All of lore.kernel.org
 help / color / mirror / Atom feed
From: Takashi Sakamoto <o-takashi@sakamocchi.jp>
To: Clemens Ladisch <clemens@ladisch.de>
Cc: alsa-devel@alsa-project.org,
	linux1394-devel@lists.sourceforge.net, ffado-devel@lists.sf.net
Subject: Re: [PATCH 52/52] bebob: Add support for M-Audio special Firewire series
Date: Wed, 26 Feb 2014 18:53:39 +0900	[thread overview]
Message-ID: <530DB9A3.1020403@sakamocchi.jp> (raw)
In-Reply-To: <530CF745.40309@ladisch.de>

[-- Attachment #1: Type: text/plain, Size: 1162 bytes --]

Clemens,

Thanks for your quick reply.

 > Maybe we should change how these functions handle timeouts.  Any other
> error code indicates that the lock transaction failed definitely, but
> after a timeout, we don't know if it was the request or the response
> that got lost.
>
> We could try to read the register, and, if we see that the change that we
> tried to do actually happend, assume that our driver's transaction did
> this change.

How is attached patch for this idea?

> Are read transactions more reliable than locks?  If not, we might need
> to increase the number of retries for these devices.

This quirk is applied for all of transactions to these devices. As long 
as I expeciment, their reliability roughly the same.

As long as I tested with this patch, the current number of retries (3 
times) seemd to be enough for these devices. ALSA applications can 
safely use PCM character devices.

> This patch (no. 52) does not change the CMP code.

I had no plan to touch CMP codes for this patch because just two models 
have this quirk. But now I want to add this patch into my series of patch.


Thanks

Takashi Sakamoto
o-takashi@sakamocchi.jp


[-- Attachment #2: 0001-firewire-lib-add-a-fallback-at-RCODE_CANCELLED.patch --]
[-- Type: text/x-patch, Size: 4212 bytes --]

>From e2a4091de645dddc116fe194b71ab20117a4f377 Mon Sep 17 00:00:00 2001
From: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Date: Wed, 26 Feb 2014 13:42:32 +0900
Subject: [PATCH] firewire-lib: add a fallback at RCODE_CANCELLED

I confirm that some devices often send no response against driver's
request even if the devices handle the request. In this case, drivers
should not retry the request because the devices already handled
an operation of the request.  The second request may bring errors or
make no sence.

This patch adds a flag, FW_RETURN_TIMEOUT, for snd_fw_transaction().
With this flag, this function returns -ETIMEOUT when fw_run_transaction()
returns RCODE_CALCELLED.,

For FCP, this patch modifies fcp_avc_transaction() with this flag. The
-ETIMEDOUT is handled as success.

For CMP, this patch modifies pcr_modify() with this flag. The -ETIMEOUT
generates another read transaction to check current value of PCR.
---
 sound/firewire/cmp.c | 23 +++++++++++++++++++++--
 sound/firewire/fcp.c |  5 +++--
 sound/firewire/lib.c |  6 +++++-
 sound/firewire/lib.h |  1 +
 4 files changed, 30 insertions(+), 5 deletions(-)

diff --git a/sound/firewire/cmp.c b/sound/firewire/cmp.c
index efdbf58..1fc332a 100644
--- a/sound/firewire/cmp.c
+++ b/sound/firewire/cmp.c
@@ -60,13 +60,32 @@ static int pcr_modify(struct cmp_connection *c,
 				c->resources.unit, TCODE_LOCK_COMPARE_SWAP,
 				CSR_REGISTER_BASE + CSR_IPCR(c->pcr_index),
 				buffer, 8,
-				FW_FIXED_GENERATION | c->resources.generation);
+				FW_FIXED_GENERATION | c->resources.generation |
+				FW_RETURN_TIMEOUT);
 
 		if (err < 0) {
 			if (err == -EAGAIN &&
 			    bus_reset_handling == SUCCEED_ON_BUS_RESET)
 				err = 0;
-			return err;
+
+			if (err != -ETIMEDOUT)
+				return err;
+
+			/* Check current PCR. */
+			err = snd_fw_transaction(
+				c->resources.unit, TCODE_READ_QUADLET_REQUEST,
+				CSR_REGISTER_BASE + CSR_IPCR(c->pcr_index),
+				buffer, 4,
+				FW_FIXED_GENERATION | c->resources.generation);
+			if (err < 0)
+				return err;
+
+			/* The lock transaction may be failed, retry */
+			if (buffer[0] != buffer[1]) {
+				buffer[0] = c->last_pcr_value;
+				continue;
+			}
+			break;
 		}
 
