| Message ID | 1412942554-752-17-git-send-email-ian.campbell@citrix.com |
|---|---|
| State | New |
| Headers | show |
Ian Campbell writes ("[PATCH RFC OSSTEST 17/19] Osstest/Debian: Remove hardcoded midway specific addresses from boot.scr"):
> This needs to go along with a change to the hostdb to add "setenv xen_addr_r
> 0x01000000" to the UBootScriptEarlyCommands property of the midway machines.
`Go along with' is not possible. Do you mean `this depends on' ?
> This is going to be a pretty common requirement on ARM/uboot
> systems, should I make it a specific property instead of reusing the
> generic hook?
Yes.
Ian.
On Fri, 2014-10-10 at 15:30 +0100, Ian Jackson wrote: > Ian Campbell writes ("[PATCH RFC OSSTEST 17/19] Osstest/Debian: Remove hardcoded midway specific addresses from boot.scr"): > > This needs to go along with a change to the hostdb to add "setenv xen_addr_r > > 0x01000000" to the UBootScriptEarlyCommands property of the midway machines. > > `Go along with' is not possible. Do you mean `this depends on' ? Yes. The hostdb change can happen first without breaking anything. > > This is going to be a pretty common requirement on ARM/uboot > > systems, should I make it a specific property instead of reusing the > > generic hook? > > Yes. Ack
Ian Campbell writes ("[PATCH OSSTEST v2 17/20] Osstest/Debian: Remove hardcoded midway specific addresses from boot.scr"):
> This depends on a change to the hostdb to add "setenv xen_addr_r
> 0x01000000" to the UBootScriptEarlyCommands property of the midway
> machines.
>
> Most platforms will need something similar. For both cubietruck and arndale
> "setenv xen_addr_r 0x41000000" is appropriate.
>
> This is going to be a pretty common requirement on ARM/uboot
> systems, should I make it a specific property instead of reusing the
> generic hook?
I don't understand why this question is still in the commit message.
I already answered it with "yes" and you said "ack" ...
Ian.
On Wed, 2014-10-29 at 16:37 +0000, Ian Jackson wrote: > Ian Campbell writes ("[PATCH OSSTEST v2 17/20] Osstest/Debian: Remove hardcoded midway specific addresses from boot.scr"): > > This depends on a change to the hostdb to add "setenv xen_addr_r > > 0x01000000" to the UBootScriptEarlyCommands property of the midway > > machines. > > > > Most platforms will need something similar. For both cubietruck and arndale > > "setenv xen_addr_r 0x41000000" is appropriate. > > > > This is going to be a pretty common requirement on ARM/uboot > > systems, should I make it a specific property instead of reusing the > > generic hook? > > I don't understand why this question is still in the commit message. > I already answered it with "yes" and you said "ack" ... Looks like I overlooked that when refreshing my memory of the last time around, sorry. Ian.
diff --git a/Osstest/Debian.pm b/Osstest/Debian.pm index eca4d7f..4539452 100644 --- a/Osstest/Debian.pm +++ b/Osstest/Debian.pm @@ -155,10 +155,6 @@ ${early_commands} fdt set /chosen \\\#address-cells <1> fdt set /chosen \\\#size-cells <1> -setenv xen_addr_r 0x01000000 -# kernel_addr_r=0x02000000 -# ramdisk_addr_r=0x04000000 - ext2load scsi 0 \\\${xen_addr_r} \$xen setenv bootargs "$xenhopt" echo Loaded \$xen to \\\${xen_addr_r} (\\\${filesize})
This needs to go along with a change to the hostdb to add "setenv xen_addr_r 0x01000000" to the UBootScriptEarlyCommands property of the midway machines. Most platforms will need something similar. For both cubietruck and arndale "setenv xen_addr_r 0x41000000" is appropriate. This is going to be a pretty common requirement on ARM/uboot systems, should I make it a specific property instead of reusing the generic hook? Signed-off-by: Ian Campbell <ian.campbell@citrix.com> --- Osstest/Debian.pm | 4 ---- 1 file changed, 4 deletions(-)