<div dir="ltr"><div>Only replying to let you know that your messages _are_ getting to the list ... Sorry, I don't know the answer.<br><br></div>Jon<br></div><div class="gmail_extra"><br><div class="gmail_quote">On 19 June 2015 at 09:51, Udo Grabowski (IMK) <span dir="ltr"><<a href="mailto:udo.grabowski@kit.edu" target="_blank">udo.grabowski@kit.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi again,<br>
<br>
(I really hate it to reply to my own posts, but there seems<br>
to be no one else left willing to reply...)<br>
<br>
Amojng the numerous different older and newer pages on how<br>
to compile illumos or OI with or without gcc or Sun cc and<br>
with this or that version I found this page today where<br>
SunStudio 12.3 is mentioned as the compiler now to be<br>
used (when stuck with Sun cc):<br>
<<a href="http://wiki.openindiana.org/oi/Building+with+oi-build" rel="noreferrer" target="_blank">http://wiki.openindiana.org/oi/Building+with+oi-build</a>><br>
<br>
Can anybody confirm that this is the version oi_151a9 was<br>
built with ? Or can anybody tell me if there's a trace of<br>
the compiler version left in any of the binaries on oi_151a9 ?<div class="HOEnZb"><div class="h5"><br>
<br>
On 18/06/2015 14:30, Udo Grabowski (IMK) wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
What I see now when disassembling the two products (e.g.,<br>
/kernel/drv/adm64/mpt_sas), there are subtle style differences<br>
in code arrangement:<br>
<br>
mptsas_iport_attach()<br>
mptsas_iport_attach: 55 pushq %rbp<br>
mptsas_iport_attach+0x1: 48 8b ec movq %rsp,%rbp<br>
mptsas_iport_attach+0x4: 48 83 ec 10 ** subq $0x10,%rsp<br>
mptsas_iport_attach+0x8: 48 89 7d f8 movq %rdi,-0x8(%rbp)<br>
mptsas_iport_attach+0xc: 48 89 75 f0 movq %rsi,-0x10(%rbp)<br>
<br>
mptsas_iport_attach()<br>
mptsas_iport_attach: 55 pushq %rbp<br>
mptsas_iport_attach+0x1: 48 8b ec movq %rsp,%rbp<br>
mptsas_iport_attach+0x4: 48 89 7c 24 f8 movq %rdi,-0x8(%rsp)<br>
mptsas_iport_attach+0x9: 48 89 74 24 f0 movq %rsi,-0x10(%rsp)<br>
mptsas_iport_attach+0xe: 48 83 ec 10 ** subq $0x10,%rsp<br>
<br>
From this I conclude that the oi_151a9 /dev/ official product is<br>
produced with a different Sun CC version than the one mentioned in<br>
the illumos build instructions. I have<br>
<br>
cc: Sun C 5.9 SunOS_i386 Patch 124868-10 2009/04/30<br>
<br>
reconstructed from a vanilla 12.0 compiler and the patches<br>
124868-10 126496-03 126498-02<br>
<br>
as well as the 12.1u1 lint checker<br>
lint: Sun C 5.10 SunOS_i386 Patch 142363-03 2009/12/03<br>
<br>
constructed from vanilla 12.1u1 with patches<br>
141858-03 142363-03<br>
<br>
Which one is the compiler oi_141a9 /dev is compiled with ?<br>
And, if different, is it approved to do The Right Thing ?<br>
<br>
On 18/06/2015 11:38, Udo Grabowski (IMK) wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi,<br>
<br>
I'm trying to reproduce the exact illumos kernel currenly used in<br>
/dev a9 (that is, a9 update 2 with the bash fixes from early this year,<br>
SUN CC compiled). I have the necessary compiler versions<br>
(reconstructing it was a fun archeological adventure into<br>
SolarisExpress paleolithicum...), and cloned this repository, which<br>
seems like the right one judging from the commit dates:<br>
<br>
<<a href="https://hg.openindiana.org/sustaining/oi_151a/illumos-gate/" rel="noreferrer" target="_blank">https://hg.openindiana.org/sustaining/oi_151a/illumos-gate/</a>><br>
<br>
illumos rev 52e13e00ba<br>
<br>
(bash fixes etc. are in sfw-gate/)<br>
<br>
After compiling (on exactly the same oi I want to reproduce) according<br>
to the instructions in<br>
<<a href="http://wiki.illumos.org/display/illumos/How+To+Build+illumos" rel="noreferrer" target="_blank">http://wiki.illumos.org/display/illumos/How+To+Build+illumos</a>>,<br>
setting LANG and LC_ALL to C to get around the locale difference,<br>
fixing the jdk7 problem (#4719), setting __SUNC="", switching<br>
off DEBUG build, NIGHTLY_OPTIONS='-nClrpt', I was lucky to see<br>
that mpt and marvell88sx were exactly reproduced (checked md5sum):<br>
<br>
size<br>
387160 2012-10-08 14:34 /kernel/drv/amd64/mpt*<br>
95112 2012-10-08 14:34 /kernel/drv/amd64/marvell88sx*<br>
<br>
But a short inspection of the other drivers quickly told me that I<br>
must have somehow pulled the wrong gate (note the date difference, only<br>
the 2012 entries and the small mouse8042 are reproduced, not the new ones):<br>
<br>
.....<br>
<br>
Where is the real gate used for the current kernel in oi 151a9 /dev ?<br>
Or is this another difference somewhere ?<br>
<br>
</blockquote>
<br>
<br>
</blockquote>
<br>
<br>
-- <br>
Dr.Udo Grabowski Inst.f.Meteorology & Climate Research IMK-ASF-SAT<br>
<a href="http://www.imk-asf.kit.edu/english/sat.php" rel="noreferrer" target="_blank">http://www.imk-asf.kit.edu/english/sat.php</a><br>
KIT - Karlsruhe Institute of Technology <a href="http://www.kit.edu" rel="noreferrer" target="_blank">http://www.kit.edu</a><br>
Postfach 3640,76021 Karlsruhe,Germany T:<a href="tel:%28%2B49%29721%20608-26026" value="+4972160826026" target="_blank">(+49)721 608-26026</a> F:-926026<br>
<br>
</div></div><br>_______________________________________________<br>
oi-dev mailing list<br>
<a href="mailto:oi-dev@openindiana.org">oi-dev@openindiana.org</a><br>
<a href="http://openindiana.org/mailman/listinfo/oi-dev" rel="noreferrer" target="_blank">http://openindiana.org/mailman/listinfo/oi-dev</a><br></blockquote></div><br></div>