summaryrefslogtreecommitdiff
path: root/cpu/mips/au1x00_eth.c
AgeCommit message (Collapse)Author
2009-05-16MIPS: Implement ethernet halt for au1x00Thomas Lange
Implement ethernet halt() by putting MAC0 in reset. If we do not do this, we will get memory corruption when ethernet frames are received during early OS boot. Signed-off-by: Thomas Lange <thomas@corelatus.se> Signed-off-by: Shinya Kuribayashi <skuribay@pobox.com>
2008-11-09net: Move initialization of Au1x00 SoC ethernet MAC to cpu_eth_initShinya Kuribayashi
This patch will move au1x00_eth_initialize from net/eth.c to cpu_eth_init as a part of ongoing eth_initialize cleanup work. The function ret value is also fixed as it should be negative on fail. Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp> Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
2008-10-18rename CFG_ macros to CONFIG_SYSJean-Christophe PLAGNIOL-VILLARD
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
2008-06-07[MIPS] cpu/mips/Makefile: Split [CS]OBJS onto separate linesShinya Kuribayashi
Also get rid of some #ifdefs in *.c files. Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
2008-06-07[MIPS] Rename Alchemy processor configs into CONFIG_SOC_*Shinya Kuribayashi
CONFIG_SOC_AU1X00 Common Alchemy Au1x00 stuff. All Alchemy processor based machines need to have this config as a system type specifier. CONFIG_SOC_AU1000, CONFIG_SOC_AU1100, CONFIG_SOC_AU1200, CONFIG_SOC_AU1500, CONFIG_SOC_AU1550 Machine type specifiers. Each port should have one of aboves. Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
2007-11-17[MIPS] au1x00_eth.c: Fixed a warning on pb1000 build.Shinya Kuribayashi
au1x00_eth.c: In function 'au1x00_miiphy_write': au1x00_eth.c:139: warning: 'return' with no value, in function returning non-void Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
2007-11-17[MIPS] au1x00_eth.c: Fix au1x00_miiphy_{read,write} build errorShinya Kuribayashi
au1x00_eth.c: In function 'au1x00_enet_initialize': au1x00_eth.c:246: error: 'au1x00_miiphy_read' undeclared (first use in this function) au1x00_eth.c:246: error: (Each undeclared identifier is reported only once au1x00_eth.c:246: error: for each function it appears in.) au1x00_eth.c:246: error: 'au1x00_miiphy_write' undeclared (first use in this function) au1x00_eth.c: In function 'au1x00_miiphy_write': au1x00_eth.c:298: warning: 'return' with no value, in function returning non-void make[1]: *** [au1x00_eth.o] Error 1 Fixed by moving these two functions forward. Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
2007-07-09cpu/m*: Remove obsolete references to CONFIG_COMMANDSJon Loeliger
Signed-off-by: Jon Loeliger <jdl@freescale.com>
2007-07-04cpu/ non-mpc*: Augment CONFIG_COMMANDS tests with defined(CONFIG_CMD_*).Jon Loeliger
This is a compatibility step that allows both the older form and the new form to co-exist for a while until the older can be removed entirely. All transformations are of the form: Before: #if (CONFIG_COMMANDS & CFG_CMD_AUTOSCRIPT) After: #if (CONFIG_COMMANDS & CFG_CMD_AUTOSCRIPT) || defined(CONFIG_CMD_AUTOSCRIPT) Signed-off-by: Jon Loeliger <jdl@freescale.com>
2006-03-13au1x00_eth.c: check malloc return value and abort if it failedWolfgang Denk
Patch by Andrew Dyer, 26 Jul 2005
2005-10-28Add support for multiple PHYs.Marian Balakowicz
2005-09-25Add support for AMD's Pb1x00 eval board;Wolfgang Denk
add MII routines to the au1x00 ethernet driver; add USB ohci driver (work in progress) Patch by Thomas Sailer, 20 Jan 2005
2005-09-24Fix ethernet timeouts on dbau1550 and other au1x00 systemsWolfgang Denk
Patch by Leif Lindholm, 29 Dec 2004
2005-01-09* Patch by Leif Lindholm, 23 Sep 2004:wdenk
add support for the AMD db1550 board * Patch by Travis Sawyer, 15 Sep 2004: Add CONFIG_SERIAL_MULTI support for ppc4xx, update README.serial_multi
2004-01-02* Patch by André Schwarz, 8 Dec 2003:wdenk
fixes for Davicom DM9102A Ethernet Chip (#define CONFIG_TULIP_FIX_DAVICOM): - TX and RX deskriptors must be quad-word aligned - does not work with only one TX deskriptor - standard reset method does not work * Patch by Masami Komiya, 08 Dec 2003: add RTL8139 ethernet driver * Patches by Ed Okerson, 07 Dec 2003: - fix ethernet for the AU1x00 processors in little-endian mode. - extend memsetup.S for the AU1x00 processors in BE and LE modes
2003-12-07* Patches by David Müller, 14 Nov 2003:wdenk
- board/mpl/common/common_util.c * implement support for BZIP2 compressed images * various cleanups (printf -> puts, ...) - board/mpl/common/flash.c * report correct errors to upper layers * check the erase fail and VPP low bits in status reg - board/mpl/vcma9/cmd_vcma9.c - board/mpl/vcma9/flash.c * various cleanups (printf -> puts, ...) - common/cmd_usb.c * fix typo in comment - cpu/arm920t/usb_ohci.c * support for S3C2410 is missing in #if line - drivers/cs8900.c * reinit some registers in case of error (cable missing, ...) - fs/fat/fat.c * support for USB/MMC devices is missing in #if line - include/configs/MIP405.h - include/configs/PIP405.h * enable BZIP2 support * enlarge malloc space to 1MiB because of BZIP2 support - include/configs/VCMA9.h * enable BZIP2 support * enlarge malloc space to 1MiB because of BZIP2 support * enable USB support - lib_arm/armlinux.c * change calling convention of ARM Linux kernel as described on http://www.arm.linux.org.uk/developer/booting.php * Patch by Thomas Lange, 14 Nov 2003: Split dbau1x00 into dbau1000, dbau1100 and dbau1500 configs to support all these AMD boards. * Patch by Thomas Lange, 14 Nov 2003: Workaround for mips au1x00 physical memory accesses (the au1x00 uses a 36 bit bus internally and cannot access physical memory directly. Use the uncached SDRAM address instead of the physical one.)
2003-10-09* Patch by Steven Scholz, 10 Oct 2003wdenk
- Add support for Altera FPGA ACEX1K * Patches by Thomas Lange, 09 Oct 2003: - Endian swap ATA identity for all big endian CPUs, not just PPC - MIPS only: New option CONFIG_MEMSIZE_IN_BYTES for passing memsize args to linux - add support for dbau1x00 board (MIPS32)