* [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit
@ 2017-02-14 21:21 Arnd Bergmann
2017-02-15 6:21 ` Sumit Saxena
` (3 more replies)
0 siblings, 4 replies; 5+ messages in thread
From: Arnd Bergmann @ 2017-02-14 21:21 UTC (permalink / raw
To: James E.J. Bottomley, Martin K. Petersen
Cc: Arnd Bergmann, Kashyap Desai, Sumit Saxena, Shivasharan S,
Tomas Henzl, Hannes Reinecke, Sasikumar Chandrasekaran,
megaraidlinux.pdl, linux-scsi, linux-kernel
When building with a dma_addr_t that is different from pointer size, we get
this warning:
drivers/scsi/megaraid/megaraid_sas_fusion.c: In function 'megasas_make_prp_nvme':
drivers/scsi/megaraid/megaraid_sas_fusion.c:1654:17: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
It's better to not pretend that the dma address is a pointer and instead
use a dma_addr_t consistently.
Fixes: 33203bc4d61b ("scsi: megaraid_sas: NVME fast path io support")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
drivers/scsi/megaraid/megaraid_sas_fusion.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c b/drivers/scsi/megaraid/megaraid_sas_fusion.c
index 750090119f81..29650ba669da 100644
--- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
@@ -1619,7 +1619,8 @@ megasas_make_prp_nvme(struct megasas_instance *instance, struct scsi_cmnd *scmd,
{
int sge_len, offset, num_prp_in_chain = 0;
struct MPI25_IEEE_SGE_CHAIN64 *main_chain_element, *ptr_first_sgl;
- u64 *ptr_sgl, *ptr_sgl_phys;
+ u64 *ptr_sgl;
+ dma_addr_t ptr_sgl_phys;
u64 sge_addr;
u32 page_mask, page_mask_result;
struct scatterlist *sg_scmd;
@@ -1651,14 +1652,14 @@ megasas_make_prp_nvme(struct megasas_instance *instance, struct scsi_cmnd *scmd,
*/
page_mask = mr_nvme_pg_size - 1;
ptr_sgl = (u64 *)cmd->sg_frame;
- ptr_sgl_phys = (u64 *)cmd->sg_frame_phys_addr;
+ ptr_sgl_phys = cmd->sg_frame_phys_addr;
memset(ptr_sgl, 0, instance->max_chain_frame_sz);
/* Build chain frame element which holds all prps except first*/
main_chain_element = (struct MPI25_IEEE_SGE_CHAIN64 *)
((u8 *)sgl_ptr + sizeof(struct MPI25_IEEE_SGE_CHAIN64));
- main_chain_element->Address = cpu_to_le64((uintptr_t)ptr_sgl_phys);
+ main_chain_element->Address = cpu_to_le64(ptr_sgl_phys);
main_chain_element->NextChainOffset = 0;
main_chain_element->Flags = IEEE_SGE_FLAGS_CHAIN_ELEMENT |
IEEE_SGE_FLAGS_SYSTEM_ADDR |
@@ -1696,16 +1697,15 @@ megasas_make_prp_nvme(struct megasas_instance *instance, struct scsi_cmnd *scmd,
scmd_printk(KERN_NOTICE,
scmd, "page boundary ptr_sgl: 0x%p\n",
ptr_sgl);
- ptr_sgl_phys++;
- *ptr_sgl =
- cpu_to_le64((uintptr_t)ptr_sgl_phys);
+ ptr_sgl_phys += 8;
+ *ptr_sgl = cpu_to_le64(ptr_sgl_phys);
ptr_sgl++;
num_prp_in_chain++;
}
*ptr_sgl = cpu_to_le64(sge_addr);
ptr_sgl++;
- ptr_sgl_phys++;
+ ptr_sgl_phys += 8;
num_prp_in_chain++;
sge_addr += mr_nvme_pg_size;
--
2.9.0
^ permalink raw reply related [flat|nested] 5+ messages in thread
* RE: [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit
