On Wed, Aug 2, 2023 at 06:38 AM, Bruce Ashfield wrote:<br />
<blockquote>On Wed, Aug 2, 2023 at 9:15 AM Joel Stanley &lt;joel@jms.id.au&gt; wrote:<br />
<blockquote>On Wed, 2 Aug 2023 at 12:45, Fr&eacute;d&eacute;ric Martinsons<br />&lt;frederic.martinsons@gmail.com&gt; wrote:<br />
<blockquote><br /><br />On Wed, 2 Aug 2023 at 13:47, Joel Stanley &lt;joel@jms.id.au&gt; wrote:<br />
<blockquote>The one liner tests using `[ ] &amp;&amp; action` need to have || true after the action,<br />otherwise the line returns false and the recipie fails.<br /><br />Signed-off-by: Joel Stanley &lt;joel@jms.id.au&gt;<br />---<br />meta/classes-recipe/kernel.bbclass | 4 ++--<br />1 file changed, 2 insertions(+), 2 deletions(-)<br /><br />diff --git a/meta/classes-recipe/kernel.bbclass b/meta/classes-recipe/kernel.bbclass<br />index 247ef4a48aa7..f1bc41e82be8 100644<br />--- a/meta/classes-recipe/kernel.bbclass<br />+++ b/meta/classes-recipe/kernel.bbclass<br />@@ -486,7 +486,7 @@ kernel_do_install() {<br />install -m 0644 System.map ${D}/${KERNEL_IMAGEDEST}/System.map-${KERNEL_VERSION}<br />install -m 0644 .config ${D}/${KERNEL_IMAGEDEST}/config-${KERNEL_VERSION}<br />install -m 0644 vmlinux ${D}/${KERNEL_IMAGEDEST}/vmlinux-${KERNEL_VERSION}<br />- [ -e Module.symvers ] &amp;&amp; install -m 0644 Module.symvers ${D}/${KERNEL_IMAGEDEST}/Module.symvers-${KERNEL_VERSION}<br />+ [ -e Module.symvers ] &amp;&amp; install -m 0644 Module.symvers ${D}/${KERNEL_IMAGEDEST}/Module.symvers-${KERNEL_VERSION} || true<br />}</blockquote>
By doing that, you simply silent possible error of the action after the test, below is a little snippet showing that</blockquote>
Sure. But the test is looking for the file, so if it doesn't exist<br />then the command install won't run.</blockquote>
Something doesn't seem quite right, that test for Module.symvers (there<br />and in the shared workdir code) were put in place for the scenario<br />you are mentioning .. if Modules are disabled and hence Module.symvers doesn't<br />exist.<br /><br />At one point, we even tested that sort of configuration.<br /><br />I didn't dig more through the logs, but maybe we had lines after that command<br />previously, so the last return code wasn't a failure and hence no task<br />failure.</blockquote>
Yep, this is what happened, there used to be these 2 lines after it:<br /><br />
<div id="left-content-487" class="gr-diff contentText" data-side="left"><span class="gr-diff tab" style="tab-size: 4; -moz-tab-size: 4;"> </span>install -d ${D}${sysconfdir}/modprobe.d</div>
<table id="diffTable" class="responsive selected-left">
<tbody class="gr-diff section delta total">
<tr class="gr-diff diff-row side-by-side" tabindex="-1" aria-labelledby="left-button-486 left-content-486">
<td class="gr-diff content left no-intraline-info remove">
<div id="left-content-486" class="gr-diff contentText" data-side="left">install -d ${D}${sysconfdir}/modules-load.d</div>
</td>
</tr>
</tbody>
</table>
<p><br />https://gerrit.openbmc.org/c/openbmc/openbmc/+/65313/3/poky/meta/classes-recipe/kernel.bbclass</p>
<br />Once these were removed we started hitting this failure over in OpenBMC.