From: Konrad Rzeszutek Wilk <konrad@darnok.org>
To: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: linux-mm@kvack.org, akpm@linux-foundation.org,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Subject: Re: [mm PATCH v2 03/26] swiotlb: Add support for DMA_ATTR_SKIP_CPU_SYNC
Date: Sat, 5 Nov 2016 15:39:30 -0400 [thread overview]
Message-ID: <20161105193929.GA26349@localhost.localdomain> (raw)
In-Reply-To: <20161102111252.79519.21950.stgit@ahduyck-blue-test.jf.intel.com>
.. snip..
> @@ -561,6 +565,7 @@ void swiotlb_tbl_unmap_single(struct device *hwdev, phys_addr_t tlb_addr,
> * First, sync the memory before unmapping the entry
> */
> if (orig_addr != INVALID_PHYS_ADDR &&
> + !(attrs & DMA_ATTR_SKIP_CPU_SYNC) &&
> ((dir == DMA_FROM_DEVICE) || (dir == DMA_BIDIRECTIONAL)))
> swiotlb_bounce(orig_addr, tlb_addr, size, DMA_FROM_DEVICE);
>
> @@ -654,7 +659,8 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
> * GFP_DMA memory; fall back on map_single(), which
> * will grab memory from the lowest available address range.
> */
> - phys_addr_t paddr = map_single(hwdev, 0, size, DMA_FROM_DEVICE);
> + phys_addr_t paddr = map_single(hwdev, 0, size,
> + DMA_FROM_DEVICE, 0);
> if (paddr == SWIOTLB_MAP_ERROR)
> goto err_warn;
>
> @@ -669,7 +675,8 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
>
> /* DMA_TO_DEVICE to avoid memcpy in unmap_single */
> swiotlb_tbl_unmap_single(hwdev, paddr,
> - size, DMA_TO_DEVICE);
> + size, DMA_TO_DEVICE,
> + DMA_ATTR_SKIP_CPU_SYNC);
This I believe is redundant. That is swiotlb_tbl_unmap_single only
does an bounce if the dir is DMA_FROM_DEVICE or DMA_BIDIRECTIONAL.
I added /* optional. */
> goto err_warn;
> }
> }
> @@ -699,7 +706,7 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
> free_pages((unsigned long)vaddr, get_order(size));
> else
> /* DMA_TO_DEVICE to avoid memcpy in swiotlb_tbl_unmap_single */
> - swiotlb_tbl_unmap_single(hwdev, paddr, size, DMA_TO_DEVICE);
> + swiotlb_tbl_unmap_single(hwdev, paddr, size, DMA_TO_DEVICE, 0);
.. but here you choose to put 0? I changed that to
DMA_ATTR_SKIP_CPU_SYNC and expanded the comment above.
Time to test the patches.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Konrad Rzeszutek Wilk <konrad@darnok.org>
To: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: linux-mm@kvack.org, akpm@linux-foundation.org,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Subject: Re: [mm PATCH v2 03/26] swiotlb: Add support for DMA_ATTR_SKIP_CPU_SYNC
Date: Sat, 5 Nov 2016 15:39:30 -0400 [thread overview]
Message-ID: <20161105193929.GA26349@localhost.localdomain> (raw)
In-Reply-To: <20161102111252.79519.21950.stgit@ahduyck-blue-test.jf.intel.com>
.. snip..
> @@ -561,6 +565,7 @@ void swiotlb_tbl_unmap_single(struct device *hwdev, phys_addr_t tlb_addr,
> * First, sync the memory before unmapping the entry
> */
> if (orig_addr != INVALID_PHYS_ADDR &&
> + !(attrs & DMA_ATTR_SKIP_CPU_SYNC) &&
> ((dir == DMA_FROM_DEVICE) || (dir == DMA_BIDIRECTIONAL)))
> swiotlb_bounce(orig_addr, tlb_addr, size, DMA_FROM_DEVICE);
>
> @@ -654,7 +659,8 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
> * GFP_DMA memory; fall back on map_single(), which
> * will grab memory from the lowest available address range.