2017-02-14 21:21 [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit Arnd Bergmann
@ 2017-02-15 6:21 ` Sumit Saxena
2017-02-15 13:10 ` Sumit Saxena
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Sumit Saxena @ 2017-02-15 6:21 UTC (permalink / raw
To: Arnd Bergmann, James E.J. Bottomley, Martin K. Petersen
Cc: Kashyap Desai, Shivasharan Srikanteshwara, Tomas Henzl,
Hannes Reinecke, Sasikumar PC, PDL,MEGARAIDLINUX, linux-scsi,
linux-kernel
>-----Original Message-----
>From: Arnd Bergmann [mailto:arnd@arndb.de]
>Sent: Wednesday, February 15, 2017 2:52 AM
>To: James E.J. Bottomley; Martin K. Petersen
>Cc: Arnd Bergmann; Kashyap Desai; Sumit Saxena; Shivasharan S; Tomas
Henzl;
>Hannes Reinecke; Sasikumar Chandrasekaran;
>megaraidlinux.pdl@broadcom.com; linux-scsi@vger.kernel.org; linux-
>kernel@vger.kernel.org
>Subject: [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit
>
>When building with a dma_addr_t that is different from pointer size, we
get this
>warning:
>
>drivers/scsi/megaraid/megaraid_sas_fusion.c: In function
>'megasas_make_prp_nvme':
>drivers/scsi/megaraid/megaraid_sas_fusion.c:1654:17: error: cast to
pointer
>from integer of different size [-Werror=int-to-pointer-cast]
>
>It's better to not pretend that the dma address is a pointer and instead
use a
>dma_addr_t consistently.
Patch looks good from review but we need to have some test runs before
acking this.
I will get back after some test runs.
Thanks,
Sumit
>
>Fixes: 33203bc4d61b ("scsi: megaraid_sas: NVME fast path io support")
>Signed-off-by: Arnd Bergmann <arnd@arndb.de>
>---
> drivers/scsi/megaraid/megaraid_sas_fusion.c | 14 +++++++-------
> 1 file changed, 7 insertions(+), 7 deletions(-)
>
>diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c
>b/drivers/scsi/megaraid/megaraid_sas_fusion.c
>index 750090119f81..29650ba669da 100644
>--- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
>+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
>@@ -1619,7 +1619,8 @@ megasas_make_prp_nvme(struct megasas_instance
>*instance, struct scsi_cmnd *scmd, {
> int sge_len, offset, num_prp_in_chain = 0;
> struct MPI25_IEEE_SGE_CHAIN64 *main_chain_element, *ptr_first_sgl;
>- u64 *ptr_sgl, *ptr_sgl_phys;
>+ u64 *ptr_sgl;
>+ dma_addr_t ptr_sgl_phys;
> u64 sge_addr;
> u32 page_mask, page_mask_result;
> struct scatterlist *sg_scmd;
>@@ -1651,14 +1652,14 @@ megasas_make_prp_nvme(struct megasas_instance
>*instance, struct scsi_cmnd *scmd,
> */
> page_mask = mr_nvme_pg_size - 1;
> ptr_sgl = (u64 *)cmd->sg_frame;
>- ptr_sgl_phys = (u64 *)cmd->sg_frame_phys_addr;
>+ ptr_sgl_phys = cmd->sg_frame_phys_addr;
> memset(ptr_sgl, 0, instance->max_chain_frame_sz);
>
> /* Build chain frame element which holds all prps except first*/
> main_chain_element = (struct MPI25_IEEE_SGE_CHAIN64 *)
> ((u8 *)sgl_ptr + sizeof(struct MPI25_IEEE_SGE_CHAIN64));
>
>- main_chain_element->Address =
cpu_to_le64((uintptr_t)ptr_sgl_phys);
>+ main_chain_element->Address = cpu_to_le64(ptr_sgl_phys);
