2.6.36 regression : swap support broken

* 2.6.36 regression : swap support broken
@ 2010-12-12 18:38 Guy Martin
  2010-12-12 19:21 ` John David Anglin
  2010-12-23  2:19 ` John David Anglin
  0 siblings, 2 replies; 10+ messages in thread
From: Guy Martin @ 2010-12-12 18:38 UTC (permalink / raw)
  To: linux-parisc; +Cc: eike-kernel

Hi all,

Recently, swap got broken. When it has to be used, the kernel issue the
following line numerous times :
swap_free: Bad swap offset entry 00100009

It eventually panic with useless backtrace. Note that the offset entry
is always the same number.

This can be reproduced easily by booting with mem=64m and start
compiling something like the kernel.

I've been bisecting this and I found out it has been introduced in
4e60c86bd9e5a7110ed28874d0b6592186550ae8 and to be precise, the
following change in include/linux/mmdebug.h :
-#define VM_BUG_ON(cond) do { } while (0)
+#define VM_BUG_ON(cond) do { (void)(cond); } while (0)

While the above change looks harmless, it introduces a slight change in
mm/swap_state.c:154 in add_to_swap() : VM_BUG_ON(!PageUptodate(page));
If you comment out that line, the problem is gone.

PageUptodate() will call smp_rmb() being the memory barrier instruction.

Those tests were done with linux-2.6.36 and gcc-4.4.4.
Having the feeling that this might have been a gcc bug, I recompiled
the whole kernel with gcc-4.2.4 and the problem was gone.

Looking at the assembly output for add_to_swap(), there is a slight
difference.

- gcc-4.4.4 output :

add_to_swap:
        .PROC
        .CALLINFO FRAME=64,CALLS,SAVE_RP,ENTRY_GR=4
        .ENTRY
        stw %r2,-20(%r30)
        stw,ma %r4,64(%r30)
        stw %r3,-60(%r30)
        ldw 0(%r26),%r28
        copy %r26,%r3
        ldw 0(%r26),%r28
        bb,>=,n %r28,28,.L61
.L61:   
        b,l get_swap_page,%r2
        ...

- gcc-4.2.4 output :

add_to_swap:
        .PROC
        .CALLINFO FRAME=64,CALLS,SAVE_RP,ENTRY_GR=4
        .ENTRY
        stw %r2,-20(%r30)
        stw,ma %r4,64(%r30)
        stw %r3,-60(%r30)
        ldw 0(%r26),%r19
        copy %r26,%r3
        ldw 0(%r26),%r28
        extrw,s,>= %r28,28,1,%r0
.L76:   
        b,l get_swap_page,%r2
        ...

As you can see, gcc-4.2 uses extrw while gcc-4.4 uses bb.

What's the next step in order to troubleshoot this ?
Is this really a gcc bug or is the problem elsewhere ?

HTH,
  Guy

^ permalink raw reply	[flat|nested] 10+ messages in thread