View Single Post
Old 09-21-2012, 03:43 PM   #688
NiLuJe
BLAM!
NiLuJe ought to be getting tired of karma fortunes by now.NiLuJe ought to be getting tired of karma fortunes by now.NiLuJe ought to be getting tired of karma fortunes by now.NiLuJe ought to be getting tired of karma fortunes by now.NiLuJe ought to be getting tired of karma fortunes by now.NiLuJe ought to be getting tired of karma fortunes by now.NiLuJe ought to be getting tired of karma fortunes by now.NiLuJe ought to be getting tired of karma fortunes by now.NiLuJe ought to be getting tired of karma fortunes by now.NiLuJe ought to be getting tired of karma fortunes by now.NiLuJe ought to be getting tired of karma fortunes by now.
 
NiLuJe's Avatar
 
Posts: 13,477
Karma: 26012492
Join Date: Jun 2010
Location: Paris, France
Device: Kindle 2i, 3g, 4, 5w, PW, PW2, PW5; Kobo H2O, Forma, Elipsa, Sage, C2E
@Kai771: 2008q1-126 is a known 'bad' release, with a buggy GCC that throws ICE in a lot of known conditions .

I tweaked it somewhat more extensively than needed because I'm using a 'bleeding-edge' TC that uses a vastly more recent GCC & glibc combo than what's used on the Kindle (and than the TC used by the devs), so I *need* some of this stuff to actually avoid pulling incompatible stuff from those newer versions (most notably the glibc's fortify support, and gcc's stack smashing protection I mentioned in my previous message).

There's also a few changes related to building mupdf in release mode, not debug, and not breaking the build of stuff targeted at the host system because I have an ARM march in the default CFLAGS, which obviously my host's x86 GCC would throw a fit about .

I could be wrong, because I'm always putting my minimal self-built system in the searchpaths (via CPPFLAGS & LDFLAGS), but I'm not sure kpdf actually needs anything out of its tree (other than a proper glibc TC), so, I personally don't tweak/set SYSROOT, and don't use stuff from the Kindle's FS or the tarball alluded to.

All that said, take a look at your binaries built with MG 2012.03 via readelf (especially the .gnu.version* sections, via readelf -V for example). That should give you a good hint of what's wrong (shouldn't pull anything more recent than glibc 2.5 & gcc 4.1).

EDIT: Well, the runtime error you get when running it on the Kindle helps, too, but readelf is faster/more detailed .

Code:
Version needs section '.gnu.version_r' contains 6 entries:
 Addr: 0x000000000000a6fc  Offset: 0x0026fc  Link: 5 (.dynstr)
  000000: Version: 1  File: libgcc_s.so.1  Cnt: 2
  0x0010:   Name: GCC_3.0  Flags: none  Version: 9
  0x0020:   Name: GCC_3.5  Flags: none  Version: 7
  0x0030: Version: 1  File: libdl.so.2  Cnt: 1
  0x0040:   Name: GLIBC_2.4  Flags: none  Version: 6
  0x0050: Version: 1  File: libstdc++.so.6  Cnt: 2
  0x0060:   Name: CXXABI_1.3  Flags: none  Version: 8
  0x0070:   Name: GLIBCXX_3.4  Flags: none  Version: 5
  0x0080: Version: 1  File: libc.so.6  Cnt: 1
  0x0090:   Name: GLIBC_2.4  Flags: none  Version: 4
  0x00a0: Version: 1  File: libpthread.so.0  Cnt: 1
  0x00b0:   Name: GLIBC_2.4  Flags: none  Version: 3
  0x00c0: Version: 1  File: libm.so.6  Cnt: 1
  0x00d0:   Name: GLIBC_2.4  Flags: none  Version: 2

Last edited by NiLuJe; 09-21-2012 at 03:51 PM.
NiLuJe is offline   Reply With Quote