* Re: [PATCH net-next 2/3] vsock: fix vsock_dequeue/enqueue_accept race
From: Stefan Hajnoczi @ 2017-08-17 14:05 UTC (permalink / raw)
To: Dexuan Cui
Cc: Michal Kubecek, joe@perches.com, olaf@aepfle.de,
Stephen Hemminger, jasowang@redhat.com, netdev@vger.kernel.org,
Haiyang Zhang, Dave Scott, linux-kernel@vger.kernel.org,
apw@canonical.com, Jorgen Hansen, Rolf Neugebauer, Marcelo Cerri,
devel@linuxdriverproject.org, Vitaly Kuznetsov,
davem@davemloft.net, George Zhang, Dan Carpenter
In-Reply-To: <KL1P15301MB00082249964D8EF1BAA48790BF8D0@KL1P15301MB0008.APCP153.PROD.OUTLOOK.COM>
[-- Attachment #1.1: Type: text/plain, Size: 474 bytes --]
On Tue, Aug 15, 2017 at 10:15:39PM +0000, Dexuan Cui wrote:
> With the current code, when vsock_dequeue_accept() is removing a sock
> from the list, nothing prevents vsock_enqueue_accept() from adding a new
> sock into the list concurrently. We should add a lock to protect the list.
The listener sock is locked, preventing concurrent modification. I have
checked both the virtio and vmci transports. Can you post an example
where the listener sock isn't locked?
Stefan
[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 455 bytes --]
[-- Attachment #2: Type: text/plain, Size: 169 bytes --]
_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
^ permalink raw reply
* Re: [PATCH] vsock: only load vmci transport on VMware hypervisor by default
From: Stefan Hajnoczi @ 2017-08-17 13:55 UTC (permalink / raw)
To: Dexuan Cui
Cc: 'Jorgen S. Hansen', 'davem@davemloft.net',
'netdev@vger.kernel.org',
'gregkh@linuxfoundation.org',
'devel@linuxdriverproject.org', KY Srinivasan,
Haiyang Zhang, Stephen Hemminger, 'George Zhang',
'Michal Kubecek', 'Asias He',
'Vitaly Kuznetsov', 'Cathy Avery',
'jasowang@redhat.com', 'Rolf Neugebauer',
'Dave Scott', 'Marcelo Cerri', "'apw@
In-Reply-To: <KL1P15301MB0008F70D4E5248AB7A296CE3BF830@KL1P15301MB0008.APCP153.PROD.OUTLOOK.COM>
[-- Attachment #1: Type: text/plain, Size: 4484 bytes --]
On Thu, Aug 17, 2017 at 08:00:29AM +0000, Dexuan Cui wrote:
>
> Without the patch, vmw_vsock_vmci_transport.ko can automatically load
> when an application creates an AF_VSOCK socket.
>
> This is the expected good behavior on VMware hypervisor, but as we
> are going to add hv_sock.ko (i.e. Hyper-V transport for AF_VSOCK), we
> should make sure vmw_vsock_vmci_transport.ko can't load on Hyper-V,
> otherwise there is a -EBUSY conflict when both vmw_vsock_vmci_transport.ko
> and hv_sock.ko try to call vsock_core_init() on Hyper-V.
>
> On the other hand, hv_sock.ko can only load on Hyper-V, because it
> depends on hv_vmbus.ko, which detects Hyper-V in hv_acpi_init().
>
> KVM's vsock_virtio_transport doesn't have the issue because it doesn't
> define MODULE_ALIAS_NETPROTO(PF_VSOCK).
Thanks for sending this patch, vmci's MODULE_ALIAS_NETPROTO(PF_VSOCK) is
a problem for vhost_vsock.ko (the virtio host driver) too. A host
userspace program can create a AF_VSOCK socket before vhost_vsock is
loaded. The vmci transport will be unconditionally loaded and that's
not the right behavior.
Putting aside nested virtualization, I want to load the transport (vmci,
Hyper-V, vsock) for which there is paravirtualized hardware present
inside the guest.
It's a little tricker on the host side (doesn't matter for Hyper-V and
probably also doesn't for VMware) because the host-side driver is a
software device with no hardware backing it. In KVM we assume the
vhost_vsock.ko kernel module will be loaded sufficiently early.
Things get trickier with nested virtualization because the VM might want
to talk to its host but also to its nested VMs. The simple way of
fixing this would be to allow two transports loaded simultaneously and
route traffic destined to CID 2 to the host transport and all other
traffic to the guest transport.
Perhaps we should discuss these cases a bit more to figure out how to
avoid conflicts over MODULE_ALIAS_NETPROTO(PF_VSOCK).
>
> The patch also adds a module parameter "skip_hypervisor_check" for
> vmw_vsock_vmci_transport.ko.
>
> Signed-off-by: Dexuan Cui <decui@microsoft.com>
> Cc: Alok Kataria <akataria@vmware.com>
> Cc: Andy King <acking@vmware.com>
> Cc: Adit Ranadive <aditr@vmware.com>
> Cc: George Zhang <georgezhang@vmware.com>
> Cc: Jorgen Hansen <jhansen@vmware.com>
> Cc: K. Y. Srinivasan <kys@microsoft.com>
> Cc: Haiyang Zhang <haiyangz@microsoft.com>
> Cc: Stephen Hemminger <sthemmin@microsoft.com>
> ---
> net/vmw_vsock/Kconfig | 2 +-
> net/vmw_vsock/vmci_transport.c | 11 +++++++++++
> 2 files changed, 12 insertions(+), 1 deletion(-)
>
> diff --git a/net/vmw_vsock/Kconfig b/net/vmw_vsock/Kconfig
> index a24369d..3f52929 100644
> --- a/net/vmw_vsock/Kconfig
> +++ b/net/vmw_vsock/Kconfig
> @@ -17,7 +17,7 @@ config VSOCKETS
>
> config VMWARE_VMCI_VSOCKETS
> tristate "VMware VMCI transport for Virtual Sockets"
> - depends on VSOCKETS && VMWARE_VMCI
> + depends on VSOCKETS && VMWARE_VMCI && HYPERVISOR_GUEST
> help
> This module implements a VMCI transport for Virtual Sockets.
>
> diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
> index 10ae782..c068873 100644
> --- a/net/vmw_vsock/vmci_transport.c
> +++ b/net/vmw_vsock/vmci_transport.c
> @@ -16,6 +16,7 @@
> #include <linux/types.h>
> #include <linux/bitops.h>
> #include <linux/cred.h>
> +#include <linux/hypervisor.h>
> #include <linux/init.h>
> #include <linux/io.h>
> #include <linux/kernel.h>
> @@ -73,6 +74,10 @@ struct vmci_transport_recv_pkt_info {
> struct vmci_transport_packet pkt;
> };
>
> +static bool skip_hypervisor_check;
> +module_param(skip_hypervisor_check, bool, 0444);
> +MODULE_PARM_DESC(hot_add, "If set, attempt to load on non-VMware platforms");
> +
> static LIST_HEAD(vmci_transport_cleanup_list);
> static DEFINE_SPINLOCK(vmci_transport_cleanup_lock);
> static DECLARE_WORK(vmci_transport_cleanup_work, vmci_transport_cleanup);
> @@ -2085,6 +2090,12 @@ static int __init vmci_transport_init(void)
> {
> int err;
>
> + /* Check if we are running on VMware's hypervisor and bail out
> + * if we are not.
> + */
> + if (!skip_hypervisor_check && x86_hyper != &x86_hyper_vmware)
> + return -ENODEV;
> +
> /* Create the datagram handle that we will use to send and receive all
> * VSocket control messages for this context.
