* [PATCH] FEC driver: patches for -next
@ 2009-04-15 11:32 Sascha Hauer
2009-04-15 11:32 ` [PATCH 01/12] fec: remove unused ifdef Sascha Hauer
2009-04-15 12:12 ` [PATCH] FEC driver: patches for -next Greg Ungerer
0 siblings, 2 replies; 29+ messages in thread
From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw)
To: netdev; +Cc: Greg Ungerer
Hi,
These patches aim to modernize the FEC ethernet driver.
Greg, can you please confirm that the ifdef in patch 01/12 is indeed unused?
Sascha
The following changes since commit 0882e8dd3aad33eca41696d463bb896e6c8817eb:
Linus Torvalds (1):
Linux 2.6.30-rc2
are available in the git repository at:
git://git.pengutronix.de/git/sha/linux-2.6.git fec
Sascha Hauer (12):
fec: remove unused ifdef
fec: switch to writel/readl
fec: do not typedef struct types
fec: remove unnecessary cast
fec: Codingstyle cleanups
fec: refactor set_multicast_list() to make it more readable
fec: refactor init function
fec: align receive packets
fec: remove debugging printks
fec: switch to net_device_ops
FEC Buffer rework
fec: call fec_restart() in fec_open()
drivers/net/fec.c | 897 ++++++++++++++++++++++-------------------------------
drivers/net/fec.h | 112 ++-----
2 files changed, 402 insertions(+), 607 deletions(-)
^ permalink raw reply [flat|nested] 29+ messages in thread* [PATCH 01/12] fec: remove unused ifdef 2009-04-15 11:32 [PATCH] FEC driver: patches for -next Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 02/12] fec: switch to writel/readl Sascha Hauer 2009-04-15 12:12 ` [PATCH] FEC driver: patches for -next Greg Ungerer 1 sibling, 1 reply; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.h | 44 -------------------------------------------- 1 files changed, 0 insertions(+), 44 deletions(-) diff --git a/drivers/net/fec.h b/drivers/net/fec.h index 76c64c9..c2f97f3 100644 --- a/drivers/net/fec.h +++ b/drivers/net/fec.h @@ -13,8 +13,6 @@ #define FEC_H /****************************************************************************/ -#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \ - defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARCH_MXC) /* * Just figures, Motorola would have to change the offsets for * registers in the same peripheral device on different models @@ -58,48 +56,6 @@ typedef struct fec { unsigned long fec_r_buff_size; /* Maximum receive buff size */ } fec_t; -#else - -/* - * Define device register set address map. - */ -typedef struct fec { - unsigned long fec_ecntrl; /* Ethernet control reg */ - unsigned long fec_ievent; /* Interrupt even reg */ - unsigned long fec_imask; /* Interrupt mask reg */ - unsigned long fec_ivec; /* Interrupt vec status reg */ - unsigned long fec_r_des_active; /* Receive descriptor reg */ - unsigned long fec_x_des_active; /* Transmit descriptor reg */ - unsigned long fec_reserved1[10]; - unsigned long fec_mii_data; /* MII manage frame reg */ - unsigned long fec_mii_speed; /* MII speed control reg */ - unsigned long fec_reserved2[17]; - unsigned long fec_r_bound; /* FIFO receive bound reg */ - unsigned long fec_r_fstart; /* FIFO receive start reg */ - unsigned long fec_reserved3[4]; - unsigned long fec_x_wmrk; /* FIFO transmit water mark */ - unsigned long fec_reserved4; - unsigned long fec_x_fstart; /* FIFO transmit start reg */ - unsigned long fec_reserved5[21]; - unsigned long fec_r_cntrl; /* Receive control reg */ - unsigned long fec_max_frm_len; /* Maximum frame length reg */ - unsigned long fec_reserved6[14]; - unsigned long fec_x_cntrl; /* Transmit Control reg */ - unsigned long fec_reserved7[158]; - unsigned long fec_addr_low; /* Low 32bits MAC address */ - unsigned long fec_addr_high; /* High 16bits MAC address */ - unsigned long fec_grp_hash_table_high;/* High 32bits hash table */ - unsigned long fec_grp_hash_table_low; /* Low 32bits hash table */ - unsigned long fec_r_des_start; /* Receive descriptor ring */ - unsigned long fec_x_des_start; /* Transmit descriptor ring */ - unsigned long fec_r_buff_size; /* Maximum receive buff size */ - unsigned long reserved8[9]; - unsigned long fec_fifo_ram[112]; /* FIFO RAM buffer */ -} fec_t; - -#endif /* CONFIG_M5272 */ - - /* * Define the buffer descriptor structure. */ -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 02/12] fec: switch to writel/readl 2009-04-15 11:32 ` [PATCH 01/12] fec: remove unused ifdef Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 03/12] fec: do not typedef struct types Sascha Hauer 2009-04-15 13:11 ` [PATCH 02/12] fec: switch to writel/readl Sascha Hauer 0 siblings, 2 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 279 ++++++++++++++++++++++------------------------------- drivers/net/fec.h | 60 +++++------- 2 files changed, 138 insertions(+), 201 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index 682e7f0..3fa3fe4 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -165,7 +165,7 @@ typedef struct { */ struct fec_enet_private { /* Hardware registers of the FEC device */ - volatile fec_t *hwp; + void __iomem *hwp; struct net_device *netdev; @@ -288,15 +288,11 @@ static int mii_queue(struct net_device *dev, int request, static int fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) { - struct fec_enet_private *fep; - volatile fec_t *fecp; + struct fec_enet_private *fep = netdev_priv(dev); volatile cbd_t *bdp; unsigned short status; unsigned long flags; - fep = netdev_priv(dev); - fecp = (volatile fec_t*)dev->base_addr; - if (!fep->link) { /* Link is down or autonegotiation is in progress. */ return 1; @@ -363,7 +359,7 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) dev->trans_start = jiffies; /* Trigger transmission start */ - fecp->fec_x_des_active = 0; + writel(0, fep->hwp + FEC_X_DES_ACTIVE); /* If this was the last BD in the ring, start at the beginning again. */ @@ -436,29 +432,25 @@ static irqreturn_t fec_enet_interrupt(int irq, void * dev_id) { struct net_device *dev = dev_id; - volatile fec_t *fecp; + struct fec_enet_private *fep = netdev_priv(dev); uint int_events; irqreturn_t ret = IRQ_NONE; - fecp = (volatile fec_t*)dev->base_addr; - - /* Get the interrupt events that caused us to be here. - */ + /* Get the interrupt events that caused us to be here. */ do { - int_events = fecp->fec_ievent; - fecp->fec_ievent = int_events; + int_events = readl(fep->hwp + FEC_IEVENT); + writel(int_events, fep->hwp + FEC_IEVENT); - /* Handle receive event in its own function. - */ + /* Handle receive event in its own function. */ if (int_events & FEC_ENET_RXF) { ret = IRQ_HANDLED; fec_enet_rx(dev); } /* Transmit OK, or non-fatal error. Update the buffer - descriptors. FEC handles all errors, we just discover - them as part of the transmit process. - */ + * descriptors. FEC handles all errors, we just discover + * them as part of the transmit process. + */ if (int_events & FEC_ENET_TXF) { ret = IRQ_HANDLED; fec_enet_tx(dev); @@ -555,8 +547,7 @@ fec_enet_tx(struct net_device *dev) static void fec_enet_rx(struct net_device *dev) { - struct fec_enet_private *fep; - volatile fec_t *fecp; + struct fec_enet_private *fep = netdev_priv(dev); volatile cbd_t *bdp; unsigned short status; struct sk_buff *skb; @@ -567,9 +558,6 @@ fec_enet_rx(struct net_device *dev) flush_cache_all(); #endif - fep = netdev_priv(dev); - fecp = (volatile fec_t*)dev->base_addr; - spin_lock_irq(&fep->hw_lock); /* First, grab all of the stats for the incoming packet. @@ -665,7 +653,7 @@ while (!((status = bdp->cbd_sc) & BD_ENET_RX_EMPTY)) { * incoming frames. On a heavily loaded network, we should be * able to keep up at the expense of system resources. */ - fecp->fec_r_des_active = 0; + writel(0, fep->hwp + FEC_R_DES_ACTIVE); #endif } /* while (!((status = bdp->cbd_sc) & BD_ENET_RX_EMPTY)) */ fep->cur_rx = (cbd_t *)bdp; @@ -690,30 +678,25 @@ static void fec_enet_mii(struct net_device *dev) { struct fec_enet_private *fep; - volatile fec_t *ep; mii_list_t *mip; - uint mii_reg; fep = netdev_priv(dev); spin_lock_irq(&fep->mii_lock); - ep = fep->hwp; - mii_reg = ep->fec_mii_data; - if ((mip = mii_head) == NULL) { printk("MII and no head!\n"); goto unlock; } if (mip->mii_func != NULL) - (*(mip->mii_func))(mii_reg, dev); + (*(mip->mii_func))(readl(fep->hwp + FEC_MII_DATA), dev); mii_head = mip->mii_next; mip->mii_next = mii_free; mii_free = mip; if ((mip = mii_head) != NULL) - ep->fec_mii_data = mip->mii_regval; + writel(mip->mii_regval, fep->hwp + FEC_MII_DATA); unlock: spin_unlock_irq(&fep->mii_lock); @@ -745,7 +728,7 @@ mii_queue(struct net_device *dev, int regval, void (*func)(uint, struct net_devi mii_tail = mip; } else { mii_head = mii_tail = mip; - fep->hwp->fec_mii_data = regval; + writel(regval, fep->hwp + FEC_MII_DATA); } } else { retval = 1; @@ -1246,11 +1229,8 @@ static void __inline__ fec_phy_ack_intr(void) static void __inline__ fec_get_mac(struct net_device *dev) { struct fec_enet_private *fep = netdev_priv(dev); - volatile fec_t *fecp; unsigned char *iap, tmpaddr[ETH_ALEN]; - fecp = fep->hwp; - if (FEC_FLASHMAC) { /* * Get MAC address from FLASH. @@ -1264,8 +1244,8 @@ static void __inline__ fec_get_mac(struct net_device *dev) (iap[3] == 0xff) && (iap[4] == 0xff) && (iap[5] == 0xff)) iap = fec_mac_default; } else { - *((unsigned long *) &tmpaddr[0]) = fecp->fec_addr_low; - *((unsigned short *) &tmpaddr[4]) = (fecp->fec_addr_high >> 16); + *((unsigned long *) &tmpaddr[0]) = readl(fep->hwp + FEC_ADDR_LOW); + *((unsigned short *) &tmpaddr[4]) = (readl(fep->hwp + FEC_ADDR_HIGH) >> 16); iap = &tmpaddr[0]; } @@ -1457,11 +1437,9 @@ static void mii_discover_phy(uint mii_reg, struct net_device *dev) { struct fec_enet_private *fep; - volatile fec_t *fecp; uint phytype; fep = netdev_priv(dev); - fecp = fep->hwp; if (fep->phy_addr < 32) { if ((phytype = (mii_reg & 0xffff)) != 0xffff && phytype != 0) { @@ -1479,7 +1457,8 @@ mii_discover_phy(uint mii_reg, struct net_device *dev) } else { printk("FEC: No PHY device found.\n"); /* Disable external MII interface */ - fecp->fec_mii_speed = fep->phy_speed = 0; + writel(0, fep->hwp + FEC_MII_SPEED); + fep->phy_speed = 0; #ifdef HAVE_mii_link_interrupt fec_disable_phy_intr(); #endif @@ -1583,32 +1562,31 @@ fec_enet_close(struct net_device *dev) static void set_multicast_list(struct net_device *dev) { - struct fec_enet_private *fep; - volatile fec_t *ep; + struct fec_enet_private *fep = netdev_priv(dev); struct dev_mc_list *dmi; - unsigned int i, j, bit, data, crc; + unsigned int i, j, bit, data, crc, tmp; unsigned char hash; - fep = netdev_priv(dev); - ep = fep->hwp; - if (dev->flags&IFF_PROMISC) { - ep->fec_r_cntrl |= 0x0008; + tmp = readl(fep->hwp + FEC_R_CNTRL); + tmp |= 0x8; + writel(tmp, fep->hwp + FEC_R_CNTRL); } else { - - ep->fec_r_cntrl &= ~0x0008; + tmp = readl(fep->hwp + FEC_R_CNTRL); + tmp &= ~0x8; + writel(tmp, fep->hwp + FEC_R_CNTRL); if (dev->flags & IFF_ALLMULTI) { /* Catch all multicast addresses, so set the * filter to all 1's. */ - ep->fec_grp_hash_table_high = 0xffffffff; - ep->fec_grp_hash_table_low = 0xffffffff; + writel(0xffffffff, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + writel(0xffffffff, fep->hwp + FEC_GRP_HASH_TABLE_LOW); } else { /* Clear filter and add the addresses in hash register. */ - ep->fec_grp_hash_table_high = 0; - ep->fec_grp_hash_table_low = 0; + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); dmi = dev->mc_list; @@ -1638,10 +1616,15 @@ static void set_multicast_list(struct net_device *dev) */ hash = (crc >> (32 - HASH_BITS)) & 0x3f; - if (hash > 31) - ep->fec_grp_hash_table_high |= 1 << (hash - 32); - else - ep->fec_grp_hash_table_low |= 1 << hash; + if (hash > 31) { + tmp = readl(fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + tmp |= 1 << (hash - 32); + writel(tmp, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + } else { + tmp = readl(fep->hwp + FEC_GRP_HASH_TABLE_LOW); + tmp |= 1 << hash; + writel(tmp, fep->hwp + FEC_GRP_HASH_TABLE_LOW); + } } } } @@ -1652,16 +1635,14 @@ static void set_multicast_list(struct net_device *dev) static void fec_set_mac_address(struct net_device *dev) { - volatile fec_t *fecp; - - fecp = ((struct fec_enet_private *)netdev_priv(dev))->hwp; + struct fec_enet_private *fep = netdev_priv(dev); /* Set station address. */ - fecp->fec_addr_low = dev->dev_addr[3] | (dev->dev_addr[2] << 8) | - (dev->dev_addr[1] << 16) | (dev->dev_addr[0] << 24); - fecp->fec_addr_high = (dev->dev_addr[5] << 16) | - (dev->dev_addr[4] << 24); - + writel(dev->dev_addr[3] | (dev->dev_addr[2] << 8) | + (dev->dev_addr[1] << 16) | (dev->dev_addr[0] << 24), + fep->hwp + FEC_ADDR_LOW); + writel((dev->dev_addr[5] << 16) | (dev->dev_addr[4] << 24), + fep + FEC_ADDR_HIGH); } /* @@ -1675,7 +1656,6 @@ int __init fec_enet_init(struct net_device *dev, int index) unsigned long mem_addr; volatile cbd_t *bdp; cbd_t *cbd_base; - volatile fec_t *fecp; int i, j; /* Allocate memory for buffer descriptors. @@ -1690,17 +1670,13 @@ int __init fec_enet_init(struct net_device *dev, int index) spin_lock_init(&fep->hw_lock); spin_lock_init(&fep->mii_lock); - /* Create an Ethernet device instance. - */ - fecp = (volatile fec_t *)dev->base_addr; - fep->index = index; - fep->hwp = fecp; + fep->hwp = (void __iomem *)dev->base_addr; fep->netdev = dev; /* Whack a reset. We should wait for this. */ - fecp->fec_ecntrl = 1; + writel(1, fep->hwp + FEC_ECNTRL); udelay(10); /* Set the Ethernet address */ @@ -1709,12 +1685,12 @@ int __init fec_enet_init(struct net_device *dev, int index) #else { unsigned long l; - l = fecp->fec_addr_low; + l = readl(fep->hwp + FEC_ADDR_LOW); dev->dev_addr[0] = (unsigned char)((l & 0xFF000000) >> 24); dev->dev_addr[1] = (unsigned char)((l & 0x00FF0000) >> 16); dev->dev_addr[2] = (unsigned char)((l & 0x0000FF00) >> 8); dev->dev_addr[3] = (unsigned char)((l & 0x000000FF) >> 0); - l = fecp->fec_addr_high; + l = readl(fep->hwp + FEC_ADDR_HIGH); dev->dev_addr[4] = (unsigned char)((l & 0xFF000000) >> 24); dev->dev_addr[5] = (unsigned char)((l & 0x00FF0000) >> 16); } @@ -1784,22 +1760,22 @@ int __init fec_enet_init(struct net_device *dev, int index) /* Set receive and transmit descriptor base. */ - fecp->fec_r_des_start = fep->bd_dma; - fecp->fec_x_des_start = (unsigned long)fep->bd_dma + sizeof(cbd_t) - * RX_RING_SIZE; + writel(fep->bd_dma, fep->hwp + FEC_R_DES_START); + writel((unsigned long)fep->bd_dma + sizeof(cbd_t) * RX_RING_SIZE, + fep->hwp + FEC_X_DES_START); #ifdef HAVE_mii_link_interrupt fec_request_mii_intr(dev); #endif - fecp->fec_grp_hash_table_high = 0; - fecp->fec_grp_hash_table_low = 0; - fecp->fec_r_buff_size = PKT_MAXBLR_SIZE; - fecp->fec_ecntrl = 2; - fecp->fec_r_des_active = 0; + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); + writel(PKT_MAXBLR_SIZE, fep->hwp + FEC_R_BUFF_SIZE); + writel(2, fep->hwp + FEC_ECNTRL); + writel(0, fep->hwp + FEC_R_DES_ACTIVE); #ifndef CONFIG_M5272 - fecp->fec_hash_table_high = 0; - fecp->fec_hash_table_low = 0; + writel(0, fep->hwp + FEC_HASH_TABLE_HIGH); + writel(0, fep->hwp + FEC_HASH_TABLE_LOW); #endif /* The FEC Ethernet specific entries in the device structure. */ @@ -1815,20 +1791,21 @@ int __init fec_enet_init(struct net_device *dev, int index) mii_free = mii_cmds; /* setup MII interface */ - fecp->fec_r_cntrl = OPT_FRAME_SIZE | 0x04; - fecp->fec_x_cntrl = 0x00; + writel(OPT_FRAME_SIZE | 0x04, fep->hwp + FEC_R_CNTRL); + writel(0, fep->hwp + FEC_X_CNTRL); /* * Set MII speed to 2.5 MHz */ fep->phy_speed = ((((clk_get_rate(fep->clk) / 2 + 4999999) / 2500000) / 2) & 0x3F) << 1; - fecp->fec_mii_speed = fep->phy_speed; + writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED); fec_restart(dev, 0); /* Clear and enable interrupts */ - fecp->fec_ievent = 0xffc00000; - fecp->fec_imask = (FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII); + writel(0xffc00000, fep->hwp + FEC_IEVENT); + writel(FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII, + fep->hwp + FEC_IMASK); /* Queue up command to detect the PHY and initialize the * remainder of the interface. @@ -1847,47 +1824,36 @@ int __init fec_enet_init(struct net_device *dev, int index) static void fec_restart(struct net_device *dev, int duplex) { - struct fec_enet_private *fep; + struct fec_enet_private *fep = netdev_priv(dev); volatile cbd_t *bdp; - volatile fec_t *fecp; int i; - fep = netdev_priv(dev); - fecp = fep->hwp; - - /* Whack a reset. We should wait for this. - */ - fecp->fec_ecntrl = 1; + /* Whack a reset. We should wait for this. */ + writel(1, fep->hwp + FEC_ECNTRL); udelay(10); - /* Clear any outstanding interrupt. - */ - fecp->fec_ievent = 0xffc00000; + /* Clear any outstanding interrupt. */ + writel(0xffc00000, fep->hwp + FEC_IEVENT); - /* Set station address. - */ + /* Set station address. */ fec_set_mac_address(dev); - /* Reset all multicast. - */ - fecp->fec_grp_hash_table_high = 0; - fecp->fec_grp_hash_table_low = 0; + /* Reset all multicast. */ + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); - /* Set maximum receive buffer size. - */ - fecp->fec_r_buff_size = PKT_MAXBLR_SIZE; + /* Set maximum receive buffer size. */ + writel(PKT_MAXBLR_SIZE, fep->hwp + FEC_R_BUFF_SIZE); - /* Set receive and transmit descriptor base. - */ - fecp->fec_r_des_start = fep->bd_dma; - fecp->fec_x_des_start = (unsigned long)fep->bd_dma + sizeof(cbd_t) - * RX_RING_SIZE; + /* Set receive and transmit descriptor base. */ + writel(fep->bd_dma, fep->hwp + FEC_R_DES_START); + writel((unsigned long)fep->bd_dma + sizeof(cbd_t) * RX_RING_SIZE, + fep->hwp + FEC_X_DES_START); fep->dirty_tx = fep->cur_tx = fep->tx_bd_base; fep->cur_rx = fep->rx_bd_base; - /* Reset SKB transmit buffers. - */ + /* Reset SKB transmit buffers. */ fep->skb_cur = fep->skb_dirty = 0; for (i=0; i<=TX_RING_MOD_MASK; i++) { if (fep->tx_skbuff[i] != NULL) { @@ -1896,96 +1862,81 @@ fec_restart(struct net_device *dev, int duplex) } } - /* Initialize the receive buffer descriptors. - */ + /* Initialize the receive buffer descriptors. */ bdp = fep->rx_bd_base; for (i=0; i<RX_RING_SIZE; i++) { - /* Initialize the BD for every fragment in the page. - */ + /* Initialize the BD for every fragment in the page. */ bdp->cbd_sc = BD_ENET_RX_EMPTY; bdp++; } - /* Set the last buffer to wrap. - */ + /* Set the last buffer to wrap. */ bdp--; bdp->cbd_sc |= BD_SC_WRAP; - /* ...and the same for transmmit. - */ + /* ...and the same for transmmit. */ bdp = fep->tx_bd_base; for (i=0; i<TX_RING_SIZE; i++) { - /* Initialize the BD for every fragment in the page. - */ + /* Initialize the BD for every fragment in the page. */ bdp->cbd_sc = 0; bdp->cbd_bufaddr = 0; bdp++; } - /* Set the last buffer to wrap. - */ + /* Set the last buffer to wrap. */ bdp--; bdp->cbd_sc |= BD_SC_WRAP; - /* Enable MII mode. - */ + /* Enable MII mode. */ if (duplex) { - fecp->fec_r_cntrl = OPT_FRAME_SIZE | 0x04;/* MII enable */ - fecp->fec_x_cntrl = 0x04; /* FD enable */ + /* MII enable / FD enable */ + writel(OPT_FRAME_SIZE | 0x04, fep->hwp + FEC_R_CNTRL); + writel(0x04, fep->hwp + FEC_X_CNTRL); } else { - /* MII enable|No Rcv on Xmit */ - fecp->fec_r_cntrl = OPT_FRAME_SIZE | 0x06; - fecp->fec_x_cntrl = 0x00; + /* MII enable / No Rcv on Xmit */ + writel(OPT_FRAME_SIZE | 0x06, fep->hwp + FEC_R_CNTRL); + writel(0x0, fep->hwp + FEC_X_CNTRL); } fep->full_duplex = duplex; - /* Set MII speed. - */ - fecp->fec_mii_speed = fep->phy_speed; + /* Set MII speed. */ + writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED); - /* And last, enable the transmit and receive processing. - */ - fecp->fec_ecntrl = 2; - fecp->fec_r_des_active = 0; + /* And last, enable the transmit and receive processing. */ + writel(2, fep->hwp + FEC_ECNTRL); + writel(0, fep->hwp + FEC_R_DES_ACTIVE); - /* Enable interrupts we wish to service. - */ - fecp->fec_imask = (FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII); + /* Enable interrupts we wish to service. */ + writel(FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII, + fep->hwp + FEC_IMASK); } static void fec_stop(struct net_device *dev) { - volatile fec_t *fecp; - struct fec_enet_private *fep; - - fep = netdev_priv(dev); - fecp = fep->hwp; + struct fec_enet_private *fep = netdev_priv(dev); /* ** We cannot expect a graceful transmit stop without link !!! */ - if (fep->link) - { - fecp->fec_x_cntrl = 0x01; /* Graceful transmit stop */ + if (fep->link) { + writel(1, fep->hwp + FEC_X_CNTRL); /* Graceful transmit stop */ udelay(10); - if (!(fecp->fec_ievent & FEC_ENET_GRA)) + if (!(readl(fep->hwp + FEC_IEVENT) & FEC_ENET_GRA)) printk("fec_stop : Graceful transmit stop did not complete !\n"); - } + } - /* Whack a reset. We should wait for this. - */ - fecp->fec_ecntrl = 1; + /* Whack a reset. We should wait for this. */ + writel(1, fep->hwp + FEC_ECNTRL); udelay(10); - /* Clear outstanding MII command interrupts. - */ - fecp->fec_ievent = FEC_ENET_MII; + /* Clear outstanding MII command interrupts. */ + writel(FEC_ENET_MII, fep->hwp + FEC_IEVENT); - fecp->fec_imask = FEC_ENET_MII; - fecp->fec_mii_speed = fep->phy_speed; + writel(FEC_ENET_MII, fep->hwp + FEC_IMASK); + writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED); } static int __devinit diff --git a/drivers/net/fec.h b/drivers/net/fec.h index c2f97f3..bb4dabc 100644 --- a/drivers/net/fec.h +++ b/drivers/net/fec.h @@ -18,43 +18,29 @@ * registers in the same peripheral device on different models * of the ColdFire! */ -typedef struct fec { - unsigned long fec_reserved0; - unsigned long fec_ievent; /* Interrupt event reg */ - unsigned long fec_imask; /* Interrupt mask reg */ - unsigned long fec_reserved1; - unsigned long fec_r_des_active; /* Receive descriptor reg */ - unsigned long fec_x_des_active; /* Transmit descriptor reg */ - unsigned long fec_reserved2[3]; - unsigned long fec_ecntrl; /* Ethernet control reg */ - unsigned long fec_reserved3[6]; - unsigned long fec_mii_data; /* MII manage frame reg */ - unsigned long fec_mii_speed; /* MII speed control reg */ - unsigned long fec_reserved4[7]; - unsigned long fec_mib_ctrlstat; /* MIB control/status reg */ - unsigned long fec_reserved5[7]; - unsigned long fec_r_cntrl; /* Receive control reg */ - unsigned long fec_reserved6[15]; - unsigned long fec_x_cntrl; /* Transmit Control reg */ - unsigned long fec_reserved7[7]; - unsigned long fec_addr_low; /* Low 32bits MAC address */ - unsigned long fec_addr_high; /* High 16bits MAC address */ - unsigned long fec_opd; /* Opcode + Pause duration */ - unsigned long fec_reserved8[10]; - unsigned long fec_hash_table_high; /* High 32bits hash table */ - unsigned long fec_hash_table_low; /* Low 32bits hash table */ - unsigned long fec_grp_hash_table_high;/* High 32bits hash table */ - unsigned long fec_grp_hash_table_low; /* Low 32bits hash table */ - unsigned long fec_reserved9[7]; - unsigned long fec_x_wmrk; /* FIFO transmit water mark */ - unsigned long fec_reserved10; - unsigned long fec_r_bound; /* FIFO receive bound reg */ - unsigned long fec_r_fstart; /* FIFO receive start reg */ - unsigned long fec_reserved11[11]; - unsigned long fec_r_des_start; /* Receive descriptor ring */ - unsigned long fec_x_des_start; /* Transmit descriptor ring */ - unsigned long fec_r_buff_size; /* Maximum receive buff size */ -} fec_t; +#define FEC_IEVENT 0x004 /* Interrupt event reg */ +#define FEC_IMASK 0x008 /* Interrupt mask reg */ +#define FEC_R_DES_ACTIVE 0x010 /* Receive descriptor reg */ +#define FEC_X_DES_ACTIVE 0x014 /* Transmit descriptor reg */ +#define FEC_ECNTRL 0x024 /* Ethernet control reg */ +#define FEC_MII_DATA 0x040 /* MII manage frame reg */ +#define FEC_MII_SPEED 0x044 /* MII speed control reg */ +#define FEC_MIB_CTRLSTAT 0x064 /* MIB control/status reg */ +#define FEC_R_CNTRL 0x084 /* Receive control reg */ +#define FEC_X_CNTRL 0x0c4 /* Transmit Control reg */ +#define FEC_ADDR_LOW 0x0e4 /* Low 32bits MAC address */ +#define FEC_ADDR_HIGH 0x0e8 /* High 16bits MAC address */ +#define FEC_OPD 0x0ec /* Opcode + Pause duration */ +#define FEC_HASH_TABLE_HIGH 0x118 /* High 32bits hash table */ +#define FEC_HASH_TABLE_LOW 0x11c /* Low 32bits hash table */ +#define FEC_GRP_HASH_TABLE_HIGH 0x120 /* High 32bits hash table */ +#define FEC_GRP_HASH_TABLE_LOW 0x124 /* Low 32bits hash table */ +#define FEC_X_WMRK 0x144 /* FIFO transmit water mark */ +#define FEC_R_BOUND 0x14c /* FIFO receive bound reg */ +#define FEC_R_FSTART 0x150 /* FIFO receive start reg */ +#define FEC_R_DES_START 0x180 /* Receive descriptor ring */ +#define FEC_X_DES_START 0x184 /* Transmit descriptor ring */ +#define FEC_R_BUFF_SIZE 0x188 /* Maximum receive buff size */ /* * Define the buffer descriptor structure. -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 03/12] fec: do not typedef struct types 2009-04-15 11:32 ` [PATCH 02/12] fec: switch to writel/readl Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 04/12] fec: remove unnecessary cast Sascha Hauer 2009-04-16 9:36 ` [PATCH 03/12] fec: do not typedef struct types David Miller 2009-04-15 13:11 ` [PATCH 02/12] fec: switch to writel/readl Sascha Hauer 1 sibling, 2 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 37 ++++++++++++++++++++----------------- drivers/net/fec.h | 8 ++++---- 2 files changed, 24 insertions(+), 21 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index 3fa3fe4..fea8e27 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -180,10 +180,14 @@ struct fec_enet_private { /* CPM dual port RAM relative addresses. */ dma_addr_t bd_dma; - cbd_t *rx_bd_base; /* Address of Rx and Tx buffers. */ - cbd_t *tx_bd_base; - cbd_t *cur_rx, *cur_tx; /* The next free ring entry */ - cbd_t *dirty_tx; /* The ring entries to be free()ed. */ + /* Address of Rx and Tx buffers. */ + struct bufdesc *rx_bd_base; + struct bufdesc *tx_bd_base; + /* The next free ring entry */ + struct bufdesc *cur_rx, *cur_tx; + /* The ring entries to be free()ed. */ + struct bufdesc *dirty_tx; + uint tx_full; /* hold while accessing the HW like ringbuffer for tx/rx but not MAC */ spinlock_t hw_lock; @@ -289,7 +293,7 @@ static int fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct fec_enet_private *fep = netdev_priv(dev); - volatile cbd_t *bdp; + struct bufdesc *bdp; unsigned short status; unsigned long flags; @@ -374,7 +378,7 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) netif_stop_queue(dev); } - fep->cur_tx = (cbd_t *)bdp; + fep->cur_tx = bdp; spin_unlock_irqrestore(&fep->hw_lock, flags); @@ -391,7 +395,7 @@ fec_timeout(struct net_device *dev) #ifndef final_version { int i; - cbd_t *bdp; + struct bufdesc *bdp; printk("Ring data dump: cur_tx %lx%s, dirty_tx %lx cur_rx: %lx\n", (unsigned long)fep->cur_tx, fep->tx_full ? " (full)" : "", @@ -471,7 +475,7 @@ static void fec_enet_tx(struct net_device *dev) { struct fec_enet_private *fep; - volatile cbd_t *bdp; + struct bufdesc *bdp; unsigned short status; struct sk_buff *skb; @@ -534,7 +538,7 @@ fec_enet_tx(struct net_device *dev) netif_wake_queue(dev); } } - fep->dirty_tx = (cbd_t *)bdp; + fep->dirty_tx = bdp; spin_unlock_irq(&fep->hw_lock); } @@ -548,7 +552,7 @@ static void fec_enet_rx(struct net_device *dev) { struct fec_enet_private *fep = netdev_priv(dev); - volatile cbd_t *bdp; + struct bufdesc *bdp; unsigned short status; struct sk_buff *skb; ushort pkt_len; @@ -656,7 +660,7 @@ while (!((status = bdp->cbd_sc) & BD_ENET_RX_EMPTY)) { writel(0, fep->hwp + FEC_R_DES_ACTIVE); #endif } /* while (!((status = bdp->cbd_sc) & BD_ENET_RX_EMPTY)) */ - fep->cur_rx = (cbd_t *)bdp; + fep->cur_rx = bdp; #if 0 /* Doing this here will allow us to process all frames in the @@ -1654,8 +1658,7 @@ int __init fec_enet_init(struct net_device *dev, int index) { struct fec_enet_private *fep = netdev_priv(dev); unsigned long mem_addr; - volatile cbd_t *bdp; - cbd_t *cbd_base; + struct bufdesc *bdp, *cbd_base; int i, j; /* Allocate memory for buffer descriptors. @@ -1696,7 +1699,7 @@ int __init fec_enet_init(struct net_device *dev, int index) } #endif - cbd_base = (cbd_t *)mem_addr; + cbd_base = (struct bufdesc *)mem_addr; /* Set receive and transmit descriptor base. */ @@ -1761,7 +1764,7 @@ int __init fec_enet_init(struct net_device *dev, int index) /* Set receive and transmit descriptor base. */ writel(fep->bd_dma, fep->hwp + FEC_R_DES_START); - writel((unsigned long)fep->bd_dma + sizeof(cbd_t) * RX_RING_SIZE, + writel((unsigned long)fep->bd_dma + sizeof(struct bufdesc) * RX_RING_SIZE, fep->hwp + FEC_X_DES_START); #ifdef HAVE_mii_link_interrupt @@ -1825,7 +1828,7 @@ static void fec_restart(struct net_device *dev, int duplex) { struct fec_enet_private *fep = netdev_priv(dev); - volatile cbd_t *bdp; + struct bufdesc *bdp; int i; /* Whack a reset. We should wait for this. */ @@ -1847,7 +1850,7 @@ fec_restart(struct net_device *dev, int duplex) /* Set receive and transmit descriptor base. */ writel(fep->bd_dma, fep->hwp + FEC_R_DES_START); - writel((unsigned long)fep->bd_dma + sizeof(cbd_t) * RX_RING_SIZE, + writel((unsigned long)fep->bd_dma + sizeof(struct bufdesc) * RX_RING_SIZE, fep->hwp + FEC_X_DES_START); fep->dirty_tx = fep->cur_tx = fep->tx_bd_base; diff --git a/drivers/net/fec.h b/drivers/net/fec.h index bb4dabc..9ebebd6 100644 --- a/drivers/net/fec.h +++ b/drivers/net/fec.h @@ -46,17 +46,17 @@ * Define the buffer descriptor structure. */ #ifdef CONFIG_ARCH_MXC -typedef struct bufdesc { +struct bufdesc { unsigned short cbd_datlen; /* Data length */ unsigned short cbd_sc; /* Control and status info */ unsigned long cbd_bufaddr; /* Buffer address */ -} cbd_t; +}; #else -typedef struct bufdesc { +struct bufdesc { unsigned short cbd_sc; /* Control and status info */ unsigned short cbd_datlen; /* Data length */ unsigned long cbd_bufaddr; /* Buffer address */ -} cbd_t; +}; #endif /* -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 04/12] fec: remove unnecessary cast 2009-04-15 11:32 ` [PATCH 03/12] fec: do not typedef struct types Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 05/12] fec: Codingstyle cleanups Sascha Hauer 2009-04-16 9:36 ` [PATCH 04/12] fec: remove unnecessary cast David Miller 2009-04-16 9:36 ` [PATCH 03/12] fec: do not typedef struct types David Miller 1 sibling, 2 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 14 +++++--------- 1 files changed, 5 insertions(+), 9 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index fea8e27..3a43d96 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -1661,11 +1661,10 @@ int __init fec_enet_init(struct net_device *dev, int index) struct bufdesc *bdp, *cbd_base; int i, j; - /* Allocate memory for buffer descriptors. - */ - mem_addr = (unsigned long)dma_alloc_coherent(NULL, PAGE_SIZE, - &fep->bd_dma, GFP_KERNEL); - if (mem_addr == 0) { + /* Allocate memory for buffer descriptors. */ + cbd_base = dma_alloc_coherent(NULL, PAGE_SIZE, &fep->bd_dma, + GFP_KERNEL); + if (!cbd_base) { printk("FEC: allocate descriptor memory failed?