svn commit: r246856 - head/etc

[Gleb Smirnoff]

On Thu, Mar 07, 2013 at 12:07:56PM +0400, Gleb Smirnoff wrote:
T> On Wed, Mar 06, 2013 at 01:42:54PM -0500, Jung-uk Kim wrote:
T> J> You know, I disagreed with the above change because this type of
T> J> quirks can be easily handled by the driver.  IMHO, /etc/rc.resume is
T> J> the last place to put it.  When I committed r233580 to fix resuming
T> J> psm(4) and MFC'd it to stable/9 (r234713), we had an agreement to back
T> J> it out if we hear no user complaints for enough time *without* merging
T> J> your /etc/rc.resume changes.  At least, that's what I remember.  In
T> J> fact, you never MFC'd your changes.  So, I thought the agreement is
T> J> still valid and it is safe to back out.  I am sorry for not asking
T> J> your confirmation, though.
T> J> 
T> J> Now, please consider fixing psm.  If you can't find a generic
T> J> solution, it is okay to add your model-specific quirk in
T> J> sys/dev/atkbdc/psm.c like this:
T> J> 
T> J> http://svnweb.freebsd.org/base/head/sys/dev/atkbdc/psm.c?annotate=233580#l1441
T> J> 
T> J> Also, please mention it in the manual page, i.e., the CAVEATS section of
T> J> psm(4).
T> 
T> On my laptop the psm(4) is a generic mouse:
T> 
T> psm0: <PS/2 Mouse> irq 12 on atkbdc0
T> psm0: [GIANT-LOCKED]
T> psm0: model Generic PS/2 mouse, device ID 0
T> 
T> Do you suggest to add PSM_CONFIG_INITAFTERSUSPEND flag for all generic mice?

P.S. I've tested this and it works.

Index: psm.c
===================================================================
--- psm.c       (revision 247911)
+++ psm.c       (working copy)
@@ -1449,6 +1449,7 @@
        case MOUSE_MODEL_SYNAPTICS:
        case MOUSE_MODEL_GLIDEPOINT:
        case MOUSE_MODEL_VERSAPAD:
+       case MOUSE_MODEL_GENERIC:
                sc->config |= PSM_CONFIG_INITAFTERSUSPEND;
                break;
        default:

-- 
Totus tuus, Glebius.