date:20170919

[PATCH 1/6] staging: fsl-dpaa2/ethsw: Add APIs for DPSW object

2017-09-19 Thread Razvan Stefanescu

Add the command build/parse APIs for operating on DPSW objects through
the DPAA2 Management Complex.

Signed-off-by: Razvan Stefanescu 
---
 drivers/staging/fsl-dpaa2/Kconfig  |8 +
 drivers/staging/fsl-dpaa2/Makefile |1 +
 drivers/staging/fsl-dpaa2/ethsw/Makefile   |7 +
 drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h |  358 +
 drivers/staging/fsl-dpaa2/ethsw/dpsw.c | 1115 
 drivers/staging/fsl-dpaa2/ethsw/dpsw.h |  579 +++
 6 files changed, 2068 insertions(+)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/Makefile
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw.c
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw.h

diff --git a/drivers/staging/fsl-dpaa2/Kconfig 
b/drivers/staging/fsl-dpaa2/Kconfig
index dfff675..8a508ef 100644
--- a/drivers/staging/fsl-dpaa2/Kconfig
+++ b/drivers/staging/fsl-dpaa2/Kconfig
@@ -16,3 +16,11 @@ config FSL_DPAA2_ETH
---help---
  Ethernet driver for Freescale DPAA2 SoCs, using the
  Freescale MC bus driver
+
+config FSL_DPAA2_ETHSW
+   tristate "Freescale DPAA2 Ethernet Switch"
+   depends on FSL_DPAA2
+   depends on NET_SWITCHDEV
+   ---help---
+   Driver for Freescale DPAA2 Ethernet Switch. Select
+   BRIDGE to have support for bridge tools.
diff --git a/drivers/staging/fsl-dpaa2/Makefile 
b/drivers/staging/fsl-dpaa2/Makefile
index 0836ba8..6cfd76b 100644
--- a/drivers/staging/fsl-dpaa2/Makefile
+++ b/drivers/staging/fsl-dpaa2/Makefile
@@ -3,3 +3,4 @@
 #
 
 obj-$(CONFIG_FSL_DPAA2_ETH)+= ethernet/
+obj-$(CONFIG_FSL_DPAA2_ETHSW)  += ethsw/
diff --git a/drivers/staging/fsl-dpaa2/ethsw/Makefile 
b/drivers/staging/fsl-dpaa2/ethsw/Makefile
new file mode 100644
index 000..db137f7
--- /dev/null
+++ b/drivers/staging/fsl-dpaa2/ethsw/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for the Freescale DPAA2 Ethernet Switch
+#
+
+obj-$(CONFIG_FSL_DPAA2_ETHSW) += dpaa2-ethsw.o
+
+dpaa2-ethsw-objs := dpsw.o
diff --git a/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h 
b/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
new file mode 100644
index 000..ddfd820
--- /dev/null
+++ b/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
@@ -0,0 +1,358 @@
+/* Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ *  notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ *  notice, this list of conditions and the following disclaimer in the
+ *  documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ *  names of any contributors may be used to endorse or promote products
+ *  derived from this software without specific prior written permission.
+ *
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+#ifndef __FSL_DPSW_CMD_H
+#define __FSL_DPSW_CMD_H
+
+/* DPSW Version */
+#define DPSW_VER_MAJOR 8
+#define DPSW_VER_MINOR 0
+
+#define DPSW_CMD_BASE_VERSION  1
+#define DPSW_CMD_ID_OFFSET 4
+
+#define DPSW_CMD_ID(id)(((id) << DPSW_CMD_ID_OFFSET) | 
DPSW_CMD_BASE_VERSION)
+
+/* Command IDs */
+#define DPSW_CMDID_CLOSEDPSW_CMD_ID(0x800)
+#define DPSW_CMDID_OPEN DPSW_CMD_ID(0x802)
+
+#define DPSW_CMDID_GET_API_VERSION  DPSW_CMD_ID(0xa02)
+
+#define DPSW_CMDID_ENABLE   DPSW_CMD_ID(0x002)
+#define DPSW_CMDID_DISABLE  DPSW_CMD_ID(0x003)
+#define DPSW_CMDID_GET_ATTR DPSW_CMD_ID(0x004)
+#define DPSW_CMDID_RESETDPSW_CMD_ID(0x005)
+
+#define DPSW_CMDID_SET_IRQ_ENABLE

[PATCH 0/6] staging: Introduce DPAA2 Ethernet Switch driver

2017-09-19 Thread Razvan Stefanescu

This patchset introduces the Ethernet Switch Driver for Freescale/NXP SoCs
with DPAA2 (DataPath Acceleration Architecture v2). The driver manages
switch objects discovered on the fsl-mc bus. A description of the driver
can be found in the associated README file.

The patchset consists of:
* A set of libraries containing APIs for configuring and controlling
  Management Complex (MC) switch objects
* The DPAA2 Ethernet Switch driver
* Patch adding ethtool support

Limitations:
* no support for control traffic to/from CPU
* only DPSW ports can be added to a bridge

Razvan Stefanescu (6):
  staging: fsl-dpaa2/ethsw: Add APIs for DPSW object
  staging: fsl-dpaa2/ethsw: Add Freescale DPAA2 Ethernet Switch driver
  staging: fsl-dpaa2/ethsw: Add ethtool support
  staging: fsl-dpaa2/ethsw: Add maintainer for Ethernet Switch driver
  staging: fsl-dpaa2/ethsw: Add README
  staging: fsl-dpaa2/ethsw: Add TODO

 MAINTAINERS |6 +
 drivers/staging/fsl-dpaa2/Kconfig   |8 +
 drivers/staging/fsl-dpaa2/Makefile  |1 +
 drivers/staging/fsl-dpaa2/ethsw/Makefile|7 +
 drivers/staging/fsl-dpaa2/ethsw/README  |  106 ++
 drivers/staging/fsl-dpaa2/ethsw/TODO|   14 +
 drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h  |  371 ++
 drivers/staging/fsl-dpaa2/ethsw/dpsw.c  | 1147 +
 drivers/staging/fsl-dpaa2/ethsw/dpsw.h  |  611 +
 drivers/staging/fsl-dpaa2/ethsw/ethsw-ethtool.c |  207 +++
 drivers/staging/fsl-dpaa2/ethsw/ethsw.c | 1526 +++
 drivers/staging/fsl-dpaa2/ethsw/ethsw.h |   91 ++
 12 files changed, 4095 insertions(+)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/Makefile
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/README
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/TODO
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw.c
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw.h
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw-ethtool.c
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw.c
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw.h

-- 
1.9.1

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH 3/6] staging: fsl-dpaa2/ethsw: Add ethtool support

2017-09-19 Thread Razvan Stefanescu

Add driver information, link details and hardware statistics to be
reported via ethtool -S.

Signed-off-by: Razvan Stefanescu 
---
 drivers/staging/fsl-dpaa2/ethsw/Makefile|   2 +-
 drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h  |  13 ++
 drivers/staging/fsl-dpaa2/ethsw/dpsw.c  |  32 
 drivers/staging/fsl-dpaa2/ethsw/dpsw.h  |  32 
 drivers/staging/fsl-dpaa2/ethsw/ethsw-ethtool.c | 207 
 drivers/staging/fsl-dpaa2/ethsw/ethsw.c |   3 +
 drivers/staging/fsl-dpaa2/ethsw/ethsw.h |   3 +
 7 files changed, 291 insertions(+), 1 deletion(-)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw-ethtool.c

diff --git a/drivers/staging/fsl-dpaa2/ethsw/Makefile 
b/drivers/staging/fsl-dpaa2/ethsw/Makefile
index a6d72d1..de92cd9 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/Makefile
+++ b/drivers/staging/fsl-dpaa2/ethsw/Makefile
@@ -4,4 +4,4 @@
 
 obj-$(CONFIG_FSL_DPAA2_ETHSW) += dpaa2-ethsw.o
 
-dpaa2-ethsw-objs := ethsw.o dpsw.o
+dpaa2-ethsw-objs := ethsw.o ethsw-ethtool.o dpsw.o
diff --git a/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h 
b/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
index ddfd820..06b71122 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
+++ b/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
@@ -74,6 +74,8 @@
 #define DPSW_CMDID_IF_SET_FLOODING  DPSW_CMD_ID(0x047)
 #define DPSW_CMDID_IF_SET_BROADCAST DPSW_CMD_ID(0x048)
 
+#define DPSW_CMDID_IF_SET_LINK_CFG  DPSW_CMD_ID(0x04C)
+
 #define DPSW_CMDID_VLAN_ADD DPSW_CMD_ID(0x060)
 #define DPSW_CMDID_VLAN_ADD_IF  DPSW_CMD_ID(0x061)
 #define DPSW_CMDID_VLAN_ADD_IF_UNTAGGED DPSW_CMD_ID(0x062)
@@ -262,6 +264,17 @@ struct dpsw_cmd_if_set_max_frame_length {
__le16 frame_length;
 };
 
+struct dpsw_cmd_if_set_link_cfg {
+   /* cmd word 0 */
+   __le16 if_id;
+   u8 pad[6];
+   /* cmd word 1 */
+   __le32 rate;
+   __le32 pad1;
+   /* cmd word 2 */
+   __le64 options;
+};
+
 struct dpsw_cmd_if_get_link_state {
__le16 if_id;
 };
diff --git a/drivers/staging/fsl-dpaa2/ethsw/dpsw.c 
b/drivers/staging/fsl-dpaa2/ethsw/dpsw.c
index f36b92b..601172a 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/dpsw.c
+++ b/drivers/staging/fsl-dpaa2/ethsw/dpsw.c
@@ -383,6 +383,38 @@ int dpsw_get_attributes(struct fsl_mc_io *mc_io,
 }
 
 /**
+ * dpsw_if_set_link_cfg() - Set the link configuration.
+ * @mc_io: Pointer to MC portal's I/O object
+ * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token: Token of DPSW object
+ * @if_id: Interface id
+ * @cfg:   Link configuration
+ *
+ * Return: '0' on Success; Error code otherwise.
+ */
+int dpsw_if_set_link_cfg(struct fsl_mc_io *mc_io,
+u32 cmd_flags,
+u16 token,
+u16 if_id,
+struct dpsw_link_cfg *cfg)
+{
+   struct mc_command cmd = { 0 };
+   struct dpsw_cmd_if_set_link_cfg *cmd_params;
+
+   /* prepare command */
+   cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_SET_LINK_CFG,
+ cmd_flags,
+ token);
+   cmd_params = (struct dpsw_cmd_if_set_link_cfg *)cmd.params;
+   cmd_params->if_id = cpu_to_le16(if_id);
+   cmd_params->rate = cpu_to_le32(cfg->rate);
+   cmd_params->options = cpu_to_le64(cfg->options);
+
+   /* send command to mc*/
+   return mc_send_command(mc_io, &cmd);
+}
+
+/**
  * dpsw_if_get_link_state - Return the link state
  * @mc_io: Pointer to MC portal's I/O object
  * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_'
diff --git a/drivers/staging/fsl-dpaa2/ethsw/dpsw.h 
b/drivers/staging/fsl-dpaa2/ethsw/dpsw.h
index e9c2906..5feadc2 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/dpsw.h
+++ b/drivers/staging/fsl-dpaa2/ethsw/dpsw.h
@@ -245,6 +245,38 @@ enum dpsw_action {
 };
 
 /**
+ * Enable auto-negotiation
+ */
+#define DPSW_LINK_OPT_AUTONEG  0x0001ULL
+/**
+ * Enable half-duplex mode
+ */
+#define DPSW_LINK_OPT_HALF_DUPLEX  0x0002ULL
+/**
+ * Enable pause frames
+ */
+#define DPSW_LINK_OPT_PAUSE0x0004ULL
+/**
+ * Enable a-symmetric pause frames
+ */
+#define DPSW_LINK_OPT_ASYM_PAUSE   0x0008ULL
+
+/**
+ * struct dpsw_link_cfg - Structure representing DPSW link configuration
+ * @rate: Rate
+ * @options: Mask of available options; use 'DPSW_LINK_OPT_' values
+ */
+struct dpsw_link_cfg {
+   u32 rate;
+   u64 options;
+};
+
+int dpsw_if_set_link_cfg(struct fsl_mc_io *mc_io,
+u32 cmd_flags,
+u16 token,
+u16 if_id,
+struct dpsw_link_cfg *cfg);
+/**
  * struct dpsw_link_state - Structure representing DPSW link state
  * @rate: Rate
  * @options: Mask of available options; use 'DPSW_LINK_OPT_' values
diff --git a/drivers/stag

[PATCH 2/6] staging: fsl-dpaa2/ethsw: Add Freescale DPAA2 Ethernet Switch driver

2017-09-19 Thread Razvan Stefanescu

Introduce the DPAA2 Ethernet Switch driver, which manages Datapath Switch
(DPSW) objects discovered on the MC bus.