\n"); return -ENOMEM; } @@ -1699,10 +1698,7 @@ int __init fec_enet_init(struct net_device *dev, int index) } #endif - cbd_base = (struct bufdesc *)mem_addr; - - /* Set receive and transmit descriptor base. - */ + /* Set receive and transmit descriptor base. */ fep->rx_bd_base = cbd_base; fep->tx_bd_base = cbd_base + RX_RING_SIZE; -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 05/12] fec: Codingstyle cleanups 2009-04-15 11:32 ` [PATCH 04/12] fec: remove unnecessary cast Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 06/12] fec: refactor set_multicast_list() to make it more readable Sascha Hauer 2009-04-16 9:36 ` [PATCH 05/12] fec: Codingstyle cleanups David Miller 2009-04-16 9:36 ` [PATCH 04/12] fec: remove unnecessary cast David Miller 1 sibling, 2 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 354 +++++++++++++++++++++-------------------------------- 1 files changed, 139 insertions(+), 215 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index 3a43d96..ab8e66b 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -86,8 +86,7 @@ static unsigned char fec_mac_default[] = { #endif #endif /* CONFIG_M5272 */ -/* Forward declarations of some structures to support different PHYs -*/ +/* Forward declarations of some structures to support different PHYs */ typedef struct { uint mii_data; @@ -123,8 +122,7 @@ typedef struct { #error "FEC: descriptor ring size constants too large" #endif -/* Interrupt events/masks. -*/ +/* Interrupt events/masks. */ #define FEC_ENET_HBERR ((uint)0x80000000) /* Heartbeat error */ #define FEC_ENET_BABR ((uint)0x40000000) /* Babbling receiver */ #define FEC_ENET_BABT ((uint)0x20000000) /* Babbling transmitter */ @@ -177,15 +175,14 @@ struct fec_enet_private { ushort skb_cur; ushort skb_dirty; - /* CPM dual port RAM relative addresses. - */ + /* CPM dual port RAM relative addresses */ dma_addr_t bd_dma; - /* Address of Rx and Tx buffers. */ + /* Address of Rx and Tx buffers */ struct bufdesc *rx_bd_base; struct bufdesc *tx_bd_base; /* The next free ring entry */ struct bufdesc *cur_rx, *cur_tx; - /* The ring entries to be free()ed. */ + /* The ring entries to be free()ed */ struct bufdesc *dirty_tx; uint tx_full; @@ -245,19 +242,16 @@ static mii_list_t *mii_tail; static int mii_queue(struct net_device *dev, int request, void (*func)(uint, struct net_device *)); -/* Make MII read/write commands for the FEC. -*/ +/* Make MII read/write commands for the FEC */ #define mk_mii_read(REG) (0x60020000 | ((REG & 0x1f) << 18)) #define mk_mii_write(REG, VAL) (0x50020000 | ((REG & 0x1f) << 18) | \ (VAL & 0xffff)) #define mk_mii_end 0 -/* Transmitter timeout. -*/ -#define TX_TIMEOUT (2*HZ) +/* Transmitter timeout */ +#define TX_TIMEOUT (2 * HZ) -/* Register definitions for the PHY. -*/ +/* Register definitions for the PHY */ #define MII_REG_CR 0 /* Control Register */ #define MII_REG_SR 1 /* Status Register */ @@ -307,7 +301,7 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) bdp = fep->cur_tx; status = bdp->cbd_sc; -#ifndef final_version + if (status & BD_ENET_TX_READY) { /* Ooops. All transmit buffers are full. Bail out. * This should not happen, since dev->tbusy should be set. @@ -316,21 +310,18 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) spin_unlock_irqrestore(&fep->hw_lock, flags); return 1; } -#endif - /* Clear all of the status flags. - */ + /* Clear all of the status flags */ status &= ~BD_ENET_TX_STATS; - /* Set buffer length and buffer pointer. - */ + /* Set buffer length and buffer pointer */ bdp->cbd_bufaddr = __pa(skb->data); bdp->cbd_datlen = skb->len; /* - * On some FEC implementations data must be aligned on - * 4-byte boundaries. Use bounce buffers to copy data - * and get it aligned. Ugh. + * On some FEC implementations data must be aligned on + * 4-byte boundaries. Use bounce buffers to copy data + * and get it aligned. Ugh. */ if (bdp->cbd_bufaddr & FEC_ALIGNMENT) { unsigned int index; @@ -339,8 +330,7 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) bdp->cbd_bufaddr = __pa(fep->tx_bounce[index]); } - /* Save skb pointer. - */ + /* Save skb pointer */ fep->tx_skbuff[fep->skb_cur] = skb; dev->stats.tx_bytes += skb->len; @@ -355,7 +345,6 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) /* Send it on its way. Tell FEC it's ready, interrupt when done, * it's the last BD of the frame, and to put the CRC on the end. */ - status |= (BD_ENET_TX_READY | BD_ENET_TX_INTR | BD_ENET_TX_LAST | BD_ENET_TX_TC); bdp->cbd_sc = status; @@ -365,13 +354,11 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) /* Trigger transmission start */ writel(0, fep->hwp + FEC_X_DES_ACTIVE); - /* If this was the last BD in the ring, start at the beginning again. - */ - if (status & BD_ENET_TX_WRAP) { + /* If this was the last BD in the ring, start at the beginning again. */ + if (status & BD_ENET_TX_WRAP) bdp = fep->tx_bd_base; - } else { + else bdp++; - } if (bdp == fep->dirty_tx) { fep->tx_full = 1; @@ -429,9 +416,6 @@ fec_timeout(struct net_device *dev) netif_wake_queue(dev); } -/* The interrupt handler. - * This is called from the MPC core interrupt. - */ static irqreturn_t fec_enet_interrupt(int irq, void * dev_id) { @@ -440,12 +424,10 @@ fec_enet_interrupt(int irq, void * dev_id) uint int_events; irqreturn_t ret = IRQ_NONE; - /* Get the interrupt events that caused us to be here. */ do { int_events = readl(fep->hwp + FEC_IEVENT); writel(int_events, fep->hwp + FEC_IEVENT); - /* Handle receive event in its own function. */ if (int_events & FEC_ENET_RXF) { ret = IRQ_HANDLED; fec_enet_rx(dev); @@ -506,31 +488,27 @@ fec_enet_tx(struct net_device *dev) dev->stats.tx_packets++; } -#ifndef final_version if (status & BD_ENET_TX_READY) printk("HEY! Enet xmit interrupt and TX_READY.\n"); -#endif + /* Deferred means some collisions occurred during transmit, * but we eventually sent the packet OK. */ if (status & BD_ENET_TX_DEF) dev->stats.collisions++; - /* Free the sk buffer associated with this last transmit. - */ + /* Free the sk buffer associated with this last transmit */ dev_kfree_skb_any(skb); fep->tx_skbuff[fep->skb_dirty] = NULL; fep->skb_dirty = (fep->skb_dirty + 1) & TX_RING_MOD_MASK; - /* Update pointer to next buffer descriptor to be transmitted. - */ + /* Update pointer to next buffer descriptor to be transmitted */ if (status & BD_ENET_TX_WRAP) bdp = fep->tx_bd_base; else bdp++; - /* Since we have freed up a buffer, the ring is no longer - * full. + /* Since we have freed up a buffer, the ring is no longer full */ if (fep->tx_full) { fep->tx_full = 0; @@ -569,114 +547,93 @@ fec_enet_rx(struct net_device *dev) */ bdp = fep->cur_rx; -while (!((status = bdp->cbd_sc) & BD_ENET_RX_EMPTY)) { + while (!((status = bdp->cbd_sc) & BD_ENET_RX_EMPTY)) { -#ifndef final_version - /* Since we have allocated space to hold a complete frame, - * the last indicator should be set. - */ - if ((status & BD_ENET_RX_LAST) == 0) - printk("FEC ENET: rcv is not +last\n"); -#endif + /* Since we have allocated space to hold a complete frame, + * the last indicator should be set. + */ + if ((status & BD_ENET_RX_LAST) == 0) + printk("FEC ENET: rcv is not +last\n"); - if (!fep->opened) - goto rx_processing_done; + if (!fep->opened) + goto rx_processing_done; - /* Check for errors. */ - if (status & (BD_ENET_RX_LG | BD_ENET_RX_SH | BD_ENET_RX_NO | + /* Check for errors. */ + if (status & (BD_ENET_RX_LG | BD_ENET_RX_SH | BD_ENET_RX_NO | BD_ENET_RX_CR | BD_ENET_RX_OV)) { - dev->stats.rx_errors++; - if (status & (BD_ENET_RX_LG | BD_ENET_RX_SH)) { - /* Frame too long or too short. */ - dev->stats.rx_length_errors++; + dev->stats.rx_errors++; + if (status & (BD_ENET_RX_LG | BD_ENET_RX_SH)) { + /* Frame too long or too short. */ + dev->stats.rx_length_errors++; + } + if (status & BD_ENET_RX_NO) /* Frame alignment */ + dev->stats.rx_frame_errors++; + if (status & BD_ENET_RX_CR) /* CRC Error */ + dev->stats.rx_crc_errors++; + if (status & BD_ENET_RX_OV) /* FIFO overrun */ + dev->stats.rx_fifo_errors++; } - if (status & BD_ENET_RX_NO) /* Frame alignment */ - dev->stats.rx_frame_errors++; - if (status & BD_ENET_RX_CR) /* CRC Error */ - dev->stats.rx_crc_errors++; - if (status & BD_ENET_RX_OV) /* FIFO overrun */ - dev->stats.rx_fifo_errors++; - } - /* Report late collisions as a frame error. - * On this error, the BD is closed, but we don't know what we - * have in the buffer. So, just drop this frame on the floor. - */ - if (status & BD_ENET_RX_CL) { - dev->stats.rx_errors++; - dev->stats.rx_frame_errors++; - goto rx_processing_done; - } + /* Report late collisions as a frame error. + * On this error, the BD is closed, but we don't know what we + * have in the buffer. So, just drop this frame on the floor. + */ + if (status & BD_ENET_RX_CL) { + dev->stats.rx_errors++; + dev->stats.rx_frame_errors++; + goto rx_processing_done; + } - /* Process the incoming frame. - */ - dev->stats.rx_packets++; - pkt_len = bdp->cbd_datlen; - dev->stats.rx_bytes += pkt_len; - data = (__u8*)__va(bdp->cbd_bufaddr); + /* Process the incoming frame. */ + dev->stats.rx_packets++; + pkt_len = bdp->cbd_datlen; + dev->stats.rx_bytes += pkt_len; + data = (__u8*)__va(bdp->cbd_bufaddr); - dma_sync_single(NULL, (unsigned long)__pa(data), + dma_sync_single(NULL, (unsigned long)__pa(data), pkt_len - 4, DMA_FROM_DEVICE); - /* This does 16 byte alignment, exactly what we need. - * The packet length includes FCS, but we don't want to - * include that when passing upstream as it messes up - * bridging applications. - */ - skb = dev_alloc_skb(pkt_len-4); - - if (skb == NULL) { - printk("%s: Memory squeeze, dropping packet.\n", dev->name); - dev->stats.rx_dropped++; - } else { - skb_put(skb,pkt_len-4); /* Make room */ - skb_copy_to_linear_data(skb, data, pkt_len-4); - skb->protocol=eth_type_trans(skb,dev); - netif_rx(skb); - } - rx_processing_done: - - /* Clear the status flags for this buffer. - */ - status &= ~BD_ENET_RX_STATS; + /* This does 16 byte alignment, exactly what we need. + * The packet length includes FCS, but we don't want to + * include that when passing upstream as it messes up + * bridging applications. + */ + skb = dev_alloc_skb(pkt_len - 4); - /* Mark the buffer empty. - */ - status |= BD_ENET_RX_EMPTY; - bdp->cbd_sc = status; + if (skb == NULL) { + printk("%s: Memory squeeze, dropping packet.\n", + dev->name); + dev->stats.rx_dropped++; + } else { + skb_put(skb, pkt_len - 4); /* Make room */ + skb_copy_to_linear_data(skb, data, pkt_len - 4); + skb->protocol = eth_type_trans(skb, dev); + netif_rx(skb); + } +rx_processing_done: + /* Clear the status flags for this buffer */ + status &= ~BD_ENET_RX_STATS; - /* Update BD pointer to next entry. - */ - if (status & BD_ENET_RX_WRAP) - bdp = fep->rx_bd_base; - else - bdp++; + /* Mark the buffer empty */ + status |= BD_ENET_RX_EMPTY; + bdp->cbd_sc = status; -#if 1 - /* Doing this here will keep the FEC running while we process - * incoming frames. On a heavily loaded network, we should be - * able to keep up at the expense of system resources. - */ - writel(0, fep->hwp + FEC_R_DES_ACTIVE); -#endif - } /* while (!((status = bdp->cbd_sc) & BD_ENET_RX_EMPTY)) */ + /* Update BD pointer to next entry */ + if (status & BD_ENET_RX_WRAP) + bdp = fep->rx_bd_base; + else + bdp++; + /* Doing this here will keep the FEC running while we process + * incoming frames. On a heavily loaded network, we should be + * able to keep up at the expense of system resources. + */ + writel(0, fep->hwp + FEC_R_DES_ACTIVE); + } fep->cur_rx = bdp; -#if 0 - /* Doing this here will allow us to process all frames in the - * ring before the FEC is allowed to put more there. On a heavily - * loaded network, some frames may be lost. Unfortunately, this - * increases the interrupt overhead since we can potentially work - * our way back to the interrupt return only to come right back - * here. - */ - fecp->fec_r_des_active = 0; -#endif - spin_unlock_irq(&fep->hw_lock); } - /* called from interrupt context */ static void fec_enet_mii(struct net_device *dev) @@ -714,8 +671,7 @@ mii_queue(struct net_device *dev, int regval, void (*func)(uint, struct net_devi mii_list_t *mip; int retval; - /* Add PHY address to register command. - */ + /* Add PHY address to register command */ fep = netdev_priv(dev); spin_lock_irqsave(&fep->mii_lock, flags); @@ -1359,11 +1315,6 @@ static void mii_relink(struct work_struct *work) fec_restart(dev, duplex); } else fec_stop(dev); - -#if 0 - enable_irq(fep->mii_irq); -#endif - } /* mii_queue_relink is called in interrupt context from mii_link_interrupt */ @@ -1372,12 +1323,12 @@ static void mii_queue_relink(uint mii_reg, struct net_device *dev) struct fec_enet_private *fep = netdev_priv(dev); /* - ** We cannot queue phy_task twice in the workqueue. It - ** would cause an endless loop in the workqueue. - ** Fortunately, if the last mii_relink entry has not yet been - ** executed now, it will do the job for the current interrupt, - ** which is just what we want. - */ + * We cannot queue phy_task twice in the workqueue. It + * would cause an endless loop in the workqueue. + * Fortunately, if the last mii_relink entry has not yet been + * executed now, it will do the job for the current interrupt, + * which is just what we want. + */ if (fep->mii_phy_task_queued) return; @@ -1408,8 +1359,7 @@ phy_cmd_t const phy_cmd_config[] = { { mk_mii_end, } }; -/* Read remainder of PHY ID. -*/ +/* Read remainder of PHY ID. */ static void mii_discover_phy3(uint mii_reg, struct net_device *dev) { @@ -1448,8 +1398,7 @@ mii_discover_phy(uint mii_reg, struct net_device *dev) if (fep->phy_addr < 32) { if ((phytype = (mii_reg & 0xffff)) != 0xffff && phytype != 0) { - /* Got first part of ID, now get remainder. - */ + /* Got first part of ID, now get remainder */ fep->phy_id = phytype << 16; mii_queue(dev, mk_mii_read(MII_REG_PHYIR2), mii_discover_phy3); @@ -1469,8 +1418,7 @@ mii_discover_phy(uint mii_reg, struct net_device *dev) } } -/* This interrupt occurs when the PHY detects a link change. -*/ +/* This interrupt occurs when the PHY detects a link change */ #ifdef HAVE_mii_link_interrupt static irqreturn_t mii_link_interrupt(int irq, void * dev_id) @@ -1480,10 +1428,6 @@ mii_link_interrupt(int irq, void * dev_id) fec_phy_ack_intr(); -#if 0 - disable_irq(fep->mii_irq); /* disable now, enable later */ -#endif - mii_do_cmd(dev, fep->phy->ack_int); mii_do_cmd(dev, phy_cmd_relink); /* restart and display status */ @@ -1534,7 +1478,7 @@ fec_enet_open(struct net_device *dev) netif_start_queue(dev); fep->opened = 1; - return 0; /* Success */ + return 0; } static int @@ -1542,8 +1486,7 @@ fec_enet_close(struct net_device *dev) { struct fec_enet_private *fep = netdev_priv(dev); - /* Don't know what to do yet. - */ + /* Don't know what to do yet. */ fep->opened = 0; netif_stop_queue(dev); fec_stop(dev); @@ -1571,7 +1514,7 @@ static void set_multicast_list(struct net_device *dev) unsigned int i, j, bit, data, crc, tmp; unsigned char hash; - if (dev->flags&IFF_PROMISC) { + if (dev->flags & IFF_PROMISC) { tmp = readl(fep->hwp + FEC_R_CNTRL); tmp |= 0x8; writel(tmp, fep->hwp + FEC_R_CNTRL); @@ -1582,42 +1525,37 @@ static void set_multicast_list(struct net_device *dev) if (dev->flags & IFF_ALLMULTI) { /* Catch all multicast addresses, so set the - * filter to all 1's. + * filter to all 1's */ writel(0xffffffff, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); writel(0xffffffff, fep->hwp + FEC_GRP_HASH_TABLE_LOW); } else { - /* Clear filter and add the addresses in hash register. - */ + /* Clear filter and add the addresses in hash register + */ writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); dmi = dev->mc_list; - for (j = 0; j < dev->mc_count; j++, dmi = dmi->next) - { - /* Only support group multicast for now. - */ + for (j = 0; j < dev->mc_count; j++, dmi = dmi->next) { + /* Only support group multicast for now */ if (!