>> But this time, BSP use memory semaphore to do that but not an IPI. >> En, maybe something I lost, hope your comments. >I do not fully understand the issue. Did the APs returned to SAL ? If my memory is right, it seems loop in OS. I will check it again in native.
Good good study,day day up ! ^_^ -Wing(zhang xin) OTC,Intel Corporation >-----Original Message----- >From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] >Sent: 2007年4月5日 15:41 >To: Zhang, Xing Z >Cc: Kouya SHIMURA; Tristan Gingold; xen-ia64-devel >Subject: RE: [Xen-ia64-devel][Patch]Add two PAL calls which >fixSMPwindowsinstallation crashing bug > >Quoting "Zhang, Xing Z" <[EMAIL PROTECTED]>: > >> Hi Kouya and Tristan: >> I still have a question. After "11.XEN stops the vcpu", APs wait in Xen, >how >> does BSP wake up them again? I don't think BSP will send an IPI to them >> again. >It should. That's the method to awake it. > >> In windows, seems BSP will wait a very short time for AP waking up. Whiles >> APs waked up, they will fall into a loop to wait another times waking up by >> BSP. >> But this time, BSP use memory semaphore to do that but not an IPI. >> En, maybe something I lost, hope your comments. >I do not fully understand the issue. Did the APs returned to SAL ? > >Tristan. _______________________________________________ Xen-ia64-devel mailing list [email protected] http://lists.xensource.com/xen-ia64-devel
