percpu_stats->tx_fifo_errors++;
return err;
}
--
2.1.0
From: Camelia Groza <[email protected]>
Signed-off-by: Camelia Groza <[email protected]>
---
drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 3e83d79..76b3c9e 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -2321,8 +2321,10 @@ static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal,
skb_len = skb->len;
- if (unlikely(netif_receive_skb(skb) == NET_RX_DROP))
+ if (unlikely(netif_receive_skb(skb) == NET_RX_DROP)) {
+ percpu_stats->rx_dropped++;
return qman_cb_dqrr_consume;
+ }
percpu_stats->rx_packets++;
percpu_stats->rx_bytes += skb_len;
--
2.1.0
From: Camelia Groza <[email protected]>
The fd_format has already been initialized at this point.
Signed-off-by: Camelia Groza <[email protected]>
---
drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 128e0a4..3e83d79 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -2289,7 +2289,6 @@ static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal,
vaddr = phys_to_virt(addr);
prefetch(vaddr + qm_fd_get_offset(fd));
- fd_format = qm_fd_get_format(fd);
/* The only FD types that we may receive are contig and S/G */
WARN_ON((fd_format != qm_fd_contig) && (fd_format != qm_fd_sg));
--
2.1.0
The wait_for_completion() call in qman_delete_cgr_safe()
was triggering a scheduling while atomic bug, replacing the
kthread with a smp_call_function_single() call to fix it.
Signed-off-by: Madalin Bucur <[email protected]>
Signed-off-by: Roy Pledge <[email protected]>
---
drivers/soc/fsl/qbman/qman.c | 28 +++++-----------------------
1 file changed, 5 insertions(+), 23 deletions(-)
diff --git a/drivers/soc/fsl/qbman/qman.c b/drivers/soc/fsl/qbman/qman.c
index e4f5bb0..ba3cfa8 100644
--- a/drivers/soc/fsl/qbman/qman.c
+++ b/drivers/soc/fsl/qbman/qman.c
@@ -2443,39 +2443,21 @@ struct cgr_comp {
struct completion completion;
};
-static int qman_delete_cgr_thread(void *p)
+static void qman_delete_cgr_smp_call(void *p)
{
- struct cgr_comp *cgr_comp = (struct cgr_comp *)p;
- int ret;
-
- ret = qman_delete_cgr(cgr_comp->cgr);
- complete(&cgr_comp->completion);
-
- return ret;
+ qman_delete_cgr((struct qman_cgr *)p);
}
void qman_delete_cgr_safe(struct qman_cgr *cgr)
{
- struct task_struct *thread;
- struct cgr_comp cgr_comp;
-
preempt_disable();
if (qman_cgr_cpus[cgr->cgrid] != smp_processor_id()) {
- init_completion(&cgr_comp.completion);
- cgr_comp.cgr = cgr;
- thread = kthread_create(qman_delete_cgr_thread, &cgr_comp,
- "cgr_del");
-
- if (IS_ERR(thread))
- goto out;
-
- kthread_bind(thread, qman_cgr_cpus[cgr->cgrid]);
- wake_up_process(thread);
- wait_for_completion(&cgr_comp.completion);
+ smp_call_function_single(qman_cgr_cpus[cgr->cgrid],
+ qman_delete_cgr_smp_call, cgr, true);
preempt_enable();
return;
}
-out:
+
qman_delete_cgr(cgr);
preempt_enable();
}
--
2.1.0
From: Madalin Bucur <[email protected]>
Date: Wed, 14 Mar 2018 08:37:27 -0500
> This patch set is addressing several issues in the DPAA Ethernet
> driver suite:
>
> - module unload crash caused by wrong reference to device being left
> in the cleanup code after the DSA related changes
> - scheduling wile atomic bug in QMan code revealed during dpaa_eth
> module unload
> - a couple of error counter fixes, a duplicated init in dpaa_eth.
Series applied, thank you.
