<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote"><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"><br>
<br>
1. GCC trunk is less stable than LLVM because the lack of general buildbots.<br></blockquote><div><br></div><div>GCC has plenty of buildbots, it has no revert-on-breakage policy.</div><div> </div><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">
* Testing a new patch means comparing the test results (including<br>
breakages) against the previous commit, and check the differences.<br>
This is a poor definition of "pass", especially when the number of<br>
failures is large</blockquote><div><br></div><div>This is an artifact of the lack of a revert-on-breakage policy.</div><div> </div><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">.<br>
* On ARM and AArch64, the number of failures is around a couple of<br>
thousand (don't know the exact figure). AFAIK, these are not marked<br>
XFAIL in any way, but are known to be broken for one reason or<br>
another.<br></blockquote><div><br></div><div>That sounds like a failure on the part of the ARM developers.</div><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"><br>
* Linaro monthly releases go out with those failures, and the fact<br>
that they keep on going means the FSF releases do, too.</blockquote><div><br></div><div>I expect this would change if someone pushed.</div><div><br></div><div>Here is, for example, the failure list for i686-pc-linux-gnu for each 4.9 release:<br></div><div><a href="https://gcc.gnu.org/gcc-4.9/buildstat.html">https://gcc.gnu.org/gcc-4.9/buildstat.html</a> <br></div><div><br></div><div><br></div><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"> This is a huge<br>
cost on the release process, since it needs complicated diff programs<br>
and often incur in manual analysis.<br></blockquote><div><br></div><div><br></div><div>You say all this as if it is a GCC testsuite issue.</div><div><br></div><div>It sounds completely like a process issue that hasn't been raised and dealt with.</div><div><br></div><div>IE something that could easily happen to LLVM.</div><div><br></div><div><br></div></div></div></div>