Suggested-by: Alexandru Marginean 
Signed-off-by: Razvan Stefanescu 
---
 drivers/staging/fsl-dpaa2/ethsw/Makefile |2 +-
 drivers/staging/fsl-dpaa2/ethsw/ethsw.c  | 1523 ++
 drivers/staging/fsl-dpaa2/ethsw/ethsw.h  |   88 ++
 3 files changed, 1612 insertions(+), 1 deletion(-)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw.c
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw.h

diff --git a/drivers/staging/fsl-dpaa2/ethsw/Makefile 
b/drivers/staging/fsl-dpaa2/ethsw/Makefile
index db137f7..a6d72d1 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/Makefile
+++ b/drivers/staging/fsl-dpaa2/ethsw/Makefile
@@ -4,4 +4,4 @@
 
 obj-$(CONFIG_FSL_DPAA2_ETHSW) += dpaa2-ethsw.o
 
-dpaa2-ethsw-objs := dpsw.o
+dpaa2-ethsw-objs := ethsw.o dpsw.o
diff --git a/drivers/staging/fsl-dpaa2/ethsw/ethsw.c 
b/drivers/staging/fsl-dpaa2/ethsw/ethsw.c
new file mode 100644
index 000..ae86078
--- /dev/null
+++ b/drivers/staging/fsl-dpaa2/ethsw/ethsw.c
@@ -0,0 +1,1523 @@
+/* Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ *  notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ *  notice, this list of conditions and the following disclaimer in the
+ *  documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ *  names of any contributors may be used to endorse or promote products
+ *  derived from this software without specific prior written permission.
+ *
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+
+#include 
+#include 
+#include 
+#include 
+
+#include "../../fsl-mc/include/mc.h"
+
+#include "ethsw.h"
+
+static struct workqueue_struct *ethsw_owq;
+
+/* Minimal supported DPSW version */
+#define DPSW_MIN_VER_MAJOR 8
+#define DPSW_MIN_VER_MINOR 0
+
+#define DEFAULT_VLAN_ID1
+
+static int ethsw_add_vlan(struct ethsw_core *ethsw, u16 vid)
+{
+   int err;
+
+   struct dpsw_vlan_cfgvcfg = {
+   .fdb_id = 0,
+   };
+
+   if (ethsw->vlans[vid]) {
+   dev_err(ethsw->dev, "VLAN already configured\n");
+   return -EEXIST;
+   }
+
+   err = dpsw_vlan_add(ethsw->mc_io, 0,
+   ethsw->dpsw_handle, vid, &vcfg);
+   if (err) {
+   dev_err(ethsw->dev, "dpsw_vlan_add err %d\n", err);
+   return err;
+   }
+   ethsw->vlans[vid] = ETHSW_VLAN_MEMBER;
+
+   return 0;
+}
+
+static int ethsw_port_add_vlan(struct ethsw_port_priv *port_priv,
+  u16 vid, u16 flags)
+{
+   struct ethsw_core *ethsw = port_priv->ethsw_data;
+   struct net_device *netdev = port_priv->netdev;
+   struct dpsw_vlan_if_cfg vcfg;
+   bool is_oper;
+   int err, err2;
+
+   if (port_priv->vlans[vid]) {
+   netdev_warn(netdev, "VLAN %d already configured\n", vid);
+   return -EEXIST;
+   }
+
+   vcfg.num_ifs = 1;
+   vcfg.if_id[0] = port_priv->idx;
+   err = dpsw_vlan_add_if(ethsw->mc_io, 0, ethsw->dpsw_handle, vid, &vcfg);
+   if (err) {
+   netdev_err(netdev, "dpsw_vlan_add_if err %d\n", err);
+   return err;
+   }
+
+   port_priv->vlans[vid] = ETHSW_VLAN_MEMBER;
+
+   if (flags & BRIDGE_VLAN_INFO_UNTAGGED) {
+   err = dpsw_vlan_add_if_untagged(ethsw->mc_io, 0,
+   ethsw->dpsw_handle,
+

[PATCH 4/6] staging: fsl-dpaa2/ethsw: Add maintainer for Ethernet Switch driver

2017-09-19 Thread Razvan Stefanescu

Signed-off-by: Razvan Stefanescu 
---
 MAINTAINERS | 6 ++
 1 file changed, 6 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 2281af4..cfd4f74 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4297,6 +4297,12 @@ L:   linux-ker...@vger.kernel.org
 S: Maintained
 F: drivers/staging/fsl-dpaa2/ethernet
 
+DPAA2 ETHERNET SWITCH DRIVER
+M: Razvan Stefanescu 
+L: linux-ker...@vger.kernel.org
+S: Maintained
+F: drivers/staging/fsl-dpaa2/ethsw
+
 DPT_I2O SCSI RAID DRIVER
 M: Adaptec OEM Raid Solutions 
 L: linux-s...@vger.kernel.org
-- 
1.9.1

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH 5/6] staging: fsl-dpaa2/ethsw: Add README

2017-09-19 Thread Razvan Stefanescu

Add a README file describing the driver architecture, components and
interfaces.

Signed-off-by: Razvan Stefanescu 
---
 drivers/staging/fsl-dpaa2/ethsw/README | 106 +
 1 file changed, 106 insertions(+)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/README

diff --git a/drivers/staging/fsl-dpaa2/ethsw/README 
b/drivers/staging/fsl-dpaa2/ethsw/README
new file mode 100644
index 000..f6fc07f
--- /dev/null
+++ b/drivers/staging/fsl-dpaa2/ethsw/README
@@ -0,0 +1,106 @@
+DPAA2 Ethernet Switch driver
+
+
+This file provides documentation for the DPAA2 Ethernet Switch driver
+
+
+Contents
+
+   Supported Platforms
+   Architecture Overview
+   Creating an Ethernet Switch
+   Features
+
+
+   Supported Platforms
+===
+This driver provides networking support for Freescale LS2085A, LS2088A
+DPAA2 SoCs.
+
+
+Architecture Overview
+=
+The Ethernet Switch in the DPAA2 architecture consists of several hardware
+resources that provide the functionality. These are allocated and
+configured via the Management Complex (MC) portals. MC abstracts most of
+these resources as DPAA2 objects and exposes ABIs through which they can
+be configured and controlled.
+
+For a more detailed description of the DPAA2 architecture and its object
+abstractions see:
+   drivers/staging/fsl-mc/README.txt
+
+The Ethernet Switch is built on top of a Datapath Switch (DPSW) object.
+
+Configuration interface:
+
+  -
+ | DPAA2 Switch driver |
+  -
+   .
+   .
+  --
+ | DPSW API |
+  --
+   .   software
+ = . ==
+   .   hardware
+  -
+ | MC hardware portals |
+  -
+   .
+   .
+ --
+| DPSW |
+ --
+
+Driver uses the switch device driver model and exposes each switch port as
+a network interface, which can be included in a bridge. Traffic switched
+between ports is offloaded into the hardware. Exposed network interfaces
+are not used for I/O, they are used just for configuration. This
+limitation is going to be addressed in the future.
+
+The DPSW can have ports connected to DPNIs or to PHYs via DPMACs.
+
+
+ [ethA] [ethB] [ethC] [ethD] [ethE] [ethF]
+:  :  :  :  :  :
+:  :  :  :  :  :
+[eth drv]  [eth drv]  [ethsw drv  ]
+:  :  :  :  :  :kernel
+
+:  :  :  :  :  :hardware
+ [DPNI]  [DPNI] [= DPSW =]
+|  |  |  |  |  |
+|   --   |   [DPMAC][DPMAC]
+ ---|  |
+|  |
+  [PHY]  [PHY]
+
+For a more detailed description of the Ethernet switch device driver model
+see:
+   Documentation/networking/switchdev.txt
+
+Creating an Ethernet Switch
+===
+A device is created for the switch objects probed on the MC bus. Each DPSW
+has a number of properties which determine the configuration options and
+associated hardware resources.
+
+A DPSW object (and the other DPAA2 objects needed for a DPAA2 switch) can
+be added to a container on the MC bus in one of two ways: statically,
+through a Datapath Layout Binary file (DPL) that is parsed by MC at boot
+time; or created dynamically at runtime, via the DPAA2 objects APIs.
+
+Features
+
+Driver configures DPSW to perform hardware switching offload of
+unicast/multicast/broadcast (VLAN tagged or untagged) traffic between its
+ports.
+
+It allows configuration of hardware learning, flooding, multicast groups,
+port VLAN configuration and STP state.
+
+Static entries can be added/removed from the FDB.
+
+Hardware statistics for each port are provided through ethtool -S option.
-- 
1.9.1

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH 6/6] staging: fsl-dpaa2/ethsw: Add TODO

2017-09-19 Thread Razvan Stefanescu

Add a TODO file describing what needs to be added/changed before the driver
can be moved out of staging.

Signed-off-by: Razvan Stefanescu 
---
 drivers/staging/fsl-dpaa2/ethsw/TODO | 14 ++
 1 file changed, 14 insertions(+)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/TODO

diff --git a/drivers/staging/fsl-dpaa2/ethsw/TODO 
b/drivers/staging/fsl-dpaa2/ethsw/TODO
new file mode 100644
index 000..d3f12c3
--- /dev/null
+++ b/drivers/staging/fsl-dpaa2/ethsw/TODO
@@ -0,0 +1,14 @@
+* Add I/O capabilities on switch port netdevices. This will allow control
+traffic to reach the CPU.
+* Add ACL to redirect control traffic to CPU.
+* Add support for displaying learned FDB entries
+* MC firmware uprev; the DPAA2 objects used by the Ethernet Switch driver
+need to   be kept in sync with binary interface changes in MC
+* refine README file
+* cleanup
+
+NOTE: At least first three of the above are required before getting the
+DPAA2 Ethernet Switch driver out of staging. Another requirement is that
+the fsl-mc bus driver is moved to drivers/bus and dpio driver is moved to
+drivers/soc (this is required for I/O).
+
-- 
1.9.1

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] staging: ion: create one device entry per heap

2017-09-19 Thread Dan Carpenter

On Mon, Sep 18, 2017 at 04:58:46PM +0200, Benjamin Gaignard wrote:
> -static int validate_ioctl_arg(unsigned int cmd, union ion_ioctl_arg *arg)
> +static int validate_ioctl_arg(struct file *filp,
> +   unsigned int cmd, union ion_ioctl_arg *arg)
>  {
>   int ret = 0;
> + int mask = 1 << iminor(filp->f_inode);
>  
>   switch (cmd) {
>   case ION_IOC_HEAP_QUERY:
> @@ -35,6 +37,9 @@ static int validate_ioctl_arg(unsigned int cmd, union 
> ion_ioctl_arg *arg)
>   ret |= arg->query.reserved1 != 0;
>   ret |= arg->query.reserved2 != 0;
>   break;
> + case ION_IOC_ALLOC:
> + ret = !(arg->allocation.heap_id_mask & mask);


validate_ioctl_arg() is really convoluted.  From reading just the patch
I at first thought we were returning 1 on failure.  Just say:

if (!(arg->allocation.heap_id_mask & mask))
return -EINVAL;
return 0;

If you want to fix the surrounding code in a separate patch that would
be good.  It would be more clear to say:

if (arg->query.reserved0 != 0 ||
arg->query.reserved1 != 0 ||
arg->query.reserved2 != 0)
return -EINVAL;

> + break;
>   default:
>   break;
>   }
> @@ -70,7 +75,7 @@ long ion_ioctl(struct file *filp, unsigned int cmd, 
> unsigned long arg)
>   if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd)))
>   return -EFAULT;
>  
> - ret = validate_ioctl_arg(cmd, &data);
> + ret = validate_ioctl_arg(filp, cmd, &data);
>   if (WARN_ON_ONCE(ret))
>   return ret;
>  
> diff --git a/drivers/staging/android/ion/ion.c 
> b/drivers/staging/android/ion/ion.c
> index 93e2c90..5144f1a 100644
> --- a/drivers/staging/android/ion/ion.c
> +++ b/drivers/staging/android/ion/ion.c
> @@ -40,6 +40,8 @@
>  
>  #include "ion.h"
>  
> +#define ION_DEV_MAX 32
> +
>  static struct ion_device *internal_dev;
>  static int heap_id;
>  
> @@ -541,11 +543,21 @@ void ion_device_add_heap(struct ion_heap *heap)
>  {
>   struct dentry *debug_file;
>   struct ion_device *dev = internal_dev;
> + int ret;
>  
>   if (!heap->ops->allocate || !heap->ops->free)
>   pr_err("%s: can not add heap with invalid ops struct.\n",
>  __func__);
>  

I don't think it can happen in current code but we should proabably have
a check here for:

if (heap_id >= ION_DEV_MAX)
return -EBUSY;

(It's possible I have missed something).


> + heap->ddev.devt = MKDEV(MAJOR(dev->devt), heap_id);
> + dev_set_name(&heap->ddev, "ion%d", heap_id);
> + device_initialize(&heap->ddev);
> + cdev_init(&heap->chrdev, &ion_fops);
> + heap->chrdev.owner = THIS_MODULE;
> + ret = cdev_device_add(&heap->chrdev, &heap->ddev);
> + if (ret < 0)
> + return;
> +
>   spin_lock_init(&heap->free_lock);
>   heap->free_list_size = 0;
>  

regards,
dan carpenter


___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] staging: ion: create one device entry per heap

2017-09-19 Thread Benjamin Gaignard

2017-09-19 11:40 GMT+02:00 Dan Carpenter :
> On Mon, Sep 18, 2017 at 04:58:46PM +0200, Benjamin Gaignard wrote:
>> -static int validate_ioctl_arg(unsigned int cmd, union ion_ioctl_arg *arg)
>> +static int validate_ioctl_arg(struct file *filp,
>> +   unsigned int cmd, union ion_ioctl_arg *arg)
>>  {
>>   int ret = 0;
>> + int mask = 1 << iminor(filp->f_inode);
>>
>>   switch (cmd) {
>>   case ION_IOC_HEAP_QUERY:
>> @@ -35,6 +37,9 @@ static int validate_ioctl_arg(unsigned int cmd, union 
>> ion_ioctl_arg *arg)
>>   ret |= arg->query.reserved1 != 0;
>>   ret |= arg->query.reserved2 != 0;
>>   break;
>> + case ION_IOC_ALLOC:
>> + ret = !(arg->allocation.heap_id_mask & mask);
>
>
> validate_ioctl_arg() is really convoluted.  From reading just the patch
> I at first thought we were returning 1 on failure.  Just say:
>
> if (!(arg->allocation.heap_id_mask & mask))
> return -EINVAL;
> return 0;
>
> If you want to fix the surrounding code in a separate patch that would
> be good.  It would be more clear to say:
>
> if (arg->query.reserved0 != 0 ||
> arg->query.reserved1 != 0 ||
> arg->query.reserved2 != 0)
> return -EINVAL;

I agree I will add a fix for that in next version

>
>> + break;
>>   default:
>>   break;
>>   }
>> @@ -70,7 +75,7 @@ long ion_ioctl(struct file *filp, unsigned int cmd, 
>> unsigned long arg)
>>   if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd)))
>>   return -EFAULT;
>>
>> - ret = validate_ioctl_arg(cmd, &data);
>> + ret = validate_ioctl_arg(filp, cmd, &data);
>>   if (WARN_ON_ONCE(ret))
>>   return ret;
>>
>> diff --git a/drivers/staging/android/ion/ion.c 
>> b/drivers/staging/android/ion/ion.c
>> index 93e2c90..5144f1a 100644
>> --- a/drivers/staging/android/ion/ion.c
>> +++ b/drivers/staging/android/ion/ion.c
>> @@ -40,6 +40,8 @@
>>
>>  #include "ion.h"
>>
>> +#define ION_DEV_MAX 32
>> +
>>  static struct ion_device *internal_dev;
>>  static int heap_id;
>>
>> @@ -541,11 +543,21 @@ void ion_device_add_heap(struct ion_heap *heap)
>>  {
>>   struct dentry *debug_file;
>>   struct ion_device *dev = internal_dev;
>> + int ret;
>>
>>   if (!heap->ops->allocate || !heap->ops->free)
>>   pr_err("%s: can not add heap with invalid ops struct.\n",
>>  __func__);
>>
>
> I don't think it can happen in current code but we should proabably have
> a check here for:
>
> if (heap_id >= ION_DEV_MAX)
> return -EBUSY;
>
> (It's possible I have missed something).
>

