On Wed, Dec 14, 2022 at 11:58:21AM +0100, Ahmad Fatoum wrote:
> Functions were kept naked because they modify the stack pointer.
> We can hoist that up and change the board-provided entry point
> to be non-naked to be more robust in face of unexpected compiler
> optimizations.
>
> Signed-off-by: A
On Wed, Dec 14, 2022 at 11:50:50AM +0100, Ahmad Fatoum wrote:
> eMMC Spec v5.1 (JEDEC Standard No. 84-B51) is quite clear that:
>
> Due to legacy reasons, a Device may still treat CMD24/25 during
> prg-state (while busy is active) as a legal or illegal command.
> A host should not send CMD24
On Fri, Dec 09, 2022 at 08:10:01AM +0100, Ahmad Fatoum wrote:
> barebox doesn't use interrupts, so it need not differentiate between a
> possibly sleeping prepare/unprepare and an atomic enable/disable.
>
> Provide a wrapper that just expands to the already existing functions to
> simplify porting
On Tue, Dec 06, 2022 at 12:26:14PM +0100, Marco Felsch wrote:
> The LZO compression hasn't been updated since 2017 and embedded build
> tools like Yocto dropping the support for it [1]. So switch to LZ4 as
> default since this seems to be the most comparable compression standard
> according Ahmads
Hi,
we are using the AM3352 processor in combination with a single 512MB Micron
DDR3 chip running the barebox bootloader in our products for several years now.
A minor amount of boards (around 5 of 2000) fail the production test, because
they do not boot properly after warm reset. In such case
Hi Sascha,
On 22-12-14, Sascha Hauer wrote:
> Several barebox struct names have a '_d' suffix. The suffix was
> originally added to struct device_d in case we may want to import
> Linux struct device as a separate struct. That never happened, instead
> barebox struct device_d basically is the same
