* Gigabit ethernet driver for Alacritechs SLIC devices (v3)
@ 2016-11-26 12:20 Lino Sanfilippo
2016-11-26 12:20 ` [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver Lino Sanfilippo
` (2 more replies)
0 siblings, 3 replies; 14+ messages in thread
From: Lino Sanfilippo @ 2016-11-26 12:20 UTC (permalink / raw)
To: davem, charrer, liodot, gregkh, andrew; +Cc: devel, linux-kernel, netdev
Hi,
this is the third version of the slicoss gigabit ethernet driver (which is a
rework of the driver from Alacritech which can currently be found under
drivers/staging/slicoss). The driver is supposed to support Mojave, Oasis and
Kalahari cards, for both copper and fiber.
If this code is accepted the staging version can be removed.
The driver has been tested on a SEN2104ET adapter (4 Port PCIe copper).
v3:
- dont add defines to pci.h but instead put it into the drivers header file
(requested by Greg Kroah-Hartman)
v2:
- remove unusual padding in statistic strings (suggested by Andrew Lunn)
- for mdio register and bit names use defines from mii.h instead of own ones
(suggested by Andrew Lunn)
- remove unused defines
- ensure PCI flush at two more places
- use mmiowb before lock to prevent mmio writes leaking out of lock
- fix some typos in comments
- add copyright and GPL header
Regards,
Lino
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-26 12:20 Gigabit ethernet driver for Alacritechs SLIC devices (v3) Lino Sanfilippo
@ 2016-11-26 12:20 ` Lino Sanfilippo
2016-11-26 15:48 ` Rami Rosen
` (3 more replies)
2016-11-26 12:20 ` [PATCH v3 net-next 2/2] MAINTAINERS: add entry for slicoss " Lino Sanfilippo
2016-11-26 12:36 ` Gigabit ethernet driver for Alacritechs SLIC devices (v3) Lino Sanfilippo
2 siblings, 4 replies; 14+ messages in thread
From: Lino Sanfilippo @ 2016-11-26 12:20 UTC (permalink / raw)
To: davem, charrer, liodot, gregkh, andrew
Cc: devel, linux-kernel, netdev, Lino Sanfilippo
Add driver for Alacritech gigabit ethernet cards with SLIC (session-layer
interface control) technology. The driver provides basic support without
SLIC for the following devices:
- Mojave cards (single port PCI Gigabit) both copper and fiber
- Oasis cards (single and dual port PCI-x Gigabit) copper and fiber
- Kalahari cards (dual and quad port PCI-e Gigabit) copper and fiber
Signed-off-by: Lino Sanfilippo <LinoSanfilippo@gmx.de>
---
drivers/net/ethernet/Kconfig | 1 +
drivers/net/ethernet/Makefile | 1 +
drivers/net/ethernet/alacritech/Kconfig | 28 +
drivers/net/ethernet/alacritech/Makefile | 4 +
drivers/net/ethernet/alacritech/slic.h | 576 +++++++++
drivers/net/ethernet/alacritech/slicoss.c | 1867 +++++++++++++++++++++++++++++
6 files changed, 2477 insertions(+)
create mode 100644 drivers/net/ethernet/alacritech/Kconfig
create mode 100644 drivers/net/ethernet/alacritech/Makefile
create mode 100644 drivers/net/ethernet/alacritech/slic.h
create mode 100644 drivers/net/ethernet/alacritech/slicoss.c
diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig
index 2ffd634..a4cc87fe 100644
--- a/drivers/net/ethernet/Kconfig
+++ b/drivers/net/ethernet/Kconfig
@@ -21,6 +21,7 @@ source "drivers/net/ethernet/3com/Kconfig"
source "drivers/net/ethernet/adaptec/Kconfig"
source "drivers/net/ethernet/aeroflex/Kconfig"
source "drivers/net/ethernet/agere/Kconfig"
+source "drivers/net/ethernet/alacritech/Kconfig"
source "drivers/net/ethernet/allwinner/Kconfig"
source "drivers/net/ethernet/alteon/Kconfig"
source "drivers/net/ethernet/altera/Kconfig"
diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile
index 1d349e9..b448027 100644
--- a/drivers/net/ethernet/Makefile
+++ b/drivers/net/ethernet/Makefile
@@ -7,6 +7,7 @@ obj-$(CONFIG_NET_VENDOR_8390) += 8390/
obj-$(CONFIG_NET_VENDOR_ADAPTEC) += adaptec/
obj-$(CONFIG_GRETH) += aeroflex/
obj-$(CONFIG_NET_VENDOR_AGERE) += agere/
+obj-$(CONFIG_NET_VENDOR_ALACRITECH) += alacritech/
obj-$(CONFIG_NET_VENDOR_ALLWINNER) += allwinner/
obj-$(CONFIG_NET_VENDOR_ALTEON) += alteon/
obj-$(CONFIG_ALTERA_TSE) += altera/
diff --git a/drivers/net/ethernet/alacritech/Kconfig b/drivers/net/ethernet/alacritech/Kconfig
new file mode 100644
index 0000000..41000a3
--- /dev/null
+++ b/drivers/net/ethernet/alacritech/Kconfig
@@ -0,0 +1,28 @@
+config NET_VENDOR_ALACRITECH
+ bool "Alacritech devices"
+ default y
+ ---help---
+ If you have a network (Ethernet) card belonging to this class, say Y.
+
+ Note that the answer to this question doesn't directly affect the
+ kernel: saying N will just cause the configurator to skip all
+ the questions about Renesas devices. If you say Y, you will be asked
+ for your specific device in the following questions.
+
+if NET_VENDOR_ALACRITECH
+
+config SLICOSS
+ tristate "Alacritech Slicoss support"
+ depends on PCI
+ select CRC32
+ ---help---
+ This driver supports Gigabit Ethernet adapters based on the
+ Session Layer Interface (SLIC) technology by Alacritech.
+
+ Supported are Mojave (1 port) and Oasis (1, 2 and 4 port) cards,
+ both copper and fiber.
+
+ To compile this driver as a module, choose M here: the module
+ will be called slicoss. This is recommended.
+
+endif # NET_VENDOR_ALACRITECH
diff --git a/drivers/net/ethernet/alacritech/Makefile b/drivers/net/ethernet/alacritech/Makefile
new file mode 100644
index 0000000..8790e9e
--- /dev/null
+++ b/drivers/net/ethernet/alacritech/Makefile
@@ -0,0 +1,4 @@
+#
+# Makefile for the Alacritech Slicoss driver
+#
+obj-$(CONFIG_SLICOSS) += slicoss.o
diff --git a/drivers/net/ethernet/alacritech/slic.h b/drivers/net/ethernet/alacritech/slic.h
new file mode 100644
index 0000000..c62d46b
--- /dev/null
+++ b/drivers/net/ethernet/alacritech/slic.h
@@ -0,0 +1,576 @@
+
+#ifndef _SLIC_H
+#define _SLIC_H
+
+#include <linux/types.h>
+#include <linux/netdevice.h>
+#include <linux/spinlock_types.h>
+#include <linux/dma-mapping.h>
+#include <linux/pci.h>
+#include <linux/netdevice.h>
+#include <linux/list.h>
+#include <linux/u64_stats_sync.h>
+
+#define SLIC_VGBSTAT_XPERR 0x40000000
+#define SLIC_VGBSTAT_XERRSHFT 25
+#define SLIC_VGBSTAT_XCSERR 0x23
+#define SLIC_VGBSTAT_XUFLOW 0x22
+#define SLIC_VGBSTAT_XHLEN 0x20
+#define SLIC_VGBSTAT_NETERR 0x01000000
+#define SLIC_VGBSTAT_NERRSHFT 16
+#define SLIC_VGBSTAT_NERRMSK 0x1ff
+#define SLIC_VGBSTAT_NCSERR 0x103
+#define SLIC_VGBSTAT_NUFLOW 0x102
+#define SLIC_VGBSTAT_NHLEN 0x100
+#define SLIC_VGBSTAT_LNKERR 0x00000080
+#define SLIC_VGBSTAT_LERRMSK 0xff
+#define SLIC_VGBSTAT_LDEARLY 0x86
+#define SLIC_VGBSTAT_LBOFLO 0x85
+#define SLIC_VGBSTAT_LCODERR 0x84
+#define SLIC_VGBSTAT_LDBLNBL 0x83
+#define SLIC_VGBSTAT_LCRCERR 0x82
+#define SLIC_VGBSTAT_LOFLO 0x81
+#define SLIC_VGBSTAT_LUFLO 0x80
+
+#define SLIC_IRHDDR_FLEN_MSK 0x0000ffff
+#define SLIC_IRHDDR_SVALID 0x80000000
+#define SLIC_IRHDDR_ERR 0x10000000
+
+#define SLIC_VRHSTAT_802OE 0x80000000
+#define SLIC_VRHSTAT_TPOFLO 0x10000000
+#define SLIC_VRHSTATB_802UE 0x80000000
+#define SLIC_VRHSTATB_RCVE 0x40000000
+#define SLIC_VRHSTATB_BUFF 0x20000000
+#define SLIC_VRHSTATB_CARRE 0x08000000
+#define SLIC_VRHSTATB_LONGE 0x02000000
+#define SLIC_VRHSTATB_PREA 0x01000000
+#define SLIC_VRHSTATB_CRC 0x00800000
+#define SLIC_VRHSTATB_DRBL 0x00400000
+#define SLIC_VRHSTATB_CODE 0x00200000
+#define SLIC_VRHSTATB_TPCSUM 0x00100000
+#define SLIC_VRHSTATB_TPHLEN 0x00080000
+#define SLIC_VRHSTATB_IPCSUM 0x00040000
+#define SLIC_VRHSTATB_IPLERR 0x00020000
+#define SLIC_VRHSTATB_IPHERR 0x00010000
+
+#define SLIC_CMD_XMT_REQ 0x01
+#define SLIC_CMD_TYPE_DUMB 3
+
+#define SLIC_RESET_MAGIC 0xDEAD
+#define SLIC_ICR_INT_OFF 0
+#define SLIC_ICR_INT_ON 1
+#define SLIC_ICR_INT_MASK 2
+
+#define SLIC_ISR_ERR 0x80000000
+#define SLIC_ISR_RCV 0x40000000
+#define SLIC_ISR_CMD 0x20000000
+#define SLIC_ISR_IO 0x60000000
+#define SLIC_ISR_UPC 0x10000000
+#define SLIC_ISR_LEVENT 0x08000000
+#define SLIC_ISR_RMISS 0x02000000
+#define SLIC_ISR_UPCERR 0x01000000
+#define SLIC_ISR_XDROP 0x00800000
+#define SLIC_ISR_UPCBSY 0x00020000
+
+#define SLIC_ISR_PING_MASK 0x00700000
+#define SLIC_ISR_UPCERR_MASK (SLIC_ISR_UPCERR | SLIC_ISR_UPCBSY)
+#define SLIC_ISR_UPC_MASK (SLIC_ISR_UPC | SLIC_ISR_UPCERR_MASK)
+#define SLIC_WCS_START 0x80000000
+#define SLIC_WCS_COMPARE 0x40000000
+#define SLIC_RCVWCS_BEGIN 0x40000000
+#define SLIC_RCVWCS_FINISH 0x80000000
+
+#define SLIC_MIICR_REG_16 0x00100000
+#define SLIC_MRV_REG16_XOVERON 0x0068
+
+#define SLIC_GIG_LINKUP 0x0001
+#define SLIC_GIG_FULLDUPLEX 0x0002
+#define SLIC_GIG_SPEED_MASK 0x000C
+#define SLIC_GIG_SPEED_1000 0x0008
+#define SLIC_GIG_SPEED_100 0x0004
+#define SLIC_GIG_SPEED_10 0x0000
+
+#define SLIC_GMCR_RESET 0x80000000
+#define SLIC_GMCR_GBIT 0x20000000
+#define SLIC_GMCR_FULLD 0x10000000
+#define SLIC_GMCR_GAPBB_SHIFT 14
+#define SLIC_GMCR_GAPR1_SHIFT 7
+#define SLIC_GMCR_GAPR2_SHIFT 0
+#define SLIC_GMCR_GAPBB_1000 0x60
+#define SLIC_GMCR_GAPR1_1000 0x2C
+#define SLIC_GMCR_GAPR2_1000 0x40
+#define SLIC_GMCR_GAPBB_100 0x70
+#define SLIC_GMCR_GAPR1_100 0x2C
+#define SLIC_GMCR_GAPR2_100 0x40
+
+#define SLIC_XCR_RESET 0x80000000
+#define SLIC_XCR_XMTEN 0x40000000
+#define SLIC_XCR_PAUSEEN 0x20000000
+#define SLIC_XCR_LOADRNG 0x10000000
+
+#define SLIC_GXCR_RESET 0x80000000
+#define SLIC_GXCR_XMTEN 0x40000000
+#define SLIC_GXCR_PAUSEEN 0x20000000
+
+#define SLIC_GRCR_RESET 0x80000000
+#define SLIC_GRCR_RCVEN 0x40000000
+#define SLIC_GRCR_RCVALL 0x20000000
+#define SLIC_GRCR_RCVBAD 0x10000000
+#define SLIC_GRCR_CTLEN 0x08000000
+#define SLIC_GRCR_ADDRAEN 0x02000000
+#define SLIC_GRCR_HASHSIZE_SHIFT 17
+#define SLIC_GRCR_HASHSIZE 14
+
+/* Reset Register */
+#define SLIC_REG_RESET 0x0000
+/* Interrupt Control Register */
+#define SLIC_REG_ICR 0x0008
+/* Interrupt status pointer */
+#define SLIC_REG_ISP 0x0010
+/* Interrupt status */
+#define SLIC_REG_ISR 0x0018
+/* Header buffer address reg
+ * 31-8 - phy addr of set of contiguous hdr buffers
+ * 7-0 - number of buffers passed
+ * Buffers are 256 bytes long on 256-byte boundaries.
+ */
+#define SLIC_REG_HBAR 0x0020
+/* Data buffer handle & address reg
+ * 4 sets of registers; Buffers are 2K bytes long 2 per 4K page.
+ */
+#define SLIC_REG_DBAR 0x0028
+/* Xmt Cmd buf addr regs.
+ * 1 per XMT interface
+ * 31-5 - phy addr of host command buffer
+ * 4-0 - length of cmd in multiples of 32 bytes
+ * Buffers are 32 bytes up to 512 bytes long
+ */
+#define SLIC_REG_CBAR 0x0030
+/* Write control store */
+#define SLIC_REG_WCS 0x0034
+/*Response buffer address reg.
+ * 31-8 - phy addr of set of contiguous response buffers
+ * 7-0 - number of buffers passed
+ * Buffers are 32 bytes long on 32-byte boundaries.
+ */
+#define SLIC_REG_RBAR 0x0038
+/* Read statistics (UPR) */
+#define SLIC_REG_RSTAT 0x0040
+/* Read link status */
+#define SLIC_REG_LSTAT 0x0048
+/* Write Mac Config */
+#define SLIC_REG_WMCFG 0x0050
+/* Write phy register */
+#define SLIC_REG_WPHY 0x0058
+/* Rcv Cmd buf addr reg */
+#define SLIC_REG_RCBAR 0x0060
+/* Read SLIC Config*/
+#define SLIC_REG_RCONFIG 0x0068
+/* Interrupt aggregation time */
+#define SLIC_REG_INTAGG 0x0070
+/* Write XMIT config reg */
+#define SLIC_REG_WXCFG 0x0078
+/* Write RCV config reg */
+#define SLIC_REG_WRCFG 0x0080
+/* Write rcv addr a low */
+#define SLIC_REG_WRADDRAL 0x0088
+/* Write rcv addr a high */
+#define SLIC_REG_WRADDRAH 0x0090
+/* Write rcv addr b low */
+#define SLIC_REG_WRADDRBL 0x0098
+/* Write rcv addr b high */
+#define SLIC_REG_WRADDRBH 0x00a0
+/* Low bits of mcast mask */
+#define SLIC_REG_MCASTLOW 0x00a8
+/* High bits of mcast mask */
+#define SLIC_REG_MCASTHIGH 0x00b0
+/* Ping the card */
+#define SLIC_REG_PING 0x00b8
+/* Dump command */
+#define SLIC_REG_DUMP_CMD 0x00c0
+/* Dump data pointer */
+#define SLIC_REG_DUMP_DATA 0x00c8
+/* Read card's pci_status register */
+#define SLIC_REG_PCISTATUS 0x00d0
+/* Write hostid field */
+#define SLIC_REG_WRHOSTID 0x00d8
+/* Put card in a low power state */
+#define SLIC_REG_LOW_POWER 0x00e0
+/* Force slic into quiescent state before soft reset */
+#define SLIC_REG_QUIESCE 0x00e8
+/* Reset interface queues */
+#define SLIC_REG_RESET_IFACE 0x00f0
+/* Register is only written when it has changed.
+ * Bits 63-32 for host i/f addrs.
