Fix virtual and physical address types

PATCH In review Ben Hutchings 7 participants 20 messages Dec 21, 2011 – Dec 22, 2011

Subsystems (11)

Networking Network drivers DRM drivers SCSI +7 more subsystems

lore

Share & tools

lore

Ben Hutchings

author

2011-12-21 01:29 UTC #0 lore raw

cover Fix virtual and physical address types

This series fixes compiler warnings on some architectures about implicit conversions and narrowing conversions between pointer and integer types.

Please apply these to the appropriate trees.

Ben.

Ben Hutchings (8):
IB/cxgb4: Fix formatting of physical address
farsync: Fix confusion about DMA address and buffer offset types
drm: Do not include page offset in argument to virt_to_page()
drm: Pass pointers to virt_to_page()
[SCSI] tgt: Pass pointers to virt_to_page(), not integers
uio: Pass pointers to virt_to_page(), not integers
rds: Pass pointers to virt_to_page(), not integers
pmcraid: Pass pointers to access_ok(), not integers

 drivers/gpu/drm/drm_pci.c            |    4 ++--
 drivers/gpu/drm/drm_vm.c             |    2 +-
 drivers/infiniband/hw/cxgb4/device.c |    4 ++--
 drivers/net/wan/farsync.c            |   19 +++++++------------
 drivers/scsi/pmcraid.c               |    3 ++-
 drivers/scsi/scsi_tgt_if.c           |    2 +-
 drivers/uio/uio.c                    |    6 ++++--
 net/rds/message.c                    |    2 +-
 8 files changed, 20 insertions(+), 22 deletions(-)

Signature · 1 line

1.7.7.3

Ben Hutchings

author

2011-12-21 01:32 UTC #1 lore raw

PATCH 1/8 IB/cxgb4: Fix formatting of physical address

Physical addresses may be wider than virtual addresses (e.g. on i386 with PAE) and must not be formatted with %p.

Signed-off-by Ben Hutchings <ben-/[email protected]>

---
The resource could alternately be formatted with %Pr.

Ben.

 drivers/infiniband/hw/cxgb4/device.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

drivers/infiniband/hw/cxgb4/device.c +2 -2

index 6d0df6e..d395623 100644
@@ -318,10 +318,10 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev)
 	     rdev->lldi.vr->qp.size,
 	     rdev->lldi.vr->cq.start,
 	     rdev->lldi.vr->cq.size);
-	PDBG("udb len 0x%x udb base %p db_reg %p gts_reg %p qpshift %lu "
+	PDBG("udb len 0x%x udb base %llx db_reg %p gts_reg %p qpshift %lu "
 	     "qpmask 0x%x cqshift %lu cqmask 0x%x\n",
 	     (unsigned)pci_resource_len(rdev->lldi.pdev, 2),
-	     (void *)pci_resource_start(rdev->lldi.pdev, 2),
+	     (u64)pci_resource_start(rdev->lldi.pdev, 2),
 	     rdev->lldi.db_reg,
 	     rdev->lldi.gts_reg,
 	     rdev->qpshift, rdev->qpmask,

Signature · 8 lines

1.7.7.3



--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Joe Perches

2011-12-21 02:35 UTC #2 lore raw

Re: [PATCH 1/8] IB/cxgb4: Fix formatting of physical address

Quoted 2 lines · Ben Hutchings

Physical addresses may be wider than virtual addresses (e.g. on i386 with PAE) and must not be formatted with %p.

[]

drivers/infiniband/hw/cxgb4/device.c

Quoted 2 lines

The resource could alternately be formatted with %Pr.
drivers/infiniband/hw/cxgb4/device.c +0 -0

[]

drivers/infiniband/hw/cxgb4/device.c static int c4iw_rdev_open(struct c4iw_rdev *rdev) · line 324

Quoted 10 lines

drivers/infiniband/hw/cxgb4/device.c +2 -2

@@ -318,10 +318,10 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev)
 	     rdev->lldi.vr->qp.size,
 	     rdev->lldi.vr->cq.start,
 	     rdev->lldi.vr->cq.size);
-	PDBG("udb len 0x%x udb base %p db_reg %p gts_reg %p qpshift %lu "
+	PDBG("udb len 0x%x udb base %llx db_reg %p gts_reg %p qpshift %lu "
 	     "qpmask 0x%x cqshift %lu cqmask 0x%x\n",
 	     (unsigned)pci_resource_len(rdev->lldi.pdev, 2),
-	     (void *)pci_resource_start(rdev->lldi.pdev, 2),
+	     (u64)pci_resource_start(rdev->lldi.pdev, 2),

