<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Sat, Apr 9, 2016 at 7:41 PM, Philip Balister <span dir="ltr"><<a href="mailto:philip@balister.org" target="_blank">philip@balister.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div class=""><div class="h5">> This might be dangerous. Even in recent version of DD (coreutils 8.25), it<br>
> might not detect in source file mapped/unmapped blocks properly,<br>
> (just use is_nul() function to check if block contains only zeros, instead<br>
> of e.g. fiemap).<br>
> Result might be so, that in output file block which must be zeros is<br>
> actually become unmapped. And as consequence, if it might not be written to<br>
> destination drive.<br>
<br>
</div></div>Does it work with bmaptool?<br></blockquote></div><div><br></div><div>bmaptool can work with practically any sparse file (even if bmap file are not provided) to flash it efficiently on raw device.</div><div>However, with such cases like "dd conv=sparse" it should be used with extra caution, as some of needed zeroed blocks might not be written to destination device.</div><div><br></div><div>Yocto bug #9099 now has two patches attached that adding initial bmap-tools support to OE-core.</div><div><div>Those patches allows to create bmap files for simpler cases, like .hddimg as well as shipping standalone version of bmaptool in deploy/tools/*</div></div><div>It probably would be content of next release, as I don't think anyone would be merging them now.<br></div><div><br></div><div>For assembling images with wic, it would require a bit more work.</div><div><br></div>-- <br><div class="gmail_signature">br, Alexander Kanevskiy</div>
</div></div>