You are right I will add that

Thanks
>
>> + heap->ddev.devt = MKDEV(MAJOR(dev->devt), heap_id);
>> + dev_set_name(&heap->ddev, "ion%d", heap_id);
>> + device_initialize(&heap->ddev);
>> + cdev_init(&heap->chrdev, &ion_fops);
>> + heap->chrdev.owner = THIS_MODULE;
>> + ret = cdev_device_add(&heap->chrdev, &heap->ddev);
>> + if (ret < 0)
>> + return;
>> +
>>   spin_lock_init(&heap->free_lock);
>>   heap->free_list_size = 0;
>>
>
> regards,
> dan carpenter
>
>



-- 
Benjamin Gaignard

Graphic Study Group

Linaro.org │ Open source software for ARM SoCs

Follow Linaro: Facebook | Twitter | Blog
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] staging: ion: create one device entry per heap

2017-09-19 Thread Tomas Winkler

On Tue, Sep 19, 2017 at 1:07 PM, Benjamin Gaignard
 wrote:
> 2017-09-19 11:40 GMT+02:00 Dan Carpenter :
>> On Mon, Sep 18, 2017 at 04:58:46PM +0200, Benjamin Gaignard wrote:
>>> -static int validate_ioctl_arg(unsigned int cmd, union ion_ioctl_arg *arg)
>>> +static int validate_ioctl_arg(struct file *filp,
>>> +   unsigned int cmd, union ion_ioctl_arg *arg)
>>>  {
>>>   int ret = 0;
>>> + int mask = 1 << iminor(filp->f_inode);
>>>
>>>   switch (cmd) {
>>>   case ION_IOC_HEAP_QUERY:
>>> @@ -35,6 +37,9 @@ static int validate_ioctl_arg(unsigned int cmd, union 
>>> ion_ioctl_arg *arg)
>>>   ret |= arg->query.reserved1 != 0;
>>>   ret |= arg->query.reserved2 != 0;
>>>   break;
>>> + case ION_IOC_ALLOC:
>>> + ret = !(arg->allocation.heap_id_mask & mask);
>>
>>
>> validate_ioctl_arg() is really convoluted.  From reading just the patch
>> I at first thought we were returning 1 on failure.  Just say:
>>
>> if (!(arg->allocation.heap_id_mask & mask))
>> return -EINVAL;
>> return 0;
>>
>> If you want to fix the surrounding code in a separate patch that would
>> be good.  It would be more clear to say:
>>
>> if (arg->query.reserved0 != 0 ||
>> arg->query.reserved1 != 0 ||
>> arg->query.reserved2 != 0)
>> return -EINVAL;
>
> I agree I will add a fix for that in next version
>
>>
>>> + break;
>>>   default:
>>>   break;
>>>   }
>>> @@ -70,7 +75,7 @@ long ion_ioctl(struct file *filp, unsigned int cmd, 
>>> unsigned long arg)
>>>   if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd)))
>>>   return -EFAULT;
>>>
>>> - ret = validate_ioctl_arg(cmd, &data);
>>> + ret = validate_ioctl_arg(filp, cmd, &data);
>>>   if (WARN_ON_ONCE(ret))
>>>   return ret;
>>>
>>> diff --git a/drivers/staging/android/ion/ion.c 
>>> b/drivers/staging/android/ion/ion.c
>>> index 93e2c90..5144f1a 100644
>>> --- a/drivers/staging/android/ion/ion.c
>>> +++ b/drivers/staging/android/ion/ion.c
>>> @@ -40,6 +40,8 @@
>>>
>>>  #include "ion.h"
>>>
>>> +#define ION_DEV_MAX 32
>>> +
>>>  static struct ion_device *internal_dev;
>>>  static int heap_id;
>>>
>>> @@ -541,11 +543,21 @@ void ion_device_add_heap(struct ion_heap *heap)
>>>  {
>>>   struct dentry *debug_file;
>>>   struct ion_device *dev = internal_dev;
>>> + int ret;
>>>
>>>   if (!heap->ops->allocate || !heap->ops->free)
>>>   pr_err("%s: can not add heap with invalid ops struct.\n",
>>>  __func__);
>>>
>>
>> I don't think it can happen in current code but we should proabably have
>> a check here for:
>>
>> if (heap_id >= ION_DEV_MAX)
>> return -EBUSY;
>>
>> (It's possible I have missed something).
>>
>
> You are right I will add that
>
> Thanks
>>
>>> + heap->ddev.devt = MKDEV(MAJOR(dev->devt), heap_id);
>>> + dev_set_name(&heap->ddev, "ion%d", heap_id);
>>> + device_initialize(&heap->ddev);
>>> + cdev_init(&heap->chrdev, &ion_fops);
>>> + heap->chrdev.owner = THIS_MODULE;
>>> + ret = cdev_device_add(&heap->chrdev, &heap->ddev);
>>> + if (ret < 0)
>>> + return;
>>> +
>>>   spin_lock_init(&heap->free_lock);
>>>   heap->free_list_size = 0;

What will happen to an application which looks for /dev/ion?

Thanks
Tomas
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] staging: ion: create one device entry per heap

2017-09-19 Thread Benjamin Gaignard

2017-09-19 12:15 GMT+02:00 Tomas Winkler :
> On Tue, Sep 19, 2017 at 1:07 PM, Benjamin Gaignard
>  wrote:
>> 2017-09-19 11:40 GMT+02:00 Dan Carpenter :
>>> On Mon, Sep 18, 2017 at 04:58:46PM +0200, Benjamin Gaignard wrote:
 -static int validate_ioctl_arg(unsigned int cmd, union ion_ioctl_arg *arg)
 +static int validate_ioctl_arg(struct file *filp,
 +   unsigned int cmd, union ion_ioctl_arg *arg)
  {
   int ret = 0;
 + int mask = 1 << iminor(filp->f_inode);

   switch (cmd) {
   case ION_IOC_HEAP_QUERY:
 @@ -35,6 +37,9 @@ static int validate_ioctl_arg(unsigned int cmd, union 
 ion_ioctl_arg *arg)
   ret |= arg->query.reserved1 != 0;
   ret |= arg->query.reserved2 != 0;
   break;
 + case ION_IOC_ALLOC:
 + ret = !(arg->allocation.heap_id_mask & mask);
>>>
>>>
>>> validate_ioctl_arg() is really convoluted.  From reading just the patch
>>> I at first thought we were returning 1 on failure.  Just say:
>>>
>>> if (!(arg->allocation.heap_id_mask & mask))
>>> return -EINVAL;
>>> return 0;
>>>
>>> If you want to fix the surrounding code in a separate patch that would
>>> be good.  It would be more clear to say:
>>>
>>> if (arg->query.reserved0 != 0 ||
>>> arg->query.reserved1 != 0 ||
>>> arg->query.reserved2 != 0)
>>> return -EINVAL;
>>
>> I agree I will add a fix for that in next version
>>
>>>
 + break;
   default:
   break;
   }
 @@ -70,7 +75,7 @@ long ion_ioctl(struct file *filp, unsigned int cmd, 
 unsigned long arg)
   if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd)))
   return -EFAULT;

 - ret = validate_ioctl_arg(cmd, &data);
 + ret = validate_ioctl_arg(filp, cmd, &data);
   if (WARN_ON_ONCE(ret))
   return ret;

 diff --git a/drivers/staging/android/ion/ion.c 
 b/drivers/staging/android/ion/ion.c
 index 93e2c90..5144f1a 100644
 --- a/drivers/staging/android/ion/ion.c
 +++ b/drivers/staging/android/ion/ion.c
 @@ -40,6 +40,8 @@

  #include "ion.h"

 +#define ION_DEV_MAX 32
 +
  static struct ion_device *internal_dev;
  static int heap_id;

 @@ -541,11 +543,21 @@ void ion_device_add_heap(struct ion_heap *heap)
  {
   struct dentry *debug_file;
   struct ion_device *dev = internal_dev;
 + int ret;

   if (!heap->ops->allocate || !heap->ops->free)
   pr_err("%s: can not add heap with invalid ops struct.\n",
  __func__);

>>>
>>> I don't think it can happen in current code but we should proabably have
>>> a check here for:
>>>
>>> if (heap_id >= ION_DEV_MAX)
>>> return -EBUSY;
>>>
>>> (It's possible I have missed something).
>>>
>>
>> You are right I will add that
>>
>> Thanks
>>>
 + heap->ddev.devt = MKDEV(MAJOR(dev->devt), heap_id);
 + dev_set_name(&heap->ddev, "ion%d", heap_id);
 + device_initialize(&heap->ddev);
 + cdev_init(&heap->chrdev, &ion_fops);
 + heap->chrdev.owner = THIS_MODULE;
 + ret = cdev_device_add(&heap->chrdev, &heap->ddev);
 + if (ret < 0)
 + return;
 +
   spin_lock_init(&heap->free_lock);
   heap->free_list_size = 0;
>
> What will happen to an application which looks for /dev/ion?

/dev/ion will no more exist with this patch.
Since ion ABI have already change a lot I don't think that could
be a problem to change also ion device.

>
> Thanks
> Tomas
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH v2 0/2] staging: ion: get one device per heap

2017-09-19 Thread Benjamin Gaignard

version 2:
- simplify ioctl check like propose by Dan
- make sure that we don't register more than ION_DEV_MAX heaps

Instead a getting one common device "/dev/ion" for
all the heaps this patch allow to create one device
entry ("/dev/ionX") per heap.
Getting an entry per heap could allow to set security rules
per heap and global ones for all heaps.

Allocation requests will be only allowed if the mask_id
match with device minor.
Query request could be done on any of the devices.

Benjamin Gaignard (2):
  staging: ion: simplify ioctl args checking function
  staging: ion: create one device entry per heap

 drivers/staging/android/ion/ion-ioctl.c | 20 +---
 drivers/staging/android/ion/ion.c   | 23 +--
 drivers/staging/android/ion/ion.h   | 10 +++---
 3 files changed, 37 insertions(+), 16 deletions(-)

-- 
2.7.4

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH v2 2/2] staging: ion: create one device entry per heap

2017-09-19 Thread Benjamin Gaignard

Instead a getting one common device "/dev/ion" for
all the heaps this patch allow to create one device
entry ("/dev/ionX") per heap.
Getting an entry per heap could allow to set security rules
per heap and global ones for all heaps.

Allocation requests will be only allowed if the mask_id
match with device minor.
Query request could be done on any of the devices.

Signed-off-by: Benjamin Gaignard 
---
 drivers/staging/android/ion/ion-ioctl.c | 11 +--
 drivers/staging/android/ion/ion.c   | 23 +--
 drivers/staging/android/ion/ion.h   | 10 +++---
 3 files changed, 33 insertions(+), 11 deletions(-)

diff --git a/drivers/staging/android/ion/ion-ioctl.c 
b/drivers/staging/android/ion/ion-ioctl.c
index e26b786..c8c906c 100644
--- a/drivers/staging/android/ion/ion-ioctl.c
+++ b/drivers/staging/android/ion/ion-ioctl.c
@@ -25,8 +25,11 @@ union ion_ioctl_arg {
struct ion_heap_query query;
 };
 
-static int validate_ioctl_arg(unsigned int cmd, union ion_ioctl_arg *arg)
+static int validate_ioctl_arg(struct file *filp,
+ unsigned int cmd, union ion_ioctl_arg *arg)
 {
+   int mask = 1 << iminor(filp->f_inode);
+
switch (cmd) {
case ION_IOC_HEAP_QUERY:
if (arg->query.reserved0 ||
@@ -34,6 +37,10 @@ static int validate_ioctl_arg(unsigned int cmd, union 
ion_ioctl_arg *arg)
arg->query.reserved2 )
return -EINVAL;
break;
+   case ION_IOC_ALLOC:
+   if (!(arg->allocation.heap_id_mask & mask))
+   return -EINVAL;
+   break;
default:
break;
}
@@ -69,7 +76,7 @@ long ion_ioctl(struct file *filp, unsigned int cmd, unsigned 
long arg)
if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd)))
return -EFAULT;
 
-   ret = validate_ioctl_arg(cmd, &data);
+   ret = validate_ioctl_arg(filp, cmd, &data);
if (WARN_ON_ONCE(ret))
return ret;
 
diff --git a/drivers/staging/android/ion/ion.c 
b/drivers/staging/android/ion/ion.c
index 93e2c90..3f8b595 100644
--- a/drivers/staging/android/ion/ion.c
+++ b/drivers/staging/android/ion/ion.c
@@ -40,6 +40,8 @@
 
 #include "ion.h"
 
+#define ION_DEV_MAX 32
+
 static struct ion_device *internal_dev;
 static int heap_id;
 
@@ -541,11 +543,24 @@ void ion_device_add_heap(struct ion_heap *heap)
 {
struct dentry *debug_file;
struct ion_device *dev = internal_dev;
+   int ret;
 
if (!heap->ops->allocate || !heap->ops->free)
pr_err("%s: can not add heap with invalid ops struct.\n",
   __func__);
 
+   if (heap_id >= ION_DEV_MAX)
+   return -EBUSY;
+
+   heap->ddev.devt = MKDEV(MAJOR(dev->devt), heap_id);
+   dev_set_name(&heap->ddev, "ion%d", heap_id);
+   device_initialize(&heap->ddev);
+   cdev_init(&heap->chrdev, &ion_fops);
+   heap->chrdev.owner = THIS_MODULE;
+   ret = cdev_device_add(&heap->chrdev, &heap->ddev);
+   if (ret < 0)
+   return;
+
spin_lock_init(&heap->free_lock);
heap->free_list_size = 0;
 
@@ -595,13 +610,9 @@ static int ion_device_create(void)
if (!idev)
return -ENOMEM;
 
-   idev->dev.minor = MISC_DYNAMIC_MINOR;
-   idev->dev.name = "ion";
-   idev->dev.fops = &ion_fops;
-   idev->dev.parent = NULL;
-   ret = misc_register(&idev->dev);
+   ret = alloc_chrdev_region(&idev->devt, 0, ION_DEV_MAX, "ion");
if (ret) {
-   pr_err("ion: failed to register misc device.\n");
+   pr_err("ion: unable to allocate major\n");
kfree(idev);
return ret;
}
diff --git a/drivers/staging/android/ion/ion.h 
b/drivers/staging/android/ion/ion.h
index 621e5f7..ef51ff5 100644
--- a/drivers/staging/android/ion/ion.h
+++ b/drivers/staging/android/ion/ion.h
@@ -17,6 +17,7 @@
 #ifndef _ION_H
 #define _ION_H
 
+#include 
 #include 
 #include 
 #include 
@@ -26,7 +27,6 @@
 #include 
 #include 
 #include 
-#include 
 
 #include "../uapi/ion.h"
 
@@ -90,13 +90,13 @@ void ion_buffer_destroy(struct ion_buffer *buffer);
 
 /**
  * struct ion_device - the metadata of the ion device node
- * @dev:   the actual misc device
+ * @dev:   the actual device
  * @buffers:   an rb tree of all the existing buffers
  * @buffer_lock:   lock protecting the tree of buffers
  * @lock:  rwsem protecting the tree of heaps and clients
  */
 struct ion_device {
-   struct miscdevice dev;
+   dev_t devt;
struct rb_root buffers;
struct mutex buffer_lock;
struct rw_semaphore lock;
@@ -152,6 +152,8 @@ struct ion_heap_ops {
  * struct ion_heap - represents a heap in the system
  * @node:  rb node to put the heap on the device's tree of heaps
  * @dev:   back pointer to the ion_device
+ *

[PATCH v2 1/2] staging: ion: simplify ioctl args checking function

2017-09-19 Thread Benjamin Gaignard

Make arguments checking more easy to read.

