platform/chrome: Expose resume result via debugfs
PATCH v2 1/1 platform/chrome: Expose resume result via debugfs
For ECs that support it, the EC returns the number of slp_s0 transitions and whether or not there was a timeout in the resume response. Expose the last resume result to usermode via debugfs so that usermode can detect and report S0ix timeouts.
---
Changes in v2:
- Moved from sysfs to debugfs (Enric)
- Added documentation (Enric)
--- Documentation/ABI/testing/debugfs-cros-ec | 22 ++++++++++++++++++++++ drivers/mfd/cros_ec.c | 6 +++++- drivers/platform/chrome/cros_ec_debugfs.c | 7 +++++++ include/linux/mfd/cros_ec.h | 1 + 4 files changed, 35 insertions(+), 1 deletion(-)
Documentation/ABI/testing/debugfs-cros-ec +22 -0
@@ -32,3 +32,25 @@ Description: is used for synchronizing the AP host time with the EC log. An error is returned if the command is not supported by the EC or there is a communication problem. + +What: /sys/kernel/debug/cros_ec/last_resume_result +Date: June 2019 +KernelVersion: 5.3 +Description: + Some ECs have a feature where they will track transitions to + the (Intel) processor's SLP_S0 line, in order to detect cases + where a system failed to go into S0ix. When the system resumes, + an EC with this feature will return a summary of SLP_S0 + transitions that occurred. The last_resume_result file returns + the most recent response from the AP's resume message to the EC. + + The bottom 31 bits contain a count of the number of SLP_S0 + transitions that occurred since the suspend message was + received. Bit 31 is set if the EC attempted to wake the + system due to a timeout when watching for SLP_S0 transitions. + Callers can use this to detect a wake from the EC due to + S0ix timeouts. The result will be zero if no suspend + transitions have been attempted, or the EC does not support + this feature. + + Output will be in the format: "0x%08x\n".
drivers/mfd/cros_ec.c +5 -1
@@ -102,12 +102,16 @@ static int cros_ec_sleep_event(struct cros_ec_device *ec_dev, u8 sleep_event)/* For now, report failure to transition to S0ix with a warning. */ if (ret >=0 && ec_dev->host_sleep_v1 && - (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) + (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) { + ec_dev->last_resume_result = + buf.u.resp1.resume_response.sleep_transitions; +WARN_ONCE (buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TIMEOUT ,"EC detected sleep transition timeout. Total slp_s0 transitions: %d" , buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TRANSITIONS_MASK ); + }return ret; }
drivers/platform/chrome/cros_ec_debugfs.c +7 -0
@@ -447,6 +447,13 @@ static int cros_ec_debugfs_probe(struct platform_device *pd)debugfs_create_file ("uptime" ,0444 , debug_info->dir, debug_info, &cros_ec_uptime_fops); +if (!strcmp (ec->class_dev.kobj.name,CROS_EC_DEV_NAME )) { +debugfs_create_x32 ("last_resume_result" , +0444 , + debug_info->dir, + &ec->ec_dev->last_resume_result); + } + ec->debug_info = debug_info;dev_set_drvdata (&pd->dev, ec);
include/linux/mfd/cros_ec.h +1 -0
@@ -155,6 +155,7 @@ struct cros_ec_device {struct ec_response_get_next_event_v1 event_data;int event_size;u32 host_event_wake_mask; +u32 last_resume_result; };/**
Signature · 1 line
2.20.1
Quoted 6 lines · Evan Green
For ECs that support it, the EC returns the number of
slp_s0transitions and whether or not there was a timeout in the resume response. Expose the last resume result to usermode via debugfs so that usermode can detect and report S0ix timeouts.Signed-off-by Evan Green <[email protected]>
This still needs a platform/chrome Ack.
