Netdev Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH][v3] i40e: optimise prefetch page refcount
@ 2020-08-18 7:07 Li RongQing
2020-08-18 18:19 ` Jesse Brandeburg
2020-09-04 23:45 ` Brown, Aaron F
0 siblings, 2 replies; 3+ messages in thread
From: Li RongQing @ 2020-08-18 7:07 UTC (permalink / raw)
To: netdev, intel-wired-lan
refcount of rx_buffer page will be added here originally, so prefetchw
is needed, but after commit 1793668c3b8c ("i40e/i40evf: Update code to
better handle incrementing page count"), and refcount is not added
everytime, so change prefetchw as prefetch,
now it mainly services page_address(), but which accesses struct page
only when WANT_PAGE_VIRTUAL or HASHED_PAGE_VIRTUAL is defined otherwise
it returns address based on offset, so we prefetch it conditionally
Jakub suggested to define prefetch_page_address in a common header
Reported-by: kernel test robot <lkp@intel.com>
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Li RongQing <lirongqing@baidu.com>
---
diff with v2: fix a build warning -Wvisibility
diff with v1: create a common function prefetch_page_address
drivers/net/ethernet/intel/i40e/i40e_txrx.c | 2 +-
include/linux/prefetch.h | 8 ++++++++
2 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 62f5b2d35f63..5f9fe55bb66d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1953,7 +1953,7 @@ static struct i40e_rx_buffer *i40e_get_rx_buffer(struct i40e_ring *rx_ring,
struct i40e_rx_buffer *rx_buffer;
rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_clean);
- prefetchw(rx_buffer->page);
+ prefetch_page_address(rx_buffer->page);
/* we are reusing so sync this buffer for CPU use */
dma_sync_single_range_for_cpu(rx_ring->dev,
diff --git a/include/linux/prefetch.h b/include/linux/prefetch.h
index 13eafebf3549..b83a3f944f28 100644
--- a/include/linux/prefetch.h
+++ b/include/linux/prefetch.h
@@ -15,6 +15,7 @@
#include <asm/processor.h>
#include <asm/cache.h>
+struct page;
/*
prefetch(x) attempts to pre-emptively get the memory pointed to
by address "x" into the CPU L1 cache.
@@ -62,4 +63,11 @@ static inline void prefetch_range(void *addr, size_t len)
#endif
}
+static inline void prefetch_page_address(struct page *page)
+{
+#if defined(WANT_PAGE_VIRTUAL) || defined(HASHED_PAGE_VIRTUAL)
+ prefetch(page);
+#endif
+}
+
#endif
--
2.16.2
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH][v3] i40e: optimise prefetch page refcount
2020-08-18 7:07 [PATCH][v3] i40e: optimise prefetch page refcount Li RongQing
@ 2020-08-18 18:19 ` Jesse Brandeburg
2020-09-04 23:45 ` Brown, Aaron F
1 sibling, 0 replies; 3+ messages in thread
From: Jesse Brandeburg @ 2020-08-18 18:19 UTC (permalink / raw)
To: Li RongQing; +Cc: netdev, intel-wired-lan
On Tue, 18 Aug 2020 15:07:57 +0800
Li RongQing <lirongqing@baidu.com> wrote:
> refcount of rx_buffer page will be added here originally, so prefetchw
> is needed, but after commit 1793668c3b8c ("i40e/i40evf: Update code to
> better handle incrementing page count"), and refcount is not added
> everytime, so change prefetchw as prefetch,
>
> now it mainly services page_address(), but which accesses struct page
> only when WANT_PAGE_VIRTUAL or HASHED_PAGE_VIRTUAL is defined
> otherwise it returns address based on offset, so we prefetch it
> conditionally
>
> Jakub suggested to define prefetch_page_address in a common header
>
> Reported-by: kernel test robot <lkp@intel.com>
> Suggested-by: Jakub Kicinski <kuba@kernel.org>
> Signed-off-by: Li RongQing <lirongqing@baidu.com>
This change looks fine to me, hopefully the more heavyweight
prefetch instruction doesn't impact performance. Were you able to test
any performance?
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
^ permalink raw reply [flat|nested] 3+ messages in thread
* RE: [PATCH][v3] i40e: optimise prefetch page refcount
2020-08-18 7:07 [PATCH][v3] i40e: optimise prefetch page refcount Li RongQing
2020-08-18 18:19 ` Jesse Brandeburg
@ 2020-09-04 23:45 ` Brown, Aaron F
1 sibling, 0 replies; 3+ messages in thread
From: Brown, Aaron F @ 2020-09-04 23:45 UTC (permalink / raw)
To: Li RongQing, netdev, intel-wired-lan
> From: netdev-owner@vger.kernel.org <netdev-owner@vger.kernel.org>
> On Behalf Of Li RongQing
> Sent: Tuesday, August 18, 2020 12:08 AM
> To: netdev@vger.kernel.org; intel-wired-lan@lists.osuosl.org
> Subject: [PATCH][v3] i40e: optimise prefetch page refcount
>
> refcount of rx_buffer page will be added here originally, so prefetchw
> is needed, but after commit 1793668c3b8c ("i40e/i40evf: Update code to
> better handle incrementing page count"), and refcount is not added
> everytime, so change prefetchw as prefetch,
>
> now it mainly services page_address(), but which accesses struct page
> only when WANT_PAGE_VIRTUAL or HASHED_PAGE_VIRTUAL is defined
> otherwise
> it returns address based on offset, so we prefetch it conditionally
>
> Jakub suggested to define prefetch_page_address in a common header
>
> Reported-by: kernel test robot <lkp@intel.com>
> Suggested-by: Jakub Kicinski <kuba@kernel.org>
> Signed-off-by: Li RongQing <lirongqing@baidu.com>
> ---
> diff with v2: fix a build warning -Wvisibility
> diff with v1: create a common function prefetch_page_address
> drivers/net/ethernet/intel/i40e/i40e_txrx.c | 2 +-
> include/linux/prefetch.h | 8 ++++++++
> 2 files changed, 9 insertions(+), 1 deletion(-)
>
Ignore my prior mail on the v2 for this patch. I really was using v3.
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2020-09-04 23:45 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-08-18 7:07 [PATCH][v3] i40e: optimise prefetch page refcount Li RongQing
2020-08-18 18:19 ` Jesse Brandeburg
2020-09-04 23:45 ` Brown, Aaron F
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).