On 02/02/11 06:51, Scott Wood wrote:
On Tue, 1 Feb 2011 20:32:29 +0100 Wolfgang Denk wd@denx.de wrote:
Dear Scott Wood,
In message 20110201102446.23b4a2e9@udp111988uds.am.freescale.net you wrote:
Prior to the introduction of LDFLAGS_u-boot, was LDFLAGS not what was used? So before, anything that board/cpu code adds directly to LDFLAGS (maybe they're supposed to use PLATFORM_LDFLAGS, but not all do) was used in the final link. After 8aba9dc, only things in PLATFORM_LDFLAGS plus -Bstatic and -T are used in the final link.
And this is correct for all boards?
By "this" do you mean the switch to PLATFORM_LDFLAGS in 8aba9dc, or the switch back to LDFLAGS? It's not obvious to me that the dropping of board/cpu modifications to LDFLAGS except during partial link was an intentional change, or a correct one for all boards.
The only case I see where it makes any difference at all is arch/i386, which does LDFLAGS += --cref. From the description of --cref in the linker manual, it probably actually belongs in LDFLAGS_FINAL, though I'm not sure if it's harmless to include it in partial link or not. Currently, with 8aba9dc, it's included *only* in partial link.
It's also not clear to me what this option has to do with i386... it looks like an arch-neutral debugging feature that doesn't affect the actual u-boot image at all (the output goes into the map file).
--cref can be dropped from x86 - I really should send a patch to rename i386 to x86 ;)
Since x86 is currently a very simple platform to maintain (one CPU, one board) and I have been dealing with breakages do to non x86 cleanups for a while, I have no problem with tweaking the x86 linker options to be more arch neutral.
Regards,
Graeme