+ */
+#define SLIC_REG_ADDR_UPPER 0x00f8
+/* 64 bit Header buffer address reg */
+#define SLIC_REG_HBAR64 0x0100
+/* 64 bit Data buffer handle & address reg */
+#define SLIC_REG_DBAR64 0x0108
+/* 64 bit Xmt Cmd buf addr regs. */
+#define SLIC_REG_CBAR64 0x0110
+/* 64 bit Response buffer address reg.*/
+#define SLIC_REG_RBAR64 0x0118
+/* 64 bit Rcv Cmd buf addr reg*/
+#define SLIC_REG_RCBAR64 0x0120
+/* Read statistics (64 bit UPR) */
+#define SLIC_REG_RSTAT64 0x0128
+/* Download Gigabit RCV sequencer ucode */
+#define SLIC_REG_RCV_WCS 0x0130
+/* Write VlanId field */
+#define SLIC_REG_WRVLANID 0x0138
+/* Read Transformer info */
+#define SLIC_REG_READ_XF_INFO 0x0140
+/* Write Transformer info */
+#define SLIC_REG_WRITE_XF_INFO 0x0148
+/* Write card ticks per second */
+#define SLIC_REG_TICKS_PER_SEC 0x0170
+#define SLIC_REG_HOSTID 0x1554
+
+#define PCI_VENDOR_ID_ALACRITECH 0x139A
+#define PCI_DEVICE_ID_ALACRITECH_MOAVE 0x0005
+#define PCI_SUBDEVICE_ID_ALACRITECH_1000X1 0x0005
+#define PCI_SUBDEVICE_ID_ALACRITECH_1000X1_2 0x0006
+#define PCI_SUBDEVICE_ID_ALACRITECH_1000X1F 0x0007
+#define PCI_SUBDEVICE_ID_ALACRITECH_CICADA 0x0008
+#define PCI_SUBDEVICE_ID_ALACRITECH_SES1001T 0x2006
+#define PCI_SUBDEVICE_ID_ALACRITECH_SES1001F 0x2007
+#define PCI_DEVICE_ID_ALACRITECH_OASIS 0x0007
+#define PCI_SUBDEVICE_ID_ALACRITECH_SEN2002XT 0x000B
+#define PCI_SUBDEVICE_ID_ALACRITECH_SEN2002XF 0x000C
+#define PCI_SUBDEVICE_ID_ALACRITECH_SEN2001XT 0x000D
+#define PCI_SUBDEVICE_ID_ALACRITECH_SEN2001XF 0x000E
+#define PCI_SUBDEVICE_ID_ALACRITECH_SEN2104EF 0x000F
+#define PCI_SUBDEVICE_ID_ALACRITECH_SEN2104ET 0x0010
+#define PCI_SUBDEVICE_ID_ALACRITECH_SEN2102EF 0x0011
+#define PCI_SUBDEVICE_ID_ALACRITECH_SEN2102ET 0x0012
+
+/* Note: power of two required for number descriptors */
+#define SLIC_NUM_RX_LES 256
+#define SLIC_RX_BUFF_SIZE 2048
+#define SLIC_RX_BUFF_ALIGN 256
+#define SLIC_RX_BUFF_HDR_SIZE 34
+#define SLIC_MAX_REQ_RX_DESCS 1
+
+#define SLIC_NUM_TX_DESCS 256
+#define SLIC_TX_DESC_ALIGN 32
+#define SLIC_MIN_TX_WAKEUP_DESCS 10
+#define SLIC_MAX_REQ_TX_DESCS 1
+#define SLIC_MAX_TX_COMPLETIONS 100
+
+#define SLIC_NUM_STAT_DESCS 128
+#define SLIC_STATS_DESC_ALIGN 256
+
+#define SLIC_NUM_STAT_DESC_ARRAYS 4
+#define SLIC_INVALID_STAT_DESC_IDX 0xffffffff
+
+#define SLIC_NAPI_WEIGHT 64
+
+#define SLIC_UPR_LSTAT 0
+#define SLIC_UPR_CONFIG 1
+
+#define SLIC_EEPROM_SIZE 128
+#define SLIC_EEPROM_MAGIC 0xa5a5
+
+#define SLIC_FIRMWARE_MOAVE "slicoss/gbdownload.sys"
+#define SLIC_FIRMWARE_OASIS "slicoss/oasisdownload.sys"
+#define SLIC_RCV_FIRMWARE_MOAVE "slicoss/gbrcvucode.sys"
+#define SLIC_RCV_FIRMWARE_OASIS "slicoss/oasisrcvucode.sys"
+#define SLIC_FIRMWARE_MIN_SIZE 64
+#define SLIC_FIRMWARE_MAX_SECTIONS 3
+
+#define SLIC_MODEL_MOJAVE 0
+#define SLIC_MODEL_OASIS 1
+
+#define SLIC_INC_STATS_COUNTER(st, counter) \
+do { \
+ u64_stats_update_begin(&(st)->syncp); \
+ (st)->counter++; \
+ u64_stats_update_end(&(st)->syncp); \
+} while (0)
+
+#define SLIC_GET_STATS_COUNTER(newst, st, counter) \
+{ \
+ unsigned int start; \
+ do { \
+ start = u64_stats_fetch_begin_irq(&(st)->syncp); \
+ newst = (st)->counter; \
+ } while (u64_stats_fetch_retry_irq(&(st)->syncp, start)); \
+}
+
+struct slic_upr {
+ dma_addr_t paddr;
+ unsigned int type;
+ struct list_head list;
+};
+
+struct slic_upr_list {
+ bool pending;
+ struct list_head list;
+ /* upr list lock */
+ spinlock_t lock;
+};
+
+/* SLIC EEPROM structure for Oasis */
+struct slic_mojave_eeprom {
+ __le16 id; /* 00 EEPROM/FLASH Magic code 'A5A5'*/
+ __le16 eeprom_code_size;/* 01 Size of EEPROM Codes (bytes * 4)*/
+ __le16 flash_size; /* 02 Flash size */
+ __le16 eeprom_size; /* 03 EEPROM Size */
+ __le16 vendor_id; /* 04 Vendor ID */
+ __le16 dev_id; /* 05 Device ID */
+ u8 rev_id; /* 06 Revision ID */
+ u8 class_code[3]; /* 07 Class Code */
+ u8 irqpin_dbg; /* 08 Debug Interrupt pin */
+ u8 irqpin; /* Network Interrupt Pin */
+ u8 min_grant; /* 09 Minimum grant */
+ u8 max_lat; /* Maximum Latency */
+ __le16 pci_stat; /* 10 PCI Status */
+ __le16 sub_vendor_id; /* 11 Subsystem Vendor Id */
+ __le16 sub_id; /* 12 Subsystem ID */
+ __le16 dev_id_dbg; /* 13 Debug Device Id */
+ __le16 ramrom; /* 14 Dram/Rom function */
+ __le16 dram_size2pci; /* 15 DRAM size to PCI (bytes * 64K) */
+ __le16 rom_size2pci; /* 16 ROM extension size to PCI (bytes * 4k) */
+ u8 pad[2]; /* 17 Padding */
+ u8 freetime; /* 18 FreeTime setting */
+ u8 ifctrl; /* 10-bit interface control (Mojave only) */
+ __le16 dram_size; /* 19 DRAM size (bytes * 64k) */
+ u8 mac[ETH_ALEN]; /* 20 MAC addresses */
+ u8 mac2[ETH_ALEN];
+ u8 pad2[6];
+ u16 dev_id2; /* Device ID for 2nd PCI function */
+ u8 irqpin2; /* Interrupt pin for 2nd PCI function */
+ u8 class_code2[3]; /* Class Code for 2nd PCI function */
+ u16 cfg_byte6; /* Config Byte 6 */
+ u16 pme_cap; /* Power Mgment capabilities */
+ u16 nwclk_ctrl; /* NetworkClockControls */
+ u8 fru_format; /* Alacritech FRU format type */
+ u8 fru_assembly[6]; /* Alacritech FRU information */
+ u8 fru_rev[2];
+ u8 fru_serial[14];
+ u8 fru_pad[3];
+ u8 oem_fru[28]; /* optional OEM FRU format type */
+ u8 pad3[4]; /* Pad to 128 bytes - includes 2 cksum bytes
+ * (if OEM FRU info exists) and two unusable
+ * bytes at the end
+ */
+};
+
+/* SLIC EEPROM structure for Oasis */
+struct slic_oasis_eeprom {
+ __le16 id; /* 00 EEPROM/FLASH Magic code 'A5A5' */
+ __le16 eeprom_code_size;/* 01 Size of EEPROM Codes (bytes * 4)*/
+ __le16 spidev0_cfg; /* 02 Flash Config for SPI device 0 */
+ __le16 spidev1_cfg; /* 03 Flash Config for SPI device 1 */
+ __le16 vendor_id; /* 04 Vendor ID */
+ __le16 dev_id; /* 05 Device ID (function 0) */
+ u8 rev_id; /* 06 Revision ID */
+ u8 class_code0[3]; /* 07 Class Code for PCI function 0 */
+ u8 irqpin1; /* 08 Interrupt pin for PCI function 1*/
+ u8 class_code1[3]; /* 09 Class Code for PCI function 1 */
+ u8 irqpin2; /* 10 Interrupt pin for PCI function 2*/
+ u8 irqpin0; /* Interrupt pin for PCI function 0*/
+ u8 min_grant; /* 11 Minimum grant */
+ u8 max_lat; /* Maximum Latency */
+ __le16 sub_vendor_id; /* 12 Subsystem Vendor Id */
+ __le16 sub_id; /* 13 Subsystem ID */
+ __le16 flash_size; /* 14 Flash size (bytes / 4K) */
+ __le16 dram_size2pci; /* 15 DRAM size to PCI (bytes / 64K) */
+ __le16 rom_size2pci; /* 16 Flash (ROM extension) size to PCI
+ * (bytes / 4K)
+ */
+ __le16 dev_id1; /* 17 Device Id (function 1) */
+ __le16 dev_id2; /* 18 Device Id (function 2) */
+ __le16 dev_stat_cfg; /* 19 Device Status Config Bytes 6-7 */
+ __le16 pme_cap; /* 20 Power Mgment capabilities */
+ u8 msi_cap; /* 21 MSI capabilities */
+ u8 clock_div; /* Clock divider */
+ __le16 pci_stat_lo; /* 22 PCI Status bits 15:0 */
+ __le16 pci_stat_hi; /* 23 PCI Status bits 31:16 */
+ __le16 dram_cfg_lo; /* 24 DRAM Configuration bits 15:0 */
+ __le16 dram_cfg_hi; /* 25 DRAM Configuration bits 31:16 */
+ __le16 dram_size; /* 26 DRAM size (bytes / 64K) */
+ __le16 gpio_tbi_ctrl; /* 27 GPIO/TBI controls for functions 1/0 */
+ __le16 eeprom_size; /* 28 EEPROM Size */
+ u8 mac[ETH_ALEN]; /* 29 MAC addresses (2 ports) */
+ u8 mac2[ETH_ALEN];
+ u8 fru_format; /* 35 Alacritech FRU format type */
+ u8 fru_assembly[6]; /* Alacritech FRU information */
+ u8 fru_rev[2];
+ u8 fru_serial[14];
+ u8 fru_pad[3];
+ u8 oem_fru[28]; /* optional OEM FRU information */
+ u8 pad[4]; /* Pad to 128 bytes - includes 2 checksum bytes
+ * (if OEM FRU info exists) and two unusable
+ * bytes at the end
+ */
+};
+
+struct slic_stats {
+ u64 rx_packets;
+ u64 rx_bytes;
+ u64 rx_mcasts;
+ u64 rx_errors;
+ u64 tx_packets;
+ u64 tx_bytes;
+ /* HW STATS */
+ u64 rx_buff_miss;
+ u64 tx_dropped;
+ u64 irq_errs;
+ /* transport layer */
+ u64 rx_tpcsum;
+ u64 rx_tpoflow;
+ u64 rx_tphlen;
+ /* ip layer */
+ u64 rx_ipcsum;
+ u64 rx_iplen;
+ u64 rx_iphlen;
+ /* link layer */
+ u64 rx_early;
+ u64 rx_buffoflow;
+ u64 rx_lcode;
+ u64 rx_drbl;
+ u64 rx_crc;
+ u64 rx_oflow802;
+ u64 rx_uflow802;
+ /* oasis only */
+ u64 tx_carrier;
+ struct u64_stats_sync syncp;
+};
+
+struct slic_shmem_data {
+ __le32 isr;
+ __le32 link;
+};
+
+struct slic_shmem {
+ dma_addr_t isr_paddr;
+ dma_addr_t link_paddr;
+ struct slic_shmem_data *shmem_data;
+};
+
+struct slic_rx_info_oasis {
+ __le32 frame_status;
+ __le32 frame_status_b;
+ __le32 time_stamp;
+ __le32 checksum;
+};
+
+struct slic_rx_info_mojave {
+ __le32 frame_status;
+ __le16 byte_cnt;
+ __le16 tp_chksum;
+ __le16 ctx_hash;
+ __le16 mac_hash;
+ __le16 buff_lnk;
+};
+
+struct slic_stat_desc {
+ __le32 hnd;
+ __u8 pad[8];
+ __le32 status;
+ __u8 pad2[16];
+};
+
+struct slic_stat_queue {
+ struct slic_stat_desc *descs[SLIC_NUM_STAT_DESC_ARRAYS];
+ dma_addr_t paddr[SLIC_NUM_STAT_DESC_ARRAYS];
+ unsigned int addr_offset[SLIC_NUM_STAT_DESC_ARRAYS];
+ unsigned int active_array;
+ unsigned int len;
+ unsigned int done_idx;
+ size_t mem_size;
+};
+
+struct slic_tx_desc {
+ __le32 hnd;
+ __le32 rsvd;
+ u8 cmd;
+ u8 flags;
+ __le16 rsvd2;
+ __le32 totlen;
+ __le32 paddrl;
+ __le32 paddrh;
+ __le32 len;
+ __le32 type;
+};
+
+struct slic_tx_buffer {
+ struct sk_buff *skb;
+ DEFINE_DMA_UNMAP_ADDR(map_addr);
+ DEFINE_DMA_UNMAP_LEN(map_len);
+ struct slic_tx_desc *desc;
+ dma_addr_t desc_paddr;
+};
+
+struct slic_tx_queue {
+ struct dma_pool *dma_pool;
+ struct slic_tx_buffer *txbuffs;
+ unsigned int len;
+ unsigned int put_idx;
+ unsigned int done_idx;
+};
+
+struct slic_rx_desc {
+ u8 pad[16];
+ __le32 buffer;
+ __le32 length;
+ __le32 status;
+};
+
+struct slic_rx_buffer {
+ struct sk_buff *skb;
+ DEFINE_DMA_UNMAP_ADDR(map_addr);
+ DEFINE_DMA_UNMAP_LEN(map_len);
+ unsigned int addr_offset;
+};
+
+struct slic_rx_queue {
+ struct slic_rx_buffer *rxbuffs;
+ unsigned int len;
+ unsigned int done_idx;
+ unsigned int put_idx;
+};
+
+struct slic_device {
+ struct pci_dev *pdev;
+ struct net_device *netdev;
+ void __iomem *regs;
+ /* upper address setting lock */
+ spinlock_t upper_lock;
+ struct slic_shmem shmem;
+ struct napi_struct napi;
+ struct slic_rx_queue rxq;
+ struct slic_tx_queue txq;
+ struct slic_stat_queue stq;
+ struct slic_stats stats;
+ struct slic_upr_list upr_list;
+ /* link configuration lock */
+ spinlock_t link_lock;
+ bool promisc;
+ bool autoneg;
+ int speed;
+ int duplex;
+ bool is_fiber;
+ unsigned char model;
+};
+
+static inline u32 slic_read(struct slic_device *sdev, unsigned int reg)
+{
+ return ioread32(sdev->regs + reg);
+}
+
+static inline void slic_write(struct slic_device *sdev, unsigned int reg,
+ u32 val)
+{
+ iowrite32(val, sdev->regs + reg);
+}
+
+static inline void slic_flush_write(struct slic_device *sdev)
+{
+ (void)ioread32(sdev->regs + SLIC_REG_HOSTID);
+}
+
+#endif /* _SLIC_H */
diff --git a/drivers/net/ethernet/alacritech/slicoss.c b/drivers/net/ethernet/alacritech/slicoss.c
new file mode 100644
index 0000000..8cd862a
--- /dev/null
+++ b/drivers/net/ethernet/alacritech/slicoss.c
@@ -0,0 +1,1867 @@
+/*
+ * Driver for Gigabit Ethernet adapters based on the Session Layer
+ * Interface (SLIC) technology by Alacritech. The driver does not
+ * support the hardware acceleration features provided by these cards.
+ *
+ * Copyright (C) 2016 Lino Sanfilippo <LinoSanfilippo@gmx.de>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/if_ether.h>
+#include <linux/crc32.h>
+#include <linux/dma-mapping.h>
+#include <linux/ethtool.h>
+#include <linux/mii.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/firmware.h>
+#include <linux/list.h>
+#include <linux/u64_stats_sync.h>
+
+#include "slic.h"
+
+#define DRV_NAME "slicoss"
+#define DRV_VERSION "1.0"
+
+static const struct pci_device_id slic_id_tbl[] = {
+ { PCI_DEVICE(PCI_VENDOR_ID_ALACRITECH,
+ PCI_DEVICE_ID_ALACRITECH_MOAVE) },
+ { PCI_DEVICE(PCI_VENDOR_ID_ALACRITECH,
+ PCI_DEVICE_ID_ALACRITECH_OASIS) },
+ { 0 }
+};
+
+static const char slic_stats_strings[][ETH_GSTRING_LEN] = {
+ "rx_packets",
+ "rx_bytes",
+ "rx_multicasts",
+ "rx_errors",
+ "rx_buff_miss",
+ "rx_tp_csum",
+ "rx_tp_oflow",
+ "rx_tp_hlen",
+ "rx_ip_csum",
+ "rx_ip_len",
+ "rx_ip_hdr_len",
+ "rx_early",
+ "rx_buff_oflow",
+ "rx_lcode",
+ "rx_drbl",
+ "rx_crc",
+ "rx_oflow_802",
+ "rx_uflow_802",
+ "tx_packets",
+ "tx_bytes",
+ "tx_carrier",
+ "tx_dropped",
+ "irq_errs",
+};
+
+static inline int slic_next_queue_idx(unsigned int idx, unsigned int qlen)
+{
+ return (idx + 1) & (qlen - 1);
+}
+
+static inline int slic_get_free_queue_descs(unsigned int put_idx,
+ unsigned int done_idx,
+ unsigned int qlen)
+{
+ if (put_idx >= done_idx)
+ return (qlen - (put_idx - done_idx) - 1);
+ return (done_idx - put_idx - 1);
+}
+
+static unsigned int slic_next_compl_idx(struct slic_device *sdev)
+{
+ struct slic_stat_queue *stq = &sdev->stq;
+ unsigned int active = stq->active_array;
+ struct slic_stat_desc *descs;
+ struct slic_stat_desc *stat;
+ unsigned int idx;
+
+ descs = stq->descs[active];
+ stat = &descs[stq->done_idx];
+
+ if (!stat->status)
+ return SLIC_INVALID_STAT_DESC_IDX;
+
+ idx = (le32_to_cpu(stat->hnd) & 0xffff) - 1;
+ /* reset desc */
+ stat->hnd = 0;
+ stat->status = 0;
+
+ stq->done_idx = slic_next_queue_idx(stq->done_idx, stq->len);
+ /* check for wraparound */
+ if (!stq->done_idx) {
+ dma_addr_t paddr = stq->paddr[active];
+
+ slic_write(sdev, SLIC_REG_RBAR, lower_32_bits(paddr) |
+ stq->len);
+ /* make sure new status descriptors are immediately available */
+ slic_flush_write(sdev);
+ active++;
+ active &= (SLIC_NUM_STAT_DESC_ARRAYS - 1);
+ stq->active_array = active;
+ }
+ return idx;
+}
+
+static unsigned int slic_get_free_tx_descs(struct slic_tx_queue *txq)
+{
+ /* ensure tail idx is updated */
+ smp_mb();
+ return slic_get_free_queue_descs(txq->put_idx, txq->done_idx, txq->len);
+}
+
+static unsigned int slic_get_free_rx_descs(struct slic_rx_queue *rxq)
+{
+ return slic_get_free_queue_descs(rxq->put_idx, rxq->done_idx, rxq->len);
+}
+
+static void slic_clear_upr_list(struct slic_upr_list *upr_list)
+{
+ struct slic_upr *upr;
+ struct slic_upr *tmp;
+
+ spin_lock_bh(&upr_list->lock);
+ list_for_each_entry_safe(upr, tmp, &upr_list->list, list) {
+ list_del(&upr->list);
+ kfree(upr);
+ }
+ upr_list->pending = false;
+ spin_unlock_bh(&upr_list->lock);
+}
+
+static void slic_start_upr(struct slic_device *sdev, struct slic_upr *upr)
+{
+ u32 reg;
+
+ reg = (upr->type == SLIC_UPR_CONFIG) ? SLIC_REG_RCONFIG :
+ SLIC_REG_LSTAT;
+ slic_write(sdev, reg, lower_32_bits(upr->paddr));
+ slic_flush_write(sdev);
+}
+
+static void slic_queue_upr(struct slic_device *sdev, struct slic_upr *upr)
+{
+ struct slic_upr_list *upr_list = &sdev->upr_list;
+ bool pending;
+
+ spin_lock_bh(&upr_list->lock);
+ pending = upr_list->pending;
+ INIT_LIST_HEAD(&upr->list);
+ list_add_tail(&upr->list, &upr_list->list);
+ upr_list->pending = true;
+ spin_unlock_bh(&upr_list->lock);
+
+ if (!pending)
+ slic_start_upr(sdev, upr);
+}
+
+static struct slic_upr *slic_dequeue_upr(struct slic_device *sdev)
+{
+ struct slic_upr_list *upr_list = &sdev->upr_list;
+ struct slic_upr *next_upr = NULL;
+ struct slic_upr *upr = NULL;
+
+ spin_lock_bh(&upr_list->lock);
+ if (!list_empty(&upr_list->list)) {
+ upr = list_first_entry(&upr_list->list, struct slic_upr, list);
+ list_del(&upr->list);
+
+ if (list_empty(&upr_list->list))
+ upr_list->pending = false;
+ else
+ next_upr = list_first_entry(&upr_list->list,
+ struct slic_upr, list);
+ }
+ spin_unlock_bh(&upr_list->lock);
+ /* trigger processing of the next upr in list */
+ if (next_upr)
+ slic_start_upr(sdev, next_upr);
+
+ return upr;
+}
+
+static int slic_new_upr(struct slic_device *sdev, unsigned int type,
+ dma_addr_t paddr)
+{
+ struct slic_upr *upr;
+
+ upr = kmalloc(sizeof(*upr), GFP_ATOMIC);
+ if (!upr)
+ return -ENOMEM;
+ upr->type = type;
+ upr->paddr = paddr;
+
+ slic_queue_upr(sdev, upr);
+
+ return 0;
+}
+
+static void slic_set_mcast_bit(u64 *mcmask, unsigned char const *addr)
+{
+ u64 mask = *mcmask;
+ u8 crc;
+ /* Get the CRC polynomial for the mac address: we use bits 1-8 (lsb),
+ * bitwise reversed, msb (= lsb bit 0 before bitrev) is automatically
+ * discarded.
+ */
+ crc = ether_crc(ETH_ALEN, addr) >> 23;
+ /* we only have space on the SLIC for 64 entries */
+ crc &= 0x3F;
+ mask |= (u64)1 << crc;
+ *mcmask = mask;
+}
+
+/* must be called with link_lock held */
+static void slic_configure_rcv(struct slic_device *sdev)
+{
+ u32 val;
+
+ val = SLIC_GRCR_RESET | SLIC_GRCR_ADDRAEN | SLIC_GRCR_RCVEN |
+ SLIC_GRCR_HASHSIZE << SLIC_GRCR_HASHSIZE_SHIFT | SLIC_GRCR_RCVBAD;
+
+ if (sdev->duplex == DUPLEX_FULL)
+ val |= SLIC_GRCR_CTLEN;
+
+ if (sdev->promisc)
+ val |= SLIC_GRCR_RCVALL;
+
+ slic_write(sdev, SLIC_REG_WRCFG, val);
+}
+
+/* must be called with link_lock held */
+static void slic_configure_xmt(struct slic_device *sdev)
+{
+ u32 val;
+
+ val = SLIC_GXCR_RESET | SLIC_GXCR_XMTEN;
+
+ if (sdev->duplex == DUPLEX_FULL)
+ val |= SLIC_GXCR_PAUSEEN;
+
+ slic_write(sdev, SLIC_REG_WXCFG, val);
+}
+
+/* must be called with link_lock held */
+static void slic_configure_mac(struct slic_device *sdev)
+{
+ u32 val;
+
+ if (sdev->speed == SPEED_1000) {
+ val = SLIC_GMCR_GAPBB_1000 << SLIC_GMCR_GAPBB_SHIFT |
+ SLIC_GMCR_GAPR1_1000 << SLIC_GMCR_GAPR1_SHIFT |
+ SLIC_GMCR_GAPR2_1000 << SLIC_GMCR_GAPR2_SHIFT |
+ SLIC_GMCR_GBIT; /* enable GMII */
+ } else {
+ val = SLIC_GMCR_GAPBB_100 << SLIC_GMCR_GAPBB_SHIFT |
+ SLIC_GMCR_GAPR1_100 << SLIC_GMCR_GAPR1_SHIFT |
+ SLIC_GMCR_GAPR2_100 << SLIC_GMCR_GAPR2_SHIFT;
+ }
+
+ if (sdev->duplex == DUPLEX_FULL)
+ val |= SLIC_GMCR_FULLD;
+
+ slic_write(sdev, SLIC_REG_WMCFG, val);
+}
+
+static void slic_configure_link_locked(struct slic_device *sdev,
+ unsigned int speed, unsigned int duplex)
+{
+ struct net_device *dev = sdev->netdev;
+
+ if (sdev->speed == speed && sdev->duplex == duplex)
+ return;
+
+ sdev->speed = speed;
+ sdev->duplex = duplex;
+
+ if (sdev->speed == SPEED_UNKNOWN) {
+ if (netif_carrier_ok(dev))
+ netif_carrier_off(dev);
+ } else {
+ /* (re)configure link settings */
+ slic_configure_mac(sdev);
+ slic_configure_xmt(sdev);
+ slic_configure_rcv(sdev);
+ slic_flush_write(sdev);
+
+ if (!netif_carrier_ok(dev))
+ netif_carrier_on(dev);
+ }
+}
+
+static void slic_configure_link(struct slic_device *sdev, unsigned int speed,
+ unsigned int duplex)
+{
+ spin_lock_bh(&sdev->link_lock);
+ slic_configure_link_locked(sdev, speed, duplex);
+ spin_unlock_bh(&sdev->link_lock);
+}
+
+static void slic_set_rx_mode(struct net_device *dev)
+{
+ struct slic_device *sdev = netdev_priv(dev);
+ struct netdev_hw_addr *hwaddr;
+ bool set_promisc;
+ u64 mcmask;
+
+ if (dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) {
+ /* Turn on all multicast addresses. We have to do this for
+ * promiscuous mode as well as ALLMCAST mode (it saves the
+ * microcode from having to keep state about the MAC
+ * configuration).
