Android-x86
Fork
Donation

  • R/O
  • HTTP
  • SSH
  • HTTPS

system-core: List of commits

system/core


RSS
Rev. Tiempo Autor
210a64b eclair-x86 2010-04-18 18:04:53 Chih-Wei Huang

init: change FIRMWARE_DIR for x86

On Android-x86, the firmware directory is /system/lib/firmware.

3408ea9 2010-04-16 12:04:18 Thorsten Glaser

Replace ash with mksh

Squashed commit for easier merging/testing.
Original commit messages before squashing follow below.

Further changes after squashing:
• updated mksh to Easter 2010 weekend (2010/04/05) version
• removed no longer compiled code from adb
• updated mksh to R39 2010/04/09 (extensively tested in Debian)

‣ This is a combination of 19 commits.
‣ The first commit's message is:
Import mksh-current as of 2010/01/12

Retrieved from: cvs -qd _anoncvs@anoncvs.mirbsd.org:/cvs co -PA mksh
Removed: Makefile, CVS/

This code is NOT A CONTRIBUTION, it shall be handled as an external
source. The licence it is under is suitable for Android; it is BSD-
like and contains an explicit copyright grant, and I know of no pa-
tents covering a shell.

‣ This is the 2nd commit message:

Remove NetBSD® ash

‣ This is the 3rd commit message:

Import printf.c from MirBSD

Retrieved via:
cvs -qd _anoncvs@anoncvs.mirbsd.org:/cvs co -PA \
src/usr.bin/printf/printf.c

This is derived from the original BSD source code and slightly
adapted to work well as optional mksh builtin.

‣ This is the 4th commit message:

Add arc4random_pushb implementation for Android

Retrieved from:
cvs -qd _anoncvs@anoncvs.mirbsd.org:/cvs co -PA \
contrib/hosted/tg/arc4rootdom.c

This is code written by me specifically for Android
and starts the series of commits that ARE contributions.

‣ This is the 5th commit message:

Add module licence and summary of legal notices of code actually compiled.

‣ This is the 6th commit message:

Add script used in (manually) creating/adjusting Android.mk

‣ This is the 7th commit message:

And finally, the (hopefully working) Android.mk file for mksh as sh

‣ This is the 8th commit message:

Now that we have mksh, nuke that ugly SH_HISTORY hack.

‣ This is the 9th commit message:

shlist.c is no longer needed without -DSH_HISTORY

‣ This is the 10th commit message:

Get rid of floating point operations.

This was only needed for the (optional) printf(1) builtin,
and the shell shouldn't ever need floating point...
besides, it will work better without in thumb mode.

‣ This is the 11th commit message:

Update printf.c from MirBSD-current

‣ This is the 12th commit message:

Update to mksh R39b (release)

R39b was released today; since the snapshot I initially began
working on is close enough, the build system requires no fur-
ther changes.

The source code is still unmodified from upstream.

‣ This is the 13th commit message:

Upgrade to mksh R39c (release)

‣ This is the 14th commit message:

remove -fstack-protector-all from CFLAGS, it breaks at least on Android-x86

‣ This is the 15th commit message:

add symlink /system/bin/mksh -> /system/bin/sh

‣ This is the 16th commit message:

Disable the mknod(8) builtin for space reasons

‣ This is the 17th commit message:

Upgrade to mksh-current as of 27.03.2010 21:44 UTC

Also, get rid of the NDK and mksh-small part of Android.mk and NOTICE;
we only build the full-features one now, which is tested on Android-x86.

‣ This is the 18th commit message:

Set the MKSH_DEFAULT_EXECSHELL to /system/bin/sh to unbreak “adb install”…

… and any other places that call a script without a shebang.
Discovered and solution suggested by: Chih-Wei Huang <cwhuang@android-x86.org>

‣ This is the 19th commit message:

Add minimal MKSHRC_PATH=/system/etc/mkshrc file for Android ($PS1, …)

Change-Id: I85a534a150197c0bc73ea4e8bde04cf6ad0811b8

Signed-off-by: Thorsten Glaser <tg@mirbsd.org>

9b4a013 2010-04-16 12:04:17 Thorsten Glaser

sys_signame[] was added to libc, remove conflicting definition

Signed-off-by: Thorsten Glaser <tg@mirbsd.org>

59a8503 2010-04-16 12:04:17 Yi Sun

enable acc

2f9c883 2010-04-16 12:04:17 Chih-Wei Huang

do not include goldfish files on x86

5612c52 2010-04-16 12:04:17 Chih-Wei Huang

init: a hack to avoid kernel warning

Change options androidboot.* to androidboot_* so kernel won't warn it.