(dmi->dmi_addr[0] & 1)) continue; - /* calculate crc32 value of mac address - */ + /* calculate crc32 value of mac address */ crc = 0xffffffff; - for (i = 0; i < dmi->dmi_addrlen; i++) - { + for (i = 0; i < dmi->dmi_addrlen; i++) { data = dmi->dmi_addr[i]; - for (bit = 0; bit < 8; bit++, data >>= 1) - { + for (bit = 0; bit < 8; bit++, data >>= 1) { crc = (crc >> 1) ^ (((crc ^ data) & 1) ? CRC32_POLY : 0); } } /* only upper 6 bits (HASH_BITS) are used - which point to specific bit in he hash registers - */ + * which point to specific bit in he hash registers + */ hash = (crc >> (32 - HASH_BITS)) & 0x3f; if (hash > 31) { @@ -1634,8 +1572,7 @@ static void set_multicast_list(struct net_device *dev) } } -/* Set a MAC change in hardware. - */ +/* Set a MAC change in hardware. */ static void fec_set_mac_address(struct net_device *dev) { @@ -1676,8 +1613,7 @@ int __init fec_enet_init(struct net_device *dev, int index) fep->hwp = (void __iomem *)dev->base_addr; fep->netdev = dev; - /* Whack a reset. We should wait for this. - */ + /* Whack a reset. We should wait for this. */ writel(1, fep->hwp + FEC_ECNTRL); udelay(10); @@ -1707,18 +1643,15 @@ int __init fec_enet_init(struct net_device *dev, int index) fep->skb_cur = fep->skb_dirty = 0; - /* Initialize the receive buffer descriptors. - */ + /* Initialize the receive buffer descriptors. */ bdp = fep->rx_bd_base; for (i=0; i<FEC_ENET_RX_PAGES; i++) { - /* Allocate a page. - */ + /* Allocate a page */ mem_addr = __get_free_page(GFP_KERNEL); /* XXX: missing check for allocation failure */ - /* Initialize the BD for every fragment in the page. - */ + /* Initialize the BD for every fragment in the page */ for (j=0; j<FEC_ENET_RX_FRPPG; j++) { bdp->cbd_sc = BD_ENET_RX_EMPTY; bdp->cbd_bufaddr = __pa(mem_addr); @@ -1727,13 +1660,11 @@ int __init fec_enet_init(struct net_device *dev, int index) } } - /* Set the last buffer to wrap. - */ + /* Set the last buffer to wrap */ bdp--; bdp->cbd_sc |= BD_SC_WRAP; - /* ...and the same for transmmit. - */ + /* ...and the same for transmit */ bdp = fep->tx_bd_base; for (i=0, j=FEC_ENET_TX_FRPPG; i<TX_RING_SIZE; i++) { if (j >= FEC_ENET_TX_FRPPG) { @@ -1745,20 +1676,17 @@ int __init fec_enet_init(struct net_device *dev, int index) } fep->tx_bounce[i] = (unsigned char *) mem_addr; - /* Initialize the BD for every fragment in the page. - */ + /* Initialize the BD for every fragment in the page */ bdp->cbd_sc = 0; bdp->cbd_bufaddr = 0; bdp++; } - /* Set the last buffer to wrap. - */ + /* Set the last buffer to wrap */ bdp--; bdp->cbd_sc |= BD_SC_WRAP; - /* Set receive and transmit descriptor base. - */ + /* Set receive and transmit descriptor base */ writel(fep->bd_dma, fep->hwp + FEC_R_DES_START); writel((unsigned long)fep->bd_dma + sizeof(struct bufdesc) * RX_RING_SIZE, fep->hwp + FEC_X_DES_START); @@ -1777,7 +1705,7 @@ int __init fec_enet_init(struct net_device *dev, int index) writel(0, fep->hwp + FEC_HASH_TABLE_LOW); #endif - /* The FEC Ethernet specific entries in the device structure. */ + /* The FEC Ethernet specific entries in the device structure */ dev->open = fec_enet_open; dev->hard_start_xmit = fec_enet_start_xmit; dev->tx_timeout = fec_timeout; @@ -1793,9 +1721,7 @@ int __init fec_enet_init(struct net_device *dev, int index) writel(OPT_FRAME_SIZE | 0x04, fep->hwp + FEC_R_CNTRL); writel(0, fep->hwp + FEC_X_CNTRL); - /* - * Set MII speed to 2.5 MHz - */ + /* Set MII speed to 2.5 MHz */ fep->phy_speed = ((((clk_get_rate(fep->clk) / 2 + 4999999) / 2500000) / 2) & 0x3F) << 1; writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED); @@ -1854,8 +1780,8 @@ fec_restart(struct net_device *dev, int duplex) /* Reset SKB transmit buffers. */ fep->skb_cur = fep->skb_dirty = 0; - for (i=0; i<=TX_RING_MOD_MASK; i++) { - if (fep->tx_skbuff[i] != NULL) { + for (i = 0; i <= TX_RING_MOD_MASK; i++) { + if (fep->tx_skbuff[i]) { dev_kfree_skb_any(fep->tx_skbuff[i]); fep->tx_skbuff[i] = NULL; } @@ -1863,20 +1789,20 @@ fec_restart(struct net_device *dev, int duplex) /* Initialize the receive buffer descriptors. */ bdp = fep->rx_bd_base; - for (i=0; i<RX_RING_SIZE; i++) { + for (i = 0; i < RX_RING_SIZE; i++) { /* Initialize the BD for every fragment in the page. */ bdp->cbd_sc = BD_ENET_RX_EMPTY; bdp++; } - /* Set the last buffer to wrap. */ + /* Set the last buffer to wrap */ bdp--; bdp->cbd_sc |= BD_SC_WRAP; - /* ...and the same for transmmit. */ + /* ...and the same for transmit */ bdp = fep->tx_bd_base; - for (i=0; i<TX_RING_SIZE; i++) { + for (i = 0; i < TX_RING_SIZE; i++) { /* Initialize the BD for every fragment in the page. */ bdp->cbd_sc = 0; @@ -1884,11 +1810,11 @@ fec_restart(struct net_device *dev, int duplex) bdp++; } - /* Set the last buffer to wrap. */ + /* Set the last buffer to wrap */ bdp--; bdp->cbd_sc |= BD_SC_WRAP; - /* Enable MII mode. */ + /* Enable MII mode */ if (duplex) { /* MII enable / FD enable */ writel(OPT_FRAME_SIZE | 0x04, fep->hwp + FEC_R_CNTRL); @@ -1900,14 +1826,14 @@ fec_restart(struct net_device *dev, int duplex) } fep->full_duplex = duplex; - /* Set MII speed. */ + /* Set MII speed */ writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED); - /* And last, enable the transmit and receive processing. */ + /* And last, enable the transmit and receive processing */ writel(2, fep->hwp + FEC_ECNTRL); writel(0, fep->hwp + FEC_R_DES_ACTIVE); - /* Enable interrupts we wish to service. */ + /* Enable interrupts we wish to service */ writel(FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII, fep->hwp + FEC_IMASK); } @@ -1917,9 +1843,7 @@ fec_stop(struct net_device *dev) { struct fec_enet_private *fep = netdev_priv(dev); - /* - ** We cannot expect a graceful transmit stop without link !!! - */ + /* We cannot expect a graceful transmit stop without link !!! */ if (fep->link) { writel(1, fep->hwp + FEC_X_CNTRL); /* Graceful transmit stop */ udelay(10); -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 06/12] fec: refactor set_multicast_list() to make it more readable 2009-04-15 11:32 ` [PATCH 05/12] fec: Codingstyle cleanups Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 07/12] fec: refactor init function Sascha Hauer 2009-04-16 9:37 ` [PATCH 06/12] fec: refactor set_multicast_list() to make it more readable David Miller 2009-04-16 9:36 ` [PATCH 05/12] fec: Codingstyle cleanups David Miller 1 sibling, 2 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 99 +++++++++++++++++++++++++++------------------------- 1 files changed, 51 insertions(+), 48 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index ab8e66b..b72df48 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -1518,57 +1518,60 @@ static void set_multicast_list(struct net_device *dev) tmp = readl(fep->hwp + FEC_R_CNTRL); tmp |= 0x8; writel(tmp, fep->hwp + FEC_R_CNTRL); - } else { - tmp = readl(fep->hwp + FEC_R_CNTRL); - tmp &= ~0x8; - writel(tmp, fep->hwp + FEC_R_CNTRL); + return; + } - if (dev->flags & IFF_ALLMULTI) { - /* Catch all multicast addresses, so set the - * filter to all 1's - */ - writel(0xffffffff, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); - writel(0xffffffff, fep->hwp + FEC_GRP_HASH_TABLE_LOW); - } else { - /* Clear filter and add the addresses in hash register - */ - writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); - writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); - - dmi = dev->mc_list; - - for (j = 0; j < dev->mc_count; j++, dmi = dmi->next) { - /* Only support group multicast for now */ - if (!(dmi->dmi_addr[0] & 1)) - continue; - - /* calculate crc32 value of mac address */ - crc = 0xffffffff; - - for (i = 0; i < dmi->dmi_addrlen; i++) { - data = dmi->dmi_addr[i]; - for (bit = 0; bit < 8; bit++, data >>= 1) { - crc = (crc >> 1) ^ - (((crc ^ data) & 1) ? CRC32_POLY : 0); - } - } - - /* only upper 6 bits (HASH_BITS) are used - * which point to specific bit in he hash registers - */ - hash = (crc >> (32 - HASH_BITS)) & 0x3f; - - if (hash > 31) { - tmp = readl(fep->hwp + FEC_GRP_HASH_TABLE_HIGH); - tmp |= 1 << (hash - 32); - writel(tmp, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); - } else { - tmp = readl(fep->hwp + FEC_GRP_HASH_TABLE_LOW); - tmp |= 1 << hash; - writel(tmp, fep->hwp + FEC_GRP_HASH_TABLE_LOW); - } + tmp = readl(fep->hwp + FEC_R_CNTRL); + tmp &= ~0x8; + writel(tmp, fep->hwp + FEC_R_CNTRL); + + if (dev->flags & IFF_ALLMULTI) { + /* Catch all multicast addresses, so set the + * filter to all 1's + */ + writel(0xffffffff, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + writel(0xffffffff, fep->hwp + FEC_GRP_HASH_TABLE_LOW); + + return; + } + + /* Clear filter and add the addresses in hash register + */ + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); + + dmi = dev->mc_list; + + for (j = 0; j < dev->mc_count; j++, dmi = dmi->next) { + /* Only support group multicast for now */ + if (!(dmi->dmi_addr[0] & 1)) + continue; + + /* calculate crc32 value of mac address */ + crc = 0xffffffff; + + for (i = 0; i < dmi->dmi_addrlen; i++) { + data = dmi->dmi_addr[i]; + for (bit = 0; bit < 8; bit++, data >>= 1) { + crc = (crc >> 1) ^ + (((crc ^ data) & 1) ? CRC32_POLY : 0); } } + + /* only upper 6 bits (HASH_BITS) are used + * which point to specific bit in he hash registers + */ + hash = (crc >> (32 - HASH_BITS)) & 0x3f; + + if (hash > 31) { + tmp = readl(fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + tmp |= 1 << (hash - 32); + writel(tmp, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + } else { + tmp = readl(fep->hwp + FEC_GRP_HASH_TABLE_LOW); + tmp |= 1 << hash; + writel(tmp, fep->hwp + FEC_GRP_HASH_TABLE_LOW); + } } } -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 07/12] fec: refactor init function 2009-04-15 11:32 ` [PATCH 06/12] fec: refactor set_multicast_list() to make it more readable Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 08/12] fec: align receive packets Sascha Hauer 2009-04-16 9:37 ` [PATCH 07/12] fec: refactor init function David Miller 2009-04-16 9:37 ` [PATCH 06/12] fec: refactor set_multicast_list() to make it more readable David Miller 1 sibling, 2 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer fec_enet_init() does the hardware initialisation and then calls fec_restart() which does the same initialisation again, so we can safely remove the initialisation from fec_enet_init(). Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 39 ++++----------------------------------- 1 files changed, 4 insertions(+), 35 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index b72df48..a3037ab 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -1616,10 +1616,6 @@ int __init fec_enet_init(struct net_device *dev, int index) fep->hwp = (void __iomem *)dev->base_addr; fep->netdev = dev; - /* Whack a reset. We should wait for this. */ - writel(1, fep->hwp + FEC_ECNTRL); - udelay(10); - /* Set the Ethernet address */ #ifdef CONFIG_M5272 fec_get_mac(dev); @@ -1641,11 +1637,6 @@ int __init fec_enet_init(struct net_device *dev, int index) fep->rx_bd_base = cbd_base; fep->tx_bd_base = cbd_base + RX_RING_SIZE; - fep->dirty_tx = fep->cur_tx = fep->tx_bd_base; - fep->cur_rx = fep->rx_bd_base; - - fep->skb_cur = fep->skb_dirty = 0; - /* Initialize the receive buffer descriptors. */ bdp = fep->rx_bd_base; for (i=0; i<FEC_ENET_RX_PAGES; i++) { @@ -1689,25 +1680,9 @@ int __init fec_enet_init(struct net_device *dev, int index) bdp--; bdp->cbd_sc |= BD_SC_WRAP; - /* Set receive and transmit descriptor base */ - writel(fep->bd_dma, fep->hwp + FEC_R_DES_START); - writel((unsigned long)fep->bd_dma + sizeof(struct bufdesc) * RX_RING_SIZE, - fep->hwp + FEC_X_DES_START); - #ifdef HAVE_mii_link_interrupt fec_request_mii_intr(dev); #endif - - writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); - writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); - writel(PKT_MAXBLR_SIZE, fep->hwp + FEC_R_BUFF_SIZE); - writel(2, fep->hwp + FEC_ECNTRL); - writel(0, fep->hwp + FEC_R_DES_ACTIVE); -#ifndef CONFIG_M5272 - writel(0, fep->hwp + FEC_HASH_TABLE_HIGH); - writel(0, fep->hwp + FEC_HASH_TABLE_LOW); -#endif - /* The FEC Ethernet specific entries in the device structure */ dev->open = fec_enet_open; dev->hard_start_xmit = fec_enet_start_xmit; @@ -1720,21 +1695,11 @@ int __init fec_enet_init(struct net_device *dev, int index) mii_cmds[i].mii_next = &mii_cmds[i+1]; mii_free = mii_cmds; - /* setup MII interface */ - writel(OPT_FRAME_SIZE | 0x04, fep->hwp + FEC_R_CNTRL); - writel(0, fep->hwp + FEC_X_CNTRL); - /* Set MII speed to 2.5 MHz */ fep->phy_speed = ((((clk_get_rate(fep->clk) / 2 + 4999999) / 2500000) / 2) & 0x3F) << 1; - writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED); fec_restart(dev, 0); - /* Clear and enable interrupts */ - writel(0xffc00000, fep->hwp + FEC_IEVENT); - writel(FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII, - fep->hwp + FEC_IMASK); - /* Queue up command to detect the PHY and initialize the * remainder of the interface. */ @@ -1769,6 +1734,10 @@ fec_restart(struct net_device *dev, int duplex) /* Reset all multicast. */ writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); +#ifndef CONFIG_M5272 + writel(0, fep->hwp + FEC_HASH_TABLE_HIGH); + writel(0, fep->hwp + FEC_HASH_TABLE_LOW); +#endif /* Set maximum receive buffer size. */ writel(PKT_MAXBLR_SIZE, fep->hwp + FEC_R_BUFF_SIZE); -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 08/12] fec: align receive packets 2009-04-15 11:32 ` [PATCH 07/12] fec: refactor init function Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 09/12] fec: remove debugging printks Sascha Hauer 2009-04-16 9:37 ` [PATCH 08/12] fec: align receive packets David Miller 2009-04-16 9:37 ` [PATCH 07/12] fec: refactor init function David Miller 1 sibling, 2 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer Otherwise we get a lot of alignment errors Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index a3037ab..be3226c 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -598,13 +598,14 @@ fec_enet_rx(struct net_device *dev) * include that when passing upstream as it messes up * bridging applications. */ - skb = dev_alloc_skb(pkt_len - 4); + skb = dev_alloc_skb(pkt_len - 4 + NET_IP_ALIGN); - if (skb == NULL) { + if (unlikely(!skb)) { printk("%s: Memory squeeze, dropping packet.