1. Patchew
  2. linux
  3. [PATCH v4 0/7] drm/msm: adreno: add support for DDR bandwidth scaling via GMU
  4. View patch

[PATCH v4 3/7] drm/msm: adreno: dynamically generate GMU bw table

Neil Armstrong posted 7 patches 1 year ago
There is a newer version of this series
[PATCH v4 3/7] drm/msm: adreno: dynamically generate GMU bw table
Posted by Neil Armstrong 1 year ago 
The Adreno GPU Management Unit (GMU) can also scale the ddr
bandwidth along the frequency and power domain level, but for
now we statically fill the bw_table with values from the
downstream driver.

Only the first entry is used, which is a disable vote, so we
currently rely on scaling via the linux interconnect paths.

Let's dynamically generate the bw_table with the vote values
previously calculated from the OPPs.

Those entries will then be used by the GMU when passing the
appropriate bandwidth level while voting for a gpu frequency.

Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_hfi.c | 41 ++++++++++++++++++++++++++++++++++-
 1 file changed, 40 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
index cb8844ed46b29c4569d05eb7a24f7b27e173190f..fc4bfad51de9a3b6617fbbd03471a5851d43ce88 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
@@ -5,7 +5,10 @@
 #include <linux/circ_buf.h>
 #include <linux/list.h>
 
+#include <dt-bindings/interconnect/qcom,icc.h>
+
 #include <soc/qcom/cmd-db.h>
+#include <soc/qcom/tcs.h>
 
 #include "a6xx_gmu.h"
 #include "a6xx_gmu.xml.h"
@@ -259,6 +262,39 @@ static int a6xx_hfi_send_perf_table(struct a6xx_gmu *gmu)
 		NULL, 0);
 }
 
+static void a6xx_generate_bw_table(const struct a6xx_info *info, struct a6xx_gmu *gmu,
+				   struct a6xx_hfi_msg_bw_table *msg)
+{
+	unsigned int i, j;
+
+	msg->ddr_wait_bitmask = QCOM_ICC_TAG_ALWAYS;
+
+	for (i = 0; i < GMU_MAX_BCMS; i++) {
+		if (!info->bcms[i].name)
+			break;
+		msg->ddr_cmds_addrs[i] = cmd_db_read_addr(info->bcms[i].name);
+	}
+	msg->ddr_cmds_num = i;
+
+	for (i = 0; i < gmu->nr_gpu_bws; ++i)
+		for (j = 0; j < msg->ddr_cmds_num; j++)
+			msg->ddr_cmds_data[i][j] = gmu->gpu_ib_votes[i][j];
+	msg->bw_level_num = gmu->nr_gpu_bws;
+
+	/*
+	 * These are the CX (CNOC) votes - these are used by the GMU
+	 * The 'CN0' BCM is used on all targets, and votes are basically
+	 * 'off' and 'on' states with first bit to enable the path.
+	 */
+
+	msg->cnoc_cmds_num = 1;
+	msg->cnoc_wait_bitmask = QCOM_ICC_TAG_AMC;
+
+	msg->cnoc_cmds_addrs[0] = cmd_db_read_addr("CN0");
+	msg->cnoc_cmds_data[0][0] = BCM_TCS_CMD(true, false, 0, 0);
+	msg->cnoc_cmds_data[1][0] = BCM_TCS_CMD(true, true, 0, BIT(0));
+}
+
 static void a618_build_bw_table(struct a6xx_hfi_msg_bw_table *msg)
 {
 	/* Send a single "off" entry since the 618 GMU doesn't do bus scaling */
@@ -664,6 +700,7 @@ static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu)
 	struct a6xx_hfi_msg_bw_table *msg;
 	struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu);
 	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
+	const struct a6xx_info *info = adreno_gpu->info->a6xx;
 
 	if (gmu->bw_table)
 		goto send;
@@ -672,7 +709,9 @@ static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu)
 	if (!msg)
 		return -ENOMEM;
 
-	if (adreno_is_a618(adreno_gpu))
+	if (info->bcms && gmu->nr_gpu_bws > 1)
+		a6xx_generate_bw_table(info, gmu, msg);
+	else if (adreno_is_a618(adreno_gpu))
 		a618_build_bw_table(msg);
 	else if (adreno_is_a619(adreno_gpu))
 		a619_build_bw_table(msg);

