Several boards are getting bigger than their specified maximum binary
size of 256Kb. Enable the TLSF allocator for them which is smaller
than the original dlmalloc. The changed defconfigs are:
eukrea_cpuimx25_defconfig
eukrea_cpuimx35_defconfig
eukrea_cpuimx51_defconfig
usb_a9260_defconfig
usb_a9263_128mib_defconfig
usb_a9263_defconfig
usb_a9g20_128mib_defconfig
usb_a9g20_defconfig
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
To get symbols like __ashrdi3.
Franck started generating these symbols from gcc assembly, which seems
to be the cleaner approach. The simpler approach for now to get additional
symbols is to link in libgcc, which is the same as the kernel does.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Franck Jullien <franck.jullien@gmail.com>
The Linux Kernel defines only one of __LITTLE_ENDIAN and
__BIG_ENDIAN. Endianess can then be tested with #ifdef __xx_ENDIAN. Userspace
always defined both __LITTLE_ENDIAN and __BIG_ENDIAN and byteorder can then
be tested with #if __BYTE_ORDER == __xx_ENDIAN.
As we tend to use a lot of Kernel code in barebox we switch to use the kernel
way of determing the byte order.
As this always causes a lot of confusion add a check to include/common.h to
make sure only one of __LITTLE_ENDIAN and __BIG_ENDIAN is defined.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Straight forward port of Synopsys Designware ethernet
driver from u-boot v2012.04.01.
Signed-off-by: Johannes Stezenbach <js@sig21.net>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
resource_size_t can be 32bit or 64bit depending on the architecture.
Add a define for it to be able to printf a resource_size_t correctly
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
On x86_64 we need CONFIG_PHYS_ADDR_T_64BIT to make the resource sizes
64bit. The kernel has this as a Kconfig variable, but on barebox sandbox
will build with whatever compiler we find, so we can't put it into Kconfig.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
With files smaller than the tftp block size the whole transfer is done
in tftp_do_open already. In this case we are in STATE_DONE, but there
is no error. Set priv->err to 0 and check for it to be able to transfer
small files.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
We already have a phys_size_t since:
commit 2f05b69256
Author: Renaud Barbier <renaud.barbier@ge.com>
Date: Fri May 11 11:58:13 2012 +0100
linux/types.h: define phys_size_t
Add this definition in preparation for the introduction of the
mpc85xx support.
Signed-off-by: Renaud Barbier <renaud.barbier@ge.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Fix build due to redefinition of this type.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Passing a too large size or offset to erase could
affect flash outside the partition boundary.
Addresses for SPI flash wrap around, thus giving a
count + offset going past the end of the flash would
wrap around and erase flash at offset 0.
Add the same check for protect.
Signed-off-by: Johannes Stezenbach <js@sig21.net>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
"erase /dev/myflash0.mypart 0xf0000+0xf0000" could erase past
the end of the partition.
Signed-off-by: Johannes Stezenbach <js@sig21.net>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Show progressbar even when erasing just a single sector,
otherwise it looks as if erase didn't do anything.
Signed-off-by: Johannes Stezenbach <js@sig21.net>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
"erase /dev/myflash0 0+1" erased the whole flash,
similar for other value of count if you guessed the
erae block size wrong.
Signed-off-by: Johannes Stezenbach <js@sig21.net>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
When barebox headers get included from HOSTCC tools they should not
include other architecture specific headers as barebox might get
cross-compiled. Instead, the tool itself should include "compiler.h".
Signed-off-by: Dirk Hörner <dirker@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
The Elfnn_Section is not available on all platforms,
noteworthy are cygwin.
Use the safe replacement _Half.
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
In the commit 40492a0c13
(MIPS: add common header file for DEBUG_LL via NS16550)
introduced common DEBUG_LL via NS16550 for MIPS
(see file arch/mips/include/debug_ll_ns16550.h).
In the commit 1cbe2b2c00
(MIPS: XBurst: add Ritmix RZX-50 board support)
the file debug_ll_ns16550.h used in
the file arch/mips/mach-xburst/include/mach/debug_ll.h.
Usage looks like this:
------------------------------------------------
+#ifdef CONFIG_BOARD_RZX50
+#include <mach/debug_ll_jz4755.h>
+#endif
+
+#include <debug_ll_ns16550.h>
------------------------------------------------
So after adding another board (e.g. A320) we will have something like this:
------------------------------------------------
#ifdef CONFIG_BOARD_RZX50
#include <mach/debug_ll_jz4755.h>
#endif
+#ifdef CONFIG_BOARD_A320
+#include <mach/debug_ll_jz4740.h>
+#endif
#include <debug_ll_ns16550.h>
------------------------------------------------
This approach has disadvantage:
* the files mach/debug_ll_jz4740.h and mach/debug_ll_jz4755.h
(they go to arch/mips/arch-xburst) are __BOARD-SPECIFIC__
(not SOC- or mach-specific!); The file mach-xburst/include/mach/debug_ll.h
is outside board directory, but it contains some board related information.
This commit introduce a more suitable solution.
Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
This patch makes possible to put a board-specific
header file (e. g. foobar.h) to arch/mips/boards/*/include/board/.
Header file usage:
#include <board/foobar.h>
Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
The tftp filesystem support is by no means dependent on the old tftp
support, it depends on NET though.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
It does not necessarily means a bug when block_get fails here,
this can also be a failure on the underlying device.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
We must check pointer to function rather than result.
Signed-off-by: Alexander Shiyan <shc_work@mail.ru>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>