linux-media.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 0/2] ngene CI TS fixup
@ 2018-03-06 17:46 Daniel Scheller
  2018-03-06 17:46 ` [PATCH 1/2] [media] ngene: move the tsin_exchange() stripcopy block into a function Daniel Scheller
  2018-03-06 17:46 ` [PATCH 2/2] [media] ngene: compensate for TS buffer offset shifts Daniel Scheller
  0 siblings, 2 replies; 3+ messages in thread
From: Daniel Scheller @ 2018-03-06 17:46 UTC (permalink / raw)
  To: linux-media, mchehab, mchehab

From: Daniel Scheller <d.scheller@gmx.net>

Remainder of the ngene updates series, taking care of the TS buffer offset
shift which occurs in conjunction with CICAM hardware during CAM inits.

Changed as requested to get rid of the DEBUG_CI_XFER ifdeffery.

This might aswell be considered a v3 of [1], but as 10 of 12 patches were
merged already, these two patches can be treated separate.

[1] https://www.spinics.net/lists/linux-media/msg129606.html

Daniel Scheller (2):
  [media] ngene: move the tsin_exchange() stripcopy block into a
    function
  [media] ngene: compensate for TS buffer offset shifts

 drivers/media/pci/ngene/ngene-dvb.c | 126 ++++++++++++++++++++++++++++--------
 drivers/media/pci/ngene/ngene.h     |   3 +
 2 files changed, 102 insertions(+), 27 deletions(-)

-- 
2.16.1

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [PATCH 1/2] [media] ngene: move the tsin_exchange() stripcopy block into a function
  2018-03-06 17:46 [PATCH 0/2] ngene CI TS fixup Daniel Scheller
@ 2018-03-06 17:46 ` Daniel Scheller
  2018-03-06 17:46 ` [PATCH 2/2] [media] ngene: compensate for TS buffer offset shifts Daniel Scheller
  1 sibling, 0 replies; 3+ messages in thread
From: Daniel Scheller @ 2018-03-06 17:46 UTC (permalink / raw)
  To: linux-media, mchehab, mchehab

From: Daniel Scheller <d.scheller@gmx.net>

Move the copy logic that will skip previously inserted TS NULL frames when
moving data to the DVB ring buffers into an own function. This is done to
not duplicate code all over the place with the following TS offset shift
fixup patch.

While we're touching this part of the code, get rid of the DEBUG_CI_XFER
debug-ifdeffery. This could be toggleable either by a Kconfig or a module
param, but in the end this will accidentally be enabled and cause lots
of kernel log messages, and such devel debug shouldn't be there anyway.

Signed-off-by: Daniel Scheller <d.scheller@gmx.net>
---
Changed since https://www.spinics.net/lists/linux-media/msg129616.html:
* DEBUG_CI_XFER ifdeffery removed

 drivers/media/pci/ngene/ngene-dvb.c | 39 +++++++++++--------------------------
 1 file changed, 11 insertions(+), 28 deletions(-)

diff --git a/drivers/media/pci/ngene/ngene-dvb.c b/drivers/media/pci/ngene/ngene-dvb.c
index 65c79f1b36f7..93a07a370cfd 100644
--- a/drivers/media/pci/ngene/ngene-dvb.c
+++ b/drivers/media/pci/ngene/ngene-dvb.c
@@ -139,12 +139,15 @@ static void swap_buffer(u32 *p, u32 len)
 /* start of filler packet */
 static u8 fill_ts[] = { 0x47, 0x1f, 0xff, 0x10, TS_FILLER };
 
-/* #define DEBUG_CI_XFER */
-#ifdef DEBUG_CI_XFER
-static u32 ok;
-static u32 overflow;
-static u32 stripped;
-#endif
+static inline void tsin_copy_stripped(struct ngene *dev, void *buf)
+{
+	if (memcmp(buf, fill_ts, sizeof(fill_ts)) != 0) {
+		if (dvb_ringbuffer_free(&dev->tsin_rbuf) >= 188) {
+			dvb_ringbuffer_write(&dev->tsin_rbuf, buf, 188);
+			wake_up(&dev->tsin_rbuf.queue);
+		}
+	}
+}
 
 void *tsin_exchange(void *priv, void *buf, u32 len, u32 clock, u32 flags)
 {
@@ -157,28 +160,8 @@ void *tsin_exchange(void *priv, void *buf, u32 len, u32 clock, u32 flags)
 
 	if (dev->ci.en && chan->number == 2) {
 		while (len >= 188) {
-			if (memcmp(buf, fill_ts, sizeof fill_ts) != 0) {
-				if (dvb_ringbuffer_free(&dev->tsin_rbuf) >= 188) {
-					dvb_ringbuffer_write(&dev->tsin_rbuf, buf, 188);
-					wake_up(&dev->tsin_rbuf.queue);
-#ifdef DEBUG_CI_XFER
-					ok++;
-#endif
-				}
-#ifdef DEBUG_CI_XFER
-				else
-					overflow++;
-#endif
-			}
-#ifdef DEBUG_CI_XFER
-			else
-				stripped++;
-
-			if (ok % 100 == 0 && overflow)
-				dev_warn(&dev->pci_dev->dev,
-					 "%s: ok %u overflow %u dropped %u\n",
-					 __func__, ok, overflow, stripped);
-#endif
+			tsin_copy_stripped(dev, buf);
+
 			buf += 188;
 			len -= 188;
 		}
-- 
2.16.1