Probably should be unsigned long long

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

Ben Hutchings

author

2011-12-21 02:47 UTC #3 lore raw

Re: [PATCH 1/8] IB/cxgb4: Fix formatting of physical address

Quoted 7 lines · Joe Perches

Quoted 2 lines · Ben Hutchings
Physical addresses may be wider than virtual addresses (e.g. on i386 with PAE) and must not be formatted with %p.
[]
drivers/infiniband/hw/cxgb4/device.c
Quoted 2 lines
The resource could alternately be formatted with %Pr.
drivers/infiniband/hw/cxgb4/device.c +0 -0
[]

Quoted 12 lines

drivers/infiniband/hw/cxgb4/device.c static int c4iw_rdev_open(struct c4iw_rdev *rdev) · line 324

Quoted 10 lines

drivers/infiniband/hw/cxgb4/device.c +2 -2

@@ -318,10 +318,10 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev)
 	     rdev->lldi.vr->qp.size,
 	     rdev->lldi.vr->cq.start,
 	     rdev->lldi.vr->cq.size);
-	PDBG("udb len 0x%x udb base %p db_reg %p gts_reg %p qpshift %lu "
+	PDBG("udb len 0x%x udb base %llx db_reg %p gts_reg %p qpshift %lu "
 	     "qpmask 0x%x cqshift %lu cqmask 0x%x\n",
 	     (unsigned)pci_resource_len(rdev->lldi.pdev, 2),
-	     (void *)pci_resource_start(rdev->lldi.pdev, 2),
+	     (u64)pci_resource_start(rdev->lldi.pdev, 2),

Probably should be unsigned long long

unsigned long long is so long long long to type. And u64 is now defined as unsigned long long on all architectures, partly because of that.

Ben.

Signature · 2 lines

Ben Hutchings
Humans are not rational beings; they are rationalising beings.

[email protected]

2011-12-22 19:52 UTC #4 lore raw

Re: [PATCH 1/8] IB/cxgb4: Fix formatting of physical address

In reply to Joe Perches

Quoted 7 lines · Joe Perches

Quoted 2 lines · Ben Hutchings
Physical addresses may be wider than virtual addresses (e.g. on i386 with PAE) and must not be formatted with %p.
[]
Quoted 1 line
The resource could alternately be formatted with %Pr.
Probably should be unsigned long long

I don't have a dog in this fight, but the %Pr sounds better to me.

Steve Wise

Acked-by on v1 1/8

2011-12-21 14:12 UTC #5 lore raw

Re: [PATCH 1/8] IB/cxgb4: Fix formatting of physical address

In reply to Ben Hutchings

Acked-by Steve Wise <[email protected]>

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

Ben Hutchings

author

2011-12-21 01:32 UTC #6 lore raw

PATCH 2/8 farsync: Fix confusion about DMA address and buffer offset types

In reply to Ben Hutchings

Consistently use dma_addr_t for DMA addresses, except in fst_{rx,tx}_dma where we truncate them to 32 bits. (We know this is OK because that's the default DMA mask for PCI devices.)

Consistently use u32 for buffer offsets within shared memory.

Compile-tested only.

Signed-off-by Ben Hutchings <[email protected]>

---
 drivers/net/wan/farsync.c |   19 +++++++------------
 1 files changed, 7 insertions(+), 12 deletions(-)

drivers/net/wan/farsync.c +7 ~~-12~~

index ebb9f24..98c04c6 100644
@@ -886,15 +886,13 @@ fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
  *      Receive a frame through the DMA
  */
 static inline void
-fst_rx_dma(struct fst_card_info *card, dma_addr_t skb,
-	   dma_addr_t mem, int len)
+fst_rx_dma(struct fst_card_info *card, dma_addr_t skb, u32 mem, int len)
 {
 	/*
 	 * This routine will setup the DMA and start it
 	 */
 
