In Linux we do not have driver for TPTCs of eDMA3 since there is no need to
do any configuration within TPTC for the eDMA3 to be operational. All
configuration is via the TPCC.
To prevent the omap_device_late_idle() to disable the TPTCs when they are
no longer bind with the edma driver, the HWMOD_INIT_NO_IDLE need to be
added.
Signed-off-by: Peter Ujfalusi <[email protected]>
---
Vinod, Olof,
This patch somehow got lost in my working branch. It was mixed within the patches
I will have for 4.5 while it should have been within the new eDMA3 binding
series..
Regards,
Peter
arch/arm/mach-omap2/omap_hwmod_33xx_43xx_ipblock_data.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/arch/arm/mach-omap2/omap_hwmod_33xx_43xx_ipblock_data.c b/arch/arm/mach-omap2/omap_hwmod_33xx_43xx_ipblock_data.c
index 907a452b78ea..3c7106a09801 100644
--- a/arch/arm/mach-omap2/omap_hwmod_33xx_43xx_ipblock_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_33xx_43xx_ipblock_data.c
@@ -1169,7 +1169,8 @@ struct omap_hwmod am33xx_tptc0_hwmod = {
.name = "tptc0",
.class = &am33xx_tptc_hwmod_class,
.clkdm_name = "l3_clkdm",
- .flags = HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY,
+ .flags = (HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY |
+ HWMOD_INIT_NO_IDLE),
.main_clk = "l3_gclk",
.prcm = {
.omap4 = {
@@ -1183,7 +1184,8 @@ struct omap_hwmod am33xx_tptc1_hwmod = {
.name = "tptc1",
.class = &am33xx_tptc_hwmod_class,
.clkdm_name = "l3_clkdm",
- .flags = (HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY),
+ .flags = (HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY |
+ HWMOD_INIT_NO_IDLE),
.main_clk = "l3_gclk",
.prcm = {
.omap4 = {
@@ -1197,7 +1199,8 @@ struct omap_hwmod am33xx_tptc2_hwmod = {
.name = "tptc2",
.class = &am33xx_tptc_hwmod_class,
.clkdm_name = "l3_clkdm",
- .flags = (HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY),
+ .flags = (HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY |
+ HWMOD_INIT_NO_IDLE),
.main_clk = "l3_gclk",
.prcm = {
.omap4 = {
--
2.6.1
On 11/02/2015 12:24 PM, Vinod Koul wrote:
> On Mon, Nov 02, 2015 at 12:11:00PM +0200, Peter Ujfalusi wrote:
>> In Linux we do not have driver for TPTCs of eDMA3 since there is no need to
>> do any configuration within TPTC for the eDMA3 to be operational. All
>> configuration is via the TPCC.
>> To prevent the omap_device_late_idle() to disable the TPTCs when they are
>> no longer bind with the edma driver, the HWMOD_INIT_NO_IDLE need to be
>> added.
>>
>> Signed-off-by: Peter Ujfalusi <[email protected]>
>> ---
>> Vinod, Olof,
>>
>> This patch somehow got lost in my working branch. It was mixed within the patches
>> I will have for 4.5 while it should have been within the new eDMA3 binding
>> series..
>
> Does this fix the issue reported by Olof? Also ACK pls for this
Yes, it is fixing the issue, I have this in my branch. everything looks fine
on AM335x/AM437x/Dra7xx where I have eDMA in use.
--
P?ter
On Mon, Nov 02, 2015 at 12:11:00PM +0200, Peter Ujfalusi wrote:
> In Linux we do not have driver for TPTCs of eDMA3 since there is no need to
> do any configuration within TPTC for the eDMA3 to be operational. All
> configuration is via the TPCC.
> To prevent the omap_device_late_idle() to disable the TPTCs when they are
> no longer bind with the edma driver, the HWMOD_INIT_NO_IDLE need to be
> added.
>
> Signed-off-by: Peter Ujfalusi <[email protected]>
> ---
> Vinod, Olof,
>
> This patch somehow got lost in my working branch. It was mixed within the patches
> I will have for 4.5 while it should have been within the new eDMA3 binding
> series..
Does this fix the issue reported by Olof? Also ACK pls for this
--
~Vinod
>
> Regards,
> Peter
>
> arch/arm/mach-omap2/omap_hwmod_33xx_43xx_ipblock_data.c | 9 ++++++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/arch/arm/mach-omap2/omap_hwmod_33xx_43xx_ipblock_data.c b/arch/arm/mach-omap2/omap_hwmod_33xx_43xx_ipblock_data.c
> index 907a452b78ea..3c7106a09801 100644
> --- a/arch/arm/mach-omap2/omap_hwmod_33xx_43xx_ipblock_data.c
> +++ b/arch/arm/mach-omap2/omap_hwmod_33xx_43xx_ipblock_data.c
> @@ -1169,7 +1169,8 @@ struct omap_hwmod am33xx_tptc0_hwmod = {
> .name = "tptc0",
> .class = &am33xx_tptc_hwmod_class,
> .clkdm_name = "l3_clkdm",
> - .flags = HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY,
> + .flags = (HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY |
> + HWMOD_INIT_NO_IDLE),
> .main_clk = "l3_gclk",
> .prcm = {
> .omap4 = {
> @@ -1183,7 +1184,8 @@ struct omap_hwmod am33xx_tptc1_hwmod = {
> .name = "tptc1",
> .class = &am33xx_tptc_hwmod_class,
> .clkdm_name = "l3_clkdm",
> - .flags = (HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY),
> + .flags = (HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY |
> + HWMOD_INIT_NO_IDLE),
> .main_clk = "l3_gclk",
> .prcm = {
> .omap4 = {
> @@ -1197,7 +1199,8 @@ struct omap_hwmod am33xx_tptc2_hwmod = {
> .name = "tptc2",
> .class = &am33xx_tptc_hwmod_class,
> .clkdm_name = "l3_clkdm",
> - .flags = (HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY),
> + .flags = (HWMOD_SWSUP_SIDLE | HWMOD_SWSUP_MSTANDBY |
> + HWMOD_INIT_NO_IDLE),
> .main_clk = "l3_gclk",
> .prcm = {
> .omap4 = {
> --
> 2.6.1
>
On 11/02/2015 12:23 PM, Peter Ujfalusi wrote:
> On 11/02/2015 12:24 PM, Vinod Koul wrote:
>> On Mon, Nov 02, 2015 at 12:11:00PM +0200, Peter Ujfalusi wrote:
>>> In Linux we do not have driver for TPTCs of eDMA3 since there is no need to
>>> do any configuration within TPTC for the eDMA3 to be operational. All
>>> configuration is via the TPCC.
>>> To prevent the omap_device_late_idle() to disable the TPTCs when they are
>>> no longer bind with the edma driver, the HWMOD_INIT_NO_IDLE need to be
>>> added.
>>>
>>> Signed-off-by: Peter Ujfalusi <[email protected]>
>>> ---
>>> Vinod, Olof,
>>>
>>> This patch somehow got lost in my working branch. It was mixed within the patches
>>> I will have for 4.5 while it should have been within the new eDMA3 binding
>>> series..
>>
>> Does this fix the issue reported by Olof? Also ACK pls for this
>
> Yes, it is fixing the issue, I have this in my branch. everything looks fine
> on AM335x/AM437x/Dra7xx where I have eDMA in use.
Vinod, can you please ignore this patch? I'm going to send a patch agains
edma.c which will handle this better.
--
P?ter