 		if (buffer[0] == old_arg) /* success? */
diff --git a/sound/firewire/fcp.c b/sound/firewire/fcp.c
index 860c080..5bb1cdf 100644
--- a/sound/firewire/fcp.c
+++ b/sound/firewire/fcp.c
@@ -90,8 +90,9 @@ int fcp_avc_transaction(struct fw_unit *unit,
 					  : TCODE_WRITE_BLOCK_REQUEST;
 		ret = snd_fw_transaction(t.unit, tcode,
 					 CSR_REGISTER_BASE + CSR_FCP_COMMAND,
-					 (void *)command, command_size, 0);
-		if (ret < 0)
+					 (void *)command, command_size,
+					 FW_RETURN_TIMEOUT);
+		if ((ret < 0) && (ret != -ETIMEDOUT))
 			break;
 
 		wait_event_timeout(t.wait, t.state != STATE_PENDING,
diff --git a/sound/firewire/lib.c b/sound/firewire/lib.c
index 7409edb..6c3278b 100644
--- a/sound/firewire/lib.c
+++ b/sound/firewire/lib.c
@@ -22,7 +22,7 @@
  * @length: length of @buffer
  * @flags: use %FW_FIXED_GENERATION and add the generation value to attempt the
  *         request only in that generation; use %FW_QUIET to suppress error
- *         messages
+ *         messages: use %FW_RETURN_TIMEOUT to avoid retry at RCODE_CANCELLED
  *
  * Submits an asynchronous request to the target device, and waits for the
  * response.  The node ID and the current generation are derived from @unit.
@@ -53,6 +53,10 @@ int snd_fw_transaction(struct fw_unit *unit, int tcode,
 		if (rcode == RCODE_GENERATION && (flags & FW_FIXED_GENERATION))
 			return -EAGAIN;
 
+		/* timeout */
+		if ((rcode == RCODE_CANCELLED) && (flags & FW_RETURN_TIMEOUT))
+			return -ETIMEDOUT;
+
 		if (rcode_is_permanent_error(rcode) || ++tries >= 3) {
 			if (!(flags & FW_QUIET))
 				dev_err(&unit->device,
diff --git a/sound/firewire/lib.h b/sound/firewire/lib.h
index 02cfabc..b8ea90b 100644
--- a/sound/firewire/lib.h
+++ b/sound/firewire/lib.h
@@ -9,6 +9,7 @@ struct fw_unit;
 #define FW_GENERATION_MASK	0x00ff
 #define FW_FIXED_GENERATION	0x0100
 #define FW_QUIET		0x0200
+#define FW_RETURN_TIMEOUT	0x0400
 
 int snd_fw_transaction(struct fw_unit *unit, int tcode,
 		       u64 offset, void *buffer, size_t length,
-- 
1.8.3.2


[-- Attachment #3: Type: text/plain, Size: 436 bytes --]

------------------------------------------------------------------------------
Flow-based real-time traffic analytics software. Cisco certified tool.
Monitor traffic, SLAs, QoS, Medianet, WAAS etc. with NetFlow Analyzer
Customize your own dashboards, set traffic alerts and generate reports.
Network behavioral analysis & security monitoring. All-in-one tool.
http://pubads.g.doubleclick.net/gampad/clk?id=126839071&iu=/4140/ostg.clktrk

