regulator: core: Clean enabling always-on regulators + their supplies
PATCH 1/1 regulator: core: Clean enabling always-on regulators + their supplies
At the end of regulator_resolve_supply() we have historically turned on our supply in some cases. This could be for one of two reasons:
1. If resolving supplies was happening before the call to set_machine_constraints() we needed to predict if set_machine_constraints() was going to turn the regulator on and we needed to preemptively turn the supply on.
2. Maybe set_machine_constraints() happened before we could resolve supplies (because we failed the first time to resolve) and thus we might need to propagate an enable that already happened up to our supply.
Historically regulator_resolve_supply() used _regulator_is_enabled() to decide whether to turn on the supply.
Let's change things a little bit. Specifically:
1. Let's try to enable the supply and the regulator in the same place, both in set_machine_constraints(). This means that we have exactly the same logic for enabling the supply and the regulator.
2. Let's properly set use_count when we enable always-on or boot-on regulators even for those that don't have supplies. The previous commit 1fc12b05895e ("regulator: core: Avoid propagating to supplies when possible") only did this right for regulators with supplies.
3. Let's make it clear that the only time we need to enable the supply in regulator_resolve_supply() is if the main regulator is currently in use. By using use_count (like the rest of the code) to decide if we're going to enable our supply we keep everything consistent.
Overall the new scheme should be cleaner and easier to reason about. In addition to fixing regulator_summary to be more correct (because of the more correct use_count), this change also has the effect of no longer using _regulator_is_enabled() in this code path. _regulator_is_enabled() could return an error code for some regulators at bootup (like RPMh) that can't read their initial state. While one can argue that the design of those regulators is sub-optimal, the new logic sidesteps this brokenness. This fix in particular fixes observed problems on Qualcomm sdm845 boards which use the above-mentioned RPMh regulator. Those problems were made worse by commit 1fc12b05895e ("regulator: core: Avoid propagating to supplies when possible") because now we'd think at bootup that the SD regulators were already enabled and we'd never try them again.
---
drivers/regulator/core.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-)
drivers/regulator/core.c +16 -3
@@ -1401,11 +1401,21 @@ static int set_machine_constraints(struct regulator_dev *rdev,* and we have control then make sure it is enabled. */ if (rdev->constraints->always_on || rdev->constraints->boot_on) { +if (rdev->supply) { + ret =regulator_enable (rdev->supply); +if (ret <0 ) { +_regulator_put (rdev->supply); + rdev->supply =NULL ; +return ret; + } + } + ret =_regulator_do_enable (rdev);if (ret <0 && ret != -EINVAL ) {rdev_err (rdev,"failed to enable\n" );return ret; } + rdev->use_count++; }print_constraints (rdev);
@@ -1822,15 +1832,18 @@ static int regulator_resolve_supply(struct regulator_dev *rdev)return ret; } -/* Cascade always-on state to supply */ -if (_regulator_is_enabled (rdev)) { +/* +* In set_machine_constraints() we may have turned this regulator on +* but we couldn't propagate to the supply if it hadn't been resolved +* yet. Do it now. +*/ +if (rdev->use_count) { ret =regulator_enable (rdev->supply);if (ret <0 ) {_regulator_put (rdev->supply); rdev->supply =NULL ;return ret; } - rdev->use_count =1 ; }return 0 ;
Signature · 1 line
2.20.0.rc2.403.gdbc3b29805-goog
Quoted 44 lines · Douglas Anderson
At the end of
regulator_resolve_supply()we have historically turned on our supply in some cases. This could be for one of two reasons:1. If resolving supplies was happening before the call to
set_machine_constraints()we needed to predict ifset_machine_constraints()was going to turn the regulator on and we needed to preemptively turn the supply on.
2. Maybeset_machine_constraints()happened before we could resolve supplies (because we failed the first time to resolve) and thus we might need to propagate an enable that already happened up to our supply.Historically
regulator_resolve_supply()used_regulator_is_enabled()to decide whether to turn on the supply.Let's change things a little bit. Specifically:
1. Let's try to enable the supply and the regulator in the same place, both in
set_machine_constraints(). This means that we have exactly the same logic for enabling the supply and the regulator.
2. Let's properly setuse_countwhen we enable always-on or boot-on regulators even for those that don't have supplies. The previous commit1fc12b05895e("regulator: core: Avoid propagating to supplies when possible") only did this right for regulators with supplies.
3. Let's make it clear that the only time we need to enable the supply inregulator_resolve_supply()is if the main regulator is currently in use. By usinguse_count(like the rest of the code) to decide if we're going to enable our supply we keep everything consistent.Overall the new scheme should be cleaner and easier to reason about. In addition to fixing
regulator_summaryto be more correct (because of the more correctuse_count), this change also has the effect of no longer using_regulator_is_enabled()in this code path._regulator_is_enabled()could return an error code for some regulators at bootup (like RPMh) that can't read their initial state. While one can argue that the design of those regulators is sub-optimal, the new logic sidesteps this brokenness. This fix in particular fixes observed problems on Qualcomm sdm845 boards which use the above-mentioned RPMh regulator. Those problems were made worse by commit1fc12b05895e("regulator: core: Avoid propagating to supplies when possible") because now we'd think at bootup that the SD regulators were already enabled and we'd never try them again.
Quoted 3 lines
Fixes 1fc12b05895e ("regulator: core: Avoid propagating to supplies when possible")Reported-by Evan Green <[email protected]>Signed-off-by Douglas Anderson <[email protected]>
Looks good on the Nexus 5 (qcom msm8974).
