summaryrefslogtreecommitdiff
path: root/cleopatre
AgeCommit message (Collapse)Author
2012-05-22cleo/u-boot: set linux to reboot on panic, closes #3084Yacine Belkadi
Ask the Linux kernel to reboot after 1 second, if a panic occurs. This is set in the boot parameters, instead of in a sysctl.conf file, in order to cover panics occuring during the boot process (from drivers for example).
2012-05-15cleopatre: make the bundle in the distant working directory, closes #3079Nélio Laranjeiro
2012-05-09cleo/devkit/plcdrv: add missing stats for data packets, closes #3001Jérémy Dufour
2012-05-09cleo/devkit/tests/plcdrv: add missing stubs for semaphoreJérémy Dufour
This commit is a fix of commit [51ac9b1b2].
2012-05-09cleo/buildroot: update _defconfig files after adding spc300dini, closes #2759MSE500 Team
2012-05-09cleo/buildroot: add new board spc300dini (2), refs #2759MSE500 Team
2012-05-09cleo/buildroot: add new board spc300dini (1), refs #2759Yacine Belkadi
spr300 was chosen because, just like the spc300dini: - it has a Gigabit phy - it has a MEDIUM_1 flash (4MiB)
2012-05-09cleo/buildroot: update the linux26.config files after adding spc300dini, ↵MSE500 Team
refs #2759
2012-05-09cleo/linux/drv: make ethernet work on spc300dini, refs #2759MSE500 Team
- Prevent advertising of 1000Base-T capability because the spc300dini doesn't support Gigabit Ethernet. - Apply a specific configuration to the phy (2ns delay for TX and RX clock).
2012-05-09cleo/linux: add board spc300dini (2), refs #2759MSE500 Team
2012-05-09cleo/linux: add board spc300dini (1), refs #2759Yacine Belkadi
Create a board file for spc300dini by copying the one from spr300. spr300 was chosen because, just like the spc300dini: - it has a Gigabit phy - it has a MEDIUM_1 flash (4MiB)
2012-05-09cleo/tools/openocd: add cfg and ocd files for spc300dini, closes #2758MSE500 Team
2012-05-09cleo/u-boot: on spc300dini, force Linux Kernel mem to 28 MiB, refs #2759Jean-Philippe SAVE
The spc300dini board has 64 MiB of RAM. 60 MiB of those should go to Linux, but a bug occurs in that case (see ticket #2842). As a workaround, the "mem" kernel parameter is used to pass only 28 MiB to the Linux kernel, just like on boards with 32 MiB of RAM.
2012-05-09cleo/u-boot: add Ethernet RX clock inversion support, refs #2758Nicolas Schodet
Might be needed for DINI prototype, not enabled by default.
2012-05-09cleo/u-boot: adapt Ethernet driver to spc300dini, refs #2758MSE500 Team
2012-05-09cleo/u-boot: fix missing nvram fields definesNicolas Schodet
2012-05-09cleo/u-boot: support spc300dini, refs #2758MSE500 Team
Support the SPC300c (MARIA2) prototype which is an FPGA based on DINI boards.
2012-05-09cleo/u-boot: remove autoconfig to sdk300 board, refs #2758Belkadi & Save
The autoconfig, added by #1249, prevents to configure a new board, so we have to remove it.
2012-05-09cleo/u-boot: add support for MIU controller, refs #2758MSE500 Team
2012-05-09cleo/u-boot: compute flash SPI speedNicolas Schodet
2012-05-09cleo/u-boot: switch to board/chip/feature configurationNicolas Schodet
2012-05-09cleo/linux: switch to board/chip/feature configurationNicolas Schodet
2012-05-04cleopatre/tools/openocd: adapt Makefile.am to last automake version, closes ↵Save & Schodet
#3070 Update Makefile.am file to be compatible to new automake version (upper to 1.11.1). New automake is more strict about directories used for specific kind of files; DATA files should be in *datadir. This does not imply any change in source code as pkgdatadir & pkglibdir are both included in the same search path (see helper/options.c).
2012-05-03cleo/devkit/plcdrv: remove /proc/net/plc/regbank, refs #3051Yacine Belkadi
Previous commits added "/proc/reset_cause" to provide the reset cause to userland. As a result, /proc/net/plc/regbank is no longer needed.
