[PATCH v4 08/11] coresight-tpdm: Add nodes to configure pattern match output
Tao Zhang
quic_taozha at quicinc.com
Fri Jun 2 01:29:26 PDT 2023
On 6/1/2023 9:28 PM, Suzuki K Poulose wrote:
> On 27/04/2023 10:00, Tao Zhang wrote:
>> Add nodes to configure trigger pattern and trigger pattern mask.
>> Each DSB subunit TPDM has maximum of n(n<7) XPR registers to
>> configure trigger pattern match output. Eight 32 bit registers
>> providing DSB interface trigger output pattern match comparison.
>> And each DSB subunit TPDM has maximum of m(m<7) XPMR registers to
>> configure trigger pattern mask match output. Eight 32 bit
>> registers providing DSB interface trigger output pattern match
>> mask.
>>
>> Signed-off-by: Tao Zhang <quic_taozha at quicinc.com>
>> ---
>> .../ABI/testing/sysfs-bus-coresight-devices-tpdm | 30 ++++++++
>> drivers/hwtracing/coresight/coresight-tpdm.c | 85
>> ++++++++++++++++++++++
>> drivers/hwtracing/coresight/coresight-tpdm.h | 8 ++
>> 3 files changed, 123 insertions(+)
>>
>> diff --git
>> a/Documentation/ABI/testing/sysfs-bus-coresight-devices-tpdm
>> b/Documentation/ABI/testing/sysfs-bus-coresight-devices-tpdm
>> index a57f000..c04c735 100644
>> --- a/Documentation/ABI/testing/sysfs-bus-coresight-devices-tpdm
>> +++ b/Documentation/ABI/testing/sysfs-bus-coresight-devices-tpdm
>> @@ -92,3 +92,33 @@ Description:
>> <integer1> : Start EDCMR register number
>> <integer2> : End EDCMR register number
>> <integer3> : The value need to be written
>> +
>> +What: /sys/bus/coresight/devices/<tpdm-name>/dsb_trig_patt_val
>> +Date: March 2023
>> +KernelVersion 6.3
>> +Contact: Jinlong Mao (QUIC) <quic_jinlmao at quicinc.com>, Tao Zhang
>> (QUIC) <quic_taozha at quicinc.com>
>> +Description:
>> + (Write) Set the trigger pattern value of DSB tpdm.
>> + Read the trigger pattern value of DSB tpdm.
>> +
>> + Expected format is the following:
>> + <integer1> <integer2>
>> +
>> + Where:
>> + <integer1> : Index number of XPR register, the range is 0 to 7
>> + <integer2> : The value need to be written
>
> I assume the values written to the registers are not special and doesn't
> have meaning and thus need not be documented ?
Sure, I will update this in the next patch series.
>
>> +
>> +What: /sys/bus/coresight/devices/<tpdm-name>/dsb_trig_patt_mask
>> +Date: March 2023
>> +KernelVersion 6.3
>
> Same as the previous one, 6.5 please
Sure, I will update this in the next patch series.
>
>> +Contact: Jinlong Mao (QUIC) <quic_jinlmao at quicinc.com>, Tao Zhang
>> (QUIC) <quic_taozha at quicinc.com>
>> +Description:
>> + (Write) Set the trigger pattern mask of DSB tpdm.
>> + Read the trigger pattern mask of DSB tpdm.
>> +
>> + Expected format is the following:
>> + <integer1> <integer2>
>> +
>> + Where:
>> + <integer1> : Index number of XPMR register, the range is 0
>> to 7
>> + <integer2> : The value need to be written
>> diff --git a/drivers/hwtracing/coresight/coresight-tpdm.c
>> b/drivers/hwtracing/coresight/coresight-tpdm.c
>> index a40e458..9387bdf 100644
>> --- a/drivers/hwtracing/coresight/coresight-tpdm.c
>> +++ b/drivers/hwtracing/coresight/coresight-tpdm.c
>> @@ -89,6 +89,13 @@ static void tpdm_enable_dsb(struct tpdm_drvdata
>> *drvdata)
>> writel_relaxed(drvdata->dsb->edge_ctrl_mask[i],
>> drvdata->base + TPDM_DSB_EDCMR(i));
>> + for (i = 0; i < TPDM_DSB_MAX_PATT; i++) {
>
> Same as the previous, can we safely assume that write to these
> registers won't trigger an Error if not impelemented ?
Yes, it won't trigger an error since these inexistent register's
addresses are not occupied and safe
for being accessed.
>
>> + writel_relaxed(drvdata->dsb->trig_patt_val[i],
>> + drvdata->base + TPDM_DSB_XPR(i));
>> + writel_relaxed(drvdata->dsb->trig_patt_mask[i],
>> + drvdata->base + TPDM_DSB_XPMR(i));
>> + }
>> +
>> val = readl_relaxed(drvdata->base + TPDM_DSB_TIER);
>> /* Set trigger timestamp */
>> if (drvdata->dsb->trig_ts)
>> @@ -444,6 +451,82 @@ static ssize_t dsb_edge_ctrl_mask_store(struct
>> device *dev,
>> }
>> static DEVICE_ATTR_RW(dsb_edge_ctrl_mask);
>> +static ssize_t dsb_trig_patt_val_show(struct device *dev,
>> + struct device_attribute *attr,
>> + char *buf)
>> +{
>> + struct tpdm_drvdata *drvdata = dev_get_drvdata(dev->parent);
>> + ssize_t size = 0;
>> + int i = 0;
>> +
>> + spin_lock(&drvdata->spinlock);
>> + for (i = 0; i < TPDM_DSB_MAX_PATT; i++) {
>> + size += sysfs_emit_at(buf, size,
>> + "Index: 0x%x Value: 0x%x\n", i,
>> + drvdata->dsb->trig_patt_val[i]);
>
> Please detect the return of 0 and break. Same below.
See my comments in patch #7 mail.
Best,
Tao
>
>
>> + }
>> + spin_unlock(&drvdata->spinlock);
>> + return size;
>> +}
>> +
>> +static ssize_t dsb_trig_patt_val_store(struct device *dev,
>> + struct device_attribute *attr,
>> + const char *buf,
>> + size_t size)
>> +{
>> + struct tpdm_drvdata *drvdata = dev_get_drvdata(dev->parent);
>> + unsigned long index, val;
>> +
>> + if (sscanf(buf, "%lx %lx", &index, &val) != 2)
>> + return -EINVAL;
>> + if (index >= TPDM_DSB_MAX_PATT)
>> + return -EPERM;
>> +
>> + spin_lock(&drvdata->spinlock);
>> + drvdata->dsb->trig_patt_val[index] = val;
>> + spin_unlock(&drvdata->spinlock);
>> + return size;
>> +}
>> +static DEVICE_ATTR_RW(dsb_trig_patt_val);
>> +
>> +static ssize_t dsb_trig_patt_mask_show(struct device *dev,
>> + struct device_attribute *attr,
>> + char *buf)
>> +{
>> + struct tpdm_drvdata *drvdata = dev_get_drvdata(dev->parent);
>> + ssize_t size = 0;
>> + int i = 0;
>> +
>> + spin_lock(&drvdata->spinlock);
>> + for (i = 0; i < TPDM_DSB_MAX_PATT; i++) {
>> + size += sysfs_emit_at(buf, size,
>> + "Index: 0x%x Value: 0x%x\n", i,
>> + drvdata->dsb->trig_patt_mask[i]);
>> + }
>> + spin_unlock(&drvdata->spinlock);
>> + return size;
>
> Suzuki
>
More information about the linux-arm-kernel
mailing list