On Wed, 2018-03-14 at 08:37 -0500, Madalin Bucur wrote:
> Hi,
>
> This patch set is addressing several issues in the DPAA Ethernet
> driver suite:
>
> - module unload crash caused by wrong reference to device being left
> in the cleanup code after the DSA related changes
> - scheduling wile atomic bug in QMan code revealed during dpaa_eth
> module unload
> - a couple of error counter fixes, a duplicated init in dpaa_eth.
hmm, some of these(all?) bugs are in stable too, CC: stable perhaps?
>
> Madalin
>
> Camelia Groza (3):
> dpaa_eth: remove duplicate initialization
> dpaa_eth: increment the RX dropped counter when needed
> dpaa_eth: remove duplicate increment of the tx_errors counter
>
> Madalin Bucur (2):
> soc/fsl/qbman: fix issue in qman_delete_cgr_safe()
> dpaa_eth: fix error in dpaa_remove()
>
> drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 8 ++++----
> drivers/soc/fsl/qbman/qman.c | 28 +++++---------------------
> 2 files changed, 9 insertions(+), 27 deletions(-)
>
> --
> 2.1.0
>
> -----Original Message-----
> From: Joakim Tjernlund [mailto:[email protected]]
> Sent: Wednesday, March 14, 2018 8:43 PM
> To: [email protected]; Madalin-cristian Bucur
> <[email protected]>
>
> On Wed, 2018-03-14 at 08:37 -0500, Madalin Bucur wrote:
> > Hi,
> >
> > This patch set is addressing several issues in the DPAA Ethernet
> > driver suite:
> >
> > - module unload crash caused by wrong reference to device being left
> > in the cleanup code after the DSA related changes
> > - scheduling wile atomic bug in QMan code revealed during dpaa_eth
> > module unload
> > - a couple of error counter fixes, a duplicated init in dpaa_eth.
>
> hmm, some of these(all?) bugs are in stable too, CC: stable perhaps?
Hi Jocke,
I did not check that, they should be added.
Thanks,
Madalin
> >
> > Madalin
> >
> > Camelia Groza (3):
> > dpaa_eth: remove duplicate initialization
> > dpaa_eth: increment the RX dropped counter when needed
> > dpaa_eth: remove duplicate increment of the tx_errors counter
> >
> > Madalin Bucur (2):
> > soc/fsl/qbman: fix issue in qman_delete_cgr_safe()
> > dpaa_eth: fix error in dpaa_remove()
> >
> > drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 8 ++++----
> > drivers/soc/fsl/qbman/qman.c | 28 +++++---------------------
> > 2 files changed, 9 insertions(+), 27 deletions(-)
> >
> > --
> > 2.1.0
> >
From: Madalin-cristian Bucur <[email protected]>
Date: Thu, 15 Mar 2018 09:32:35 +0000
>> -----Original Message-----
>> From: Joakim Tjernlund [mailto:[email protected]]
>> Sent: Wednesday, March 14, 2018 8:43 PM
>> To: [email protected]; Madalin-cristian Bucur
>> <[email protected]>
>>
>> On Wed, 2018-03-14 at 08:37 -0500, Madalin Bucur wrote:
>> > Hi,
>> >
>> > This patch set is addressing several issues in the DPAA Ethernet
>> > driver suite:
>> >
>> > - module unload crash caused by wrong reference to device being left
>> > in the cleanup code after the DSA related changes
>> > - scheduling wile atomic bug in QMan code revealed during dpaa_eth
>> > module unload
>> > - a couple of error counter fixes, a duplicated init in dpaa_eth.
>>
>> hmm, some of these(all?) bugs are in stable too, CC: stable perhaps?
>
> Hi Jocke,
>
> I did not check that, they should be added.
Networking patches are not queued to stable by CC:'ing stable.
Instead you just simply need to ask me explicitly for -stable submission
which I will do at the appropriate time after the fixes have been able
to cook in Linus's tree for a little while.
I've queued this series up for that purpose, thanks.