> */
> - phys_addr_t paddr = map_single(hwdev, 0, size, DMA_FROM_DEVICE);
> + phys_addr_t paddr = map_single(hwdev, 0, size,
> + DMA_FROM_DEVICE, 0);
> if (paddr == SWIOTLB_MAP_ERROR)
> goto err_warn;
>
> @@ -669,7 +675,8 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
>
> /* DMA_TO_DEVICE to avoid memcpy in unmap_single */
> swiotlb_tbl_unmap_single(hwdev, paddr,
> - size, DMA_TO_DEVICE);
> + size, DMA_TO_DEVICE,
> + DMA_ATTR_SKIP_CPU_SYNC);
This I believe is redundant. That is swiotlb_tbl_unmap_single only
does an bounce if the dir is DMA_FROM_DEVICE or DMA_BIDIRECTIONAL.
I added /* optional. */
> goto err_warn;
> }
> }
> @@ -699,7 +706,7 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
> free_pages((unsigned long)vaddr, get_order(size));
> else
> /* DMA_TO_DEVICE to avoid memcpy in swiotlb_tbl_unmap_single */
> - swiotlb_tbl_unmap_single(hwdev, paddr, size, DMA_TO_DEVICE);
> + swiotlb_tbl_unmap_single(hwdev, paddr, size, DMA_TO_DEVICE, 0);
.. but here you choose to put 0? I changed that to
DMA_ATTR_SKIP_CPU_SYNC and expanded the comment above.
Time to test the patches.
next prev parent reply other threads:[~2016-11-05 19:39 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-02 11:12 [mm PATCH v2 00/26] Add support for DMA writable pages being writable by the network stack Alexander Duyck
2016-11-02 11:12 ` Alexander Duyck
2016-11-02 11:12 ` Alexander Duyck
2016-11-02 11:12 ` [mm PATCH v2 01/26] swiotlb: Drop unused functions swiotlb_map_sg and swiotlb_unmap_sg Alexander Duyck
2016-11-02 11:12 ` Alexander Duyck
2016-11-03 14:14 ` Christoph Hellwig
2016-11-03 14:14 ` Christoph Hellwig
2016-11-03 14:29 ` Konrad Rzeszutek Wilk
2016-11-03 14:29 ` Konrad Rzeszutek Wilk
2016-11-03 14:45 ` Christoph Hellwig
2016-11-03 14:45 ` Christoph Hellwig
2016-11-03 15:02 ` Duyck, Alexander H
2016-11-03 15:02 ` Duyck, Alexander H
2016-11-02 11:12 ` [mm PATCH v2 02/26] swiotlb-xen: Enforce return of DMA_ERROR_CODE in mapping function Alexander Duyck
2016-11-02 11:12 ` Alexander Duyck
2016-11-02 11:13 ` [mm PATCH v2 03/26] swiotlb: Add support for DMA_ATTR_SKIP_CPU_SYNC Alexander Duyck
2016-11-02 11:13 ` Alexander Duyck
2016-11-05 19:39 ` Konrad Rzeszutek Wilk [this message]
2016-11-05 19:39 ` Konrad Rzeszutek Wilk
2016-11-05 23:13 ` Alexander Duyck
2016-11-05 23:13 ` Alexander Duyck
2016-11-02 11:13 ` [mm PATCH v2 04/26] arch/arc: Add option to skip sync on DMA mapping Alexander Duyck
2016-11-02 11:13 ` Alexander Duyck
2016-11-02 11:13 ` [mm PATCH v2 05/26] arch/arm: Add option to skip sync on DMA map and unmap Alexander Duyck
2016-11-02 11:13 ` Alexander Duyck
2016-11-02 11:13 ` [mm PATCH v2 06/26] arch/avr32: Add option to skip sync on DMA map Alexander Duyck
2016-11-02 11:13 ` Alexander Duyck
2016-11-02 11:13 ` [mm PATCH v2 07/26] arch/blackfin: " Alexander Duyck
2016-11-02 11:13 ` Alexander Duyck
2016-11-02 11:13 ` [mm PATCH v2 08/26] arch/c6x: Add option to skip sync on DMA map and unmap Alexander Duyck
2016-11-02 11:13 ` Alexander Duyck
