From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeremy Fitzhardinge Subject: Re: [PATCH 16/16] make vsmp a paravirt client Date: Wed, 31 Oct 2007 21:38:30 -0700 Message-ID: <47295846.8000700@goop.org> References: <1193858101367-git-send-email-gcosta@redhat.com> <11938581073775-git-send-email-gcosta@redhat.com> <11938581133479-git-send-email-gcosta@redhat.com> <1193858118284-git-send-email-gcosta@redhat.com> <11938581242810-git-send-email-gcosta@redhat.com> <1193858130443-git-send-email-gcosta@redhat.com> <11938581351064-git-send-email-gcosta@redhat.com> <11938581403478-git-send-email-gcosta@redhat.com> <11938581451804-git-send-email-gcosta@redhat.com> <11938581512730-git-send-email-gcosta@redhat.com> <11938581562552-git-send-email-gcosta@redhat.com> <11938581612389-git-send-email-gcosta@redhat.com> <11938581663639-git-send-email-gcosta@redhat.com> <11938581713998-git-send-email-gcosta@redhat.com> <11938581762150-git-send-email-gcosta@redhat.com> <11938581812241-git-send-email-gcosta@redhat.com> <1193858187364-git-send-email-gcosta@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: zach-pghWNbHTmq7QT0dZR+AlfA@public.gmane.org, lguest-mnsaURCQ41sdnm+yROfE0A@public.gmane.org, kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org, ak-l3A5Bk7waGM@public.gmane.org, akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org, virtualization-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org To: Glauber de Oliveira Costa Return-path: In-Reply-To: <1193858187364-git-send-email-gcosta-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Errors-To: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Id: kvm.vger.kernel.org Glauber de Oliveira Costa wrote: > This patch makes vsmp a paravirt client. It now uses the whole > infrastructure provided by pvops. When we detect we're running > a vsmp box, we change the irq-related paravirt operations (and so, > it have to happen quite early), and the patching function > > Signed-off-by: Glauber de Oliveira Costa > Signed-off-by: Steven Rostedt > Acked-by: Jeremy Fitzhardinge > --- > arch/x86/Kconfig.x86_64 | 3 +- > arch/x86/kernel/setup_64.c | 3 ++ > arch/x86/kernel/vsmp_64.c | 72 +++++++++++++++++++++++++++++++++++++++---- > include/asm-x86/setup.h | 3 +- > 4 files changed, 71 insertions(+), 10 deletions(-) > > diff --git a/arch/x86/Kconfig.x86_64 b/arch/x86/Kconfig.x86_64 > index 04734dd..544bad5 100644 > --- a/arch/x86/Kconfig.x86_64 > +++ b/arch/x86/Kconfig.x86_64 > @@ -148,15 +148,14 @@ config X86_PC > bool "PC-compatible" > help > Choose this option if your computer is a standard PC or compatible. > - > config X86_VSMP > bool "Support for ScaleMP vSMP" > depends on PCI > + select PARAVIRT > help > Support for ScaleMP vSMP systems. Say 'Y' here if this kernel is > supposed to run on these EM64T-based machines. Only choose this option > if you have one of these machines. > - > endchoice > > choice > diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c > index 44a11e3..c522549 100644 > --- a/arch/x86/kernel/setup_64.c > +++ b/arch/x86/kernel/setup_64.c > @@ -335,6 +335,9 @@ void __init setup_arch(char **cmdline_p) > > init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT)); > > +#ifdef CONFIG_VSMP > + vsmp_init(); > +#endif > dmi_scan_machine(); > > #ifdef CONFIG_SMP > diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c > index 