Re: [BUG,2.6.28,s390] Fails to boot in Hercules S/390 emulator -hang traced

[john stultz]

On Tue, 2009-03-17 at 19:26 -0700, john stultz wrote:
> So two issues that I've found so far:
> 1) do_div() isn't dividing correctly in this case. Not sure why.

Patch to prove this is part of the issue below. Disables the s390
specific __div64_32() implementation and uses the generic code.

Frans: Mind giving this a try to verify the issue goes away with this?

Martin: I'm not sure if the problem here is the __div64_32
implementation or if the emulator is doing something wrong here. Might
need some help from you in sorting this out.



> 2) We *really* shouldn't be using oneshot mode while the jiffies
> clocksource is in use. I believe x86 avoids this specifically, so I'm
> not sure why s390 isn't doing the same. Still have to investigate.

Still working this.

thanks
-john


diff --git a/arch/s390/lib/div64.c b/arch/s390/lib/div64.c
index a5f8300..6c249fa 100644
--- a/arch/s390/lib/div64.c
+++ b/arch/s390/lib/div64.c
@@ -68,7 +68,7 @@ static uint32_t __div64_31(uint64_t *n, uint32_t base)
 	words[1] = reg3;
 	return reg2;
 }
-
+#if 0
 /*
  * Function to divide an unsigned 64 bit integer by an unsigned
  * 32 bit integer using the unsigned 64/31 bit division.
@@ -123,9 +123,9 @@ uint32_t __div64_32(uint64_t *n, uint32_t base)
 	}
 	return r;
 }
-
+#endif
 #else /* MARCH_G5 */
-
+#if 0
 uint32_t __div64_32(uint64_t *n, uint32_t base)
 {
 	register uint32_t reg2 asm("2");
@@ -145,5 +145,5 @@ uint32_t __div64_32(uint64_t *n, uint32_t base)
 	words[1] = reg3;
 	return reg2;
 }
-
+#endif
 #endif /* MARCH_G5 */


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/