[PATCH] parisc: fix longstanding build failure on prefetch

[PATCH] parisc: fix longstanding build failure on prefetch

[Paul Gortmaker]

Upstream commit e66eed651fd18a961f11cda62f3b5286c8cc4f9f

    "list: remove prefetching from regular list iterators"

removed <linux/prefetch.h> from <linux/list.h> very early in the
v3.0 dev cycle (v2.6.39-2-ge66eed6), thereby causing build failures
when it exposed the implicit prefetch users in parisc.

Reported-by: David Rientjes <rientjes@google.com>
Reported-by: WANG Cong <xiyou.wangcong@gmail.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
---

[ The parisc defconfig has been failing to build everywhere since
  last May!  http://kisskb.ellerman.id.au/kisskb/target/2059/   And this
  one liner fix https://lkml.org/lkml/2011/10/15/121 was reported last
  Oct, and again http://www.spinics.net/lists/linux-parisc/msg03945.html
  in early Feb.  So, now with a clear reasoning in the commit header as
  to why it fails, can we finally get this in tree please? ]

diff --git a/drivers/parisc/iommu-helpers.h b/drivers/parisc/iommu-helpers.h
index a9c46cc..8c33491 100644
--- a/drivers/parisc/iommu-helpers.h
+++ b/drivers/parisc/iommu-helpers.h
@@ -1,3 +1,5 @@
+#include <linux/prefetch.h>
+
 /**
  * iommu_fill_pdir - Insert coalesced scatter/gather chunks into the I/O Pdir.
  * @ioc: The I/O Controller.
-- 
1.7.9.1

Re: [PATCH] parisc: fix longstanding build failure on prefetch

[James Bottomley]

On Sun, 2012-02-26 at 15:51 -0500, Paul Gortmaker wrote:
> Upstream commit e66eed651fd18a961f11cda62f3b5286c8cc4f9f
> 
>     "list: remove prefetching from regular list iterators"
> 
> removed <linux/prefetch.h> from <linux/list.h> very early in the
> v3.0 dev cycle (v2.6.39-2-ge66eed6), thereby causing build failures
> when it exposed the implicit prefetch users in parisc.

I already have this in my internal patch queue:

http://marc.info/?l=linux-parisc&m=132825467010662

James

Re: [PATCH] parisc: fix longstanding build failure on prefetch

[Paul Gortmaker]

On 12-02-27 10:24 AM, James Bottomley wrote:
> On Sun, 2012-02-26 at 15:51 -0500, Paul Gortmaker wrote:
>> Upstream commit e66eed651fd18a961f11cda62f3b5286c8cc4f9f
>>
>>     "list: remove prefetching from regular list iterators"
>>
>> removed <linux/prefetch.h> from <linux/list.h> very early in the
>> v3.0 dev cycle (v2.6.39-2-ge66eed6), thereby causing build failures
>> when it exposed the implicit prefetch users in parisc.
> 
> I already have this in my internal patch queue:
> 
> http://marc.info/?l=linux-parisc&m=132825467010662

Is this in a different queue than what is in linux next?

It would be nice to see such build fixes get into linux-next
ASAP, so folks doing multi-arch coverage builds on their work
can know if they've introduced a regression.  At the moment,
the parisc builds in linux-next have been worthless for over
six months, and I'm sure I'm not the 1st one to waste time
"re-solving" old bugs like this.

Thanks,
Paul.

> 
> James
> 
> 

Re: [PATCH] parisc: fix longstanding build failure on prefetch

[James Bottomley]

On Mon, 2012-02-27 at 11:02 -0500, Paul Gortmaker wrote:
> On 12-02-27 10:24 AM, James Bottomley wrote:
> > On Sun, 2012-02-26 at 15:51 -0500, Paul Gortmaker wrote:
> >> Upstream commit e66eed651fd18a961f11cda62f3b5286c8cc4f9f
> >>
> >>     "list: remove prefetching from regular list iterators"
> >>
> >> removed <linux/prefetch.h> from <linux/list.h> very early in the
> >> v3.0 dev cycle (v2.6.39-2-ge66eed6), thereby causing build failures
> >> when it exposed the implicit prefetch users in parisc.
> > 
> > I already have this in my internal patch queue:
> > 
> > http://marc.info/?l=linux-parisc&m=132825467010662
> 
> Is this in a different queue than what is in linux next?

Yes, sorry ... I hadn't got around to building the parisc tree for the
current version.  I've done that now.

> It would be nice to see such build fixes get into linux-next
> ASAP, so folks doing multi-arch coverage builds on their work
> can know if they've introduced a regression.  At the moment,
> the parisc builds in linux-next have been worthless for over
> six months, and I'm sure I'm not the 1st one to waste time
> "re-solving" old bugs like this.

So if you recheck with the next build of linux-next, they should all be
there (modulo the cross compile patch which is still under discussion).

James