<html><head><meta http-equiv="Content-Type" content="text/html; charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class=""><div><blockquote type="cite" class=""><div class="">On 4 Feb 2019, at 23:20, Sylvestre Ledru <<a href="mailto:sylvestre@debian.org" class="">sylvestre@debian.org</a>> wrote:</div><br class="Apple-interchange-newline"><div class="">
<meta http-equiv="Content-Type" content="text/html;
charset=windows-1252" class="">
<div class=""><p class="">Hello</p><p class="">Could you please provide a simple test case with the command line
used?</p><p class="">And do you get an error ?</p><p class="">Thanks,<br class="">
</p><p class="">S</p><p class=""><br class=""></p></div></div></blockquote><div>Hello Sylvestre</div><div><br class=""></div>Unfortunately thats not so easy as the stuff is done in ./configure tests.<div class="">Here is a text file where I compared the attmept to get gnustep-base compiled under debian9 (stretch). Something which used to work fine but now seems broken due to clang7 or due to clang8 (or due to libobjc2 or due to gnustep-make or gnustep-base)</div><div class=""><br class=""></div><div class="">So what I just did was install a minimal VM with clang7 and another VM with clang8 and did the exact same steps on both machines to get a working gnustep base environment.</div><div class="">Both fail at the end but due to different reasons which seem to be related to clang / libobjc2 runtime interworking with each other somehow.</div><div class=""><br class=""></div><div class="">The result is however a test case which produces a linking error at the end which I think is the problem for clang7.</div><div class="">For clang8 it looks more like some internal libobjc2 initialisation [NSObject init] goes into a recursive exception.</div><div class=""><br class=""></div><div class=""><div class=""><br class=""></div><div class=""></div></div></div></body></html>