[-- Attachment #4: Type: text/plain, Size: 0 bytes --]



  parent reply	other threads:[~2014-02-26  9:53 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-29 13:44 [RFC v3] [PATCH 00/52] Enhancement for support of firewire devices Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 01/52] firewire-lib: Rename functions, structure, member for AMDTP Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 02/52] firewire-lib: Add macros instead of fixed value " Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 03/52] firewire-lib: Add 'direction' member to 'amdtp_stream' structure Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 04/52] firewire-lib: Split some codes into functions to reuse for both streams Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 05/52] firewire-lib: Add support for AMDTP in-stream and PCM capture Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 06/52] firewire-lib: Add support for MIDI capture/playback Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 07/52] firewire-lib: Give syt value as parameter to handle_out_packet() Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 08/52] firewire-lib: Add support for duplex streams synchronization in blocking mode Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 09/52] firewire-lib: Add sort function for transmitted packet Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 10/52] firewire-lib: Add transfer delay to synchronized duplex streams Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 11/52] firewire-lib: Add support for channel mapping Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 12/52] firewire-lib: Rename macros, variables and functions for CMP Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 13/52] firewire-lib: Add 'direction' member to 'cmp_connection' structure Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 14/52] firewire-lib: Add handling output connection by CMP Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 15/52] firewire-lib: Add a new function to check others' connection Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 16/52] firewire-lib: Add some AV/C general commands Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 17/52] firewire-lib: Add quirks for Fireworks Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 18/52] fireworks: Add skelton for Fireworks based devices Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 19/52] fireworks: Add transaction and some commands Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 20/52] fireworks: Add connection and stream management Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 21/52] fireworks: Add proc interface for debugging purpose Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 22/52] fireworks: Add MIDI interface Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 23/52] fireworks: Add PCM interface Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 24/52] fireworks: Add hwdep interface Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 25/52] fireworks: Add command/response functionality into " Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 26/52] bebob: Add skelton for BeBoB based devices Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 27/52] bebob: Add commands and connections/streams management Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 28/52] bebob: Add proc interface for debugging purpose Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 29/52] bebob: Add MIDI interface Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 30/52] bebob: Add PCM interface Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 31/52] bebob: Add hwdep interface Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 32/52] bebob: Prepare for device specific operations Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 33/52] bebob: Add support for Terratec PHASE, EWS series and Aureon Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 34/52] bebob: Add support for Yamaha GO series Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 35/52] bebob: Add support for Focusrite Saffire/SaffirePro series Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 36/52] bebob: Add support for M-Audio usual Firewire series Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 37/52] bebob: Send a cue to load firmware for M-Audio " Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 38/52] speakers: Rename to oxfw Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 39/52] oxfw: Move to its own directory Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 40/52] oxfw: Split stream functionality to a new file and add a header file Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 41/52] oxfw: Split PCM functionality to a new file Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 42/52] oxfw: Split control " Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 43/52] oxfw: Change the way to name card Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 44/52] oxfw: Change the way to make PCM rules/constraints Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 45/52] oxfw: Add proc interface for debugging purpose Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 46/52] oxfw: Change the way to start stream Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 47/52] oxfw: Add some AV/C commands to get stream formation and supported sample rate Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 48/52] oxfw: Add support for Behringer/Mackie devices Takashi Sakamoto
2014-01-29 13:44 ` [PATCH 49/52] oxfw: Add support AMDTP in-stream and PCM capture Takashi Sakamoto
2014-01-29 14:49 ` [PATCH 50/52] oxfw: Add support for capture/playback MIDI messages Takashi Sakamoto
2014-01-29 14:49 ` [PATCH 51/52] oxfw: Add hwdep interface Takashi Sakamoto
2014-01-29 14:49 ` [PATCH 52/52] bebob: Add support for M-Audio special Firewire series Takashi Sakamoto
2014-02-25 14:02   ` Takashi Sakamoto
2014-02-25 20:04     ` Clemens Ladisch
2014-02-25 20:12       ` [FFADO-devel] [alsa-devel] " Малышев Михаил
2014-02-26  9:53       ` Takashi Sakamoto [this message]
2014-01-30  0:16 ` About impacts to user-land Takashi Sakamoto
2014-02-21 16:54 ` [FFADO-devel] [RFC v3] [PATCH 00/52] Enhancement for support of firewire devices Jay Fenlason
2014-02-21 20:28   ` Stefan Richter
2014-02-21 20:39     ` Stefan Richter
2014-02-22  2:34   ` Takashi Sakamoto
2014-02-24 17:28     ` Jay Fenlason
2014-02-25  3:41       ` Takashi Sakamoto

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=530DB9A3.1020403@sakamocchi.jp \
    --to=o-takashi@sakamocchi.jp \
    --cc=alsa-devel@alsa-project.org \
    --cc=clemens@ladisch.de \
    --cc=ffado-devel@lists.sf.net \
    --cc=linux1394-devel@lists.sourceforge.net \
    /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.