+ */
+ mcmask = ~(u64)0;
+ } else {
+ mcmask = 0;
+
+ netdev_for_each_mc_addr(hwaddr, dev) {
+ slic_set_mcast_bit(&mcmask, hwaddr->addr);
+ }
+ }
+
+ slic_write(sdev, SLIC_REG_MCASTLOW, lower_32_bits(mcmask));
+ slic_write(sdev, SLIC_REG_MCASTHIGH, upper_32_bits(mcmask));
+
+ set_promisc = !!(dev->flags & IFF_PROMISC);
+
+ spin_lock_bh(&sdev->link_lock);
+ if (sdev->promisc != set_promisc) {
+ sdev->promisc = set_promisc;
+ slic_configure_rcv(sdev);
+ /* make sure writes to receiver cant leak out of the lock */
+ mmiowb();
+ }
+ spin_unlock_bh(&sdev->link_lock);
+}
+
+static void slic_xmit_complete(struct slic_device *sdev)
+{
+ struct slic_tx_queue *txq = &sdev->txq;
+ struct net_device *dev = sdev->netdev;
+ unsigned int idx = txq->done_idx;
+ struct slic_tx_buffer *buff;
+ unsigned int frames = 0;
+ unsigned int bytes = 0;
+
+ do {
+ idx = slic_next_compl_idx(sdev);
+ if (idx == SLIC_INVALID_STAT_DESC_IDX)
+ break;
+
+ txq->done_idx = idx;
+ buff = &txq->txbuffs[idx];
+
+ if (unlikely(!buff->skb)) {
+ netdev_warn(dev,
+ "no skb found for desc idx %i\n", idx);
+ continue;
+ }
+ dma_unmap_single(&sdev->pdev->dev,
+ dma_unmap_addr(buff, map_addr),
+ dma_unmap_len(buff, map_len), DMA_TO_DEVICE);
+
+ bytes += buff->skb->len;
+ frames++;
+
+ dev_kfree_skb_any(buff->skb);
+ buff->skb = NULL;
+ } while (frames < SLIC_MAX_TX_COMPLETIONS);
+ /* make sure xmit sees the new value for done_idx */
+ smp_wmb();
+
+ u64_stats_update_begin(&sdev->stats.syncp);
+ sdev->stats.tx_bytes += bytes;
+ sdev->stats.tx_packets += frames;
+ u64_stats_update_end(&sdev->stats.syncp);
+
+ netif_tx_lock(dev);
+ if (netif_queue_stopped(dev) &&
+ (slic_get_free_tx_descs(txq) >= SLIC_MIN_TX_WAKEUP_DESCS))
+ netif_wake_queue(dev);
+ netif_tx_unlock(dev);
+}
+
+static void slic_refill_rx_queue(struct slic_device *sdev, gfp_t gfp)
+{
+ const unsigned int ALIGN_MASK = SLIC_RX_BUFF_ALIGN - 1;
+ unsigned int maplen = SLIC_RX_BUFF_SIZE;
+ struct slic_rx_queue *rxq = &sdev->rxq;
+ struct net_device *dev = sdev->netdev;
+ struct slic_rx_buffer *buff;
+ struct slic_rx_desc *desc;
+ unsigned int misalign;
+ unsigned int offset;
+ struct sk_buff *skb;
+ dma_addr_t paddr;
+
+ while (slic_get_free_rx_descs(rxq) > SLIC_MAX_REQ_RX_DESCS) {
+ skb = alloc_skb(maplen + ALIGN_MASK, gfp);
+ if (!skb)
+ break;
+
+ paddr = dma_map_single(&sdev->pdev->dev, skb->data, maplen,
+ DMA_FROM_DEVICE);
+ if (dma_mapping_error(&sdev->pdev->dev, paddr)) {
+ netdev_err(dev, "mapping rx packet failed\n");
+ /* drop skb */
+ dev_kfree_skb_any(skb);
+ break;
+ }
+ /* ensure head buffer descriptors are 256 byte aligned */
+ offset = 0;
+ misalign = paddr & ALIGN_MASK;
+ if (misalign) {
+ offset = SLIC_RX_BUFF_ALIGN - misalign;
+ skb_reserve(skb, offset);
+ }
+ /* the HW expects dma chunks for descriptor + frame data */
+ desc = (struct slic_rx_desc *)skb->data;
+ memset(desc, 0, sizeof(*desc));
+
+ buff = &rxq->rxbuffs[rxq->put_idx];
+ buff->skb = skb;
+ dma_unmap_addr_set(buff, map_addr, paddr);
+ dma_unmap_len_set(buff, map_len, maplen);
+ buff->addr_offset = offset;
+ /* head buffer descriptors are placed immediately before skb */
+ slic_write(sdev, SLIC_REG_HBAR, lower_32_bits(paddr) +
+ offset);
+ rxq->put_idx = slic_next_queue_idx(rxq->put_idx, rxq->len);
+ }
+}
+
+static void slic_handle_frame_error(struct slic_device *sdev,
+ struct sk_buff *skb)
+{
+ struct slic_stats *stats = &sdev->stats;
+
+ if (sdev->model == SLIC_MODEL_OASIS) {
+ struct slic_rx_info_oasis *info;
+ u32 status_b;
+ u32 status;
+
+ info = (struct slic_rx_info_oasis *)skb->data;
+ status = le32_to_cpu(info->frame_status);
+ status_b = le32_to_cpu(info->frame_status_b);
+ /* transport layer */
+ if (status_b & SLIC_VRHSTATB_TPCSUM)
+ SLIC_INC_STATS_COUNTER(stats, rx_tpcsum);
+ if (status & SLIC_VRHSTAT_TPOFLO)
+ SLIC_INC_STATS_COUNTER(stats, rx_tpoflow);
+ if (status_b & SLIC_VRHSTATB_TPHLEN)
+ SLIC_INC_STATS_COUNTER(stats, rx_tphlen);
+ /* ip layer */
+ if (status_b & SLIC_VRHSTATB_IPCSUM)
+ SLIC_INC_STATS_COUNTER(stats, rx_ipcsum);
+ if (status_b & SLIC_VRHSTATB_IPLERR)
+ SLIC_INC_STATS_COUNTER(stats, rx_iplen);
+ if (status_b & SLIC_VRHSTATB_IPHERR)
+ SLIC_INC_STATS_COUNTER(stats, rx_iphlen);
+ /* link layer */
+ if (status_b & SLIC_VRHSTATB_RCVE)
+ SLIC_INC_STATS_COUNTER(stats, rx_early);
+ if (status_b & SLIC_VRHSTATB_BUFF)
+ SLIC_INC_STATS_COUNTER(stats, rx_buffoflow);
+ if (status_b & SLIC_VRHSTATB_CODE)
+ SLIC_INC_STATS_COUNTER(stats, rx_lcode);
+ if (status_b & SLIC_VRHSTATB_DRBL)
+ SLIC_INC_STATS_COUNTER(stats, rx_drbl);
+ if (status_b & SLIC_VRHSTATB_CRC)
+ SLIC_INC_STATS_COUNTER(stats, rx_crc);
+ if (status & SLIC_VRHSTAT_802OE)
+ SLIC_INC_STATS_COUNTER(stats, rx_oflow802);
+ if (status_b & SLIC_VRHSTATB_802UE)
+ SLIC_INC_STATS_COUNTER(stats, rx_uflow802);
+ if (status_b & SLIC_VRHSTATB_CARRE)
+ SLIC_INC_STATS_COUNTER(stats, tx_carrier);
+ } else { /* mojave */
+ struct slic_rx_info_mojave *info;
+ u32 status;
+
+ info = (struct slic_rx_info_mojave *)skb->data;
+ status = le32_to_cpu(info->frame_status);
+ /* transport layer */
+ if (status & SLIC_VGBSTAT_XPERR) {
+ u32 xerr = status >> SLIC_VGBSTAT_XERRSHFT;
+
+ if (xerr == SLIC_VGBSTAT_XCSERR)
+ SLIC_INC_STATS_COUNTER(stats, rx_tpcsum);
+ if (xerr == SLIC_VGBSTAT_XUFLOW)
+ SLIC_INC_STATS_COUNTER(stats, rx_tpoflow);
+ if (xerr == SLIC_VGBSTAT_XHLEN)
+ SLIC_INC_STATS_COUNTER(stats, rx_tphlen);
+ }
+ /* ip layer */
+ if (status & SLIC_VGBSTAT_NETERR) {
+ u32 nerr = status >> SLIC_VGBSTAT_NERRSHFT &
+ SLIC_VGBSTAT_NERRMSK;
+
+ if (nerr == SLIC_VGBSTAT_NCSERR)
+ SLIC_INC_STATS_COUNTER(stats, rx_ipcsum);
+ if (nerr == SLIC_VGBSTAT_NUFLOW)
+ SLIC_INC_STATS_COUNTER(stats, rx_iplen);
+ if (nerr == SLIC_VGBSTAT_NHLEN)
+ SLIC_INC_STATS_COUNTER(stats, rx_iphlen);
+ }
+ /* link layer */
+ if (status & SLIC_VGBSTAT_LNKERR) {
+ u32 lerr = status & SLIC_VGBSTAT_LERRMSK;
+
+ if (lerr == SLIC_VGBSTAT_LDEARLY)
+ SLIC_INC_STATS_COUNTER(stats, rx_early);
+ if (lerr == SLIC_VGBSTAT_LBOFLO)
+ SLIC_INC_STATS_COUNTER(stats, rx_buffoflow);
+ if (lerr == SLIC_VGBSTAT_LCODERR)
+ SLIC_INC_STATS_COUNTER(stats, rx_lcode);
+ if (lerr == SLIC_VGBSTAT_LDBLNBL)
+ SLIC_INC_STATS_COUNTER(stats, rx_drbl);
+ if (lerr == SLIC_VGBSTAT_LCRCERR)
+ SLIC_INC_STATS_COUNTER(stats, rx_crc);
+ if (lerr == SLIC_VGBSTAT_LOFLO)
+ SLIC_INC_STATS_COUNTER(stats, rx_oflow802);
+ if (lerr == SLIC_VGBSTAT_LUFLO)
+ SLIC_INC_STATS_COUNTER(stats, rx_uflow802);
+ }
+ }
+ SLIC_INC_STATS_COUNTER(stats, rx_errors);
+}
+
+static void slic_handle_receive(struct slic_device *sdev, unsigned int todo,
+ unsigned int *done)
+{
+ struct slic_rx_queue *rxq = &sdev->rxq;
+ struct net_device *dev = sdev->netdev;
+ struct slic_rx_buffer *buff;
+ struct slic_rx_desc *desc;
+ unsigned int frames = 0;
+ unsigned int bytes = 0;
+ struct sk_buff *skb;
+ u32 status;
+ u32 len;
+
+ while (todo && (rxq->done_idx != rxq->put_idx)) {
+ buff = &rxq->rxbuffs[rxq->done_idx];
+
+ skb = buff->skb;
+ if (!skb)
+ break;
+
+ desc = (struct slic_rx_desc *)skb->data;
+
+ dma_sync_single_for_cpu(&sdev->pdev->dev,
+ dma_unmap_addr(buff, map_addr),
+ buff->addr_offset + sizeof(*desc),
+ DMA_FROM_DEVICE);
+
+ status = le32_to_cpu(desc->status);
+ if (!(status & SLIC_IRHDDR_SVALID))
+ break;
+
+ buff->skb = NULL;
+
+ dma_unmap_single(&sdev->pdev->dev,
+ dma_unmap_addr(buff, map_addr),
+ dma_unmap_len(buff, map_len),
+ DMA_FROM_DEVICE);
+
+ /* skip rx descriptor that is placed before the frame data */
+ skb_reserve(skb, SLIC_RX_BUFF_HDR_SIZE);
+
+ if (unlikely(status & SLIC_IRHDDR_ERR)) {
+ slic_handle_frame_error(sdev, skb);
+ dev_kfree_skb_any(skb);
+ } else {
+ struct ethhdr *eh = (struct ethhdr *)skb->data;
+
+ if (is_multicast_ether_addr(eh->h_dest))
+ SLIC_INC_STATS_COUNTER(&sdev->stats, rx_mcasts);
+
+ len = le32_to_cpu(desc->length) & SLIC_IRHDDR_FLEN_MSK;
+ skb_put(skb, len);
+ skb->protocol = eth_type_trans(skb, dev);
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
+ skb->dev = dev;
+
+ napi_gro_receive(&sdev->napi, skb);
+
+ bytes += len;
+ frames++;
+ }
+ rxq->done_idx = slic_next_queue_idx(rxq->done_idx, rxq->len);
+ todo--;
+ }
+
+ u64_stats_update_begin(&sdev->stats.syncp);
+ sdev->stats.rx_bytes += bytes;
+ sdev->stats.rx_packets += frames;
+ u64_stats_update_end(&sdev->stats.syncp);
+
+ slic_refill_rx_queue(sdev, GFP_ATOMIC);
+}
+
+static void slic_handle_link_irq(struct slic_device *sdev)
+{
+ struct slic_shmem *sm = &sdev->shmem;
+ struct slic_shmem_data *sm_data = sm->shmem_data;
+ unsigned int duplex;
+ unsigned int speed;
+ u32 link;
+
+ link = le32_to_cpu(sm_data->link);
+
+ if (link & SLIC_GIG_LINKUP) {
+ if (link & SLIC_GIG_SPEED_1000)
+ speed = SPEED_1000;
+ else if (link & SLIC_GIG_SPEED_100)
+ speed = SPEED_100;
+ else
+ speed = SPEED_10;
+
+ duplex = (link & SLIC_GIG_FULLDUPLEX) ? DUPLEX_FULL :
+ DUPLEX_HALF;
+ } else {
+ duplex = DUPLEX_UNKNOWN;
+ speed = SPEED_UNKNOWN;
+ }
+ slic_configure_link(sdev, speed, duplex);
+}
+
+static void slic_handle_upr_irq(struct slic_device *sdev, u32 irqs)
+{
+ struct slic_upr *upr;
+
+ /* remove upr that caused this irq (always the first entry in list) */
+ upr = slic_dequeue_upr(sdev);
+ if (!upr) {
+ netdev_warn(sdev->netdev, "no upr found on list\n");
+ return;
+ }
+
+ if (upr->type == SLIC_UPR_LSTAT) {
+ if (unlikely(irqs & SLIC_ISR_UPCERR_MASK)) {
+ /* try again */
+ slic_queue_upr(sdev, upr);
+ return;
+ }
+ slic_handle_link_irq(sdev);
+ }
+ kfree(upr);
+}
+
+static int slic_handle_link_change(struct slic_device *sdev)
+{
+ return slic_new_upr(sdev, SLIC_UPR_LSTAT, sdev->shmem.link_paddr);
+}
+
+static void slic_handle_err_irq(struct slic_device *sdev, u32 isr)
+{
+ struct slic_stats *stats = &sdev->stats;
+
+ if (isr & SLIC_ISR_RMISS)
+ SLIC_INC_STATS_COUNTER(stats, rx_buff_miss);
+ if (isr & SLIC_ISR_XDROP)
+ SLIC_INC_STATS_COUNTER(stats, tx_dropped);
+ if (!(isr & (SLIC_ISR_RMISS | SLIC_ISR_XDROP)))
+ SLIC_INC_STATS_COUNTER(stats, irq_errs);
+}
+
+static void slic_handle_irq(struct slic_device *sdev, u32 isr,
+ unsigned int todo, unsigned int *done)
+{
+ if (isr & SLIC_ISR_ERR)
+ slic_handle_err_irq(sdev, isr);
+
+ if (isr & SLIC_ISR_LEVENT)
+ slic_handle_link_change(sdev);
+
+ if (isr & SLIC_ISR_UPC_MASK)
+ slic_handle_upr_irq(sdev, isr);
+
+ if (isr & SLIC_ISR_RCV)
+ slic_handle_receive(sdev, todo, done);
+
+ if (isr & SLIC_ISR_CMD)
+ slic_xmit_complete(sdev);
+}
+
+static int slic_poll(struct napi_struct *napi, int todo)
+{
+ struct slic_device *sdev = container_of(napi, struct slic_device, napi);
+ struct slic_shmem *sm = &sdev->shmem;
+ struct slic_shmem_data *sm_data = sm->shmem_data;
+ u32 isr = le32_to_cpu(sm_data->isr);
+ unsigned int done = 0;
+
+ slic_handle_irq(sdev, isr, todo, &done);
+
+ if (done < todo) {
+ napi_complete(napi);
+ /* reenable irqs */
+ sm_data->isr = 0;
+ /* make sure sm_data->isr is cleard before irqs are reenabled */
+ wmb();
+ slic_write(sdev, SLIC_REG_ISR, 0);
+ slic_flush_write(sdev);
+ }
+
+ return done;
+}
+
+static irqreturn_t slic_irq(int irq, void *dev_id)
+{
+ struct slic_device *sdev = dev_id;
+ struct slic_shmem *sm = &sdev->shmem;
+ struct slic_shmem_data *sm_data = sm->shmem_data;
+
+ slic_write(sdev, SLIC_REG_ICR, SLIC_ICR_INT_MASK);
+ slic_flush_write(sdev);
+ /* make sure sm_data->isr is read after ICR_INT_MASK is set */
+ wmb();
+
+ if (!sm_data->isr) {
+ dma_rmb();
+ /* spurious interrupt */
+ slic_write(sdev, SLIC_REG_ISR, 0);
+ slic_flush_write(sdev);
+ return IRQ_NONE;
+ }
+
+ napi_schedule(&sdev->napi);
+
+ return IRQ_HANDLED;
+}
+
+static void slic_card_reset(struct slic_device *sdev)
+{
+ u16 cmd;
+
+ slic_write(sdev, SLIC_REG_RESET, SLIC_RESET_MAGIC);
+ /* flush write by means of config space */
+ pci_read_config_word(sdev->pdev, PCI_COMMAND, &cmd);
+ mdelay(1);
+}
+
+static int slic_init_stat_queue(struct slic_device *sdev)
+{
+ const unsigned int DESC_ALIGN_MASK = SLIC_STATS_DESC_ALIGN - 1;
+ struct slic_stat_queue *stq = &sdev->stq;
+ struct slic_stat_desc *descs;
+ unsigned int misalign;
+ unsigned int offset;
+ dma_addr_t paddr;
+ size_t size;
+ int err;
+ int i;
+
+ stq->len = SLIC_NUM_STAT_DESCS;
+ stq->active_array = 0;
+ stq->done_idx = 0;
+
+ size = stq->len * sizeof(*descs) + DESC_ALIGN_MASK;
+
+ for (i = 0; i < SLIC_NUM_STAT_DESC_ARRAYS; i++) {
+ descs = dma_zalloc_coherent(&sdev->pdev->dev, size, &paddr,
+ GFP_KERNEL);
+ if (!descs) {
+ netdev_err(sdev->netdev,
+ "failed to allocate status descriptors\n");
+ err = -ENOMEM;
+ goto free_descs;
+ }
+ /* ensure correct alignment */
+ offset = 0;
+ misalign = paddr & DESC_ALIGN_MASK;
+ if (misalign) {
+ offset = SLIC_STATS_DESC_ALIGN - misalign;
+ descs += offset;
+ paddr += offset;
+ }
+
+ slic_write(sdev, SLIC_REG_RBAR, lower_32_bits(paddr) |
+ stq->len);
+ stq->descs[i] = descs;
+ stq->paddr[i] = paddr;
+ stq->addr_offset[i] = offset;
+ }
+
+ stq->mem_size = size;
+
+ return 0;
+
+free_descs:
+ while (i--) {
+ dma_free_coherent(&sdev->pdev->dev, stq->mem_size,
+ stq->descs[i] - stq->addr_offset[i],
+ stq->paddr[i] - stq->addr_offset[i]);
+ }
+
+ return err;
+}
+
+static void slic_free_stat_queue(struct slic_device *sdev)
+{
+ struct slic_stat_queue *stq = &sdev->stq;
+ int i;
+
+ for (i = 0; i < SLIC_NUM_STAT_DESC_ARRAYS; i++) {
+ dma_free_coherent(&sdev->pdev->dev, stq->mem_size,
+ stq->descs[i] - stq->addr_offset[i],
+ stq->paddr[i] - stq->addr_offset[i]);
+ }
+}
+
+static int slic_init_tx_queue(struct slic_device *sdev)
+{
+ struct slic_tx_queue *txq = &sdev->txq;
+ struct slic_tx_buffer *buff;
+ struct slic_tx_desc *desc;
+ int err;
+ int i;
+
+ txq->len = SLIC_NUM_TX_DESCS;
+ txq->put_idx = 0;
+ txq->done_idx = 0;
+
+ txq->txbuffs = kcalloc(txq->len, sizeof(*buff), GFP_KERNEL);
+ if (!txq->txbuffs)
+ return -ENOMEM;
+
+ txq->dma_pool = dma_pool_create("slic_pool", &sdev->pdev->dev,
+ sizeof(*desc), SLIC_TX_DESC_ALIGN,
+ 4096);
+ if (!txq->dma_pool) {
+ err = -ENOMEM;
+ netdev_err(sdev->netdev, "failed to create dma pool\n");
+ goto free_buffs;
+ }
+
+ for (i = 0; i < txq->len; i++) {
+ buff = &txq->txbuffs[i];
+ desc = dma_pool_zalloc(txq->dma_pool, GFP_KERNEL,
+ &buff->desc_paddr);
+ if (!desc) {
+ netdev_err(sdev->netdev,
+ "failed to alloc pool chunk (%i)\n", i);
+ err = -ENOMEM;
+ goto free_descs;
+ }
+
+ desc->hnd = cpu_to_le32((u32)(i + 1));
+ desc->cmd = SLIC_CMD_XMT_REQ;
+ desc->flags = 0;
+ desc->type = cpu_to_le32(SLIC_CMD_TYPE_DUMB);
+ buff->desc = desc;
+ }
+
+ return 0;
+
+free_descs:
+ while (i--) {
+ buff = &txq->txbuffs[i];
+ dma_pool_free(txq->dma_pool, buff->desc, buff->desc_paddr);
+ }
+ dma_pool_destroy(txq->dma_pool);
+
+free_buffs:
+ kfree(txq->txbuffs);
+
+ return err;
+}
+
+static void slic_free_tx_queue(struct slic_device *sdev)
+{
+ struct slic_tx_queue *txq = &sdev->txq;
+ struct slic_tx_buffer *buff;
+ int i;
+
+ for (i = 0; i < txq->len; i++) {
+ buff = &txq->txbuffs[i];
+ dma_pool_free(txq->dma_pool, buff->desc, buff->desc_paddr);
+ if (!buff->skb)
+ continue;
+
+ dma_unmap_single(&sdev->pdev->dev,
+ dma_unmap_addr(buff, map_addr),
+ dma_unmap_len(buff, map_len), DMA_TO_DEVICE);
+ consume_skb(buff->skb);
+ }
+ dma_pool_destroy(txq->dma_pool);
+
+ kfree(txq->txbuffs);
+}
+
+static int slic_init_rx_queue(struct slic_device *sdev)
+{
+ struct slic_rx_queue *rxq = &sdev->rxq;
+ struct slic_rx_buffer *buff;
+
+ rxq->len = SLIC_NUM_RX_LES;
+ rxq->done_idx = 0;
+ rxq->put_idx = 0;
+
+ buff = kcalloc(rxq->len, sizeof(*buff), GFP_KERNEL);
+ if (!buff)
+ return -ENOMEM;
+
+ rxq->rxbuffs = buff;
+ slic_refill_rx_queue(sdev, GFP_KERNEL);
+
+ return 0;
+}
+
+static void slic_free_rx_queue(struct slic_device *sdev)
+{
+ struct slic_rx_queue *rxq = &sdev->rxq;
+ struct slic_rx_buffer *buff;
+ int i;
+
+ /* free rx buffers */
+ for (i = 0; i < rxq->len; i++) {
+ buff = &rxq->rxbuffs[i];
+
+ if (!buff->skb)
+ continue;
+
+ dma_unmap_single(&sdev->pdev->dev,
+ dma_unmap_addr(buff, map_addr),
+ dma_unmap_len(buff, map_len),
+ DMA_FROM_DEVICE);
+ consume_skb(buff->skb);
+ }
+ kfree(rxq->rxbuffs);
+}
+
+static void slic_set_link_autoneg(struct slic_device *sdev)
+{
+ unsigned int subid = sdev->pdev->subsystem_device;
+ u32 val;
+
+ if (sdev->is_fiber) {
+ /* We've got a fiber gigabit interface, and register 4 is
+ * different in fiber mode than in copper mode.
+ */
+ /* advertise FD only @1000 Mb */
+ val = MII_ADVERTISE << 16 | ADVERTISE_1000XFULL |
+ ADVERTISE_1000XPAUSE | ADVERTISE_1000XPSE_ASYM;
+ /* enable PAUSE frames */
+ slic_write(sdev, SLIC_REG_WPHY, val);
+ /* reset phy, enable auto-neg */
+ val = MII_BMCR << 16 | BMCR_RESET | BMCR_ANENABLE |
+ BMCR_ANRESTART;
+ slic_write(sdev, SLIC_REG_WPHY, val);
+ } else { /* copper gigabit */
+ /* We've got a copper gigabit interface, and register 4 is
+ * different in copper mode than in fiber mode.