Signed-off-by: Benjamin Gaignard 
---
 drivers/staging/android/ion/ion-ioctl.c | 11 +--
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/staging/android/ion/ion-ioctl.c 
b/drivers/staging/android/ion/ion-ioctl.c
index d9f8b14..e26b786 100644
--- a/drivers/staging/android/ion/ion-ioctl.c
+++ b/drivers/staging/android/ion/ion-ioctl.c
@@ -27,19 +27,18 @@ union ion_ioctl_arg {
 
 static int validate_ioctl_arg(unsigned int cmd, union ion_ioctl_arg *arg)
 {
-   int ret = 0;
-
switch (cmd) {
case ION_IOC_HEAP_QUERY:
-   ret = arg->query.reserved0 != 0;
-   ret |= arg->query.reserved1 != 0;
-   ret |= arg->query.reserved2 != 0;
+   if (arg->query.reserved0 ||
+   arg->query.reserved1 ||
+   arg->query.reserved2 )
+   return -EINVAL;
break;
default:
break;
}
 
-   return ret ? -EINVAL : 0;
+   return 0;
 }
 
 /* fix up the cases where the ioctl direction bits are incorrect */
-- 
2.7.4

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] staging: ion: create one device entry per heap

2017-09-19 Thread Greg Kroah-Hartman

On Tue, Sep 19, 2017 at 12:20:15PM +0200, Benjamin Gaignard wrote:
> 2017-09-19 12:15 GMT+02:00 Tomas Winkler :
> > On Tue, Sep 19, 2017 at 1:07 PM, Benjamin Gaignard
> >  wrote:
> >> 2017-09-19 11:40 GMT+02:00 Dan Carpenter :
> >>> On Mon, Sep 18, 2017 at 04:58:46PM +0200, Benjamin Gaignard wrote:
>  -static int validate_ioctl_arg(unsigned int cmd, union ion_ioctl_arg 
>  *arg)
>  +static int validate_ioctl_arg(struct file *filp,
>  +   unsigned int cmd, union ion_ioctl_arg *arg)
>   {
>    int ret = 0;
>  + int mask = 1 << iminor(filp->f_inode);
> 
>    switch (cmd) {
>    case ION_IOC_HEAP_QUERY:
>  @@ -35,6 +37,9 @@ static int validate_ioctl_arg(unsigned int cmd, union 
>  ion_ioctl_arg *arg)
>    ret |= arg->query.reserved1 != 0;
>    ret |= arg->query.reserved2 != 0;
>    break;
>  + case ION_IOC_ALLOC:
>  + ret = !(arg->allocation.heap_id_mask & mask);
> >>>
> >>>
> >>> validate_ioctl_arg() is really convoluted.  From reading just the patch
> >>> I at first thought we were returning 1 on failure.  Just say:
> >>>
> >>> if (!(arg->allocation.heap_id_mask & mask))
> >>> return -EINVAL;
> >>> return 0;
> >>>
> >>> If you want to fix the surrounding code in a separate patch that would
> >>> be good.  It would be more clear to say:
> >>>
> >>> if (arg->query.reserved0 != 0 ||
> >>> arg->query.reserved1 != 0 ||
> >>> arg->query.reserved2 != 0)
> >>> return -EINVAL;
> >>
> >> I agree I will add a fix for that in next version
> >>
> >>>
>  + break;
>    default:
>    break;
>    }
>  @@ -70,7 +75,7 @@ long ion_ioctl(struct file *filp, unsigned int cmd, 
>  unsigned long arg)
>    if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd)))
>    return -EFAULT;
> 
>  - ret = validate_ioctl_arg(cmd, &data);
>  + ret = validate_ioctl_arg(filp, cmd, &data);
>    if (WARN_ON_ONCE(ret))
>    return ret;
> 
>  diff --git a/drivers/staging/android/ion/ion.c 
>  b/drivers/staging/android/ion/ion.c
>  index 93e2c90..5144f1a 100644
>  --- a/drivers/staging/android/ion/ion.c
>  +++ b/drivers/staging/android/ion/ion.c
>  @@ -40,6 +40,8 @@
> 
>   #include "ion.h"
> 
>  +#define ION_DEV_MAX 32
>  +
>   static struct ion_device *internal_dev;
>   static int heap_id;
> 
>  @@ -541,11 +543,21 @@ void ion_device_add_heap(struct ion_heap *heap)
>   {
>    struct dentry *debug_file;
>    struct ion_device *dev = internal_dev;
>  + int ret;
> 
>    if (!heap->ops->allocate || !heap->ops->free)
>    pr_err("%s: can not add heap with invalid ops struct.\n",
>   __func__);
> 
> >>>
> >>> I don't think it can happen in current code but we should proabably have
> >>> a check here for:
> >>>
> >>> if (heap_id >= ION_DEV_MAX)
> >>> return -EBUSY;
> >>>
> >>> (It's possible I have missed something).
> >>>
> >>
> >> You are right I will add that
> >>
> >> Thanks
> >>>
>  + heap->ddev.devt = MKDEV(MAJOR(dev->devt), heap_id);
>  + dev_set_name(&heap->ddev, "ion%d", heap_id);
>  + device_initialize(&heap->ddev);
>  + cdev_init(&heap->chrdev, &ion_fops);
>  + heap->chrdev.owner = THIS_MODULE;
>  + ret = cdev_device_add(&heap->chrdev, &heap->ddev);
>  + if (ret < 0)
>  + return;
>  +
>    spin_lock_init(&heap->free_lock);
>    heap->free_list_size = 0;
> >
> > What will happen to an application which looks for /dev/ion?
> 
> /dev/ion will no more exist with this patch.
> Since ion ABI have already change a lot I don't think that could
> be a problem to change also ion device.

So, what did you just break in userspace?  :(

Do you also have userspace patches submitted anywhere to handle this
change?

thanks,

greg k-h
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH v2 2/2] staging: ion: create one device entry per heap

2017-09-19 Thread Greg KH

On Tue, Sep 19, 2017 at 12:25:38PM +0200, Benjamin Gaignard wrote:
> Instead a getting one common device "/dev/ion" for
> all the heaps this patch allow to create one device
> entry ("/dev/ionX") per heap.
> Getting an entry per heap could allow to set security rules
> per heap and global ones for all heaps.
> 
> Allocation requests will be only allowed if the mask_id
> match with device minor.
> Query request could be done on any of the devices.
> 
> Signed-off-by: Benjamin Gaignard 
> ---
>  drivers/staging/android/ion/ion-ioctl.c | 11 +--
>  drivers/staging/android/ion/ion.c   | 23 +--
>  drivers/staging/android/ion/ion.h   | 10 +++---
>  3 files changed, 33 insertions(+), 11 deletions(-)
> 
> diff --git a/drivers/staging/android/ion/ion-ioctl.c 
> b/drivers/staging/android/ion/ion-ioctl.c
> index e26b786..c8c906c 100644
> --- a/drivers/staging/android/ion/ion-ioctl.c
> +++ b/drivers/staging/android/ion/ion-ioctl.c
> @@ -25,8 +25,11 @@ union ion_ioctl_arg {
>   struct ion_heap_query query;
>  };
>  
> -static int validate_ioctl_arg(unsigned int cmd, union ion_ioctl_arg *arg)
> +static int validate_ioctl_arg(struct file *filp,
> +   unsigned int cmd, union ion_ioctl_arg *arg)
>  {
> + int mask = 1 << iminor(filp->f_inode);
> +
>   switch (cmd) {
>   case ION_IOC_HEAP_QUERY:
>   if (arg->query.reserved0 ||
> @@ -34,6 +37,10 @@ static int validate_ioctl_arg(unsigned int cmd, union 
> ion_ioctl_arg *arg)
>   arg->query.reserved2 )
>   return -EINVAL;
>   break;
> + case ION_IOC_ALLOC:
> + if (!(arg->allocation.heap_id_mask & mask))
> + return -EINVAL;
> + break;
>   default:
>   break;
>   }
> @@ -69,7 +76,7 @@ long ion_ioctl(struct file *filp, unsigned int cmd, 
> unsigned long arg)
>   if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd)))
>   return -EFAULT;
>  
> - ret = validate_ioctl_arg(cmd, &data);
> + ret = validate_ioctl_arg(filp, cmd, &data);
>   if (WARN_ON_ONCE(ret))
>   return ret;
>  
> diff --git a/drivers/staging/android/ion/ion.c 
> b/drivers/staging/android/ion/ion.c
> index 93e2c90..3f8b595 100644
> --- a/drivers/staging/android/ion/ion.c
> +++ b/drivers/staging/android/ion/ion.c
> @@ -40,6 +40,8 @@
>  
>  #include "ion.h"
>  
> +#define ION_DEV_MAX 32
> +
>  static struct ion_device *internal_dev;
>  static int heap_id;
>  
> @@ -541,11 +543,24 @@ void ion_device_add_heap(struct ion_heap *heap)
>  {
>   struct dentry *debug_file;
>   struct ion_device *dev = internal_dev;
> + int ret;
>  
>   if (!heap->ops->allocate || !heap->ops->free)
>   pr_err("%s: can not add heap with invalid ops struct.\n",
>  __func__);
>  
> + if (heap_id >= ION_DEV_MAX)
> + return -EBUSY;
> +
> + heap->ddev.devt = MKDEV(MAJOR(dev->devt), heap_id);
> + dev_set_name(&heap->ddev, "ion%d", heap_id);
> + device_initialize(&heap->ddev);
> + cdev_init(&heap->chrdev, &ion_fops);
> + heap->chrdev.owner = THIS_MODULE;
> + ret = cdev_device_add(&heap->chrdev, &heap->ddev);
> + if (ret < 0)
> + return;

No cleanup needed?  No reporting an error happened back up the chain?
Not nice :(

> +
>   spin_lock_init(&heap->free_lock);
>   heap->free_list_size = 0;
>  
> @@ -595,13 +610,9 @@ static int ion_device_create(void)
>   if (!idev)
>   return -ENOMEM;
>  
> - idev->dev.minor = MISC_DYNAMIC_MINOR;
> - idev->dev.name = "ion";
> - idev->dev.fops = &ion_fops;
> - idev->dev.parent = NULL;
> - ret = misc_register(&idev->dev);
> + ret = alloc_chrdev_region(&idev->devt, 0, ION_DEV_MAX, "ion");

Did you just change the major number for the device node as well?

Wow, that's a lot of userspace breakage (both major number, and name),
how did you test this?

Have you gotten "upstream" to agree to these changes?  We can't take
these until they think it's ok as well.

thanks,

greg k-h
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH v2 2/2] staging: ion: create one device entry per heap

2017-09-19 Thread Benjamin Gaignard

2017-09-19 13:02 GMT+02:00 Greg KH :
> On Tue, Sep 19, 2017 at 12:25:38PM +0200, Benjamin Gaignard wrote:
>> Instead a getting one common device "/dev/ion" for
>> all the heaps this patch allow to create one device
>> entry ("/dev/ionX") per heap.
>> Getting an entry per heap could allow to set security rules
>> per heap and global ones for all heaps.
>>
>> Allocation requests will be only allowed if the mask_id
>> match with device minor.
>> Query request could be done on any of the devices.
>>
>> Signed-off-by: Benjamin Gaignard 
>> ---
>>  drivers/staging/android/ion/ion-ioctl.c | 11 +--
>>  drivers/staging/android/ion/ion.c   | 23 +--
>>  drivers/staging/android/ion/ion.h   | 10 +++---
>>  3 files changed, 33 insertions(+), 11 deletions(-)
>>
>> diff --git a/drivers/staging/android/ion/ion-ioctl.c 
>> b/drivers/staging/android/ion/ion-ioctl.c
>> index e26b786..c8c906c 100644
>> --- a/drivers/staging/android/ion/ion-ioctl.c
>> +++ b/drivers/staging/android/ion/ion-ioctl.c
>> @@ -25,8 +25,11 @@ union ion_ioctl_arg {
>>   struct ion_heap_query query;
>>  };
>>
>> -static int validate_ioctl_arg(unsigned int cmd, union ion_ioctl_arg *arg)
>> +static int validate_ioctl_arg(struct file *filp,
>> +   unsigned int cmd, union ion_ioctl_arg *arg)
>>  {
>> + int mask = 1 << iminor(filp->f_inode);
>> +
>>   switch (cmd) {
>>   case ION_IOC_HEAP_QUERY:
>>   if (arg->query.reserved0 ||
>> @@ -34,6 +37,10 @@ static int validate_ioctl_arg(unsigned int cmd, union 
>> ion_ioctl_arg *arg)
>>   arg->query.reserved2 )
>>   return -EINVAL;
>>   break;
>> + case ION_IOC_ALLOC:
>> + if (!(arg->allocation.heap_id_mask & mask))
>> + return -EINVAL;
>> + break;
>>   default:
>>   break;
>>   }
>> @@ -69,7 +76,7 @@ long ion_ioctl(struct file *filp, unsigned int cmd, 
>> unsigned long arg)
>>   if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd)))
>>   return -EFAULT;
>>
>> - ret = validate_ioctl_arg(cmd, &data);
>> + ret = validate_ioctl_arg(filp, cmd, &data);
>>   if (WARN_ON_ONCE(ret))
>>   return ret;
>>
>> diff --git a/drivers/staging/android/ion/ion.c 
>> b/drivers/staging/android/ion/ion.c
>> index 93e2c90..3f8b595 100644
>> --- a/drivers/staging/android/ion/ion.c
>> +++ b/drivers/staging/android/ion/ion.c
>> @@ -40,6 +40,8 @@
>>
>>  #include "ion.h"
>>
>> +#define ION_DEV_MAX 32
>> +
>>  static struct ion_device *internal_dev;
>>  static int heap_id;
>>
>> @@ -541,11 +543,24 @@ void ion_device_add_heap(struct ion_heap *heap)
>>  {
>>   struct dentry *debug_file;
>>   struct ion_device *dev = internal_dev;
>> + int ret;
>>
>>   if (!heap->ops->allocate || !heap->ops->free)
>>   pr_err("%s: can not add heap with invalid ops struct.\n",
>>  __func__);
>>
>> + if (heap_id >= ION_DEV_MAX)
>> + return -EBUSY;
>> +
>> + heap->ddev.devt = MKDEV(MAJOR(dev->devt), heap_id);
>> + dev_set_name(&heap->ddev, "ion%d", heap_id);
>> + device_initialize(&heap->ddev);
>> + cdev_init(&heap->chrdev, &ion_fops);
>> + heap->chrdev.owner = THIS_MODULE;
>> + ret = cdev_device_add(&heap->chrdev, &heap->ddev);
>> + if (ret < 0)
>> + return;
>
> No cleanup needed?  No reporting an error happened back up the chain?
> Not nice :(

I will change that

>> +
>>   spin_lock_init(&heap->free_lock);
>>   heap->free_list_size = 0;
>>
>> @@ -595,13 +610,9 @@ static int ion_device_create(void)
>>   if (!idev)
>>   return -ENOMEM;
>>
>> - idev->dev.minor = MISC_DYNAMIC_MINOR;
>> - idev->dev.name = "ion";
>> - idev->dev.fops = &ion_fops;
>> - idev->dev.parent = NULL;
>> - ret = misc_register(&idev->dev);
>> + ret = alloc_chrdev_region(&idev->devt, 0, ION_DEV_MAX, "ion");
>
> Did you just change the major number for the device node as well?
>
My understanding of alloc_chrdev_region() is that major number is chosen
dynamically but I don't understand the link with device node, sorry.