Quoted 96 lines
---
Changes in v2:
- Moved from sysfs to debugfs (Enric)
- Added documentation (Enric)--- Documentation/ABI/testing/debugfs-cros-ec | 22 ++++++++++++++++++++++ drivers/mfd/cros_ec.c | 6 +++++- drivers/platform/chrome/cros_ec_debugfs.c | 7 +++++++ include/linux/mfd/cros_ec.h | 1 + 4 files changed, 35 insertions(+), 1 deletion(-)Documentation/ABI/testing/debugfs-cros-ec +22
-0@@ -32,3 +32,25 @@ Description: is used for synchronizing the AP host time with the EC log. An error is returned if the command is not supported by the EC or there is a communication problem. + +What: /sys/kernel/debug/cros_ec/last_resume_result +Date: June 2019 +KernelVersion: 5.3 +Description: + Some ECs have a feature where they will track transitions to + the (Intel) processor's SLP_S0 line, in order to detect cases + where a system failed to go into S0ix. When the system resumes, + an EC with this feature will return a summary of SLP_S0 + transitions that occurred. The last_resume_result file returns + the most recent response from the AP's resume message to the EC. + + The bottom 31 bits contain a count of the number of SLP_S0 + transitions that occurred since the suspend message was + received. Bit 31 is set if the EC attempted to wake the + system due to a timeout when watching for SLP_S0 transitions. + Callers can use this to detect a wake from the EC due to + S0ix timeouts. The result will be zero if no suspend + transitions have been attempted, or the EC does not support + this feature. + + Output will be in the format: "0x%08x\n".drivers/mfd/cros_ec.c +5
-1@@ -102,12 +102,16 @@ static int cros_ec_sleep_event(struct cros_ec_device *ec_dev, u8 sleep_event)/* For now, report failure to transition to S0ix with a warning. */ if (ret >=0 && ec_dev->host_sleep_v1 && - (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) + (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) { + ec_dev->last_resume_result = + buf.u.resp1.resume_response.sleep_transitions; +WARN_ONCE (buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TIMEOUT ,"EC detected sleep transition timeout. Total slp_s0 transitions: %d" , buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TRANSITIONS_MASK ); + }return ret; }drivers/platform/chrome/cros_ec_debugfs.c +7
-0@@ -447,6 +447,13 @@ static int cros_ec_debugfs_probe(struct platform_device *pd)debugfs_create_file ("uptime" ,0444 , debug_info->dir, debug_info, &cros_ec_uptime_fops); +if (!strcmp (ec->class_dev.kobj.name,CROS_EC_DEV_NAME )) { +debugfs_create_x32 ("last_resume_result" , +0444 , + debug_info->dir, + &ec->ec_dev->last_resume_result); + } + ec->debug_info = debug_info;dev_set_drvdata (&pd->dev, ec);include/linux/mfd/cros_ec.h +1
-0@@ -155,6 +155,7 @@ struct cros_ec_device {struct ec_response_get_next_event_v1 event_data;int event_size;u32 host_event_wake_mask; +u32 last_resume_result; };/**
Signature · 4 lines
Lee Jones [李琼斯] Linaro Services Technical Lead Linaro.org │ Open source software for ARM SoCs Follow Linaro: Facebook | Twitter | Blog
Hi Evan,
Two few comments and I think I'm fine with it.
Quoted 25 lines · Lee Jones
Quoted 6 lines · Evan Green
For ECs that support it, the EC returns the number of
slp_s0transitions and whether or not there was a timeout in the resume response. Expose the last resume result to usermode via debugfs so that usermode can detect and report S0ix timeouts.Signed-off-by Evan Green <[email protected]>This still needs a platform/chrome Ack.
Quoted 14 lines
---
Changes in v2:
- Moved from sysfs to debugfs (Enric)
- Added documentation (Enric)--- Documentation/ABI/testing/debugfs-cros-ec | 22 ++++++++++++++++++++++ drivers/mfd/cros_ec.c | 6 +++++- drivers/platform/chrome/cros_ec_debugfs.c | 7 +++++++ include/linux/mfd/cros_ec.h | 1 + 4 files changed, 35 insertions(+), 1 deletion(-)
Quoted 10 lines
Documentation/ABI/testing/debugfs-cros-ec Description: · line 36
Quoted 10 lines
Documentation/ABI/testing/debugfs-cros-ec +2
-0@@ -32,3 +32,25 @@ Description: is used for synchronizing the AP host time with the EC log. An error is returned if the command is not supported by the EC or there is a communication problem. + +What: /sys/kernel/debug/cros_ec/last_resume_result
Thinking about it, as other the other interfaces, I'd do
s/cros_ec/<cros-ec-device>/
I know that for now only cros_ec supports that, but we don't know what will happen in the future, specially now that the number of cros devices is incrementing.
