linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: dri-devel@lists.freedesktop.org, thomas@shipmail.org,
	airlied@linux.ie, linux-kernel@vger.kernel.org
Cc: konrad@darnok.org, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Subject: [PATCH 2/5] tm: Utilize the dma_addr_t array for pages that are to in DMA32 pool.
Date: Fri,  7 Jan 2011 12:11:41 -0500	[thread overview]
Message-ID: <1294420304-24811-3-git-send-email-konrad.wilk@oracle.com> (raw)
In-Reply-To: <1294420304-24811-1-git-send-email-konrad.wilk@oracle.com>

We only use the "if (pool == NULL)" path for right now.

Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
---
 drivers/gpu/drm/ttm/ttm_page_alloc.c |   26 +++++++++++++++++++++++---
 1 files changed, 23 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc.c b/drivers/gpu/drm/ttm/ttm_page_alloc.c
index 6859288..5d09677 100644
--- a/drivers/gpu/drm/ttm/ttm_page_alloc.c
+++ b/drivers/gpu/drm/ttm/ttm_page_alloc.c
@@ -683,14 +683,22 @@ int ttm_get_pages(struct list_head *pages, int flags,
 			gfp_flags |= GFP_HIGHUSER;
 
 		for (r = 0; r < count; ++r) {
-			p = alloc_page(gfp_flags);
+			if ((flags & TTM_PAGE_FLAG_DMA32) && dma_address) {
+				void *addr;
+				addr = dma_alloc_coherent(NULL, PAGE_SIZE,
+							&dma_address[r],
+							gfp_flags);
+				if (addr == NULL)
+					return -ENOMEM;
+				p = virt_to_page(addr);
+			} else
+				p = alloc_page(gfp_flags);
 			if (!p) {
 
 				printk(KERN_ERR TTM_PFX
 				       "Unable to allocate page.");
 				return -ENOMEM;
 			}
-
 			list_add(&p->lru, pages);
 		}
 		return 0;
@@ -739,12 +747,24 @@ void ttm_put_pages(struct list_head *pages, unsigned page_count, int flags,
 	unsigned long irq_flags;
 	struct ttm_page_pool *pool = ttm_get_pool(flags, cstate);
 	struct page *p, *tmp;
+	unsigned r;
 
 	if (pool == NULL) {
 		/* No pool for this memory type so free the pages */
 
+		r = page_count-1;
 		list_for_each_entry_safe(p, tmp, pages, lru) {
-			__free_page(p);
+			if ((flags & TTM_PAGE_FLAG_DMA32) && dma_address) {
+				void *addr = page_address(p);
+				WARN_ON(!addr || !dma_address[r]);
+				if (addr)
+					dma_free_coherent(NULL, PAGE_SIZE,
+							addr,
+							dma_address[r]);
+				dma_address[r] = 0;
+			} else
+				__free_page(p);
+			r--;
 		}
 		/* Make the pages list empty */
 		INIT_LIST_HEAD(pages);
-- 
1.7.1


  parent reply	other threads:[~2011-01-07 17:14 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-01-07 17:11 [RFC PATCH v2] Utilize the PCI API in the TTM framework Konrad Rzeszutek Wilk
2011-01-07 17:11 ` [PATCH 1/5] ttm: Introduce a placeholder for DMA (bus) addresses Konrad Rzeszutek Wilk
2011-01-07 17:11 ` Konrad Rzeszutek Wilk [this message]
2011-01-07 17:11 ` [PATCH 3/5] ttm: Expand (*populate) to support an array of DMA addresses Konrad Rzeszutek Wilk
2011-01-07 17:11 ` [PATCH 4/5] radeon/ttm/PCIe: Use dma_addr if TTM has set it Konrad Rzeszutek Wilk
2011-01-27 21:20   ` Konrad Rzeszutek Wilk
2011-01-28 14:42     ` Jerome Glisse
2011-01-28 15:03       ` Konrad Rzeszutek Wilk
2011-02-16 15:54       ` Konrad Rzeszutek Wilk
2011-02-16 18:51         ` Jerome Glisse
2011-01-07 17:11 ` [PATCH 5/5] nouveau/ttm/PCIe: " Konrad Rzeszutek Wilk
2011-01-27 21:22   ` Konrad Rzeszutek Wilk
2011-01-07 22:21 ` [RFC PATCH v2] Utilize the PCI API in the TTM framework Ian Campbell
2011-01-08 10:41 ` Thomas Hellstrom
2011-01-10 14:25 ` Thomas Hellstrom
2011-01-10 15:21   ` Konrad Rzeszutek Wilk
2011-01-10 15:58     ` Thomas Hellstrom
2011-01-10 16:45       ` Konrad Rzeszutek Wilk
2011-01-10 20:50         ` Thomas Hellstrom
2011-01-11 15:55           ` Konrad Rzeszutek Wilk
2011-01-11 16:21             ` Alex Deucher
2011-01-11 16:59               ` Konrad Rzeszutek Wilk
2011-01-11 18:12                 ` Alex Deucher
2011-01-11 18:28                   ` Konrad Rzeszutek Wilk
2011-01-11 19:28                     ` Alex Deucher
2011-01-12  9:12             ` Thomas Hellstrom
2011-01-12 15:19               ` Konrad Rzeszutek Wilk
2011-01-24 14:49                 ` Konrad Rzeszutek Wilk
2011-03-21 13:11 ` Michel Dänzer
2011-03-21 23:18   ` Konrad Rzeszutek Wilk
2011-03-22 13:13     ` Michel Dänzer
2011-03-22 14:54       ` Konrad Rzeszutek Wilk
2011-03-22 15:10         ` Michel Dänzer

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=1294420304-24811-3-git-send-email-konrad.wilk@oracle.com \
    --to=konrad.wilk@oracle.com \
    --cc=airlied@linux.ie \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=konrad@darnok.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=thomas@shipmail.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 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).