> Wow, that's a lot of userspace breakage (both major number, and name),
> how did you test this?

I had to write a test by myself:
https://git.linaro.org/people/benjamin.gaignard/ion_test_application.git/log/?h=one_device_per_heap

Laura have tried to push a test VGEM but I believe it hasn't be
accepted yet (I will check)

>
> Have you gotten "upstream" to agree to these changes?  We can't take
> these until they think it's ok as well.

Split /dev/ion into multiple nodes is one of the task listed in
staging/android/TODO
file before been able to de-stage ion.

Since it has been a big bang in ion ABI on 4.12 and the fact  that
ion.h is still in
staging/android/uapi/ directory I do believe that userland is still unstable.
I hope this kind of patch will help to clarify what is still need

Re: [PATCH v2 2/2] staging: ion: create one device entry per heap

2017-09-19 Thread Dan Carpenter

If you had spelled out how this patch breaks user space in the changelog
then you wouldn't be catching so much flak for it now...  You should
fix that in v3.

regards,
dan carpenter

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH v2 1/2] include: linux: sysfs: Add __ATTR_NAMED macro

2017-09-19 Thread Lars-Peter Clausen

On 09/13/2017 11:29 PM, Greg KH wrote:
> On Wed, Sep 13, 2017 at 09:23:31PM +0200, Lars-Peter Clausen wrote:
>> On 09/13/2017 08:58 PM, Greg KH wrote:
>>> On Wed, Sep 13, 2017 at 06:03:10PM +0100, Jonathan Cameron wrote:
 On Wed, 13 Sep 2017 14:14:07 +0530
 Himanshi Jain  wrote:

> Add __ATTR_NAMED macro similar to __ATTR but taking name as a
> string instead of implicit conversion of argument to string using
> the macro _stringify(_name).
>
> Signed-off-by: Himanshi Jain 
> ---
>  include/linux/sysfs.h | 7 +++
>  1 file changed, 7 insertions(+)
>
> diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
> index aa02c32..20321cf 100644
> --- a/include/linux/sysfs.h
> +++ b/include/linux/sysfs.h
> @@ -104,6 +104,13 @@ struct attribute_group {
>   .store  = _store,   \
>  }
>  
> +#define __ATTR_NAMED(_name, _mode, _show, _store) {  
> \

 I'm not sure about the naming here.  The normal __ATTR macro is also
 'named'.  Maybe something as awful as

 __ATTR_STRING_NAME ? 

 Greg what do you think?
>>>
>>> ick ick ick.
>>>
 This is all to allow us to have names with operators in them without
 checkpatch complaining about them... A worthwhile aim just to stop
 more people wasting time trying to 'fix' those cases by adding spaces.
>>>
>>> Yeah, but this really seems "heavy" for just a crazy sysfs name in a
>>> macro.  Adding a whole new "core" define for that is a hard sell...
>>>
>>> I also want to get rid of the "generic" __ATTR type macros, and force
>>> people to use the proper _RW and friends instead.  I don't want to add
>>> another new one that people will start to use that I later have to
>>> change...
>>>
>>> So no, I don't like this, how about just changing your macros instead?
>>> No one else has this problem :)
>>
>> Nobody else realized they have this problem yet. E.g. there are a few users
>> of __ATTR in block/genhd.c that have the same issue and are likely to
>> generate the same false positives from static checkers.
> 
> Then fix the broken static checkers :)

The static checkers aren't broken, the macro is. It takes a string
parameter, but instead of a string it is passed as an expression and then
transformed to string using preprocessor magic in the macro. That's not very
good semantics. And hence this gets detected as a false positive, because
nobody expects such strange behavior.

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH 0/6] staging: Introduce DPAA2 Ethernet Switch driver

2017-09-19 Thread Andrew Lunn

On Tue, Sep 19, 2017 at 12:01:32PM +0300, Razvan Stefanescu wrote:
> This patchset introduces the Ethernet Switch Driver for Freescale/NXP SoCs
> with DPAA2 (DataPath Acceleration Architecture v2). The driver manages
> switch objects discovered on the fsl-mc bus. A description of the driver
> can be found in the associated README file.

Hi Razvan

You should probably Cc: netdev with these patches, if you want people
who have written switch drivers to review you code. You can also drop
linux-arm-kernel, since there is nothing ARM specific in these
patches.

Andrew
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH 5/6] staging: fsl-dpaa2/ethsw: Add README

2017-09-19 Thread Andrew Lunn

On Tue, Sep 19, 2017 at 12:01:37PM +0300, Razvan Stefanescu wrote:
> +Driver uses the switch device driver model and exposes each switch port as
> +a network interface, which can be included in a bridge. Traffic switched
> +between ports is offloaded into the hardware. Exposed network interfaces
> +are not used for I/O, they are used just for configuration. This
> +limitation is going to be addressed in the future.

Hi Razvan

Could you briefly describe how Ethernet frames get from the CPU to the
switch. This is what decided if you should write a plain switchdev
driver, or a DSA driver.

Andrew
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH v2 04/11] usb: xhci: Add Intel cherrytrail extended cap / otg phy mux handling

2017-09-19 Thread Mathias Nyman

Hi,

sorry about the long delay

On 07.09.2017 18:49, Hans de Goede wrote:

Hi,

On 07-09-17 15:14, Mathias Nyman wrote:

On 05.09.2017 19:42, Hans de Goede wrote:

The Intel cherrytrail xhci controller has an extended cap mmio-range
which contains registers to control the muxing to the xhci (host mode)
or the dwc3 (device mode) and vbus-detection for the otg usb-phy.

Having a mux driver included in the xhci code (or under drivers/usb/host)
is not desirable. So this commit adds a simple handler for this extended
capability, which creates a platform device with the caps mmio region as
resource, this allows us to write a separate platform mux driver for the
mux.

I think it would be better to have one place where we add handlers for
vendor specific extended capabilities.

Something like xhci-vendor-ext-caps.c, or just xhci-ext-caps.c as
there's a xhci-ext-caps.h header already

We could walk through the capability list once and add the needed handlers.
Something like:

+int xhci_ext_cap_init(void __iomem *base)

This will need to take a struct xhci_hcd *xhci param instead
as some of the ext_cap handling (including the cht mux code)
will need access to this.

yes, sample code added in second patch for reference/testing.

So I see 2 options here (without making this function PCI specific)
1) Add an u32 product_id field to struct xhci_hcd; or
2) Use a quirk flag as my current code is doing.

I'm fine with doing this either way, please let me know your preference.

Lets go with the quirk for now, I'll sort that out later

Can you do a "git format-patch" of that and send it to me? If you
can give me that + your preference for how to check if we're
dealing with a cht xhci hcd in xhci_ext_cap_init I can do a v3
with your suggestions applied.

Ended up modifying xhci_find_next_ext_cap() using id = 0 for
the next capability in list. Patch attached,

Second patch is just for reference how to use it.

Thanks
-Mathias

>From d5f26c1595f211ea7d46fca91551f70d1207330d Mon Sep 17 00:00:00 2001
From: Mathias Nyman 
Date: Tue, 19 Sep 2017 14:54:58 +0300
Subject: [PATCH 1/2] xhci: Add option to get next extended capability in list
 by passing id = 0

Modify xhci_find_next_ext_cap(base, offset, id) to return the next
capability offset if 0 is passed for id. Otherwise it will behave as
previously and return the offset of the next capability with matching id

capability id 0 is not used by xhci (reserved)

This is useful when we want to loop through all capabilities.

Signed-off-by: Mathias Nyman 
---
 drivers/usb/host/xhci-ext-caps.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/host/xhci-ext-caps.h b/drivers/usb/host/xhci-ext-caps.h
index 28deea5..c1b4042 100644
--- a/drivers/usb/host/xhci-ext-caps.h
+++ b/drivers/usb/host/xhci-ext-caps.h
@@ -96,7 +96,8 @@
  * @base	PCI MMIO registers base address.
  * @start	address at which to start looking, (0 or HCC_PARAMS to start at
  *		beginning of list)
- * @id		Extended capability ID to search for.
+ * @id		Extended capability ID to search for, or 0 for the next
+ *		capability
  *
  * Returns the offset of the next matching extended capability structure.
  * Some capabilities can occur several times, e.g., the XHCI_EXT_CAPS_PROTOCOL,
@@ -122,7 +123,7 @@ static inline int xhci_find_next_ext_cap(void __iomem *base, u32 start, int id)
 		val = readl(base + offset);
 		if (val == ~0)
 			return 0;
-		if (XHCI_EXT_CAPS_ID(val) == id && offset != start)
+		if (offset != start && (id == 0 || XHCI_EXT_CAPS_ID(val) == id))
 			return offset;

 		next = XHCI_EXT_CAPS_NEXT(val);
-- 
1.9.1

>From da44e961605d382829f90fdcfb90b61fa5ca9590 Mon Sep 17 00:00:00 2001
From: Mathias Nyman 
Date: Tue, 19 Sep 2017 14:58:40 +0300
Subject: [PATCH 2/2] xhci: test xhci_find_next_ext_cap with 0 id

NOT for UPSTREAM, just testing/reference code

Signed-off-by: Mathias Nyman 
---
 drivers/usb/host/xhci-ext-caps.h |  3 +++
 drivers/usb/host/xhci.c  | 30 ++
 2 files changed, 33 insertions(+)

diff --git a/drivers/usb/host/xhci-ext-caps.h b/drivers/usb/host/xhci-ext-caps.h
index c1b4042..7deb9e7 100644
--- a/drivers/usb/host/xhci-ext-caps.h
+++ b/drivers/usb/host/xhci-ext-caps.h
@@ -51,6 +51,9 @@
 #define XHCI_EXT_CAPS_ROUTE	5
 /* IDs 6-9 reserved */
 #define XHCI_EXT_CAPS_DEBUG	10
+/* IDs 192 - 255 vendor specific extensions */
+#define XHCI_EXT_CAPS_VENDOR_INTEL 192
+
 /* USB Legacy Support Capability - section 7.1.1 */
 #define XHCI_HC_BIOS_OWNED	(1 << 16)
 #define XHCI_HC_OS_OWNED	(1 << 24)
diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
index 870093a..99c804a 100644
--- a/drivers/usb/host/xhci.c
+++ b/drivers/usb/host/xhci.c
@@ -4772,6 +4772,34 @@ static int xhci_get_frame(struct usb_hcd *hcd)
 	return readl(&xhci->run_regs->microframe_index) >> 3;
 }

+int xhci_ext_cap_init(struct xhci_hcd *xhci)
+{
+	void __iomem *base;
+	u32 cap_offset;
+	u32 val;
+
+	base = &xhci->cap_regs->hc_capbase;
+	cap_offset = xh

Re: [PATCH v2 2/2] staging: ion: create one device entry per heap

2017-09-19 Thread Greg KH

On Tue, Sep 19, 2017 at 01:55:36PM +0200, Benjamin Gaignard wrote:
> >> +
> >>   spin_lock_init(&heap->free_lock);
> >>   heap->free_list_size = 0;
> >>
> >> @@ -595,13 +610,9 @@ static int ion_device_create(void)
> >>   if (!idev)
> >>   return -ENOMEM;
> >>
> >> - idev->dev.minor = MISC_DYNAMIC_MINOR;
> >> - idev->dev.name = "ion";
> >> - idev->dev.fops = &ion_fops;
> >> - idev->dev.parent = NULL;
> >> - ret = misc_register(&idev->dev);
> >> + ret = alloc_chrdev_region(&idev->devt, 0, ION_DEV_MAX, "ion");
> >
> > Did you just change the major number for the device node as well?
> >
> My understanding of alloc_chrdev_region() is that major number is chosen
> dynamically but I don't understand the link with device node, sorry.

Yes, the major is chosen dynamically if you ask for it (like you are
here), but previously you were using the misc major number.  That might
break userspace really badly if it had hard-coded /dev (like lots of
android devices do...)

> > Wow, that's a lot of userspace breakage (both major number, and name),
> > how did you test this?
> 
> I had to write a test by myself:
> https://git.linaro.org/people/benjamin.gaignard/ion_test_application.git/log/?h=one_device_per_heap
> 
> Laura have tried to push a test VGEM but I believe it hasn't be
> accepted yet (I will check)

A "test" program is a bit different than "boots and runs a working
system", right?  Please work with the correct graphics people to ensure
this change works with their code, before resending it.

thanks,

greg k-h
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[RESEND PATCH 0/6] staging: Introduce DPAA2 Ethernet Switch driver

2017-09-19 Thread Razvan Stefanescu

This patchset introduces the Ethernet Switch Driver for Freescale/NXP SoCs
with DPAA2 (DataPath Acceleration Architecture v2). The driver manages
switch objects discovered on the fsl-mc bus. A description of the driver
can be found in the associated README file.

The patchset consists of:
* A set of libraries containing APIs for configuring and controlling
  Management Complex (MC) switch objects
* The DPAA2 Ethernet Switch driver
* Patch adding ethtool support

Limitations:
* no support for control traffic to/from CPU
* only DPSW ports can be added to a bridge

Resending in order to add the netdev list (and remove the arm one), based
on Andrew Lunn's suggestion.