^ permalink raw reply related	[flat|nested] 3+ messages in thread

* [PATCH 2/2] [media] ngene: compensate for TS buffer offset shifts
  2018-03-06 17:46 [PATCH 0/2] ngene CI TS fixup Daniel Scheller
  2018-03-06 17:46 ` [PATCH 1/2] [media] ngene: move the tsin_exchange() stripcopy block into a function Daniel Scheller
@ 2018-03-06 17:46 ` Daniel Scheller
  1 sibling, 0 replies; 3+ messages in thread
From: Daniel Scheller @ 2018-03-06 17:46 UTC (permalink / raw)
  To: linux-media, mchehab, mchehab

From: Daniel Scheller <d.scheller@gmx.net>

A possible hardware bug was discovered when using CA addon hardware
attached to the ngene hardware, in that the TS input buffer much likely
will shift and thus become unaligned to 188 byte blocks (a full TS frame)
when things like CA module initialisation (which happens via differing
communication paths) take place. This causes the TS NULL removal in
tsin_exchange() to fail to detect this previously inserted data and thus
causes userspace applications to receive data they didn't sent beforehand
and ultimately cause troubles.

On driver load with an inserted CAM, buffers are fine at first (note that
the driver has to keep the communication running from/to the card by
inserting TS NULL frames, this is done in tsout_exchange() via
FillTSBuffer() - that data is simply sent back by the hardware):

  offset | 0    1   2   3   4   5 .... 188 189 190 191 192 193 .... 376
  data   | 47  1f  ff  10  6f  6f ....  47  1f  ff  10  6f  6f ....  47

After a few seconds, the CA module is recognised and initialised, which is
signalled by

  dvb_ca_en50221: dvb_ca adapter X: DVB CAM detected and initialised successfully

This is where the first shift happens (this is always four bytes), buffer
becomes like this:

  offset | 0    1   2   3   4   5 .... 188 189 190 191 192 193 .... 376
  data   | 6f  6f  6f  6f  47  1f ....  6f  6f  6f  6f  47  1f ....  6f

Next, VDR, TVHeadend or any other CI aware application is started, buffers
will shift by even more bytes. It is believed this is due to the hardware
not handling control and data bytes properly distinct, and control data
having an influence on the actual data stream, which we cannot properly
detect at the driver level.

Workaround this hardware quirk by adding a detection for the TS sync byte
0x47 before each TS frame copy, scan for a new SYNC byte and a TS NULL
packet if buffers become unaligned, take note of that offset and apply
that when copying data to the DVB ring buffers. The last <188 bytes from
the hardware buffers are stored in a temp buffer (tsin_buffer), for which
the remainder will be in the beginning of the next hardware buffer (next
iteration of tsin_exchange()). That remainder will be appended to the
temp buffer and finally sent to the DVB ring buffer. The resulting TS
stream is perfectly fine, and the TS NULL packets inserted by the driver
which are sent back are properly removed. The resulting offset is being
clamped to 188 byte segments (one TS packet). Though this can result in
a repeated TS packet if the overall offset grows beyond this (and it
will grow only on CA initialisation), this is still way better than
unaligned TS frames and data sent to userspace that just isn't supposed
to be there.

This compensation can be toggled by the ci_tsfix modparam, which defaults
to 1 (enabled). In the case of problems, this can be turned off by setting
the parameter to 0 to restore the old behaviour.

Signed-off-by: Daniel Scheller <d.scheller@gmx.net>
---
 drivers/media/pci/ngene/ngene-dvb.c | 91 ++++++++++++++++++++++++++++++++++++-
 drivers/media/pci/ngene/ngene.h     |  3 ++
 2 files changed, 93 insertions(+), 1 deletion(-)

diff --git a/drivers/media/pci/ngene/ngene-dvb.c b/drivers/media/pci/ngene/ngene-dvb.c
index 93a07a370cfd..fee89b9ed9c1 100644
--- a/drivers/media/pci/ngene/ngene-dvb.c
+++ b/drivers/media/pci/ngene/ngene-dvb.c
@@ -38,6 +38,9 @@
 
 #include "ngene.h"
 
