Commit f62fb99941 fixed handling of all rodata sections by using a
wildcard combined with calls to ld's builtin functions SORT_BY_ALIGNMENT()
and SORT_BY_NAME(). Unfortunately these functions were only
introduced with biunutils version 2.16, so the modification broke
building with all tool chains using older binutils.
This patch makes it work again. This is done by omitting the use of
these functions for such old tool chains. This will result in
slightly larger target binaries, as the rodata sections are no longer
in optimal order alignment-wise which reauls in unused gaps, but the
effect was found to be insignificant - especially compared to the fact
that you cannot build U-Boot at all in the current state.
As ld seems to have no support for conditionals we run the linker
script through the C preprocessor which can be easily used to remove
the unwanted function calls.
Note that the C preprocessor must be run with the "-ansi" (or a
"-std=") option to make sure all the system-specific predefined
macros outside the reserved namespace are suppressed. Otherise, cpp
might for example substitute "powerpc" to "1", thus corrupting for
example "OUTPUT_ARCH(powerpc)" etc.
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Mike Frysinger <vapier@gentoo.org>
A recent gcc added a new unaligned rodata section called '.rodata.str1.1',
which needs to be added the the linker script. Instead of just adding this
one section, we use a wildcard ".rodata*" to get all rodata linker section
gcc has now and might add in the future.
However, '*(.rodata*)' by itself will result in sub-optimal section
ordering. The sections will be sorted by object file, which causes extra
padding between the unaligned rodata.str.1.1 of one object file and the
aligned rodata of the next object file. This is easy to fix by using the
SORT_BY_ALIGNMENT command.
This patch has not be tested one most of the boards modified. Some boards
have a linker script that looks something like this:
*(.text)
. = ALIGN(16);
*(.rodata)
*(.rodata.str1.4)
*(.eh_frame)
I change this to:
*(.text)
. = ALIGN(16);
*(.eh_frame)
*(SORT_BY_ALIGNMENT(SORT_BY_NAME(.rodata*)))
This means the start of rodata will no longer be 16 bytes aligned.
However, the boundary between text and rodata/eh_frame is still aligned to
16 bytes, which is what I think the real purpose of the ALIGN call is.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
We load the secondary stage u-boot image from NAND to
system memory by nand_load, but we did not flush d-cache
to memory, nor invalidate i-cache before we jump to RAM.
When the system has cache enabled and the TLB/page attribute
of system memory is cacheable, it will cause issues.
- 83xx family is using the d-cache lock, so all of d-cache
access is cache-inhibited. so you can't see the issue.
- 85xx family is using d-cache, i-cache enable, partial
cache lock. you will see the issue.
This patch fixes the cache issue.
Signed-off-by: Dave Liu <daveliu@freescale.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
Note that with older board revisions, NAND boot may only work after a
power-on reset, and not after a warm reset. I don't have a newer board
to test on; if you have a board with a 33MHz crystal, please let me know
if it works after a warm reset.
Signed-off-by: Scott Wood <scottwood@freescale.com>
Wolfgang Denk
Trent Piepho
Dave Liu
Scott Wood