iommu/omap: Fix pointer cast -Wpointer-to-int-cast warnings on 64 bit
PATCH RESEND 1/4 iommu/omap: Fix pointer cast -Wpointer-to-int-cast warnings on 64 bit
pointers should be casted to unsigned long to avoid -Wpointer-to-int-cast warnings when compiling on 64-bit platform (e.g. with COMPILE_TEST):
drivers/iommu/omap-iommu.c: In function ‘omap2_iommu_enable’:
drivers/iommu/omap-iommu.c:170:25: warning:
cast from pointer to integer of different size [-Wpointer-to-int-cast]
if (!obj->iopgd || !IS_ALIGNED((u32)obj->iopgd, SZ_16K))
^---
Not tested on hardware.
--- drivers/iommu/omap-iommu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iommu/omap-iommu.c +1 -1
@@ -167,7 +167,7 @@ static int omap2_iommu_enable(struct omap_iommu *obj) {u32 l, pa; -if (!obj->iopgd || !IS_ALIGNED ((u32 )obj->iopgd,SZ_16K )) +if (!obj->iopgd || !IS_ALIGNED ((unsigned long )obj->iopgd,SZ_16K ))return -EINVAL ; pa =virt_to_phys (obj->iopgd);
Signature · 1 line
2.17.1
PATCH RESEND 2/4 iommu/omap: Fix printing format for size_t on 64-bit
Print size_t as %zu or %zx to fix -Wformat warnings when compiling on 64-bit platform (e.g. with COMPILE_TEST):
drivers/iommu/omap-iommu.c: In function ‘flush_iotlb_page’:
drivers/iommu/omap-iommu.c:437:47: warning:
format ‘%x’ expects argument of type ‘unsigned int’,
but argument 7 has type ‘size_t {aka long unsigned int}’ [-Wformat=]
---
Not tested on hardware.
--- drivers/iommu/omap-iommu.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/iommu/omap-iommu.c +4 -4
@@ -434,7 +434,7 @@ static void flush_iotlb_page(struct omap_iommu *obj, u32 da) bytes =iopgsz_to_bytes (cr.cam &3 );if ((start <= da) && (da < start + bytes)) { -dev_dbg (obj->dev,"%s: %08x<=%08x(%x)\n" , +dev_dbg (obj->dev,"%s: %08x<=%08x(%zx)\n" , __func__, start, da, bytes);iotlb_load_cr (obj, &cr);iommu_write_reg (obj,1 ,MMU_FLUSH_ENTRY );
@@ -1352,11 +1352,11 @@ static int omap_iommu_map(struct iommu_domain *domain, unsigned long da, omap_pgsz =bytes_to_iopgsz (bytes);if (omap_pgsz <0 ) { -dev_err (dev,"invalid size to map: %d\n" , bytes); +dev_err (dev,"invalid size to map: %zu\n" , bytes);return -EINVAL ; } -dev_dbg (dev,"mapping da 0x%lx to pa %pa size 0x%x\n" , da, &pa, bytes); +dev_dbg (dev,"mapping da 0x%lx to pa %pa size 0x%zx\n" , da, &pa, bytes);iotlb_init_entry (&e, da, pa, omap_pgsz);
@@ -1393,7 +1393,7 @@ static size_t omap_iommu_unmap(struct iommu_domain *domain, unsigned long da,size_t bytes =0 ;int i; -dev_dbg (dev,"unmapping da 0x%lx size %u\n" , da, size); +dev_dbg (dev,"unmapping da 0x%lx size %zu\n" , da, size); iommu = omap_domain->iommus;for (i =0 ; i < omap_domain->num_iommus; i++, iommu++) {
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
PATCH RESEND 3/4 iommu/omap: Fix -Woverflow warnings when compiling on 64-bit architectures
In reply to Krzysztof KozlowskiAlthough the OMAP IOMMU driver supports only ARMv7 (32-bit) platforms, it can be compile tested for other architectures, including 64-bit ones. In such case the warning appears:
In file included from drivers/iommu/omap-iommu.c:33:0:
drivers/iommu/omap-iommu.c: In function 'omap_iommu_iova_to_phys':
>> drivers/iommu/omap-iopgtable.h:44:21: warning: large integer implicitly truncated to unsigned type [-Woverflow]
#define IOPTE_MASK (~(IOPTE_SIZE - 1))
^
>> drivers/iommu/omap-iommu.c:1641:41: note: in expansion of macro 'IOPTE_MASK'
ret = omap_iommu_translate(*pte, da, IOPTE_MASK);
^~~~~~~~~~
Fix this by using architecture-depending types in omap_iommu_translate():
1. Pointer should be cast to unsigned long,
2. Virtual addresses should be cast to dma_addr_t.
On 32-bit this will be the same as original code (using u32). On 64-bit it should produce meaningful result, although it does not really matter.
