<div dir="ltr">Hi Petr,<div><br></div><div>Thank you for reverting the patches.</div><div><br></div><div>I'm trying to replicate the issue but so far I haven't had any luck, and just by looking at the code I'm not seeing where/what is could be breaking.</div><div><br></div><div>Regarding the upstream patch, that one is just an enabler (read: it only extends the "mtd_add_partition" function but it does not add any code to force the access mode on sub-partitions).</div><div>The reason for this is because this enforcement is done on the mtd parser code that is OpenWrt specific and implemented via kernel patches (not present on upstream).</div><div><br></div><div>Best regards,</div><div>Bruno Pena</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Jan 21, 2020 at 7:57 PM Petr Štetiar <<a href="mailto:ynezz@true.cz">ynezz@true.cz</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Bruno Pena <<a href="mailto:brunompena@gmail.com" target="_blank">brunompena@gmail.com</a>> [2020-01-21 14:53:54]:<br>
<br>
Hi,<br>
<br>
> Based on the last comment from Steve (fstools patch was not reverted), I'm<br>
> not sure if that's the root cause.<br>
<br>
you need to find out, but that Daniel's remark seems legit to me, your patch<br>
likely changed the mtd device open order/flags.<br>
<br>
> The kernel patch (which when reverted makes rootfs_data writable again)<br>
> only enforces the parent mtd access mode on the sub-partitions.<br>
<br>
FYI I currently dont have time to fix that regression myself and since its<br>
likely affecting a lot of users, so I've reverted those changes. I think, that<br>
this change is useful, so I'm still willing to merge it once fixed, no<br>
worries. Having some upstream feedback beforehand would be a plus.<br>
<br>
BTW it would be fair to inform upstream about this possible issue as well, so<br>
the patch wont get merged in its current state, unless its double checked (or<br>
just write them, that you're planning v2, so the patch is removed from their<br>
Patchwork).<br>
<br>
-- ynezz<br>
</blockquote></div>