October 2016 - U-Boot

[U-Boot] Pull request: u-boot-net.git master
by Joe Hershberger 14 Oct '16

14 Oct '16

Hi Tom, The following changes since commit 5ebd27d860ec0c6e36f1b0f973653fe66a7360be: Merge branch 'master' of git://git.denx.de/u-boot-x86 (2016-10-12 13:59:26 -0400) are available in the git repository at: git://git.denx.de/u-boot-net.git master for you to fetch changes up to c9abfbdd66ed97e1cac8553d0b31bd23f68480a9: net: smsc95xx: fix DM MAC address reading (2016-10-13 12:25:40 -0500) ---------------------------------------------------------------- Andrea Merello (1): phy: atheros: add support for RGMII_ID, RGMII_TXID and RGMII_RXID Chris Packham (3): net: Add support for mv88e609x switches net: mv88e61xx: Add support for fixed links net: mvneta: fix typo in comment Guillaume GARDET (1): test: add NFS download test Hannes Schmelzer (1): net: write enetaddr down to hardware on env_callback Mugunthan V N (3): drivers: net: phy: atheros: add separate config for AR8031 include: configs: am335x: add Atheros phy support driver: net: cpsw: add support for RGMII id mode support and RMII clock source selection Peter Chubb (2): rtl8169: fix cache misalignment message on transmit. net: Fix cache misalignment message after network load operations Roger Quadros (2): board: am335x-icev2: add ethernet phy mode detection logic board: am335x: Always set eth/eth1addr environment variable Stephen Warren (1): net: smsc95xx: fix DM MAC address reading board/ti/am335x/board.c | 154 +++++++++++++++++++++++++++++++++--------- cmd/net.c | 3 - drivers/net/cpsw.c | 157 +++++++++++++++++++++++++++++++++++++++---- drivers/net/mvneta.c | 2 +- drivers/net/phy/atheros.c | 53 ++++++++++++++- drivers/net/phy/mv88e61xx.c | 61 ++++++++++++++++- drivers/net/rtl8169.c | 3 +- drivers/usb/eth/smsc95xx.c | 16 +++++ include/configs/am335x_evm.h | 2 + include/cpsw.h | 1 + net/eth-uclass.c | 1 + test/py/tests/test_net.py | 49 ++++++++++++++ 12 files changed, 447 insertions(+), 55 deletions(-) Thanks! -Joe

2 1

[U-Boot] [PATCH] cmd/nand.c: Remove unnecessarily repeated step
by Ahmed Samir Khalil 13 Oct '16

13 Oct '16

Getting the current NAND device is already done once as part of nand command. Therefore, repeating this step as part of the sub-commands is unnecessary. Signed-off-by: Ahmed Samir Khalil <engkhalil86(a)gmail.com> --- cmd/nand.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/cmd/nand.c b/cmd/nand.c index ec7f1df..71ffe85 100644 --- a/cmd/nand.c +++ b/cmd/nand.c @@ -528,8 +528,6 @@ static int do_nand(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) if (set_dev(dev)) return 1; - mtd = nand_info[dev]; - memset(&opts, 0, sizeof(opts)); opts.offset = off; opts.length = size; @@ -597,8 +595,6 @@ static int do_nand(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) if (set_dev(dev)) return 1; - mtd = nand_info[dev]; - if (argc > 4 && !str2long(argv[4], &pagecount)) { printf("'%s' is not a number\n", argv[4]); return 1; @@ -626,8 +622,6 @@ static int do_nand(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) rwsize = size; } - mtd = nand_info[dev]; - if (!s || !strcmp(s, ".jffs2") || !strcmp(s, ".e") || !strcmp(s, ".i")) { if (read) -- 2.7.4

2 1

[U-Boot] [PATCH] net: smsc95xx: fix DM MAC address reading
by Stephen Warren 13 Oct '16