Razvan Stefanescu (6):
  staging: fsl-dpaa2/ethsw: Add APIs for DPSW object
  staging: fsl-dpaa2/ethsw: Add Freescale DPAA2 Ethernet Switch driver
  staging: fsl-dpaa2/ethsw: Add ethtool support
  staging: fsl-dpaa2/ethsw: Add maintainer for Ethernet Switch driver
  staging: fsl-dpaa2/ethsw: Add README
  staging: fsl-dpaa2/ethsw: Add TODO

 MAINTAINERS |6 +
 drivers/staging/fsl-dpaa2/Kconfig   |8 +
 drivers/staging/fsl-dpaa2/Makefile  |1 +
 drivers/staging/fsl-dpaa2/ethsw/Makefile|7 +
 drivers/staging/fsl-dpaa2/ethsw/README  |  106 ++
 drivers/staging/fsl-dpaa2/ethsw/TODO|   14 +
 drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h  |  371 ++
 drivers/staging/fsl-dpaa2/ethsw/dpsw.c  | 1147 +
 drivers/staging/fsl-dpaa2/ethsw/dpsw.h  |  611 +
 drivers/staging/fsl-dpaa2/ethsw/ethsw-ethtool.c |  207 +++
 drivers/staging/fsl-dpaa2/ethsw/ethsw.c | 1526 +++
 drivers/staging/fsl-dpaa2/ethsw/ethsw.h |   91 ++
 12 files changed, 4095 insertions(+)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/Makefile
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/README
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/TODO
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw.c
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw.h
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw-ethtool.c
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw.c
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw.h

-- 
1.9.1

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[RESEND PATCH 3/6] staging: fsl-dpaa2/ethsw: Add ethtool support

2017-09-19 Thread Razvan Stefanescu

Add driver information, link details and hardware statistics to be
reported via ethtool -S.

Signed-off-by: Razvan Stefanescu 
---
 drivers/staging/fsl-dpaa2/ethsw/Makefile|   2 +-
 drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h  |  13 ++
 drivers/staging/fsl-dpaa2/ethsw/dpsw.c  |  32 
 drivers/staging/fsl-dpaa2/ethsw/dpsw.h  |  32 
 drivers/staging/fsl-dpaa2/ethsw/ethsw-ethtool.c | 207 
 drivers/staging/fsl-dpaa2/ethsw/ethsw.c |   3 +
 drivers/staging/fsl-dpaa2/ethsw/ethsw.h |   3 +
 7 files changed, 291 insertions(+), 1 deletion(-)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw-ethtool.c

diff --git a/drivers/staging/fsl-dpaa2/ethsw/Makefile 
b/drivers/staging/fsl-dpaa2/ethsw/Makefile
index a6d72d1..de92cd9 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/Makefile
+++ b/drivers/staging/fsl-dpaa2/ethsw/Makefile
@@ -4,4 +4,4 @@
 
 obj-$(CONFIG_FSL_DPAA2_ETHSW) += dpaa2-ethsw.o
 
-dpaa2-ethsw-objs := ethsw.o dpsw.o
+dpaa2-ethsw-objs := ethsw.o ethsw-ethtool.o dpsw.o
diff --git a/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h 
b/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
index ddfd820..06b71122 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
+++ b/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
@@ -74,6 +74,8 @@
 #define DPSW_CMDID_IF_SET_FLOODING  DPSW_CMD_ID(0x047)
 #define DPSW_CMDID_IF_SET_BROADCAST DPSW_CMD_ID(0x048)
 
+#define DPSW_CMDID_IF_SET_LINK_CFG  DPSW_CMD_ID(0x04C)
+
 #define DPSW_CMDID_VLAN_ADD DPSW_CMD_ID(0x060)
 #define DPSW_CMDID_VLAN_ADD_IF  DPSW_CMD_ID(0x061)
 #define DPSW_CMDID_VLAN_ADD_IF_UNTAGGED DPSW_CMD_ID(0x062)
@@ -262,6 +264,17 @@ struct dpsw_cmd_if_set_max_frame_length {
__le16 frame_length;
 };
 
+struct dpsw_cmd_if_set_link_cfg {
+   /* cmd word 0 */
+   __le16 if_id;
+   u8 pad[6];
+   /* cmd word 1 */
+   __le32 rate;
+   __le32 pad1;
+   /* cmd word 2 */
+   __le64 options;
+};
+
 struct dpsw_cmd_if_get_link_state {
__le16 if_id;
 };
diff --git a/drivers/staging/fsl-dpaa2/ethsw/dpsw.c 
b/drivers/staging/fsl-dpaa2/ethsw/dpsw.c
index f36b92b..601172a 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/dpsw.c
+++ b/drivers/staging/fsl-dpaa2/ethsw/dpsw.c
@@ -383,6 +383,38 @@ int dpsw_get_attributes(struct fsl_mc_io *mc_io,
 }
 
 /**
+ * dpsw_if_set_link_cfg() - Set the link configuration.
+ * @mc_io: Pointer to MC portal's I/O object
+ * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token: Token of DPSW object
+ * @if_id: Interface id
+ * @cfg:   Link configuration
+ *
+ * Return: '0' on Success; Error code otherwise.
+ */
+int dpsw_if_set_link_cfg(struct fsl_mc_io *mc_io,
+u32 cmd_flags,
+u16 token,
+u16 if_id,
+struct dpsw_link_cfg *cfg)
+{
+   struct mc_command cmd = { 0 };
+   struct dpsw_cmd_if_set_link_cfg *cmd_params;
+
+   /* prepare command */
+   cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_SET_LINK_CFG,
+ cmd_flags,
+ token);
+   cmd_params = (struct dpsw_cmd_if_set_link_cfg *)cmd.params;
+   cmd_params->if_id = cpu_to_le16(if_id);
+   cmd_params->rate = cpu_to_le32(cfg->rate);
+   cmd_params->options = cpu_to_le64(cfg->options);
+
+   /* send command to mc*/
+   return mc_send_command(mc_io, &cmd);
+}
+
+/**
  * dpsw_if_get_link_state - Return the link state
  * @mc_io: Pointer to MC portal's I/O object
  * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_'
diff --git a/drivers/staging/fsl-dpaa2/ethsw/dpsw.h 
b/drivers/staging/fsl-dpaa2/ethsw/dpsw.h
index e9c2906..5feadc2 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/dpsw.h
+++ b/drivers/staging/fsl-dpaa2/ethsw/dpsw.h
@@ -245,6 +245,38 @@ enum dpsw_action {
 };
 
 /**
+ * Enable auto-negotiation
+ */
+#define DPSW_LINK_OPT_AUTONEG  0x0001ULL
+/**
+ * Enable half-duplex mode
+ */
+#define DPSW_LINK_OPT_HALF_DUPLEX  0x0002ULL
+/**
+ * Enable pause frames
+ */
+#define DPSW_LINK_OPT_PAUSE0x0004ULL
+/**
+ * Enable a-symmetric pause frames
+ */
+#define DPSW_LINK_OPT_ASYM_PAUSE   0x0008ULL
+
+/**
+ * struct dpsw_link_cfg - Structure representing DPSW link configuration
+ * @rate: Rate
+ * @options: Mask of available options; use 'DPSW_LINK_OPT_' values
+ */
+struct dpsw_link_cfg {
+   u32 rate;
+   u64 options;
+};
+
+int dpsw_if_set_link_cfg(struct fsl_mc_io *mc_io,
+u32 cmd_flags,
+u16 token,
+u16 if_id,
+struct dpsw_link_cfg *cfg);
+/**
  * struct dpsw_link_state - Structure representing DPSW link state
  * @rate: Rate
  * @options: Mask of available options; use 'DPSW_LINK_OPT_' values
diff --git a/drivers/stag

[RESEND PATCH 1/6] staging: fsl-dpaa2/ethsw: Add APIs for DPSW object

2017-09-19 Thread Razvan Stefanescu

Add the command build/parse APIs for operating on DPSW objects through
the DPAA2 Management Complex.

Signed-off-by: Razvan Stefanescu 
---
 drivers/staging/fsl-dpaa2/Kconfig  |8 +
 drivers/staging/fsl-dpaa2/Makefile |1 +
 drivers/staging/fsl-dpaa2/ethsw/Makefile   |7 +
 drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h |  358 +
 drivers/staging/fsl-dpaa2/ethsw/dpsw.c | 1115 
 drivers/staging/fsl-dpaa2/ethsw/dpsw.h |  579 +++
 6 files changed, 2068 insertions(+)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/Makefile
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw.c
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/dpsw.h

diff --git a/drivers/staging/fsl-dpaa2/Kconfig 
b/drivers/staging/fsl-dpaa2/Kconfig
index dfff675..8a508ef 100644
--- a/drivers/staging/fsl-dpaa2/Kconfig
+++ b/drivers/staging/fsl-dpaa2/Kconfig
@@ -16,3 +16,11 @@ config FSL_DPAA2_ETH
---help---
  Ethernet driver for Freescale DPAA2 SoCs, using the
  Freescale MC bus driver
+
+config FSL_DPAA2_ETHSW
+   tristate "Freescale DPAA2 Ethernet Switch"
+   depends on FSL_DPAA2
+   depends on NET_SWITCHDEV
+   ---help---
+   Driver for Freescale DPAA2 Ethernet Switch. Select
+   BRIDGE to have support for bridge tools.
diff --git a/drivers/staging/fsl-dpaa2/Makefile 
b/drivers/staging/fsl-dpaa2/Makefile
index 0836ba8..6cfd76b 100644
--- a/drivers/staging/fsl-dpaa2/Makefile
+++ b/drivers/staging/fsl-dpaa2/Makefile
@@ -3,3 +3,4 @@
 #
 
 obj-$(CONFIG_FSL_DPAA2_ETH)+= ethernet/
+obj-$(CONFIG_FSL_DPAA2_ETHSW)  += ethsw/
diff --git a/drivers/staging/fsl-dpaa2/ethsw/Makefile 
b/drivers/staging/fsl-dpaa2/ethsw/Makefile
new file mode 100644
index 000..db137f7
--- /dev/null
+++ b/drivers/staging/fsl-dpaa2/ethsw/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for the Freescale DPAA2 Ethernet Switch
+#
+
+obj-$(CONFIG_FSL_DPAA2_ETHSW) += dpaa2-ethsw.o
+
+dpaa2-ethsw-objs := dpsw.o
diff --git a/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h 
b/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
new file mode 100644
index 000..ddfd820
--- /dev/null
+++ b/drivers/staging/fsl-dpaa2/ethsw/dpsw-cmd.h
@@ -0,0 +1,358 @@
+/* Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ *  notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ *  notice, this list of conditions and the following disclaimer in the
+ *  documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ *  names of any contributors may be used to endorse or promote products
+ *  derived from this software without specific prior written permission.
+ *
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+#ifndef __FSL_DPSW_CMD_H
+#define __FSL_DPSW_CMD_H
+
+/* DPSW Version */
+#define DPSW_VER_MAJOR 8
+#define DPSW_VER_MINOR 0
+
+#define DPSW_CMD_BASE_VERSION  1
+#define DPSW_CMD_ID_OFFSET 4
+
+#define DPSW_CMD_ID(id)(((id) << DPSW_CMD_ID_OFFSET) | 
DPSW_CMD_BASE_VERSION)
+
+/* Command IDs */
+#define DPSW_CMDID_CLOSEDPSW_CMD_ID(0x800)
+#define DPSW_CMDID_OPEN DPSW_CMD_ID(0x802)
+
+#define DPSW_CMDID_GET_API_VERSION  DPSW_CMD_ID(0xa02)
+
+#define DPSW_CMDID_ENABLE   DPSW_CMD_ID(0x002)
+#define DPSW_CMDID_DISABLE  DPSW_CMD_ID(0x003)
+#define DPSW_CMDID_GET_ATTR DPSW_CMD_ID(0x004)
+#define DPSW_CMDID_RESETDPSW_CMD_ID(0x005)
+
+#define DPSW_CMDID_SET_IRQ_ENABLE

[RESEND PATCH 6/6] staging: fsl-dpaa2/ethsw: Add TODO

2017-09-19 Thread Razvan Stefanescu

Add a TODO file describing what needs to be added/changed before the driver
can be moved out of staging.

Signed-off-by: Razvan Stefanescu 
---
 drivers/staging/fsl-dpaa2/ethsw/TODO | 14 ++
 1 file changed, 14 insertions(+)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/TODO

diff --git a/drivers/staging/fsl-dpaa2/ethsw/TODO 
b/drivers/staging/fsl-dpaa2/ethsw/TODO
new file mode 100644
index 000..d3f12c3
--- /dev/null
+++ b/drivers/staging/fsl-dpaa2/ethsw/TODO
@@ -0,0 +1,14 @@
+* Add I/O capabilities on switch port netdevices. This will allow control
+traffic to reach the CPU.
+* Add ACL to redirect control traffic to CPU.
+* Add support for displaying learned FDB entries
+* MC firmware uprev; the DPAA2 objects used by the Ethernet Switch driver
+need to   be kept in sync with binary interface changes in MC
+* refine README file
+* cleanup
+
+NOTE: At least first three of the above are required before getting the
+DPAA2 Ethernet Switch driver out of staging. Another requirement is that
+the fsl-mc bus driver is moved to drivers/bus and dpio driver is moved to
+drivers/soc (this is required for I/O).
+
-- 
1.9.1

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[RESEND PATCH 5/6] staging: fsl-dpaa2/ethsw: Add README

2017-09-19 Thread Razvan Stefanescu

Add a README file describing the driver architecture, components and
interfaces.

