From 22eca6468b4aea47c783770ec0739d1e13bf3bfc Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Fri, 10 Feb 2017 12:16:12 +0100
Subject: [PATCH] pcm: dmix: Allow disabling x86 optimizations

The dmix plugin has some optimized implementations for x86 using the
direct memory accesses, which was rather the original version, in
addition to the "generic" implementation using the semaphore
blocking.  The x86 implementation relies on the memory coherency *and*
the fast read/write on it.

For other architectures, this has been always disabled just because of
memory coherency.  But, the recent LPE audio development revealed
that, even on x86 platforms, the read/write performance might become
extremely bad when the buffer is marked as uncached.  Some drivers
already know the buffer is uncached, we need to switch to the generic
mode in such a case.

This patch introduces yet another flag to dmix configuration,
direct_memory_access, that indicates whether the x86-specific
optimization can be used or not.  Each driver can set the flag in its
cards config namespace, and the default dmix config refers to it.

As of this patch, only HDMI LPE Audio driver sets it.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 src/conf/cards/HdmiLpeAudio.conf |  3 +++
 src/conf/pcm/dmix.conf           | 15 +++++++++++++++
 src/pcm/pcm_direct.c             |  8 ++++++++
 src/pcm/pcm_direct.h             |  2 ++
 src/pcm/pcm_dmix.c               |  1 +
 src/pcm/pcm_dmix_i386.c          |  5 +++++
 src/pcm/pcm_dmix_x86_64.c        |  5 +++++
 7 files changed, 39 insertions(+)

diff --git a/src/conf/cards/HdmiLpeAudio.conf b/src/conf/cards/HdmiLpeAudio.conf
index f5836c05..61bdfeae 100644
--- a/src/conf/cards/HdmiLpeAudio.conf
+++ b/src/conf/cards/HdmiLpeAudio.conf
@@ -20,6 +20,9 @@ HdmiLpeAudio.pcm.front.0 {
 	}
 }
 
+# uncached memory reads have a high penalty
+HdmiLpeAudio.dmix.direct_memory_access false
+
 # default with dmix+softvol
 HdmiLpeAudio.pcm.default {
 	@args [ CARD ]
diff --git a/src/conf/pcm/dmix.conf b/src/conf/pcm/dmix.conf
index 7d0aa015..2d3b329e 100644
--- a/src/conf/pcm/dmix.conf
+++ b/src/conf/pcm/dmix.conf
@@ -49,6 +49,21 @@ pcm.!dmix {
 		@func refer
 		name defaults.pcm.ipc_perm
 	}
+	direct_memory_access {
+		@func refer
+		name {
+			@func concat
+			strings [
+				"cards."
+				{
+					@func card_driver
+					card $CARD
+				}
+				".dmix.direct_memory_access"
+			]
+		}
+		default true
+	}
 	slave {
 		pcm {
 			type hw
diff --git a/src/pcm/pcm_direct.c b/src/pcm/pcm_direct.c
index e4bef786..364191b2 100644
--- a/src/pcm/pcm_direct.c
+++ b/src/pcm/pcm_direct.c
@@ -1861,6 +1861,7 @@ int snd_pcm_direct_parse_open_conf(snd_config_t *root, snd_config_t *conf,
 	rec->slowptr = 1;
 	rec->max_periods = 0;
 	rec->var_periodsize = 1;
+	rec->direct_memory_access = 1;
 
 	/* read defaults */
 	if (snd_config_search(root, "defaults.pcm.dmix_max_periods", &n) >= 0) {
@@ -1974,6 +1975,13 @@ int snd_pcm_direct_parse_open_conf(snd_config_t *root, snd_config_t *conf,
 			rec->var_periodsize = err;
 			continue;
 		}
+		if (strcmp(id, "direct_memory_access") == 0) {
+			err = snd_config_get_bool(n);
+			if (err < 0)
+				return err;
+			rec->direct_memory_access = err;
+			continue;
+		}
 		SNDERR("Unknown field %s", id);
 		return -EINVAL;
 	}
diff --git a/src/pcm/pcm_direct.h b/src/pcm/pcm_direct.h
index 66107ec0..fba55fdc 100644
--- a/src/pcm/pcm_direct.h
+++ b/src/pcm/pcm_direct.h
@@ -159,6 +159,7 @@ struct snd_pcm_direct {
 	unsigned int channels;		/* client's channels */
 	unsigned int *bindings;
 	unsigned int recoveries;	/* mirror of executed recoveries on slave */
+	int direct_memory_access;	/* use arch-optimized buffer RW */
 	union {
 		struct {
 			int shmid_sum;			/* IPC global sum ring buffer memory identification */
@@ -340,6 +341,7 @@ struct snd_pcm_direct_open_conf {
 	int slowptr;
 	int max_periods;
 	int var_periodsize;
+	int direct_memory_access;
 	snd_config_t *slave;
 	snd_config_t *bindings;
 };
diff --git a/src/pcm/pcm_dmix.c b/src/pcm/pcm_dmix.c
index 2ef61599..dd0356e9 100644
--- a/src/pcm/pcm_dmix.c
+++ b/src/pcm/pcm_dmix.c
@@ -1065,6 +1065,7 @@ int snd_pcm_dmix_open(snd_pcm_t **pcmp, const char *name,
 	dmix->max_periods = opts->max_periods;
 	dmix->var_periodsize = opts->var_periodsize;
 	dmix->sync_ptr = snd_pcm_dmix_sync_ptr;
+	dmix->direct_memory_access = opts->direct_memory_access;
 
  retry:
 	if (first_instance) {
diff --git a/src/pcm/pcm_dmix_i386.c b/src/pcm/pcm_dmix_i386.c
index dcc6b9a4..1ab983a8 100644
--- a/src/pcm/pcm_dmix_i386.c
+++ b/src/pcm/pcm_dmix_i386.c
@@ -87,6 +87,11 @@ static void mix_select_callbacks(snd_pcm_direct_t *dmix)
 {
 	static int smp = 0, mmx = 0, cmov = 0;
 
+	if (!dmix->direct_memory_access) {
+		generic_mix_select_callbacks(dmix);
+		return;
+	}
+
 	if (!((1ULL<< dmix->shmptr->s.format) & i386_dmix_supported_format)) {
 		generic_mix_select_callbacks(dmix);
 		return;
diff --git a/src/pcm/pcm_dmix_x86_64.c b/src/pcm/pcm_dmix_x86_64.c
index 831046d9..34c40d4e 100644
--- a/src/pcm/pcm_dmix_x86_64.c
+++ b/src/pcm/pcm_dmix_x86_64.c
@@ -70,6 +70,11 @@ static void mix_select_callbacks(snd_pcm_direct_t *dmix)
 {
 	static int smp = 0;
 	
+	if (!dmix->direct_memory_access) {
+		generic_mix_select_callbacks(dmix);
+		return;
+	}
+
 	if (!((1ULL<< dmix->shmptr->s.format) & x86_64_dmix_supported_format)) {
 		generic_mix_select_callbacks(dmix);
 		return;
-- 
2.47.1