Re: [PATCH] intel-iommu: fix compile warnings

[Ingo Molnar <mingo@elte.hu>]

* FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> wrote:

> I got the following warnings on IA64:
> 
> linux-2.6/drivers/pci/intel-iommu.c: In function 'init_dmars':
> linux-2.6/drivers/pci/intel-iommu.c:1658: warning: format '%Lx' expects type 'long long unsigned int', but argument 2 has type 'u64'
> linux-2.6/drivers/pci/intel-iommu.c:1663: warning: format '%Lx' expects type 'long long unsigned int', but argument 2 has type 'u64'

applied to tip/core/urgent, thanks!

Note, i extended the changelog, see the commit attached below.

I think in 2.6.29 we want to convert struct dmar_drhd_unit's 
->reg_base_addr to the phys_addr_t type, and use the %pR extension to 
print it out.

There's other files affected as well by this type problem: 
drivers/pci/intr_remapping.c for example - it's just not built on ia64 
right now because it depends on CONFIG_INTR_REMAP which is only 64-bit 
x86.

Some goes for other fields as well:

	u64	base_address;		/* reserved base address*/
	u64	end_address;		/* reserved end address */

(and some other variables in dmar.c as well)

Would you be interested in sending a patch for this wider cleanup?

It will be a pure cleanup with no functional complications as DMAR is 
only available on 64-bit platforms so phys_addr_t is always 64 bits 
wide - but it will help avoid such problems that u64 types 
historically have.

	Ingo

------------->
>From b4e0f9eb8aeceb22c48fee005378bd19e25216fc Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Wed, 19 Nov 2008 13:53:42 +0900
Subject: [PATCH] intel-iommu: fix compile warnings

Impact: cleanup

I got the following warnings on IA64:

  linux-2.6/drivers/pci/intel-iommu.c: In function 'init_dmars':
  linux-2.6/drivers/pci/intel-iommu.c:1658: warning: format '%Lx' expects type 'long long unsigned int', but argument 2 has type 'u64'
  linux-2.6/drivers/pci/intel-iommu.c:1663: warning: format '%Lx' expects type 'long long unsigned int', but argument 2 has type 'u64'

Another victim of int-ll64.h versus int-l64.h confusion between platforms.

->reg_base_addr has a type of u64 - which can only be printed out
consistently if we cast its type up to LL.

[ Eventually reg_base_addr should be converted to phys_addr_t, for which
  we have the %pR printk helper - but that is out of the scope of late
  -rc's. ]

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 drivers/pci/intel-iommu.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index a269272..5c8baa4 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1655,12 +1655,14 @@ int __init init_dmars(void)
 			iommu->flush.flush_context = __iommu_flush_context;
 			iommu->flush.flush_iotlb = __iommu_flush_iotlb;
 			printk(KERN_INFO "IOMMU 0x%Lx: using Register based "
-			       "invalidation\n", drhd->reg_base_addr);
+			       "invalidation\n",
+			       (unsigned long long)drhd->reg_base_addr);
 		} else {
 			iommu->flush.flush_context = qi_flush_context;
 			iommu->flush.flush_iotlb = qi_flush_iotlb;
 			printk(KERN_INFO "IOMMU 0x%Lx: using Queued "
-			       "invalidation\n", drhd->reg_base_addr);
+			       "invalidation\n",
+			       (unsigned long long)drhd->reg_base_addr);
 		}
 	}