Quoted 50 lines
Quoted 50 lines
Documentation/ABI/testing/debugfs-cros-ec +20
-0+Date: June 2019 +KernelVersion: 5.3 +Description: + Some ECs have a feature where they will track transitions to + the (Intel) processor's SLP_S0 line, in order to detect cases + where a system failed to go into S0ix. When the system resumes, + an EC with this feature will return a summary of SLP_S0 + transitions that occurred. The last_resume_result file returns + the most recent response from the AP's resume message to the EC. + + The bottom 31 bits contain a count of the number of SLP_S0 + transitions that occurred since the suspend message was + received. Bit 31 is set if the EC attempted to wake the + system due to a timeout when watching for SLP_S0 transitions. + Callers can use this to detect a wake from the EC due to + S0ix timeouts. The result will be zero if no suspend + transitions have been attempted, or the EC does not support + this feature. + + Output will be in the format: "0x%08x\n".drivers/mfd/cros_ec.c +5
-1@@ -102,12 +102,16 @@ static int cros_ec_sleep_event(struct cros_ec_device *ec_dev, u8 sleep_event)/* For now, report failure to transition to S0ix with a warning. */ if (ret >=0 && ec_dev->host_sleep_v1 && - (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) + (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) { + ec_dev->last_resume_result = + buf.u.resp1.resume_response.sleep_transitions; +WARN_ONCE (buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TIMEOUT ,"EC detected sleep transition timeout. Total slp_s0 transitions: %d" , buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TRANSITIONS_MASK ); + }return ret; }drivers/platform/chrome/cros_ec_debugfs.c +0
-0@@ -447,6 +447,13 @@ static int cros_ec_debugfs_probe(struct platform_device *pd)debugfs_create_file ("uptime" ,0444 , debug_info->dir, debug_info, &cros_ec_uptime_fops);
Quoted 1 line
drivers/platform/chrome/cros_ec_debugfs.c static int cros_ec_debugfs_probe(struct platform_device *pd) · line 450
Quoted 1 line
drivers/platform/chrome/cros_ec_debugfs.c +1
-0+if (!strcmp (ec->class_dev.kobj.name,CROS_EC_DEV_NAME )) {
For debugfs I don't care having the file exposed even is not supported, anyway there are some CROS_EC_DEV_NAME that won't support it, so just make this simple and create the file always.
Quoted 20 lines
Quoted 20 lines
drivers/platform/chrome/cros_ec_debugfs.c +6
-0+debugfs_create_x32 ("last_resume_result" , +0444 , + debug_info->dir, + &ec->ec_dev->last_resume_result); + } + ec->debug_info = debug_info;dev_set_drvdata (&pd->dev, ec);include/linux/mfd/cros_ec.h +1
-0@@ -155,6 +155,7 @@ struct cros_ec_device {struct ec_response_get_next_event_v1 event_data;int event_size;u32 host_event_wake_mask; +u32 last_resume_result; };
Quoted 2 lines
include/linux/mfd/cros_ec.h struct cros_ec_device { · line 161
Quoted 1 line
include/linux/mfd/cros_ec.h +0
-0/**
Thanks,
~ Enric
Quoted 44 lines · Enric Balletbo i Serra
Hi Evan,
Two few comments and I think I'm fine with it.
Quoted 25 lines · Lee Jones
Quoted 6 lines · Evan Green
For ECs that support it, the EC returns the number of
slp_s0transitions and whether or not there was a timeout in the resume response. Expose the last resume result to usermode via debugfs so that usermode can detect and report S0ix timeouts.Signed-off-by Evan Green <[email protected]>This still needs a platform/chrome Ack.