2012-05-03cleo/linux/spc300: export the reset_cause strings, refs #3051Yacine Belkadi
Export to userland, the strings returned by reading "/proc/reset_cause"
2012-05-03cleo/app/fw_wd: identify resets due to firmware watchdog, refs #3051Yacine Belkadi
When the firmware watchdog complains, fw_wd may be configured to reboot. The problem is that "/proc/reset_cause" will say it was due to software reset. So there is no distinction between 1) a soft reset and 2) a soft reset due to the complain of the firmware watchdog. With this commit, fw_wd informs that the real reason for the reset is the firmware watchdog.
2012-05-03cleo/devkit/plcdrv: identify resets due to firmware watchdog, refs #3051Yacine Belkadi
When the firmware watchdog complains, and there is no one to handle that, plcdrv performs a reset. The problem is that "/proc/reset_cause" will say it was due to software reset. So there is no distinction between 1) a soft reset and 2) a soft reset due to the complain of the firmware watchdog. With this commit, plcdrv informs that the real reason for the reset is the firmware watchdog.
2012-05-03cleo/linux/spc300: add /proc/reset_cause, refs #3051Yacine Belkadi
Add "/proc/reset_cause" to expose what caused the last reset. The objective is to replace the current way for applications to get that info, which is to analyze some hardware registers exposed via /proc/net/plc/regbank. Because determining the reset cause is low level and hardware related, it seems appropriate to hide that in the kernel, and only expose the final cause to userspace.
2012-05-03cleo/linux/spc300/hardware/import: fix values of RB_PACKAGEYacine Belkadi
See: - ticket #561 on the Hardware Team's Trac (http://stestephe/trac/ticket/561). - Changeset r9120 (http://stestephe/trac/changeset/9120).
2012-05-03cleo/devkit/plcdrv: use arch_reset() instead of writing to hardware regYacine Belkadi
Refactor: In order to reset, use arch_reset() instead of writing directly to the hardware reset register.
2012-04-26cleo/linux/drivers/net/arm: accept giant frames, closes #2992Nicolas Schodet
2012-04-26cleo/linux: bridge 802.1ad and 802.1QinQ frames, refs #2992Nicolas Schodet
2012-04-19cleo/devkit/doc[eoc]: create training presentationCeline Buret
2012-04-18cleo/{linux,tools}: add new flashes Winbond / Eon, closes #2919Tom Yang
Support Winbond 25Q32 / Eon en25q32 / Eon en25q128. Conflicts: cleopatre/linux-2.6.25.10-spc300/drivers/mtd/devices/m25p80.c
2012-04-12cleo/devkit/doc: update managerd documentationCeline Buret
2012-04-10cleopatre/linux: raise SPI flash speed to 20 MHz, closes #3068Nicolas Schodet
2012-04-10cleopatre/u-boot: raise SPI speed to 18.375 MHz, closes #3067Nicolas Schodet
Speed computed for the maximum master clock, better setup will be done with mse500 commits.
2012-04-10cleopatre/linux/drivers/spi: round down SPI speed, closes #3066Nicolas Schodet
2012-04-10cleopatre: link libspid with libmme, refs #3065Nélio Laranjeiro
2012-04-04cleopatre/tools/genNVRAM: remove useless code noiseNicolas Schodet
2012-04-04cleopatre/tools/genNVRAM: fix command line inconsistenciesNicolas Schodet
2012-04-04cleopatre/tools/genNVRAM: use --ethernet instead of --ether, refs #3052Nicolas Schodet
2012-04-04cleopatre/tools/genNVRAM: fix argument parsingNicolas Schodet
2012-03-30cleo/app: fix sub make invocations, refs #3056Yacine Belkadi
When building the bundle, warnings similar to the one below appear: make[2]: warning: jobserver unavailable: using -j1. Add `+' to parent make rule. This comes from the recursive "make" calls in the makefiles. "$(MAKE)" or "+make" should be used instead.
2012-03-30cleo/linux/driver/net: redefine PKT_BUF_SZ using linux definesJérémy Dufour
2012-03-30cleo/linux/driver/net: use default ethernet change mtu function, closes #2969Jérémy Dufour
2012-03-30cleo/devkit/plcdrv: redefine PKT_BUF_SZ using linux definesJérémy Dufour
2012-03-30cleo/devkit/plcdrv: use default ethernet change mtu function, refs #2969Jérémy Dufour
2012-03-30cleopatre/buildroot/target: remove faulty MTU setting, refs #2969Nicolas Schodet
2012-03-30cleopatre/linux/arch/arm/mach-spc300: set IPMBOX priority, closes #2988Nélio Laranjeiro