\n", dev->name); dev->stats.rx_dropped++; } else { + skb_reserve(skb, NET_IP_ALIGN); skb_put(skb, pkt_len - 4); /* Make room */ skb_copy_to_linear_data(skb, data, pkt_len - 4); skb->protocol = eth_type_trans(skb, dev); -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 09/12] fec: remove debugging printks 2009-04-15 11:32 ` [PATCH 08/12] fec: align receive packets Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 10/12] fec: switch to net_device_ops Sascha Hauer 2009-04-16 9:37 ` [PATCH 09/12] fec: remove debugging printks David Miller 2009-04-16 9:37 ` [PATCH 08/12] fec: align receive packets David Miller 1 sibling, 2 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer These printks in fec_timeout do not give useful information in a production kernel. Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 33 --------------------------------- 1 files changed, 0 insertions(+), 33 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index be3226c..d96aba9 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -377,41 +377,8 @@ fec_timeout(struct net_device *dev) { struct fec_enet_private *fep = netdev_priv(dev); - printk("%s: transmit timed out.\n", dev->name); dev->stats.tx_errors++; -#ifndef final_version - { - int i; - struct bufdesc *bdp; - - printk("Ring data dump: cur_tx %lx%s, dirty_tx %lx cur_rx: %lx\n", - (unsigned long)fep->cur_tx, fep->tx_full ? " (full)" : "", - (unsigned long)fep->dirty_tx, - (unsigned long)fep->cur_rx); - - bdp = fep->tx_bd_base; - printk(" tx: %u buffers\n", TX_RING_SIZE); - for (i = 0 ; i < TX_RING_SIZE; i++) { - printk(" %08x: %04x %04x %08x\n", - (uint) bdp, - bdp->cbd_sc, - bdp->cbd_datlen, - (int) bdp->cbd_bufaddr); - bdp++; - } - bdp = fep->rx_bd_base; - printk(" rx: %lu buffers\n", (unsigned long) RX_RING_SIZE); - for (i = 0 ; i < RX_RING_SIZE; i++) { - printk(" %08x: %04x %04x %08x\n", - (uint) bdp, - bdp->cbd_sc, - bdp->cbd_datlen, - (int) bdp->cbd_bufaddr); - bdp++; - } - } -#endif fec_restart(dev, fep->full_duplex); netif_wake_queue(dev); } -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 10/12] fec: switch to net_device_ops 2009-04-15 11:32 ` [PATCH 09/12] fec: remove debugging printks Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 11/12] FEC Buffer rework Sascha Hauer 2009-04-16 9:37 ` [PATCH 10/12] fec: switch to net_device_ops David Miller 2009-04-16 9:37 ` [PATCH 09/12] fec: remove debugging printks David Miller 1 sibling, 2 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 36 ++++++++++++++++++++---------------- 1 files changed, 20 insertions(+), 16 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index d96aba9..f4afbe9 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -210,17 +210,13 @@ struct fec_enet_private { int full_duplex; }; -static int fec_enet_open(struct net_device *dev); -static int fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev); static void fec_enet_mii(struct net_device *dev); static irqreturn_t fec_enet_interrupt(int irq, void * dev_id); static void fec_enet_tx(struct net_device *dev); static void fec_enet_rx(struct net_device *dev); static int fec_enet_close(struct net_device *dev); -static void set_multicast_list(struct net_device *dev); static void fec_restart(struct net_device *dev, int duplex); static void fec_stop(struct net_device *dev); -static void fec_set_mac_address(struct net_device *dev); /* MII processing. We keep this as simple as possible. Requests are @@ -1411,7 +1407,6 @@ fec_enet_open(struct net_device *dev) /* I should reset the ring buffers here, but I don't yet know * a simple way to do that. */ - fec_set_mac_address(dev); fep->sequence_done = 0; fep->link = 0; @@ -1544,19 +1539,35 @@ static void set_multicast_list(struct net_device *dev) } /* Set a MAC change in hardware. */ -static void -fec_set_mac_address(struct net_device *dev) +static int +fec_set_mac_address(struct net_device *dev, void *p) { struct fec_enet_private *fep = netdev_priv(dev); + struct sockaddr *addr = p; + + if (!is_valid_ether_addr(addr->sa_data)) + return -EADDRNOTAVAIL; + + memcpy(dev->dev_addr, addr->sa_data, dev->addr_len); - /* Set station address. */ writel(dev->dev_addr[3] | (dev->dev_addr[2] << 8) | (dev->dev_addr[1] << 16) | (dev->dev_addr[0] << 24), fep->hwp + FEC_ADDR_LOW); writel((dev->dev_addr[5] << 16) | (dev->dev_addr[4] << 24), fep + FEC_ADDR_HIGH); + return 0; } +static const struct net_device_ops fec_netdev_ops = { + .ndo_open = fec_enet_open, + .ndo_stop = fec_enet_close, + .ndo_start_xmit = fec_enet_start_xmit, + .ndo_set_multicast_list = set_multicast_list, + .ndo_validate_addr = eth_validate_addr, + .ndo_tx_timeout = fec_timeout, + .ndo_set_mac_address = fec_set_mac_address, +}; + /* * XXX: We need to clean up on failure exits here. * @@ -1652,12 +1663,8 @@ int __init fec_enet_init(struct net_device *dev, int index) fec_request_mii_intr(dev); #endif /* The FEC Ethernet specific entries in the device structure */ - dev->open = fec_enet_open; - dev->hard_start_xmit = fec_enet_start_xmit; - dev->tx_timeout = fec_timeout; dev->watchdog_timeo = TX_TIMEOUT; - dev->stop = fec_enet_close; - dev->set_multicast_list = set_multicast_list; + dev->netdev_ops = &fec_netdev_ops; for (i=0; i<NMII-1; i++) mii_cmds[i].mii_next = &mii_cmds[i+1]; @@ -1696,9 +1703,6 @@ fec_restart(struct net_device *dev, int duplex) /* Clear any outstanding interrupt. */ writel(0xffc00000, fep->hwp + FEC_IEVENT); - /* Set station address. */ - fec_set_mac_address(dev); - /* Reset all multicast. */ writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 11/12] FEC Buffer rework 2009-04-15 11:32 ` [PATCH 10/12] fec: switch to net_device_ops Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-15 11:32 ` [PATCH 12/12] fec: call fec_restart() in fec_open() Sascha Hauer ` (2 more replies) 2009-04-16 9:37 ` [PATCH 10/12] fec: switch to net_device_ops David Miller 1 sibling, 3 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer Allocate buffers in fec_open and free them again in fec_close. This makes it possible to use this driver as a module. Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 139 +++++++++++++++++++++++++++++++++------------------- 1 files changed, 88 insertions(+), 51 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index f4afbe9..0e1d268 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -172,6 +172,7 @@ struct fec_enet_private { /* The saved address of a sent-in-place packet/buffer, for skfree(). */ unsigned char *tx_bounce[TX_RING_SIZE]; struct sk_buff* tx_skbuff[TX_RING_SIZE]; + struct sk_buff* rx_skbuff[RX_RING_SIZE]; ushort skb_cur; ushort skb_dirty; @@ -335,8 +336,8 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) /* Push the data cache so the CPM does not get stale memory * data. */ - dma_sync_single(NULL, bdp->cbd_bufaddr, - bdp->cbd_datlen, DMA_TO_DEVICE); + bdp->cbd_bufaddr = dma_map_single(&dev->dev, skb->data, + FEC_ENET_TX_FRSIZE, DMA_TO_DEVICE); /* Send it on its way. Tell FEC it's ready, interrupt when done, * it's the last BD of the frame, and to put the CRC on the end. @@ -429,7 +430,11 @@ fec_enet_tx(struct net_device *dev) bdp = fep->dirty_tx; while (((status = bdp->cbd_sc) & BD_ENET_TX_READY) == 0) { - if (bdp == fep->cur_tx && fep->tx_full == 0) break; + if (bdp == fep->cur_tx && fep->tx_full == 0) + break; + + dma_unmap_single(&dev->dev, bdp->cbd_bufaddr, FEC_ENET_TX_FRSIZE, DMA_TO_DEVICE); + bdp->cbd_bufaddr = 0; skb = fep->tx_skbuff[fep->skb_dirty]; /* Check for errors. */ @@ -553,8 +558,8 @@ fec_enet_rx(struct net_device *dev) dev->stats.rx_bytes += pkt_len; data = (__u8*)__va(bdp->cbd_bufaddr); - dma_sync_single(NULL, (unsigned long)__pa(data), - pkt_len - 4, DMA_FROM_DEVICE); + dma_unmap_single(NULL, bdp->cbd_bufaddr, bdp->cbd_datlen, + DMA_FROM_DEVICE); /* This does 16 byte alignment, exactly what we need. * The packet length includes FCS, but we don't want to @@ -574,6 +579,9 @@ fec_enet_rx(struct net_device *dev) skb->protocol = eth_type_trans(skb, dev); netif_rx(skb); } + + bdp->cbd_bufaddr = dma_map_single(NULL, data, bdp->cbd_datlen, + DMA_FROM_DEVICE); rx_processing_done: /* Clear the status flags for this buffer */ status &= ~BD_ENET_RX_STATS; @@ -1399,15 +1407,86 @@ mii_link_interrupt(int irq, void * dev_id) } #endif +static void fec_enet_free_buffers(struct net_device *dev) +{ + struct fec_enet_private *fep = netdev_priv(dev); + int i; + struct sk_buff *skb; + struct bufdesc *bdp; + + bdp = fep->rx_bd_base; + for (i = 0; i < RX_RING_SIZE; i++) { + skb = fep->rx_skbuff[i]; + + if (bdp->cbd_bufaddr) + dma_unmap_single(&dev->dev, bdp->cbd_bufaddr, + FEC_ENET_RX_FRSIZE, DMA_FROM_DEVICE); + if (skb) + dev_kfree_skb(skb); + bdp++; + } + + bdp = fep->tx_bd_base; + for (i = 0; i < TX_RING_SIZE; i++) + kfree(fep->tx_bounce[i]); +} + +static int fec_enet_alloc_buffers(struct net_device *dev) +{ + struct fec_enet_private *fep = netdev_priv(dev); + int i; + struct sk_buff *skb; + struct bufdesc *bdp; + + bdp = fep->rx_bd_base; + for (i = 0; i < RX_RING_SIZE; i++) { + skb = dev_alloc_skb(FEC_ENET_RX_FRSIZE); + if (!skb) { + fec_enet_free_buffers(dev); + return -ENOMEM; + } + fep->rx_skbuff[i] = skb; + + bdp->cbd_bufaddr = dma_map_single(&dev->dev, skb->data, + FEC_ENET_RX_FRSIZE, DMA_FROM_DEVICE); + bdp->cbd_sc = BD_ENET_RX_EMPTY; + bdp++; + } + + /* Set the last buffer to wrap. */ + bdp--; + bdp->cbd_sc |= BD_SC_WRAP; + + bdp = fep->tx_bd_base; + for (i = 0; i < TX_RING_SIZE; i++) { + fep->tx_bounce[i] = kmalloc(FEC_ENET_TX_FRSIZE, GFP_KERNEL); + + bdp->cbd_sc = 0; + bdp->cbd_bufaddr = 0; + bdp++; + } + + /* Set the last buffer to wrap. */ + bdp--; + bdp->cbd_sc |= BD_SC_WRAP; + + return 0; +} + static int fec_enet_open(struct net_device *dev) { struct fec_enet_private *fep = netdev_priv(dev); + int ret; /* I should reset the ring buffers here, but I don't yet know * a simple way to do that. */ + ret = fec_enet_alloc_buffers(dev); + if (ret) + return ret; + fep->sequence_done = 0; fep->link = 0; @@ -1454,6 +1533,8 @@ fec_enet_close(struct net_device *dev) netif_stop_queue(dev); fec_stop(dev); + fec_enet_free_buffers(dev); + return 0; } @@ -1576,9 +1657,8 @@ static const struct net_device_ops fec_netdev_ops = { int __init fec_enet_init(struct net_device *dev, int index) { struct fec_enet_private *fep = netdev_priv(dev); - unsigned long mem_addr; - struct bufdesc *bdp, *cbd_base; - int i, j; + struct bufdesc *cbd_base; + int i; /* Allocate memory for buffer descriptors. */ cbd_base = dma_alloc_coherent(NULL, PAGE_SIZE, &fep->bd_dma, @@ -1616,49 +1696,6 @@ int __init fec_enet_init(struct net_device *dev, int index) fep->rx_bd_base = cbd_base; fep->tx_bd_base = cbd_base + RX_RING_SIZE; - /* Initialize the receive buffer descriptors. */ - bdp = fep->rx_bd_base; - for (i=0; i<FEC_ENET_RX_PAGES; i++) { - - /* Allocate a page */ - mem_addr = __get_free_page(GFP_KERNEL); - /* XXX: missing check for allocation failure */ - - /* Initialize the BD for every fragment in the page */ - for (j=0; j<FEC_ENET_RX_FRPPG; j++) { - bdp->cbd_sc = BD_ENET_RX_EMPTY; - bdp->cbd_bufaddr = __pa(mem_addr); - mem_addr += FEC_ENET_RX_FRSIZE; - bdp++; - } - } - - /* Set the last buffer to wrap */ - bdp--; - bdp->cbd_sc |= BD_SC_WRAP; - - /* ...and the same for transmit */ - bdp = fep->tx_bd_base; - for (i=0, j=FEC_ENET_TX_FRPPG; i<TX_RING_SIZE; i++) { - if (j >= FEC_ENET_TX_FRPPG) { - mem_addr = __get_free_page(GFP_KERNEL); - j = 1; - } else { - mem_addr += FEC_ENET_TX_FRSIZE; - j++; - } - fep->tx_bounce[i] = (unsigned char *) mem_addr; - - /* Initialize the BD for every fragment in the page */ - bdp->cbd_sc = 0; - bdp->cbd_bufaddr = 0; - bdp++; - } - - /* Set the last buffer to wrap */ - bdp--; - bdp->cbd_sc |= BD_SC_WRAP; - #ifdef HAVE_mii_link_interrupt fec_request_mii_intr(dev); #endif -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH 12/12] fec: call fec_restart() in fec_open() 2009-04-15 11:32 ` [PATCH 11/12] FEC Buffer rework Sascha Hauer @ 2009-04-15 11:32 ` Sascha Hauer 2009-04-16 9:38 ` David Miller 2009-04-16 9:38 ` [PATCH 11/12] FEC Buffer rework David Miller 2009-04-17 10:07 ` Greg Ungerer 2 siblings, 1 reply; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 11:32 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer, Sascha Hauer We called fec_stop() in fec_enet_close(), thus we have to call fec_restart() in fec_enet_open(). Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 18 ++++++++---------- 1 files changed, 8 insertions(+), 10 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index 0e1d268..28db691 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -1490,6 +1490,8 @@ fec_enet_open(struct net_device *dev) fep->sequence_done = 0; fep->link = 0; + fec_restart(dev, 1); + if (fep->phy) { mii_do_cmd(dev, fep->phy->ack_int); mii_do_cmd(dev, fep->phy->config); @@ -1506,18 +1508,14 @@ fec_enet_open(struct net_device *dev) schedule(); mii_do_cmd(dev, fep->phy->startup); - - /* Set the initial link state to true. A lot of hardware - * based on this device does not implement a PHY interrupt, - * so we are never notified of link change. - */ - fep->link = 1; - } else { - fep->link = 1; /* lets just try it and see */ - /* no phy, go full duplex, it's most likely a hub chip */ - fec_restart(dev, 1); } + /* Set the initial link state to true. A lot of hardware + * based on this device does not implement a PHY interrupt, + * so we are never notified of link change. + */ + fep->link = 1; + netif_start_queue(dev); fep->opened = 1; return 0; -- 1.6.2.1 ^ permalink raw reply related [flat|nested] 29+ messages in thread