76383ab 2010-04-16 12:04:17 Chih-Wei Huang

vold: rfkill bootstrapping

Simulate rfkill add uevents to change permission of bluetooth devices on boot.
This is better since it works after suspend and wakeup.

8c9f0ec 2010-04-16 12:04:17 Chih-Wei Huang

vold: usb drive bootstrapping

The usb drive plugged before booting can be mounted by this change.

Original author: Luke Chen <jschen.cse@gmail.com>

77dc70b 2010-04-16 12:02:42 Chih-Wei Huang

init: add an option to disable 3D acceleration

If the hardware OpenGL support doesn't work for you,
disable it by adding HWACCEL=0 to cmdline.

d122853 2010-04-14 19:34:35 Chia-I Wu

init: Create subdirectory for drm subsystem.

b35476a 2010-04-14 19:33:45 Yi Sun

fixed a bug caused by the changes in the prop name size

f4d4c8f 2010-04-14 19:33:44 Yi SUn

add etherenet state tracker support.

The dhcpcd somehow can not handle multiple interfaces with on command line.
I have to create different services for different interfaces

cf55f25 2010-04-14 19:33:42 Chih-Wei Huang

fix permission of /dev/videoX, and remove unnecessary /dev/snd/*

b74386d 2010-04-14 19:33:07 Chih-Wei Huang

patch from Yi Sun to fix power status issue

a24f3a0 2010-04-14 19:19:54 Chih-Wei Huang

vold: add fat partition type detection, and fix a crash bug

7744356 2010-04-14 19:19:11 Chih-Wei Huang

vold: auto-mount MMC/SD and USB drive

Based on Yi Sun's work: add vold support on usb drive

ad305d3 2010-04-14 19:18:38 Chih-Wei Huang

vold: validate fw_sock before trying FD_ISSET it

72a8d1a 2010-04-14 19:18:31 Chih-Wei Huang

vold: reimplement get_uevent_param in the correct way

The original implementation can not find correct parameters for certain
cases. For example, if you have both foo_bar and foo, try to find foo
may be mismatched. The correct way is to match with "=".

Modify from original patch of Yi Sun.

613d75f 2010-04-14 19:18:14 Jim Huang

Let console applications have a controlling tty.

Since the logic in open_console() is changed, terminating console
application within a shell is now possible via Ctrl-C.

4db6705 2010-04-14 19:17:28 Dan Bornstein

Define HAVE_SYS_SENDFILE_H.

Platforms differ on whether sys/sendfile.h or sys/socket.h is where to
find the declaration for sendfile().

870f3df 2010-04-14 19:12:28 Alexey Tarasov

Make get_my_path() safer

Adds maxLen parameter to get_my_path().
Some small cosmetic fixes.

c9cb031 2010-04-14 19:12:28 San Mehat

libsysutils: Fix some bugs in NetlinkListener and NetlinkEvent

Signed-off-by: San Mehat <san@android.com>

d4be7c7 2010-04-14 19:12:28 Sean McNeil

Set permissions properly for any ALSA device under /dev/snd

d939ad5 2010-04-14 19:12:28 Ben Gruver

Allow an onrestart command to stop the current service

For example,

service recovery /sbin/recovery
onrestart start loadkeys
onrestart class_start con
onrestart stop recovery

Previously, if you had a service like the above, the "onrestart stop
recovery" clause would not have any effect, because the restart flag
would be re-set after the stop command had executed.

This is similar to a service with the "oneshot" keyword, with the critical
difference being that it executes the other onrestart commands when the
service dies.

21d1a11 2010-04-14 19:12:28 Jay Freeman (saurik)

added chdir/chroot commands to init for mount switching/pivoting

3265251 2010-04-14 19:12:28 Jay Freeman (saurik)

Trivial implementation of init's import command.

97778f4 2010-04-14 19:12:28 Jay Freeman (saurik)

completed implementation of "trigger" for init

6abf935 2010-04-14 19:12:28 Chuck Tuffli

Fix incorrect check for LOG_UEVENTS

LOG_UEVENTS is a boolean value as defined in system/core/init/init.h.
Therefore, code should use an #if check and not an #ifdef check as the
macro will always be defined.

26771a2 2010-04-14 19:12:28 Hiroshi Takekawa

init.rc: Fixes typo: sytem -> system.

e5d5a57 2010-04-14 19:12:28 Xinyu Chen

Fix bug of mmc.c not checking read_file result.
This bug causes segment fault when reading name node of
SDIO mmc device's /sys file, which is not existed at all.

Signed-off-by: Xinyu Chen <xinyu.chen@freescale.com>

Show on old repository browser