+ */
+ /* advertise 10/100 Mb modes */
+ val = MII_ADVERTISE << 16 | ADVERTISE_100FULL |
+ ADVERTISE_100HALF | ADVERTISE_10FULL | ADVERTISE_10HALF;
+ /* enable PAUSE frames */
+ val |= ADVERTISE_PAUSE_CAP | ADVERTISE_PAUSE_ASYM;
+ /* required by the Cicada PHY */
+ val |= ADVERTISE_CSMA;
+ slic_write(sdev, SLIC_REG_WPHY, val);
+
+ /* advertise FD only @1000 Mb */
+ val = MII_CTRL1000 << 16 | ADVERTISE_1000FULL;
+ slic_write(sdev, SLIC_REG_WPHY, val);
+
+ if (subid != PCI_SUBDEVICE_ID_ALACRITECH_CICADA) {
+ /* if a Marvell PHY enable auto crossover */
+ val = SLIC_MIICR_REG_16 | SLIC_MRV_REG16_XOVERON;
+ slic_write(sdev, SLIC_REG_WPHY, val);
+
+ /* reset phy, enable auto-neg */
+ val = MII_BMCR << 16 | BMCR_RESET | BMCR_ANENABLE |
+ BMCR_ANRESTART;
+ slic_write(sdev, SLIC_REG_WPHY, val);
+ } else {
+ /* enable and restart auto-neg (don't reset) */
+ val = MII_BMCR << 16 | BMCR_ANENABLE | BMCR_ANRESTART;
+ slic_write(sdev, SLIC_REG_WPHY, val);
+ }
+ }
+ sdev->autoneg = true;
+}
+
+static void slic_set_mac_address(struct slic_device *sdev)
+{
+ u8 *addr = sdev->netdev->dev_addr;
+ u32 val;
+
+ val = addr[5] | addr[4] << 8 | addr[3] << 16 | addr[2] << 24;
+
+ slic_write(sdev, SLIC_REG_WRADDRAL, val);
+ slic_write(sdev, SLIC_REG_WRADDRBL, val);
+
+ val = addr[0] << 8 | addr[1];
+
+ slic_write(sdev, SLIC_REG_WRADDRAH, val);
+ slic_write(sdev, SLIC_REG_WRADDRBH, val);
+ slic_flush_write(sdev);
+}
+
+static u32 slic_read_dword_from_firmware(const struct firmware *fw, int *offset)
+{
+ int idx = *offset;
+ __le32 val;
+
+ memcpy(&val, fw->data + *offset, sizeof(val));
+ idx += 4;
+ *offset = idx;
+
+ return le32_to_cpu(val);
+}
+
+MODULE_FIRMWARE(SLIC_RCV_FIRMWARE_MOAVE);
+MODULE_FIRMWARE(SLIC_RCV_FIRMWARE_OASIS);
+
+static int slic_load_rcvseq_firmware(struct slic_device *sdev)
+{
+ const struct firmware *fw;
+ const char *file;
+ u32 codelen;
+ int idx = 0;
+ u32 instr;
+ u32 addr;
+ int err;
+
+ file = (sdev->model == SLIC_MODEL_OASIS) ? SLIC_RCV_FIRMWARE_OASIS :
+ SLIC_RCV_FIRMWARE_MOAVE;
+ err = request_firmware(&fw, file, &sdev->pdev->dev);
+ if (err) {
+ dev_err(&sdev->pdev->dev,
+ "failed to load receive sequencer firmware %s\n", file);
+ return err;
+ }
+ /* Do an initial sanity check concerning firmware size now. A further
+ * check follows below.
+ */
+ if (fw->size < SLIC_FIRMWARE_MIN_SIZE) {
+ dev_err(&sdev->pdev->dev,
+ "invalid firmware size %zu (min %u expected)\n",
+ fw->size, SLIC_FIRMWARE_MIN_SIZE);
+ err = -EINVAL;
+ goto release;
+ }
+
+ codelen = slic_read_dword_from_firmware(fw, &idx);
+
+ /* do another sanity check against firmware size */
+ if ((codelen + 4) > fw->size) {
+ dev_err(&sdev->pdev->dev,
+ "invalid rcv-sequencer firmware size %zu\n", fw->size);
+ err = -EINVAL;
+ goto release;
+ }
+
+ /* download sequencer code to card */
+ slic_write(sdev, SLIC_REG_RCV_WCS, SLIC_RCVWCS_BEGIN);
+ for (addr = 0; addr < codelen; addr++) {
+ __le32 val;
+ /* write out instruction address */
+ slic_write(sdev, SLIC_REG_RCV_WCS, addr);
+
+ instr = slic_read_dword_from_firmware(fw, &idx);
+ /* write out the instruction data low addr */
+ slic_write(sdev, SLIC_REG_RCV_WCS, instr);
+
+ val = (__le32)fw->data[idx];
+ instr = le32_to_cpu(val);
+ idx++;
+ /* write out the instruction data high addr */
+ slic_write(sdev, SLIC_REG_RCV_WCS, instr);
+ }
+ /* finish download */
+ slic_write(sdev, SLIC_REG_RCV_WCS, SLIC_RCVWCS_FINISH);
+ slic_flush_write(sdev);
+release:
+ release_firmware(fw);
+
+ return 0;
+}
+
+MODULE_FIRMWARE(SLIC_FIRMWARE_MOAVE);
+MODULE_FIRMWARE(SLIC_FIRMWARE_OASIS);
+
+static int slic_load_firmware(struct slic_device *sdev)
+{
+ u32 sectstart[SLIC_FIRMWARE_MAX_SECTIONS];
+ u32 sectsize[SLIC_FIRMWARE_MAX_SECTIONS];
+ const struct firmware *fw;
+ unsigned int datalen;
+ const char *file;
+ int code_start;
+ u32 numsects;
+ int idx = 0;
+ u32 sect;
+ u32 instr;
+ u32 addr;
+ u32 base;
+ int err;
+ int i;
+
+ file = (sdev->model == SLIC_MODEL_OASIS) ? SLIC_FIRMWARE_OASIS :
+ SLIC_FIRMWARE_MOAVE;
+ err = request_firmware(&fw, file, &sdev->pdev->dev);
+ if (err) {
+ dev_err(&sdev->pdev->dev, "failed to load firmware %s\n", file);
+ return err;
+ }
+ /* Do an initial sanity check concerning firmware size now. A further
+ * check follows below.
+ */
+ if (fw->size < SLIC_FIRMWARE_MIN_SIZE) {
+ dev_err(&sdev->pdev->dev,
+ "invalid firmware size %zu (min is %u)\n", fw->size,
+ SLIC_FIRMWARE_MIN_SIZE);
+ err = -EINVAL;
+ goto release;
+ }
+
+ numsects = slic_read_dword_from_firmware(fw, &idx);
+ if (numsects == 0 || numsects > SLIC_FIRMWARE_MAX_SECTIONS) {
+ dev_err(&sdev->pdev->dev,
+ "invalid number of sections in firmware: %u", numsects);
+ err = -EINVAL;
+ goto release;
+ }
+
+ datalen = numsects * 8 + 4;
+ for (i = 0; i < numsects; i++) {
+ sectsize[i] = slic_read_dword_from_firmware(fw, &idx);
+ datalen += sectsize[i];
+ }
+
+ /* do another sanity check against firmware size */
+ if (datalen > fw->size) {
+ dev_err(&sdev->pdev->dev,
+ "invalid firmware size %zu (expected >= %u)\n",
+ fw->size, datalen);
+ err = -EINVAL;
+ goto release;
+ }
+ /* get sections */
+ for (i = 0; i < numsects; i++)
+ sectstart[i] = slic_read_dword_from_firmware(fw, &idx);
+
+ code_start = idx;
+ instr = slic_read_dword_from_firmware(fw, &idx);
+
+ for (sect = 0; sect < numsects; sect++) {
+ unsigned int ssize = sectsize[sect] >> 3;
+
+ base = sectstart[sect];
+
+ for (addr = 0; addr < ssize; addr++) {
+ /* write out instruction address */
+ slic_write(sdev, SLIC_REG_WCS, base + addr);
+ /* write out instruction to low addr */
+ slic_write(sdev, SLIC_REG_WCS, instr);
+ instr = slic_read_dword_from_firmware(fw, &idx);
+ /* write out instruction to high addr */
+ slic_write(sdev, SLIC_REG_WCS, instr);
+ instr = slic_read_dword_from_firmware(fw, &idx);
+ }
+ }
+
+ idx = code_start;
+
+ for (sect = 0; sect < numsects; sect++) {
+ unsigned int ssize = sectsize[sect] >> 3;
+
+ instr = slic_read_dword_from_firmware(fw, &idx);
+ base = sectstart[sect];
+ if (base < 0x8000)
+ continue;
+
+ for (addr = 0; addr < ssize; addr++) {
+ /* write out instruction address */
+ slic_write(sdev, SLIC_REG_WCS,
+ SLIC_WCS_COMPARE | (base + addr));
+ /* write out instruction to low addr */
+ slic_write(sdev, SLIC_REG_WCS, instr);
+ instr = slic_read_dword_from_firmware(fw, &idx);
+ /* write out instruction to high addr */
+ slic_write(sdev, SLIC_REG_WCS, instr);
+ instr = slic_read_dword_from_firmware(fw, &idx);
+ }
+ }
+ slic_flush_write(sdev);
+ mdelay(10);
+ /* everything OK, kick off the card */
+ slic_write(sdev, SLIC_REG_WCS, SLIC_WCS_START);
+ slic_flush_write(sdev);
+ /* wait long enough for ucode to init card and reach the mainloop */
+ mdelay(20);
+release:
+ release_firmware(fw);
+
+ return err;
+}
+
+static int slic_init_shmem(struct slic_device *sdev)
+{
+ struct slic_shmem *sm = &sdev->shmem;
+ struct slic_shmem_data *sm_data;
+ dma_addr_t paddr;
+
+ sm_data = dma_zalloc_coherent(&sdev->pdev->dev, sizeof(*sm_data),
+ &paddr, GFP_KERNEL);
+ if (!sm_data) {
+ dev_err(&sdev->pdev->dev, "failed to allocate shared memory\n");
+ return -ENOMEM;
+ }
+
+ sm->shmem_data = sm_data;
+ sm->isr_paddr = paddr;
+ sm->link_paddr = paddr + offsetof(struct slic_shmem_data, link);
+
+ return 0;
+}
+
+static void slic_free_shmem(struct slic_device *sdev)
+{
+ struct slic_shmem *sm = &sdev->shmem;
+ struct slic_shmem_data *sm_data = sm->shmem_data;
+
+ dma_free_coherent(&sdev->pdev->dev, sizeof(*sm_data), sm_data,
+ sm->isr_paddr);
+}
+
+static int slic_init_iface(struct slic_device *sdev)
+{
+ struct slic_shmem *sm = &sdev->shmem;
+ int err;
+
+ sdev->upr_list.pending = false;
+
+ err = slic_init_shmem(sdev);
+ if (err) {
+ netdev_err(sdev->netdev, "failed to load firmware\n");
+ return err;
+ }
+
+ err = slic_load_firmware(sdev);
+ if (err) {
+ netdev_err(sdev->netdev, "failed to load firmware\n");
+ goto free_sm;
+ }
+
+ err = slic_load_rcvseq_firmware(sdev);
+ if (err) {
+ netdev_err(sdev->netdev,
+ "failed to load firmware for receive sequencer\n");
+ goto free_sm;
+ }
+
+ slic_write(sdev, SLIC_REG_ICR, SLIC_ICR_INT_OFF);
+ slic_flush_write(sdev);
+ mdelay(1);
+
+ err = slic_init_rx_queue(sdev);
+ if (err) {
+ netdev_err(sdev->netdev, "failed to init rx queue: %u\n", err);
+ goto free_sm;
+ }
+
+ err = slic_init_tx_queue(sdev);
+ if (err) {
+ netdev_err(sdev->netdev, "failed to init tx queue: %u\n", err);
+ goto free_rxq;
+ }
+
+ err = slic_init_stat_queue(sdev);
+ if (err) {
+ netdev_err(sdev->netdev, "failed to init status queue: %u\n",
+ err);
+ goto free_txq;
+ }
+
+ slic_write(sdev, SLIC_REG_ISP, lower_32_bits(sm->isr_paddr));
+ napi_enable(&sdev->napi);
+ /* disable irq mitigation */
+ slic_write(sdev, SLIC_REG_INTAGG, 0);
+ slic_write(sdev, SLIC_REG_ISR, 0);
+ slic_flush_write(sdev);
+
+ slic_set_mac_address(sdev);
+
+ spin_lock_bh(&sdev->link_lock);
+ sdev->duplex = DUPLEX_UNKNOWN;
+ sdev->speed = SPEED_UNKNOWN;
+ spin_unlock_bh(&sdev->link_lock);
+
+ slic_set_link_autoneg(sdev);
+
+ err = request_irq(sdev->pdev->irq, slic_irq, IRQF_SHARED, DRV_NAME,
+ sdev);
+ if (err) {
+ netdev_err(sdev->netdev, "failed to request irq: %u\n", err);
+ goto disable_napi;
+ }
+
+ slic_write(sdev, SLIC_REG_ICR, SLIC_ICR_INT_ON);
+ slic_flush_write(sdev);
+ /* request initial link status */
+ err = slic_handle_link_change(sdev);
+ if (err)
+ netdev_warn(sdev->netdev,
+ "failed to set initial link state: %u\n", err);
+ return 0;
+
+disable_napi:
+ napi_disable(&sdev->napi);
+ slic_free_stat_queue(sdev);
+free_txq:
+ slic_free_tx_queue(sdev);
+free_rxq:
+ slic_free_rx_queue(sdev);
+free_sm:
+ slic_free_shmem(sdev);
+ slic_card_reset(sdev);
+
+ return err;
+}
+
+static int slic_open(struct net_device *dev)
+{
+ struct slic_device *sdev = netdev_priv(dev);
+ int err;
+
+ netif_carrier_off(dev);
+
+ err = slic_init_iface(sdev);
+ if (err) {
+ netdev_err(dev, "failed to initialize interface: %i\n", err);
+ return err;
+ }
+
+ netif_start_queue(dev);
+
+ return 0;
+}
+
+static int slic_close(struct net_device *dev)
+{
+ struct slic_device *sdev = netdev_priv(dev);
+ u32 val;
+
+ netif_stop_queue(dev);
+
+ /* stop irq handling */
+ napi_disable(&sdev->napi);
+ slic_write(sdev, SLIC_REG_ICR, SLIC_ICR_INT_OFF);
+ slic_write(sdev, SLIC_REG_ISR, 0);
+ slic_flush_write(sdev);
+
+ free_irq(sdev->pdev->irq, sdev);
+ /* turn off RCV and XMT and power down PHY */
+ val = SLIC_GXCR_RESET | SLIC_GXCR_PAUSEEN;
+ slic_write(sdev, SLIC_REG_WXCFG, val);
+
+ val = SLIC_GRCR_RESET | SLIC_GRCR_CTLEN | SLIC_GRCR_ADDRAEN |
+ SLIC_GRCR_HASHSIZE << SLIC_GRCR_HASHSIZE_SHIFT;
+ slic_write(sdev, SLIC_REG_WRCFG, val);
+
+ val = MII_BMCR << 16 | BMCR_PDOWN;
+ slic_write(sdev, SLIC_REG_WPHY, val);
+ slic_flush_write(sdev);
+
+ slic_clear_upr_list(&sdev->upr_list);
+ slic_write(sdev, SLIC_REG_QUIESCE, 0);
+
+ slic_free_stat_queue(sdev);
+ slic_free_tx_queue(sdev);
+ slic_free_rx_queue(sdev);
+ slic_free_shmem(sdev);
+
+ slic_card_reset(sdev);
+ netif_carrier_off(dev);
+
+ return 0;
+}
+
+static netdev_tx_t slic_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+ struct slic_device *sdev = netdev_priv(dev);
+ struct slic_tx_queue *txq = &sdev->txq;
+ struct slic_tx_buffer *buff;
+ struct slic_tx_desc *desc;
+ dma_addr_t paddr;
+ u32 cbar_val;
+ u32 maplen;
+
+ if (unlikely(slic_get_free_tx_descs(txq) < SLIC_MAX_REQ_TX_DESCS)) {
+ netdev_err(dev, "BUG! not enought tx LEs left: %u\n",
+ slic_get_free_tx_descs(txq));
+ return NETDEV_TX_BUSY;
+ }
+
+ maplen = skb_headlen(skb);
+ paddr = dma_map_single(&sdev->pdev->dev, skb->data, maplen,
+ DMA_TO_DEVICE);
+ if (dma_mapping_error(&sdev->pdev->dev, paddr)) {
+ netdev_err(dev, "failed to map tx buffer\n");
+ goto drop_skb;
+ }
+
+ buff = &txq->txbuffs[txq->put_idx];
+ buff->skb = skb;
+ dma_unmap_addr_set(buff, map_addr, paddr);
+ dma_unmap_len_set(buff, map_len, maplen);
+
+ desc = buff->desc;
+ desc->totlen = cpu_to_le32(maplen);
+ desc->paddrl = cpu_to_le32(lower_32_bits(paddr));
+ desc->paddrh = cpu_to_le32(upper_32_bits(paddr));
+ desc->len = cpu_to_le32(maplen);
+
+ txq->put_idx = slic_next_queue_idx(txq->put_idx, txq->len);
+
+ cbar_val = lower_32_bits(buff->desc_paddr) | 1;
+ /* complete writes to RAM and DMA before hardware is informed */
+ wmb();
+
+ slic_write(sdev, SLIC_REG_CBAR, cbar_val);
+
+ if (slic_get_free_tx_descs(txq) < SLIC_MAX_REQ_TX_DESCS)
+ netif_stop_queue(dev);
+ /* make sure writes to io-memory cant leak out of tx queue lock */
+ mmiowb();
+
+ return NETDEV_TX_OK;
+drop_skb:
+ dev_kfree_skb_any(skb);
+
+ return NETDEV_TX_OK;
+}
+
+static struct rtnl_link_stats64 *slic_get_stats(struct net_device *dev,
+ struct rtnl_link_stats64 *lst)
+{
+ struct slic_device *sdev = netdev_priv(dev);
+ struct slic_stats *stats = &sdev->stats;
+
+ SLIC_GET_STATS_COUNTER(lst->rx_packets, stats, rx_packets);
+ SLIC_GET_STATS_COUNTER(lst->tx_packets, stats, tx_packets);
+ SLIC_GET_STATS_COUNTER(lst->rx_bytes, stats, rx_bytes);
+ SLIC_GET_STATS_COUNTER(lst->tx_bytes, stats, tx_bytes);
+ SLIC_GET_STATS_COUNTER(lst->rx_errors, stats, rx_errors);
+ SLIC_GET_STATS_COUNTER(lst->rx_dropped, stats, rx_buff_miss);
+ SLIC_GET_STATS_COUNTER(lst->tx_dropped, stats, tx_dropped);
+ SLIC_GET_STATS_COUNTER(lst->multicast, stats, rx_mcasts);
+ SLIC_GET_STATS_COUNTER(lst->rx_over_errors, stats, rx_buffoflow);
+ SLIC_GET_STATS_COUNTER(lst->rx_crc_errors, stats, rx_crc);
+ SLIC_GET_STATS_COUNTER(lst->rx_fifo_errors, stats, rx_oflow802);
+ SLIC_GET_STATS_COUNTER(lst->tx_carrier_errors, stats, tx_carrier);
+
+ return lst;
+}
+
+static int slic_get_sset_count(struct net_device *dev, int sset)
+{
+ switch (sset) {
+ case ETH_SS_STATS:
+ return ARRAY_SIZE(slic_stats_strings);
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
+static void slic_get_ethtool_stats(struct net_device *dev,
+ struct ethtool_stats *eth_stats, u64 *data)
+{
+ struct slic_device *sdev = netdev_priv(dev);
+ struct slic_stats *stats = &sdev->stats;
+
+ SLIC_GET_STATS_COUNTER(data[0], stats, rx_packets);
+ SLIC_GET_STATS_COUNTER(data[1], stats, rx_bytes);
+ SLIC_GET_STATS_COUNTER(data[2], stats, rx_mcasts);
+ SLIC_GET_STATS_COUNTER(data[3], stats, rx_errors);
+ SLIC_GET_STATS_COUNTER(data[4], stats, rx_buff_miss);
+ SLIC_GET_STATS_COUNTER(data[5], stats, rx_tpcsum);
+ SLIC_GET_STATS_COUNTER(data[6], stats, rx_tpoflow);
+ SLIC_GET_STATS_COUNTER(data[7], stats, rx_tphlen);
+ SLIC_GET_STATS_COUNTER(data[8], stats, rx_ipcsum);
+ SLIC_GET_STATS_COUNTER(data[9], stats, rx_iplen);
+ SLIC_GET_STATS_COUNTER(data[10], stats, rx_iphlen);
+ SLIC_GET_STATS_COUNTER(data[11], stats, rx_early);
+ SLIC_GET_STATS_COUNTER(data[12], stats, rx_buffoflow);
+ SLIC_GET_STATS_COUNTER(data[13], stats, rx_lcode);
+ SLIC_GET_STATS_COUNTER(data[14], stats, rx_drbl);
+ SLIC_GET_STATS_COUNTER(data[15], stats, rx_crc);
+ SLIC_GET_STATS_COUNTER(data[16], stats, rx_oflow802);
+ SLIC_GET_STATS_COUNTER(data[17], stats, rx_uflow802);
+ SLIC_GET_STATS_COUNTER(data[18], stats, tx_packets);
+ SLIC_GET_STATS_COUNTER(data[19], stats, tx_bytes);
+ SLIC_GET_STATS_COUNTER(data[20], stats, tx_carrier);
+ SLIC_GET_STATS_COUNTER(data[21], stats, tx_dropped);
+ SLIC_GET_STATS_COUNTER(data[22], stats, irq_errs);
+}
+
+static void slic_get_strings(struct net_device *dev, u32 stringset, u8 *data)
+{
+ if (stringset == ETH_SS_STATS) {
+ memcpy(data, slic_stats_strings, sizeof(slic_stats_strings));
+ data += sizeof(slic_stats_strings);
+ }
+}
+
+static void slic_get_drvinfo(struct net_device *dev,
+ struct ethtool_drvinfo *info)
+{
+ struct slic_device *sdev = netdev_priv(dev);
+
+ strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
+ strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+ strlcpy(info->bus_info, pci_name(sdev->pdev), sizeof(info->bus_info));
+}
+
+static const struct ethtool_ops slic_ethtool_ops = {
+ .get_drvinfo = slic_get_drvinfo,
+ .get_link = ethtool_op_get_link,
+ .get_strings = slic_get_strings,
+ .get_ethtool_stats = slic_get_ethtool_stats,
+ .get_sset_count = slic_get_sset_count,
+};
+
+static const struct net_device_ops slic_netdev_ops = {
+ .ndo_open = slic_open,
+ .ndo_stop = slic_close,
+ .ndo_start_xmit = slic_xmit,
+ .ndo_set_mac_address = eth_mac_addr,
+ .ndo_get_stats64 = slic_get_stats,
+ .ndo_set_rx_mode = slic_set_rx_mode,
+ .ndo_change_mtu = eth_change_mtu,
+ .ndo_validate_addr = eth_validate_addr,
+};
+
+static u16 slic_eeprom_csum(unsigned char *eeprom, unsigned int len)
+{
+ unsigned char *ptr = eeprom;
+ u32 csum = 0;
+ __le16 data;
+
+ while (len > 1) {
+ memcpy(&data, ptr, sizeof(data));
+ csum += le16_to_cpu(data);
+ ptr += 2;
+ len -= 2;
+ }
+ if (len > 0)
+ csum += *(u8 *)ptr;
+ while (csum >> 16)
+ csum = (csum & 0xFFFF) + ((csum >> 16) & 0xFFFF);
+ return ~csum;
+}
+
+/* check eeprom size, magic and checksum */
+static bool slic_eeprom_valid(unsigned char *eeprom, unsigned int size)