* Re: [PATCH 12/12] fec: call fec_restart() in fec_open() 2009-04-15 11:32 ` [PATCH 12/12] fec: call fec_restart() in fec_open() Sascha Hauer @ 2009-04-16 9:38 ` David Miller 0 siblings, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:38 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 13:32:25 +0200 > We called fec_stop() in fec_enet_close(), thus we have to call > fec_restart() in fec_enet_open(). > > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 11/12] FEC Buffer rework 2009-04-15 11:32 ` [PATCH 11/12] FEC Buffer rework Sascha Hauer 2009-04-15 11:32 ` [PATCH 12/12] fec: call fec_restart() in fec_open() Sascha Hauer @ 2009-04-16 9:38 ` David Miller 2009-04-17 10:07 ` Greg Ungerer 2 siblings, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:38 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 13:32:24 +0200 > Allocate buffers in fec_open and free them again in fec_close. This makes > it possible to use this driver as a module. > > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 11/12] FEC Buffer rework 2009-04-15 11:32 ` [PATCH 11/12] FEC Buffer rework Sascha Hauer 2009-04-15 11:32 ` [PATCH 12/12] fec: call fec_restart() in fec_open() Sascha Hauer 2009-04-16 9:38 ` [PATCH 11/12] FEC Buffer rework David Miller @ 2009-04-17 10:07 ` Greg Ungerer 2009-04-17 10:12 ` Sascha Hauer 2 siblings, 1 reply; 29+ messages in thread From: Greg Ungerer @ 2009-04-17 10:07 UTC (permalink / raw) To: Sascha Hauer; +Cc: netdev Hi Sascha, Sascha Hauer wrote: > Allocate buffers in fec_open and free them again in fec_close. This makes > it possible to use this driver as a module. > > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> This is the only one that doesn't work for ColdFire. There is no dma_map_single() or dma_unmap_single() currently. I will fix that, so I am fine with this anyways. With all the other patches applied it still basically works on ColdFire (at least on initial testing on a 5208). I have seen a few spurious interrupts though with these changes applied, that didn't happen before. I will investigate further when get a few minutes. Regards Greg > --- > drivers/net/fec.c | 139 +++++++++++++++++++++++++++++++++------------------- > 1 files changed, 88 insertions(+), 51 deletions(-) > > diff --git a/drivers/net/fec.c b/drivers/net/fec.c > index f4afbe9..0e1d268 100644 > --- a/drivers/net/fec.c > +++ b/drivers/net/fec.c > @@ -172,6 +172,7 @@ struct fec_enet_private { > /* The saved address of a sent-in-place packet/buffer, for skfree(). */ > unsigned char *tx_bounce[TX_RING_SIZE]; > struct sk_buff* tx_skbuff[TX_RING_SIZE]; > + struct sk_buff* rx_skbuff[RX_RING_SIZE]; > ushort skb_cur; > ushort skb_dirty; > > @@ -335,8 +336,8 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) > /* Push the data cache so the CPM does not get stale memory > * data. > */ > - dma_sync_single(NULL, bdp->cbd_bufaddr, > - bdp->cbd_datlen, DMA_TO_DEVICE); > + bdp->cbd_bufaddr = dma_map_single(&dev->dev, skb->data, > + FEC_ENET_TX_FRSIZE, DMA_TO_DEVICE); > > /* Send it on its way. Tell FEC it's ready, interrupt when done, > * it's the last BD of the frame, and to put the CRC on the end. > @@ -429,7 +430,11 @@ fec_enet_tx(struct net_device *dev) > bdp = fep->dirty_tx; > > while (((status = bdp->cbd_sc) & BD_ENET_TX_READY) == 0) { > - if (bdp == fep->cur_tx && fep->tx_full == 0) break; > + if (bdp == fep->cur_tx && fep->tx_full == 0) > + break; > + > + dma_unmap_single(&dev->dev, bdp->cbd_bufaddr, FEC_ENET_TX_FRSIZE, DMA_TO_DEVICE); > + bdp->cbd_bufaddr = 0; > > skb = fep->tx_skbuff[fep->skb_dirty]; > /* Check for errors. */ > @@ -553,8 +558,8 @@ fec_enet_rx(struct net_device *dev) > dev->stats.rx_bytes += pkt_len; > data = (__u8*)__va(bdp->cbd_bufaddr); > > - dma_sync_single(NULL, (unsigned long)__pa(data), > - pkt_len - 4, DMA_FROM_DEVICE); > + dma_unmap_single(NULL, bdp->cbd_bufaddr, bdp->cbd_datlen, > + DMA_FROM_DEVICE); > > /* This does 16 byte alignment, exactly what we need. > * The packet length includes FCS, but we don't want to > @@ -574,6 +579,9 @@ fec_enet_rx(struct net_device *dev) > skb->protocol = eth_type_trans(skb, dev); > netif_rx(skb); > } > + > + bdp->cbd_bufaddr = dma_map_single(NULL, data, bdp->cbd_datlen, > + DMA_FROM_DEVICE); > rx_processing_done: > /* Clear the status flags for this buffer */ > status &= ~BD_ENET_RX_STATS; > @@ -1399,15 +1407,86 @@ mii_link_interrupt(int irq, void * dev_id) > } > #endif > > +static void fec_enet_free_buffers(struct net_device *dev) > +{ > + struct fec_enet_private *fep = netdev_priv(dev); > + int i; > + struct sk_buff *skb; > + struct bufdesc *bdp; > + > + bdp = fep->rx_bd_base; > + for (i = 0; i < RX_RING_SIZE; i++) { > + skb = fep->rx_skbuff[i]; > + > + if (bdp->cbd_bufaddr) > + dma_unmap_single(&dev->dev, bdp->cbd_bufaddr, > + FEC_ENET_RX_FRSIZE, DMA_FROM_DEVICE); > + if (skb) > + dev_kfree_skb(skb); > + bdp++; > + } > + > + bdp = fep->tx_bd_base; > + for (i = 0; i < TX_RING_SIZE; i++) > + kfree(fep->tx_bounce[i]); > +} > + > +static int fec_enet_alloc_buffers(struct net_device *dev) > +{ > + struct fec_enet_private *fep = netdev_priv(dev); > + int i; > + struct sk_buff *skb; > + struct bufdesc *bdp; > + > + bdp = fep->rx_bd_base; > + for (i = 0; i < RX_RING_SIZE; i++) { > + skb = dev_alloc_skb(FEC_ENET_RX_FRSIZE); > + if (!skb) { > + fec_enet_free_buffers(dev); > + return -ENOMEM; > + } > + fep->rx_skbuff[i] = skb; > + > + bdp->cbd_bufaddr = dma_map_single(&dev->dev, skb->data, > + FEC_ENET_RX_FRSIZE, DMA_FROM_DEVICE); > + bdp->cbd_sc = BD_ENET_RX_EMPTY; > + bdp++; > + } > + > + /* Set the last buffer to wrap. */ > + bdp--; > + bdp->cbd_sc |= BD_SC_WRAP; > + > + bdp = fep->tx_bd_base; > + for (i = 0; i < TX_RING_SIZE; i++) { > + fep->tx_bounce[i] = kmalloc(FEC_ENET_TX_FRSIZE, GFP_KERNEL); > + > + bdp->cbd_sc = 0; > + bdp->cbd_bufaddr = 0; > + bdp++; > + } > + > + /* Set the last buffer to wrap. */ > + bdp--; > + bdp->cbd_sc |= BD_SC_WRAP; > + > + return 0; > +} > + > static int > fec_enet_open(struct net_device *dev) > { > struct fec_enet_private *fep = netdev_priv(dev); > + int ret; > > /* I should reset the ring buffers here, but I don't yet know > * a simple way to do that. > */ > > + ret = fec_enet_alloc_buffers(dev); > + if (ret) > + return ret; > + > fep->sequence_done = 0; > fep->link = 0; > > @@ -1454,6 +1533,8 @@ fec_enet_close(struct net_device *dev) > netif_stop_queue(dev); > fec_stop(dev); > > + fec_enet_free_buffers(dev); > + > return 0; > } > > @@ -1576,9 +1657,8 @@ static const struct net_device_ops fec_netdev_ops = { > int __init fec_enet_init(struct net_device *dev, int index) > { > struct fec_enet_private *fep = netdev_priv(dev); > - unsigned long mem_addr; > - struct bufdesc *bdp, *cbd_base; > - int i, j; > + struct bufdesc *cbd_base; > + int i; > > /* Allocate memory for buffer descriptors. */ > cbd_base = dma_alloc_coherent(NULL, PAGE_SIZE, &fep->bd_dma, > @@ -1616,49 +1696,6 @@ int __init fec_enet_init(struct net_device *dev, int index) > fep->rx_bd_base = cbd_base; > fep->tx_bd_base = cbd_base + RX_RING_SIZE; > > - /* Initialize the receive buffer descriptors. */ > - bdp = fep->rx_bd_base; > - for (i=0; i<FEC_ENET_RX_PAGES; i++) { > - > - /* Allocate a page */ > - mem_addr = __get_free_page(GFP_KERNEL); > - /* XXX: missing check for allocation failure */ > - > - /* Initialize the BD for every fragment in the page */ > - for (j=0; j<FEC_ENET_RX_FRPPG; j++) { > - bdp->cbd_sc = BD_ENET_RX_EMPTY; > - bdp->cbd_bufaddr = __pa(mem_addr); > - mem_addr += FEC_ENET_RX_FRSIZE; > - bdp++; > - } > - } > - > - /* Set the last buffer to wrap */ > - bdp--; > - bdp->cbd_sc |= BD_SC_WRAP; > - > - /* ...and the same for transmit */ > - bdp = fep->tx_bd_base; > - for (i=0, j=FEC_ENET_TX_FRPPG; i<TX_RING_SIZE; i++) { > - if (j >= FEC_ENET_TX_FRPPG) { > - mem_addr = __get_free_page(GFP_KERNEL); > - j = 1; > - } else { > - mem_addr += FEC_ENET_TX_FRSIZE; > - j++; > - } > - fep->tx_bounce[i] = (unsigned char *) mem_addr; > - > - /* Initialize the BD for every fragment in the page */ > - bdp->cbd_sc = 0; > - bdp->cbd_bufaddr = 0; > - bdp++; > - } > - > - /* Set the last buffer to wrap */ > - bdp--; > - bdp->cbd_sc |= BD_SC_WRAP; > - > #ifdef HAVE_mii_link_interrupt > fec_request_mii_intr(dev); > #endif -- ------------------------------------------------------------------------ Greg Ungerer -- Principal Engineer EMAIL: gerg@snapgear.com SnapGear Group, McAfee PHONE: +61 7 3435 2888 825 Stanley St, FAX: +61 7 3891 3630 Woolloongabba, QLD, 4102, Australia WEB: http://www.SnapGear.com ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 11/12] FEC Buffer rework 2009-04-17 10:07 ` Greg Ungerer @ 2009-04-17 10:12 ` Sascha Hauer 0 siblings, 0 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-17 10:12 UTC (permalink / raw) To: Greg Ungerer; +Cc: netdev On Fri, Apr 17, 2009 at 08:07:26PM +1000, Greg Ungerer wrote: > > Hi Sascha, > > Sascha Hauer wrote: >> Allocate buffers in fec_open and free them again in fec_close. This makes >> it possible to use this driver as a module. >> >> Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> > > This is the only one that doesn't work for ColdFire. There is no > dma_map_single() or dma_unmap_single() currently. I will fix that, > so I am fine with this anyways. > > With all the other patches applied it still basically works on ColdFire > (at least on initial testing on a 5208). I have seen a few spurious > interrupts though with these changes applied, that didn't happen > before. I will investigate further when get a few minutes. Ok, thanks for testing Regards, Sascha -- Pengutronix e.K. | | Industrial Linux Solutions | http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 10/12] fec: switch to net_device_ops 2009-04-15 11:32 ` [PATCH 10/12] fec: switch to net_device_ops Sascha Hauer 2009-04-15 11:32 ` [PATCH 11/12] FEC Buffer rework Sascha Hauer @ 2009-04-16 9:37 ` David Miller 1 sibling, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:37 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 13:32:23 +0200 > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 09/12] fec: remove debugging printks 2009-04-15 11:32 ` [PATCH 09/12] fec: remove debugging printks Sascha Hauer 2009-04-15 11:32 ` [PATCH 10/12] fec: switch to net_device_ops Sascha Hauer @ 2009-04-16 9:37 ` David Miller 1 sibling, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:37 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 13:32:22 +0200 > These printks in fec_timeout do not give useful information in > a production kernel. > > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 08/12] fec: align receive packets 2009-04-15 11:32 ` [PATCH 08/12] fec: align receive packets Sascha Hauer 2009-04-15 11:32 ` [PATCH 09/12] fec: remove debugging printks Sascha Hauer @ 2009-04-16 9:37 ` David Miller 1 sibling, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:37 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 13:32:21 +0200 > Otherwise we get a lot of alignment errors > > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 07/12] fec: refactor init function 2009-04-15 11:32 ` [PATCH 07/12] fec: refactor init function Sascha Hauer 2009-04-15 11:32 ` [PATCH 08/12] fec: align receive packets Sascha Hauer @ 2009-04-16 9:37 ` David Miller 1 sibling, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:37 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 13:32:20 +0200 > fec_enet_init() does the hardware initialisation and then calls > fec_restart() which does the same initialisation again, so we > can safely remove the initialisation from fec_enet_init(). > > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 06/12] fec: refactor set_multicast_list() to make it more readable 2009-04-15 11:32 ` [PATCH 06/12] fec: refactor set_multicast_list() to make it more readable Sascha Hauer 2009-04-15 11:32 ` [PATCH 07/12] fec: refactor init function Sascha Hauer @ 2009-04-16 9:37 ` David Miller 1 sibling, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:37 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 13:32:19 +0200 > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 05/12] fec: Codingstyle cleanups 2009-04-15 11:32 ` [PATCH 05/12] fec: Codingstyle cleanups Sascha Hauer 2009-04-15 11:32 ` [PATCH 06/12] fec: refactor set_multicast_list() to make it more readable Sascha Hauer @ 2009-04-16 9:36 ` David Miller 1 sibling, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:36 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 13:32:18 +0200 > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 04/12] fec: remove unnecessary cast 2009-04-15 11:32 ` [PATCH 04/12] fec: remove unnecessary cast Sascha Hauer 2009-04-15 11:32 ` [PATCH 05/12] fec: Codingstyle cleanups Sascha Hauer @ 2009-04-16 9:36 ` David Miller 1 sibling, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:36 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 13:32:17 +0200 > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 03/12] fec: do not typedef struct types 2009-04-15 11:32 ` [PATCH 03/12] fec: do not typedef struct types Sascha Hauer 2009-04-15 11:32 ` [PATCH 04/12] fec: remove unnecessary cast Sascha Hauer @ 2009-04-16 9:36 ` David Miller 1 sibling, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:36 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 13:32:16 +0200 > Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 02/12] fec: switch to writel/readl 2009-04-15 11:32 ` [PATCH 02/12] fec: switch to writel/readl Sascha Hauer 2009-04-15 11:32 ` [PATCH 03/12] fec: do not typedef struct types Sascha Hauer @ 2009-04-15 13:11 ` Sascha Hauer 2009-04-16 9:36 ` David Miller 1 sibling, 1 reply; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 13:11 UTC (permalink / raw) To: netdev; +Cc: Greg Ungerer Since the second register set is used for M5272 we cannot remove it. Here is an updated version. Sascha >From 4330c85a8902007a6fb726333a85b9e981579992 Mon Sep 17 00:00:00 2001 From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 09:51:34 +0200 Subject: [PATCH 01/11] fec: switch to writel/readl Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> --- drivers/net/fec.c | 279 ++++++++++++++++++++++------------------------------- drivers/net/fec.h | 119 +++++++++-------------- 2 files changed, 161 insertions(+), 237 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index 682e7f0..3fa3fe4 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -165,7 +165,7 @@ typedef struct { */ struct fec_enet_private { /* Hardware registers of the FEC device */ - volatile fec_t *hwp; + void __iomem *hwp; struct net_device *netdev; @@ -288,15 +288,11 @@ static int mii_queue(struct net_device *dev, int request, static int fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) { - struct fec_enet_private *fep; - volatile fec_t *fecp; + struct fec_enet_private *fep = netdev_priv(dev); volatile cbd_t *bdp; unsigned short status; unsigned long flags; - fep = netdev_priv(dev); - fecp = (volatile fec_t*)dev->base_addr; - if (!fep->link) { /* Link is down or autonegotiation is in progress. */ return 1; @@ -363,7 +359,7 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) dev->trans_start = jiffies; /* Trigger transmission start */ - fecp->fec_x_des_active = 0; + writel(0, fep->hwp + FEC_X_DES_ACTIVE); /* If this was the last BD in the ring, start at the beginning again. */ @@ -436,29 +432,25 @@ static irqreturn_t fec_enet_interrupt(int irq, void * dev_id) { struct net_device *dev = dev_id; - volatile fec_t *fecp; + struct fec_enet_private *fep = netdev_priv(dev); uint int_events; irqreturn_t ret = IRQ_NONE; - fecp = (volatile fec_t*)dev->base_addr; - - /* Get the interrupt events that caused us to be here. - */ + /* Get the interrupt events that caused us to be here. */ do { - int_events = fecp->fec_ievent; - fecp->fec_ievent = int_events; + int_events = readl(fep->hwp + FEC_IEVENT); + writel(int_events, fep->hwp + FEC_IEVENT); - /* Handle receive event in its own function. - */ + /* Handle receive event in its own function. */ if (int_events & FEC_ENET_RXF) { ret = IRQ_HANDLED; fec_enet_rx(dev); } /* Transmit OK, or non-fatal error. Update the buffer - descriptors. FEC handles all errors, we just discover - them as part of the transmit process. - */ + * descriptors. FEC handles all errors, we just discover + * them as part of the transmit process. + */ if (int_events & FEC_ENET_TXF) { ret = IRQ_HANDLED; fec_enet_tx(dev); @@ -555,8 +547,7 @@ fec_enet_tx(struct net_device *dev) static void fec_enet_rx(struct net_device *dev) { - struct fec_enet_private *fep; - volatile fec_t *fecp; + struct fec_enet_private *fep = netdev_priv(dev); volatile cbd_t *bdp; unsigned short status; struct sk_buff *skb; @@ -567,9 +558,6 @@ fec_enet_rx(struct net_device *dev) flush_cache_all(); #endif - fep = netdev_priv(dev); - fecp = (volatile fec_t*)dev->base_addr; - spin_lock_irq(&fep->hw_lock); /* First, grab all of the stats for the incoming packet. @@ -665,7 +653,7 @@ while (!((status = bdp->cbd_sc) & BD_ENET_RX_EMPTY)) { * incoming frames. On a heavily loaded network, we should be * able to keep up at the expense of system resources. */ - fecp->fec_r_des_active = 0; + writel(0, fep->hwp + FEC_R_DES_ACTIVE); #endif } /* while (!((status = bdp->cbd_sc) & BD_ENET_RX_EMPTY)) */ fep->cur_rx = (cbd_t *)bdp; @@ -690,30 +678,25 @@ static void fec_enet_mii(struct net_device *dev) { struct fec_enet_private *fep; - volatile fec_t *ep; mii_list_t *mip; - uint mii_reg; fep = netdev_priv(dev); spin_lock_irq(&fep->mii_lock); - ep = fep->hwp; - mii_reg = ep->fec_mii_data; - if ((mip = mii_head) == NULL) { printk("MII and no head!\n"); goto unlock; } if (mip->mii_func != NULL) - (*(mip->mii_func))(mii_reg, dev); + (*(mip->mii_func))(readl(fep->hwp + FEC_MII_DATA), dev); mii_head = mip->mii_next; mip->mii_next = mii_free; mii_free = mip; if ((mip = mii_head) != NULL) - ep->fec_mii_data = mip->mii_regval; + writel(mip->mii_regval, fep->hwp + FEC_MII_DATA); unlock: spin_unlock_irq(&fep->mii_lock); @@ -745,7 +728,7 @@ mii_queue(struct net_device *dev, int regval, void (*func)(uint, struct net_devi mii_tail = mip; } else { mii_head = mii_tail = mip; - fep->hwp->fec_mii_data = regval; + writel(regval, fep->hwp + FEC_MII_DATA); } } else { retval = 1; @@ -1246,11 +1229,8 @@ static void __inline__ fec_phy_ack_intr(void) static void __inline__ fec_get_mac(struct net_device *dev) { struct fec_enet_private *fep = netdev_priv(dev); - volatile fec_t *fecp; unsigned char *iap, tmpaddr[ETH_ALEN]; - fecp = fep->hwp; - if (FEC_FLASHMAC) { /* * Get MAC address from FLASH. @@ -1264,8 +1244,8 @@ static void __inline__ fec_get_mac(struct net_device *dev) (iap[3] == 0xff) && (iap[4] == 0xff) && (iap[5] == 0xff)) iap = fec_mac_default; } else { - *((unsigned long *) &tmpaddr[0]) = fecp->fec_addr_low; - *((unsigned short *) &tmpaddr[4]) = (fecp->fec_addr_high >> 16); + *((unsigned long *) &tmpaddr[0]) = readl(fep->hwp + FEC_ADDR_LOW); + *((unsigned short *) &tmpaddr[4]) = (readl(fep->hwp + FEC_ADDR_HIGH) >> 16); iap = &tmpaddr[0]; } @@ -1457,11 +1437,9 @@ static void mii_discover_phy(uint mii_reg, struct net_device *dev) { struct fec_enet_private *fep; - volatile fec_t *fecp; uint phytype; fep = netdev_priv(dev); - fecp = fep->hwp; if (fep->phy_addr < 32) { if ((phytype = (mii_reg & 0xffff)) != 0xffff && phytype != 0) { @@ -1479,7 +1457,8 @@ mii_discover_phy(uint mii_reg, struct net_device *dev) } else { printk("FEC: No PHY device found.\n"); /* Disable external MII interface */ - fecp->fec_mii_speed = fep->phy_speed = 0; + writel(0, fep->hwp + FEC_MII_SPEED); + fep->phy_speed = 0; #ifdef HAVE_mii_link_interrupt fec_disable_phy_intr(); #endif @@ -1583,32 +1562,31 @@ fec_enet_close(struct net_device *dev) static void set_multicast_list(struct net_device *dev) { - struct fec_enet_private *fep; - volatile fec_t *ep; + struct fec_enet_private *fep = netdev_priv(dev); struct dev_mc_list *dmi; - unsigned int i, j, bit, data, crc; + unsigned int i, j, bit, data, crc, tmp; unsigned char hash; - fep = netdev_priv(dev); - ep = fep->hwp; - if (dev->flags&IFF_PROMISC) { - ep->fec_r_cntrl |= 0x0008; + tmp = readl(fep->hwp + FEC_R_CNTRL); + tmp |= 0x8; + writel(tmp, fep->hwp + FEC_R_CNTRL); } else { - - ep->fec_r_cntrl &= ~0x0008; + tmp = readl(fep->hwp + FEC_R_CNTRL); + tmp &= ~0x8; + writel(tmp, fep->hwp + FEC_R_CNTRL); if (dev->flags & IFF_ALLMULTI) { /* Catch all multicast addresses, so set the * filter to all 1's. */ - ep->fec_grp_hash_table_high = 0xffffffff; - ep->fec_grp_hash_table_low = 0xffffffff; + writel(0xffffffff, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + writel(0xffffffff, fep->hwp + FEC_GRP_HASH_TABLE_LOW); } else { /* Clear filter and add the addresses in hash register. */ - ep->fec_grp_hash_table_high = 0; - ep->fec_grp_hash_table_low = 0; + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); dmi = dev->mc_list; @@ -1638,10 +1616,15 @@ static void set_multicast_list(struct net_device *dev) */ hash = (crc >> (32 - HASH_BITS)) & 0x3f; - if (hash > 31) - ep->fec_grp_hash_table_high |= 1 << (hash - 32); - else - ep->fec_grp_hash_table_low |= 1 << hash; + if (hash > 31) { + tmp = readl(fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + tmp |= 1 << (hash - 32); + writel(tmp, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + } else { + tmp = readl(fep->hwp + FEC_GRP_HASH_TABLE_LOW); + tmp |= 1 << hash; + writel(tmp, fep->hwp + FEC_GRP_HASH_TABLE_LOW); + } } } } @@ -1652,16 +1635,14 @@ static void set_multicast_list(struct net_device *dev) static void fec_set_mac_address(struct net_device *dev) { - volatile fec_t *fecp; - - fecp = ((struct fec_enet_private *)netdev_priv(dev))->hwp; + struct fec_enet_private *fep = netdev_priv(dev); /* Set station address. */ - fecp->fec_addr_low = dev->dev_addr[3] | (dev->dev_addr[2] << 8) | - (dev->dev_addr[1] << 16) | (dev->dev_addr[0] << 24); - fecp->fec_addr_high = (dev->dev_addr[5] << 16) | - (dev->dev_addr[4] << 24); - + writel(dev->dev_addr[3] | (dev->dev_addr[2] << 8) | + (dev->dev_addr[1] << 16) | (dev->dev_addr[0] << 24), + fep->hwp + FEC_ADDR_LOW); + writel((dev->dev_addr[5] << 16) | (dev->dev_addr[4] << 24), + fep + FEC_ADDR_HIGH); } /* @@ -1675,7 +1656,6 @@ int __init fec_enet_init(struct net_device *dev, int index) unsigned long mem_addr; volatile cbd_t *bdp; cbd_t *cbd_base; - volatile fec_t *fecp; int i, j; /* Allocate memory for buffer descriptors. @@ -1690,17 +1670,13 @@ int __init fec_enet_init(struct net_device *dev, int index) spin_lock_init(&fep->hw_lock); spin_lock_init(&fep->mii_lock); - /* Create an Ethernet device instance. - */ - fecp = (volatile fec_t *)dev->base_addr; - fep->index = index; - fep->hwp = fecp; + fep->hwp = (void __iomem *)dev->base_addr; fep->netdev = dev; /* Whack a reset. We should wait for this. */ - fecp->fec_ecntrl = 1; + writel(1, fep->hwp + FEC_ECNTRL); udelay(10); /* Set the Ethernet address */ @@ -1709,12 +1685,12 @@ int __init fec_enet_init(struct net_device *dev, int index) #else { unsigned long l; - l = fecp->fec_addr_low; + l = readl(fep->hwp + FEC_ADDR_LOW); dev->dev_addr[0] = (unsigned char)((l & 0xFF000000) >> 24); dev->dev_addr[1] = (unsigned char)((l & 0x00FF0000) >> 16); dev->dev_addr[2] = (unsigned char)((l & 0x0000FF00) >> 8); dev->dev_addr[3] = (unsigned char)((l & 0x000000FF) >> 0); - l = fecp->fec_addr_high; + l = readl(fep->hwp + FEC_ADDR_HIGH); dev->dev_addr[4] = (unsigned char)((l & 0xFF000000) >> 24); dev->dev_addr[5] = (unsigned char)((l & 0x00FF0000) >> 16); } @@ -1784,22 +1760,22 @@ int __init fec_enet_init(struct net_device *dev, int index) /* Set receive and transmit descriptor base. */ - fecp->fec_r_des_start = fep->bd_dma; - fecp->fec_x_des_start = (unsigned long)fep->bd_dma + sizeof(cbd_t) - * RX_RING_SIZE; + writel(fep->bd_dma, fep->hwp + FEC_R_DES_START); + writel((unsigned long)fep->bd_dma + sizeof(cbd_t) * RX_RING_SIZE, + fep->hwp + FEC_X_DES_START); #ifdef HAVE_mii_link_interrupt fec_request_mii_intr(dev); #endif - fecp->fec_grp_hash_table_high = 0; - fecp->fec_grp_hash_table_low = 0; - fecp->fec_r_buff_size = PKT_MAXBLR_SIZE; - fecp->fec_ecntrl = 2; - fecp->fec_r_des_active = 0; + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); + writel(PKT_MAXBLR_SIZE, fep->hwp + FEC_R_BUFF_SIZE); + writel(2, fep->hwp + FEC_ECNTRL); + writel(0, fep->hwp + FEC_R_DES_ACTIVE); #ifndef CONFIG_M5272 - fecp->fec_hash_table_high = 0; - fecp->fec_hash_table_low = 0; + writel(0, fep->hwp + FEC_HASH_TABLE_HIGH); + writel(0, fep->hwp + FEC_HASH_TABLE_LOW); #endif /* The FEC Ethernet specific entries in the device structure. */ @@ -1815,20 +1791,21 @@ int __init fec_enet_init(struct net_device *dev, int index) mii_free = mii_cmds; /* setup MII interface */ - fecp->fec_r_cntrl = OPT_FRAME_SIZE | 0x04; - fecp->fec_x_cntrl = 0x00; + writel(OPT_FRAME_SIZE | 0x04, fep->hwp + FEC_R_CNTRL); + writel(0, fep->hwp + FEC_X_CNTRL); /* * Set MII speed to 2.5 MHz */ fep->phy_speed = ((((clk_get_rate(fep->clk) / 2 + 4999999) / 2500000) / 2) & 0x3F) << 1; - fecp->fec_mii_speed = fep->phy_speed; + writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED); fec_restart(dev, 0); /* Clear and enable interrupts */ - fecp->fec_ievent = 0xffc00000; - fecp->fec_imask = (FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII); + writel(0xffc00000, fep->hwp + FEC_IEVENT); + writel(FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII, + fep->hwp + FEC_IMASK); /* Queue up command to detect the PHY and initialize the * remainder of the interface. @@ -1847,47 +1824,36 @@ int __init fec_enet_init(struct net_device *dev, int index) static void fec_restart(struct net_device *dev, int duplex) { - struct fec_enet_private *fep; + struct fec_enet_private *fep = netdev_priv(dev); volatile cbd_t *bdp; - volatile fec_t *fecp; int i; - fep = netdev_priv(dev); - fecp = fep->hwp; - - /* Whack a reset. We should wait for this. - */ - fecp->fec_ecntrl = 1; + /* Whack a reset. We should wait for this. */ + writel(1, fep->hwp + FEC_ECNTRL); udelay(10); - /* Clear any outstanding interrupt. - */ - fecp->fec_ievent = 0xffc00000; + /* Clear any outstanding interrupt. */ + writel(0xffc00000, fep->hwp + FEC_IEVENT); - /* Set station address. - */ + /* Set station address. */ fec_set_mac_address(dev); - /* Reset all multicast. - */ - fecp->fec_grp_hash_table_high = 0; - fecp->fec_grp_hash_table_low = 0; + /* Reset all multicast. */ + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); + writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); - /* Set maximum receive buffer size. - */ - fecp->fec_r_buff_size = PKT_MAXBLR_SIZE; + /* Set maximum receive buffer size. */ + writel(PKT_MAXBLR_SIZE, fep->hwp + FEC_R_BUFF_SIZE); - /* Set receive and transmit descriptor base. - */ - fecp->fec_r_des_start = fep->bd_dma; - fecp->fec_x_des_start = (unsigned long)fep->bd_dma + sizeof(cbd_t) - * RX_RING_SIZE; + /* Set receive and transmit descriptor base. */ + writel(fep->bd_dma, fep->hwp + FEC_R_DES_START); + writel((unsigned long)fep->bd_dma + sizeof(cbd_t) * RX_RING_SIZE, + fep->hwp + FEC_X_DES_START); fep->dirty_tx = fep->cur_tx = fep->tx_bd_base; fep->cur_rx = fep->rx_bd_base; - /* Reset SKB transmit buffers. - */ + /* Reset SKB transmit buffers. */ fep->skb_cur = fep->skb_dirty = 0; for (i=0; i<=TX_RING_MOD_MASK; i++) { if (fep->tx_skbuff[i] != NULL) { @@ -1896,96 +1862,81 @@ fec_restart(struct net_device *dev, int duplex) } } - /* Initialize the receive buffer descriptors. - */ + /* Initialize the receive buffer descriptors. */ bdp = fep->rx_bd_base; for (i=0; i<RX_RING_SIZE; i++) { - /* Initialize the BD for every fragment in the page. - */ + /* Initialize the BD for every fragment in the page. */ bdp->cbd_sc = BD_ENET_RX_EMPTY; bdp++; } - /* Set the last buffer to wrap. - */ + /* Set the last buffer to wrap. */ bdp--; bdp->cbd_sc |= BD_SC_WRAP; - /* ...and the same for transmmit. - */ + /* ...and the same for transmmit. */ bdp = fep->tx_bd_base; for (i=0; i<TX_RING_SIZE; i++) { - /* Initialize the BD for every fragment in the page. - */ + /* Initialize the BD for every fragment in the page. */ bdp->cbd_sc = 0; bdp->cbd_bufaddr = 0; bdp++; } - /* Set the last buffer to wrap. - */ + /* Set the last buffer to wrap. */ bdp--; bdp->cbd_sc |= BD_SC_WRAP; - /* Enable MII mode. - */ + /* Enable MII mode. */ if (duplex) { - fecp->fec_r_cntrl = OPT_FRAME_SIZE | 0x04;/* MII enable */ - fecp->fec_x_cntrl = 0x04; /* FD enable */ + /* MII enable / FD enable */ + writel(OPT_FRAME_SIZE | 0x04, fep->hwp + FEC_R_CNTRL); + writel(0x04, fep->hwp + FEC_X_CNTRL); } else { - /* MII enable|No Rcv on Xmit */ - fecp->fec_r_cntrl = OPT_FRAME_SIZE | 0x06; - fecp->fec_x_cntrl = 0x00; + /* MII enable / No Rcv on Xmit */ + writel(OPT_FRAME_SIZE | 0x06, fep->hwp + FEC_R_CNTRL); + writel(0x0, fep->hwp + FEC_X_CNTRL); } fep->full_duplex = duplex; - /* Set MII speed. - */ - fecp->fec_mii_speed = fep->phy_speed; + /* Set MII speed. */ + writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED); - /* And last, enable the transmit and receive processing. - */ - fecp->fec_ecntrl = 2; - fecp->fec_r_des_active = 0; + /* And last, enable the transmit and receive processing. */ + writel(2, fep->hwp + FEC_ECNTRL); + writel(0, fep->hwp + FEC_R_DES_ACTIVE); - /* Enable interrupts we wish to service. - */ - fecp->fec_imask = (FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII); + /* Enable interrupts we wish to service. */ + writel(FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII, + fep->hwp + FEC_IMASK); } static void fec_stop(struct net_device *dev) { - volatile fec_t *fecp; - struct fec_enet_private *fep; - - fep = netdev_priv(dev); - fecp = fep->hwp; + struct fec_enet_private *fep = netdev_priv(dev); /* ** We cannot expect a graceful transmit stop without link !!! */ - if (fep->link) - { - fecp->fec_x_cntrl = 0x01; /* Graceful transmit stop */ + if (fep->link) { + writel(1, fep->hwp + FEC_X_CNTRL); /* Graceful transmit stop */ udelay(10); - if (!