Skip to content
This repository has been archived by the owner on Nov 6, 2023. It is now read-only.

keeping adb server/daemon running causes dmesg PME spam every 1 second #18103

Closed
ghost opened this issue Jun 19, 2019 · 0 comments
Closed

keeping adb server/daemon running causes dmesg PME spam every 1 second #18103

ghost opened this issue Jun 19, 2019 · 0 comments

Comments

@ghost
Copy link

ghost commented Jun 19, 2019

Type: code issue

This is the dmesg spam when adb server(aka daemon) is running:

[ 1460.763742] pcieport 0000:00:1c.4: restoring config space at offset 0x2c (was 0x0, writing 0x0)
[ 1460.763748] pcieport 0000:00:1c.4: restoring config space at offset 0x28 (was 0x0, writing 0x0)
[ 1460.763752] pcieport 0000:00:1c.4: restoring config space at offset 0x24 (was 0x1fff1, writing 0x1fff1)
[ 1460.763813] pcieport 0000:00:1c.4: PME# disabled
[ 1460.763920] xhci_hcd 0000:03:00.0: PME# disabled
[ 1460.763928] xhci_hcd 0000:03:00.0: enabling bus mastering
[ 1460.770239] xhci_hcd 0000:03:00.0: PME# enabled
[ 1460.770377] xhci_hcd 0000:03:00.0: PME# disabled
[ 1460.770383] xhci_hcd 0000:03:00.0: enabling bus mastering
[ 1460.884813] xhci_hcd 0000:03:00.0: PME# enabled
[ 1460.884925] pcieport 0000:00:1c.4: PME# enabled
[ 1461.896751] pcieport 0000:00:1c.4: restoring config space at offset 0x2c (was 0x0, writing 0x0)
[ 1461.896764] pcieport 0000:00:1c.4: restoring config space at offset 0x28 (was 0x0, writing 0x0)
[ 1461.896772] pcieport 0000:00:1c.4: restoring config space at offset 0x24 (was 0x1fff1, writing 0x1fff1)
[ 1461.896863] pcieport 0000:00:1c.4: PME# disabled
[ 1461.897001] xhci_hcd 0000:03:00.0: PME# disabled
[ 1461.897014] xhci_hcd 0000:03:00.0: enabling bus mastering
[ 1461.903257] xhci_hcd 0000:03:00.0: PME# enabled
[ 1461.903477] xhci_hcd 0000:03:00.0: PME# disabled
[ 1461.903490] xhci_hcd 0000:03:00.0: enabling bus mastering
[ 1462.012901] xhci_hcd 0000:03:00.0: PME# enabled
[ 1462.013021] pcieport 0000:00:1c.4: PME# enabled
[ 1463.025739] pcieport 0000:00:1c.4: restoring config space at offset 0x2c (was 0x0, writing 0x0)
[ 1463.025751] pcieport 0000:00:1c.4: restoring config space at offset 0x28 (was 0x0, writing 0x0)
[ 1463.025760] pcieport 0000:00:1c.4: restoring config space at offset 0x24 (was 0x1fff1, writing 0x1fff1)
[ 1463.025844] pcieport 0000:00:1c.4: PME# disabled
[ 1463.025980] xhci_hcd 0000:03:00.0: PME# disabled
[ 1463.025992] xhci_hcd 0000:03:00.0: enabling bus mastering
[ 1463.032196] xhci_hcd 0000:03:00.0: PME# enabled
[ 1463.032313] xhci_hcd 0000:03:00.0: PME# disabled
[ 1463.032318] xhci_hcd 0000:03:00.0: enabling bus mastering
[ 1463.140830] xhci_hcd 0000:03:00.0: PME# enabled
[ 1463.140956] pcieport 0000:00:1c.4: PME# enabled

and it's happening every second unless either of the two devices are set to stay awake full time (via powertop). Details in this gist: https://gist.github.com/howaboutsynergy/35d0d2de2f7a692f874216887a9e4fcd

adb server/daemon is left running by utils/android-push.sh when it executes adb devices (or any adb command really)

I'll make a PR to adb kill-server iff the script utils/android-push.sh is the one that started it. Ergo, if adb server was already running(started by user, manually, for example) we do not kill it. (user wouldn't want that, for sure)

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

1 participant