+{
+ const unsigned int MAX_SIZE = 128;
+ const unsigned int MIN_SIZE = 98;
+ __le16 magic;
+ __le16 csum;
+
+ if (size < MIN_SIZE || size > MAX_SIZE)
+ return false;
+ memcpy(&magic, eeprom, sizeof(magic));
+ if (le16_to_cpu(magic) != SLIC_EEPROM_MAGIC)
+ return false;
+ /* cut checksum bytes */
+ size -= 2;
+ memcpy(&csum, eeprom + size, sizeof(csum));
+
+ return (le16_to_cpu(csum) == slic_eeprom_csum(eeprom, size));
+}
+
+static int slic_read_eeprom(struct slic_device *sdev)
+{
+ unsigned int devfn = PCI_FUNC(sdev->pdev->devfn);
+ struct slic_shmem *sm = &sdev->shmem;
+ struct slic_shmem_data *sm_data = sm->shmem_data;
+ const unsigned int MAX_LOOPS = 5000;
+ unsigned int codesize;
+ unsigned char *eeprom;
+ struct slic_upr *upr;
+ dma_addr_t paddr;
+ int err = 0;
+ u8 *mac[2];
+ int i = 0;
+
+ eeprom = dma_zalloc_coherent(&sdev->pdev->dev, SLIC_EEPROM_SIZE,
+ &paddr, GFP_KERNEL);
+ if (!eeprom)
+ return -ENOMEM;
+
+ slic_write(sdev, SLIC_REG_ICR, SLIC_ICR_INT_OFF);
+ /* setup ISP temporarily */
+ slic_write(sdev, SLIC_REG_ISP, lower_32_bits(sm->isr_paddr));
+
+ err = slic_new_upr(sdev, SLIC_UPR_CONFIG, paddr);
+ if (!err) {
+ for (i = 0; i < MAX_LOOPS; i++) {
+ if (le32_to_cpu(sm_data->isr) & SLIC_ISR_UPC)
+ break;
+ mdelay(1);
+ }
+ if (i == MAX_LOOPS) {
+ dev_err(&sdev->pdev->dev,
+ "timed out while waiting for eeprom data\n");
+ err = -ETIMEDOUT;
+ }
+ upr = slic_dequeue_upr(sdev);
+ kfree(upr);
+ }
+
+ slic_write(sdev, SLIC_REG_ISP, 0);
+ slic_write(sdev, SLIC_REG_ISR, 0);
+ slic_flush_write(sdev);
+
+ if (err)
+ goto free_eeprom;
+
+ if (sdev->model == SLIC_MODEL_OASIS) {
+ struct slic_oasis_eeprom *oee;
+
+ oee = (struct slic_oasis_eeprom *)eeprom;
+ mac[0] = oee->mac;
+ mac[1] = oee->mac2;
+ codesize = le16_to_cpu(oee->eeprom_code_size);
+ } else {
+ struct slic_mojave_eeprom *mee;
+
+ mee = (struct slic_mojave_eeprom *)eeprom;
+ mac[0] = mee->mac;
+ mac[1] = mee->mac2;
+ codesize = le16_to_cpu(mee->eeprom_code_size);
+ }
+
+ if (!slic_eeprom_valid(eeprom, codesize)) {
+ dev_err(&sdev->pdev->dev, "invalid checksum in eeprom\n");
+ err = -EINVAL;
+ goto free_eeprom;
+ }
+ /* set mac address */
+ ether_addr_copy(sdev->netdev->dev_addr, mac[devfn]);
+free_eeprom:
+ dma_free_coherent(&sdev->pdev->dev, SLIC_EEPROM_SIZE, eeprom, paddr);
+
+ return err;
+}
+
+static int slic_init(struct slic_device *sdev)
+{
+ int err;
+
+ spin_lock_init(&sdev->upper_lock);
+ spin_lock_init(&sdev->link_lock);
+ INIT_LIST_HEAD(&sdev->upr_list.list);
+ spin_lock_init(&sdev->upr_list.lock);
+ u64_stats_init(&sdev->stats.syncp);
+
+ slic_card_reset(sdev);
+
+ err = slic_load_firmware(sdev);
+ if (err) {
+ dev_err(&sdev->pdev->dev, "failed to load firmware\n");
+ return err;
+ }
+
+ /* we need the shared memory to read EEPROM so set it up temporarily */
+ err = slic_init_shmem(sdev);
+ if (err) {
+ dev_err(&sdev->pdev->dev, "failed to init shared memory\n");
+ return err;
+ }
+
+ err = slic_read_eeprom(sdev);
+ if (err) {
+ dev_err(&sdev->pdev->dev, "failed to read eeprom\n");
+ goto free_sm;
+ }
+
+ slic_card_reset(sdev);
+ slic_free_shmem(sdev);
+
+ return 0;
+free_sm:
+ slic_free_shmem(sdev);
+
+ return err;
+}
+
+static bool slic_is_fiber(unsigned short subdev)
+{
+ switch (subdev) {
+ /* Mojave */
+ case PCI_SUBDEVICE_ID_ALACRITECH_1000X1F: /* fallthrough */
+ case PCI_SUBDEVICE_ID_ALACRITECH_SES1001F: /* fallthrough */
+ /* Oasis */
+ case PCI_SUBDEVICE_ID_ALACRITECH_SEN2002XF: /* fallthrough */
+ case PCI_SUBDEVICE_ID_ALACRITECH_SEN2001XF: /* fallthrough */
+ case PCI_SUBDEVICE_ID_ALACRITECH_SEN2104EF: /* fallthrough */
+ case PCI_SUBDEVICE_ID_ALACRITECH_SEN2102EF: /* fallthrough */
+ return true;
+ }
+ return false;
+}
+
+static void slic_configure_pci(struct pci_dev *pdev)
+{
+ u16 old;
+ u16 cmd;
+
+ pci_read_config_word(pdev, PCI_COMMAND, &old);
+
+ cmd = old | PCI_COMMAND_PARITY | PCI_COMMAND_SERR;
+ if (old != cmd)
+ pci_write_config_word(pdev, PCI_COMMAND, cmd);
+}
+
+static int slic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+ struct slic_device *sdev;
+ struct net_device *dev;
+ int err;
+
+ err = pci_enable_device(pdev);
+ if (err) {
+ dev_err(&pdev->dev, "failed to enable PCI device\n");
+ return err;
+ }
+
+ pci_set_master(pdev);
+ pci_try_set_mwi(pdev);
+
+ slic_configure_pci(pdev);
+
+ err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32));
+ if (err) {
+ dev_err(&pdev->dev, "failed to setup DMA\n");
+ goto disable;
+ }
+
+ dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
+
+ err = pci_request_regions(pdev, DRV_NAME);
+ if (err) {
+ dev_err(&pdev->dev, "failed to obtain PCI regions\n");
+ goto disable;
+ }
+
+ dev = alloc_etherdev(sizeof(*sdev));
+ if (!dev) {
+ dev_err(&pdev->dev, "failed to alloc ethernet device\n");
+ err = -ENOMEM;
+ goto free_regions;
+ }
+
+ SET_NETDEV_DEV(dev, &pdev->dev);
+ pci_set_drvdata(pdev, dev);
+ dev->irq = pdev->irq;
+ dev->netdev_ops = &slic_netdev_ops;
+ dev->hw_features = NETIF_F_RXCSUM;
+ dev->features |= dev->hw_features;
+
+ dev->ethtool_ops = &slic_ethtool_ops;
+
+ sdev = netdev_priv(dev);
+ sdev->model = (pdev->device == PCI_DEVICE_ID_ALACRITECH_OASIS) ?
+ SLIC_MODEL_OASIS : SLIC_MODEL_MOJAVE;
+ sdev->is_fiber = slic_is_fiber(pdev->subsystem_device);
+ sdev->pdev = pdev;
+ sdev->netdev = dev;
+ sdev->regs = ioremap_nocache(pci_resource_start(pdev, 0),
+ pci_resource_len(pdev, 0));
+ if (!sdev->regs) {
+ dev_err(&pdev->dev, "failed to map registers\n");
+ err = -ENOMEM;
+ goto free_netdev;
+ }
+
+ err = slic_init(sdev);
+ if (err) {
+ dev_err(&pdev->dev, "failed to initialize driver\n");
+ err = -ENOMEM;
+ goto unmap;
+ }
+
+ netif_napi_add(dev, &sdev->napi, slic_poll, SLIC_NAPI_WEIGHT);
+ netif_carrier_off(dev);
+
+ err = register_netdev(dev);
+ if (err) {
+ dev_err(&pdev->dev, "failed to register net device: %i\n",
+ err);
+ goto unmap;
+ }
+
+ return 0;
+
+unmap:
+ iounmap(sdev->regs);
+free_netdev:
+ free_netdev(dev);
+free_regions:
+ pci_release_regions(pdev);
+disable:
+ pci_disable_device(pdev);
+
+ return err;
+}
+
+static void slic_remove(struct pci_dev *pdev)
+{
+ struct net_device *dev = pci_get_drvdata(pdev);
+ struct slic_device *sdev = netdev_priv(dev);
+
+ unregister_netdev(dev);
+ iounmap(sdev->regs);
+ free_netdev(dev);
+ pci_release_regions(pdev);
+ pci_disable_device(pdev);
+}
+
+static struct pci_driver slic_driver = {
+ .name = DRV_NAME,
+ .id_table = slic_id_tbl,
+ .probe = slic_probe,
+ .remove = slic_remove,
+};
+
+static int __init slic_init_module(void)
+{
+ return pci_register_driver(&slic_driver);
+}
+
+static void __exit slic_cleanup_module(void)
+{
+ pci_unregister_driver(&slic_driver);
+}
+
+module_init(slic_init_module);
+module_exit(slic_cleanup_module);
+
+MODULE_DESCRIPTION("Alacritech non-accelerated SLIC driver");
+MODULE_AUTHOR("Lino Sanfilippo <LinoSanfilippo@gmx.de>");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_VERSION);
--
1.9.1
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH v3 net-next 2/2] MAINTAINERS: add entry for slicoss ethernet driver
2016-11-26 12:20 Gigabit ethernet driver for Alacritechs SLIC devices (v3) Lino Sanfilippo
2016-11-26 12:20 ` [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver Lino Sanfilippo
@ 2016-11-26 12:20 ` Lino Sanfilippo
2016-11-26 12:36 ` Gigabit ethernet driver for Alacritechs SLIC devices (v3) Lino Sanfilippo
2 siblings, 0 replies; 14+ messages in thread
From: Lino Sanfilippo @ 2016-11-26 12:20 UTC (permalink / raw)
To: davem, charrer, liodot, gregkh, andrew
Cc: devel, linux-kernel, netdev, Lino Sanfilippo
Add myself as maintainer for the slicoss ethernet driver.
Signed-off-by: Lino Sanfilippo <LinoSanfilippo@gmx.de>
---
MAINTAINERS | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 6781a3f..bb9af28 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -562,6 +562,11 @@ T: git git://linuxtv.org/anttip/media_tree.git
S: Maintained
F: drivers/media/usb/airspy/
+ALACRITECH GIGABIT ETHERNET DRIVER
+M: Lino Sanfilippo <LinoSanfilippo@gmx.de>
+S: Maintained
+F: drivers/net/ethernet/alacritech/*
+
ALCATEL SPEEDTOUCH USB DRIVER
M: Duncan Sands <duncan.sands@free.fr>
L: linux-usb@vger.kernel.org
--
1.9.1
^ permalink raw reply related [flat|nested] 14+ messages in thread
* Re: Gigabit ethernet driver for Alacritechs SLIC devices (v3)
2016-11-26 12:20 Gigabit ethernet driver for Alacritechs SLIC devices (v3) Lino Sanfilippo
2016-11-26 12:20 ` [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver Lino Sanfilippo
2016-11-26 12:20 ` [PATCH v3 net-next 2/2] MAINTAINERS: add entry for slicoss " Lino Sanfilippo
@ 2016-11-26 12:36 ` Lino Sanfilippo
2 siblings, 0 replies; 14+ messages in thread
From: Lino Sanfilippo @ 2016-11-26 12:36 UTC (permalink / raw)
To: davem, charrer, liodot, gregkh, andrew; +Cc: devel, linux-kernel, netdev
On 26.11.2016 13:20, Lino Sanfilippo wrote:
> v3:
> - dont add defines to pci.h but instead put it into the drivers header file
This should of course be "pci_ids.h".
Regards,
Lino
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-26 12:20 ` [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver Lino Sanfilippo
@ 2016-11-26 15:48 ` Rami Rosen
2016-11-27 1:02 ` Lino Sanfilippo
2016-11-27 17:59 ` Markus Böhme
` (2 subsequent siblings)
3 siblings, 1 reply; 14+ messages in thread
From: Rami Rosen @ 2016-11-26 15:48 UTC (permalink / raw)
To: Lino Sanfilippo
Cc: David Miller, charrer, liodot, gregkh, andrew, devel,
linux-kernel, Netdev
Hi, Lino,
...
> @@ -0,0 +1,28 @@
> +config NET_VENDOR_ALACRITECH
> + bool "Alacritech devices"
> + default y
> + ---help---
> + If you have a network (Ethernet) card belonging to this class, say Y.
> +
> + Note that the answer to this question doesn't directly affect the
> + kernel: saying N will just cause the configurator to skip all
Shouldn't it be "Alacritech devices" here, as appears earlier ?
> + the questions about Renesas devices. If you say Y, you will be asked
> + for your specific device in the following questions.
> +
...
...
...
> +struct slic_device {
> + struct pci_dev *pdev;
...
> + bool promisc;
Seems that the autoneg boolean is not used anywhere, apart from
setting it once to true in
the slic_set_link_autoneg() method. Apart from this member it is not
accessed anywhere, so it seems it should be removed.
> + bool autoneg;
> + int speed;
...
...
> +static int slic_load_rcvseq_firmware(struct slic_device *sdev)
> +{
> + const struct firmware *fw;
> + const char *file;
> + u32 codelen;
> + int idx = 0;
> + u32 instr;
> + u32 addr;
> + int err;
> +
...
> + /* Do an initial sanity check concerning firmware size now. A further
> + * check follows below.
> + */
> + if (fw->size < SLIC_FIRMWARE_MIN_SIZE) {
> + dev_err(&sdev->pdev->dev,
> + "invalid firmware size %zu (min %u expected)\n",
> + fw->size, SLIC_FIRMWARE_MIN_SIZE);
> + err = -EINVAL;
in the release label, always 0 is returned:
> + goto release;
> + }
> +
> + codelen = slic_read_dword_from_firmware(fw, &idx);
> +
> + /* do another sanity check against firmware size */
> + if ((codelen + 4) > fw->size) {
> + dev_err(&sdev->pdev->dev,
> + "invalid rcv-sequencer firmware size %zu\n", fw->size);
> + err = -EINVAL;
Again, in the release label, always 0 is returned:
> + goto release;
> + }
> +
>
> +release:
> + release_firmware(fw);
> +
> + return 0;
> +}
> +
Regards,
Rami Rosen
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-26 15:48 ` Rami Rosen
@ 2016-11-27 1:02 ` Lino Sanfilippo
0 siblings, 0 replies; 14+ messages in thread
From: Lino Sanfilippo @ 2016-11-27 1:02 UTC (permalink / raw)
To: Rami Rosen
Cc: David Miller, charrer, liodot, gregkh, andrew, devel,
linux-kernel, Netdev
Hi Rami,
On 26.11.2016 16:48, Rami Rosen wrote:
>> @@ -0,0 +1,28 @@
>> +config NET_VENDOR_ALACRITECH
>> + bool "Alacritech devices"
>> + default y
>> + ---help---
>> + If you have a network (Ethernet) card belonging to this class, say Y.
>> +
>> + Note that the answer to this question doesn't directly affect the
>> + kernel: saying N will just cause the configurator to skip all
>
> Shouldn't it be "Alacritech devices" here, as appears earlier ?
>
>> + the questions about Renesas devices. If you say Y, you will be asked
Yes, it definitely should not be Renesas :). This is a stupid copy and paste error, I will fix it,
thank you!
>> + for your specific device in the following questions.
>> +
>
> ...
> ...
> ...
>> +struct slic_device {
>> + struct pci_dev *pdev;
> ...
>> + bool promisc;
>
> Seems that the autoneg boolean is not used anywhere, apart from
> setting it once to true in
> the slic_set_link_autoneg() method. Apart from this member it is not
> accessed anywhere, so it seems it should be removed.
>
>> + bool autoneg;
>> + int speed;
Agreed, this variable can be removed.
> ...
>
>> +static int slic_load_rcvseq_firmware(struct slic_device *sdev)
>> +{
>> + const struct firmware *fw;
>> + const char *file;
>> + u32 codelen;
>> + int idx = 0;
>> + u32 instr;
>> + u32 addr;
>> + int err;
>> +
> ...
>> + /* Do an initial sanity check concerning firmware size now. A further
>> + * check follows below.
>> + */
>> + if (fw->size < SLIC_FIRMWARE_MIN_SIZE) {
>> + dev_err(&sdev->pdev->dev,
>> + "invalid firmware size %zu (min %u expected)\n",
>> + fw->size, SLIC_FIRMWARE_MIN_SIZE);
>> + err = -EINVAL;
>
> in the release label, always 0 is returned:
>
>> + goto release;
>> + }
>> +
>> + codelen = slic_read_dword_from_firmware(fw, &idx);
>> +
>> + /* do another sanity check against firmware size */
>> + if ((codelen + 4) > fw->size) {
>> + dev_err(&sdev->pdev->dev,
>> + "invalid rcv-sequencer firmware size %zu\n", fw->size);
>> + err = -EINVAL;
>
> Again, in the release label, always 0 is returned:
>
>> + goto release;
>> + }
>> +
>>
>> +release:
>> + release_firmware(fw);
>> +
>> + return 0;
>> +}
This should return "err", I will fix it.
Thanks a lot for the review!
Regards,
Lino
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-26 12:20 ` [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver Lino Sanfilippo
2016-11-26 15:48 ` Rami Rosen
@ 2016-11-27 17:59 ` Markus Böhme
2016-11-28 20:46 ` Lino Sanfilippo
2016-11-28 4:56 ` Florian Fainelli
2016-12-04 2:13 ` kbuild test robot
3 siblings, 1 reply; 14+ messages in thread
From: Markus Böhme @ 2016-11-27 17:59 UTC (permalink / raw)
To: Lino Sanfilippo, davem, charrer, liodot, gregkh, andrew
Cc: devel, netdev, linux-kernel
Hello Lino,
just some things barely worth mentioning:
On 11/26/2016 01:20 PM, Lino Sanfilippo wrote:
> Add driver for Alacritech gigabit ethernet cards with SLIC (session-layer
> interface control) technology. The driver provides basic support without
> SLIC for the following devices:
>
> - Mojave cards (single port PCI Gigabit) both copper and fiber
> - Oasis cards (single and dual port PCI-x Gigabit) copper and fiber
> - Kalahari cards (dual and quad port PCI-e Gigabit) copper and fiber
>
> Signed-off-by: Lino Sanfilippo <LinoSanfilippo@gmx.de>
> ---
> drivers/net/ethernet/Kconfig | 1 +
> drivers/net/ethernet/Makefile | 1 +
> drivers/net/ethernet/alacritech/Kconfig | 28 +
> drivers/net/ethernet/alacritech/Makefile | 4 +
> drivers/net/ethernet/alacritech/slic.h | 576 +++++++++
> drivers/net/ethernet/alacritech/slicoss.c | 1867 +++++++++++++++++++++++++++++
> 6 files changed, 2477 insertions(+)
> create mode 100644 drivers/net/ethernet/alacritech/Kconfig
> create mode 100644 drivers/net/ethernet/alacritech/Makefile
> create mode 100644 drivers/net/ethernet/alacritech/slic.h
> create mode 100644 drivers/net/ethernet/alacritech/slicoss.c
>
[...]
> diff --git a/drivers/net/ethernet/alacritech/slic.h b/drivers/net/ethernet/alacritech/slic.h
> new file mode 100644
> index 0000000..c62d46b
> --- /dev/null
> +++ b/drivers/net/ethernet/alacritech/slic.h
> @@ -0,0 +1,576 @@
> +
> +#ifndef _SLIC_H
> +#define _SLIC_H
I found a bunch of unused #defines in slic.h. I cannot judge if they are
worth keeping:
SLIC_VRHSTATB_LONGE
SLIC_VRHSTATB_PREA
SLIC_ISR_IO
SLIC_ISR_PING_MASK
SLIC_GIG_SPEED_MASK
SLIC_GMCR_RESET
SLIC_XCR_RESET
SLIC_XCR_XMTEN
SLIC_XCR_PAUSEEN
SLIC_XCR_LOADRNG
SLIC_REG_DBAR
SLIC_REG_PING
SLIC_REG_DUMP_CMD
SLIC_REG_DUMP_DATA
SLIC_REG_WRHOSTID
SLIC_REG_LOW_POWER
SLIC_REG_RESET_IFACE
SLIC_REG_ADDR_UPPER
SLIC_REG_HBAR64
SLIC_REG_DBAR64
SLIC_REG_CBAR64
SLIC_REG_RBAR64
SLIC_REG_WRVLANID
SLIC_REG_READ_XF_INFO
SLIC_REG_WRITE_XF_INFO
SLIC_REG_TICKS_PER_SEC
These device IDs are not used, either, but maybe it's good to keep them
for documentation purposes:
PCI_SUBDEVICE_ID_ALACRITECH_1000X1_2
PCI_SUBDEVICE_ID_ALACRITECH_SES1001T
PCI_SUBDEVICE_ID_ALACRITECH_SEN2002XT
PCI_SUBDEVICE_ID_ALACRITECH_SEN2001XT
PCI_SUBDEVICE_ID_ALACRITECH_SEN2104ET
PCI_SUBDEVICE_ID_ALACRITECH_SEN2102ET
[...]
> +
> +/* SLIC EEPROM structure for Oasis */
> +struct slic_mojave_eeprom {
Comment: "for Mojave".
[...]