-	dbg(DBG_RX, "In fst_rx_dma %lx %lx %d\n",
-	    (unsigned long) skb, (unsigned long) mem, len);
+	dbg(DBG_RX, "In fst_rx_dma %x %x %d\n", (u32)skb, mem, len);
 	if (card->dmarx_in_progress) {
 		dbg(DBG_ASS, "In fst_rx_dma while dma in progress\n");
 	}

@@ -915,20 +913,19 @@ fst_rx_dma(struct fst_card_info *card, dma_addr_t skb,
  *      Send a frame through the DMA
  */
 static inline void
-fst_tx_dma(struct fst_card_info *card, unsigned char *skb,
-	   unsigned char *mem, int len)
+fst_tx_dma(struct fst_card_info *card, dma_addr_t skb, u32 mem, int len)
 {
 	/*
 	 * This routine will setup the DMA and start it.
 	 */
 
-	dbg(DBG_TX, "In fst_tx_dma %p %p %d\n", skb, mem, len);
+	dbg(DBG_TX, "In fst_tx_dma %x %x %d\n", (u32)skb, mem, len);
 	if (card->dmatx_in_progress) {
 		dbg(DBG_ASS, "In fst_tx_dma while dma in progress\n");
 	}
 
-	outl((unsigned long) skb, card->pci_conf + DMAPADR1);	/* Copy from here */
-	outl((unsigned long) mem, card->pci_conf + DMALADR1);	/* to here */
+	outl(skb, card->pci_conf + DMAPADR1);	/* Copy from here */
+	outl(mem, card->pci_conf + DMALADR1);	/* to here */
 	outl(len, card->pci_conf + DMASIZ1);	/* for this length */
 	outl(0x000000004, card->pci_conf + DMADPR1);	/* In this direction */

@@ -1405,9 +1402,7 @@ do_bottom_half_tx(struct fst_card_info *card)
 					card->dma_len_tx = skb->len;
 					card->dma_txpos = port->txpos;
 					fst_tx_dma(card,
-						   (char *) card->
-						   tx_dma_handle_card,
-						   (char *)
+						   card->tx_dma_handle_card,
 						   BUF_OFFSET(txBuffer[pi]
 							      [port->txpos][0]),
 						   skb->len);

Signature · 1 line

1.7.7.3

Ben Hutchings

author

2011-12-21 01:42 UTC #7 lore raw

Re: [PATCH 2/8] farsync: Fix confusion about DMA address and buffer offset types

I accidentally left signing enabled for this one, which I had intended to disable since you've said that patchwork doesn't deal with signatures. Except this looks OK on patchwork:
<http://patchwork.ozlabs.org/patch/132548/>.

If that also looks OK to you, is there any reason to avoid signing networking patches?

Ben.

Signature · 2 lines

Ben Hutchings
Humans are not rational beings; they are rationalising beings.

David Miller

maintainer

2011-12-21 04:48 UTC #8 lore raw

Re: [PATCH 2/8] farsync: Fix confusion about DMA address and buffer offset types

From: Ben Hutchings <[email protected]>
Date: Wed, 21 Dec 2011 01:42:26 +0000

Quoted 7 lines

I accidentally left signing enabled for this one, which I had intended to disable since you've said that patchwork doesn't deal with signatures. Except this looks OK on patchwork:
<http://patchwork.ozlabs.org/patch/132548/>.
If that also looks OK to you, is there any reason to avoid signing networking patches?

What do you mean exactly by "signing".

Do you mean simply the "Signed-off-by: " tag, which you should always add to your own patches, or something else?

Ben Hutchings

author

2011-12-21 05:41 UTC #9 lore raw

Re: [PATCH 2/8] farsync: Fix confusion about DMA address and buffer offset types

Quoted 8 lines · David Miller

From: Ben Hutchings <[email protected]>
Date: Wed, 21 Dec 2011 01:42:26 +0000
Quoted 5 lines
I accidentally left signing enabled for this one, which I had intended to disable since you've said that patchwork doesn't deal with signatures. Except this looks OK on patchwork:
<http://patchwork.ozlabs.org/patch/132548/>.

Quoted 7 lines

Quoted 2 lines
If that also looks OK to you, is there any reason to avoid signing networking patches?
What do you mean exactly by "signing".
Do you mean simply the "Signed-off-by: " tag, which you should always add to your own patches, or something else?

GPG signing.

Ben.

Signature · 2 lines

Ben Hutchings
Humans are not rational beings; they are rationalising beings.