Signed-off-by: Razvan Stefanescu 
---
 drivers/staging/fsl-dpaa2/ethsw/README | 106 +
 1 file changed, 106 insertions(+)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/README

diff --git a/drivers/staging/fsl-dpaa2/ethsw/README 
b/drivers/staging/fsl-dpaa2/ethsw/README
new file mode 100644
index 000..f6fc07f
--- /dev/null
+++ b/drivers/staging/fsl-dpaa2/ethsw/README
@@ -0,0 +1,106 @@
+DPAA2 Ethernet Switch driver
+
+
+This file provides documentation for the DPAA2 Ethernet Switch driver
+
+
+Contents
+
+   Supported Platforms
+   Architecture Overview
+   Creating an Ethernet Switch
+   Features
+
+
+   Supported Platforms
+===
+This driver provides networking support for Freescale LS2085A, LS2088A
+DPAA2 SoCs.
+
+
+Architecture Overview
+=
+The Ethernet Switch in the DPAA2 architecture consists of several hardware
+resources that provide the functionality. These are allocated and
+configured via the Management Complex (MC) portals. MC abstracts most of
+these resources as DPAA2 objects and exposes ABIs through which they can
+be configured and controlled.
+
+For a more detailed description of the DPAA2 architecture and its object
+abstractions see:
+   drivers/staging/fsl-mc/README.txt
+
+The Ethernet Switch is built on top of a Datapath Switch (DPSW) object.
+
+Configuration interface:
+
+  -
+ | DPAA2 Switch driver |
+  -
+   .
+   .
+  --
+ | DPSW API |
+  --
+   .   software
+ = . ==
+   .   hardware
+  -
+ | MC hardware portals |
+  -
+   .
+   .
+ --
+| DPSW |
+ --
+
+Driver uses the switch device driver model and exposes each switch port as
+a network interface, which can be included in a bridge. Traffic switched
+between ports is offloaded into the hardware. Exposed network interfaces
+are not used for I/O, they are used just for configuration. This
+limitation is going to be addressed in the future.
+
+The DPSW can have ports connected to DPNIs or to PHYs via DPMACs.
+
+
+ [ethA] [ethB] [ethC] [ethD] [ethE] [ethF]
+:  :  :  :  :  :
+:  :  :  :  :  :
+[eth drv]  [eth drv]  [ethsw drv  ]
+:  :  :  :  :  :kernel
+
+:  :  :  :  :  :hardware
+ [DPNI]  [DPNI] [= DPSW =]
+|  |  |  |  |  |
+|   --   |   [DPMAC][DPMAC]
+ ---|  |
+|  |
+  [PHY]  [PHY]
+
+For a more detailed description of the Ethernet switch device driver model
+see:
+   Documentation/networking/switchdev.txt
+
+Creating an Ethernet Switch
+===
+A device is created for the switch objects probed on the MC bus. Each DPSW
+has a number of properties which determine the configuration options and
+associated hardware resources.
+
+A DPSW object (and the other DPAA2 objects needed for a DPAA2 switch) can
+be added to a container on the MC bus in one of two ways: statically,
+through a Datapath Layout Binary file (DPL) that is parsed by MC at boot
+time; or created dynamically at runtime, via the DPAA2 objects APIs.
+
+Features
+
+Driver configures DPSW to perform hardware switching offload of
+unicast/multicast/broadcast (VLAN tagged or untagged) traffic between its
+ports.
+
+It allows configuration of hardware learning, flooding, multicast groups,
+port VLAN configuration and STP state.
+
+Static entries can be added/removed from the FDB.
+
+Hardware statistics for each port are provided through ethtool -S option.
-- 
1.9.1

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[RESEND PATCH 2/6] staging: fsl-dpaa2/ethsw: Add Freescale DPAA2 Ethernet Switch driver

2017-09-19 Thread Razvan Stefanescu

Introduce the DPAA2 Ethernet Switch driver, which manages Datapath Switch
(DPSW) objects discovered on the MC bus.

Suggested-by: Alexandru Marginean 
Signed-off-by: Razvan Stefanescu 
---
 drivers/staging/fsl-dpaa2/ethsw/Makefile |2 +-
 drivers/staging/fsl-dpaa2/ethsw/ethsw.c  | 1523 ++
 drivers/staging/fsl-dpaa2/ethsw/ethsw.h  |   88 ++
 3 files changed, 1612 insertions(+), 1 deletion(-)
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw.c
 create mode 100644 drivers/staging/fsl-dpaa2/ethsw/ethsw.h

diff --git a/drivers/staging/fsl-dpaa2/ethsw/Makefile 
b/drivers/staging/fsl-dpaa2/ethsw/Makefile
index db137f7..a6d72d1 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/Makefile
+++ b/drivers/staging/fsl-dpaa2/ethsw/Makefile
@@ -4,4 +4,4 @@
 
 obj-$(CONFIG_FSL_DPAA2_ETHSW) += dpaa2-ethsw.o
 
-dpaa2-ethsw-objs := dpsw.o
+dpaa2-ethsw-objs := ethsw.o dpsw.o
diff --git a/drivers/staging/fsl-dpaa2/ethsw/ethsw.c 
b/drivers/staging/fsl-dpaa2/ethsw/ethsw.c
new file mode 100644
index 000..ae86078
--- /dev/null
+++ b/drivers/staging/fsl-dpaa2/ethsw/ethsw.c
@@ -0,0 +1,1523 @@
+/* Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ *  notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ *  notice, this list of conditions and the following disclaimer in the
+ *  documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ *  names of any contributors may be used to endorse or promote products
+ *  derived from this software without specific prior written permission.
+ *
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+
+#include 
+#include 
+#include 
+#include 
+
+#include "../../fsl-mc/include/mc.h"
+
+#include "ethsw.h"
+
+static struct workqueue_struct *ethsw_owq;
+
+/* Minimal supported DPSW version */
+#define DPSW_MIN_VER_MAJOR 8
+#define DPSW_MIN_VER_MINOR 0
+
+#define DEFAULT_VLAN_ID1
+
+static int ethsw_add_vlan(struct ethsw_core *ethsw, u16 vid)
+{
+   int err;
+
+   struct dpsw_vlan_cfgvcfg = {
+   .fdb_id = 0,
+   };
+
+   if (ethsw->vlans[vid]) {
+   dev_err(ethsw->dev, "VLAN already configured\n");
+   return -EEXIST;
+   }
+
+   err = dpsw_vlan_add(ethsw->mc_io, 0,
+   ethsw->dpsw_handle, vid, &vcfg);
+   if (err) {
+   dev_err(ethsw->dev, "dpsw_vlan_add err %d\n", err);
+   return err;
+   }
+   ethsw->vlans[vid] = ETHSW_VLAN_MEMBER;
+
+   return 0;
+}
+
+static int ethsw_port_add_vlan(struct ethsw_port_priv *port_priv,
+  u16 vid, u16 flags)
+{
+   struct ethsw_core *ethsw = port_priv->ethsw_data;
+   struct net_device *netdev = port_priv->netdev;
+   struct dpsw_vlan_if_cfg vcfg;
+   bool is_oper;
+   int err, err2;
+
+   if (port_priv->vlans[vid]) {
+   netdev_warn(netdev, "VLAN %d already configured\n", vid);
+   return -EEXIST;
+   }
+
+   vcfg.num_ifs = 1;
+   vcfg.if_id[0] = port_priv->idx;
+   err = dpsw_vlan_add_if(ethsw->mc_io, 0, ethsw->dpsw_handle, vid, &vcfg);
+   if (err) {
+   netdev_err(netdev, "dpsw_vlan_add_if err %d\n", err);
+   return err;
+   }
+
+   port_priv->vlans[vid] = ETHSW_VLAN_MEMBER;
+
+   if (flags & BRIDGE_VLAN_INFO_UNTAGGED) {
+   err = dpsw_vlan_add_if_untagged(ethsw->mc_io, 0,
+   ethsw->dpsw_handle,
+

[RESEND PATCH 4/6] staging: fsl-dpaa2/ethsw: Add maintainer for Ethernet Switch driver

2017-09-19 Thread Razvan Stefanescu

Signed-off-by: Razvan Stefanescu 
---
 MAINTAINERS | 6 ++
 1 file changed, 6 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 2281af4..cfd4f74 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4297,6 +4297,12 @@ L:   linux-ker...@vger.kernel.org
 S: Maintained
 F: drivers/staging/fsl-dpaa2/ethernet
 
+DPAA2 ETHERNET SWITCH DRIVER
+M: Razvan Stefanescu 
+L: linux-ker...@vger.kernel.org
+S: Maintained
+F: drivers/staging/fsl-dpaa2/ethsw
+
 DPT_I2O SCSI RAID DRIVER
 M: Adaptec OEM Raid Solutions 
 L: linux-s...@vger.kernel.org
-- 
1.9.1

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH v2 05/11] mux: Add Intel Cherrytrail USB mux driver

2017-09-19 Thread Hans de Goede


Hi,

Thank you for the reviews and patches!

On 09/08/2017 05:45 PM, Peter Rosin wrote:

On 2017-09-05 18:42, Hans de Goede wrote:

Intel Cherrytrail SoCs have an internal USB mux for muxing the otg-port
USB data lines between the xHCI host controller and the dwc3 gadget
controller. On some Cherrytrail systems this mux is controlled through
AML code reacting on a GPIO IRQ connected to the USB OTG id pin (through
an _AIE ACPI method) so things just work.

But on other Cherrytrail systems we need to control the mux ourselves
this driver exports the mux through the mux subsys, so that other drivers
can control it if necessary.

This driver also updates the vbus-valid reporting to the dwc3 gadget
controller, as this uses the same registers as the mux. This is needed
for gadget/device mode to work properly (even on systems which control
the mux from their AML code).

Note this depends on the xhci driver registering a platform device
named "intel_cht_usb_mux", which has an IOMEM resource 0 which points
to the Intel Vendor Defined XHCI extended capabilities region.

Signed-off-by: Hans de Goede 
---
Changes in v2:
-Complete rewrite as a stand-alone platform-driver rather then as a phy
  driver, since this is just a mux, not a phy

Changes in v3:
-Make this a mux subsys driver instead of listening to USB_HOST extcon
  cable events and responding to those

Changes in v4 (of patch, v2 of new mux based series):
-Rename C-file to use - in name
-Add MAINTAINERS entry
-Various code-style fixes
---
  MAINTAINERS |   5 +
  drivers/mux/Kconfig |  11 ++
  drivers/mux/Makefile|  10 +-
  drivers/mux/intel-cht-usb-mux.c | 257 
  4 files changed, 279 insertions(+), 4 deletions(-)
  create mode 100644 drivers/mux/intel-cht-usb-mux.c

diff --git a/MAINTAINERS b/MAINTAINERS
index 14a2fd905217..dfaed958db85 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8991,6 +8991,11 @@ F:   include/linux/dt-bindings/mux/
  F:include/linux/mux/
  F:drivers/mux/
  
+MULTIPLEXER SUBSYSTEM INTEL CHT USB MUX DRIVER

+M: Hans de Goede 
+S: Maintained
+F: drivers/mix/intel-cht-usb-mux.c


s/mix/mux/

(also in patch 06/11)


+
  MULTISOUND SOUND DRIVER
  M:Andrew Veliath 
  S:Maintained
diff --git a/drivers/mux/Kconfig b/drivers/mux/Kconfig
index 19e4e904c9bf..947cfd7af02a 100644
--- a/drivers/mux/Kconfig
+++ b/drivers/mux/Kconfig
@@ -34,6 +34,17 @@ config MUX_GPIO
  To compile the driver as a module, choose M here: the module will
  be called mux-gpio.
  
+config MUX_INTEL_CHT_USB_MUX

+   tristate "Intel Cherrytrail USB Multiplexer"
+   depends on ACPI && X86 && EXTCON
+   help
+ This driver adds support for the internal USB mux for muxing the OTG
+ USB data lines between the xHCI host controller and the dwc3 gadget
+ controller found on Intel Cherrytrail SoCs.
+
+ To compile the driver as a module, choose M here: the module will
+ be called mux-intel_cht_usb_mux.


s/_/-/g


+
  config MUX_MMIO
tristate "MMIO register bitfield-controlled Multiplexer"
depends on (OF && MFD_SYSCON) || COMPILE_TEST
diff --git a/drivers/mux/Makefile b/drivers/mux/Makefile
index 0e1e59760e3f..6cf41be2754f 100644
--- a/drivers/mux/Makefile
+++ b/drivers/mux/Makefile
@@ -5,9 +5,11 @@
  mux-core-objs := core.o
  mux-adg792a-objs  := adg792a.o
  mux-gpio-objs := gpio.o
+mux-intel-cht-usb-mux-objs := intel-cht-usb-mux.o
  mux-mmio-objs := mmio.o
  
-obj-$(CONFIG_MULTIPLEXER)	+= mux-core.o

-obj-$(CONFIG_MUX_ADG792A)  += mux-adg792a.o
-obj-$(CONFIG_MUX_GPIO) += mux-gpio.o
-obj-$(CONFIG_MUX_MMIO) += mux-mmio.o
+obj-$(CONFIG_MULTIPLEXER)  += mux-core.o
+obj-$(CONFIG_MUX_ADG792A)  += mux-adg792a.o
+obj-$(CONFIG_MUX_GPIO) += mux-gpio.o
+obj-$(CONFIG_MUX_INTEL_CHT_USB_MUX)+= mux-intel-cht-usb-mux.o
+obj-$(CONFIG_MUX_MMIO) += mux-mmio.o
diff --git a/drivers/mux/intel-cht-usb-mux.c b/drivers/mux/intel-cht-usb-mux.c
new file mode 100644
index ..9cd1a1f5027f
--- /dev/null
+++ b/drivers/mux/intel-cht-usb-mux.c
@@ -0,0 +1,257 @@
+/*
+ * Intel Cherrytrail USB OTG MUX driver
+ *
+ * Copyright (c) 2016 Hans de Goede 


2017?


Actually I stated working on this (in a different form before the mux framework 
was
merge) quite a while back I will make this 2016-2017.




+ *
+ * Loosely based on android x86 kernel code which is:
+ *
+ * Copyright (C) 2014 Intel Corp.
+ *
+ * Author: Wu, Hao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation, or (at your option)
+ * any later version.
+ */
+
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#inclu

Re: [PATCH 1/2] mux: add mux_control_get_optional() API

2017-09-19 Thread Hans de Goede


Hi,

On 09/08/2017 05:54 PM, Peter Rosin wrote:

On 2017-09-08 17:45, Peter Rosin wrote:

From: Stephen Boyd 

Sometimes drivers only use muxes under certain scenarios. For
example, the chipidea usb controller may be connected to a usb
switch on some platforms, and connected directly to a usb port on
others. The driver won't know one way or the other though, so add
a mux_control_get_optional() API that allows the driver to
differentiate errors getting the mux from there not being a mux
for the driver to use at all.
---
  Documentation/driver-model/devres.txt |   1 +
  drivers/mux/core.c| 104 +++---
  include/linux/mux/consumer.h  |   4 ++
  3 files changed, 89 insertions(+), 20 deletions(-)

I haven't tested this patch, and hence I have not signed it and I also
removed the sign-off from Stephen...


Huh, I definitely intended to mention that this patch is based on [1]
from Stephen, but that I've made changes according to the comments in
that thread (and more). And those changes are what made me remove the
sign-off from Stephen...


AFAIK normally a Signed-off-by is kept if some (small-ish) changes
are made. The S-o-b is mostly an indication that the author is
ok with adding the code to the kernel under the GPL.

Anyways, Stephen are you ok with re-adding your S-o-b to
the version modified by Peter?

Regards,

Hans
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] [media] staging: atomisp: use clock framework for camera clocks

2017-09-19 Thread Pierre-Louis Bossart

The Atom ISP driver initializes and configures PMC clocks which are
already handled by the clock framework.

