From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michael Schmitz Subject: [PATCH 0/4] Atari kernel-in-FastRAM patches, take three Date: Sun, 30 Mar 2014 13:01:22 +1300 Message-ID: <1396137686-32678-1-git-send-email-schmitz@debian.org> Return-path: Received: from mail-pd0-f182.google.com ([209.85.192.182]:57261 "EHLO mail-pd0-f182.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751991AbaC3ABg (ORCPT ); Sat, 29 Mar 2014 20:01:36 -0400 Received: by mail-pd0-f182.google.com with SMTP id y10so6449809pdj.27 for ; Sat, 29 Mar 2014 17:01:36 -0700 (PDT) Sender: linux-m68k-owner@vger.kernel.org List-Id: linux-m68k@vger.kernel.org To: linux-m68k@vger.kernel.org Cc: geert@linux-m68k.org, debian-68k@lists.debian.org Geert, see the following patch series for the hopefully final word on the 'make ST-RAM pool accessible for kernels running from FastRAM' story. ST-RAM pool is allocated from an arch initcall if the kernel runs in FastRAM, at which time memory mapping support is up and running and ioremap works as it should. This should make current kernels run again on TTs with insufficent ST-RAM but enough TT-RAM to hold the kernel and user data. It does speed up kernel operations a great deal on Falcon hardware with FastRAM on accelerators. Side effects: rebooting from a kernel in FastRAM will go through a hard reboot -no idea why. SCSI operations seem to fail sooner than otherwise on my system (the SCSI patches are not in my atafb-stram branch). Need to investigate on both accounts. Please review and apply. Cheers, Michael