13 Oct '16

From: Stephen Warren <swarren(a)nvidia.com> eth-uclass.c expects DM-capable Ethernet adapters to implement ops-> read_rom_hwaddr(), or for some other mechanism to set pdata->enetaddr, or for the user to set environment variable $usbethaddr. Without any of these, it will refuse to initialize the device since no valid MAC address is known. Implement this function for the smsc95xx driver. With this feature implemented, there is no point smsc95xx_init_common() re-reading the MAC address from ROM, so ifdef out this code when DM_ETH is enabled. This allows (at least) the built-in Ethernet on the NVIDIA Harmony board to operate again. Fixes: 0990fcb77219 ("net: smsc95xx: Add driver-model support") Signed-off-by: Stephen Warren <swarren(a)nvidia.com> --- drivers/usb/eth/smsc95xx.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/usb/eth/smsc95xx.c b/drivers/usb/eth/smsc95xx.c index 7d9abfda3be1..d4c8ea4a98aa 100644 --- a/drivers/usb/eth/smsc95xx.c +++ b/drivers/usb/eth/smsc95xx.c @@ -519,9 +519,11 @@ static int smsc95xx_init_common(struct usb_device *udev, struct ueth_data *dev, debug("timeout waiting for PHY Reset\n"); return -ETIMEDOUT; } +#ifndef CONFIG_DM_ETH if (!priv->have_hwaddr && smsc95xx_init_mac_address(enetaddr, udev) == 0) priv->have_hwaddr = 1; +#endif if (!priv->have_hwaddr) { puts("Error: SMSC95xx: No MAC address set - set usbethaddr\n"); return -EADDRNOTAVAIL; @@ -1022,6 +1024,19 @@ int smsc95xx_write_hwaddr(struct udevice *dev) return smsc95xx_write_hwaddr_common(udev, priv, pdata->enetaddr); } +int smsc95xx_read_rom_hwaddr(struct udevice *dev) +{ + struct usb_device *udev = dev_get_parent_priv(dev); + struct eth_pdata *pdata = dev_get_platdata(dev); + int ret; + + ret = smsc95xx_init_mac_address(pdata->enetaddr, udev); + if (ret) + memset(pdata->enetaddr, 0, 6); + + return 0; +} + static int smsc95xx_eth_probe(struct udevice *dev) { struct smsc95xx_private *priv = dev_get_priv(dev); @@ -1037,6 +1052,7 @@ static const struct eth_ops smsc95xx_eth_ops = { .free_pkt = smsc95xx_free_pkt, .stop = smsc95xx_eth_stop, .write_hwaddr = smsc95xx_write_hwaddr, + .read_rom_hwaddr = smsc95xx_read_rom_hwaddr, }; U_BOOT_DRIVER(smsc95xx_eth) = { -- 2.9.3

3 6

[U-Boot] [PATCH] test: add NFS download test
by Guillaume GARDET 13 Oct '16

13 Oct '16

Add a NFS download test, based on TFTP test. Tested on i.MX6 SabreLite board. Signed-off-by: Guillaume GARDET <guillaume.gardet(a)free.fr> Cc: Tom Rini <trini(a)konsulko.com> Cc: Joe Hershberger <joe.hershberger(a)ni.com> Cc: Stephen Warren <swarren(a)wwwdotorg.org> Cc: Simon Glass <sjg(a)chromium.org> --- test/py/tests/test_net.py | 49 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 49 insertions(+) diff --git a/test/py/tests/test_net.py b/test/py/tests/test_net.py index 4ab58b4..0884051 100644 --- a/test/py/tests/test_net.py +++ b/test/py/tests/test_net.py @@ -47,6 +47,15 @@ env__net_tftp_readable_file = { "size": 5058624, "crc32": "c2244b26", } + +# Details regarding a file that may be read from a NFS server. This variable +# may be omitted or set to None if NFS testing is not possible or desired. +env__net_nfs_readable_file = { + "fn": "ubtest-readable.bin", + "addr": 0x10000000, + "size": 5058624, + "crc32": "c2244b26", +} """ net_set_up = False @@ -157,3 +166,43 @@ def test_net_tftpboot(u_boot_console): output = u_boot_console.run_command('crc32 %x $filesize' % addr) assert expected_crc in output + +(a)pytest.mark.buildconfigspec('cmd_nfs') +def test_net_nfs(u_boot_console): + """Test the nfs command. + + A file is downloaded from the NFS server, its size and optionally its + CRC32 are validated. + + The details of the file to download are provided by the boardenv_* file; + see the comment at the beginning of this file. + """ + + if not net_set_up: + pytest.skip('Network not initialized') + + f = u_boot_console.config.env.get('env__net_nfs_readable_file', None) + if not f: + pytest.skip('No NFS readable file to read') + + addr = f.get('addr', None) + if not addr: + addr = u_boot_utils.find_ram_base(u_boot_console) + + fn = f['fn'] + output = u_boot_console.run_command('nfs %x %s' % (addr, fn)) + expected_text = 'Bytes transferred = ' + sz = f.get('size', None) + if sz: + expected_text += '%d' % sz + assert expected_text in output + + expected_crc = f.get('crc32', None) + if not expected_crc: + return + + if u_boot_console.config.buildconfig.get('config_cmd_crc32', 'n') != 'y': + return + + output = u_boot_console.run_command('crc32 %x $filesize' % addr) + assert expected_crc in output -- 1.8.4.5