2016-11-02 11:14 ` [mm PATCH v2 09/26] arch/frv: Add option to skip sync on DMA map Alexander Duyck
2016-11-02 11:14 ` Alexander Duyck
2016-11-02 11:14 ` [mm PATCH v2 10/26] arch/hexagon: Add option to skip DMA sync as a part of mapping Alexander Duyck
2016-11-02 11:14 ` Alexander Duyck
2016-11-02 11:14 ` [mm PATCH v2 11/26] arch/m68k: " Alexander Duyck
2016-11-02 11:14 ` Alexander Duyck
2016-11-02 11:14 ` Alexander Duyck
2016-11-02 11:14 ` Alexander Duyck
2016-11-02 11:14 ` [mm PATCH v2 12/26] arch/metag: Add option to skip DMA sync as a part of map and unmap Alexander Duyck
2016-11-02 11:14 ` Alexander Duyck
2016-11-02 11:14 ` [mm PATCH v2 13/26] arch/microblaze: " Alexander Duyck
2016-11-02 11:14 ` Alexander Duyck
2016-11-02 11:14 ` [mm PATCH v2 14/26] arch/mips: " Alexander Duyck
2016-11-02 11:14 ` Alexander Duyck
2016-11-02 11:14 ` [mm PATCH v2 15/26] arch/nios2: " Alexander Duyck
2016-11-02 11:14 ` Alexander Duyck
2016-11-02 11:14 ` [mm PATCH v2 16/26] arch/openrisc: Add option to skip DMA sync as a part of mapping Alexander Duyck
2016-11-02 11:14 ` Alexander Duyck
2016-11-02 11:15 ` [mm PATCH v2 17/26] arch/parisc: Add option to skip DMA sync as a part of map and unmap Alexander Duyck
2016-11-02 11:15 ` Alexander Duyck
2016-11-02 11:15 ` [mm PATCH v2 18/26] arch/powerpc: Add option to skip DMA sync as a part of mapping Alexander Duyck
2016-11-02 11:15 ` Alexander Duyck
2016-11-04 3:16 ` Michael Ellerman
2016-11-04 3:16 ` Michael Ellerman
2016-11-02 11:15 ` [mm PATCH v2 19/26] arch/sh: " Alexander Duyck
2016-11-02 11:15 ` Alexander Duyck
2016-11-02 11:15 ` Alexander Duyck
2016-11-02 11:15 ` [mm PATCH v2 20/26] arch/sparc: Add option to skip DMA sync as a part of map and unmap Alexander Duyck
2016-11-02 11:15 ` Alexander Duyck
2016-11-02 11:15 ` Alexander Duyck
2016-11-02 11:15 ` [mm PATCH v2 21/26] arch/tile: " Alexander Duyck
2016-11-02 11:15 ` Alexander Duyck
2016-11-02 11:15 ` [mm PATCH v2 22/26] arch/xtensa: Add option to skip DMA sync as a part of mapping Alexander Duyck
2016-11-02 11:15 ` Alexander Duyck
2016-11-02 11:16 ` [mm PATCH v2 23/26] dma: Add calls for dma_map_page_attrs and dma_unmap_page_attrs Alexander Duyck
2016-11-02 11:16 ` Alexander Duyck
2016-11-02 11:16 ` [mm PATCH v2 24/26] mm: Add support for releasing multiple instances of a page Alexander Duyck
2016-11-02 11:16 ` Alexander Duyck
2016-11-02 11:16 ` [mm PATCH v2 25/26] igb: Update driver to make use of DMA_ATTR_SKIP_CPU_SYNC Alexander Duyck
2016-11-02 11:16 ` Alexander Duyck
2016-11-02 11:16 ` [mm PATCH v2 26/26] igb: Update code to better handle incrementing page count Alexander Duyck
2016-11-02 11:16 ` Alexander Duyck
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20161105193929.GA26349@localhost.localdomain \
--to=konrad@darnok.org \
--cc=akpm@linux-foundation.org \
--cc=alexander.h.duyck@intel.com \
--cc=konrad.wilk@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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.