(fecp->fec_ievent & FEC_ENET_GRA)) + if (!(readl(fep->hwp + FEC_IEVENT) & FEC_ENET_GRA)) printk("fec_stop : Graceful transmit stop did not complete !\n"); - } + } - /* Whack a reset. We should wait for this. - */ - fecp->fec_ecntrl = 1; + /* Whack a reset. We should wait for this. */ + writel(1, fep->hwp + FEC_ECNTRL); udelay(10); - /* Clear outstanding MII command interrupts. - */ - fecp->fec_ievent = FEC_ENET_MII; + /* Clear outstanding MII command interrupts. */ + writel(FEC_ENET_MII, fep->hwp + FEC_IEVENT); - fecp->fec_imask = FEC_ENET_MII; - fecp->fec_mii_speed = fep->phy_speed; + writel(FEC_ENET_MII, fep->hwp + FEC_IMASK); + writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED); } static int __devinit diff --git a/drivers/net/fec.h b/drivers/net/fec.h index 76c64c9..5474ba3 100644 --- a/drivers/net/fec.h +++ b/drivers/net/fec.h @@ -20,82 +20,55 @@ * registers in the same peripheral device on different models * of the ColdFire! */ -typedef struct fec { - unsigned long fec_reserved0; - unsigned long fec_ievent; /* Interrupt event reg */ - unsigned long fec_imask; /* Interrupt mask reg */ - unsigned long fec_reserved1; - unsigned long fec_r_des_active; /* Receive descriptor reg */ - unsigned long fec_x_des_active; /* Transmit descriptor reg */ - unsigned long fec_reserved2[3]; - unsigned long fec_ecntrl; /* Ethernet control reg */ - unsigned long fec_reserved3[6]; - unsigned long fec_mii_data; /* MII manage frame reg */ - unsigned long fec_mii_speed; /* MII speed control reg */ - unsigned long fec_reserved4[7]; - unsigned long fec_mib_ctrlstat; /* MIB control/status reg */ - unsigned long fec_reserved5[7]; - unsigned long fec_r_cntrl; /* Receive control reg */ - unsigned long fec_reserved6[15]; - unsigned long fec_x_cntrl; /* Transmit Control reg */ - unsigned long fec_reserved7[7]; - unsigned long fec_addr_low; /* Low 32bits MAC address */ - unsigned long fec_addr_high; /* High 16bits MAC address */ - unsigned long fec_opd; /* Opcode + Pause duration */ - unsigned long fec_reserved8[10]; - unsigned long fec_hash_table_high; /* High 32bits hash table */ - unsigned long fec_hash_table_low; /* Low 32bits hash table */ - unsigned long fec_grp_hash_table_high;/* High 32bits hash table */ - unsigned long fec_grp_hash_table_low; /* Low 32bits hash table */ - unsigned long fec_reserved9[7]; - unsigned long fec_x_wmrk; /* FIFO transmit water mark */ - unsigned long fec_reserved10; - unsigned long fec_r_bound; /* FIFO receive bound reg */ - unsigned long fec_r_fstart; /* FIFO receive start reg */ - unsigned long fec_reserved11[11]; - unsigned long fec_r_des_start; /* Receive descriptor ring */ - unsigned long fec_x_des_start; /* Transmit descriptor ring */ - unsigned long fec_r_buff_size; /* Maximum receive buff size */ -} fec_t; +#define FEC_IEVENT 0x004 /* Interrupt event reg */ +#define FEC_IMASK 0x008 /* Interrupt mask reg */ +#define FEC_R_DES_ACTIVE 0x010 /* Receive descriptor reg */ +#define FEC_X_DES_ACTIVE 0x014 /* Transmit descriptor reg */ +#define FEC_ECNTRL 0x024 /* Ethernet control reg */ +#define FEC_MII_DATA 0x040 /* MII manage frame reg */ +#define FEC_MII_SPEED 0x044 /* MII speed control reg */ +#define FEC_MIB_CTRLSTAT 0x064 /* MIB control/status reg */ +#define FEC_R_CNTRL 0x084 /* Receive control reg */ +#define FEC_X_CNTRL 0x0c4 /* Transmit Control reg */ +#define FEC_ADDR_LOW 0x0e4 /* Low 32bits MAC address */ +#define FEC_ADDR_HIGH 0x0e8 /* High 16bits MAC address */ +#define FEC_OPD 0x0ec /* Opcode + Pause duration */ +#define FEC_HASH_TABLE_HIGH 0x118 /* High 32bits hash table */ +#define FEC_HASH_TABLE_LOW 0x11c /* Low 32bits hash table */ +#define FEC_GRP_HASH_TABLE_HIGH 0x120 /* High 32bits hash table */ +#define FEC_GRP_HASH_TABLE_LOW 0x124 /* Low 32bits hash table */ +#define FEC_X_WMRK 0x144 /* FIFO transmit water mark */ +#define FEC_R_BOUND 0x14c /* FIFO receive bound reg */ +#define FEC_R_FSTART 0x150 /* FIFO receive start reg */ +#define FEC_R_DES_START 0x180 /* Receive descriptor ring */ +#define FEC_X_DES_START 0x184 /* Transmit descriptor ring */ +#define FEC_R_BUFF_SIZE 0x188 /* Maximum receive buff size */ #else -/* - * Define device register set address map. - */ -typedef struct fec { - unsigned long fec_ecntrl; /* Ethernet control reg */ - unsigned long fec_ievent; /* Interrupt even reg */ - unsigned long fec_imask; /* Interrupt mask reg */ - unsigned long fec_ivec; /* Interrupt vec status reg */ - unsigned long fec_r_des_active; /* Receive descriptor reg */ - unsigned long fec_x_des_active; /* Transmit descriptor reg */ - unsigned long fec_reserved1[10]; - unsigned long fec_mii_data; /* MII manage frame reg */ - unsigned long fec_mii_speed; /* MII speed control reg */ - unsigned long fec_reserved2[17]; - unsigned long fec_r_bound; /* FIFO receive bound reg */ - unsigned long fec_r_fstart; /* FIFO receive start reg */ - unsigned long fec_reserved3[4]; - unsigned long fec_x_wmrk; /* FIFO transmit water mark */ - unsigned long fec_reserved4; - unsigned long fec_x_fstart; /* FIFO transmit start reg */ - unsigned long fec_reserved5[21]; - unsigned long fec_r_cntrl; /* Receive control reg */ - unsigned long fec_max_frm_len; /* Maximum frame length reg */ - unsigned long fec_reserved6[14]; - unsigned long fec_x_cntrl; /* Transmit Control reg */ - unsigned long fec_reserved7[158]; - unsigned long fec_addr_low; /* Low 32bits MAC address */ - unsigned long fec_addr_high; /* High 16bits MAC address */ - unsigned long fec_grp_hash_table_high;/* High 32bits hash table */ - unsigned long fec_grp_hash_table_low; /* Low 32bits hash table */ - unsigned long fec_r_des_start; /* Receive descriptor ring */ - unsigned long fec_x_des_start; /* Transmit descriptor ring */ - unsigned long fec_r_buff_size; /* Maximum receive buff size */ - unsigned long reserved8[9]; - unsigned long fec_fifo_ram[112]; /* FIFO RAM buffer */ -} fec_t; +#define FEC_ECNTRL; 0x000 /* Ethernet control reg */ +#define FEC_IEVENT; 0x004 /* Interrupt even reg */ +#define FEC_IMASK; 0x008 /* Interrupt mask reg */ +#define FEC_IVEC; 0x00c /* Interrupt vec status reg */ +#define FEC_R_DES_ACTIVE; 0x010 /* Receive descriptor reg */ +#define FEC_X_DES_ACTIVE; 0x01c /* Transmit descriptor reg */ +#define FEC_MII_DATA 0x040 /* MII manage frame reg */ +#define FEC_MII_SPEED 0x044 /* MII speed control reg */ +#define FEC_R_BOUND 0x08c /* FIFO receive bound reg */ +#define FEC_R_FSTART 0x090 /* FIFO receive start reg */ +#define FEC_X_WMRK 0x0a4 /* FIFO transmit water mark */ +#define FEC_X_FSTART 0x0ac /* FIFO transmit start reg */ +#define FEC_R_CNTRL 0x104 /* Receive control reg */ +#define FEC_MAX_FRM_LEN 0x108 /* Maximum frame length reg */ +#define FEC_X_CNTRL 0x144 /* Transmit Control reg */ +#define FEC_ADDR_LOW 0x3c0 /* Low 32bits MAC address */ +#define FEC_ADDR_HIGH 0x3c4 /* High 16bits MAC address */ +#define FEC_GRP_HASH_TABLE_HIGH 0x3c8 /* High 32bits hash table */ +#define FEC_GRP_HASH_TABLE_LOW 0x3cc /* Low 32bits hash table */ +#define FEC_R_DES_START 0x3d0 /* Receive descriptor ring */ +#define FEC_X_DES_START 0x3d4 /* Transmit descriptor ring */ +#define FEC_R_BUFF_SIZE 0x3d8 /* Maximum receive buff size */ +#define FEC_FIFO_RAM 0x400 /* FIFO RAM buffer */ #endif /* CONFIG_M5272 */ -- 1.6.2.1 -- Pengutronix e.K. | | Industrial Linux Solutions | http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | ^ permalink raw reply related [flat|nested] 29+ messages in thread
* Re: [PATCH 02/12] fec: switch to writel/readl 2009-04-15 13:11 ` [PATCH 02/12] fec: switch to writel/readl Sascha Hauer @ 2009-04-16 9:36 ` David Miller 0 siblings, 0 replies; 29+ messages in thread From: David Miller @ 2009-04-16 9:36 UTC (permalink / raw) To: s.hauer; +Cc: netdev, gerg From: Sascha Hauer <s.hauer@pengutronix.de> Date: Wed, 15 Apr 2009 15:11:30 +0200 > Since the second register set is used for M5272 we cannot remove it. > Here is an updated version. Applied. ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH] FEC driver: patches for -next 2009-04-15 11:32 [PATCH] FEC driver: patches for -next Sascha Hauer 2009-04-15 11:32 ` [PATCH 01/12] fec: remove unused ifdef Sascha Hauer @ 2009-04-15 12:12 ` Greg Ungerer 2009-04-15 12:55 ` Sascha Hauer 1 sibling, 1 reply; 29+ messages in thread From: Greg Ungerer @ 2009-04-15 12:12 UTC (permalink / raw) To: Sascha Hauer; +Cc: netdev Hi Sascha, Sascha Hauer wrote: > These patches aim to modernize the FEC ethernet driver. > Greg, can you please confirm that the ifdef in patch 01/12 is indeed unused? It is used for the CONFIG_M5272 case. The offsets to some registers are slightly different for the version of the FEC hardware in that SoC CPU. Regards Greg > The following changes since commit 0882e8dd3aad33eca41696d463bb896e6c8817eb: > Linus Torvalds (1): > Linux 2.6.30-rc2 > > are available in the git repository at: > > git://git.pengutronix.de/git/sha/linux-2.6.git fec > > Sascha Hauer (12): > fec: remove unused ifdef > fec: switch to writel/readl > fec: do not typedef struct types > fec: remove unnecessary cast > fec: Codingstyle cleanups > fec: refactor set_multicast_list() to make it more readable > fec: refactor init function > fec: align receive packets > fec: remove debugging printks > fec: switch to net_device_ops > FEC Buffer rework > fec: call fec_restart() in fec_open() > > drivers/net/fec.c | 897 ++++++++++++++++++++++------------------------------- > drivers/net/fec.h | 112 ++----- > 2 files changed, 402 insertions(+), 607 deletions(-) > -- ------------------------------------------------------------------------ Greg Ungerer -- Principal Engineer EMAIL: gerg@snapgear.com SnapGear Group, McAfee PHONE: +61 7 3435 2888 825 Stanley St, FAX: +61 7 3891 3630 Woolloongabba, QLD, 4102, Australia WEB: http://www.SnapGear.com ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH] FEC driver: patches for -next 2009-04-15 12:12 ` [PATCH] FEC driver: patches for -next Greg Ungerer @ 2009-04-15 12:55 ` Sascha Hauer 0 siblings, 0 replies; 29+ messages in thread From: Sascha Hauer @ 2009-04-15 12:55 UTC (permalink / raw) To: Greg Ungerer; +Cc: netdev On Wed, Apr 15, 2009 at 10:12:05PM +1000, Greg Ungerer wrote: > > Hi Sascha, > > Sascha Hauer wrote: >> These patches aim to modernize the FEC ethernet driver. >> Greg, can you please confirm that the ifdef in patch 01/12 is indeed unused? > > It is used for the CONFIG_M5272 case. The offsets to some registers > are slightly different for the version of the FEC hardware in that > SoC CPU. Ok, then I'll skip the first patch and update the second one accordingly. Regards Sascha > > Regards > Greg > > >> The following changes since commit 0882e8dd3aad33eca41696d463bb896e6c8817eb: >> Linus Torvalds (1): >> Linux 2.6.30-rc2 >> >> are available in the git repository at: >> >> git://git.pengutronix.de/git/sha/linux-2.6.git fec >> >> Sascha Hauer (12): >> fec: remove unused ifdef >> fec: switch to writel/readl >> fec: do not typedef struct types >> fec: remove unnecessary cast >> fec: Codingstyle cleanups >> fec: refactor set_multicast_list() to make it more readable >> fec: refactor init function >> fec: align receive packets >> fec: remove debugging printks >> fec: switch to net_device_ops >> FEC Buffer rework >> fec: call fec_restart() in fec_open() >> >> drivers/net/fec.c | 897 ++++++++++++++++++++++------------------------------- >> drivers/net/fec.h | 112 ++----- >> 2 files changed, 402 insertions(+), 607 deletions(-) >> > > > -- > ------------------------------------------------------------------------ > Greg Ungerer -- Principal Engineer EMAIL: gerg@snapgear.com > SnapGear Group, McAfee PHONE: +61 7 3435 2888 > 825 Stanley St, FAX: +61 7 3891 3630 > Woolloongabba, QLD, 4102, Australia WEB: http://www.SnapGear.com > -- Pengutronix e.K. | | Industrial Linux Solutions | http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | ^ permalink raw reply [flat|nested] 29+ messages in thread
end of thread, other threads:[~2009-04-17 10:12 UTC | newest] Thread overview: 29+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2009-04-15 11:32 [PATCH] FEC driver: patches for -next Sascha Hauer 2009-04-15 11:32 ` [PATCH 01/12] fec: remove unused ifdef Sascha Hauer 2009-04-15 11:32 ` [PATCH 02/12] fec: switch to writel/readl Sascha Hauer 2009-04-15 11:32 ` [PATCH 03/12] fec: do not typedef struct types Sascha Hauer 2009-04-15 11:32 ` [PATCH 04/12] fec: remove unnecessary cast Sascha Hauer 2009-04-15 11:32 ` [PATCH 05/12] fec: Codingstyle cleanups Sascha Hauer 2009-04-15 11:32 ` [PATCH 06/12] fec: refactor set_multicast_list() to make it more readable Sascha Hauer 2009-04-15 11:32 ` [PATCH 07/12] fec: refactor init function Sascha Hauer 2009-04-15 11:32 ` [PATCH 08/12] fec: align receive packets Sascha Hauer 2009-04-15 11:32 ` [PATCH 09/12] fec: remove debugging printks Sascha Hauer 2009-04-15 11:32 ` [PATCH 10/12] fec: switch to net_device_ops Sascha Hauer 2009-04-15 11:32 ` [PATCH 11/12] FEC Buffer rework Sascha Hauer 2009-04-15 11:32 ` [PATCH 12/12] fec: call fec_restart() in fec_open() Sascha Hauer 2009-04-16 9:38 ` David Miller 2009-04-16 9:38 ` [PATCH 11/12] FEC Buffer rework David Miller 2009-04-17 10:07 ` Greg Ungerer 2009-04-17 10:12 ` Sascha Hauer 2009-04-16 9:37 ` [PATCH 10/12] fec: switch to net_device_ops David Miller 2009-04-16 9:37 ` [PATCH 09/12] fec: remove debugging printks David Miller 2009-04-16 9:37 ` [PATCH 08/12] fec: align receive packets David Miller 2009-04-16 9:37 ` [PATCH 07/12] fec: refactor init function David Miller 2009-04-16 9:37 ` [PATCH 06/12] fec: refactor set_multicast_list() to make it more readable David Miller 2009-04-16 9:36 ` [PATCH 05/12] fec: Codingstyle cleanups David Miller 2009-04-16 9:36 ` [PATCH 04/12] fec: remove unnecessary cast David Miller 2009-04-16 9:36 ` [PATCH 03/12] fec: do not typedef struct types David Miller 2009-04-15 13:11 ` [PATCH 02/12] fec: switch to writel/readl Sascha Hauer 2009-04-16 9:36 ` David Miller 2009-04-15 12:12 ` [PATCH] FEC driver: patches for -next Greg Ungerer 2009-04-15 12:55 ` Sascha Hauer
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).