Quoted 14 lines
---
Changes in v2:
- Moved from sysfs to debugfs (Enric)
- Added documentation (Enric)--- Documentation/ABI/testing/debugfs-cros-ec | 22 ++++++++++++++++++++++ drivers/mfd/cros_ec.c | 6 +++++- drivers/platform/chrome/cros_ec_debugfs.c | 7 +++++++ include/linux/mfd/cros_ec.h | 1 + 4 files changed, 35 insertions(+), 1 deletion(-)Quoted 10 lines
Documentation/ABI/testing/debugfs-cros-ec Description: · line 36
Quoted 10 lines
Documentation/ABI/testing/debugfs-cros-ec +2
-0@@ -32,3 +32,25 @@ Description: is used for synchronizing the AP host time with the EC log. An error is returned if the command is not supported by the EC or there is a communication problem. + +What: /sys/kernel/debug/cros_ec/last_resume_resultThinking about it, as other the other interfaces, I'd do
Quoted 4 lines
s/cros_ec/<cros-ec-device>/
I know that for now only
cros_ecsupports that, but we don't know what will happen in the future, specially now that the number of cros devices is incrementing.
See my comment below, I suppose the fate of these two comments are tied together.
Quoted 51 lines
Quoted 51 lines
Quoted 51 lines
Documentation/ABI/testing/debugfs-cros-ec +20
-0+Date: June 2019 +KernelVersion: 5.3 +Description: + Some ECs have a feature where they will track transitions to + the (Intel) processor's SLP_S0 line, in order to detect cases + where a system failed to go into S0ix. When the system resumes, + an EC with this feature will return a summary of SLP_S0 + transitions that occurred. The last_resume_result file returns + the most recent response from the AP's resume message to the EC. + + The bottom 31 bits contain a count of the number of SLP_S0 + transitions that occurred since the suspend message was + received. Bit 31 is set if the EC attempted to wake the + system due to a timeout when watching for SLP_S0 transitions. + Callers can use this to detect a wake from the EC due to + S0ix timeouts. The result will be zero if no suspend + transitions have been attempted, or the EC does not support + this feature. + + Output will be in the format: "0x%08x\n".drivers/mfd/cros_ec.c +5
-1@@ -102,12 +102,16 @@ static int cros_ec_sleep_event(struct cros_ec_device *ec_dev, u8 sleep_event)/* For now, report failure to transition to S0ix with a warning. */ if (ret >=0 && ec_dev->host_sleep_v1 && - (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) + (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) { + ec_dev->last_resume_result = + buf.u.resp1.resume_response.sleep_transitions; +WARN_ONCE (buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TIMEOUT ,"EC detected sleep transition timeout. Total slp_s0 transitions: %d" , buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TRANSITIONS_MASK ); + }return ret; }drivers/platform/chrome/cros_ec_debugfs.c +0
-0@@ -447,6 +447,13 @@ static int cros_ec_debugfs_probe(struct platform_device *pd)debugfs_create_file ("uptime" ,0444 , debug_info->dir, debug_info, &cros_ec_uptime_fops);
Quoted 5 lines
Quoted 1 line
drivers/platform/chrome/cros_ec_debugfs.c static int cros_ec_debugfs_probe(struct platform_device *pd) · line 450
Quoted 1 line
drivers/platform/chrome/cros_ec_debugfs.c +1
-0+if (!strcmp (ec->class_dev.kobj.name,CROS_EC_DEV_NAME )) {For debugfs I don't care having the file exposed even is not supported, anyway there are some
CROS_EC_DEV_NAMEthat won't support it, so just make this simple and create the file always.
Aw, really? This file is very specific to system suspend/resume. My original patch had it everywhere, but I was finding it very ugly that this was showing up on things like the fingerprint device. I can change it if you think it's better to have it everywhere, but it also seems like an easy change to make in the future if this file is for some reason needed on other EC types.