> +struct slic_device {
> + struct pci_dev *pdev;
> + struct net_device *netdev;
> + void __iomem *regs;
> + /* upper address setting lock */
> + spinlock_t upper_lock;
> + struct slic_shmem shmem;
> + struct napi_struct napi;
> + struct slic_rx_queue rxq;
> + struct slic_tx_queue txq;
> + struct slic_stat_queue stq;
> + struct slic_stats stats;
> + struct slic_upr_list upr_list;
> + /* link configuration lock */
> + spinlock_t link_lock;
> + bool promisc;
> + bool autoneg;
> + int speed;
> + int duplex;
Maybe make speed and duplex unsigned? They are assigned and compared
against unsigned values in slicoss.c, so this would get rid of some
(benign, because of the range of the values) -Wsign-compare warnings in
slic_configure_link_locked. However, in a comparison there SPEED_UNKNOWN
would need to be casted to unsigned to prevent another one popping up.
[...]
> +#endif /* _SLIC_H */
> diff --git a/drivers/net/ethernet/alacritech/slicoss.c b/drivers/net/ethernet/alacritech/slicoss.c
> new file mode 100644
> index 0000000..8cd862a
> --- /dev/null
> +++ b/drivers/net/ethernet/alacritech/slicoss.c
> @@ -0,0 +1,1867 @@
[...]
> +
> +static const struct pci_device_id slic_id_tbl[] = {
> + { PCI_DEVICE(PCI_VENDOR_ID_ALACRITECH,
> + PCI_DEVICE_ID_ALACRITECH_MOAVE) },
I missed this in slic.h, but is this a typo and "MOAVE" should be
"MOJAVE"? There are a couple similar #defines in slic.h.
[...]
> +static void slic_refill_rx_queue(struct slic_device *sdev, gfp_t gfp)
> +{
> + const unsigned int ALIGN_MASK = SLIC_RX_BUFF_ALIGN - 1;
> + unsigned int maplen = SLIC_RX_BUFF_SIZE;
> + struct slic_rx_queue *rxq = &sdev->rxq;
> + struct net_device *dev = sdev->netdev;
> + struct slic_rx_buffer *buff;
> + struct slic_rx_desc *desc;
> + unsigned int misalign;
> + unsigned int offset;
> + struct sk_buff *skb;
> + dma_addr_t paddr;
> +
> + while (slic_get_free_rx_descs(rxq) > SLIC_MAX_REQ_RX_DESCS) {
> + skb = alloc_skb(maplen + ALIGN_MASK, gfp);
> + if (!skb)
> + break;
> +
> + paddr = dma_map_single(&sdev->pdev->dev, skb->data, maplen,
> + DMA_FROM_DEVICE);
> + if (dma_mapping_error(&sdev->pdev->dev, paddr)) {
> + netdev_err(dev, "mapping rx packet failed\n");
> + /* drop skb */
> + dev_kfree_skb_any(skb);
> + break;
> + }
> + /* ensure head buffer descriptors are 256 byte aligned */
> + offset = 0;
> + misalign = paddr & ALIGN_MASK;
> + if (misalign) {
> + offset = SLIC_RX_BUFF_ALIGN - misalign;
> + skb_reserve(skb, offset);
> + }
> + /* the HW expects dma chunks for descriptor + frame data */
> + desc = (struct slic_rx_desc *)skb->data;
> + memset(desc, 0, sizeof(*desc));
> +
> + buff = &rxq->rxbuffs[rxq->put_idx];
> + buff->skb = skb;
> + dma_unmap_addr_set(buff, map_addr, paddr);
> + dma_unmap_len_set(buff, map_len, maplen);
> + buff->addr_offset = offset;
> + /* head buffer descriptors are placed immediately before skb */
> + slic_write(sdev, SLIC_REG_HBAR, lower_32_bits(paddr) +
> + offset);
This fits nicely on one line. :-)
[...]
> +static int slic_init_tx_queue(struct slic_device *sdev)
> +{
> + struct slic_tx_queue *txq = &sdev->txq;
> + struct slic_tx_buffer *buff;
> + struct slic_tx_desc *desc;
> + int err;
> + int i;
You could make i unsigned...
> +
> + txq->len = SLIC_NUM_TX_DESCS;
> + txq->put_idx = 0;
> + txq->done_idx = 0;
> +
> + txq->txbuffs = kcalloc(txq->len, sizeof(*buff), GFP_KERNEL);
> + if (!txq->txbuffs)
> + return -ENOMEM;
> +
> + txq->dma_pool = dma_pool_create("slic_pool", &sdev->pdev->dev,
> + sizeof(*desc), SLIC_TX_DESC_ALIGN,
> + 4096);
> + if (!txq->dma_pool) {
> + err = -ENOMEM;
> + netdev_err(sdev->netdev, "failed to create dma pool\n");
> + goto free_buffs;
> + }
> +
> + for (i = 0; i < txq->len; i++) {
...to fix a signed/unsigned comparison warning here, but...
> + buff = &txq->txbuffs[i];
> + desc = dma_pool_zalloc(txq->dma_pool, GFP_KERNEL,
> + &buff->desc_paddr);
> + if (!desc) {
> + netdev_err(sdev->netdev,
> + "failed to alloc pool chunk (%i)\n", i);
> + err = -ENOMEM;
> + goto free_descs;
> + }
> +
> + desc->hnd = cpu_to_le32((u32)(i + 1));
> + desc->cmd = SLIC_CMD_XMT_REQ;
> + desc->flags = 0;
> + desc->type = cpu_to_le32(SLIC_CMD_TYPE_DUMB);
> + buff->desc = desc;
> + }
> +
> + return 0;
> +
> +free_descs:
> + while (i--) {
...this would require reworking this logic to prevent an endless loop,
so probably not worth bothering, considering that txq->len is well
within the positive signed range.
> + buff = &txq->txbuffs[i];
> + dma_pool_free(txq->dma_pool, buff->desc, buff->desc_paddr);
> + }
> + dma_pool_destroy(txq->dma_pool);
> +
> +free_buffs:
> + kfree(txq->txbuffs);
> +
> + return err;
> +}
> +
> +static void slic_free_tx_queue(struct slic_device *sdev)
> +{
> + struct slic_tx_queue *txq = &sdev->txq;
> + struct slic_tx_buffer *buff;
> + int i;
Make i unsigned? One warning less, almost no work invested.
> +
> + for (i = 0; i < txq->len; i++) {
> + buff = &txq->txbuffs[i];
> + dma_pool_free(txq->dma_pool, buff->desc, buff->desc_paddr);
> + if (!buff->skb)
> + continue;
> +
> + dma_unmap_single(&sdev->pdev->dev,
> + dma_unmap_addr(buff, map_addr),
> + dma_unmap_len(buff, map_len), DMA_TO_DEVICE);
> + consume_skb(buff->skb);
> + }
> + dma_pool_destroy(txq->dma_pool);
> +
> + kfree(txq->txbuffs);
> +}
> +
[...]
> +static void slic_free_rx_queue(struct slic_device *sdev)
> +{
> + struct slic_rx_queue *rxq = &sdev->rxq;
> + struct slic_rx_buffer *buff;
> + int i;
Unsigned?
> +
> + /* free rx buffers */
> + for (i = 0; i < rxq->len; i++) {
> + buff = &rxq->rxbuffs[i];
> +
> + if (!buff->skb)
> + continue;
> +
> + dma_unmap_single(&sdev->pdev->dev,
> + dma_unmap_addr(buff, map_addr),
> + dma_unmap_len(buff, map_len),
> + DMA_FROM_DEVICE);
> + consume_skb(buff->skb);
> + }
> + kfree(rxq->rxbuffs);
> +}
[...]
> +static int slic_load_firmware(struct slic_device *sdev)
> +{
> + u32 sectstart[SLIC_FIRMWARE_MAX_SECTIONS];
> + u32 sectsize[SLIC_FIRMWARE_MAX_SECTIONS];
> + const struct firmware *fw;
> + unsigned int datalen;
> + const char *file;
> + int code_start;
> + u32 numsects;
> + int idx = 0;
> + u32 sect;
> + u32 instr;
> + u32 addr;
> + u32 base;
> + int err;
> + int i;
Make i unsigned?
> +
> + file = (sdev->model == SLIC_MODEL_OASIS) ? SLIC_FIRMWARE_OASIS :
> + SLIC_FIRMWARE_MOAVE;
> + err = request_firmware(&fw, file, &sdev->pdev->dev);
> + if (err) {
> + dev_err(&sdev->pdev->dev, "failed to load firmware %s\n", file);
> + return err;
> + }
> + /* Do an initial sanity check concerning firmware size now. A further
> + * check follows below.
> + */
> + if (fw->size < SLIC_FIRMWARE_MIN_SIZE) {
> + dev_err(&sdev->pdev->dev,
> + "invalid firmware size %zu (min is %u)\n", fw->size,
> + SLIC_FIRMWARE_MIN_SIZE);
> + err = -EINVAL;
> + goto release;
> + }
> +
> + numsects = slic_read_dword_from_firmware(fw, &idx);
> + if (numsects == 0 || numsects > SLIC_FIRMWARE_MAX_SECTIONS) {
> + dev_err(&sdev->pdev->dev,
> + "invalid number of sections in firmware: %u", numsects);
> + err = -EINVAL;
> + goto release;
> + }
> +
> + datalen = numsects * 8 + 4;
> + for (i = 0; i < numsects; i++) {
> + sectsize[i] = slic_read_dword_from_firmware(fw, &idx);
> + datalen += sectsize[i];
> + }
> +
> + /* do another sanity check against firmware size */
> + if (datalen > fw->size) {
> + dev_err(&sdev->pdev->dev,
> + "invalid firmware size %zu (expected >= %u)\n",
> + fw->size, datalen);
> + err = -EINVAL;
> + goto release;
> + }
> + /* get sections */
> + for (i = 0; i < numsects; i++)
> + sectstart[i] = slic_read_dword_from_firmware(fw, &idx);
> +
> + code_start = idx;
> + instr = slic_read_dword_from_firmware(fw, &idx);
> +
> + for (sect = 0; sect < numsects; sect++) {
> + unsigned int ssize = sectsize[sect] >> 3;
> +
> + base = sectstart[sect];
> +
> + for (addr = 0; addr < ssize; addr++) {
> + /* write out instruction address */
> + slic_write(sdev, SLIC_REG_WCS, base + addr);
> + /* write out instruction to low addr */
> + slic_write(sdev, SLIC_REG_WCS, instr);
> + instr = slic_read_dword_from_firmware(fw, &idx);
> + /* write out instruction to high addr */
> + slic_write(sdev, SLIC_REG_WCS, instr);
> + instr = slic_read_dword_from_firmware(fw, &idx);
> + }
> + }
> +
> + idx = code_start;
> +
> + for (sect = 0; sect < numsects; sect++) {
> + unsigned int ssize = sectsize[sect] >> 3;
> +
> + instr = slic_read_dword_from_firmware(fw, &idx);
> + base = sectstart[sect];
> + if (base < 0x8000)
> + continue;
> +
> + for (addr = 0; addr < ssize; addr++) {
> + /* write out instruction address */
> + slic_write(sdev, SLIC_REG_WCS,
> + SLIC_WCS_COMPARE | (base + addr));
> + /* write out instruction to low addr */
> + slic_write(sdev, SLIC_REG_WCS, instr);
> + instr = slic_read_dword_from_firmware(fw, &idx);
> + /* write out instruction to high addr */
> + slic_write(sdev, SLIC_REG_WCS, instr);
> + instr = slic_read_dword_from_firmware(fw, &idx);
> + }
> + }
> + slic_flush_write(sdev);
> + mdelay(10);
> + /* everything OK, kick off the card */
> + slic_write(sdev, SLIC_REG_WCS, SLIC_WCS_START);
> + slic_flush_write(sdev);
> + /* wait long enough for ucode to init card and reach the mainloop */
> + mdelay(20);
> +release:
> + release_firmware(fw);
> +
> + return err;
> +}
[...]
> +static int slic_init_iface(struct slic_device *sdev)
> +{
> + struct slic_shmem *sm = &sdev->shmem;
> + int err;
> +
> + sdev->upr_list.pending = false;
> +
> + err = slic_init_shmem(sdev);
> + if (err) {
> + netdev_err(sdev->netdev, "failed to load firmware\n");
Wrong error message.
> + return err;
> + }
[...]
> +static netdev_tx_t slic_xmit(struct sk_buff *skb, struct net_device *dev)
> +{
> + struct slic_device *sdev = netdev_priv(dev);
> + struct slic_tx_queue *txq = &sdev->txq;
> + struct slic_tx_buffer *buff;
> + struct slic_tx_desc *desc;
> + dma_addr_t paddr;
> + u32 cbar_val;
> + u32 maplen;
> +
> + if (unlikely(slic_get_free_tx_descs(txq) < SLIC_MAX_REQ_TX_DESCS)) {
> + netdev_err(dev, "BUG! not enought tx LEs left: %u\n",
"Enough"?
> + slic_get_free_tx_descs(txq));
> + return NETDEV_TX_BUSY;
> + }
[...]
> +static int slic_read_eeprom(struct slic_device *sdev)
> +{
> + unsigned int devfn = PCI_FUNC(sdev->pdev->devfn);
> + struct slic_shmem *sm = &sdev->shmem;
> + struct slic_shmem_data *sm_data = sm->shmem_data;
> + const unsigned int MAX_LOOPS = 5000;
Another benign -Wsign-compare warning can be fixed by either dropping
the unsigned here or making i below unsigned, too.
> + unsigned int codesize;
> + unsigned char *eeprom;
> + struct slic_upr *upr;
> + dma_addr_t paddr;
> + int err = 0;
> + u8 *mac[2];
> + int i = 0;
> +
> + eeprom = dma_zalloc_coherent(&sdev->pdev->dev, SLIC_EEPROM_SIZE,
> + &paddr, GFP_KERNEL);
> + if (!eeprom)
> + return -ENOMEM;
> +
> + slic_write(sdev, SLIC_REG_ICR, SLIC_ICR_INT_OFF);
> + /* setup ISP temporarily */
> + slic_write(sdev, SLIC_REG_ISP, lower_32_bits(sm->isr_paddr));
> +
> + err = slic_new_upr(sdev, SLIC_UPR_CONFIG, paddr);
> + if (!err) {
> + for (i = 0; i < MAX_LOOPS; i++) {
> + if (le32_to_cpu(sm_data->isr) & SLIC_ISR_UPC)
> + break;
> + mdelay(1);
> + }
> + if (i == MAX_LOOPS) {
> + dev_err(&sdev->pdev->dev,
> + "timed out while waiting for eeprom data\n");
> + err = -ETIMEDOUT;
> + }
> + upr = slic_dequeue_upr(sdev);
> + kfree(upr);
> + }
> +
> + slic_write(sdev, SLIC_REG_ISP, 0);
> + slic_write(sdev, SLIC_REG_ISR, 0);
> + slic_flush_write(sdev);
> +
> + if (err)
> + goto free_eeprom;
> +
> + if (sdev->model == SLIC_MODEL_OASIS) {
> + struct slic_oasis_eeprom *oee;
> +
> + oee = (struct slic_oasis_eeprom *)eeprom;
> + mac[0] = oee->mac;
> + mac[1] = oee->mac2;
> + codesize = le16_to_cpu(oee->eeprom_code_size);
> + } else {
> + struct slic_mojave_eeprom *mee;
> +
> + mee = (struct slic_mojave_eeprom *)eeprom;
> + mac[0] = mee->mac;
> + mac[1] = mee->mac2;
> + codesize = le16_to_cpu(mee->eeprom_code_size);
> + }
> +
> + if (!slic_eeprom_valid(eeprom, codesize)) {
> + dev_err(&sdev->pdev->dev, "invalid checksum in eeprom\n");
> + err = -EINVAL;
> + goto free_eeprom;
> + }
> + /* set mac address */
> + ether_addr_copy(sdev->netdev->dev_addr, mac[devfn]);
> +free_eeprom:
> + dma_free_coherent(&sdev->pdev->dev, SLIC_EEPROM_SIZE, eeprom, paddr);
> +
> + return err;
> +}
[...]
> +static int slic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
> +{
[...]
> + err = register_netdev(dev);
> + if (err) {
> + dev_err(&pdev->dev, "failed to register net device: %i\n",
> + err);
Could be on one line.
Regards,
Markus
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-26 12:20 ` [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver Lino Sanfilippo
2016-11-26 15:48 ` Rami Rosen
2016-11-27 17:59 ` Markus Böhme
@ 2016-11-28 4:56 ` Florian Fainelli
2016-11-28 21:41 ` Lino Sanfilippo
2016-12-04 2:13 ` kbuild test robot
3 siblings, 1 reply; 14+ messages in thread
From: Florian Fainelli @ 2016-11-28 4:56 UTC (permalink / raw)
To: Lino Sanfilippo, davem, charrer, liodot, gregkh, andrew
Cc: devel, linux-kernel, netdev
On 11/26/2016 04:20 AM, Lino Sanfilippo wrote:
> Add driver for Alacritech gigabit ethernet cards with SLIC (session-layer
> interface control) technology. The driver provides basic support without
> SLIC for the following devices:
>
> - Mojave cards (single port PCI Gigabit) both copper and fiber
> - Oasis cards (single and dual port PCI-x Gigabit) copper and fiber
> - Kalahari cards (dual and quad port PCI-e Gigabit) copper and fiber
This looks great, a few nits below:
> +#define SLIC_MAX_TX_COMPLETIONS 100
You usually don't want to limit the number of TX completion, if the
entire TX ring needs to be cleaned, you would want to allow that.
[snip]
> + while (slic_get_free_rx_descs(rxq) > SLIC_MAX_REQ_RX_DESCS) {
> + skb = alloc_skb(maplen + ALIGN_MASK, gfp);
> + if (!skb)
> + break;
> +
> + paddr = dma_map_single(&sdev->pdev->dev, skb->data, maplen,
> + DMA_FROM_DEVICE);
> + if (dma_mapping_error(&sdev->pdev->dev, paddr)) {
> + netdev_err(dev, "mapping rx packet failed\n");
> + /* drop skb */
> + dev_kfree_skb_any(skb);
> + break;
> + }
> + /* ensure head buffer descriptors are 256 byte aligned */
> + offset = 0;
> + misalign = paddr & ALIGN_MASK;
> + if (misalign) {
> + offset = SLIC_RX_BUFF_ALIGN - misalign;
> + skb_reserve(skb, offset);
> + }
> + /* the HW expects dma chunks for descriptor + frame data */
> + desc = (struct slic_rx_desc *)skb->data;
> + memset(desc, 0, sizeof(*desc));
Do you really need to zero-out the prepending RX descriptor? Are not you
missing a write barrier here?
[snip]
> +
> + dma_sync_single_for_cpu(&sdev->pdev->dev,
> + dma_unmap_addr(buff, map_addr),
> + buff->addr_offset + sizeof(*desc),
> + DMA_FROM_DEVICE);
> +
> + status = le32_to_cpu(desc->status);
> + if (!(status & SLIC_IRHDDR_SVALID))
> + break;
> +
> + buff->skb = NULL;
> +
> + dma_unmap_single(&sdev->pdev->dev,
> + dma_unmap_addr(buff, map_addr),
> + dma_unmap_len(buff, map_len),
> + DMA_FROM_DEVICE);
This is potentially inefficient, you already did a cache invalidation
for the RX descriptor here, you could be more efficient with just
invalidating the packet length, minus the descriptor length.
> +
> + /* skip rx descriptor that is placed before the frame data */
> + skb_reserve(skb, SLIC_RX_BUFF_HDR_SIZE);
> +
> + if (unlikely(status & SLIC_IRHDDR_ERR)) {
> + slic_handle_frame_error(sdev, skb);
> + dev_kfree_skb_any(skb);
> + } else {
> + struct ethhdr *eh = (struct ethhdr *)skb->data;
> +
> + if (is_multicast_ether_addr(eh->h_dest))
> + SLIC_INC_STATS_COUNTER(&sdev->stats, rx_mcasts);
> +
> + len = le32_to_cpu(desc->length) & SLIC_IRHDDR_FLEN_MSK;
> + skb_put(skb, len);
> + skb->protocol = eth_type_trans(skb, dev);
> + skb->ip_summed = CHECKSUM_UNNECESSARY;
> + skb->dev = dev;
eth_type_trans() already assigns skb->dev = dev;
> +static int slic_poll(struct napi_struct *napi, int todo)
> +{
> + struct slic_device *sdev = container_of(napi, struct slic_device, napi);
> + struct slic_shmem *sm = &sdev->shmem;
> + struct slic_shmem_data *sm_data = sm->shmem_data;
> + u32 isr = le32_to_cpu(sm_data->isr);
> + unsigned int done = 0;
> +
> + slic_handle_irq(sdev, isr, todo, &done);
> +
> + if (done < todo) {
> + napi_complete(napi);
napi_complete_done() since you know how many packets you completed.
> + /* reenable irqs */
> + sm_data->isr = 0;
> + /* make sure sm_data->isr is cleard before irqs are reenabled */
> + wmb();
> + slic_write(sdev, SLIC_REG_ISR, 0);
> + slic_flush_write(sdev);
> + }
> +
> + return done;
> +}
> +
> +static irqreturn_t slic_irq(int irq, void *dev_id)
> +{
> + struct slic_device *sdev = dev_id;
> + struct slic_shmem *sm = &sdev->shmem;
> + struct slic_shmem_data *sm_data = sm->shmem_data;
> +
> + slic_write(sdev, SLIC_REG_ICR, SLIC_ICR_INT_MASK);
> + slic_flush_write(sdev);
> + /* make sure sm_data->isr is read after ICR_INT_MASK is set */
> + wmb();
> +
> + if (!sm_data->isr) {
> + dma_rmb();
> + /* spurious interrupt */
> + slic_write(sdev, SLIC_REG_ISR, 0);
> + slic_flush_write(sdev);
> + return IRQ_NONE;
> + }
> +
> + napi_schedule(&sdev->napi);
Likewise napi_schedule_irqoff() can be used here.
--
Florian
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-27 17:59 ` Markus Böhme
@ 2016-11-28 20:46 ` Lino Sanfilippo
2016-11-29 12:15 ` Markus Böhme
0 siblings, 1 reply; 14+ messages in thread
From: Lino Sanfilippo @ 2016-11-28 20:46 UTC (permalink / raw)
To: Markus Böhme, davem, charrer, liodot, gregkh, andrew
Cc: devel, netdev, linux-kernel
Hi Markus,
On 27.11.2016 18:59, Markus Böhme wrote:
> Hello Lino,
>
> just some things barely worth mentioning:
>
>
> I found a bunch of unused #defines in slic.h. I cannot judge if they are
> worth keeping:
>
> SLIC_VRHSTATB_LONGE
> SLIC_VRHSTATB_PREA
> SLIC_ISR_IO
> SLIC_ISR_PING_MASK
> SLIC_GIG_SPEED_MASK
> SLIC_GMCR_RESET
> SLIC_XCR_RESET
> SLIC_XCR_XMTEN
> SLIC_XCR_PAUSEEN
> SLIC_XCR_LOADRNG
> SLIC_REG_DBAR
> SLIC_REG_PING
> SLIC_REG_DUMP_CMD
> SLIC_REG_DUMP_DATA
> SLIC_REG_WRHOSTID
> SLIC_REG_LOW_POWER
> SLIC_REG_RESET_IFACE
> SLIC_REG_ADDR_UPPER
> SLIC_REG_HBAR64
> SLIC_REG_DBAR64
> SLIC_REG_CBAR64
> SLIC_REG_RBAR64
> SLIC_REG_WRVLANID
> SLIC_REG_READ_XF_INFO
> SLIC_REG_WRITE_XF_INFO
> SLIC_REG_TICKS_PER_SEC
>
> These device IDs are not used, either, but maybe it's good to keep them
> for documentation purposes:
>
> PCI_SUBDEVICE_ID_ALACRITECH_1000X1_2
> PCI_SUBDEVICE_ID_ALACRITECH_SES1001T
> PCI_SUBDEVICE_ID_ALACRITECH_SEN2002XT
> PCI_SUBDEVICE_ID_ALACRITECH_SEN2001XT
> PCI_SUBDEVICE_ID_ALACRITECH_SEN2104ET
> PCI_SUBDEVICE_ID_ALACRITECH_SEN2102ET
>
I left these defines in for both documentation and to avoid gaps in
register ranges. I would like to keep this as it is.
