linux-kernel - `ohci_rh_resume()` called during `usb_dev

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <4d24d2f7-6123-ae18-7055-b5d2f50bb964@molgen.mpg.de>
Date:   Tue, 24 Jul 2018 01:19:19 +0200
From:   Paul Menzel <pmenzel+linux-usb@...gen.mpg.de>
To:     Alan Stern <stern@...land.harvard.edu>
Cc:     linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: `ohci_rh_resume()` called during `usb_dev_suspend()`

Dear Linux folks,


Profiling the suspend to and resume/wake-up from ACPI S3 with 
`sleepgraph.py` on an ASRock E350M1, I noticed that resume methods are 
called during suspend.

Here is an excerpt from the callgraph.

>   600.376906 |   0)   kworker-81   |               |  /* device_pm_callback_start: usb usb5, parent: 0000:00:14.5, type [suspend] */
>   600.376909 |   0)   kworker-81   |               |    usb_dev_suspend() {
>   600.376911 |   0)   kworker-81   |               |      usb_suspend() {
>   600.376913 |   0)   kworker-81   |               |        __pm_runtime_resume() {
>   600.376915 |   0)   kworker-81   |               |          _cond_resched() {
>   600.376917 |   0)   kworker-81   |   0.565 us    |            rcu_all_qs();
>   600.376921 |   0)   kworker-81   |   4.034 us    |          } /* _cond_resched */
>   600.376922 |   0)   kworker-81   |   0.505 us    |          _raw_spin_lock_irqsave();
>   600.376926 |   0)   kworker-81   |               |          rpm_resume() {
>   600.376928 |   0)   kworker-81   |   0.573 us    |            _raw_spin_lock();
>   600.376934 |   0)   kworker-81   |   0.706 us    |            rpm_resume();
>   600.376937 |   0)   kworker-81   |   0.556 us    |            _raw_spin_lock();
>   600.376942 |   0)   kworker-81   |   0.721 us    |            __rpm_get_callback();
>   600.376946 |   0)   kworker-81   |   0.564 us    |            dev_pm_disable_wake_irq_check();
>   600.376949 |   0)   kworker-81   |               |            rpm_callback() {
>   600.376952 |   0)   kworker-81   |               |              __rpm_callback() {
>   600.376954 |   0)   kworker-81   |               |                usb_runtime_resume() {
>   600.376956 |   0)   kworker-81   |               |                  usb_resume_both() {
>   600.376959 |   0)   kworker-81   |               |                    generic_resume() {
>   600.376960 |   0)   kworker-81   |               |                      hcd_bus_resume() {
>   600.376963 |   0)   kworker-81   |               |                        ohci_bus_resume [ohci_hcd]() {
>   600.376964 |   0)   kworker-81   |   0.588 us    |                          _raw_spin_lock_irq();
>   600.376968 |   0)   kworker-81   |               |                          ohci_rh_resume [ohci_hcd]() {
>   600.377043 |   0)   kworker-81   |               |                            msleep() {

Please find the full callgraph and the HTML output attached.

Is that expected?


Kind regards,

Paul


[1]: https://01.org/suspendresume

Download attachment "kodi_mem.html.7z" of type "application/x-7z-compressed" (41725 bytes)

View attachment "kodi_mem_ftrace-ohci.txt" of type "text/plain" (35634 bytes)