Bug 108083 – [GTK] Building fails in an armv5tel board

• Component: New Bugs
• Status: RESOLVED FIXED

[Adrian Perez]

[GTK] Building fails in an armv5tel board

[Adrian Perez]

More precisely, the “configure” script claims that the C++ compiler
cannot be found, examining “configure.log” this can be observed:

configure:5761: g++ -c -march=pentium4 -msse2 -mfpmath=sse   conftest.cpp >&5
g++: error: unrecognized argument in option '-march=pentium4'
g++: note: valid arguments to '-march=' are: armv2 armv2a armv3 armv3m armv4 armv4t armv5 armv5e armv5t armv5te armv6 armv6-m armv6j armv6k armv6s-m armv6t2 armv6z armv6zk armv7 armv7-a armv7-m armv7-r armv7e-m ep9312 iwmmxt iwmmxt2 native
g++: error: unrecognized command line option '-msse2'
g++: error: unrecognized command line option '-mfpmath=sse'

Of course, passing those options does not make sense in ARM.

[Adrian Perez]

For reference, the board is a Dreamplug, which reports the following:

% cat /proc/cpuinfo
processor       : 0
model name      : Feroceon 88FR131 rev 1 (v5l)
BogoMIPS        : 1191.11
Features        : swp half thumb fastmult edsp
CPU implementer : 0x56
CPU architecture: 5TE
CPU variant     : 0x2
CPU part        : 0x131
CPU revision    : 1

Hardware        : Marvell Kirkwood (Flattened Device Tree)
Revision        : 0000
Serial          : 0000000000000000

% uname -m
armv5tel

[Adrian Perez]

Created attachment 184983 [details]
Patch

[Daniel Bates]

Comment on attachment 184983 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=184983&action=review

> Tools/Scripts/webkitdirs.pm:1323
> +    return $Config{archname} =~ /^arm(-|v[567])/;

Would it be sufficient to match a string that begins with "arm-" or "armv"? In particular, would it be sufficient to match a string that begins with "armv" as opposed to {armv5, armv6, armv7}?

[Adrian Perez]

(In reply to comment #4)
> (From update of attachment 184983 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=184983&action=review
> 
> > Tools/Scripts/webkitdirs.pm:1323
> > +    return $Config{archname} =~ /^arm(-|v[567])/;
> 
> Would it be sufficient to match a string that begins with "arm-" or "armv"? In particular, would it be sufficient to match a string that begins with "armv" as opposed to {armv5, armv6, armv7}?

After thinking a bit more about this, I think you are right: it may
be better to go for the broader /^arm[v-]/ regexp here, so it won't
be needed in the future to edit this if some new ARM incarnation
happens to be reported e.g. as “armv8” (or whatever the version is).

[Adrian Perez]

Created attachment 185226 [details]
Patch

[Adrian Perez]

Could the second version of the patch (which checks for "^arm[v-]")
be landed? Probably that's the best we can to get builds being done
in the broadest amount of ARM machines possible.

[WebKit Review Bot]

Comment on attachment 185226 [details]
Patch

Clearing flags on attachment: 185226

Committed r141683: <http://trac.webkit.org/changeset/141683>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.