414caf0..547d3b3 100644 > --- a/arch/x86/kernel/vsmp_64.c > +++ b/arch/x86/kernel/vsmp_64.c > @@ -8,18 +8,70 @@ > * > * Ravikiran Thirumalai , > * Shai Fultheim > + * Paravirt ops integration: Glauber de Oliveira Costa > */ > - > #include > #include > #include > #include > #include > +#include > + > +/* > + * Interrupt control for the VSMP architecture: > + */ > + > +static inline unsigned long vsmp_save_fl(void) > No point being inline. > +{ > + unsigned long flags = native_save_fl(); > + > + if (flags & X86_EFLAGS_IF) > + return X86_EFLAGS_IF; > Is this right, or should the if be testing _AC? Otherwise, why not just "return flags & X86_EFLAGS_IF"? > + return 0; > +} > > -static int __init vsmp_init(void) > +static inline void vsmp_restore_fl(unsigned long flags) > +{ > + if (flags & X86_EFLAGS_IF) > + flags &= ~X86_EFLAGS_AC; > + if (!(flags & X86_EFLAGS_IF)) > + flags &= X86_EFLAGS_AC; > Just use "else"? > + native_restore_fl(flags); > +} > + > +static inline void vsmp_irq_disable(void) > +{ > + unsigned long flags = native_save_fl(); > + > + vsmp_restore_fl((flags & ~X86_EFLAGS_IF)); > ((double paren?)) > +} > + > +static inline void vsmp_irq_enable(void) > +{ > + unsigned long flags = native_save_fl(); > + > + vsmp_restore_fl((flags | X86_EFLAGS_IF)); > +} > + > +static unsigned __init vsmp_patch(u8 type, u16 clobbers, void *ibuf, > + unsigned long addr, unsigned len) > +{ > + switch (type) { > + case PARAVIRT_PATCH(pv_irq_ops.irq_enable): > + case PARAVIRT_PATCH(pv_irq_ops.irq_disable): > + case PARAVIRT_PATCH(pv_irq_ops.save_fl): > + case PARAVIRT_PATCH(pv_irq_ops.restore_fl): > + return paravirt_patch_default(type, clobbers, ibuf, addr, len); > + default: > + return native_patch(type, clobbers, ibuf, addr, len); > + } > + > +} > + > +int __init vsmp_init(void) > { > void *address; > - unsigned int cap, ctl; > + unsigned int cap, ctl, cfg; > > if (!early_pci_allowed()) > return 0; > @@ -29,8 +81,16 @@ static int __init vsmp_init(void) > (read_pci_config_16(0, 0x1f, 0, PCI_DEVICE_ID) != PCI_DEVICE_ID_SCALEMP_VSMP_CTL)) > return 0; > > + /* If we are, use the distinguished irq functions */ > + pv_irq_ops.irq_disable = vsmp_irq_disable; > + pv_irq_ops.irq_enable = vsmp_irq_enable; > + pv_irq_ops.save_fl = vsmp_save_fl; > + pv_irq_ops.restore_fl = vsmp_restore_fl; > + pv_init_ops.patch = vsmp_patch; > + > /* set vSMP magic bits to indicate vSMP capable kernel */ > - address = ioremap(read_pci_config(0, 0x1f, 0, PCI_BASE_ADDRESS_0), 8); > + cfg = read_pci_config(0, 0x1f, 0, PCI_BASE_ADDRESS_0); > + address = early_ioremap(cfg, 8); > cap = readl(address); > ctl = readl(address + 4); > printk("vSMP CTL: capabilities:0x%08x control:0x%08x\n", cap, ctl); > @@ -42,8 +102,6 @@ static int __init vsmp_init(void) > printk("vSMP CTL: control set to:0x%08x\n", ctl); > } > > - iounmap(address); > + early_iounmap(address, 8); > return 0; > } > - > -core_initcall(vsmp_init); > diff --git a/include/asm-x86/setup.h b/include/asm-x86/setup.h > index 071e054..dd7996c 100644 > --- a/include/asm-x86/setup.h > +++ b/include/asm-x86/setup.h > @@ -58,7 +58,8 @@ void __init add_memory_region(unsigned long long start, > > extern unsigned long init_pg_tables_end; > > - > +/* For EM64T-based VSMP machines */ > +int vsmp_init(void); > > #endif /* __i386__ */ > #endif /* _SETUP */ > ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/