David Miller

maintainer

2011-12-21 04:47 UTC #10 lore raw

Re: [PATCH 2/8] farsync: Fix confusion about DMA address and buffer offset types

In reply to Ben Hutchings

From: Ben Hutchings <[email protected]>
Date: Wed, 21 Dec 2011 01:32:52 +0000

drivers/net/wan/farsync.c fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port, · line 895

Quoted 3 lines

drivers/net/wan/farsync.c +1 -2

-	dbg(DBG_RX, "In fst_rx_dma %lx %lx %d\n",
-	    (unsigned long) skb, (unsigned long) mem, len);
+	dbg(DBG_RX, "In fst_rx_dma %x %x %d\n", (u32)skb, mem, len);

This is more appropriately fixed by using "%p" instead of casting to a 32-bit int.

Ben Hutchings

author

2011-12-21 05:41 UTC #11 lore raw

Re: [PATCH 2/8] farsync: Fix confusion about DMA address and buffer offset types

Quoted 9 lines · David Miller

From: Ben Hutchings <[email protected]>
Date: Wed, 21 Dec 2011 01:32:52 +0000
drivers/net/wan/farsync.c fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port, · line 895
Quoted 3 lines
drivers/net/wan/farsync.c +1 -2
-	dbg(DBG_RX, "In fst_rx_dma %lx %lx %d\n",
-	    (unsigned long) skb, (unsigned long) mem, len);
+	dbg(DBG_RX, "In fst_rx_dma %x %x %d\n", (u32)skb, mem, len);
This is more appropriately fixed by using "%p" instead of casting to a 32-bit int.

'skb' is the physical address of the data in the skb. Whereas 'mem' is the offset of the corresponding buffer in shared memory.

All completely clear, right? ;-)

Ben.

Signature · 2 lines

Ben Hutchings
Humans are not rational beings; they are rationalising beings.

Kevin Curtis

2011-12-21 08:48 UTC #12 lore raw

RE: [PATCH 2/8] farsync: Fix confusion about DMA address and buffer offset types

Hi,
Thanks for the patches. I will apply them to the master copy of the source code here.

Regards

Kevin Curtis
Linux Development
FarSite Communications Ltd http://www.farsite.com
Winner of The Queen's Award for Enterprise 2009
tel: +44 1256 330461
fax: +44 1256 854931

-----Original Message-----
From: Ben Hutchings [mailto:[email protected]]
Sent: 21 December 2011 05:41
To: David Miller
Cc: Kevin Curtis; [email protected]; [email protected]
Subject: Re: [PATCH 2/8] farsync: Fix confusion about DMA address and buffer offset types

Quoted 9 lines · David Miller

From: Ben Hutchings <[email protected]>
Date: Wed, 21 Dec 2011 01:32:52 +0000
drivers/net/wan/farsync.c fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port, · line 895
Quoted 3 lines
drivers/net/wan/farsync.c +1 -2
-	dbg(DBG_RX, "In fst_rx_dma %lx %lx %d\n",
-	    (unsigned long) skb, (unsigned long) mem, len);
+	dbg(DBG_RX, "In fst_rx_dma %x %x %d\n", (u32)skb, mem, len);
This is more appropriately fixed by using "%p" instead of casting to a 32-bit int.

'skb' is the physical address of the data in the skb. Whereas 'mem' is the offset of the corresponding buffer in shared memory.

All completely clear, right? ;-)

Ben.

--
Ben Hutchings
Humans are not rational beings; they are rationalising beings.

Ben Hutchings

author

2011-12-21 01:33 UTC #13 lore raw

PATCH 3/8 drm: Do not include page offset in argument to virt_to_page()

In reply to Ben Hutchings

By definition, the page offset will not affect the result.

Compile-tested only.

Signed-off-by Ben Hutchings <[email protected]>

---
 drivers/gpu/drm/drm_vm.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

drivers/gpu/drm/drm_vm.c +1 -1

index 8c03eaf..98c3922 100644
@@ -306,7 +306,7 @@ static int drm_do_vm_dma_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 	offset = (unsigned long)vmf->virtual_address - vma->vm_start;	/* vm_[pg]off[set] should be 0 */
 	page_nr = offset >> PAGE_SHIFT; /* page_nr could just be vmf->pgoff */
-	page = virt_to_page((dma->pagelist[page_nr] + (offset & (~PAGE_MASK))));
+	page = virt_to_page(dma->pagelist[page_nr]);
 
 	get_page(page);
 	vmf->page = page;

Signature · 1 line

1.7.7.3

Ben Hutchings

author

2011-12-21 01:33 UTC #14 lore raw

PATCH 4/8 drm: Pass pointers to virt_to_page()

In reply to Ben Hutchings

Most architectures define virt_to_page() as a macro that casts its argument such that an argument of type unsigned long will be accepted without complaint. However, the proper type is void *, and passing unsigned long results in a warning on MIPS.

