Messages in this thread | | | From | Bartlomiej Zolnierkiewicz <> | Subject | Re: [PATCH 1/2] Revert "intel_idle: mark states tables with __initdata tag" | Date | Fri, 10 Jan 2014 10:23:36 +0100 |
| |
On Thursday, January 09, 2014 02:38:14 PM Rafael J. Wysocki wrote: > Hi, > > On Thursday, January 09, 2014 02:20:22 PM Bartlomiej Zolnierkiewicz wrote: > > > > Hi, > > > > On Thursday, January 09, 2014 03:30:26 PM Jiang Liu wrote: > > > This reverts commit 9d046ccb98085f1d437585f84748c783a04ba240. > > > > > > Commit 9d046ccb98085 marks all state tables with __initdata, but > > > the state table may be accessed when doing CPU online, which then > > > causing system crash as below: > > > > Uh, sorry for that - it most likely got missed since I tested it > > together with intel_idle_cpu_init() removal patches (they are in > > Rafael's PM tree now). > > > > Anyway, better than reverting it altogether would be to fix it by > > backporting the following patch: > > > > http://lkml.org/lkml/2013/12/20/372 > > > > Could you please try to solve the issue this way? > > No, it's too late for that as far as 3.13 is concerned. > > I have the patch above queued up for 3.14, though, so do you think that the > commit being reverted here can be safely re-applied on top of it?
Yes, the reverted commit can be safely re-applied on top of this patch.
Best regards, -- Bartlomiej Zolnierkiewicz Samsung R&D Institute Poland Samsung Electronics
| |