+static int ci_tsfix = 1;
+module_param(ci_tsfix, int, 0444);
+MODULE_PARM_DESC(ci_tsfix, "Detect and fix TS buffer offset shifs in conjunction with CI expansions (default: 1/enabled)");
 
 /****************************************************************************/
 /* COMMAND API interface ****************************************************/
@@ -139,6 +142,24 @@ static void swap_buffer(u32 *p, u32 len)
 /* start of filler packet */
 static u8 fill_ts[] = { 0x47, 0x1f, 0xff, 0x10, TS_FILLER };
 
+static int tsin_find_offset(void *buf, u32 len)
+{
+	int i, l;
+
+	l = len - sizeof(fill_ts);
+	if (l <= 0)
+		return -1;
+
+	for (i = 0; i < l; i++) {
+		if (((char *)buf)[i] == 0x47) {
+			if (!memcmp(buf + i, fill_ts, sizeof(fill_ts)))
+				return i % 188;
+		}
+	}
+
+	return -1;
+}
+
 static inline void tsin_copy_stripped(struct ngene *dev, void *buf)
 {
 	if (memcmp(buf, fill_ts, sizeof(fill_ts)) != 0) {
@@ -153,18 +174,86 @@ void *tsin_exchange(void *priv, void *buf, u32 len, u32 clock, u32 flags)
 {
 	struct ngene_channel *chan = priv;
 	struct ngene *dev = chan->dev;
-
+	int tsoff;
 
 	if (flags & DF_SWAP32)
 		swap_buffer(buf, len);
 
 	if (dev->ci.en && chan->number == 2) {
+		/* blindly copy buffers if ci_tsfix is disabled */
+		if (!ci_tsfix) {
+			while (len >= 188) {
+				tsin_copy_stripped(dev, buf);
+
+				buf += 188;
+				len -= 188;
+			}
+			return NULL;
+		}
+
+		/* ci_tsfix = 1 */
+
+		/*
+		 * since the remainder of the TS packet which got cut off
+		 * in the previous tsin_exchange() run is at the beginning
+		 * of the new TS buffer, append this to the temp buffer and
+		 * send it to the DVB ringbuffer afterwards.
+		 */
+		if (chan->tsin_offset) {
+			memcpy(&chan->tsin_buffer[(188 - chan->tsin_offset)],
+			       buf, chan->tsin_offset);
+			tsin_copy_stripped(dev, &chan->tsin_buffer);
+
+			buf += chan->tsin_offset;
+			len -= chan->tsin_offset;
+		}
+
+		/*
+		 * copy TS packets to the DVB ringbuffer and detect new offset
+		 * shifts by checking for a valid TS SYNC byte
+		 */
 		while (len >= 188) {
+			if (*((char *)buf) != 0x47) {
+				/*
+				 * no SYNC header, find new offset shift
+				 * (max. 188 bytes, tsoff will be mod 188)
+				 */
+				tsoff = tsin_find_offset(buf, len);
+				if (tsoff > 0) {
+					chan->tsin_offset += tsoff;
+					chan->tsin_offset %= 188;
+
+					buf += tsoff;
+					len -= tsoff;
+
+					dev_info(&dev->pci_dev->dev,
+						 "%s(): tsin_offset shift by %d on channel %d\n",
+						 __func__, tsoff,
+						 chan->number);
+
+					/*
+					 * offset corrected. re-check remaining
+					 * len for a full TS frame, break and
+					 * skip to fragment handling if < 188.
+					 */
+					if (len < 188)
+						break;
+				}
+			}
+
 			tsin_copy_stripped(dev, buf);
 
 			buf += 188;
 			len -= 188;
 		}
+
+		/*
+		 * if a fragment is left, copy to temp buffer. The remainder
+		 * will be appended in the next tsin_exchange() iteration.
+		 */
+		if (len > 0 && len < 188)
+			memcpy(&chan->tsin_buffer, buf, len);
+
 		return NULL;
 	}
 
diff --git a/drivers/media/pci/ngene/ngene.h b/drivers/media/pci/ngene/ngene.h
index 66d8eaa28549..01d9f1b58fcb 100644
--- a/drivers/media/pci/ngene/ngene.h
+++ b/drivers/media/pci/ngene/ngene.h
@@ -732,6 +732,9 @@ struct ngene_channel {
 #endif
 
 	int running;
+
+	int tsin_offset;
+	u8  tsin_buffer[188];
 };
 
 
-- 
2.16.1

^ permalink raw reply related	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2018-03-06 17:46 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-03-06 17:46 [PATCH 0/2] ngene CI TS fixup Daniel Scheller
2018-03-06 17:46 ` [PATCH 1/2] [media] ngene: move the tsin_exchange() stripcopy block into a function Daniel Scheller
2018-03-06 17:46 ` [PATCH 2/2] [media] ngene: compensate for TS buffer offset shifts Daniel Scheller

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).