-Evan
Hi Evan,
Quoted 111 lines · Evan Green
Quoted 44 lines · Enric Balletbo i Serra
Hi Evan,
Two few comments and I think I'm fine with it.
Quoted 25 lines · Lee Jones
Quoted 6 lines · Evan Green
For ECs that support it, the EC returns the number of
slp_s0transitions and whether or not there was a timeout in the resume response. Expose the last resume result to usermode via debugfs so that usermode can detect and report S0ix timeouts.Signed-off-by Evan Green <[email protected]>This still needs a platform/chrome Ack.
Quoted 14 lines
---
Changes in v2:
- Moved from sysfs to debugfs (Enric)
- Added documentation (Enric)--- Documentation/ABI/testing/debugfs-cros-ec | 22 ++++++++++++++++++++++ drivers/mfd/cros_ec.c | 6 +++++- drivers/platform/chrome/cros_ec_debugfs.c | 7 +++++++ include/linux/mfd/cros_ec.h | 1 + 4 files changed, 35 insertions(+), 1 deletion(-)Quoted 10 lines
Documentation/ABI/testing/debugfs-cros-ec Description: · line 36
Quoted 10 lines
Documentation/ABI/testing/debugfs-cros-ec +2
-0@@ -32,3 +32,25 @@ Description: is used for synchronizing the AP host time with the EC log. An error is returned if the command is not supported by the EC or there is a communication problem. + +What: /sys/kernel/debug/cros_ec/last_resume_resultThinking about it, as other the other interfaces, I'd do
Quoted 4 lines
s/cros_ec/<cros-ec-device>/
I know that for now only
cros_ecsupports that, but we don't know what will happen in the future, specially now that the number of cros devices is incrementing.See my comment below, I suppose the fate of these two comments are tied together.
Quoted 51 lines
Quoted 51 lines
Quoted 51 lines
Documentation/ABI/testing/debugfs-cros-ec +20
-0+Date: June 2019 +KernelVersion: 5.3 +Description: + Some ECs have a feature where they will track transitions to + the (Intel) processor's SLP_S0 line, in order to detect cases + where a system failed to go into S0ix. When the system resumes, + an EC with this feature will return a summary of SLP_S0 + transitions that occurred. The last_resume_result file returns + the most recent response from the AP's resume message to the EC. + + The bottom 31 bits contain a count of the number of SLP_S0 + transitions that occurred since the suspend message was + received. Bit 31 is set if the EC attempted to wake the + system due to a timeout when watching for SLP_S0 transitions. + Callers can use this to detect a wake from the EC due to + S0ix timeouts. The result will be zero if no suspend + transitions have been attempted, or the EC does not support + this feature. + + Output will be in the format: "0x%08x\n".drivers/mfd/cros_ec.c +5
-1@@ -102,12 +102,16 @@ static int cros_ec_sleep_event(struct cros_ec_device *ec_dev, u8 sleep_event)/* For now, report failure to transition to S0ix with a warning. */ if (ret >=0 && ec_dev->host_sleep_v1 && - (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) + (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) { + ec_dev->last_resume_result = + buf.u.resp1.resume_response.sleep_transitions; +WARN_ONCE (buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TIMEOUT ,"EC detected sleep transition timeout. Total slp_s0 transitions: %d" , buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TRANSITIONS_MASK ); + }return ret; }drivers/platform/chrome/cros_ec_debugfs.c +0
-0@@ -447,6 +447,13 @@ static int cros_ec_debugfs_probe(struct platform_device *pd)debugfs_create_file ("uptime" ,0444 , debug_info->dir, debug_info, &cros_ec_uptime_fops);Quoted 5 lines
Quoted 1 line
drivers/platform/chrome/cros_ec_debugfs.c static int cros_ec_debugfs_probe(struct platform_device *pd) · line 450
Quoted 1 line
drivers/platform/chrome/cros_ec_debugfs.c +1
-0+if (!strcmp (ec->class_dev.kobj.name,CROS_EC_DEV_NAME )) {For debugfs I don't care having the file exposed even is not supported, anyway there are some
CROS_EC_DEV_NAMEthat won't support it, so just make this simple and create the file always.
