Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754950AbYHVS4T (ORCPT ); Fri, 22 Aug 2008 14:56:19 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751355AbYHVS4K (ORCPT ); Fri, 22 Aug 2008 14:56:10 -0400 Received: from rv-out-0506.google.com ([209.85.198.234]:59818 "EHLO rv-out-0506.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751928AbYHVS4I (ORCPT ); Fri, 22 Aug 2008 14:56:08 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version :content-type:content-transfer-encoding:content-disposition :references; b=ofhot60jiHmsKHAxJSR3mNuh+z/XwcGmjz2ZISBKIXUhVmTn8g5OI6scWI8vTkgDWf S5NjZ7osJtCJtHSzoxfWaNmZ1oFEYeBG8kHClwGKxlt9vH8Pt7fa5L47RB2iEdORI5kT Sf7P/NsyNGjdioRGe2/gNv0g4JoGRpH6c5gns= Message-ID: <86802c440808221156h7ef48973hb5db8cf2b01ac27b@mail.gmail.com> Date: Fri, 22 Aug 2008 11:56:08 -0700 From: "Yinghai Lu" To: "Robert Richter" Subject: Re: [PATCH] x86: fix: make PCI ECS for AMD CPUs hotplug capable Cc: "Ingo Molnar" , "Thomas Gleixner" , LKML , "Jan Beulich" In-Reply-To: <1219429418-29023-2-git-send-email-robert.richter@amd.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <48AD82D5.76E4.0078.0@novell.com> <1219429418-29023-2-git-send-email-robert.richter@amd.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3567 Lines: 108 On Fri, Aug 22, 2008 at 11:23 AM, Robert Richter wrote: > Until now, PCI ECS setup was performed at boot time only and for cpus > that are enabled then. This patch fixes this and adds cpu hotplug. > > Tests sequence (check if ECS bit is set when bringing cpu online again): > > # ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) < /dev/cpu/1/msr > 00000008 00404010 > # ( perl -e 'sysseek(STDOUT, 0xC001001F, 0); print pack "l*", 8, 0x00400010' ) > /dev/cpu/1/msr > # ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) < /dev/cpu/1/msr > 00000008 00400010 > # echo 0 > /sys/devices/system/cpu/cpu1/online > # echo 1 > /sys/devices/system/cpu/cpu1/online > # ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) < /dev/cpu/1/msr > 00000008 00404010 > > Cc: Jan Beulich > Cc: Yinghai Lu > Signed-off-by: Robert Richter > --- > arch/x86/pci/amd_bus.c | 35 +++++++++++++++++++++++++++++++---- > 1 files changed, 31 insertions(+), 4 deletions(-) > > diff --git a/arch/x86/pci/amd_bus.c b/arch/x86/pci/amd_bus.c > index 4a6f1a6..6a0fca7 100644 > --- a/arch/x86/pci/amd_bus.c > +++ b/arch/x86/pci/amd_bus.c > @@ -1,6 +1,7 @@ > #include > #include > #include > +#include > #include "pci.h" > > #ifdef CONFIG_X86_64 > @@ -565,7 +566,7 @@ static int __init early_fill_mp_bus_info(void) { return 0; } > > #define ENABLE_CF8_EXT_CFG (1ULL << 46) > > -static void enable_pci_io_ecs_per_cpu(void *unused) > +static void enable_pci_io_ecs(void *unused) > { > u64 reg; > rdmsrl(MSR_AMD64_NB_CFG, reg); > @@ -575,13 +576,39 @@ static void enable_pci_io_ecs_per_cpu(void *unused) > } > } > > -static int __init enable_pci_io_ecs(void) > +static int __cpuinit amd_cpu_notify(struct notifier_block *self, > + unsigned long action, void *hcpu) > { > + int cpu = (long)hcpu; > + switch(action) { > + case CPU_ONLINE: > + case CPU_ONLINE_FROZEN: > + smp_call_function_single(cpu, enable_pci_io_ecs, NULL, 0); > + break; > + default: > + break; > + } > + return NOTIFY_OK; > +} > + > +static struct notifier_block __cpuinitdata amd_cpu_notifier = { > + .notifier_call = amd_cpu_notify, > +}; > + > +static int __init pci_io_ecs_init(void) > +{ > + int cpu; > + > /* assume all cpus from fam10h have IO ECS */ > if (boot_cpu_data.x86 < 0x10) > return 0; > - on_each_cpu(enable_pci_io_ecs_per_cpu, NULL, 1); > + > + register_cpu_notifier(&amd_cpu_notifier); > + for_each_online_cpu(cpu) > + amd_cpu_notify(&amd_cpu_notifier, (unsigned long)CPU_ONLINE, > + (void *)(long)cpu); wonder if those two lines should be reversed. > pci_probe |= PCI_HAS_IO_ECS; > + > return 0; > } > > @@ -591,7 +618,7 @@ static int __init amd_postcore_init(void) > return 0; > > early_fill_mp_bus_info(); > - enable_pci_io_ecs(); > + pci_io_ecs_init(); > > return 0; > } > -- > 1.5.6.4 > > > -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/