> main_chain_element->NextChainOffset = 0;
> main_chain_element->Flags = IEEE_SGE_FLAGS_CHAIN_ELEMENT |
> IEEE_SGE_FLAGS_SYSTEM_ADDR |
>@@ -1696,16 +1697,15 @@ megasas_make_prp_nvme(struct megasas_instance
>*instance, struct scsi_cmnd *scmd,
> scmd_printk(KERN_NOTICE,
> scmd, "page boundary ptr_sgl: 0x%p\n",
> ptr_sgl);
>- ptr_sgl_phys++;
>- *ptr_sgl =
>- cpu_to_le64((uintptr_t)ptr_sgl_phys);
>+ ptr_sgl_phys += 8;
>+ *ptr_sgl = cpu_to_le64(ptr_sgl_phys);
> ptr_sgl++;
> num_prp_in_chain++;
> }
>
> *ptr_sgl = cpu_to_le64(sge_addr);
> ptr_sgl++;
>- ptr_sgl_phys++;
>+ ptr_sgl_phys += 8;
> num_prp_in_chain++;
>
> sge_addr += mr_nvme_pg_size;
>--
>2.9.0
^ permalink raw reply [flat|nested] 5+ messages in thread
* RE: [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit
2017-02-14 21:21 [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit Arnd Bergmann
2017-02-15 6:21 ` Sumit Saxena
@ 2017-02-15 13:10 ` Sumit Saxena
2017-02-15 14:15 ` Sumit Saxena
2017-02-16 2:40 ` Martin K. Petersen
3 siblings, 0 replies; 5+ messages in thread
From: Sumit Saxena @ 2017-02-15 13:10 UTC (permalink / raw
To: Arnd Bergmann, James E.J. Bottomley, Martin K. Petersen
Cc: Kashyap Desai, Shivasharan Srikanteshwara, Tomas Henzl,
Hannes Reinecke, Sasikumar PC, PDL,MEGARAIDLINUX, linux-scsi,
linux-kernel
>-----Original Message-----
>From: Arnd Bergmann [mailto:arnd@arndb.de]
>Sent: Wednesday, February 15, 2017 2:52 AM
>To: James E.J. Bottomley; Martin K. Petersen
>Cc: Arnd Bergmann; Kashyap Desai; Sumit Saxena; Shivasharan S; Tomas
Henzl;
>Hannes Reinecke; Sasikumar Chandrasekaran;
>megaraidlinux.pdl@broadcom.com; linux-scsi@vger.kernel.org; linux-
>kernel@vger.kernel.org
>Subject: [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit
>
>When building with a dma_addr_t that is different from pointer size, we
get this
>warning:
>
>drivers/scsi/megaraid/megaraid_sas_fusion.c: In function
>'megasas_make_prp_nvme':
>drivers/scsi/megaraid/megaraid_sas_fusion.c:1654:17: error: cast to
pointer
>from integer of different size [-Werror=int-to-pointer-cast]
>
>It's better to not pretend that the dma address is a pointer and instead
use a
>dma_addr_t consistently.
>
>Fixes: 33203bc4d61b ("scsi: megaraid_sas: NVME fast path io support")
>Signed-off-by: Arnd Bergmann <arnd@arndb.de>
>---
> drivers/scsi/megaraid/megaraid_sas_fusion.c | 14 +++++++-------
> 1 file changed, 7 insertions(+), 7 deletions(-)
>
>diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c
>b/drivers/scsi/megaraid/megaraid_sas_fusion.c
>index 750090119f81..29650ba669da 100644
>--- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
>+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
>@@ -1619,7 +1619,8 @@ megasas_make_prp_nvme(struct megasas_instance
>*instance, struct scsi_cmnd *scmd, {
> int sge_len, offset, num_prp_in_chain = 0;
> struct MPI25_IEEE_SGE_CHAIN64 *main_chain_element, *ptr_first_sgl;
>- u64 *ptr_sgl, *ptr_sgl_phys;
>+ u64 *ptr_sgl;
>+ dma_addr_t ptr_sgl_phys;
> u64 sge_addr;