-- 
2.34.1
Re: [PATCH v4 3/7] drm/msm: adreno: dynamically generate GMU bw table
Posted by Akhil P Oommen 1 year ago 
On 12/5/2024 8:31 PM, Neil Armstrong wrote:
> The Adreno GPU Management Unit (GMU) can also scale the ddr
> bandwidth along the frequency and power domain level, but for
> now we statically fill the bw_table with values from the
> downstream driver.
> 
> Only the first entry is used, which is a disable vote, so we
> currently rely on scaling via the linux interconnect paths.
> 
> Let's dynamically generate the bw_table with the vote values
> previously calculated from the OPPs.
> 
> Those entries will then be used by the GMU when passing the
> appropriate bandwidth level while voting for a gpu frequency.
> 
> Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
> ---
>  drivers/gpu/drm/msm/adreno/a6xx_hfi.c | 41 ++++++++++++++++++++++++++++++++++-
>  1 file changed, 40 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
> index cb8844ed46b29c4569d05eb7a24f7b27e173190f..fc4bfad51de9a3b6617fbbd03471a5851d43ce88 100644
> --- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
> +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
> @@ -5,7 +5,10 @@
>  #include <linux/circ_buf.h>
>  #include <linux/list.h>
>  
> +#include <dt-bindings/interconnect/qcom,icc.h>
> +
>  #include <soc/qcom/cmd-db.h>
> +#include <soc/qcom/tcs.h>
>  
>  #include "a6xx_gmu.h"
>  #include "a6xx_gmu.xml.h"
> @@ -259,6 +262,39 @@ static int a6xx_hfi_send_perf_table(struct a6xx_gmu *gmu)
>  		NULL, 0);
>  }
>  
> +static void a6xx_generate_bw_table(const struct a6xx_info *info, struct a6xx_gmu *gmu,
> +				   struct a6xx_hfi_msg_bw_table *msg)
> +{
> +	unsigned int i, j;
> +
> +	msg->ddr_wait_bitmask = QCOM_ICC_TAG_ALWAYS;

Why this is QCOM_ICC_TAG_ALWAYS?

IIRC, this bitmask informs RPMH whether it should wait for previous BCM
vote to complete. Can we implement the same logic from kgsl to create
this bitmask?


> +
> +	for (i = 0; i < GMU_MAX_BCMS; i++) {
> +		if (!info->bcms[i].name)
> +			break;
> +		msg->ddr_cmds_addrs[i] = cmd_db_read_addr(info->bcms[i].name);
> +	}
> +	msg->ddr_cmds_num = i;
> +
> +	for (i = 0; i < gmu->nr_gpu_bws; ++i)
> +		for (j = 0; j < msg->ddr_cmds_num; j++)
> +			msg->ddr_cmds_data[i][j] = gmu->gpu_ib_votes[i][j];
> +	msg->bw_level_num = gmu->nr_gpu_bws;
> +
> +	/*
> +	 * These are the CX (CNOC) votes - these are used by the GMU
> +	 * The 'CN0' BCM is used on all targets, and votes are basically
> +	 * 'off' and 'on' states with first bit to enable the path.
> +	 */
> +
> +	msg->cnoc_cmds_num = 1;
> +	msg->cnoc_wait_bitmask = QCOM_ICC_TAG_AMC;

Same here.

Rest looks fine to me.