Quoted 7 lines
Aw, really? This file is very specific to system suspend/resume. My original patch had it everywhere, but I was finding it very ugly that this was showing up on things like the fingerprint device. I can change it if you think it's better to have it everywhere, but it also seems like an easy change to make in the future if this file is for some reason needed on other EC types.
I'd think different if it was a sysfs property but it's a debugfs.
Right now we have:
#define CROS_EC_DEV_NAME "cros_ec" #define CROS_EC_DEV_FP_NAME "cros_fp" #define CROS_EC_DEV_ISH_NAME "cros_ish" #define CROS_EC_DEV_PD_NAME "cros_pd" #define CROS_EC_DEV_SCP_NAME "cros_scp" #define CROS_EC_DEV_TP_NAME "cros_tp"
Is really the named cros_ec the only that has this feature? What about cros_scp (is not supposed to run the same base code as cros_ec)? And cros_ish ?
And all the named cros_ec devices have this feature? Maybe is supported by Nocturne but not Veyron? Wouldn't be exposed also on those cros_ec that doesn't support it?
And don't have the answer to all these questions, hence my concerns.
As per your documentation
+ ... The result will be zero if no suspend
+ transitions have been attempted, or the EC does not support
+ this feature.
I can accept if _all_ the CROS_EC_DEV_NAMEs and _only_ the CROS_EC_DEV_NAME supports it, is that the case?
But if you are anyway exposing this on CROS_EC_DEV_NAMEs that doesn't support it why not just expose to all and skip a future bunch of code to filter, it is clear what a 0 means and at the end it's just a debugfs file.
~ Enric
Quoted 2 lines
-Evan
Quoted 149 lines · Enric Balletbo i Serra
Hi Evan,
Quoted 111 lines · Evan Green
Quoted 44 lines · Enric Balletbo i Serra
Hi Evan,
Two few comments and I think I'm fine with it.
Quoted 25 lines · Lee Jones
Quoted 6 lines · Evan Green
For ECs that support it, the EC returns the number of
slp_s0transitions and whether or not there was a timeout in the resume response. Expose the last resume result to usermode via debugfs so that usermode can detect and report S0ix timeouts.Signed-off-by Evan Green <[email protected]>This still needs a platform/chrome Ack.
Quoted 14 lines
---
Changes in v2:
- Moved from sysfs to debugfs (Enric)
- Added documentation (Enric)--- Documentation/ABI/testing/debugfs-cros-ec | 22 ++++++++++++++++++++++ drivers/mfd/cros_ec.c | 6 +++++- drivers/platform/chrome/cros_ec_debugfs.c | 7 +++++++ include/linux/mfd/cros_ec.h | 1 + 4 files changed, 35 insertions(+), 1 deletion(-)Quoted 10 lines
Documentation/ABI/testing/debugfs-cros-ec Description: · line 36
Quoted 10 lines
Documentation/ABI/testing/debugfs-cros-ec +2
-0@@ -32,3 +32,25 @@ Description: is used for synchronizing the AP host time with the EC log. An error is returned if the command is not supported by the EC or there is a communication problem. + +What: /sys/kernel/debug/cros_ec/last_resume_resultThinking about it, as other the other interfaces, I'd do
Quoted 4 lines
s/cros_ec/<cros-ec-device>/
I know that for now only
cros_ecsupports that, but we don't know what will happen in the future, specially now that the number of cros devices is incrementing.See my comment below, I suppose the fate of these two comments are tied together.