> u32 page_mask, page_mask_result;
> struct scatterlist *sg_scmd;
>@@ -1651,14 +1652,14 @@ megasas_make_prp_nvme(struct megasas_instance
>*instance, struct scsi_cmnd *scmd,
> */
> page_mask = mr_nvme_pg_size - 1;
> ptr_sgl = (u64 *)cmd->sg_frame;
>- ptr_sgl_phys = (u64 *)cmd->sg_frame_phys_addr;
>+ ptr_sgl_phys = cmd->sg_frame_phys_addr;
> memset(ptr_sgl, 0, instance->max_chain_frame_sz);
>
> /* Build chain frame element which holds all prps except first*/
> main_chain_element = (struct MPI25_IEEE_SGE_CHAIN64 *)
> ((u8 *)sgl_ptr + sizeof(struct MPI25_IEEE_SGE_CHAIN64));
>
>- main_chain_element->Address =
cpu_to_le64((uintptr_t)ptr_sgl_phys);
>+ main_chain_element->Address = cpu_to_le64(ptr_sgl_phys);
> main_chain_element->NextChainOffset = 0;
> main_chain_element->Flags = IEEE_SGE_FLAGS_CHAIN_ELEMENT |
> IEEE_SGE_FLAGS_SYSTEM_ADDR |
>@@ -1696,16 +1697,15 @@ megasas_make_prp_nvme(struct megasas_instance
>*instance, struct scsi_cmnd *scmd,
> scmd_printk(KERN_NOTICE,
> scmd, "page boundary ptr_sgl: 0x%p\n",
> ptr_sgl);
>- ptr_sgl_phys++;
>- *ptr_sgl =
>- cpu_to_le64((uintptr_t)ptr_sgl_phys);
>+ ptr_sgl_phys += 8;
>+ *ptr_sgl = cpu_to_le64(ptr_sgl_phys);
> ptr_sgl++;
> num_prp_in_chain++;
> }
>
> *ptr_sgl = cpu_to_le64(sge_addr);
> ptr_sgl++;
>- ptr_sgl_phys++;
>+ ptr_sgl_phys += 8;
> num_prp_in_chain++;
>
> sge_addr += mr_nvme_pg_size;
Looks good. Tested and it works as expected.
Acked-by: Sumit Saxena<sumit.saxena@broadcom.com>
>--
>2.9.0
^ permalink raw reply [flat|nested] 5+ messages in thread
* RE: [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit
2017-02-14 21:21 [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit Arnd Bergmann
2017-02-15 6:21 ` Sumit Saxena
2017-02-15 13:10 ` Sumit Saxena
@ 2017-02-15 14:15 ` Sumit Saxena
2017-02-16 2:40 ` Martin K. Petersen
3 siblings, 0 replies; 5+ messages in thread
From: Sumit Saxena @ 2017-02-15 14:15 UTC (permalink / raw
To: Arnd Bergmann, James E.J. Bottomley, Martin K. Petersen
Cc: Kashyap Desai, Shivasharan Srikanteshwara, Tomas Henzl,
Hannes Reinecke, Sasikumar PC, PDL,MEGARAIDLINUX, linux-scsi,
linux-kernel
>-----Original Message-----
>From: Arnd Bergmann [mailto:arnd@arndb.de]
>Sent: Wednesday, February 15, 2017 2:52 AM
>To: James E.J. Bottomley; Martin K. Petersen
>Cc: Arnd Bergmann; Kashyap Desai; Sumit Saxena; Shivasharan S; Tomas
Henzl;
>Hannes Reinecke; Sasikumar Chandrasekaran;
>megaraidlinux.pdl@broadcom.com; linux-scsi@vger.kernel.org; linux-
>kernel@vger.kernel.org
>Subject: [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit
>
>When building with a dma_addr_t that is different from pointer size, we
get this
>warning:
>
>drivers/scsi/megaraid/megaraid_sas_fusion.c: In function
>'megasas_make_prp_nvme':
>drivers/scsi/megaraid/megaraid_sas_fusion.c:1654:17: error: cast to
pointer
>from integer of different size [-Werror=int-to-pointer-cast]
>
>It's better to not pretend that the dma address is a pointer and instead
use a
>dma_addr_t consistently.