> */
> --
> 2.7.4
>
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 455 bytes --]
^ permalink raw reply
* Re: [PATCH net] PCI: fix the return value for the pci_find_pcie_root_port()
From: Ding Tianhong @ 2017-08-17 13:38 UTC (permalink / raw)
To: Thierry Reding
Cc: mark.rutland, gabriele.paoloni, asit.k.mallick, catalin.marinas,
will.deacon, linuxarm, alexander.duyck, ashok.raj, eric.dumazet,
helgaas, jeffrey.t.kirsher, linux-pci, ganeshgr, Bob.Shaw, leedom,
patrick.j.cramer, bhelgaas, werner, linux-arm-kernel, amira,
netdev, linux-kernel, David.Laight, Suravee.Suthikulpanit,
robin.murphy, davem, l.stach
In-Reply-To: <20170817133012.GC6854@ulmo>
On 2017/8/17 21:30, Thierry Reding wrote:
> On Thu, Aug 17, 2017 at 08:40:16PM +0800, Ding Tianhong wrote:
>>
>>
>> On 2017/8/17 18:51, Thierry Reding wrote:
>>> On Thu, Aug 17, 2017 at 10:25:30AM +0800, Ding Tianhong wrote:
>>>> The pci_find_pcie_root_port() would return NULL if the given
>>>> dev is already a Root Port, it looks like unfriendly to the
>>>> PCIe Root Port device, Thierry and Bjorn suggest to let this
>>>> function return the given dev under this circumstances.
>>>>
>>>> Fixes: 0e405232871d6 ("PCI: fix oops when try to find Root Port for a PCI device")
>>>> Suggested-by: Thierry Reding <thierry.reding@gmail.com>
>>>> Suggested-by: Bjorn Helgaas <helgaas@kernel.org>
>>>> Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
>>>> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com>
>>>> ---
>>>> drivers/pci/pci.c | 2 +-
>>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
>>>> index 7e2022f..352bb53 100644
>>>> --- a/drivers/pci/pci.c
>>>> +++ b/drivers/pci/pci.c
>>>> @@ -514,7 +514,7 @@ struct resource *pci_find_resource(struct pci_dev *dev, struct resource *res)
>>>> */
>>>> struct pci_dev *pci_find_pcie_root_port(struct pci_dev *dev)
>>>> {
>>>> - struct pci_dev *bridge, *highest_pcie_bridge = NULL;
>>>> + struct pci_dev *bridge, *highest_pcie_bridge = dev;
>>>>
>>>> bridge = pci_upstream_bridge(dev);
>>>> while (bridge && pci_is_pcie(bridge)) {
>>>
>>> I think this should actually be this change on top of a revert of commit
>>> 0e405232871d6 ("PCI: fix oops when try to find Root Port for a PCI
>>> device"). After the above change, the previous fix will have a redundant
>>> check because highest_pcie_bridge will never be NULL.
>>>
>>> Let me send out that version to clarify what I mean.
>>>
>>
>> Hi Thierry:
>>
>> The patch ("PCI: fix oops when try to find Root Port for a PCI device")
>> has been merge to the linus mainline tree before you found this deficiencies....
>
> I understand that. I'm just saying that there's no point keeping that
> change around because it no longer makes sense after we initialize the
> highest_pcie_bridge variable to dev.
>
Ok, NO problem.:)
> Thierry
>
^ permalink raw reply
* [net-next 12/15] net/mlx5e: Avoid using multiple blank lines
From: Saeed Mahameed @ 2017-08-17 13:30 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
To fix these checkpatch complaints:
CHECK: Please don't use multiple blank lines
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 1 -
drivers/net/ethernet/mellanox/mlx5/core/main.c | 1 -
2 files changed, 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index 31cbe5e86a01..0ef68a7c051e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -802,7 +802,6 @@ static void cmd_work_handler(struct work_struct *work)
bool poll_cmd = ent->polling;
int alloc_ret;
-
sem = ent->page_queue ? &cmd->pages_sem : &cmd->sem;
down(sem);
if (!ent->page_queue) {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 7e6e24398926..514c22d21729 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -836,7 +836,6 @@ static int mlx5_core_set_issi(struct mlx5_core_dev *dev)
return -EOPNOTSUPP;
}
-
static int mlx5_pci_init(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
{
struct pci_dev *pdev = dev->pdev;
--
2.13.0
^ permalink raw reply related
* [net-next 10/15] net/mlx5: Add a blank line after declarations
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
To fix these checkpatch complaints:
WARNING: Missing a blank line after declarations
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/alloc.c | 1 +
drivers/net/ethernet/mellanox/mlx5/core/eq.c | 1 +
2 files changed, 2 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
index 3c95f7f53802..47239bf7bf43 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
@@ -258,6 +258,7 @@ EXPORT_SYMBOL_GPL(mlx5_db_alloc);
void mlx5_db_free(struct mlx5_core_dev *dev, struct mlx5_db *db)
{
u32 db_per_page = PAGE_SIZE / cache_line_size();
+
mutex_lock(&dev->priv.pgdir_mutex);
__set_bit(db->index, db->u.pgdir->bitmap);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index de704ff5619a..a08027b8f3ce 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -188,6 +188,7 @@ static enum mlx5_dev_event port_subtype_event(u8 subtype)
static void eq_update_ci(struct mlx5_eq *eq, int arm)
{
__be32 __iomem *addr = eq->doorbell + (arm ? 0 : 2);
+
u32 val = (eq->cons_index & 0xffffff) | (eq->eqn << 24);
__raw_writel((__force u32)cpu_to_be32(val), addr);
/* We still want ordering, just not swabbing, so add a barrier */
--
2.13.0
^ permalink raw reply related
* [net-next 15/15] net/mlx5e: Use size_t to store byte offset in statistics descriptors
From: Saeed Mahameed @ 2017-08-17 13:30 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Gal Pressman, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
The byte offset of counter descriptors should be stored in size_t variable
instead of an integer.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 40b5c73e5e26..6761796e803c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -47,7 +47,7 @@
struct counter_desc {
char format[ETH_GSTRING_LEN];
- int offset; /* Byte offset */
+ size_t offset; /* Byte offset */
};
struct mlx5e_sw_stats {
--
2.13.0
^ permalink raw reply related
* [net-next 13/15] net/mlx5e: Place constants on the right side of comparisons
From: Saeed Mahameed @ 2017-08-17 13:30 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
To fix these checkpatch complaints:
WARNING: Comparisons should place the constant on the right side of the test
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 8e224bcbc6a6..55a6786d3c4c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -509,8 +509,8 @@ static void mlx5e_lro_update_hdr(struct sk_buff *skb, struct mlx5_cqe64 *cqe,
u16 tot_len;
u8 l4_hdr_type = get_cqe_l4_hdr_type(cqe);
- int tcp_ack = ((CQE_L4_HDR_TYPE_TCP_ACK_NO_DATA == l4_hdr_type) ||
