Bug#598893: linux-2.6 [m68k]: please work around gcc-4.3 ICE bug (37052)
Dixi quod…
>>Please see to that -fno-omit-frame-pointer is used during compilation
>>of a Debian m68k kernel as long as it is built with gcc-4.3 (I think
>>defining CONFIG_FRAME_POINTER should be enough, I have yet to build
>>any kernel at all though and have not tested this).
>
>ARCH_WANT_FRAME_POINTERS might be needed as well.
This part¹ of a debdiff did the trick for me:
diff -u linux-2.6-2.6.32/debian/config/m68k/config linux-2.6-2.6.32/debian/config/m68k/config
--- linux-2.6-2.6.32/debian/config/m68k/config
+++ linux-2.6-2.6.32/debian/config/m68k/config
@@ -312,0 +313,3 @@
+# Debian #598893
+CONFIG_ARCH_WANT_FRAME_POINTERS=y
+CONFIG_FRAME_POINTER=y
Corresponding changelog entry:
[ Thorsten Glaser ]
* debian/config/m68k/config: define CONFIG_ARCH_WANT_FRAME_POINTERS
and CONFIG_FRAME_POINTER to yes (Closes: #598893)
Note that this *only* applies to the sid kernel, as the experimental
kernel uses gcc-4.4 where I did not run into this problem last time.
(The current¹ problem however does apply to both sid and experimental.)
① as mentioned, I’m running into other issues; patches are being drafted
and submitted, however, even emulated, m68k is slow so bear with me…
bye,
//mirabilos
--
Solange man keine schmutzigen Tricks macht, und ich meine *wirklich*
schmutzige Tricks, wie bei einer doppelt verketteten Liste beide
Pointer XORen und in nur einem Word speichern, funktioniert Boehm ganz
hervorragend. -- Andreas Bogk über boehm-gc in d.a.s.r
Reply to: