Re: [Qemu-devel] [PATCH 06/10] r2d: Flash memory creation is confused about size, mark FIXME

[Markus Armbruster]

Philippe Mathieu-Daudé <address@hidden> writes:

> On 3/5/19 6:25 PM, Peter Maydell wrote:
>> On Tue, 5 Mar 2019 at 17:21, Philippe Mathieu-Daudé <address@hidden> wrote:
>>> But I'd recommend changing/fixing the sector size during the next dev
>>> cycle, so we have more time for testing.
>> 
>> Nobody in the upstream dev community is using or testing this board.
>
> Well I submitted a Avocado test last year:
> https://lists.gnu.org/archive/html/qemu-devel/2018-10/msg02749.html
>
> And I rebase/run it from time to time.
>
>> The only way we'll find out if there's a problem with changing the
>> sector size is to put the change in and get it into a release.
>> I would vote for just making the change now.
>
> I'm happy with this vote, and am sure Markus will be too :)
>
> Markus: the last field I wasn't sure about without double checking the
> code is the @width one. I find it misleading, is that the size of the
> data bus or the size of the flash words? Answer: this is the size of the
> words in byte. NOR flash devices can not write less data than their word
> boundary.
>
> The S29PL127J60TFI130 only support 16bit words, so using @width=2 is
> correct.
>
> And the winner is.... ta-da!
>
>     pflash_cfi02_register(0x0, NULL, "r2d.flash", FLASH_SIZE,
>                           dinfo ? blk_by_legacy_dinfo(dinfo) : NULL,
> -                         16 * KiB, FLASH_SIZE >> 16,
> -                         1, 4, 0x0000, 0x0000, 0x0000, 0x0000,
> +                         64 * KiB, FLASH_SIZE >> 16 /* will get removed
> later */,
> +                         1, 2, 0x0001, 0x227e, 0x2220, 0x2200
>                           0x555, 0x2aa, 0);

Sold!

[...]