- (CQE_L4_HDR_TYPE_TCP_ACK_AND_DATA == l4_hdr_type));
+ int tcp_ack = ((l4_hdr_type == CQE_L4_HDR_TYPE_TCP_ACK_NO_DATA) ||
+ (l4_hdr_type == CQE_L4_HDR_TYPE_TCP_ACK_AND_DATA));
skb->mac_len = ETH_HLEN;
proto = __vlan_get_protocol(skb, eth->h_proto, &network_depth);
--
2.13.0
^ permalink raw reply related
* [net-next 14/15] net/mlx5e: Use kernel types instead of uint*_t in ethtool callbacks
From: Saeed Mahameed @ 2017-08-17 13:30 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Gal Pressman, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
Fix checkpatch errors:
CHECK:PREFER_KERNEL_TYPES: Prefer kernel type 'u32' over 'uint32_t'
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 8 +++-----
drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c | 3 +--
2 files changed, 4 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index ed161312a773..03b130408359 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -206,7 +206,7 @@ static int mlx5e_get_sset_count(struct net_device *dev, int sset)
return mlx5e_ethtool_get_sset_count(priv, sset);
}
-static void mlx5e_fill_stats_strings(struct mlx5e_priv *priv, uint8_t *data)
+static void mlx5e_fill_stats_strings(struct mlx5e_priv *priv, u8 *data)
{
int i, j, tc, prio, idx = 0;
unsigned long pfc_combined;
@@ -308,8 +308,7 @@ static void mlx5e_fill_stats_strings(struct mlx5e_priv *priv, uint8_t *data)
priv->channel_tc2txq[i][tc]);
}
-void mlx5e_ethtool_get_strings(struct mlx5e_priv *priv,
- uint32_t stringset, uint8_t *data)
+void mlx5e_ethtool_get_strings(struct mlx5e_priv *priv, u32 stringset, u8 *data)
{
int i;
@@ -331,8 +330,7 @@ void mlx5e_ethtool_get_strings(struct mlx5e_priv *priv,
}
}
-static void mlx5e_get_strings(struct net_device *dev,
- uint32_t stringset, uint8_t *data)
+static void mlx5e_get_strings(struct net_device *dev, u32 stringset, u8 *data)
{
struct mlx5e_priv *priv = netdev_priv(dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
index dd49a59854e5..43c126c63955 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
@@ -43,8 +43,7 @@ static void mlx5i_get_drvinfo(struct net_device *dev,
sizeof(drvinfo->driver));
}
-static void mlx5i_get_strings(struct net_device *dev,
- uint32_t stringset, uint8_t *data)
+static void mlx5i_get_strings(struct net_device *dev, u32 stringset, u8 *data)
{
struct mlx5e_priv *priv = mlx5i_epriv(dev);
--
2.13.0
^ permalink raw reply related
* [net-next 06/15] net/mlx5: Add RX buffer fullness counters infrastructure
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Gal Pressman, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
Add capability bit in PCAM register and counters to PPCNT register.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
include/linux/mlx5/mlx5_ifc.h | 15 +++++++++++++--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index ba533b39c885..cf7ff52c594e 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -1538,7 +1538,17 @@ struct mlx5_ifc_eth_extended_cntrs_grp_data_layout_bits {
u8 port_transmit_wait_low[0x20];
- u8 reserved_at_40[0x780];
+ u8 reserved_at_40[0x100];
+
+ u8 rx_buffer_almost_full_high[0x20];
+
+ u8 rx_buffer_almost_full_low[0x20];
+
+ u8 rx_buffer_full_high[0x20];
+
+ u8 rx_buffer_full_low[0x20];
+
+ u8 reserved_at_1c0[0x600];
};
struct mlx5_ifc_eth_3635_cntrs_grp_data_layout_bits {
@@ -7723,8 +7733,9 @@ struct mlx5_ifc_peir_reg_bits {
};
struct mlx5_ifc_pcam_enhanced_features_bits {
- u8 reserved_at_0[0x7c];
+ u8 reserved_at_0[0x7b];
+ u8 rx_buffer_fullness_counters[0x1];
u8 ptys_connector_type[0x1];
u8 reserved_at_7d[0x1];
u8 ppcnt_discard_group[0x1];
--
2.13.0
^ permalink raw reply related
* [net-next 11/15] net/mlx5e: Properly indent within conditional statements
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
To fix these checkpatch complaints:
WARNING: suspect code indent for conditional statements (8, 24)
+ if (eth_proto & (MLX5E_PROT_MASK(MLX5E_10GBASE_SR)
[...]
+ return PORT_FIBRE;
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 22 +++++++++++-----------
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index a75ac4d11c5b..ed161312a773 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -988,23 +988,23 @@ static u8 get_connector_port(u32 eth_proto, u8 connector_type)
return ptys2connector_type[connector_type];
if (eth_proto & (MLX5E_PROT_MASK(MLX5E_10GBASE_SR)
- | MLX5E_PROT_MASK(MLX5E_40GBASE_SR4)
- | MLX5E_PROT_MASK(MLX5E_100GBASE_SR4)
- | MLX5E_PROT_MASK(MLX5E_1000BASE_CX_SGMII))) {
- return PORT_FIBRE;
+ | MLX5E_PROT_MASK(MLX5E_40GBASE_SR4)
+ | MLX5E_PROT_MASK(MLX5E_100GBASE_SR4)
+ | MLX5E_PROT_MASK(MLX5E_1000BASE_CX_SGMII))) {
+ return PORT_FIBRE;
}
if (eth_proto & (MLX5E_PROT_MASK(MLX5E_40GBASE_CR4)
- | MLX5E_PROT_MASK(MLX5E_10GBASE_CR)
- | MLX5E_PROT_MASK(MLX5E_100GBASE_CR4))) {
- return PORT_DA;
+ | MLX5E_PROT_MASK(MLX5E_10GBASE_CR)
+ | MLX5E_PROT_MASK(MLX5E_100GBASE_CR4))) {
+ return PORT_DA;
}
if (eth_proto & (MLX5E_PROT_MASK(MLX5E_10GBASE_KX4)
- | MLX5E_PROT_MASK(MLX5E_10GBASE_KR)
- | MLX5E_PROT_MASK(MLX5E_40GBASE_KR4)
- | MLX5E_PROT_MASK(MLX5E_100GBASE_KR4))) {
- return PORT_NONE;
+ | MLX5E_PROT_MASK(MLX5E_10GBASE_KR)
+ | MLX5E_PROT_MASK(MLX5E_40GBASE_KR4)
+ | MLX5E_PROT_MASK(MLX5E_100GBASE_KR4))) {
+ return PORT_NONE;
}
return PORT_OTHER;
--
2.13.0
^ permalink raw reply related
* [net-next 07/15] net/mlx5e: Add RX buffer fullness counters
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Gal Pressman, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
rx_buffer_passed_thres_phy - The number of events where the port RX
buffer has passed a fullness threshold.
rx_buffer_full_phy - The number of events where the port RX buffer has
reached 100% fullness.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 8 ++++++++
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 6 ++++++
drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 17 ++++++++++++++++-
3 files changed, 30 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 07202f7322fc..8c013a521319 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -242,6 +242,10 @@ static void mlx5e_fill_stats_strings(struct mlx5e_priv *priv, uint8_t *data)
strcpy(data + (idx++) * ETH_GSTRING_LEN,
pport_phy_statistical_stats_desc[i].format);
+ for (i = 0; i < NUM_PPORT_ETH_EXT_COUNTERS(priv); i++)
+ strcpy(data + (idx++) * ETH_GSTRING_LEN,
+ pport_eth_ext_stats_desc[i].format);
+
for (i = 0; i < NUM_PCIE_PERF_COUNTERS(priv); i++)
strcpy(data + (idx++) * ETH_GSTRING_LEN,
pcie_perf_stats_desc[i].format);