>> +
>> +/* SLIC EEPROM structure for Oasis */
>> +struct slic_mojave_eeprom {
>
> Comment: "for Mojave".
Will fix, thanks,
>
> [...]
>
>> +struct slic_device {
>> + struct pci_dev *pdev;
>> + struct net_device *netdev;
>> + void __iomem *regs;
>> + /* upper address setting lock */
>> + spinlock_t upper_lock;
>> + struct slic_shmem shmem;
>> + struct napi_struct napi;
>> + struct slic_rx_queue rxq;
>> + struct slic_tx_queue txq;
>> + struct slic_stat_queue stq;
>> + struct slic_stats stats;
>> + struct slic_upr_list upr_list;
>> + /* link configuration lock */
>> + spinlock_t link_lock;
>> + bool promisc;
>> + bool autoneg;
>> + int speed;
>> + int duplex;
>
> Maybe make speed and duplex unsigned? They are assigned and compared
> against unsigned values in slicoss.c, so this would get rid of some
> (benign, because of the range of the values) -Wsign-compare warnings in
> slic_configure_link_locked. However, in a comparison there SPEED_UNKNOWN
> would need to be casted to unsigned to prevent another one popping up.
>
There is indeed a bunch of warnings concerning signedness. Will have a look
at all of them. However I think I will keep "speed" as an int, because casting
SPEED_UNKNOWN to an unsigned int is IMHO an ugly thing to do.
> [...]
>
>> +#endif /* _SLIC_H */
>> diff --git a/drivers/net/ethernet/alacritech/slicoss.c b/drivers/net/ethernet/alacritech/slicoss.c
>> new file mode 100644
>> index 0000000..8cd862a
>> --- /dev/null
>> +++ b/drivers/net/ethernet/alacritech/slicoss.c
>> @@ -0,0 +1,1867 @@
>
> [...]
>
>> +
>> +static const struct pci_device_id slic_id_tbl[] = {
>> + { PCI_DEVICE(PCI_VENDOR_ID_ALACRITECH,
>> + PCI_DEVICE_ID_ALACRITECH_MOAVE) },
>
> I missed this in slic.h, but is this a typo and "MOAVE" should be
> "MOJAVE"? There are a couple similar #defines in slic.h.
This should definitely be "Mojave". Will fix it.
>
> [...]
>
>> +static void slic_refill_rx_queue(struct slic_device *sdev, gfp_t gfp)
>> +{
>> + const unsigned int ALIGN_MASK = SLIC_RX_BUFF_ALIGN - 1;
>> + unsigned int maplen = SLIC_RX_BUFF_SIZE;
>> + struct slic_rx_queue *rxq = &sdev->rxq;
>> + struct net_device *dev = sdev->netdev;
>> + struct slic_rx_buffer *buff;
>> + struct slic_rx_desc *desc;
>> + unsigned int misalign;
>> + unsigned int offset;
>> + struct sk_buff *skb;
>> + dma_addr_t paddr;
>> +
>> + while (slic_get_free_rx_descs(rxq) > SLIC_MAX_REQ_RX_DESCS) {
>> + skb = alloc_skb(maplen + ALIGN_MASK, gfp);
>> + if (!skb)
>> + break;
>> +
>> + paddr = dma_map_single(&sdev->pdev->dev, skb->data, maplen,
>> + DMA_FROM_DEVICE);
>> + if (dma_mapping_error(&sdev->pdev->dev, paddr)) {
>> + netdev_err(dev, "mapping rx packet failed\n");
>> + /* drop skb */
>> + dev_kfree_skb_any(skb);
>> + break;
>> + }
>> + /* ensure head buffer descriptors are 256 byte aligned */
>> + offset = 0;
>> + misalign = paddr & ALIGN_MASK;
>> + if (misalign) {
>> + offset = SLIC_RX_BUFF_ALIGN - misalign;
>> + skb_reserve(skb, offset);
>> + }
>> + /* the HW expects dma chunks for descriptor + frame data */
>> + desc = (struct slic_rx_desc *)skb->data;
>> + memset(desc, 0, sizeof(*desc));
>> +
>> + buff = &rxq->rxbuffs[rxq->put_idx];
>> + buff->skb = skb;
>> + dma_unmap_addr_set(buff, map_addr, paddr);
>> + dma_unmap_len_set(buff, map_len, maplen);
>> + buff->addr_offset = offset;
>> + /* head buffer descriptors are placed immediately before skb */
>> + slic_write(sdev, SLIC_REG_HBAR, lower_32_bits(paddr) +
>> + offset);
>
> This fits nicely on one line. :-)
Right, will fix.
>
> [...]
>
>> +static int slic_init_tx_queue(struct slic_device *sdev)
>> +{
>> + struct slic_tx_queue *txq = &sdev->txq;
>> + struct slic_tx_buffer *buff;
>> + struct slic_tx_desc *desc;
>> + int err;
>> + int i;
>
> You could make i unsigned...
>
>> +
>> + txq->len = SLIC_NUM_TX_DESCS;
>> + txq->put_idx = 0;
>> + txq->done_idx = 0;
>> +
>> + txq->txbuffs = kcalloc(txq->len, sizeof(*buff), GFP_KERNEL);
>> + if (!txq->txbuffs)
>> + return -ENOMEM;
>> +
>> + txq->dma_pool = dma_pool_create("slic_pool", &sdev->pdev->dev,
>> + sizeof(*desc), SLIC_TX_DESC_ALIGN,
>> + 4096);
>> + if (!txq->dma_pool) {
>> + err = -ENOMEM;
>> + netdev_err(sdev->netdev, "failed to create dma pool\n");
>> + goto free_buffs;
>> + }
>> +
>> + for (i = 0; i < txq->len; i++) {
>
> ...to fix a signed/unsigned comparison warning here, but...
>
>> + buff = &txq->txbuffs[i];
>> + desc = dma_pool_zalloc(txq->dma_pool, GFP_KERNEL,
>> + &buff->desc_paddr);
>> + if (!desc) {
>> + netdev_err(sdev->netdev,
>> + "failed to alloc pool chunk (%i)\n", i);
>> + err = -ENOMEM;
>> + goto free_descs;
>> + }
>> +
>> + desc->hnd = cpu_to_le32((u32)(i + 1));
>> + desc->cmd = SLIC_CMD_XMT_REQ;
>> + desc->flags = 0;
>> + desc->type = cpu_to_le32(SLIC_CMD_TYPE_DUMB);
>> + buff->desc = desc;
>> + }
>> +
>> + return 0;
>> +
>> +free_descs:
>> + while (i--) {
>
> ...this would require reworking this logic to prevent an endless loop,
> so probably not worth bothering, considering that txq->len is well
> within the positive signed range.
AFAICS the logic does not have to be changed. The while loop will also work
fine if "i" is unsigned.
>
>> + buff = &txq->txbuffs[i];
>> + dma_pool_free(txq->dma_pool, buff->desc, buff->desc_paddr);
>> + }
>> + dma_pool_destroy(txq->dma_pool);
>> +
>> +free_buffs:
>> + kfree(txq->txbuffs);
>> +
>> + return err;
>> +}
>> +
>> +static void slic_free_tx_queue(struct slic_device *sdev)
>> +{
>> + struct slic_tx_queue *txq = &sdev->txq;
>> + struct slic_tx_buffer *buff;
>> + int i;
>
> Make i unsigned? One warning less, almost no work invested.
>
>> +
>> + for (i = 0; i < txq->len; i++) {
>> + buff = &txq->txbuffs[i];
>> + dma_pool_free(txq->dma_pool, buff->desc, buff->desc_paddr);
>> + if (!buff->skb)
>> + continue;
>> +
>> + dma_unmap_single(&sdev->pdev->dev,
>> + dma_unmap_addr(buff, map_addr),
>> + dma_unmap_len(buff, map_len), DMA_TO_DEVICE);
>> + consume_skb(buff->skb);
>> + }
>> + dma_pool_destroy(txq->dma_pool);
>> +
>> + kfree(txq->txbuffs);
>> +}
>> +
>
> [...]
>
>> +static void slic_free_rx_queue(struct slic_device *sdev)
>> +{
>> + struct slic_rx_queue *rxq = &sdev->rxq;
>> + struct slic_rx_buffer *buff;
>> + int i;
>
> Unsigned?
>
>> +
>> + /* free rx buffers */
>> + for (i = 0; i < rxq->len; i++) {
>> + buff = &rxq->rxbuffs[i];
>> +
>> + if (!buff->skb)
>> + continue;
>> +
>> + dma_unmap_single(&sdev->pdev->dev,
>> + dma_unmap_addr(buff, map_addr),
>> + dma_unmap_len(buff, map_len),
>> + DMA_FROM_DEVICE);
>> + consume_skb(buff->skb);
>> + }
>> + kfree(rxq->rxbuffs);
>> +}
>
> [...]
>
>> +static int slic_load_firmware(struct slic_device *sdev)
>> +{
>> + u32 sectstart[SLIC_FIRMWARE_MAX_SECTIONS];
>> + u32 sectsize[SLIC_FIRMWARE_MAX_SECTIONS];
>> + const struct firmware *fw;
>> + unsigned int datalen;
>> + const char *file;
>> + int code_start;
>> + u32 numsects;
>> + int idx = 0;
>> + u32 sect;
>> + u32 instr;
>> + u32 addr;
>> + u32 base;
>> + int err;
>> + int i;
>
> Make i unsigned?
>
>> +
>> + file = (sdev->model == SLIC_MODEL_OASIS) ? SLIC_FIRMWARE_OASIS :
>> + SLIC_FIRMWARE_MOAVE;
>> + err = request_firmware(&fw, file, &sdev->pdev->dev);
>> + if (err) {
>> + dev_err(&sdev->pdev->dev, "failed to load firmware %s\n", file);
>> + return err;
>> + }
>> + /* Do an initial sanity check concerning firmware size now. A further
>> + * check follows below.
>> + */
>> + if (fw->size < SLIC_FIRMWARE_MIN_SIZE) {
>> + dev_err(&sdev->pdev->dev,
>> + "invalid firmware size %zu (min is %u)\n", fw->size,
>> + SLIC_FIRMWARE_MIN_SIZE);
>> + err = -EINVAL;
>> + goto release;
>> + }
>> +
>> + numsects = slic_read_dword_from_firmware(fw, &idx);
>> + if (numsects == 0 || numsects > SLIC_FIRMWARE_MAX_SECTIONS) {
>> + dev_err(&sdev->pdev->dev,
>> + "invalid number of sections in firmware: %u", numsects);
>> + err = -EINVAL;
>> + goto release;
>> + }
>> +
>> + datalen = numsects * 8 + 4;
>> + for (i = 0; i < numsects; i++) {
>> + sectsize[i] = slic_read_dword_from_firmware(fw, &idx);
>> + datalen += sectsize[i];
>> + }
>> +
>> + /* do another sanity check against firmware size */
>> + if (datalen > fw->size) {
>> + dev_err(&sdev->pdev->dev,
>> + "invalid firmware size %zu (expected >= %u)\n",
>> + fw->size, datalen);
>> + err = -EINVAL;
>> + goto release;
>> + }
>> + /* get sections */
>> + for (i = 0; i < numsects; i++)
>> + sectstart[i] = slic_read_dword_from_firmware(fw, &idx);
>> +
>> + code_start = idx;
>> + instr = slic_read_dword_from_firmware(fw, &idx);
>> +
>> + for (sect = 0; sect < numsects; sect++) {
>> + unsigned int ssize = sectsize[sect] >> 3;
>> +
>> + base = sectstart[sect];
>> +
>> + for (addr = 0; addr < ssize; addr++) {
>> + /* write out instruction address */
>> + slic_write(sdev, SLIC_REG_WCS, base + addr);
>> + /* write out instruction to low addr */
>> + slic_write(sdev, SLIC_REG_WCS, instr);
>> + instr = slic_read_dword_from_firmware(fw, &idx);
>> + /* write out instruction to high addr */
>> + slic_write(sdev, SLIC_REG_WCS, instr);
>> + instr = slic_read_dword_from_firmware(fw, &idx);
>> + }
>> + }
>> +
>> + idx = code_start;
>> +
>> + for (sect = 0; sect < numsects; sect++) {
>> + unsigned int ssize = sectsize[sect] >> 3;
>> +
>> + instr = slic_read_dword_from_firmware(fw, &idx);
>> + base = sectstart[sect];
>> + if (base < 0x8000)
>> + continue;
>> +
>> + for (addr = 0; addr < ssize; addr++) {
>> + /* write out instruction address */
>> + slic_write(sdev, SLIC_REG_WCS,
>> + SLIC_WCS_COMPARE | (base + addr));
>> + /* write out instruction to low addr */
>> + slic_write(sdev, SLIC_REG_WCS, instr);
>> + instr = slic_read_dword_from_firmware(fw, &idx);
>> + /* write out instruction to high addr */
>> + slic_write(sdev, SLIC_REG_WCS, instr);
>> + instr = slic_read_dword_from_firmware(fw, &idx);
>> + }
>> + }
>> + slic_flush_write(sdev);
>> + mdelay(10);
>> + /* everything OK, kick off the card */
>> + slic_write(sdev, SLIC_REG_WCS, SLIC_WCS_START);
>> + slic_flush_write(sdev);
>> + /* wait long enough for ucode to init card and reach the mainloop */
>> + mdelay(20);
>> +release:
>> + release_firmware(fw);
>> +
>> + return err;
>> +}
>
> [...]
>
>> +static int slic_init_iface(struct slic_device *sdev)
>> +{
>> + struct slic_shmem *sm = &sdev->shmem;
>> + int err;
>> +
>> + sdev->upr_list.pending = false;
>> +
>> + err = slic_init_shmem(sdev);
>> + if (err) {
>> + netdev_err(sdev->netdev, "failed to load firmware\n");
>
> Wrong error message.
Yep, will fix.
>
>> + return err;
>> + }
>
> [...]
>
>> +static netdev_tx_t slic_xmit(struct sk_buff *skb, struct net_device *dev)
>> +{
>> + struct slic_device *sdev = netdev_priv(dev);
>> + struct slic_tx_queue *txq = &sdev->txq;
>> + struct slic_tx_buffer *buff;
>> + struct slic_tx_desc *desc;
>> + dma_addr_t paddr;
>> + u32 cbar_val;
>> + u32 maplen;
>> +
>> + if (unlikely(slic_get_free_tx_descs(txq) < SLIC_MAX_REQ_TX_DESCS)) {
>> + netdev_err(dev, "BUG! not enought tx LEs left: %u\n",
>
> "Enough"?
Will fix.
>> + slic_get_free_tx_descs(txq));
>> + return NETDEV_TX_BUSY;
>> + }
>
> [...]
>
>> +static int slic_read_eeprom(struct slic_device *sdev)
>> +{
>> + unsigned int devfn = PCI_FUNC(sdev->pdev->devfn);
>> + struct slic_shmem *sm = &sdev->shmem;
>> + struct slic_shmem_data *sm_data = sm->shmem_data;
>> + const unsigned int MAX_LOOPS = 5000;
>
> Another benign -Wsign-compare warning can be fixed by either dropping
> the unsigned here or making i below unsigned, too.
>
>> + unsigned int codesize;
>> + unsigned char *eeprom;
>> + struct slic_upr *upr;
>> + dma_addr_t paddr;
>> + int err = 0;
>> + u8 *mac[2];
>> + int i = 0;
>> +
>> + eeprom = dma_zalloc_coherent(&sdev->pdev->dev, SLIC_EEPROM_SIZE,
>> + &paddr, GFP_KERNEL);
>> + if (!eeprom)
>> + return -ENOMEM;
>> +
>> + slic_write(sdev, SLIC_REG_ICR, SLIC_ICR_INT_OFF);
>> + /* setup ISP temporarily */
>> + slic_write(sdev, SLIC_REG_ISP, lower_32_bits(sm->isr_paddr));
>> +
>> + err = slic_new_upr(sdev, SLIC_UPR_CONFIG, paddr);
>> + if (!err) {
>> + for (i = 0; i < MAX_LOOPS; i++) {
>> + if (le32_to_cpu(sm_data->isr) & SLIC_ISR_UPC)
>> + break;
>> + mdelay(1);
>> + }
>> + if (i == MAX_LOOPS) {
>> + dev_err(&sdev->pdev->dev,
>> + "timed out while waiting for eeprom data\n");
>> + err = -ETIMEDOUT;
>> + }
>> + upr = slic_dequeue_upr(sdev);
>> + kfree(upr);
>> + }
>> +
>> + slic_write(sdev, SLIC_REG_ISP, 0);
>> + slic_write(sdev, SLIC_REG_ISR, 0);
>> + slic_flush_write(sdev);
>> +
>> + if (err)
>> + goto free_eeprom;
>> +
>> + if (sdev->model == SLIC_MODEL_OASIS) {
>> + struct slic_oasis_eeprom *oee;
>> +
>> + oee = (struct slic_oasis_eeprom *)eeprom;
>> + mac[0] = oee->mac;
>> + mac[1] = oee->mac2;
>> + codesize = le16_to_cpu(oee->eeprom_code_size);
>> + } else {
>> + struct slic_mojave_eeprom *mee;
>> +
>> + mee = (struct slic_mojave_eeprom *)eeprom;
>> + mac[0] = mee->mac;
>> + mac[1] = mee->mac2;
>> + codesize = le16_to_cpu(mee->eeprom_code_size);
>> + }
>> +
>> + if (!slic_eeprom_valid(eeprom, codesize)) {
>> + dev_err(&sdev->pdev->dev, "invalid checksum in eeprom\n");
>> + err = -EINVAL;
>> + goto free_eeprom;
>> + }
>> + /* set mac address */
>> + ether_addr_copy(sdev->netdev->dev_addr, mac[devfn]);
>> +free_eeprom:
>> + dma_free_coherent(&sdev->pdev->dev, SLIC_EEPROM_SIZE, eeprom, paddr);
>> +
>> + return err;
>> +}
>
> [...]
>
>> +static int slic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>> +{
>
> [...]
>
>> + err = register_netdev(dev);
>> + if (err) {
>> + dev_err(&pdev->dev, "failed to register net device: %i\n",
>> + err);
>
> Could be on one line.
Right, will adjust it.
> Regards,
> Markus
>
Thanks Markus!
Regards,
Lino
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-28 4:56 ` Florian Fainelli
@ 2016-11-28 21:41 ` Lino Sanfilippo
2016-11-29 17:14 ` Florian Fainelli
0 siblings, 1 reply; 14+ messages in thread
From: Lino Sanfilippo @ 2016-11-28 21:41 UTC (permalink / raw)
To: Florian Fainelli, davem, charrer, liodot, gregkh, andrew
Cc: devel, linux-kernel, netdev
Hi Florian,
On 28.11.2016 05:56, Florian Fainelli wrote:
> On 11/26/2016 04:20 AM, Lino Sanfilippo wrote:
>> Add driver for Alacritech gigabit ethernet cards with SLIC (session-layer
>> interface control) technology. The driver provides basic support without
>> SLIC for the following devices:
>>
>> - Mojave cards (single port PCI Gigabit) both copper and fiber
>> - Oasis cards (single and dual port PCI-x Gigabit) copper and fiber
>> - Kalahari cards (dual and quad port PCI-e Gigabit) copper and fiber
>
> This looks great, a few nits below:
>
>
>> +#define SLIC_MAX_TX_COMPLETIONS 100
>
> You usually don't want to limit the number of TX completion, if the
> entire TX ring needs to be cleaned, you would want to allow that.
>
The problem is that the HW does not provide a tx completion index. Instead we have to
iterate the status descriptors until we get an invalid idx which indicates that there
are no further tx descriptors done for now. I am afraid that if we do not limit the
number of descriptors processed in the tx completion handler, a continuous transmission
of frames could keep the loop in xmit_complete() run endlessly. I dont know if this
can actually happen but I wanted to make sure that this is avoided.
> [snip]
>
>> + while (slic_get_free_rx_descs(rxq) > SLIC_MAX_REQ_RX_DESCS) {
>> + skb = alloc_skb(maplen + ALIGN_MASK, gfp);
>> + if (!skb)
>> + break;
>> +
>> + paddr = dma_map_single(&sdev->pdev->dev, skb->data, maplen,
>> + DMA_FROM_DEVICE);
>> + if (dma_mapping_error(&sdev->pdev->dev, paddr)) {
>> + netdev_err(dev, "mapping rx packet failed\n");
>> + /* drop skb */
>> + dev_kfree_skb_any(skb);
>> + break;
>> + }
>> + /* ensure head buffer descriptors are 256 byte aligned */
>> + offset = 0;
>> + misalign = paddr & ALIGN_MASK;
>> + if (misalign) {
>> + offset = SLIC_RX_BUFF_ALIGN - misalign;
>> + skb_reserve(skb, offset);
>> + }
>> + /* the HW expects dma chunks for descriptor + frame data */
>> + desc = (struct slic_rx_desc *)skb->data;
>> + memset(desc, 0, sizeof(*desc));
>
> Do you really need to zero-out the prepending RX descriptor? Are not you
> missing a write barrier here?
Indeed, it should be sufficient to make sure that the bit SLIC_IRHDDR_SVALID is not set.
I will adjust it.
Concerning the write barrier: You mean a wmb() before slic_write() to ensure that the zeroing
of the status desc is done before the descriptor is passed to the HW, right?
> [snip]
>
>> +
>> + dma_sync_single_for_cpu(&sdev->pdev->dev,
>> + dma_unmap_addr(buff, map_addr),
>> + buff->addr_offset + sizeof(*desc),
>> + DMA_FROM_DEVICE);
>> +
>> + status = le32_to_cpu(desc->status);
>> + if (!(status & SLIC_IRHDDR_SVALID))
>> + break;
>> +
>> + buff->skb = NULL;
>> +
>> + dma_unmap_single(&sdev->pdev->dev,
>> + dma_unmap_addr(buff, map_addr),
>> + dma_unmap_len(buff, map_len),
>> + DMA_FROM_DEVICE);
>
> This is potentially inefficient, you already did a cache invalidation
> for the RX descriptor here, you could be more efficient with just
> invalidating the packet length, minus the descriptor length.
>
I am not sure I understand: We have to unmap the complete dma area, no matter if we synced
part of it before, dont we? AFAIK a dma sync is different from unmapping dma, or do I miss
something?