Quoted 51 lines
Quoted 51 lines
Quoted 51 lines
Documentation/ABI/testing/debugfs-cros-ec +20
-0+Date: June 2019 +KernelVersion: 5.3 +Description: + Some ECs have a feature where they will track transitions to + the (Intel) processor's SLP_S0 line, in order to detect cases + where a system failed to go into S0ix. When the system resumes, + an EC with this feature will return a summary of SLP_S0 + transitions that occurred. The last_resume_result file returns + the most recent response from the AP's resume message to the EC. + + The bottom 31 bits contain a count of the number of SLP_S0 + transitions that occurred since the suspend message was + received. Bit 31 is set if the EC attempted to wake the + system due to a timeout when watching for SLP_S0 transitions. + Callers can use this to detect a wake from the EC due to + S0ix timeouts. The result will be zero if no suspend + transitions have been attempted, or the EC does not support + this feature. + + Output will be in the format: "0x%08x\n".drivers/mfd/cros_ec.c +5
-1@@ -102,12 +102,16 @@ static int cros_ec_sleep_event(struct cros_ec_device *ec_dev, u8 sleep_event)/* For now, report failure to transition to S0ix with a warning. */ if (ret >=0 && ec_dev->host_sleep_v1 && - (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) + (sleep_event ==HOST_SLEEP_EVENT_S0IX_RESUME )) { + ec_dev->last_resume_result = + buf.u.resp1.resume_response.sleep_transitions; +WARN_ONCE (buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TIMEOUT ,"EC detected sleep transition timeout. Total slp_s0 transitions: %d" , buf.u.resp1.resume_response.sleep_transitions &EC_HOST_RESUME_SLEEP_TRANSITIONS_MASK ); + }return ret; }drivers/platform/chrome/cros_ec_debugfs.c +0
-0@@ -447,6 +447,13 @@ static int cros_ec_debugfs_probe(struct platform_device *pd)debugfs_create_file ("uptime" ,0444 , debug_info->dir, debug_info, &cros_ec_uptime_fops);Quoted 5 lines
Quoted 1 line
drivers/platform/chrome/cros_ec_debugfs.c static int cros_ec_debugfs_probe(struct platform_device *pd) · line 450
Quoted 1 line
drivers/platform/chrome/cros_ec_debugfs.c +1
-0+if (!strcmp (ec->class_dev.kobj.name,CROS_EC_DEV_NAME )) {For debugfs I don't care having the file exposed even is not supported, anyway there are some
CROS_EC_DEV_NAMEthat won't support it, so just make this simple and create the file always.Quoted 7 lines
Aw, really? This file is very specific to system suspend/resume. My original patch had it everywhere, but I was finding it very ugly that this was showing up on things like the fingerprint device. I can change it if you think it's better to have it everywhere, but it also seems like an easy change to make in the future if this file is for some reason needed on other EC types.
I'd think different if it was a sysfs property but it's a debugfs.
Right now we have:
#define CROS_EC_DEV_NAME "cros_ec" #define CROS_EC_DEV_FP_NAME "cros_fp" #define CROS_EC_DEV_ISH_NAME "cros_ish" #define CROS_EC_DEV_PD_NAME "cros_pd" #define CROS_EC_DEV_SCP_NAME "cros_scp" #define CROS_EC_DEV_TP_NAME "cros_tp"Is really the named
cros_ecthe only that has this feature? What aboutcros_scp(is not supposed to run the same base code ascros_ec)? Andcros_ish?And all the named
cros_ecdevices have this feature? Maybe is supported by Nocturne but not Veyron? Wouldn't be exposed also on thosecros_ecthat doesn't support it?And don't have the answer to all these questions, hence my concerns.
As per your documentation
+ ... The result will be zero if no suspend
+ transitions have been attempted, or the EC does not support
+ this feature.
Quoted 6 lines
I can accept if _all_ the
CROS_EC_DEV_NAMEsand _only_ theCROS_EC_DEV_NAMEsupports it, is that the case?But if you are anyway exposing this on
CROS_EC_DEV_NAMEsthat doesn't support it why not just expose to all and skip a future bunch of code to filter, it is clear what a 0 means and at the end it's just a debugfs file.
Yes, my patch was attempting to expose it for all "cros_ec" devices, even if the EC didn't support that feature. I was hoping to avoid polluting other devices like cros_fp and cros_pd, but I think you may be correct about cros_scp. That's something that will act as the system EC but is named differently because it can send IPIs as well? Weird, but ok.
Ok, I'll add this debugfs file to all EC devices.
-Evan