@@ -377,6 +381,10 @@ void mlx5e_ethtool_get_ethtool_stats(struct mlx5e_priv *priv,
data[idx++] = MLX5E_READ_CTR64_BE(&priv->stats.pport.phy_statistical_counters,
pport_phy_statistical_stats_desc, i);
+ for (i = 0; i < NUM_PPORT_ETH_EXT_COUNTERS(priv); i++)
+ data[idx++] = MLX5E_READ_CTR64_BE(&priv->stats.pport.eth_ext_counters,
+ pport_eth_ext_stats_desc, i);
+
for (i = 0; i < NUM_PCIE_PERF_COUNTERS(priv); i++)
data[idx++] = MLX5E_READ_CTR32_BE(&priv->stats.pcie.pcie_perf_counters,
pcie_perf_stats_desc, i);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 6921c6fe3087..2b0c891ad2a9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -288,6 +288,12 @@ static void mlx5e_update_pport_counters(struct mlx5e_priv *priv, bool full)
mlx5_core_access_reg(mdev, in, sz, out, sz, MLX5_REG_PPCNT, 0, 0);
}
+ if (MLX5_CAP_PCAM_FEATURE(mdev, rx_buffer_fullness_counters)) {
+ out = pstats->eth_ext_counters;
+ MLX5_SET(ppcnt_reg, in, grp, MLX5_ETHERNET_EXTENDED_COUNTERS_GROUP);
+ mlx5_core_access_reg(mdev, in, sz, out, sz, MLX5_REG_PPCNT, 0, 0);
+ }
+
MLX5_SET(ppcnt_reg, in, grp, MLX5_PER_PRIORITY_COUNTERS_GROUP);
for (prio = 0; prio < NUM_PPORT_PRIO; prio++) {
out = pstats->per_prio_counters[prio];
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index bdddddc46170..be49df4bedd9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -216,6 +216,12 @@ static const struct counter_desc vport_stats_desc[] = {
MLX5_GET64(ppcnt_reg, pstats->per_prio_counters[prio], \
counter_set.eth_per_prio_grp_data_layout.c##_high)
#define NUM_PPORT_PRIO 8
+#define PPORT_ETH_EXT_OFF(c) \
+ MLX5_BYTE_OFF(ppcnt_reg, \
+ counter_set.eth_extended_cntrs_grp_data_layout.c##_high)
+#define PPORT_ETH_EXT_GET(pstats, c) \
+ MLX5_GET64(ppcnt_reg, (pstats)->eth_ext_counters, \
+ counter_set.eth_extended_cntrs_grp_data_layout.c##_high)
struct mlx5e_pport_stats {
__be64 IEEE_802_3_counters[MLX5_ST_SZ_QW(ppcnt_reg)];
@@ -224,6 +230,7 @@ struct mlx5e_pport_stats {
__be64 per_prio_counters[NUM_PPORT_PRIO][MLX5_ST_SZ_QW(ppcnt_reg)];
__be64 phy_counters[MLX5_ST_SZ_QW(ppcnt_reg)];
__be64 phy_statistical_counters[MLX5_ST_SZ_QW(ppcnt_reg)];
+ __be64 eth_ext_counters[MLX5_ST_SZ_QW(ppcnt_reg)];
};
static const struct counter_desc pport_802_3_stats_desc[] = {
@@ -290,6 +297,10 @@ static const struct counter_desc pport_per_prio_pfc_stats_desc[] = {
{ "rx_%s_pause_transition", PPORT_PER_PRIO_OFF(rx_pause_transition) },
};
+static const struct counter_desc pport_eth_ext_stats_desc[] = {
+ { "rx_buffer_passed_thres_phy", PPORT_ETH_EXT_OFF(rx_buffer_almost_full) },
+};
+
#define PCIE_PERF_OFF(c) \
MLX5_BYTE_OFF(mpcnt_reg, counter_set.pcie_perf_cntrs_grp_data_layout.c)
#define PCIE_PERF_GET(pcie_stats, c) \
@@ -411,12 +422,16 @@ static const struct counter_desc sq_stats_desc[] = {
ARRAY_SIZE(pport_per_prio_traffic_stats_desc)
#define NUM_PPORT_PER_PRIO_PFC_COUNTERS \
ARRAY_SIZE(pport_per_prio_pfc_stats_desc)
+#define NUM_PPORT_ETH_EXT_COUNTERS(priv) \
+ (ARRAY_SIZE(pport_eth_ext_stats_desc) * \
+ MLX5_CAP_PCAM_FEATURE((priv)->mdev, rx_buffer_fullness_counters))
#define NUM_PPORT_COUNTERS(priv) (NUM_PPORT_802_3_COUNTERS + \
NUM_PPORT_2863_COUNTERS + \
NUM_PPORT_2819_COUNTERS + \
NUM_PPORT_PHY_STATISTICAL_COUNTERS(priv) + \
NUM_PPORT_PER_PRIO_TRAFFIC_COUNTERS * \
- NUM_PPORT_PRIO)
+ NUM_PPORT_PRIO + \
+ NUM_PPORT_ETH_EXT_COUNTERS(priv))
#define NUM_PCIE_COUNTERS(priv) (NUM_PCIE_PERF_COUNTERS(priv) + \
NUM_PCIE_PERF_STALL_COUNTERS(priv))
#define NUM_RQ_STATS ARRAY_SIZE(rq_stats_desc)
--
2.13.0
^ permalink raw reply related
* [net-next 09/15] net/mlx5: Avoid blank lines after/before open/close brace
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
To fix these checkpatch complaints:
CHECK: Blank lines aren't necessary after an open brace '{'
CHECK: Blank lines aren't necessary before a close brace '}'
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 1 -
drivers/net/ethernet/mellanox/mlx5/core/sriov.c | 1 -
2 files changed, 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index d453a11f41fe..a75ac4d11c5b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -176,7 +176,6 @@ static bool mlx5e_query_global_pause_combined(struct mlx5e_priv *priv)
int mlx5e_ethtool_get_sset_count(struct mlx5e_priv *priv, int sset)
{
-
switch (sset) {
case ETH_SS_STATS:
return NUM_SW_COUNTERS +
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
index 5e7ffc9fad78..55b07c5ecd12 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
@@ -71,7 +71,6 @@ static int mlx5_device_enable_sriov(struct mlx5_core_dev *dev, int num_vfs)
sriov->vfs_ctx[vf].enabled = 1;
sriov->enabled_vfs++;
mlx5_core_dbg(dev, "successfully enabled VF* %d\n", vf);
-
}
return 0;
--
2.13.0
^ permalink raw reply related
* [net-next 08/15] net/mlx5e: Add outbound PCI buffer overflow counter
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Eran Ben Elisha, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Eran Ben Elisha <eranbe@mellanox.com>
Add outbound_pci_buffer_overflow to ethtool output for monitoring the
number of packets that were dropped due to lack of PCIe buffers on
receive path from NIC port toward the host(s).
This counter is valid only in case that tx_overflow_buffer_pkt is
supported in MCAM enhanced features.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 12 ++++++++++--
drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 14 ++++++++++++++
include/linux/mlx5/mlx5_ifc.h | 6 ++++--
3 files changed, 28 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 8c013a521319..d453a11f41fe 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -250,9 +250,13 @@ static void mlx5e_fill_stats_strings(struct mlx5e_priv *priv, uint8_t *data)
strcpy(data + (idx++) * ETH_GSTRING_LEN,
pcie_perf_stats_desc[i].format);
- for (i = 0; i < NUM_PCIE_PERF_STALL_COUNTERS(priv); i++)
+ for (i = 0; i < NUM_PCIE_PERF_COUNTERS64(priv); i++)
strcpy(data + (idx++) * ETH_GSTRING_LEN,
- pcie_perf_stall_stats_desc[i].format);
+ pcie_perf_stats_desc64[i].format);
+
+ for (i = 0; i < NUM_PCIE_PERF_STALL_COUNTERS(priv); i++)
+ strcpy(data + (idx++) * ETH_GSTRING_LEN,
+ pcie_perf_stall_stats_desc[i].format);
for (prio = 0; prio < NUM_PPORT_PRIO; prio++) {
for (i = 0; i < NUM_PPORT_PER_PRIO_TRAFFIC_COUNTERS; i++)
@@ -389,6 +393,10 @@ void mlx5e_ethtool_get_ethtool_stats(struct mlx5e_priv *priv,
data[idx++] = MLX5E_READ_CTR32_BE(&priv->stats.pcie.pcie_perf_counters,
pcie_perf_stats_desc, i);
+ for (i = 0; i < NUM_PCIE_PERF_COUNTERS64(priv); i++)
+ data[idx++] = MLX5E_READ_CTR64_BE(&priv->stats.pcie.pcie_perf_counters,