5 4

[U-Boot] [PATCH v3] net: Fix cache misalignment message after network load operations
by Peter.Chubb＠data61.csiro.au 13 Oct '16

13 Oct '16

After any operation that downloads a file (e.g., pxe get, or dhcp), the buffer containing the downloaded data is flushed. This is unnecessary and annoying. Unnecessary, because the network driver should already have fliushed the cache for the DMAed area, and annoying because it generates a cache misalignment message. Signed-off-by: Peter Chubb <peter.chubb(a)data61.csiro.au> --- cmd/net.c | 3 --- 1 file changed, 3 deletions(-) Index: u-boot/cmd/net.c =================================================================== --- u-boot.orig/cmd/net.c 2016-09-07 13:50:46.616156851 +1000 +++ u-boot/cmd/net.c 2016-09-07 19:18:18.962450874 +1000 @@ -243,9 +243,6 @@ static int netboot_common(enum proto_t p return CMD_RET_SUCCESS; } - /* flush cache */ - flush_cache(load_addr, size); - bootstage_mark(BOOTSTAGE_ID_NET_LOADED); rcode = bootm_maybe_autostart(cmdtp, argv[0]);

5 6

[U-Boot] [PATCH v2] rtl8169: fix cache misalignment message on transmit.
by Peter.Chubb＠data61.csiro.au 13 Oct '16

13 Oct '16

The call to flush cache on the transmit buffer was misplaced (for very short packets) and asked to flush less than a cacheline. Move the flush cache call to after a short packet has been padded to minimum length (so the padding is flushed too), and round the size up to a cacheline. Signed-off-by: Peter Chubb <peter.chubb(a)data61.csiro.au> Acked-by: Joe Hershberger <joe.hershberger(a)ni.com> --- drivers/net/rtl8169.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/rtl8169.c b/drivers/net/rtl8169.c index 1cc0b40..a3f4423 100644 --- a/drivers/net/rtl8169.c +++ b/drivers/net/rtl8169.c @@ -629,11 +629,12 @@ static int rtl_send_common(pci_dev_t dev, unsigned long dev_iobase, /* point to the current txb incase multiple tx_rings are used */ ptxb = tpc->Tx_skbuff[entry * MAX_ETH_FRAME_SIZE]; memcpy(ptxb, (char *)packet, (int)length); - rtl_flush_buffer(ptxb, length); while (len < ETH_ZLEN) ptxb[len++] = '\0'; + rtl_flush_buffer(ptxb, ALIGN(len, RTL8169_ALIGN)); + tpc->TxDescArray[entry].buf_Haddr = 0; #ifdef CONFIG_DM_ETH tpc->TxDescArray[entry].buf_addr = cpu_to_le32( -- 2.9.3

3 2

[U-Boot] [PATCH] net: write enetaddr down to hardware on env_callback
by Hannes Schmelzer 13 Oct '16

13 Oct '16

If mac-address is changed using "setenv ethaddr ...." command the new mac-adress also must be written into the responsible ethernet driver. Signed-off-by: Hannes Schmelzer <oe5hpm(a)oevsv.at> --- net/eth-uclass.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/eth-uclass.c b/net/eth-uclass.c index c15cc4d..a32961e 100644 --- a/net/eth-uclass.c +++ b/net/eth-uclass.c @@ -227,6 +227,7 @@ static int on_ethaddr(const char *name, const char *value, enum env_op op, case env_op_create: case env_op_overwrite: eth_parse_enetaddr(value, pdata->enetaddr); + eth_write_hwaddr(dev); break; case env_op_delete: memset(pdata->enetaddr, 0, 6); -- 1.9.1

5 8

[U-Boot] [PATCH] net: mvneta: fix typo in comment
by Chris Packham 13 Oct '16

13 Oct '16

Signed-off-by: Chris Packham <judge.packham(a)gmail.com> --- drivers/net/mvneta.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/mvneta.c b/drivers/net/mvneta.c index c23fe9a9d5a7..bbfce9604710 100644 --- a/drivers/net/mvneta.c +++ b/drivers/net/mvneta.c @@ -1589,7 +1589,7 @@ static int mvneta_probe(struct udevice *dev) /* * Allocate buffer area for descs and rx_buffers. This is only * done once for all interfaces. As only one interface can - * be active. Make this area DMA save by disabling the D-cache + * be active. Make this area DMA safe by disabling the D-cache */ if (!buffer_loc.tx_descs) { /* Align buffer area for descs and rx_buffers to 1MiB */ -- 2.9.2.518.ged577c6.dirty