-Akhil

> +
> +	msg->cnoc_cmds_addrs[0] = cmd_db_read_addr("CN0");
> +	msg->cnoc_cmds_data[0][0] = BCM_TCS_CMD(true, false, 0, 0);
> +	msg->cnoc_cmds_data[1][0] = BCM_TCS_CMD(true, true, 0, BIT(0));
> +}
> +
>  static void a618_build_bw_table(struct a6xx_hfi_msg_bw_table *msg)
>  {
>  	/* Send a single "off" entry since the 618 GMU doesn't do bus scaling */
> @@ -664,6 +700,7 @@ static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu)
>  	struct a6xx_hfi_msg_bw_table *msg;
>  	struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu);
>  	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
> +	const struct a6xx_info *info = adreno_gpu->info->a6xx;
>  
>  	if (gmu->bw_table)
>  		goto send;
> @@ -672,7 +709,9 @@ static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu)
>  	if (!msg)
>  		return -ENOMEM;
>  
> -	if (adreno_is_a618(adreno_gpu))
> +	if (info->bcms && gmu->nr_gpu_bws > 1)
> +		a6xx_generate_bw_table(info, gmu, msg);
> +	else if (adreno_is_a618(adreno_gpu))
>  		a618_build_bw_table(msg);
>  	else if (adreno_is_a619(adreno_gpu))
>  		a619_build_bw_table(msg);
>
Re: [PATCH v4 3/7] drm/msm: adreno: dynamically generate GMU bw table
Posted by Neil Armstrong 1 year ago 
On 09/12/2024 13:11, Akhil P Oommen wrote:
> On 12/5/2024 8:31 PM, Neil Armstrong wrote:
>> The Adreno GPU Management Unit (GMU) can also scale the ddr
>> bandwidth along the frequency and power domain level, but for
>> now we statically fill the bw_table with values from the
>> downstream driver.
>>
>> Only the first entry is used, which is a disable vote, so we
>> currently rely on scaling via the linux interconnect paths.
>>
>> Let's dynamically generate the bw_table with the vote values
>> previously calculated from the OPPs.
>>
>> Those entries will then be used by the GMU when passing the
>> appropriate bandwidth level while voting for a gpu frequency.
>>
>> Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
>> ---
>>   drivers/gpu/drm/msm/adreno/a6xx_hfi.c | 41 ++++++++++++++++++++++++++++++++++-
>>   1 file changed, 40 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
>> index cb8844ed46b29c4569d05eb7a24f7b27e173190f..fc4bfad51de9a3b6617fbbd03471a5851d43ce88 100644
>> --- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
>> +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
>> @@ -5,7 +5,10 @@
>>   #include <linux/circ_buf.h>
>>   #include <linux/list.h>
>>   
>> +#include <dt-bindings/interconnect/qcom,icc.h>
>> +
>>   #include <soc/qcom/cmd-db.h>
>> +#include <soc/qcom/tcs.h>
>>   
>>   #include "a6xx_gmu.h"
>>   #include "a6xx_gmu.xml.h"
>> @@ -259,6 +262,39 @@ static int a6xx_hfi_send_perf_table(struct a6xx_gmu *gmu)
>>   		NULL, 0);
>>   }
>>   
>> +static void a6xx_generate_bw_table(const struct a6xx_info *info, struct a6xx_gmu *gmu,
>> +				   struct a6xx_hfi_msg_bw_table *msg)
>> +{
>> +	unsigned int i, j;
>> +
>> +	msg->ddr_wait_bitmask = QCOM_ICC_TAG_ALWAYS;
> 
> Why this is QCOM_ICC_TAG_ALWAYS?
> 
> IIRC, this bitmask informs RPMH whether it should wait for previous BCM
> vote to complete. Can we implement the same logic from kgsl to create
> this bitmask?

Ack, Let me check

> 
> 
>> +
>> +	for (i = 0; i < GMU_MAX_BCMS; i++) {
>> +		if (!info->bcms[i].name)
>> +			break;
>> +		msg->ddr_cmds_addrs[i] = cmd_db_read_addr(info->bcms[i].name);
>> +	}
>> +	msg->ddr_cmds_num = i;
>> +
>> +	for (i = 0; i < gmu->nr_gpu_bws; ++i)
>> +		for (j = 0; j < msg->ddr_cmds_num; j++)
>> +			msg->ddr_cmds_data[i][j] = gmu->gpu_ib_votes[i][j];
>> +	msg->bw_level_num = gmu->nr_gpu_bws;
>> +
>> +	/*
>> +	 * These are the CX (CNOC) votes - these are used by the GMU
>> +	 * The 'CN0' BCM is used on all targets, and votes are basically
>> +	 * 'off' and 'on' states with first bit to enable the path.
>> +	 */
>> +
>> +	msg->cnoc_cmds_num = 1;
>> +	msg->cnoc_wait_bitmask = QCOM_ICC_TAG_AMC;
> 
> Same here.
> 
> Rest looks fine to me.

Thanks,
Neil

> 
> -Akhil
> 
>> +
>> +	msg->cnoc_cmds_addrs[0] = cmd_db_read_addr("CN0");
>> +	msg->cnoc_cmds_data[0][0] = BCM_TCS_CMD(true, false, 0, 0);
>> +	msg->cnoc_cmds_data[1][0] = BCM_TCS_CMD(true, true, 0, BIT(0));
>> +}
>> +
>>   static void a618_build_bw_table(struct a6xx_hfi_msg_bw_table *msg)
>>   {
>>   	/* Send a single "off" entry since the 618 GMU doesn't do bus scaling */
>> @@ -664,6 +700,7 @@ static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu)
>>   	struct a6xx_hfi_msg_bw_table *msg;
>>   	struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu);
>>   	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
>> +	const struct a6xx_info *info = adreno_gpu->info->a6xx;
>>   
>>   	if (gmu->bw_table)
>>   		goto send;
>> @@ -672,7 +709,9 @@ static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu)
>>   	if (!msg)
>>   		return -ENOMEM;
>>   
>> -	if (adreno_is_a618(adreno_gpu))
>> +	if (info->bcms && gmu->nr_gpu_bws > 1)
>> +		a6xx_generate_bw_table(info, gmu, msg);
>> +	else if (adreno_is_a618(adreno_gpu))
>>   		a618_build_bw_table(msg);
>>   	else if (adreno_is_a619(adreno_gpu))
>>   		a619_build_bw_table(msg);
>>
>

Patchew 2.1-devel-b67e4c2

© 2016 - 2025 Red Hat, Inc.