Mark Brown
maintainerQuoted 2 lines · Douglas Anderson
At the end of
regulator_resolve_supply()we have historically turned on our supply in some cases. This could be for one of two reasons:
This doesn't apply against current code, please check and resend.
Mark,
Quoted 6 lines · Mark Brown
Quoted 2 lines · Douglas Anderson
At the end of
regulator_resolve_supply()we have historically turned on our supply in some cases. This could be for one of two reasons:This doesn't apply against current code, please check and resend.
Can you clarify? See below where it's applying cleanly to "for-next" for me:
$ git remote -v | grep linuxregulator linuxregulator git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator.git (fetch) linuxregulator git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator.git (push)
$ git fetch linuxregulator
$ git log --oneline -1 linuxregulator/for-next f00e93a1224f (linuxregulator/for-next) Merge remote-tracking branch 'regulator/topic/coupled' into regulator-next
$ git checkout linuxregulator/for-next [ ... cut ... ] HEAD is now at f00e93a1224f Merge remote-tracking branch 'regulator/topic/coupled' into regulator-next
$ pwclient git-am -p lkml 1022439 Applying patch #1022439 using u'git am' Description: regulator: core: Clean enabling always-on regulators + their supplies Applying: regulator: core: Clean enabling always-on regulators + their supplies
Thanks!
-Doug
Mark Brown
maintainerQuoted 9 lines · Doug Anderson
Quoted 5 lines · Mark Brown
Quoted 2 lines · Douglas Anderson
At the end of
regulator_resolve_supply()we have historically turned on our supply in some cases. This could be for one of two reasons:This doesn't apply against current code, please check and resend.
Can you clarify? See below where it's applying cleanly to "for-next" for me:
I tried a git am on my for-4.21 and for-next branches and it didn't work on either, I didn't investigate beyond that.
Mark Brown
maintainerQuoted 6 lines · Mark Brown
Quoted 1 line · Doug Anderson
Can you clarify? See below where it's applying cleanly to "for-next" for me:
I tried a git am on my for-4.21 and for-next branches and it didn't work on either, I didn't investigate beyond that.
Applying: regulator: core: Clean enabling always-on regulators + their supplies Using index info to reconstruct a base tree...
M drivers/regulator/core.c
Falling back to patching base and 3-way merge...
Auto-merging drivers/regulator/core.c
CONFLICT (content): Merge conflict in drivers/regulator/core.c
Recorded preimage for 'drivers/regulator/core.c'
error: Failed to merge in the changes.
Patch failed at 0001 regulator: core: Clean enabling always-on regulators + their supplies
hint: Use 'git am --show-current-patch' to see the failed patch
Mark,
Quoted 9 lines · Mark Brown
Quoted 6 lines · Mark Brown
Quoted 1 line · Doug Anderson
Can you clarify? See below where it's applying cleanly to "for-next" for me:
I tried a git am on my for-4.21 and for-next branches and it didn't work on either, I didn't investigate beyond that.
Quoted 10 lines
Applying: regulator: core: Clean enabling always-on regulators + their supplies Using index info to reconstruct a base tree...
Mdrivers/regulator/core.c
Falling back to patching base and 3-way merge...
Auto-mergingdrivers/regulator/core.c
CONFLICT (content): Merge conflict indrivers/regulator/core.c
Recorded preimage for 'drivers/regulator/core.c'
error: Failed to merge in the changes.
Patch failed at 0001 regulator: core: Clean enabling always-on regulators + their supplies
hint: Use 'git am --show-current-patch' to see the failed patch
Can you confirm my patch applies cleanly to your for-next branch? I can see that it doesn't apply to your 4.21 branch but I can't see it fail on your for-next branch.
The problem is that commit 2bb166636933 ("regulator: core: enable power when setting up constraints") touches the same code but isn't in the 4.21 branch. If you pick that to your 4.21 then my patch applies cleanly.
If you'd prefer that I post my patch atop 4.21 I can do that tomorrow.
-Doug
Mark Brown
maintainerQuoted 3 lines · Doug Anderson
Can you confirm my patch applies cleanly to your for-next branch? I can see that it doesn't apply to your 4.21 branch but I can't see it fail on your for-next branch.
No, I tried both and it didn't work on either.
Quoted 6 lines
The problem is that commit
2bb166636933("regulator: core: enable power when setting up constraints") touches the same code but isn't in the 4.21 branch. If you pick that to your 4.21 then my patch applies cleanly.If you'd prefer that I post my patch atop 4.21 I can do that tomorrow.
I'd prefer that you resend like I said :(
Hi,
Quoted 8 lines · Mark Brown
Quoted 3 lines · Doug Anderson
Can you confirm my patch applies cleanly to your for-next branch? I can see that it doesn't apply to your 4.21 branch but I can't see it fail on your for-next branch.
No, I tried both and it didn't work on either.
I'm totally confused because I can't reproduce, but OK.
Quoted 8 lines
Quoted 6 lines
The problem is that commit
2bb166636933("regulator: core: enable power when setting up constraints") touches the same code but isn't in the 4.21 branch. If you pick that to your 4.21 then my patch applies cleanly.If you'd prefer that I post my patch atop 4.21 I can do that tomorrow.
I'd prefer that you resend like I said :(
If I can't reproduce the merge conflict that you're seeing I'm worried that I'll just be sending you the exact same patch I sent before and it'll have the same problem. That's why I've been so focused on trying to understand.
In any case it sounds as if sending it up against 4.21 should be fine and maybe is preferred (since the patch it Fixes is in 4.21), so here comes v2 against for-4.21. OK, posted at
<https://lkml.kernel.org/r/[email protected]>
-Doug