* [PATCH] dmaengine: xgene-dma: Disable memcpy operation due to performance drop
@ 2015-10-08 8:25 Rameshwar Prasad Sahu
2015-10-08 8:53 ` Arnd Bergmann
0 siblings, 1 reply; 3+ messages in thread
From: Rameshwar Prasad Sahu @ 2015-10-08 8:25 UTC (permalink / raw)
To: linux-arm-kernel
The DMA engine supports memory copy, RAID5 XOR, RAID6 PQ, and other
computations. But the bandwidth of the entire DMA engine is shared
among all channels. This patch re-configures operations availability
such that one can achieve maximum performance for XOR and PQ
computation by removing the memory offload operations.
Signed-off-by: Rameshwar Prasad Sahu <rsahu@apm.com>
---
drivers/dma/xgene-dma.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/dma/xgene-dma.c b/drivers/dma/xgene-dma.c
index 8d57b1b..2998fcb 100644
--- a/drivers/dma/xgene-dma.c
+++ b/drivers/dma/xgene-dma.c
@@ -154,6 +154,7 @@
#define XGENE_DMA_MAX_XOR_SRC 5
#define XGENE_DMA_16K_BUFFER_LEN_CODE 0x0
#define XGENE_DMA_INVALID_LEN_CODE 0x7800000000000000ULL
+#undef XGENE_DMA_MEMCPY_ENABLE
/* X-Gene DMA descriptor error codes */
#define ERR_DESC_AXI 0x01
@@ -1707,7 +1708,9 @@ static void xgene_dma_set_caps(struct xgene_dma_chan *chan,
dma_cap_zero(dma_dev->cap_mask);
/* Set DMA device capability */
+#ifdef XGENE_DMA_MEMCPY_ENABLE
dma_cap_set(DMA_MEMCPY, dma_dev->cap_mask);
+#endif
dma_cap_set(DMA_SG, dma_dev->cap_mask);
/* Basically here, the X-Gene SoC DMA engine channel 0 supports XOR
--
1.8.2.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* [PATCH] dmaengine: xgene-dma: Disable memcpy operation due to performance drop
2015-10-08 8:25 [PATCH] dmaengine: xgene-dma: Disable memcpy operation due to performance drop Rameshwar Prasad Sahu
@ 2015-10-08 8:53 ` Arnd Bergmann
2015-10-08 8:57 ` Rameshwar Sahu
0 siblings, 1 reply; 3+ messages in thread
From: Arnd Bergmann @ 2015-10-08 8:53 UTC (permalink / raw)
To: linux-arm-kernel
On Thursday 08 October 2015 13:55:16 Rameshwar Prasad Sahu wrote:
> The DMA engine supports memory copy, RAID5 XOR, RAID6 PQ, and other
> computations. But the bandwidth of the entire DMA engine is shared
> among all channels. This patch re-configures operations availability
> such that one can achieve maximum performance for XOR and PQ
> computation by removing the memory offload operations.
>
> Signed-off-by: Rameshwar Prasad Sahu <rsahu@apm.com>
> ---
> drivers/dma/xgene-dma.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/dma/xgene-dma.c b/drivers/dma/xgene-dma.c
> index 8d57b1b..2998fcb 100644
> --- a/drivers/dma/xgene-dma.c
> +++ b/drivers/dma/xgene-dma.c
> @@ -154,6 +154,7 @@
> #define XGENE_DMA_MAX_XOR_SRC 5
> #define XGENE_DMA_16K_BUFFER_LEN_CODE 0x0
> #define XGENE_DMA_INVALID_LEN_CODE 0x7800000000000000ULL
> +#undef XGENE_DMA_MEMCPY_ENABLE
>
> /* X-Gene DMA descriptor error codes */
> #define ERR_DESC_AXI 0x01
> @@ -1707,7 +1708,9 @@ static void xgene_dma_set_caps(struct xgene_dma_chan *chan,
> dma_cap_zero(dma_dev->cap_mask);
>
> /* Set DMA device capability */
> +#ifdef XGENE_DMA_MEMCPY_ENABLE
> dma_cap_set(DMA_MEMCPY, dma_dev->cap_mask);
> +#endif
> dma_cap_set(DMA_SG, dma_dev->cap_mask);
>
> /* Basically here, the X-Gene SoC DMA engine channel 0 supports XOR
I don't see what the #ifdef gains you here when the setting is
hardcoded. Why not just remove that DMA_MEMCPY capability completely
if you don't want to use it?
Arnd
^ permalink raw reply [flat|nested] 3+ messages in thread
* [PATCH] dmaengine: xgene-dma: Disable memcpy operation due to performance drop
2015-10-08 8:53 ` Arnd Bergmann
@ 2015-10-08 8:57 ` Rameshwar Sahu
0 siblings, 0 replies; 3+ messages in thread
From: Rameshwar Sahu @ 2015-10-08 8:57 UTC (permalink / raw)
To: linux-arm-kernel
On Thu, Oct 8, 2015 at 2:23 PM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Thursday 08 October 2015 13:55:16 Rameshwar Prasad Sahu wrote:
>> The DMA engine supports memory copy, RAID5 XOR, RAID6 PQ, and other
>> computations. But the bandwidth of the entire DMA engine is shared
>> among all channels. This patch re-configures operations availability
>> such that one can achieve maximum performance for XOR and PQ
>> computation by removing the memory offload operations.
>>
>> Signed-off-by: Rameshwar Prasad Sahu <rsahu@apm.com>
>> ---
>> drivers/dma/xgene-dma.c | 3 +++
>> 1 file changed, 3 insertions(+)
>>
>> diff --git a/drivers/dma/xgene-dma.c b/drivers/dma/xgene-dma.c
>> index 8d57b1b..2998fcb 100644
>> --- a/drivers/dma/xgene-dma.c
>> +++ b/drivers/dma/xgene-dma.c
>> @@ -154,6 +154,7 @@
>> #define XGENE_DMA_MAX_XOR_SRC 5
>> #define XGENE_DMA_16K_BUFFER_LEN_CODE 0x0
>> #define XGENE_DMA_INVALID_LEN_CODE 0x7800000000000000ULL
>> +#undef XGENE_DMA_MEMCPY_ENABLE
>>
>> /* X-Gene DMA descriptor error codes */
>> #define ERR_DESC_AXI 0x01
>> @@ -1707,7 +1708,9 @@ static void xgene_dma_set_caps(struct xgene_dma_chan *chan,
>> dma_cap_zero(dma_dev->cap_mask);
>>
>> /* Set DMA device capability */
>> +#ifdef XGENE_DMA_MEMCPY_ENABLE
>> dma_cap_set(DMA_MEMCPY, dma_dev->cap_mask);
>> +#endif
>> dma_cap_set(DMA_SG, dma_dev->cap_mask);
>>
>> /* Basically here, the X-Gene SoC DMA engine channel 0 supports XOR
>
> I don't see what the #ifdef gains you here when the setting is
> hardcoded. Why not just remove that DMA_MEMCPY capability completely
> if you don't want to use it?
Okay Arnd,
>
> Arnd
> --
> To unsubscribe from this list: send the line "unsubscribe dmaengine" in
> the body of a message to majordomo at vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2015-10-08 8:57 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-10-08 8:25 [PATCH] dmaengine: xgene-dma: Disable memcpy operation due to performance drop Rameshwar Prasad Sahu
2015-10-08 8:53 ` Arnd Bergmann
2015-10-08 8:57 ` Rameshwar Sahu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).