Signed-off-by Ben Hutchings <[email protected]>

---
 drivers/gpu/drm/drm_pci.c |    4 ++--
 drivers/gpu/drm/drm_vm.c  |    2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

drivers/gpu/drm/drm_pci.c +2 -2

index d4d10b7..2e19fcc 100644
@@ -82,7 +82,7 @@ drm_dma_handle_t *drm_pci_alloc(struct drm_device * dev, size_t size, size_t ali
 	/* Reserve */
 	for (addr = (unsigned long)dmah->vaddr, sz = size;
 	     sz > 0; addr += PAGE_SIZE, sz -= PAGE_SIZE) {
-		SetPageReserved(virt_to_page(addr));
+		SetPageReserved(virt_to_page((void *)addr));
 	}
 
 	return dmah;

@@ -107,7 +107,7 @@ void __drm_pci_free(struct drm_device * dev, drm_dma_handle_t * dmah)
 		/* Unreserve */
 		for (addr = (unsigned long)dmah->vaddr, sz = dmah->size;
 		     sz > 0; addr += PAGE_SIZE, sz -= PAGE_SIZE) {
-			ClearPageReserved(virt_to_page(addr));
+			ClearPageReserved(virt_to_page((void *)addr));
 		}
 		dma_free_coherent(&dev->pdev->dev, dmah->size, dmah->vaddr,
 				  dmah->busaddr);

drivers/gpu/drm/drm_vm.c +1 -1

index 98c3922..8cc1c06 100644
@@ -306,7 +306,7 @@ static int drm_do_vm_dma_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 	offset = (unsigned long)vmf->virtual_address - vma->vm_start;	/* vm_[pg]off[set] should be 0 */
 	page_nr = offset >> PAGE_SHIFT; /* page_nr could just be vmf->pgoff */
-	page = virt_to_page(dma->pagelist[page_nr]);
+	page = virt_to_page((void *)dma->pagelist[page_nr]);
 
 	get_page(page);
 	vmf->page = page;

Signature · 1 line

1.7.7.3

Ben Hutchings

author

2011-12-21 01:34 UTC #15 lore raw

PATCH 5/8 tgt: Pass pointers to virt_to_page(), not integers

In reply to Ben Hutchings

Signed-off-by Ben Hutchings <[email protected]>

---
 drivers/scsi/scsi_tgt_if.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

drivers/scsi/scsi_tgt_if.c +1 -1

index 6209110..7199753 100644
@@ -286,7 +286,7 @@ static int uspace_ring_map(struct vm_area_struct *vma, unsigned long addr,
 	int i, err;
 
 	for (i = 0; i < TGT_RING_PAGES; i++) {
-		struct page *page = virt_to_page(ring->tr_pages[i]);
+		struct page *page = virt_to_page((void *)ring->tr_pages[i]);
 		err = vm_insert_page(vma, addr, page);
 		if (err)
 			return err;

Signature · 1 line

1.7.7.3

Ben Hutchings

author

2011-12-21 01:36 UTC #16 lore raw

PATCH 6/8 uio: Pass pointers to virt_to_page(), not integers

In reply to Ben Hutchings

Signed-off-by Ben Hutchings <[email protected]>

---
It might be clearer to assign:

	addr = (void *)((unsigned long)idev->info->mem[mi].addr + offset);

since arithmetic on void pointers is weird. Either way should work though.