>
>Fixes: 33203bc4d61b ("scsi: megaraid_sas: NVME fast path io support")
>Signed-off-by: Arnd Bergmann <arnd@arndb.de>
>---
> drivers/scsi/megaraid/megaraid_sas_fusion.c | 14 +++++++-------
> 1 file changed, 7 insertions(+), 7 deletions(-)
>
>diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c
>b/drivers/scsi/megaraid/megaraid_sas_fusion.c
>index 750090119f81..29650ba669da 100644
>--- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
>+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
>@@ -1619,7 +1619,8 @@ megasas_make_prp_nvme(struct megasas_instance
>*instance, struct scsi_cmnd *scmd, {
> int sge_len, offset, num_prp_in_chain = 0;
> struct MPI25_IEEE_SGE_CHAIN64 *main_chain_element, *ptr_first_sgl;
>- u64 *ptr_sgl, *ptr_sgl_phys;
>+ u64 *ptr_sgl;
>+ dma_addr_t ptr_sgl_phys;
> u64 sge_addr;
> u32 page_mask, page_mask_result;
> struct scatterlist *sg_scmd;
>@@ -1651,14 +1652,14 @@ megasas_make_prp_nvme(struct megasas_instance
>*instance, struct scsi_cmnd *scmd,
> */
> page_mask = mr_nvme_pg_size - 1;
> ptr_sgl = (u64 *)cmd->sg_frame;
>- ptr_sgl_phys = (u64 *)cmd->sg_frame_phys_addr;
>+ ptr_sgl_phys = cmd->sg_frame_phys_addr;
> memset(ptr_sgl, 0, instance->max_chain_frame_sz);
>
> /* Build chain frame element which holds all prps except first*/
> main_chain_element = (struct MPI25_IEEE_SGE_CHAIN64 *)
> ((u8 *)sgl_ptr + sizeof(struct MPI25_IEEE_SGE_CHAIN64));
>
>- main_chain_element->Address =
cpu_to_le64((uintptr_t)ptr_sgl_phys);
>+ main_chain_element->Address = cpu_to_le64(ptr_sgl_phys);
> main_chain_element->NextChainOffset = 0;
> main_chain_element->Flags = IEEE_SGE_FLAGS_CHAIN_ELEMENT |
> IEEE_SGE_FLAGS_SYSTEM_ADDR |
>@@ -1696,16 +1697,15 @@ megasas_make_prp_nvme(struct megasas_instance
>*instance, struct scsi_cmnd *scmd,
> scmd_printk(KERN_NOTICE,
> scmd, "page boundary ptr_sgl: 0x%p\n",
> ptr_sgl);
>- ptr_sgl_phys++;
>- *ptr_sgl =
>- cpu_to_le64((uintptr_t)ptr_sgl_phys);
>+ ptr_sgl_phys += 8;
>+ *ptr_sgl = cpu_to_le64(ptr_sgl_phys);
> ptr_sgl++;
> num_prp_in_chain++;
> }
>
> *ptr_sgl = cpu_to_le64(sge_addr);
> ptr_sgl++;
>- ptr_sgl_phys++;
>+ ptr_sgl_phys += 8;
> num_prp_in_chain++;
>
> sge_addr += mr_nvme_pg_size;
Patch looks good. In last reply, Acked-by tag was not in proper format.
Fixing it now. Sorry for inconvenience.
Acked-by: Sumit Saxena <sumit.saxena@broadcom.com>
>--
>2.9.0
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit
2017-02-14 21:21 [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit Arnd Bergmann
` (2 preceding siblings ...)
2017-02-15 14:15 ` Sumit Saxena
@ 2017-02-16 2:40 ` Martin K. Petersen
3 siblings, 0 replies; 5+ messages in thread
From: Martin K. Petersen @ 2017-02-16 2:40 UTC (permalink / raw
To: Arnd Bergmann
Cc: James E.J. Bottomley, Martin K. Petersen, Kashyap Desai,
Sumit Saxena, Shivasharan S, Tomas Henzl, Hannes Reinecke,
Sasikumar Chandrasekaran, megaraidlinux.pdl, linux-scsi,
linux-kernel
>>>>> "Arnd" == Arnd Bergmann <arnd@arndb.de> writes:
Arnd> When building with a dma_addr_t that is different from pointer
Arnd> size, we get this warning:
Applied to 4.11/scsi-queue.
--
Martin K. Petersen Oracle Linux Engineering
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2017-02-16 2:41 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-02-14 21:21 [PATCH] scsi: megaraid_sas: handle dma_addr_t right on 32-bit Arnd Bergmann
2017-02-15 6:21 ` Sumit Saxena
2017-02-15 13:10 ` Sumit Saxena
2017-02-15 14:15 ` Sumit Saxena
2017-02-16 2:40 ` Martin K. Petersen
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.