Remove all legacy vlv2_platform_clock stuff and move to the clk API to
avoid conflicts, e.g. with audio machine drivers enabling the MCLK for
external codecs

Tested-by: Carlo Caione 
Signed-off-by: Pierre-Louis Bossart 
---
 drivers/staging/media/atomisp/Kconfig  |   1 +
 drivers/staging/media/atomisp/platform/Makefile|   1 -
 .../staging/media/atomisp/platform/clock/Makefile  |   6 -
 .../platform/clock/platform_vlv2_plat_clk.c|  40 
 .../platform/clock/platform_vlv2_plat_clk.h|  27 ---
 .../media/atomisp/platform/clock/vlv2_plat_clock.c | 247 -
 .../platform/intel-mid/atomisp_gmin_platform.c |  63 +-
 7 files changed, 52 insertions(+), 333 deletions(-)
 delete mode 100644 drivers/staging/media/atomisp/platform/clock/Makefile
 delete mode 100644 
drivers/staging/media/atomisp/platform/clock/platform_vlv2_plat_clk.c
 delete mode 100644 
drivers/staging/media/atomisp/platform/clock/platform_vlv2_plat_clk.h
 delete mode 100644 
drivers/staging/media/atomisp/platform/clock/vlv2_plat_clock.c

diff --git a/drivers/staging/media/atomisp/Kconfig 
b/drivers/staging/media/atomisp/Kconfig
index 8eb13c3ba29c..7cdebea35ccf 100644
--- a/drivers/staging/media/atomisp/Kconfig
+++ b/drivers/staging/media/atomisp/Kconfig
@@ -1,6 +1,7 @@
 menuconfig INTEL_ATOMISP
 bool "Enable support to Intel MIPI camera drivers"
 depends on X86 && EFI && MEDIA_CONTROLLER && PCI && ACPI
+   select COMMON_CLK
 help
   Enable support for the Intel ISP2 camera interfaces and MIPI
   sensor drivers.
diff --git a/drivers/staging/media/atomisp/platform/Makefile 
b/drivers/staging/media/atomisp/platform/Makefile
index df157630bda9..0e3b7e1c81c6 100644
--- a/drivers/staging/media/atomisp/platform/Makefile
+++ b/drivers/staging/media/atomisp/platform/Makefile
@@ -2,5 +2,4 @@
 # Makefile for camera drivers.
 #
 
-obj-$(CONFIG_INTEL_ATOMISP) += clock/
 obj-$(CONFIG_INTEL_ATOMISP) += intel-mid/
diff --git a/drivers/staging/media/atomisp/platform/clock/Makefile 
b/drivers/staging/media/atomisp/platform/clock/Makefile
deleted file mode 100644
index 82fbe8b6968a..
diff --git 
a/drivers/staging/media/atomisp/platform/clock/platform_vlv2_plat_clk.c 
b/drivers/staging/media/atomisp/platform/clock/platform_vlv2_plat_clk.c
deleted file mode 100644
index 0aae9b0283bb..
diff --git 
a/drivers/staging/media/atomisp/platform/clock/platform_vlv2_plat_clk.h 
b/drivers/staging/media/atomisp/platform/clock/platform_vlv2_plat_clk.h
deleted file mode 100644
index b730ab0e8223..
diff --git a/drivers/staging/media/atomisp/platform/clock/vlv2_plat_clock.c 
b/drivers/staging/media/atomisp/platform/clock/vlv2_plat_clock.c
deleted file mode 100644
index f96789a31819..
diff --git 
a/drivers/staging/media/atomisp/platform/intel-mid/atomisp_gmin_platform.c 
b/drivers/staging/media/atomisp/platform/intel-mid/atomisp_gmin_platform.c
index edaae93af8f9..17b4cfae5abf 100644
--- a/drivers/staging/media/atomisp/platform/intel-mid/atomisp_gmin_platform.c
+++ b/drivers/staging/media/atomisp/platform/intel-mid/atomisp_gmin_platform.c
@@ -4,10 +4,10 @@
 #include 
 #include 
 #include 
+#include 
 #include 
 #include 
 #include 
-#include "../../include/linux/vlv2_plat_clock.h"
 #include 
 #include 
 #include 
@@ -17,11 +17,7 @@
 
 #define MAX_SUBDEVS 8
 
-/* Should be defined in vlv2_plat_clock API, isn't: */
-#define VLV2_CLK_PLL_19P2MHZ 1
-#define VLV2_CLK_XTAL_19P2MHZ 0
-#define VLV2_CLK_ON  1
-#define VLV2_CLK_OFF 2
+#define VLV2_CLK_PLL_19P2MHZ 1 /* XTAL on CHT */
 #define ELDO1_SEL_REG  0x19
 #define ELDO1_1P8V 0x16
 #define ELDO1_CTRL_SHIFT 0x00
@@ -33,6 +29,7 @@ struct gmin_subdev {
struct v4l2_subdev *subdev;
int clock_num;
int clock_src;
+   struct clk *pmc_clk;
struct gpio_desc *gpio0;
struct gpio_desc *gpio1;
struct regulator *v1p8_reg;
@@ -344,6 +341,9 @@ static int gmin_platform_deinit(void)
return 0;
 }
 
+#define GMIN_PMC_CLK_NAME 14 /* "pmc_plt_clk_[0..5]" */
+static char gmin_pmc_clk_name[GMIN_PMC_CLK_NAME];
+
 static struct gmin_subdev *gmin_subdev_add(struct v4l2_subdev *subdev)
 {
int i, ret;
@@ -377,6 +377,37 @@ static struct gmin_subdev *gmin_subdev_add(struct 
v4l2_subdev *subdev)
gmin_subdevs[i].gpio0 = gpiod_get_index(dev, NULL, 0, GPIOD_OUT_LOW);
gmin_subdevs[i].gpio1 = gpiod_get_index(dev, NULL, 1, GPIOD_OUT_LOW);
 
+   /* get PMC clock with clock framework */
+   snprintf(gmin_pmc_clk_name,
+sizeof(gmin_pmc_clk_name),
+"%s_%d", "pmc_plt_clk", gmin_subdevs[i].clock_num);
+
+   gmin_subdevs[i].pmc_clk = devm_clk_get(dev, gmin_pmc_clk_name);
+   if (IS_ERR(gmin_subdevs[i].pmc_clk)) {
+   ret = PTR_ERR(gmin_subdevs[i].pmc_clk);
+
+   dev_

Re: [PATCH v2 2/2] staging: ion: create one device entry per heap

2017-09-19 Thread Laura Abbott


On 09/19/2017 04:55 AM, Benjamin Gaignard wrote:

2017-09-19 13:02 GMT+02:00 Greg KH :

On Tue, Sep 19, 2017 at 12:25:38PM +0200, Benjamin Gaignard wrote:

Instead a getting one common device "/dev/ion" for
all the heaps this patch allow to create one device
entry ("/dev/ionX") per heap.
Getting an entry per heap could allow to set security rules
per heap and global ones for all heaps.

Allocation requests will be only allowed if the mask_id
match with device minor.
Query request could be done on any of the devices.

Signed-off-by: Benjamin Gaignard 
---
  drivers/staging/android/ion/ion-ioctl.c | 11 +--
  drivers/staging/android/ion/ion.c   | 23 +--
  drivers/staging/android/ion/ion.h   | 10 +++---
  3 files changed, 33 insertions(+), 11 deletions(-)

diff --git a/drivers/staging/android/ion/ion-ioctl.c 
b/drivers/staging/android/ion/ion-ioctl.c
index e26b786..c8c906c 100644
--- a/drivers/staging/android/ion/ion-ioctl.c
+++ b/drivers/staging/android/ion/ion-ioctl.c
@@ -25,8 +25,11 @@ union ion_ioctl_arg {
   struct ion_heap_query query;
  };

-static int validate_ioctl_arg(unsigned int cmd, union ion_ioctl_arg *arg)
+static int validate_ioctl_arg(struct file *filp,
+   unsigned int cmd, union ion_ioctl_arg *arg)
  {
+ int mask = 1 << iminor(filp->f_inode);
+
   switch (cmd) {
   case ION_IOC_HEAP_QUERY:
   if (arg->query.reserved0 ||
@@ -34,6 +37,10 @@ static int validate_ioctl_arg(unsigned int cmd, union 
ion_ioctl_arg *arg)
   arg->query.reserved2 )
   return -EINVAL;
   break;
+ case ION_IOC_ALLOC:
+ if (!(arg->allocation.heap_id_mask & mask))
+ return -EINVAL;
+ break;
   default:
   break;
   }
@@ -69,7 +76,7 @@ long ion_ioctl(struct file *filp, unsigned int cmd, unsigned 
long arg)
   if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd)))
   return -EFAULT;

- ret = validate_ioctl_arg(cmd, &data);
+ ret = validate_ioctl_arg(filp, cmd, &data);
   if (WARN_ON_ONCE(ret))
   return ret;

diff --git a/drivers/staging/android/ion/ion.c 
b/drivers/staging/android/ion/ion.c
index 93e2c90..3f8b595 100644
--- a/drivers/staging/android/ion/ion.c
+++ b/drivers/staging/android/ion/ion.c
@@ -40,6 +40,8 @@

  #include "ion.h"

+#define ION_DEV_MAX 32
+
  static struct ion_device *internal_dev;
  static int heap_id;

@@ -541,11 +543,24 @@ void ion_device_add_heap(struct ion_heap *heap)
  {
   struct dentry *debug_file;
   struct ion_device *dev = internal_dev;
+ int ret;

   if (!heap->ops->allocate || !heap->ops->free)
   pr_err("%s: can not add heap with invalid ops struct.\n",
  __func__);

+ if (heap_id >= ION_DEV_MAX)
+ return -EBUSY;
+
+ heap->ddev.devt = MKDEV(MAJOR(dev->devt), heap_id);
+ dev_set_name(&heap->ddev, "ion%d", heap_id);
+ device_initialize(&heap->ddev);
+ cdev_init(&heap->chrdev, &ion_fops);
+ heap->chrdev.owner = THIS_MODULE;
+ ret = cdev_device_add(&heap->chrdev, &heap->ddev);
+ if (ret < 0)
+ return;


No cleanup needed?  No reporting an error happened back up the chain?
Not nice :(


I will change that


+
   spin_lock_init(&heap->free_lock);
   heap->free_list_size = 0;

@@ -595,13 +610,9 @@ static int ion_device_create(void)
   if (!idev)
   return -ENOMEM;

- idev->dev.minor = MISC_DYNAMIC_MINOR;
- idev->dev.name = "ion";
- idev->dev.fops = &ion_fops;
- idev->dev.parent = NULL;
- ret = misc_register(&idev->dev);
+ ret = alloc_chrdev_region(&idev->devt, 0, ION_DEV_MAX, "ion");


Did you just change the major number for the device node as well?


My understanding of alloc_chrdev_region() is that major number is chosen
dynamically but I don't understand the link with device node, sorry.


Wow, that's a lot of userspace breakage (both major number, and name),
how did you test this?


I had to write a test by myself:
https://git.linaro.org/people/benjamin.gaignard/ion_test_application.git/log/?h=one_device_per_heap

Laura have tried to push a test VGEM but I believe it hasn't be
accepted yet (I will check)



Have you gotten "upstream" to agree to these changes?  We can't take
these until they think it's ok as well.


Split /dev/ion into multiple nodes is one of the task listed in
staging/android/TODO
file before been able to de-stage ion.

Since it has been a big bang in ion ABI on 4.12 and the fact  that
ion.h is still in
staging/android/uapi/ directory I do believe that userland is still unstable.
I hope this kind of patch will help to clarify what is still need to
be done to de-stage ion
even if this patch is NACK-ed we can at least the item from the TODO list.



Thanks for sending this Benjamin.

At plumbers, it was requested to not break the ABI too much or do it in
one last big bang

[PATCH 1/6] staging: fsl-dpaa2/ethsw: Add APIs for DPSW object

[PATCH 0/6] staging: Introduce DPAA2 Ethernet Switch driver

[PATCH 3/6] staging: fsl-dpaa2/ethsw: Add ethtool support

[PATCH 2/6] staging: fsl-dpaa2/ethsw: Add Freescale DPAA2 Ethernet Switch driver

[PATCH 4/6] staging: fsl-dpaa2/ethsw: Add maintainer for Ethernet Switch driver

[PATCH 5/6] staging: fsl-dpaa2/ethsw: Add README

[PATCH 6/6] staging: fsl-dpaa2/ethsw: Add TODO

Re: [PATCH] staging: ion: create one device entry per heap

Re: [PATCH] staging: ion: create one device entry per heap

Re: [PATCH] staging: ion: create one device entry per heap

Re: [PATCH] staging: ion: create one device entry per heap

[PATCH v2 0/2] staging: ion: get one device per heap

[PATCH v2 2/2] staging: ion: create one device entry per heap

[PATCH v2 1/2] staging: ion: simplify ioctl args checking function

Re: [PATCH] staging: ion: create one device entry per heap

Re: [PATCH v2 2/2] staging: ion: create one device entry per heap

Re: [PATCH v2 2/2] staging: ion: create one device entry per heap

Re: [PATCH v2 2/2] staging: ion: create one device entry per heap

Re: [PATCH v2 1/2] include: linux: sysfs: Add __ATTR_NAMED macro

Re: [PATCH 0/6] staging: Introduce DPAA2 Ethernet Switch driver

Re: [PATCH 5/6] staging: fsl-dpaa2/ethsw: Add README

Re: [PATCH v2 04/11] usb: xhci: Add Intel cherrytrail extended cap / otg phy mux handling

Re: [PATCH v2 2/2] staging: ion: create one device entry per heap

[RESEND PATCH 0/6] staging: Introduce DPAA2 Ethernet Switch driver

[RESEND PATCH 3/6] staging: fsl-dpaa2/ethsw: Add ethtool support

[RESEND PATCH 1/6] staging: fsl-dpaa2/ethsw: Add APIs for DPSW object

[RESEND PATCH 6/6] staging: fsl-dpaa2/ethsw: Add TODO

[RESEND PATCH 5/6] staging: fsl-dpaa2/ethsw: Add README

[RESEND PATCH 2/6] staging: fsl-dpaa2/ethsw: Add Freescale DPAA2 Ethernet Switch driver

[RESEND PATCH 4/6] staging: fsl-dpaa2/ethsw: Add maintainer for Ethernet Switch driver

Re: [PATCH v2 05/11] mux: Add Intel Cherrytrail USB mux driver

Re: [PATCH 1/2] mux: add mux_control_get_optional() API

[PATCH] [media] staging: atomisp: use clock framework for camera clocks

Re: [PATCH v2 2/2] staging: ion: create one device entry per heap

34 matches

Site Navigation

Mail list logo

Footer information