2008-10-28 23:43:31

by Gary Hade

[permalink] [raw]
Subject: x86_64: remove debug code from arch_add_memory()


Gets rid of dmesg spam created during physical memory hot-add which
will very likely confuse users. The change removes what appears to
be debugging code which I assume was unintentionally included in:
x86: arch/x86/mm/init_64.c printk fixes
commit 10f22dde556d1ed41d55355d1fb8ad495f9810c8

Signed-off-by: Gary Hade <[email protected]>

---
arch/x86/mm/init_64.c | 1 -
1 file changed, 1 deletion(-)

Index: linux-2.6.28-rc2/arch/x86/mm/init_64.c
===================================================================
--- linux-2.6.28-rc2.orig/arch/x86/mm/init_64.c 2008-10-28 15:39:05.000000000 -0700
+++ linux-2.6.28-rc2/arch/x86/mm/init_64.c 2008-10-28 15:39:32.000000000 -0700
@@ -836,7 +836,6 @@
max_pfn_mapped = last_mapped_pfn;

ret = __add_pages(zone, start_pfn, nr_pages);
- WARN_ON(1);

return ret;
}


2008-10-29 08:31:24

by Ingo Molnar

[permalink] [raw]
Subject: Re: x86_64: remove debug code from arch_add_memory()


* Gary Hade <[email protected]> wrote:

>
> Gets rid of dmesg spam created during physical memory hot-add which
> will very likely confuse users. The change removes what appears to
> be debugging code which I assume was unintentionally included in:
> x86: arch/x86/mm/init_64.c printk fixes
> commit 10f22dde556d1ed41d55355d1fb8ad495f9810c8
>
> Signed-off-by: Gary Hade <[email protected]>
>
> ---
> arch/x86/mm/init_64.c | 1 -
> 1 file changed, 1 deletion(-)

applied to tip/x86/urgent, thanks Gary!

Note, i changed it slightly: instead of removing it completely i
changed it to WARN_ON_ONCE(ret), to show us when __add_pages() fails.
(which is what the original intention was there) This does not trigger
in your tests, right?

Ingo

--------------->
>From fe8b868eccb9f85a0e231e35f0abac5b39bac801 Mon Sep 17 00:00:00 2001
From: Gary Hade <[email protected]>
Date: Tue, 28 Oct 2008 16:43:14 -0700
Subject: [PATCH] x86: remove debug code from arch_add_memory()

Impact: remove incorrect WARN_ON(1)

Gets rid of dmesg spam created during physical memory hot-add which
will very likely confuse users. The change removes what appears to
be debugging code which I assume was unintentionally included in:

x86: arch/x86/mm/init_64.c printk fixes
commit 10f22dde556d1ed41d55355d1fb8ad495f9810c8

Signed-off-by: Gary Hade <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
---
arch/x86/mm/init_64.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index ebe1811..9db01db 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -858,7 +858,7 @@ int arch_add_memory(int nid, u64 start, u64 size)
max_pfn_mapped = last_mapped_pfn;

ret = __add_pages(zone, start_pfn, nr_pages);
- WARN_ON(1);
+ WARN_ON_ONCE(ret);

return ret;
}

2008-10-29 16:43:35

by Gary Hade

[permalink] [raw]
Subject: Re: x86_64: remove debug code from arch_add_memory()

On Wed, Oct 29, 2008 at 09:30:53AM +0100, Ingo Molnar wrote:
>
> * Gary Hade <[email protected]> wrote:
>
> >
> > Gets rid of dmesg spam created during physical memory hot-add which
> > will very likely confuse users. The change removes what appears to
> > be debugging code which I assume was unintentionally included in:
> > x86: arch/x86/mm/init_64.c printk fixes
> > commit 10f22dde556d1ed41d55355d1fb8ad495f9810c8
> >
> > Signed-off-by: Gary Hade <[email protected]>
> >
> > ---
> > arch/x86/mm/init_64.c | 1 -
> > 1 file changed, 1 deletion(-)
>
> applied to tip/x86/urgent, thanks Gary!
>
> Note, i changed it slightly: instead of removing it completely i
> changed it to WARN_ON_ONCE(ret), to show us when __add_pages() fails.
> (which is what the original intention was there)

No problem! That's obviously better.

> This does not trigger in your tests, right?

Correct. I just did a memory hot-add with that change and the
WARN_ON_ONCE(ret) did not write anything to dmesg.

Thanks,
Gary

--
Gary Hade
System x Enablement
IBM Linux Technology Center
503-578-4503 IBM T/L: 775-4503
[email protected]
http://www.ibm.com/linux/ltc