summaryrefslogtreecommitdiffstats
path: root/getting-started/machines/porter.md
diff options
context:
space:
mode:
authorMartin Kelly <mkelly@xevo.com>2017-05-02 10:21:26 -0700
committerMartin Kelly <mkelly@xevo.com>2017-05-02 10:35:51 -0700
commit0c70d7729bd20308a42fcbec32599698841b2c25 (patch)
tree366b3136e4481875cf8afa9aae5604199b3af20b /getting-started/machines/porter.md
parent4e9dc8be544e2a107d35860d1a9852e16d60cc03 (diff)
r-car: boot with ro, not rw
Booting with ro is safer in the case of filesystem corruption because it prevents processes from writing to a broken filesystem before the kernel has a chance to fsck and then remount rw. For this reason, it seems to be the standard in other Linux distros. From my testing, I don't see any drawback to it. Signed-off-by: Martin Kelly <mkelly@xevo.com>
Diffstat (limited to 'getting-started/machines/porter.md')
-rw-r--r--getting-started/machines/porter.md2
1 files changed, 1 insertions, 1 deletions
diff --git a/getting-started/machines/porter.md b/getting-started/machines/porter.md
index f3f6d15..85ff695 100644
--- a/getting-started/machines/porter.md
+++ b/getting-started/machines/porter.md
@@ -364,7 +364,7 @@ setenv ethaddr 2e:09:0a:00:75:b5
```
setenv bootargs_console 'console=ttySC6,38400 ignore_loglevel'
setenv bootargs_video 'vmalloc=384M video=HDMI-A-1:1920x1080-32@60'
-setenv bootargs_root 'root=/dev/mmcblk0p1 rootdelay=3 rw rootfstype=ext4 rootwait'
+setenv bootargs_root 'root=/dev/mmcblk0p1 rootdelay=3 ro rootfstype=ext4 rootwait'
setenv bootmmc '1:1'
setenv bootcmd_sd 'ext4load mmc ${bootmmc} 0x40007fc0 boot/uImage+dtb'
setenv bootcmd 'setenv bootargs ${bootargs_console} ${bootargs_video} ${bootargs_root}; run bootcmd_sd; bootm 0x40007fc0'