+ pcie_perf_stats_desc64, i);
+
for (i = 0; i < NUM_PCIE_PERF_STALL_COUNTERS(priv); i++)
data[idx++] = MLX5E_READ_CTR32_BE(&priv->stats.pcie.pcie_perf_counters,
pcie_perf_stall_stats_desc, i);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index be49df4bedd9..40b5c73e5e26 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -307,6 +307,12 @@ static const struct counter_desc pport_eth_ext_stats_desc[] = {
MLX5_GET(mpcnt_reg, (pcie_stats)->pcie_perf_counters, \
counter_set.pcie_perf_cntrs_grp_data_layout.c)
+#define PCIE_PERF_OFF64(c) \
+ MLX5_BYTE_OFF(mpcnt_reg, counter_set.pcie_perf_cntrs_grp_data_layout.c##_high)
+#define PCIE_PERF_GET64(pcie_stats, c) \
+ MLX5_GET64(mpcnt_reg, (pcie_stats)->pcie_perf_counters, \
+ counter_set.pcie_perf_cntrs_grp_data_layout.c##_high)
+
struct mlx5e_pcie_stats {
__be64 pcie_perf_counters[MLX5_ST_SZ_QW(mpcnt_reg)];
};
@@ -316,6 +322,10 @@ static const struct counter_desc pcie_perf_stats_desc[] = {
{ "tx_pci_signal_integrity", PCIE_PERF_OFF(tx_errors) },
};
+static const struct counter_desc pcie_perf_stats_desc64[] = {
+ { "outbound_pci_buffer_overflow", PCIE_PERF_OFF64(tx_overflow_buffer_pkt) },
+};
+
static const struct counter_desc pcie_perf_stall_stats_desc[] = {
{ "outbound_pci_stalled_rd", PCIE_PERF_OFF(outbound_stalled_reads) },
{ "outbound_pci_stalled_wr", PCIE_PERF_OFF(outbound_stalled_writes) },
@@ -415,6 +425,9 @@ static const struct counter_desc sq_stats_desc[] = {
#define NUM_PCIE_PERF_COUNTERS(priv) \
(ARRAY_SIZE(pcie_perf_stats_desc) * \
MLX5_CAP_MCAM_FEATURE((priv)->mdev, pcie_performance_group))
+#define NUM_PCIE_PERF_COUNTERS64(priv) \
+ (ARRAY_SIZE(pcie_perf_stats_desc64) * \
+ MLX5_CAP_MCAM_FEATURE((priv)->mdev, tx_overflow_buffer_pkt))
#define NUM_PCIE_PERF_STALL_COUNTERS(priv) \
(ARRAY_SIZE(pcie_perf_stall_stats_desc) * \
MLX5_CAP_MCAM_FEATURE((priv)->mdev, pcie_outbound_stalled))
@@ -433,6 +446,7 @@ static const struct counter_desc sq_stats_desc[] = {
NUM_PPORT_PRIO + \
NUM_PPORT_ETH_EXT_COUNTERS(priv))
#define NUM_PCIE_COUNTERS(priv) (NUM_PCIE_PERF_COUNTERS(priv) + \
+ NUM_PCIE_PERF_COUNTERS64(priv) +\
NUM_PCIE_PERF_STALL_COUNTERS(priv))
#define NUM_RQ_STATS ARRAY_SIZE(rq_stats_desc)
#define NUM_SQ_STATS ARRAY_SIZE(sq_stats_desc)
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index cf7ff52c594e..ae7d09b9c52f 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -1864,7 +1864,9 @@ struct mlx5_ifc_pcie_perf_cntrs_grp_data_layout_bits {
u8 crc_error_tlp[0x20];
- u8 reserved_at_140[0x40];
+ u8 tx_overflow_buffer_pkt_high[0x20];
+
+ u8 tx_overflow_buffer_pkt_low[0x20];
u8 outbound_stalled_reads[0x20];
@@ -7767,7 +7769,7 @@ struct mlx5_ifc_pcam_reg_bits {
struct mlx5_ifc_mcam_enhanced_features_bits {
u8 reserved_at_0[0x7b];
u8 pcie_outbound_stalled[0x1];
- u8 reserved_at_7c[0x1];
+ u8 tx_overflow_buffer_pkt[0x1];
u8 mtpps_enh_out_per_adj[0x1];
u8 mtpps_fs[0x1];
u8 pcie_performance_group[0x1];
--
2.13.0
^ permalink raw reply related
* [pull request][net-next 00/15] Mellanox, mlx5 updates 2017-08-17
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Saeed Mahameed
Hi Dave,
Tthe following changes provide updates for mlx5 ethernet and IPoIB
netdevice driver.
For more details please see tag log message below.
Please pull and let me know if there's any problem.
Thanks,
Saeed.
---
The following changes since commit f03b488b2fb1af20a290509b1ed25db08391b4e7:
Merge branch 'vmbus-sendpacket-cleanups' (2017-08-16 16:27:45 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-updates-2017-08-17
for you to fetch changes up to 9a1ceef8a42cb90406b6adc620e63e9a11e04dbb:
net/mlx5e: Use size_t to store byte offset in statistics descriptors (2017-08-17 15:57:15 +0300)
----------------------------------------------------------------
mlx5-updates-2017-08-17
Some updates for mlx5 ethernet and IPoIB device driver.
Eran added the support for manage physical link state from netdevice upon
interface open/close requests.
Feras fixed the driver name showed in ethtool for IPoIB interfaces.
Shalom Added the support for IPoIB netdevice ethtool get link settings.
Gal and Eran exposed new diagnostic counters for outbound PCIe stalls and overflow
and RX buffer fullness statistics.
Code cleanups from Or Gerlitz.
Variable types cleanup from Gal.
Thanks,
Saeed.
----------------------------------------------------------------
Eran Ben Elisha (2):
net/mlx5e: Send PAOS command on interface up/down
net/mlx5e: Add outbound PCI buffer overflow counter
Feras Daoud (1):
net/mlx5e: IPoIB, Fix driver name retrieved by ethtool
Gal Pressman (6):
net/mlx5: Add PCIe outbound stalls counters infrastructure
net/mlx5e: Add PCIe outbound stalls counters
net/mlx5: Add RX buffer fullness counters infrastructure
net/mlx5e: Add RX buffer fullness counters
net/mlx5e: Use kernel types instead of uint*_t in ethtool callbacks
net/mlx5e: Use size_t to store byte offset in statistics descriptors
Or Gerlitz (5):
net/mlx5: Avoid blank lines after/before open/close brace
net/mlx5: Add a blank line after declarations
net/mlx5e: Properly indent within conditional statements
net/mlx5e: Avoid using multiple blank lines
net/mlx5e: Place constants on the right side of comparisons
Shalom Lagziel (1):
net/mlx5e: IPoIB, Add support for get_link_ksettings in ethtool
drivers/net/ethernet/mellanox/mlx5/core/alloc.c | 1 +
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 1 -
.../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 55 ++++++---
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 13 ++
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 20 +--
drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 4 +-
drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 46 ++++++-
drivers/net/ethernet/mellanox/mlx5/core/eq.c | 1 +
.../ethernet/mellanox/mlx5/core/ipoib/ethtool.c | 135 ++++++++++++++++++---
drivers/net/ethernet/mellanox/mlx5/core/main.c | 1 -
drivers/net/ethernet/mellanox/mlx5/core/sriov.c | 1 -
include/linux/mlx5/mlx5_ifc.h | 34 +++++-
12 files changed, 261 insertions(+), 51 deletions(-)
^ permalink raw reply
* [net-next 02/15] net/mlx5e: IPoIB, Fix driver name retrieved by ethtool
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Feras Daoud, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
Printing an enhanced IPoIB device information using
"ethtool -i DEVNAME", prints the low level driver name: mlx5_core.
This commit changes the name to mlx5_core [ib_ipoib], to include the
ipoib device driver infromation.
