From: Jon Burgess <jburgess777@googlemail.com>
To: linux-dvb@linuxtv.org, v4l-dvb-maintainer@linuxtv.org
Cc: Markus Rechberger <mrechberger@gmail.com>,
Gregoire Favre <Gregoire.Favre@gmail.com>,
linux-kernel@vger.kernel.org, Lee Revell <rlrevell@joe-job.com>
Subject: [PATCH] Re: More than 2Gb problem (dvb related) ?
Date: Sat, 28 Apr 2007 18:17:32 +0100 [thread overview]
Message-ID: <1177780652.7152.14.camel@localhost.localdomain> (raw)
In-Reply-To: <75b66ecd0704271506x595dfb6bqe0ef4fd153818501@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 747 bytes --]
On Fri, 2007-04-27 at 18:06 -0400, Lee Revell wrote:
> On 4/27/07, Jon Burgess <jburgess777@googlemail.com> wrote:
> > Interesting - I see similar symptoms after upgrading my PC:
> > * old PC was AMD Athlon 64 3000 w/ 2GB of RAM which had no issues
> > * new PC is a Intel Core 2 Duo w/ 4GB of RAM and fails in the way you
> > describe.
>
> Driver using an incorrect DMA mask?
>
> Lee
It appears the problem is that the driver is using streamed PCI and
needs to explicitly sync the data otherwise it breaks when the SWIOTLB
is in use. A call to pci_unmap_sg() was missing too.
This patch is against the current v4l-dvb HG tree.
I've only tested the change with a budget-ci card.
Jon
Signed-off-by: Jon Burgess <jburgess777@googlemail.com>
[-- Attachment #2: dvb-dma-sync.patch --]
[-- Type: text/x-patch, Size: 2416 bytes --]
diff -r a8b2a0738f4e linux/drivers/media/common/saa7146_core.c
--- a/linux/drivers/media/common/saa7146_core.c Thu Apr 19 17:07:37 2007 -0300
+++ b/linux/drivers/media/common/saa7146_core.c Sat Apr 28 17:44:38 2007 +0100
@@ -152,6 +152,7 @@ char *saa7146_vmalloc_build_pgtable(stru
return NULL;
}
+ pt->nents = pages;
slen = pci_map_sg(pci,pt->slist,pages,PCI_DMA_FROMDEVICE);
if (0 != saa7146_pgtable_build_single(pci, pt, pt->slist, slen)) {
return NULL;
@@ -165,6 +166,7 @@ void saa7146_pgtable_free(struct pci_dev
if (NULL == pt->cpu)
return;
pci_free_consistent(pci, pt->size, pt->cpu, pt->dma);
+ pci_unmap_sg(pci, pt->slist, pt->nents, PCI_DMA_FROMDEVICE);
pt->cpu = NULL;
kfree(pt->slist);
pt->slist = NULL;
diff -r a8b2a0738f4e linux/drivers/media/dvb/ttpci/av7110.c
--- a/linux/drivers/media/dvb/ttpci/av7110.c Thu Apr 19 17:07:37 2007 -0300
+++ b/linux/drivers/media/dvb/ttpci/av7110.c Sat Apr 28 17:10:38 2007 +0100
@@ -1246,6 +1246,9 @@ static void vpeirq(unsigned long data)
if (!budget->feeding1 || (newdma == olddma))
return;
+ /* Ensure streamed PCI data is synced to CPU */
+ pci_dma_sync_sg_for_cpu(budget->dev->pci, budget->pt.slist, budget->pt.nents, PCI_DMA_FROMDEVICE);
+
#if 0 /* keep */
/* track rps1 activity */
printk("vpeirq: %02x Event Counter 1 0x%04x\n",
diff -r a8b2a0738f4e linux/drivers/media/dvb/ttpci/budget-core.c
--- a/linux/drivers/media/dvb/ttpci/budget-core.c Thu Apr 19 17:07:37 2007 -0300
+++ b/linux/drivers/media/dvb/ttpci/budget-core.c Sat Apr 28 17:09:13 2007 +0100
@@ -184,6 +184,9 @@ static void vpeirq(unsigned long data)
if (budget->feeding == 0 || newdma == olddma)
return;
+ /* Ensure streamed PCI data is synced to CPU */
+ pci_dma_sync_sg_for_cpu(budget->dev->pci, budget->pt.slist, budget->pt.nents, PCI_DMA_FROMDEVICE);
+
if (newdma > olddma) { /* no wraparound, dump olddma..newdma */
count = newdma - olddma;
dvb_dmx_swfilter_packets(&budget->demux, mem + olddma, count / 188);
diff -r a8b2a0738f4e linux/include/media/saa7146.h
--- a/linux/include/media/saa7146.h Thu Apr 19 17:07:37 2007 -0300
+++ b/linux/include/media/saa7146.h Sat Apr 28 17:27:36 2007 +0100
@@ -70,6 +70,7 @@ struct saa7146_pgtable {
unsigned long offset;
/* used for custom pagetables (used for example by budget dvb cards) */
struct scatterlist *slist;
+ int nents;
};
struct saa7146_pci_extension_data {
next prev parent reply other threads:[~2007-04-28 17:17 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-04-27 11:10 More than 2Gb problem (dvb related) ? Gregoire Favre
2007-04-27 20:46 ` Markus Rechberger
2007-04-27 21:03 ` Gregoire Favre
2007-04-27 22:00 ` [linux-dvb] " Jon Burgess
2007-04-27 22:06 ` Lee Revell
2007-04-27 22:37 ` Jon Burgess
2007-04-28 17:17 ` Jon Burgess [this message]
2007-04-28 20:14 ` [PATCH] " Jon Burgess
2007-04-29 14:07 ` Gregoire Favre
2007-04-30 0:22 ` [linux-dvb] " Oliver Endriss
2007-04-30 16:52 ` Gregoire Favre
2007-04-30 18:05 ` Jon Burgess
2007-05-01 9:39 ` Gregoire Favre
2007-05-01 9:41 ` Gregoire Favre
2007-05-01 22:11 ` Jon Burgess
2007-05-02 9:35 ` Andi Kleen
2007-05-02 12:17 ` Gregoire Favre
2007-05-05 7:36 ` Gregoire Favre
2007-05-01 15:58 ` [linux-dvb] " Oliver Endriss
2007-05-01 20:03 ` Gregoire Favre
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=1177780652.7152.14.camel@localhost.localdomain \
--to=jburgess777@googlemail.com \
--cc=Gregoire.Favre@gmail.com \
--cc=linux-dvb@linuxtv.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mrechberger@gmail.com \
--cc=rlrevell@joe-job.com \
--cc=v4l-dvb-maintainer@linuxtv.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