[Libwebsockets] Openssl is too old to support lws_tls_vhost_cert_info

Kyle Dias kyledias at vurbox.com
Fri Mar 4 06:35:35 CET 2022


Hi Andy,

Thank you so much. I saw that error way early on but didn't think
anything of it :(, I'll take it as a learning experience. Anyway I can
donate to the Libwebsockets project to show my appreciation?

Thanks,
Kyle

On Thu, Mar 3, 2022 at 11:48 PM Andy Green <andy at warmcat.com> wrote:

>
>
> On 3/4/22 04:36, Kyle Dias wrote:
>
> > Yes I had but I didn't see any direct error: here is the output:
>
> ...
>
> >     -Wignored-qualifiers -DGCC_VER="\"\"" *-fPIC* -ffunction-sections
>
> So you're building lws with -fPIC which is good if you want to end up
> with a .so.
>
> >
>  /usr/lib/gcc-cross/arm-linux-gnueabihf/9/../../../../arm-linux-gnueabihf/bin/ld:
> >
>  /media/diask5/NVMe/Cross_Builds/cross-root-*generic32/usr/local/lib/libcrypto.a(hmac.o):
> >     relocation R_ARM_THM_MOVW_ABS_NC against `a local symbol' can not be
> >     used when making a shared object; recompile with -fPIC*
> >
> >
> > Also I did build openssl as *static* if that is an issue?
>
> If you want to be able relocate lws (as a .so, you don't know where the
> loader will put it in memory) then you have to be able to relocate all
> the pieces it links to, too.
>
> So it is telling you to rebuild openssl with -fPIC and this problem will
> go away.  I don't think it matters for this if static (it means a copy
> of the object will get composed into lws), or shared (it means lws will
> bind to it at load-time) so long as the code in there is built to be
> relocatable.
>
> -Andy
>
> > Not sure exactly what this means and couldn't find anything online if
> > you know?:
> >
> >     * relocation R_ARM_THM_MOVW_ABS_NC against `a local symbol' can not
> >     be used when making a shared object; recompile with -fPIC*
> >
> >
> > It is also mainly these relocation errors: evp_lib.c:(.text+0x29c):
> > dangerous relocation: unsupported relocation
> >
> > -Kyle
> >
> > On Thu, Mar 3, 2022 at 11:10 PM Andy Green <andy at warmcat.com
> > <mailto:andy at warmcat.com>> wrote:
> >
> >
> >
> >     On 3/4/22 04:06, Kyle Dias wrote:
> >
> >      > I have tried everything I could think of. I don't know why I
> cannot
> >
> >     ... just do what I suggested last time which should help you
> understand
> >     why cmake feels like that.  It's more likely to guide you in the
> right
> >     direction than jamming down all the buttons until you are burned
> >     out. --->
> >
> >      >>     Delete build/CMakeFiles/CMakeError.log, rebuild and look in
> >     that file
> >      >>     for what happened when cmake tried to test for, eg,
> >     HMAC_CTX_new.
> >
> >     -Andy
> >
> >
> >
> > --
> > Kyle Dias
> > CEO & Founder
> > kyledias at vurbox.com <mailto:kyledias at vurbox.com>
> >
> >
> >
> > VurBox Inc.
> > Suite 201, 1300 Cornwall Rd. Oakville, ON | L6J 7W5
>


-- 
Kyle Dias
CEO & Founder
kyledias at vurbox.com



VurBox Inc.
Suite 201, 1300 Cornwall Rd. Oakville, ON | L6J 7W5
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://libwebsockets.org/pipermail/libwebsockets/attachments/20220304/43bf1ab2/attachment-0001.htm>


More information about the Libwebsockets mailing list