Fixes: 076b0936e5fb ("net/mlx5e: IPoIB, Add ethtool support")
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
index eb04e97d8765..b080fabfe8de 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
@@ -39,6 +39,8 @@ static void mlx5i_get_drvinfo(struct net_device *dev,
struct mlx5e_priv *priv = mlx5i_epriv(dev);
mlx5e_ethtool_get_drvinfo(priv, drvinfo);
+ strlcpy(drvinfo->driver, DRIVER_NAME "[ib_ipoib]",
+ sizeof(drvinfo->driver));
}
static void mlx5i_get_strings(struct net_device *dev,
--
2.13.0
^ permalink raw reply related
* [net-next 04/15] net/mlx5: Add PCIe outbound stalls counters infrastructure
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Gal Pressman, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
Add capability bit in MCAM register and counters to MPCNT register.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
include/linux/mlx5/mlx5_ifc.h | 17 ++++++++++++++---
1 file changed, 14 insertions(+), 3 deletions(-)
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index c99daffc3c3c..ba533b39c885 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -1854,7 +1854,17 @@ struct mlx5_ifc_pcie_perf_cntrs_grp_data_layout_bits {
u8 crc_error_tlp[0x20];
- u8 reserved_at_140[0x680];
+ u8 reserved_at_140[0x40];
+
+ u8 outbound_stalled_reads[0x20];
+
+ u8 outbound_stalled_writes[0x20];
+
+ u8 outbound_stalled_reads_events[0x20];
+
+ u8 outbound_stalled_writes_events[0x20];
+
+ u8 reserved_at_200[0x5c0];
};
struct mlx5_ifc_cmd_inter_comp_event_bits {
@@ -7744,8 +7754,9 @@ struct mlx5_ifc_pcam_reg_bits {
};
struct mlx5_ifc_mcam_enhanced_features_bits {
- u8 reserved_at_0[0x7d];
-
+ u8 reserved_at_0[0x7b];
+ u8 pcie_outbound_stalled[0x1];
+ u8 reserved_at_7c[0x1];
u8 mtpps_enh_out_per_adj[0x1];
u8 mtpps_fs[0x1];
u8 pcie_performance_group[0x1];
--
2.13.0
^ permalink raw reply related
* [net-next 03/15] net/mlx5e: IPoIB, Add support for get_link_ksettings in ethtool
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Shalom Lagziel, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Shalom Lagziel <shaloml@mellanox.com>
Add support for "ethtool DEVNAME" over ipoib ports,
Display standard port information for IPoIB netdevices using ethtool
For example:
$ ethtool ib2
> Settings for ib2:
Supported ports: [ ]
Supported link modes: Not reported
Supported pause frame use: No
Supports auto-negotiation: No
Advertised link modes: Not reported
Advertised pause frame use: No
Advertised auto-negotiation: No
Speed: 100000Mb/s
Duplex: Full
Port: Other
PHYAD: 0
Transceiver: internal
Auto-negotiation: off
Link detected: yes
Signed-off-by: Shalom Lagziel <shaloml@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../ethernet/mellanox/mlx5/core/ipoib/ethtool.c | 130 +++++++++++++++++++--
1 file changed, 118 insertions(+), 12 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
index b080fabfe8de..dd49a59854e5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
@@ -131,17 +131,123 @@ static int mlx5i_flash_device(struct net_device *netdev,
return mlx5e_ethtool_flash_device(priv, flash);
}
+enum mlx5_ptys_width {
+ MLX5_PTYS_WIDTH_1X = 1 << 0,
+ MLX5_PTYS_WIDTH_2X = 1 << 1,
+ MLX5_PTYS_WIDTH_4X = 1 << 2,
+ MLX5_PTYS_WIDTH_8X = 1 << 3,
+ MLX5_PTYS_WIDTH_12X = 1 << 4,
+};
+
+static inline int mlx5_ptys_width_enum_to_int(enum mlx5_ptys_width width)
+{
+ switch (width) {
+ case MLX5_PTYS_WIDTH_1X: return 1;
+ case MLX5_PTYS_WIDTH_2X: return 2;
+ case MLX5_PTYS_WIDTH_4X: return 4;
+ case MLX5_PTYS_WIDTH_8X: return 8;
+ case MLX5_PTYS_WIDTH_12X: return 12;
+ default: return -1;
+ }
+}
+
+enum mlx5_ptys_rate {
+ MLX5_PTYS_RATE_SDR = 1 << 0,
+ MLX5_PTYS_RATE_DDR = 1 << 1,
+ MLX5_PTYS_RATE_QDR = 1 << 2,
+ MLX5_PTYS_RATE_FDR10 = 1 << 3,
+ MLX5_PTYS_RATE_FDR = 1 << 4,
+ MLX5_PTYS_RATE_EDR = 1 << 5,
+ MLX5_PTYS_RATE_HDR = 1 << 6,
+};
+
+static inline int mlx5_ptys_rate_enum_to_int(enum mlx5_ptys_rate rate)
+{
+ switch (rate) {
+ case MLX5_PTYS_RATE_SDR: return 2500;
+ case MLX5_PTYS_RATE_DDR: return 5000;
+ case MLX5_PTYS_RATE_QDR:
+ case MLX5_PTYS_RATE_FDR10: return 10000;
+ case MLX5_PTYS_RATE_FDR: return 14000;
+ case MLX5_PTYS_RATE_EDR: return 25000;
+ case MLX5_PTYS_RATE_HDR: return 50000;
+ default: return -1;
+ }
+}
+
+static int mlx5i_get_port_settings(struct net_device *netdev,
+ u16 *ib_link_width_oper, u16 *ib_proto_oper)
+{
+ struct mlx5e_priv *priv = mlx5i_epriv(netdev);
+ struct mlx5_core_dev *mdev = priv->mdev;
+ u32 out[MLX5_ST_SZ_DW(ptys_reg)] = {0};
+ int ret;
+
+ ret = mlx5_query_port_ptys(mdev, out, sizeof(out), MLX5_PTYS_IB, 1);
+ if (ret)
+ return ret;
+
+ *ib_link_width_oper = MLX5_GET(ptys_reg, out, ib_link_width_oper);
+ *ib_proto_oper = MLX5_GET(ptys_reg, out, ib_proto_oper);
+
+ return 0;
+}
+
+static int mlx5i_get_speed_settings(u16 ib_link_width_oper, u16 ib_proto_oper)
+{
+ int rate, width;
+
+ rate = mlx5_ptys_rate_enum_to_int(ib_proto_oper);
+ if (rate < 0)
+ return -EINVAL;
+ width = mlx5_ptys_width_enum_to_int(ib_link_width_oper);
+ if (width < 0)
+ return -EINVAL;
+
+ return rate * width;
+}
+
+static int mlx5i_get_link_ksettings(struct net_device *netdev,
+ struct ethtool_link_ksettings *link_ksettings)
+{
+ u16 ib_link_width_oper;
+ u16 ib_proto_oper;
+ int speed, ret;
+
+ ret = mlx5i_get_port_settings(netdev, &ib_link_width_oper, &ib_proto_oper);
+ if (ret)
+ return ret;
+
+ ethtool_link_ksettings_zero_link_mode(link_ksettings, supported);
+ ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising);
+
+ speed = mlx5i_get_speed_settings(ib_link_width_oper, ib_proto_oper);
+ if (speed < 0)
+ return -EINVAL;
+
+ link_ksettings->base.duplex = DUPLEX_FULL;
+ link_ksettings->base.port = PORT_OTHER;
+
+ link_ksettings->base.autoneg = AUTONEG_DISABLE;
+
+ link_ksettings->base.speed = speed;
+
+ return 0;
+}
+
const struct ethtool_ops mlx5i_ethtool_ops = {
- .get_drvinfo = mlx5i_get_drvinfo,
- .get_strings = mlx5i_get_strings,
- .get_sset_count = mlx5i_get_sset_count,
- .get_ethtool_stats = mlx5i_get_ethtool_stats,
- .get_ringparam = mlx5i_get_ringparam,
- .set_ringparam = mlx5i_set_ringparam,
- .flash_device = mlx5i_flash_device,
- .get_channels = mlx5i_get_channels,
- .set_channels = mlx5i_set_channels,
- .get_coalesce = mlx5i_get_coalesce,
- .set_coalesce = mlx5i_set_coalesce,
- .get_ts_info = mlx5i_get_ts_info,
+ .get_drvinfo = mlx5i_get_drvinfo,
+ .get_strings = mlx5i_get_strings,
+ .get_sset_count = mlx5i_get_sset_count,
+ .get_ethtool_stats = mlx5i_get_ethtool_stats,
+ .get_ringparam = mlx5i_get_ringparam,
+ .set_ringparam = mlx5i_set_ringparam,
+ .flash_device = mlx5i_flash_device,
+ .get_channels = mlx5i_get_channels,
+ .set_channels = mlx5i_set_channels,
+ .get_coalesce = mlx5i_get_coalesce,
+ .set_coalesce = mlx5i_set_coalesce,
+ .get_ts_info = mlx5i_get_ts_info,
+ .get_link_ksettings = mlx5i_get_link_ksettings,
+ .get_link = ethtool_op_get_link,
};
--
2.13.0
^ permalink raw reply related
* [net-next 05/15] net/mlx5e: Add PCIe outbound stalls counters
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Gal Pressman, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
outbound_pci_stalled_rd - The percentage of time within the last second
that the NIC had outbound non-posted read requests but could not perform
the operation due to insufficient non-posted credits.