Ben.

 drivers/uio/uio.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)

drivers/uio/uio.c +4 -2

index a783d53..1ff7927 100644
@@ -620,6 +620,7 @@ static int uio_vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	struct uio_device *idev = vma->vm_private_data;
 	struct page *page;
 	unsigned long offset;
+	void *addr;
 
 	int mi = uio_find_mem_index(vma);
 	if (mi < 0)

@@ -631,10 +632,11 @@ static int uio_vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	 */
 	offset = (vmf->pgoff - mi) << PAGE_SHIFT;
 
+	addr = (void *)(unsigned long)idev->info->mem[mi].addr + offset;
 	if (idev->info->mem[mi].memtype == UIO_MEM_LOGICAL)
-		page = virt_to_page(idev->info->mem[mi].addr + offset);
+		page = virt_to_page(addr);
 	else
-		page = vmalloc_to_page((void *)(unsigned long)idev->info->mem[mi].addr + offset);
+		page = vmalloc_to_page(addr);
 	get_page(page);
 	vmf->page = page;
 	return 0;

Signature · 1 line

1.7.7.3

Hans J. Koch

2011-12-21 02:00 UTC #17 lore raw

Re: [PATCH 6/8] uio: Pass pointers to virt_to_page(), not integers

Quoted 4 lines · Ben Hutchings

Most architectures define virt_to_page() as a macro that casts its argument such that an argument of type unsigned long will be accepted without complaint. However, the proper type is void *, and passing unsigned long results in a warning on MIPS.

Not only on MIPS... This is already fixed in Greg's tree and will go to mainline in the next merge window.

Thanks,
Hans

Quoted 46 lines

Signed-off-by Ben Hutchings <[email protected]>

---
It might be clearer to assign:

	addr = (void *)((unsigned long)idev->info->mem[mi].addr + offset);

since arithmetic on void pointers is weird. Either way should work though.

Ben.

 drivers/uio/uio.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)

drivers/uio/uio.c +4 -2

index a783d53..1ff7927 100644
@@ -620,6 +620,7 @@ static int uio_vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	struct uio_device *idev = vma->vm_private_data;
 	struct page *page;
 	unsigned long offset;
+	void *addr;
 
 	int mi = uio_find_mem_index(vma);
 	if (mi < 0)

@@ -631,10 +632,11 @@ static int uio_vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	 */
 	offset = (vmf->pgoff - mi) << PAGE_SHIFT;
 
+	addr = (void *)(unsigned long)idev->info->mem[mi].addr + offset;
 	if (idev->info->mem[mi].memtype == UIO_MEM_LOGICAL)
-		page = virt_to_page(idev->info->mem[mi].addr + offset);
+		page = virt_to_page(addr);
 	else
-		page = vmalloc_to_page((void *)(unsigned long)idev->info->mem[mi].addr + offset);
+		page = vmalloc_to_page(addr);
 	get_page(page);
 	vmf->page = page;
 	return 0;

Signature · 1 line

1.7.7.3

Ben Hutchings

author

2011-12-21 01:37 UTC #18 lore raw

PATCH 7/8 rds: Pass pointers to virt_to_page(), not integers

In reply to Ben Hutchings

Signed-off-by Ben Hutchings <[email protected]>

---
 net/rds/message.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

net/rds/message.c +1 -1

index f0a4658..ccd3de5 100644
@@ -257,7 +257,7 @@ struct rds_message *rds_message_map_pages(unsigned long *page_addrs, unsigned in
 
 	for (i = 0; i < rm->data.op_nents; ++i) {
 		sg_set_page(&rm->data.op_sg[i],
-				virt_to_page(page_addrs[i]),
+				virt_to_page((void *)page_addrs[i]),
 				PAGE_SIZE, 0);
 	}

Signature · 1 line

1.7.7.3

Ben Hutchings

author

2011-12-21 01:37 UTC #19 lore raw

PATCH 8/8 pmcraid: Pass pointers to access_ok(), not integers

In reply to Ben Hutchings

Most architectures define access_ok() as a macro that casts its argument such that an argument of type unsigned long will be accepted without complaint. However, the proper type is void *, and passing unsigned long results in a warning on sparc64.

Signed-off-by Ben Hutchings <[email protected]>

---
 drivers/scsi/pmcraid.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

drivers/scsi/pmcraid.c +2 -1

index 5163edb..01b29a6 100644
@@ -3807,7 +3807,8 @@ static long pmcraid_ioctl_passthrough(
 	}
 
 	if (request_size > 0) {
-		rc = access_ok(access, arg, request_offset + request_size);
+		rc = access_ok(access, (void *)arg,
+			       request_offset + request_size);
 
 		if (!rc) {
 			rc = -EFAULT;

Signature · 1 line

1.7.7.3