>> +
>> + /* skip rx descriptor that is placed before the frame data */
>> + skb_reserve(skb, SLIC_RX_BUFF_HDR_SIZE);
>> +
>> + if (unlikely(status & SLIC_IRHDDR_ERR)) {
>> + slic_handle_frame_error(sdev, skb);
>> + dev_kfree_skb_any(skb);
>> + } else {
>> + struct ethhdr *eh = (struct ethhdr *)skb->data;
>> +
>> + if (is_multicast_ether_addr(eh->h_dest))
>> + SLIC_INC_STATS_COUNTER(&sdev->stats, rx_mcasts);
>> +
>> + len = le32_to_cpu(desc->length) & SLIC_IRHDDR_FLEN_MSK;
>> + skb_put(skb, len);
>> + skb->protocol = eth_type_trans(skb, dev);
>> + skb->ip_summed = CHECKSUM_UNNECESSARY;
>> + skb->dev = dev;
>
> eth_type_trans() already assigns skb->dev = dev;
>
Right, this is unnecessary, I will fix it.
>> +static int slic_poll(struct napi_struct *napi, int todo)
>> +{
>> + struct slic_device *sdev = container_of(napi, struct slic_device, napi);
>> + struct slic_shmem *sm = &sdev->shmem;
>> + struct slic_shmem_data *sm_data = sm->shmem_data;
>> + u32 isr = le32_to_cpu(sm_data->isr);
>> + unsigned int done = 0;
>> +
>> + slic_handle_irq(sdev, isr, todo, &done);
>> +
>> + if (done < todo) {
>> + napi_complete(napi);
>
> napi_complete_done() since you know how many packets you completed.
>
Ok, will change it.
>> + /* reenable irqs */
>> + sm_data->isr = 0;
>> + /* make sure sm_data->isr is cleard before irqs are reenabled */
>> + wmb();
>> + slic_write(sdev, SLIC_REG_ISR, 0);
>> + slic_flush_write(sdev);
>> + }
>> +
>> + return done;
>> +}
>> +
>> +static irqreturn_t slic_irq(int irq, void *dev_id)
>> +{
>> + struct slic_device *sdev = dev_id;
>> + struct slic_shmem *sm = &sdev->shmem;
>> + struct slic_shmem_data *sm_data = sm->shmem_data;
>> +
>> + slic_write(sdev, SLIC_REG_ICR, SLIC_ICR_INT_MASK);
>> + slic_flush_write(sdev);
>> + /* make sure sm_data->isr is read after ICR_INT_MASK is set */
>> + wmb();
>> +
>> + if (!sm_data->isr) {
>> + dma_rmb();
>> + /* spurious interrupt */
>> + slic_write(sdev, SLIC_REG_ISR, 0);
>> + slic_flush_write(sdev);
>> + return IRQ_NONE;
>> + }
>> +
>> + napi_schedule(&sdev->napi);
>
> Likewise napi_schedule_irqoff() can be used here.
>
Ok, will change it.
Thanks a lot Florian!
Regards,
Lino
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-28 20:46 ` Lino Sanfilippo
@ 2016-11-29 12:15 ` Markus Böhme
0 siblings, 0 replies; 14+ messages in thread
From: Markus Böhme @ 2016-11-29 12:15 UTC (permalink / raw)
To: Lino Sanfilippo, davem, charrer, liodot, gregkh, andrew
Cc: devel, netdev, linux-kernel
On 11/28/2016 09:46 PM, Lino Sanfilippo wrote:
> Hi Markus,
>
> On 27.11.2016 18:59, Markus Böhme wrote:
>> Hello Lino,
>>
>> just some things barely worth mentioning:
>>
>
>>
>> I found a bunch of unused #defines in slic.h. I cannot judge if they are
>> worth keeping:
>>
>> SLIC_VRHSTATB_LONGE
>> SLIC_VRHSTATB_PREA
>> SLIC_ISR_IO
>> SLIC_ISR_PING_MASK
>> SLIC_GIG_SPEED_MASK
>> SLIC_GMCR_RESET
>> SLIC_XCR_RESET
>> SLIC_XCR_XMTEN
>> SLIC_XCR_PAUSEEN
>> SLIC_XCR_LOADRNG
>> SLIC_REG_DBAR
>> SLIC_REG_PING
>> SLIC_REG_DUMP_CMD
>> SLIC_REG_DUMP_DATA
>> SLIC_REG_WRHOSTID
>> SLIC_REG_LOW_POWER
>> SLIC_REG_RESET_IFACE
>> SLIC_REG_ADDR_UPPER
>> SLIC_REG_HBAR64
>> SLIC_REG_DBAR64
>> SLIC_REG_CBAR64
>> SLIC_REG_RBAR64
>> SLIC_REG_WRVLANID
>> SLIC_REG_READ_XF_INFO
>> SLIC_REG_WRITE_XF_INFO
>> SLIC_REG_TICKS_PER_SEC
>>
>> These device IDs are not used, either, but maybe it's good to keep them
>> for documentation purposes:
>>
>> PCI_SUBDEVICE_ID_ALACRITECH_1000X1_2
>> PCI_SUBDEVICE_ID_ALACRITECH_SES1001T
>> PCI_SUBDEVICE_ID_ALACRITECH_SEN2002XT
>> PCI_SUBDEVICE_ID_ALACRITECH_SEN2001XT
>> PCI_SUBDEVICE_ID_ALACRITECH_SEN2104ET
>> PCI_SUBDEVICE_ID_ALACRITECH_SEN2102ET
>>
>
> I left these defines in for both documentation and to avoid gaps in
> register ranges. I would like to keep this as it is.
Seems reasonable.
[...]
>>> +static int slic_init_tx_queue(struct slic_device *sdev)
>>> +{
>>> + struct slic_tx_queue *txq = &sdev->txq;
>>> + struct slic_tx_buffer *buff;
>>> + struct slic_tx_desc *desc;
>>> + int err;
>>> + int i;
>>
>> You could make i unsigned...
>>
>
>>> +
>>> + txq->len = SLIC_NUM_TX_DESCS;
>>> + txq->put_idx = 0;
>>> + txq->done_idx = 0;
>>> +
>>> + txq->txbuffs = kcalloc(txq->len, sizeof(*buff), GFP_KERNEL);
>>> + if (!txq->txbuffs)
>>> + return -ENOMEM;
>>> +
>>> + txq->dma_pool = dma_pool_create("slic_pool", &sdev->pdev->dev,
>>> + sizeof(*desc), SLIC_TX_DESC_ALIGN,
>>> + 4096);
>>> + if (!txq->dma_pool) {
>>> + err = -ENOMEM;
>>> + netdev_err(sdev->netdev, "failed to create dma pool\n");
>>> + goto free_buffs;
>>> + }
>>> +
>>> + for (i = 0; i < txq->len; i++) {
>>
>> ...to fix a signed/unsigned comparison warning here, but...
>>
>>> + buff = &txq->txbuffs[i];
>>> + desc = dma_pool_zalloc(txq->dma_pool, GFP_KERNEL,
>>> + &buff->desc_paddr);
>>> + if (!desc) {
>>> + netdev_err(sdev->netdev,
>>> + "failed to alloc pool chunk (%i)\n", i);
>>> + err = -ENOMEM;
>>> + goto free_descs;
>>> + }
>>> +
>>> + desc->hnd = cpu_to_le32((u32)(i + 1));
>>> + desc->cmd = SLIC_CMD_XMT_REQ;
>>> + desc->flags = 0;
>>> + desc->type = cpu_to_le32(SLIC_CMD_TYPE_DUMB);
>>> + buff->desc = desc;
>>> + }
>>> +
>>> + return 0;
>>> +
>>> +free_descs:
>>> + while (i--) {
>>
>> ...this would require reworking this logic to prevent an endless loop,
>> so probably not worth bothering, considering that txq->len is well
>> within the positive signed range.
>
> AFAICS the logic does not have to be changed. The while loop will also work
> fine if "i" is unsigned.
>
My bad! Of course you are right.
Regards,
Markus
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-28 21:41 ` Lino Sanfilippo
@ 2016-11-29 17:14 ` Florian Fainelli
2016-11-30 19:48 ` Lino Sanfilippo
0 siblings, 1 reply; 14+ messages in thread
From: Florian Fainelli @ 2016-11-29 17:14 UTC (permalink / raw)
To: Lino Sanfilippo, davem, charrer, liodot, gregkh, andrew
Cc: devel, linux-kernel, netdev
On 11/28/2016 01:41 PM, Lino Sanfilippo wrote:
> The problem is that the HW does not provide a tx completion index. Instead we have to
> iterate the status descriptors until we get an invalid idx which indicates that there
> are no further tx descriptors done for now. I am afraid that if we do not limit the
> number of descriptors processed in the tx completion handler, a continuous transmission
> of frames could keep the loop in xmit_complete() run endlessly. I dont know if this
> can actually happen but I wanted to make sure that this is avoided.
OK, it might be a good idea to put that comment somewhere around the tx
completion handler to understand why it is bounded with a specific value.
>
>> [snip]
>>
>>> + while (slic_get_free_rx_descs(rxq) > SLIC_MAX_REQ_RX_DESCS) {
>>> + skb = alloc_skb(maplen + ALIGN_MASK, gfp);
>>> + if (!skb)
>>> + break;
>>> +
>>> + paddr = dma_map_single(&sdev->pdev->dev, skb->data, maplen,
>>> + DMA_FROM_DEVICE);
>>> + if (dma_mapping_error(&sdev->pdev->dev, paddr)) {
>>> + netdev_err(dev, "mapping rx packet failed\n");
>>> + /* drop skb */
>>> + dev_kfree_skb_any(skb);
>>> + break;
>>> + }
>>> + /* ensure head buffer descriptors are 256 byte aligned */
>>> + offset = 0;
>>> + misalign = paddr & ALIGN_MASK;
>>> + if (misalign) {
>>> + offset = SLIC_RX_BUFF_ALIGN - misalign;
>>> + skb_reserve(skb, offset);
>>> + }
>>> + /* the HW expects dma chunks for descriptor + frame data */
>>> + desc = (struct slic_rx_desc *)skb->data;
>>> + memset(desc, 0, sizeof(*desc));
>>
>> Do you really need to zero-out the prepending RX descriptor? Are not you
>> missing a write barrier here?
>
> Indeed, it should be sufficient to make sure that the bit SLIC_IRHDDR_SVALID is not set.
> I will adjust it.
> Concerning the write barrier: You mean a wmb() before slic_write() to ensure that the zeroing
> of the status desc is done before the descriptor is passed to the HW, right?
Correct, that's what I meant here.
>
>
>> [snip]
>>
>>> +
>>> + dma_sync_single_for_cpu(&sdev->pdev->dev,
>>> + dma_unmap_addr(buff, map_addr),
>>> + buff->addr_offset + sizeof(*desc),
>>> + DMA_FROM_DEVICE);
>>> +
>>> + status = le32_to_cpu(desc->status);
>>> + if (!(status & SLIC_IRHDDR_SVALID))
>>> + break;
>>> +
>>> + buff->skb = NULL;
>>> +
>>> + dma_unmap_single(&sdev->pdev->dev,
>>> + dma_unmap_addr(buff, map_addr),
>>> + dma_unmap_len(buff, map_len),
>>> + DMA_FROM_DEVICE);
>>
>> This is potentially inefficient, you already did a cache invalidation
>> for the RX descriptor here, you could be more efficient with just
>> invalidating the packet length, minus the descriptor length.
>>
>
> I am not sure I understand: We have to unmap the complete dma area, no matter if we synced
> part of it before, dont we? AFAIK a dma sync is different from unmapping dma, or do I miss
> something?
Sorry, I was not very clear, what I meant is that you can allocate and
do the initial dma_map_single() of your RX skbs during ndo_open(), and
then, in your RX path, you can only do dma_sync_single_for_cpu() twice
(once for the RX descriptor status, second time for the actual packet
contents), and when you return the SKB to the HW, do a
dma_sync_single_for_device(). The advantage of doing that, is that if
your cache operations are slow, you only do them on exactly packet
length, and not the actual RX buffer size (e.g: 2KB).
--
Florian
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-29 17:14 ` Florian Fainelli
@ 2016-11-30 19:48 ` Lino Sanfilippo
0 siblings, 0 replies; 14+ messages in thread
From: Lino Sanfilippo @ 2016-11-30 19:48 UTC (permalink / raw)
To: Florian Fainelli, davem, charrer, liodot, gregkh, andrew
Cc: devel, linux-kernel, netdev
On 29.11.2016 18:14, Florian Fainelli wrote:
> On 11/28/2016 01:41 PM, Lino Sanfilippo wrote:
>> The problem is that the HW does not provide a tx completion index. Instead we have to
>> iterate the status descriptors until we get an invalid idx which indicates that there
>> are no further tx descriptors done for now. I am afraid that if we do not limit the
>> number of descriptors processed in the tx completion handler, a continuous transmission
>> of frames could keep the loop in xmit_complete() run endlessly. I dont know if this
>> can actually happen but I wanted to make sure that this is avoided.
>
> OK, it might be a good idea to put that comment somewhere around the tx
> completion handler to understand why it is bounded with a specific value.
>
Agreed, I will add such a comment.
>>
>>> [snip]
>>>
>>>> + while (slic_get_free_rx_descs(rxq) > SLIC_MAX_REQ_RX_DESCS) {
>>>> + skb = alloc_skb(maplen + ALIGN_MASK, gfp);
>>>> + if (!skb)
>>>> + break;
>>>> +
>>>> + paddr = dma_map_single(&sdev->pdev->dev, skb->data, maplen,
>>>> + DMA_FROM_DEVICE);
>>>> + if (dma_mapping_error(&sdev->pdev->dev, paddr)) {
>>>> + netdev_err(dev, "mapping rx packet failed\n");
>>>> + /* drop skb */
>>>> + dev_kfree_skb_any(skb);
>>>> + break;
>>>> + }
>>>> + /* ensure head buffer descriptors are 256 byte aligned */
>>>> + offset = 0;
>>>> + misalign = paddr & ALIGN_MASK;
>>>> + if (misalign) {
>>>> + offset = SLIC_RX_BUFF_ALIGN - misalign;
>>>> + skb_reserve(skb, offset);
>>>> + }
>>>> + /* the HW expects dma chunks for descriptor + frame data */
>>>> + desc = (struct slic_rx_desc *)skb->data;
>>>> + memset(desc, 0, sizeof(*desc));
>>>
>>> Do you really need to zero-out the prepending RX descriptor? Are not you
>>> missing a write barrier here?
>>
>> Indeed, it should be sufficient to make sure that the bit SLIC_IRHDDR_SVALID is not set.
>> I will adjust it.
>> Concerning the write barrier: You mean a wmb() before slic_write() to ensure that the zeroing
>> of the status desc is done before the descriptor is passed to the HW, right?
>
> Correct, that's what I meant here.
>
Ok, will fix this. Good catch BTW!
>>
>>> [snip]
>>>
>>>> +
>>>> + dma_sync_single_for_cpu(&sdev->pdev->dev,
>>>> + dma_unmap_addr(buff, map_addr),
>>>> + buff->addr_offset + sizeof(*desc),
>>>> + DMA_FROM_DEVICE);
>>>> +
>>>> + status = le32_to_cpu(desc->status);
>>>> + if (!(status & SLIC_IRHDDR_SVALID))
>>>> + break;
>>>> +
>>>> + buff->skb = NULL;
>>>> +
>>>> + dma_unmap_single(&sdev->pdev->dev,
>>>> + dma_unmap_addr(buff, map_addr),
>>>> + dma_unmap_len(buff, map_len),
>>>> + DMA_FROM_DEVICE);
>>>
>>> This is potentially inefficient, you already did a cache invalidation
>>> for the RX descriptor here, you could be more efficient with just
>>> invalidating the packet length, minus the descriptor length.
>>>
>>
>> I am not sure I understand: We have to unmap the complete dma area, no matter if we synced
>> part of it before, dont we? AFAIK a dma sync is different from unmapping dma, or do I miss
>> something?
>
> Sorry, I was not very clear, what I meant is that you can allocate and
> do the initial dma_map_single() of your RX skbs during ndo_open(), and
> then, in your RX path, you can only do dma_sync_single_for_cpu() twice
> (once for the RX descriptor status, second time for the actual packet
> contents), and when you return the SKB to the HW, do a
> dma_sync_single_for_device(). The advantage of doing that, is that if
> your cache operations are slow, you only do them on exactly packet
> length, and not the actual RX buffer size (e.g: 2KB).
Um. In the rx path the SKB will be consumed (by napi_gro_receive()).
AFAIK we _have_ to unmap it before this call. Doing only a dma_sync_single_for_cpu()
for the packet contents does IMHO only make sense if the corresponding SKB is
reused somehow. But this is not the case. The rx buffers are refilled with newly
allocated SKBs each time, and thus we need to create a new dma mapping for each of them.
Or do I still misunderstand when to call the dma sync functions?
BTW: I just realized that if the descriptor has not been used by the HW yet, see:
+ dma_sync_single_for_cpu(&sdev->pdev->dev,
+ dma_unmap_addr(buff, map_addr),
+ buff->addr_offset + sizeof(*desc),
+ DMA_FROM_DEVICE);
+
+ status = le32_to_cpu(desc->status);
+ if (!(status & SLIC_IRHDDR_SVALID))
+ break;
+
^^^^^^^^^^^^^^^^^^^^^^^^^^^ dma_sync_single_for_device missing
there has to be a dma_sync_single_for_device to undo the sync for cpu (since the
HW will write to this descr when the next rx packet arrives), right? But this is racy:
What if the HW writes to that descr after we synced it for cpu but before we synced
it for the HW again? Any ideas?
Regards,
Lino
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver
2016-11-26 12:20 ` [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver Lino Sanfilippo
` (2 preceding siblings ...)
2016-11-28 4:56 ` Florian Fainelli
@ 2016-12-04 2:13 ` kbuild test robot
3 siblings, 0 replies; 14+ messages in thread
From: kbuild test robot @ 2016-12-04 2:13 UTC (permalink / raw)
To: Lino Sanfilippo
Cc: kbuild-all, davem, charrer, liodot, gregkh, andrew, devel,
linux-kernel, netdev, Lino Sanfilippo
[-- Attachment #1: Type: text/plain, Size: 3184 bytes --]
Hi Lino,
[auto build test ERROR on net-next/master]
url: https://github.com/0day-ci/linux/commits/Lino-Sanfilippo/net-ethernet-slicoss-add-slicoss-gigabit-ethernet-driver/20161126-202438
config: sparc64-allyesconfig (attached as .config)
compiler: sparc64-linux-gnu-gcc (Debian 6.1.1-9) 6.1.1 20160705
reproduce:
wget https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
make.cross ARCH=sparc64
All errors (new ones prefixed by >>):
drivers/staging/slicoss/slicoss.c: In function 'slic_cmdq_addcmdpage':
>> drivers/staging/slicoss/slicoss.c:1258:14: error: implicit declaration of function 'virt_to_bus' [-Werror=implicit-function-declaration]
phys_addr = virt_to_bus((void *)page);
^~~~~~~~~~~
cc1: some warnings being treated as errors
vim +/virt_to_bus +1258 drivers/staging/slicoss/slicoss.c
4d6ea9c3 Denis Kirjanov 2010-07-10 1242 struct slic_hostcmd *cmd;
4d6ea9c3 Denis Kirjanov 2010-07-10 1243 struct slic_hostcmd *prev;
4d6ea9c3 Denis Kirjanov 2010-07-10 1244 struct slic_hostcmd *tail;
4d6ea9c3 Denis Kirjanov 2010-07-10 1245 struct slic_cmdqueue *cmdq;
4d6ea9c3 Denis Kirjanov 2010-07-10 1246 int cmdcnt;
4d6ea9c3 Denis Kirjanov 2010-07-10 1247 void *cmdaddr;
4d6ea9c3 Denis Kirjanov 2010-07-10 1248 ulong phys_addr;
4d6ea9c3 Denis Kirjanov 2010-07-10 1249 u32 phys_addrl;
4d6ea9c3 Denis Kirjanov 2010-07-10 1250 u32 phys_addrh;
4d6ea9c3 Denis Kirjanov 2010-07-10 1251 struct slic_handle *pslic_handle;
eafe6002 David Matlack 2015-05-11 1252 unsigned long flags;
4d6f6af8 Greg Kroah-Hartman 2008-03-19 1253
4d6ea9c3 Denis Kirjanov 2010-07-10 1254 cmdaddr = page;
dd146d21 Shraddha Barke 2015-10-15 1255 cmd = cmdaddr;
4d6ea9c3 Denis Kirjanov 2010-07-10 1256 cmdcnt = 0;
4d6f6af8 Greg Kroah-Hartman 2008-03-19 1257
4d6ea9c3 Denis Kirjanov 2010-07-10 @1258 phys_addr = virt_to_bus((void *)page);
4d6ea9c3 Denis Kirjanov 2010-07-10 1259 phys_addrl = SLIC_GET_ADDR_LOW(phys_addr);
4d6ea9c3 Denis Kirjanov 2010-07-10 1260 phys_addrh = SLIC_GET_ADDR_HIGH(phys_addr);
4d6f6af8 Greg Kroah-Hartman 2008-03-19 1261
4d6ea9c3 Denis Kirjanov 2010-07-10 1262 prev = NULL;
4d6ea9c3 Denis Kirjanov 2010-07-10 1263 tail = cmd;
4d6ea9c3 Denis Kirjanov 2010-07-10 1264 while ((cmdcnt < SLIC_CMDQ_CMDSINPAGE) &&
4d6ea9c3 Denis Kirjanov 2010-07-10 1265 (adapter->slic_handle_ix < 256)) {
4d6ea9c3 Denis Kirjanov 2010-07-10 1266 /* Allocate and initialize a SLIC_HANDLE for this command */
:::::: The code at line 1258 was first introduced by commit
:::::: 4d6ea9c3223da8d8dc91b369087fa40cc53edd36 Staging: slicoss: kill functions prototypes and reorder functions
:::::: TO: Denis Kirjanov <dkirjanov@hera.kernel.org>
:::::: CC: Greg Kroah-Hartman <gregkh@suse.de>
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 48196 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2016-12-04 2:14 UTC | newest]
Thread overview: 14+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-11-26 12:20 Gigabit ethernet driver for Alacritechs SLIC devices (v3) Lino Sanfilippo
2016-11-26 12:20 ` [PATCH v3 net-next 1/2] net: ethernet: slicoss: add slicoss gigabit ethernet driver Lino Sanfilippo
2016-11-26 15:48 ` Rami Rosen
2016-11-27 1:02 ` Lino Sanfilippo
2016-11-27 17:59 ` Markus Böhme
2016-11-28 20:46 ` Lino Sanfilippo
2016-11-29 12:15 ` Markus Böhme
2016-11-28 4:56 ` Florian Fainelli
2016-11-28 21:41 ` Lino Sanfilippo
2016-11-29 17:14 ` Florian Fainelli
2016-11-30 19:48 ` Lino Sanfilippo
2016-12-04 2:13 ` kbuild test robot
2016-11-26 12:20 ` [PATCH v3 net-next 2/2] MAINTAINERS: add entry for slicoss " Lino Sanfilippo
2016-11-26 12:36 ` Gigabit ethernet driver for Alacritechs SLIC devices (v3) Lino Sanfilippo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).