outbound_pci_stalled_wr - The percentage of time within the
last second that the NIC had outbound posted writes requests but could
not perform the operation due to insufficient posted credits.
outbound_pci_stalled_rd_events - The number of events where
outbound_pci_stalled_rd was above the threshold.
outbound_pci_stalled_wr_events - The number of events where
outbound_pci_stalled_wr was above the threshold.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 8 ++++++++
drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 13 ++++++++++++-
2 files changed, 20 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 917fade5f5d5..07202f7322fc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -246,6 +246,10 @@ static void mlx5e_fill_stats_strings(struct mlx5e_priv *priv, uint8_t *data)
strcpy(data + (idx++) * ETH_GSTRING_LEN,
pcie_perf_stats_desc[i].format);
+ for (i = 0; i < NUM_PCIE_PERF_STALL_COUNTERS(priv); i++)
+ strcpy(data + (idx++) * ETH_GSTRING_LEN,
+ pcie_perf_stall_stats_desc[i].format);
+
for (prio = 0; prio < NUM_PPORT_PRIO; prio++) {
for (i = 0; i < NUM_PPORT_PER_PRIO_TRAFFIC_COUNTERS; i++)
sprintf(data + (idx++) * ETH_GSTRING_LEN,
@@ -377,6 +381,10 @@ void mlx5e_ethtool_get_ethtool_stats(struct mlx5e_priv *priv,
data[idx++] = MLX5E_READ_CTR32_BE(&priv->stats.pcie.pcie_perf_counters,
pcie_perf_stats_desc, i);
+ for (i = 0; i < NUM_PCIE_PERF_STALL_COUNTERS(priv); i++)
+ data[idx++] = MLX5E_READ_CTR32_BE(&priv->stats.pcie.pcie_perf_counters,
+ pcie_perf_stall_stats_desc, i);
+
for (prio = 0; prio < NUM_PPORT_PRIO; prio++) {
for (i = 0; i < NUM_PPORT_PER_PRIO_TRAFFIC_COUNTERS; i++)
data[idx++] = MLX5E_READ_CTR64_BE(&priv->stats.pport.per_prio_counters[prio],
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index e65517eafc58..bdddddc46170 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -305,6 +305,13 @@ static const struct counter_desc pcie_perf_stats_desc[] = {
{ "tx_pci_signal_integrity", PCIE_PERF_OFF(tx_errors) },
};
+static const struct counter_desc pcie_perf_stall_stats_desc[] = {
+ { "outbound_pci_stalled_rd", PCIE_PERF_OFF(outbound_stalled_reads) },
+ { "outbound_pci_stalled_wr", PCIE_PERF_OFF(outbound_stalled_writes) },
+ { "outbound_pci_stalled_rd_events", PCIE_PERF_OFF(outbound_stalled_reads_events) },
+ { "outbound_pci_stalled_wr_events", PCIE_PERF_OFF(outbound_stalled_writes_events) },
+};
+
struct mlx5e_rq_stats {
u64 packets;
u64 bytes;
@@ -397,6 +404,9 @@ static const struct counter_desc sq_stats_desc[] = {
#define NUM_PCIE_PERF_COUNTERS(priv) \
(ARRAY_SIZE(pcie_perf_stats_desc) * \
MLX5_CAP_MCAM_FEATURE((priv)->mdev, pcie_performance_group))
+#define NUM_PCIE_PERF_STALL_COUNTERS(priv) \
+ (ARRAY_SIZE(pcie_perf_stall_stats_desc) * \
+ MLX5_CAP_MCAM_FEATURE((priv)->mdev, pcie_outbound_stalled))
#define NUM_PPORT_PER_PRIO_TRAFFIC_COUNTERS \
ARRAY_SIZE(pport_per_prio_traffic_stats_desc)
#define NUM_PPORT_PER_PRIO_PFC_COUNTERS \
@@ -407,7 +417,8 @@ static const struct counter_desc sq_stats_desc[] = {
NUM_PPORT_PHY_STATISTICAL_COUNTERS(priv) + \
NUM_PPORT_PER_PRIO_TRAFFIC_COUNTERS * \
NUM_PPORT_PRIO)
-#define NUM_PCIE_COUNTERS(priv) NUM_PCIE_PERF_COUNTERS(priv)
+#define NUM_PCIE_COUNTERS(priv) (NUM_PCIE_PERF_COUNTERS(priv) + \
+ NUM_PCIE_PERF_STALL_COUNTERS(priv))
#define NUM_RQ_STATS ARRAY_SIZE(rq_stats_desc)
#define NUM_SQ_STATS ARRAY_SIZE(sq_stats_desc)
--
2.13.0
^ permalink raw reply related
* [net-next 01/15] net/mlx5e: Send PAOS command on interface up/down
From: Saeed Mahameed @ 2017-08-17 13:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Eran Ben Elisha, Saeed Mahameed
In-Reply-To: <20170817133003.16900-1-saeedm@mellanox.com>
From: Eran Ben Elisha <eranbe@mellanox.com>
Upon interface up/down, driver will send POAS (Ports Administrative and
Operational Status Register) in order to inform the Firmware on the
desired status of the port by the driver.
Since now we might change physical link status on mlx5e_open/close,
logical VF representor should not use mlx5e_open/close ndos as is, and
should call the logical version mlx5e_open/closed_locked.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 7 +++++++
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 20 +++++++++++++-------
2 files changed, 20 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 8633ca5af6ed..6921c6fe3087 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -2682,6 +2682,8 @@ int mlx5e_open(struct net_device *netdev)
mutex_lock(&priv->state_lock);
err = mlx5e_open_locked(netdev);
+ if (!err)
+ mlx5_set_port_admin_status(priv->mdev, MLX5_PORT_UP);
mutex_unlock(&priv->state_lock);
return err;
@@ -2716,6 +2718,7 @@ int mlx5e_close(struct net_device *netdev)
return -ENODEV;
mutex_lock(&priv->state_lock);
+ mlx5_set_port_admin_status(priv->mdev, MLX5_PORT_DOWN);
err = mlx5e_close_locked(netdev);
mutex_unlock(&priv->state_lock);
@@ -4187,6 +4190,10 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv)
mlx5e_init_l2_addr(priv);
+ /* Marking the link as currently not needed by the Driver */
+ if (!netif_running(netdev))
+ mlx5_set_port_admin_status(mdev, MLX5_PORT_DOWN);
+
/* MTU range: 68 - hw-specific max */
netdev->min_mtu = ETH_MIN_MTU;
mlx5_query_port_max_mtu(priv->mdev, &max_mtu, 1);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index f34c00fbf78c..5e0f2d234ace 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -613,15 +613,18 @@ static int mlx5e_rep_open(struct net_device *dev)
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
int err;
- err = mlx5e_open(dev);
+ mutex_lock(&priv->state_lock);
+ err = mlx5e_open_locked(dev);
if (err)
- return err;
+ goto unlock;
- err = mlx5_eswitch_set_vport_state(esw, rep->vport, MLX5_ESW_VPORT_ADMIN_STATE_UP);
- if (!err)
+ if (!mlx5_eswitch_set_vport_state(esw, rep->vport,
+ MLX5_ESW_VPORT_ADMIN_STATE_UP))
netif_carrier_on(dev);
- return 0;
+unlock:
+ mutex_unlock(&priv->state_lock);
+ return err;
}
static int mlx5e_rep_close(struct net_device *dev)
@@ -630,10 +633,13 @@ static int mlx5e_rep_close(struct net_device *dev)
struct mlx5e_rep_priv *rpriv = priv->ppriv;
struct mlx5_eswitch_rep *rep = rpriv->rep;
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
+ int ret;
+ mutex_lock(&priv->state_lock);
(void)mlx5_eswitch_set_vport_state(esw, rep->vport, MLX5_ESW_VPORT_ADMIN_STATE_DOWN);
-
- return mlx5e_close(dev);
+ ret = mlx5e_close_locked(dev);
+ mutex_unlock(&priv->state_lock);
+ return ret;
}
static int mlx5e_rep_get_phys_port_name(struct net_device *dev,
--
2.13.0
^ permalink raw reply related
* Re: [PATCH net] PCI: fix the return value for the pci_find_pcie_root_port()
From: Thierry Reding @ 2017-08-17 13:30 UTC (permalink / raw)
To: Ding Tianhong
Cc: leedom, ashok.raj, bhelgaas, helgaas, werner, ganeshgr,
asit.k.mallick, patrick.j.cramer, Suravee.Suthikulpanit, Bob.Shaw,
l.stach, amira, gabriele.paoloni, David.Laight, jeffrey.t.kirsher,
catalin.marinas, will.deacon, mark.rutland, robin.murphy, davem,
alexander.duyck, eric.dumazet, linux-arm-kernel, netdev,
linux-pci, linux-kernel, linuxarm
In-Reply-To: <ea83e7a8-39a4-5b94-1148-5bf4ae4ed214@huawei.com>
[-- Attachment #1: Type: text/plain, Size: 2188 bytes --]
On Thu, Aug 17, 2017 at 08:40:16PM +0800, Ding Tianhong wrote:
>
>
> On 2017/8/17 18:51, Thierry Reding wrote:
> > On Thu, Aug 17, 2017 at 10:25:30AM +0800, Ding Tianhong wrote:
> >> The pci_find_pcie_root_port() would return NULL if the given
> >> dev is already a Root Port, it looks like unfriendly to the
> >> PCIe Root Port device, Thierry and Bjorn suggest to let this
> >> function return the given dev under this circumstances.
