Skip to contentSkip to conversation

iram pool not available for MX27

DISCUSSION Chris Ruehl 2 participants 5 messages Sep 30, 2013 – Oct 29, 2013 Media (V4L/DVB)

lore
Share & tools
lore

Chris Ruehl

author
#0 lore raw

Hi Phillipp,

hope things doing OK.

I recently update to the 3.12-rc kernel and hit this problem below.

[ 3.377790] coda coda-imx27.0: iram pool not available
[ 3.383363] coda: probe of coda-imx27.0 failed with error -12

I read your comments of the patch-set using platform data rather then hard coded addresses to get the ocram from a SoC.

I checked the imx27.dtsi for the iram (coda: coda@..) definition and compare with the former hard coded address and size it matches.

My .config also has the CONFIG_OF set.

Any Idea what's go wrong?

Chris

Signature · 5 lines
GTSYS -
Unit A01, 24/F Gold King Industrial Building,
35-41 Tai Lin Pai Road, Kwai Chung,
Hong Kong
新界葵涌大連排道35-41號金基工業大廈24樓A01室
#1 lore raw

Hi Chris,

Am Montag, den 30.09.2013, 13:40 +0800 schrieb Chris Ruehl:

Quoted 13 lines

Hi Phillipp,

hope things doing OK.

I recently update to the 3.12-rc kernel and hit this problem below.

[ 3.377790] coda coda-imx27.0: iram pool not available
[ 3.383363] coda: probe of coda-imx27.0 failed with error -12

I read your comments of the patch-set using platform data rather then hard coded addresses to get
the ocram from a SoC.

Quoted 6 lines

I checked the imx27.dtsi for the iram (coda: coda@..) definition and compare with the former hard coded address and size it matches.

My .config also has the CONFIG_OF set.

Any Idea what's go wrong?

do you have the mmio-sram driver enabled (CONFIG_SRAM=y)?

regards
Philipp

author
#2 lore raw

Hi Philipp,

Quoted 19 lines · Philipp Zabel

Hi Chris,

Am Montag, den 30.09.2013, 13:40 +0800 schrieb Chris Ruehl:

Quoted 16 lines

Hi Phillipp,

hope things doing OK.

I recently update to the 3.12-rc kernel and hit this problem below.

[ 3.377790] coda coda-imx27.0: iram pool not available
[ 3.383363] coda: probe of coda-imx27.0 failed with error -12

I read your comments of the patch-set using platform data rather then hard coded addresses to get
the ocram from a SoC.

I checked the imx27.dtsi for the iram (coda: coda@..) definition and compare with the former hard coded address and size it matches.

Quoted 8 lines
Quoted 3 lines

My .config also has the CONFIG_OF set.

Any Idea what's go wrong?

do you have the mmio-sram driver enabled (CONFIG_SRAM=y)?

regards
Philipp

No, I didn't, and I found out that my device is not yet ported to use "Device Tree Support"

for the moment I will quick add the CONFIG_SRAM and see what happen, but on the long term I move my code (clone of mach-mx27ads.c) to DTS which makes absolute sense when I see how nice that code works.

Thanks for the reply!
Chris

author
#3 lore raw
Quoted 31 lines · Chris Ruehl

Hi Philipp,

Quoted 19 lines · Philipp Zabel

Hi Chris,

Am Montag, den 30.09.2013, 13:40 +0800 schrieb Chris Ruehl:

Quoted 16 lines

Hi Phillipp,

hope things doing OK.

I recently update to the 3.12-rc kernel and hit this problem below.

[ 3.377790] coda coda-imx27.0: iram pool not available
[ 3.383363] coda: probe of coda-imx27.0 failed with error -12

I read your comments of the patch-set using platform data rather then hard coded addresses to get
the ocram from a SoC.

I checked the imx27.dtsi for the iram (coda: coda@..) definition and compare with the former hard coded address and size it matches.

Quoted 8 lines
Quoted 3 lines

My .config also has the CONFIG_OF set.

Any Idea what's go wrong?

do you have the mmio-sram driver enabled (CONFIG_SRAM=y)?

regards
Philipp

Quoted 9 lines

No, I didn't, and I found out that my device is not yet ported to use "Device Tree Support"

for the moment I will quick add the CONFIG_SRAM and see what happen, but on the long term I move my code (clone of mach-mx27ads.c) to DTS which makes absolute sense when I see how nice that code works.

Thanks for the reply!
Chris

CONFIG_SRAM not solve my problem, I must port the code to Device Tree Support and call the of_ functions to make the iram config available.

thank you.
Chris

author
#4 lore raw
Quoted 38 lines · Chris Ruehl
Quoted 37 lines · Chris Ruehl

Hi Philipp,

Quoted 19 lines · Philipp Zabel

Hi Chris,

Am Montag, den 30.09.2013, 13:40 +0800 schrieb Chris Ruehl:

Quoted 16 lines

Hi Phillipp,

hope things doing OK.

I recently update to the 3.12-rc kernel and hit this problem below.

[ 3.377790] coda coda-imx27.0: iram pool not available
[ 3.383363] coda: probe of coda-imx27.0 failed with error -12

I read your comments of the patch-set using platform data rather then hard coded addresses to get
the ocram from a SoC.

I checked the imx27.dtsi for the iram (coda: coda@..) definition and compare with the former hard coded address and size it matches.

Quoted 8 lines
Quoted 3 lines

My .config also has the CONFIG_OF set.

Any Idea what's go wrong?

do you have the mmio-sram driver enabled (CONFIG_SRAM=y)?

regards
Philipp

No, I didn't, and I found out that my device is not yet ported to use "Device Tree Support"

for the moment I will quick add the CONFIG_SRAM and see what happen, but on the long term I move my code (clone of mach-mx27ads.c) to DTS which makes absolute sense when I see how nice that code works.

Quoted 11 lines
Quoted 3 lines

Thanks for the reply!
Chris

CONFIG_SRAM not solve my problem, I must port the code to Device Tree Support and call the of_ functions to make the iram config available.

thank you.
Chris

Signature · 3 lines
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Hi,

almost done but the coda works now with my board :-) and more questions coming up but 1st:

[    4.626900] udevd[717]: starting version 175
[    7.190348] coda 10023000.coda: Initialized CodaDx6.
[    7.195486] coda 10023000.coda: Firmware version: 2.2.5
[    7.313178] coda 10023000.coda: codec registered as /dev/video0
[   13.999803] EXT4-fs (mmcblk0p1): re-mounted. Opts: (null)

:-) )))

Chris