From: han.lu@intel.com
To: broonie@kernel.org, alsa-devel@alsa-project.org,
liam.r.girdwood@intel.com
Cc: "Lu, Han" <han.lu@intel.com>
Subject: [PATCH 1/1] ASoC: Intel: remove conflicts when load/unload multiple firmware images
Date: Wed, 25 Feb 2015 08:26:21 +0800 [thread overview]
Message-ID: <1424823981-14133-1-git-send-email-han.lu@intel.com> (raw)
From: "Lu, Han" <han.lu@intel.com>
Details:
1. Unload all modules on fw_list of dsp when suspend, and reload all
modules on fw_list when resume.
2. A DSP expects only one scratch, but hsw_parse_fw_image() allocates
scratch blocks for each firmware image it parses. Move the allocate function
sst_block_alloc_scratch() out of hsw_parse_fw_image() to make sure a scratch
be allocated only after all firmware images be parsed.
Signed-off-by: Lu, Han <han.lu@intel.com>
diff --git a/sound/soc/intel/sst-haswell-dsp.c b/sound/soc/intel/sst-haswell-dsp.c
index c42ffae..402b728 100644
--- a/sound/soc/intel/sst-haswell-dsp.c
+++ b/sound/soc/intel/sst-haswell-dsp.c
@@ -207,9 +207,6 @@ static int hsw_parse_fw_image(struct sst_fw *sst_fw)
module = (void *)module + sizeof(*module) + module->mod_size;
}
- /* allocate scratch mem regions */
- sst_block_alloc_scratch(dsp);
-
return 0;
}
diff --git a/sound/soc/intel/sst-haswell-ipc.c b/sound/soc/intel/sst-haswell-ipc.c
index 394af56..863a9ca 100644
--- a/sound/soc/intel/sst-haswell-ipc.c
+++ b/sound/soc/intel/sst-haswell-ipc.c
@@ -1732,6 +1732,7 @@ static void sst_hsw_drop_all(struct sst_hsw *hsw)
int sst_hsw_dsp_load(struct sst_hsw *hsw)
{
struct sst_dsp *dsp = hsw->dsp;
+ struct sst_fw *sst_fw, *t;
int ret;
dev_dbg(hsw->dev, "loading audio DSP....");
@@ -1748,12 +1749,17 @@ int sst_hsw_dsp_load(struct sst_hsw *hsw)
return ret;
}
- ret = sst_fw_reload(hsw->sst_fw);
- if (ret < 0) {
- dev_err(hsw->dev, "error: SST FW reload failed\n");
- sst_dsp_dma_put_channel(dsp);
- return -ENOMEM;
+ list_for_each_entry_safe_reverse(sst_fw, t, &dsp->fw_list, list) {
+ ret = sst_fw_reload(sst_fw);
+ if (ret < 0) {
+ dev_err(hsw->dev, "error: SST FW reload failed\n");
+ sst_dsp_dma_put_channel(dsp);
+ return -ENOMEM;
+ }
}
+ ret = sst_block_alloc_scratch(hsw->dsp);
+ if (ret < 0)
+ return -EINVAL;
sst_dsp_dma_put_channel(dsp);
return 0;
@@ -1809,12 +1815,17 @@ int sst_hsw_dsp_runtime_suspend(struct sst_hsw *hsw)
int sst_hsw_dsp_runtime_sleep(struct sst_hsw *hsw)
{
- sst_fw_unload(hsw->sst_fw);
- sst_block_free_scratch(hsw->dsp);
+ struct sst_fw *sst_fw, *t;
+ struct sst_dsp *dsp = hsw->dsp;
+
+ list_for_each_entry_safe(sst_fw, t, &dsp->fw_list, list) {
+ sst_fw_unload(sst_fw);
+ }
+ sst_block_free_scratch(dsp);
hsw->boot_complete = false;
- sst_dsp_sleep(hsw->dsp);
+ sst_dsp_sleep(dsp);
return 0;
}
@@ -1943,6 +1954,11 @@ int sst_hsw_dsp_init(struct device *dev, struct sst_pdata *pdata)
goto fw_err;
}
+ /* allocate scratch mem regions */
+ ret = sst_block_alloc_scratch(hsw->dsp);
+ if (ret < 0)
+ goto boot_err;
+
/* wait for DSP boot completion */
sst_dsp_boot(hsw->dsp);
ret = wait_event_timeout(hsw->boot_wait, hsw->boot_complete,
--
1.9.1
next reply other threads:[~2015-02-25 0:28 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-02-25 0:26 han.lu [this message]
2015-03-06 21:08 ` [PATCH 1/1] ASoC: Intel: remove conflicts when load/unload multiple firmware images Mark Brown
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=1424823981-14133-1-git-send-email-han.lu@intel.com \
--to=han.lu@intel.com \
--cc=alsa-devel@alsa-project.org \
--cc=broonie@kernel.org \
--cc=liam.r.girdwood@intel.com \
/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