Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1312893
| From | Jeff Merkey <linux.mdb@gmail.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [PATCH v2] Fix Hard Lockup in ktime_get_ts64 with debuggers |
| Date | 2016-01-20 06:50 +0100 |
| Message-ID | <qSTOy-25l-9@gated-at.bofh.it> (permalink) |
| References | <qSSpv-1b4-89@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
I just finished a 4 hour regression test of this patch and please drop it. It does fix the symptom but not the illness. It worked on i386 and x86_64 for my first test but failed the second test. I did a 4 hour debugging session and the problem just shows up another way. This patch gets rid of the hard lockup but it results in the gnome gui and systemd eventually locking up after entering and exiting the debugger and holding the system suspended for several hours. This tells me that whatever is causing that value to get so large with the clock skew is affecting other areas of the timekeeping logic. I will keep working on it since I need this to work. kgdb also experiences the same problem. So drop this patch and I'll keep looking for the sickness rather than treat the symtoms. If you guys have any ideas let me know. Jeff
Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread
[PATCH v2] Fix Hard Lockup in ktime_get_ts64 with debuggers Jeff Merkey <linux.mdb@gmail.com> - 2016-01-20 05:20 +0100 Re: [PATCH v2] Fix Hard Lockup in ktime_get_ts64 with debuggers Jeff Merkey <linux.mdb@gmail.com> - 2016-01-20 06:50 +0100
csiph-web