Eugene Surovegin wrote: > On Fri, Jun 03, 2005 at 04:22:40PM -0700, Geoff Levand wrote: > > [snip] > > >>+ /* save current CPM */ >>+ cpm_save_er = mfdcr(DCRN_CPC0_ER); >>+ >>+ /* save UIC0 enable registers */ >>+ uic_save_er = mfdcr(DCRN_UIC_ER(UIC0)); >>+ >>+#ifdef USE_ETHER_TO_RESUME >>+ mtdcr(DCRN_UIC_ER(UIC0), UIC0_EIR5_BIT|UIC0_UIC1NC_BIT); >>+#else >>+ /* mask UIC0 interrupts, except External Intr #5 */ >>+ mtdcr(DCRN_UIC_ER(UIC0), UIC0_EIR5_BIT); >>+#endif > > > Why UIC PM code is here and not in ppc4xx_pic.c? I don't think this is > the right place to mess with UIC registers. >
Yes, that is resonable. Maybe something like ppc4xx_pic_suspend() and ppc4xx_pic_resume(). > > >>=================================================================== >>--- linux-2.6.12-bhpm.orig/arch/ppc/platforms/4xx/ibm440gp_sleep.S > > 2005-06-01 08:52:49.947684744 -0700 > >>+++ linux-2.6.12-bhpm/arch/ppc/platforms/4xx/ibm440gp_sleep.S > > 2005-06-03 16:15:07.000000000 -0700 > > I think it should be in arch/ppc/syslib not in arch/ppc/platforms/4xx. > I was thinking the same... I want to do some cleanup of the debugging code and reformat the text, then I'll post a new version with these changes. -Geoff