---
Not tested on hardware.
--- drivers/iommu/omap-iopgtable.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/iommu/omap-iopgtable.h +2 -1
@@ -63,7 +63,8 @@* * va to pa translation */ -static inline phys_addr_t omap_iommu_translate (u32 d,u32 va,u32 mask) +static inline phys_addr_t omap_iommu_translate (unsigned long d,dma_addr_t va, +dma_addr_t mask) {return (d & mask) | (va & (~mask)); }
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
PATCH RESEND 4/4 iommu: Enable compile testing for some of drivers
In reply to Krzysztof KozlowskiSome of the IOMMU drivers can be compile tested to increase build coverage. The OMAP, Rockchip and Exynos drivers use device.dev_archdata.iommu field which does not exist on all platforms. The sPAPR TCE and ARM SMMU have also restrictions where they can be built.
--- drivers/iommu/Kconfig | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/iommu/Kconfig +8 -8
@@ -273,7 +273,7 @@ config IRQ_REMAP # OMAP IOMMU support config OMAP_IOMMU bool "OMAP IOMMU Support" - depends on ARM && MMU + depends on ARM && MMU || (COMPILE_TEST && (ARM || ARM64 || IA64 || SPARC)) depends on ARCH_OMAP2PLUS || COMPILE_TEST select IOMMU_API ---help---
@@ -291,7 +291,7 @@ config OMAP_IOMMU_DEBUG config ROCKCHIP_IOMMU bool "Rockchip IOMMU Support" - depends on ARM || ARM64 + depends on ARM || ARM64 || (COMPILE_TEST && (ARM64 || IA64 || SPARC)) depends on ARCH_ROCKCHIP || COMPILE_TEST select IOMMU_API select ARM_DMA_USE_IOMMU
@@ -325,7 +325,7 @@ config TEGRA_IOMMU_SMMU config EXYNOS_IOMMU bool "Exynos IOMMU Support" - depends on ARCH_EXYNOS && MMU + depends on ARCH_EXYNOS && MMU || (COMPILE_TEST && (ARM || ARM64 || IA64 || SPARC)) depends on !CPU_BIG_ENDIAN # revisit driver if we can enable big-endian ptes select IOMMU_API select ARM_DMA_USE_IOMMU
@@ -361,7 +361,7 @@ config IPMMU_VMSA config SPAPR_TCE_IOMMU bool "sPAPR TCE IOMMU Support" - depends on PPC_POWERNV || PPC_PSERIES + depends on PPC_POWERNV || PPC_PSERIES || (PPC && COMPILE_TEST) select IOMMU_API help Enables bits of IOMMU API required by VFIO. The iommu_ops
@@ -370,7 +370,7 @@ config SPAPR_TCE_IOMMU # ARM IOMMU support config ARM_SMMU tristate "ARM Ltd. System MMU (SMMU) Support" - depends on (ARM64 || ARM) && MMU + depends on (ARM64 || ARM || (COMPILE_TEST && !GENERIC_ATOMIC64)) && MMU select IOMMU_API select IOMMU_IO_PGTABLE_LPAE select ARM_DMA_USE_IOMMU if ARM
@@ -440,7 +440,7 @@ config S390_IOMMU config S390_CCW_IOMMU bool "S390 CCW IOMMU Support" - depends on S390 && CCW + depends on S390 && CCW || COMPILE_TEST select IOMMU_API help Enables bits of IOMMU API required by VFIO. The iommu_ops
@@ -448,7 +448,7 @@ config S390_CCW_IOMMU config S390_AP_IOMMU bool "S390 AP IOMMU Support" - depends on S390 && ZCRYPT + depends on S390 && ZCRYPT || COMPILE_TEST select IOMMU_API help Enables bits of IOMMU API required by VFIO. The iommu_ops
@@ -456,7 +456,7 @@ config S390_AP_IOMMU config MTK_IOMMU bool "MTK IOMMU Support" - depends on ARM || ARM64 + depends on ARM || ARM64 || COMPILE_TEST depends on ARCH_MEDIATEK || COMPILE_TEST select ARM_DMA_USE_IOMMU select IOMMU_API
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Quoted 11 lines · Krzysztof Kozlowski
pointers should be casted to unsigned long to avoid -Wpointer-to-int-cast warnings when compiling on 64-bit platform (e.g. with
COMPILE_TEST):drivers/iommu/omap-iommu.c: In function ‘omap2_iommu_enable’: drivers/iommu/omap-iommu.c:170:25: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast] if (!obj->iopgd || !IS_ALIGNED((u32)obj->iopgd, SZ_16K)) ^Signed-off-by Krzysztof Kozlowski <[email protected]>
Applied all, thanks.