3 2

[U-Boot] [PATCH v1 1/2] net: Add support for mv88e609x switches
by Chris Packham 13 Oct '16

13 Oct '16

The Marvell Link Street mv88e60xx is a series of FastEthernet switch chips, some of which also support Gigabit ports. It is similar to the mv88e61xx series which support Gigabit on all ports. The main difference is the number of ports. Which affects the PORT_COUNT define and the size of the mask passed to mv88e61xx_port_set_vlan(). Other than that it's just a matter of adding the appropriate chip IDs. Signed-off-by: Chris Packham <judge.packham(a)gmail.com> Cc: Joshua Scott <joshua.scott(a)alliedtelesis.co.nz> --- This is a re-work of Joshua's patch http://patchwork.ozlabs.org/patch/425556/ Thanks to the work from Kevin Smith the changes to support the 88E609x are quite trivial. I've tested this on a board with a 88E6097. It should also work for a 88E6096 based on the datasheet. drivers/net/phy/mv88e61xx.c | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/mv88e61xx.c b/drivers/net/phy/mv88e61xx.c index 74d5609..7385848 100644 --- a/drivers/net/phy/mv88e61xx.c +++ b/drivers/net/phy/mv88e61xx.c @@ -40,7 +40,7 @@ #define PHY_AUTONEGOTIATE_TIMEOUT 5000 -#define PORT_COUNT 7 +#define PORT_COUNT 11 #define PORT_MASK ((1 << PORT_COUNT) - 1) /* Device addresses */ @@ -167,6 +167,8 @@ #endif /* ID register values for different switch models */ +#define PORT_SWITCH_ID_6096 0x0980 +#define PORT_SWITCH_ID_6097 0x0990 #define PORT_SWITCH_ID_6172 0x1720 #define PORT_SWITCH_ID_6176 0x1760 #define PORT_SWITCH_ID_6240 0x2400 @@ -580,7 +582,7 @@ static int mv88e61xx_port_enable(struct phy_device *phydev, u8 port) } static int mv88e61xx_port_set_vlan(struct phy_device *phydev, u8 port, - u8 mask) + u16 mask) { int val; @@ -974,9 +976,21 @@ static struct phy_driver mv88e61xx_driver = { .shutdown = &genphy_shutdown, }; +static struct phy_driver mv88e609x_driver = { + .name = "Marvell MV88E609x", + .uid = 0x1410c89, + .mask = 0xfffffff0, + .features = PHY_GBIT_FEATURES, + .probe = mv88e61xx_probe, + .config = mv88e61xx_phy_config, + .startup = mv88e61xx_phy_startup, + .shutdown = &genphy_shutdown, +}; + int phy_mv88e61xx_init(void) { phy_register(&mv88e61xx_driver); + phy_register(&mv88e609x_driver); return 0; } -- 2.9.2.518.ged577c6.dirty

3 5

[U-Boot] [PATCH v2 0/3] Add Atheros phy delay support and CPSW RGMII id support
by Mugunthan V N 13 Oct '16

13 Oct '16

This patch series adds support for RGMII phy internal delay support for Atheros 8031 phy driver and also add support for the phy mode configuration in control module in cpsw driver. Tested the series on the following EVMs with a zImage download from server, logs [1] and pushed a branch for testing [2]. * DRA 72 Rev B EVM * DRA 72 Rev C EVM * DRA 74 EVM * AM437x GP EVM * AM335x BBB * AM335x GP EVM Changes from initial version: * Added RMII refl clock "rmii-clock-ext" DT parsing. [1] - http://pastebin.ubuntu.com/23317809/ [2] - git://git.ti.com/~mugunthanvnm/ti-u-boot/mugunth-ti-u-boot.git cpsw-rgmii-v2 Mugunthan V N (3): drivers: net: phy: atheros: add separate config for AR8031 include: configs: am335x: add Atheros phy support driver: net: cpsw: add support for RGMII id mode support and RMII clock source selection drivers/net/cpsw.c | 157 +++++++++++++++++++++++++++++++++++++++---- drivers/net/phy/atheros.c | 37 +++++++++- include/configs/am335x_evm.h | 2 + include/cpsw.h | 1 + 4 files changed, 182 insertions(+), 15 deletions(-) -- 2.10.1.445.g3cdd5d1

2 6