This series [1] reducing the kmalloc() minimum alignment on arm64 to 8
(from 128). In libsas, this will cause SMP requests to be 8-byte-aligned
through kmalloc() allocation. However, for the hisi_sas hardware, all
commands address must be 16-byte-aligned. Otherwise, the commands fail to
be executed.
So use 16B as the alignment for SMP request.
Link: https://lkml.kernel.org/r/[email protected] [1]
Signed-off-by: Yihang Li <[email protected]>
---
Changes since v2:
- Use 16B as alignment for SMP requests instead of ARCH_DMA_MINALIGN.
Changes since v1:
- Directly modify alloc_smp_req() instead of using handler callback.
---
drivers/scsi/libsas/sas_expander.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c
index a2204674b680..5ddbd00d5c76 100644
--- a/drivers/scsi/libsas/sas_expander.c
+++ b/drivers/scsi/libsas/sas_expander.c
@@ -135,7 +135,7 @@ static int smp_execute_task(struct domain_device *dev, void *req, int req_size,
static inline void *alloc_smp_req(int size)
{
- u8 *p = kzalloc(size, GFP_KERNEL);
+ u8 *p = kzalloc(ALIGN(size, 16), GFP_KERNEL);
if (p)
p[0] = SMP_REQUEST;
return p;
--
2.33.0
On 3/28/24 15:26, Yihang Li wrote:
> This series [1] reducing the kmalloc() minimum alignment on arm64 to 8
> (from 128). In libsas, this will cause SMP requests to be 8-byte-aligned
> through kmalloc() allocation. However, for the hisi_sas hardware, all
> commands address must be 16-byte-aligned. Otherwise, the commands fail to
> be executed.
>
> So use 16B as the alignment for SMP request.
>
> Link: https://lkml.kernel.org/r/[email protected] [1]
> Signed-off-by: Yihang Li <[email protected]>
Looks OK to me.
Reviewed-by: Damien Le Moal <[email protected]>
> ---
> Changes since v2:
> - Use 16B as alignment for SMP requests instead of ARCH_DMA_MINALIGN.
>
> Changes since v1:
> - Directly modify alloc_smp_req() instead of using handler callback.
> ---
> drivers/scsi/libsas/sas_expander.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c
> index a2204674b680..5ddbd00d5c76 100644
> --- a/drivers/scsi/libsas/sas_expander.c
> +++ b/drivers/scsi/libsas/sas_expander.c
> @@ -135,7 +135,7 @@ static int smp_execute_task(struct domain_device *dev, void *req, int req_size,
>
> static inline void *alloc_smp_req(int size)
> {
> - u8 *p = kzalloc(size, GFP_KERNEL);
> + u8 *p = kzalloc(ALIGN(size, 16), GFP_KERNEL);
> if (p)
> p[0] = SMP_REQUEST;
> return p;
--
Damien Le Moal
Western Digital Research
> + u8 *p = kzalloc(ALIGN(size, 16), GFP_KERNEL);
No, a magic undocumented 16 here max no sense at all. See the
discussion on the previous iteration.