> >>
> >> Fixes: 0e405232871d6 ("PCI: fix oops when try to find Root Port for a PCI device")
> >> Suggested-by: Thierry Reding <thierry.reding@gmail.com>
> >> Suggested-by: Bjorn Helgaas <helgaas@kernel.org>
> >> Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
> >> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com>
> >> ---
> >> drivers/pci/pci.c | 2 +-
> >> 1 file changed, 1 insertion(+), 1 deletion(-)
> >>
> >> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> >> index 7e2022f..352bb53 100644
> >> --- a/drivers/pci/pci.c
> >> +++ b/drivers/pci/pci.c
> >> @@ -514,7 +514,7 @@ struct resource *pci_find_resource(struct pci_dev *dev, struct resource *res)
> >> */
> >> struct pci_dev *pci_find_pcie_root_port(struct pci_dev *dev)
> >> {
> >> - struct pci_dev *bridge, *highest_pcie_bridge = NULL;
> >> + struct pci_dev *bridge, *highest_pcie_bridge = dev;
> >>
> >> bridge = pci_upstream_bridge(dev);
> >> while (bridge && pci_is_pcie(bridge)) {
> >
> > I think this should actually be this change on top of a revert of commit
> > 0e405232871d6 ("PCI: fix oops when try to find Root Port for a PCI
> > device"). After the above change, the previous fix will have a redundant
> > check because highest_pcie_bridge will never be NULL.
> >
> > Let me send out that version to clarify what I mean.
> >
>
> Hi Thierry:
>
> The patch ("PCI: fix oops when try to find Root Port for a PCI device")
> has been merge to the linus mainline tree before you found this deficiencies....
I understand that. I'm just saying that there's no point keeping that
change around because it no longer makes sense after we initialize the
highest_pcie_bridge variable to dev.
Thierry
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply
* [PATCH] net: ibm: ibmvnic: constify vio_device_id
From: Arvind Yadav @ 2017-08-17 13:22 UTC (permalink / raw)
To: tlfalcon, benh, jallen, mpe, davem; +Cc: linux-kernel, linuxppc-dev, netdev
In-Reply-To: <b8393cdf646fae5ccbb64aab482c133fd54a7339.1502975929.git.arvind.yadav.cs@gmail.com>
vio_device_id are not supposed to change at runtime. All functions
working with vio_device_id provided by <asm/vio.h> work with
const vio_device_id. So mark the non-const structs as const.
Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
---
drivers/net/ethernet/ibm/ibmvnic.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c
index a3e6946..d5372b5 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.c
+++ b/drivers/net/ethernet/ibm/ibmvnic.c
@@ -3859,7 +3859,7 @@ static int ibmvnic_resume(struct device *dev)
return 0;
}
-static struct vio_device_id ibmvnic_device_table[] = {
+static const struct vio_device_id ibmvnic_device_table[] = {
{"network", "IBM,vnic"},
{"", "" }
};
--
2.7.4
^ permalink raw reply related
* [PATCH] net: ibm: ibmveth: constify vio_device_id
From: Arvind Yadav @ 2017-08-17 13:22 UTC (permalink / raw)
To: tlfalcon, benh, jallen, mpe, davem; +Cc: linux-kernel, linuxppc-dev, netdev
vio_device_id are not supposed to change at runtime. All functions
working with vio_device_id provided by <asm/vio.h> work with
const vio_device_id. So mark the non-const structs as const.
Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
---
drivers/net/ethernet/ibm/ibmveth.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c
index d17c2b0..f210398 100644
--- a/drivers/net/ethernet/ibm/ibmveth.c
+++ b/drivers/net/ethernet/ibm/ibmveth.c
@@ -1897,7 +1897,7 @@ static int ibmveth_resume(struct device *dev)
return 0;
}
-static struct vio_device_id ibmveth_device_table[] = {
+static const struct vio_device_id ibmveth_device_table[] = {
{ "network", "IBM,l-lan"},
{ "", "" }
};
--
2.7.4
^ permalink raw reply related
* Re: TCP fast retransmit issues
From: Jeremy Harris @ 2017-08-17 13:20 UTC (permalink / raw)
Cc: netdev
In-Reply-To: <20170728072750.GA6437@1wt.eu>
On 28/07/17 08:27, Willy Tarreau wrote:
> I didn't want to suggest names but since you did it first ;-) Indeed it's
> mostly on the same device that I've been bothered a lot by their annoying
> randomization. I used to know by memory the exact command to type to disable
> it, but I don't anymore (something along "no randomization").
https://supportforums.cisco.com/document/48551/single-tcp-flow-performance-firewall-services-module-fwsm#TCP_Sequence_Number_Randomization_and_SACK
https://supportcenter.checkpoint.com/supportcenter/portal?eventSubmit_doGoviewsolutiondetails=&solutionid=sk74640
--
Cheers,
Jeremy
^ permalink raw reply
* [PATCH net-next] bpf: no need to nullify ri->map in xdp_do_redirect
From: Daniel Borkmann @ 2017-08-17 13:07 UTC (permalink / raw)
To: davem; +Cc: john.fastabend, ast, netdev, Daniel Borkmann
We are guaranteed to have a NULL ri->map in this branch since
we test for it earlier, so we don't need to reset it here.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
net/core/filter.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/net/core/filter.c b/net/core/filter.c
index e9f8dce..ea3ca34 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -2568,7 +2568,6 @@ int xdp_do_redirect(struct net_device *dev, struct xdp_buff *xdp,
fwd = dev_get_by_index_rcu(dev_net(dev), index);
ri->ifindex = 0;
- ri->map = NULL;
if (unlikely(!fwd)) {
bpf_warn_invalid_xdp_redirect(index);
return -EINVAL;
--
1.9.3
^ permalink raw reply related
* LL,
From: adeola chambers @ 2017-08-17 13:07 UTC (permalink / raw)
Dear friend, I am Barrister Wilson, I want you to work with me as a
foreign partner also as family members or close relatives to my late
client who deposited USD$14.3Million dollars in a bank here my country
before his death, and a nationality of your country, The bank has
given me notice as his Lawyer who sign and witness everything to
present the next of kin of the fund or the money will be confiscated
if i fail, I need your immediate reply to enable me send more Detailed
information about this project to you.
Thanks
Danis Wilson.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox