Re: [Xen-devel] [RFC PATCH V3 2/2] qemu: support xen hvm direct kernel boot

["Chun Yan Liu" <cyliu@xxxxxxxx>]

>>> On 6/20/2014 at 08:08 PM, in message
<alpine.DEB.2.02.1406201300410.6594@xxxxxxxxxxxxxxxxxxxxxxx>, Stefano
Stabellini <stefano.stabellini@xxxxxxxxxxxxx> wrote: 
> On Fri, 20 Jun 2014, Chunyan Liu wrote: 
> > qemu side patch to support xen HVM direct kernel boot: 
> > if -kernel exists, calls xen_load_linux(), which will read kernel/initrd 
> > and add a linuxboot.bin or multiboot.bin option rom. The 
> > linuxboot.bin/multiboot.bin will load kernel/initrd and jump to execute 
> > kernel directly. It's working when xen uses seabios. 
> >  
> > Signed-off-by: Chunyan Liu <cyliu@xxxxxxxx> 
> > --- 
> >  hw/i386/pc.c           | 22 ++++++++++++++++++++++ 
> >  hw/i386/pc_piix.c      |  7 +++++++ 
> >  hw/i386/xen/xen_apic.c |  1 + 
> >  include/hw/i386/pc.h   |  5 +++++ 
> >  4 files changed, 35 insertions(+) 
> >  
> > diff --git a/hw/i386/pc.c b/hw/i386/pc.c 
> > index 3e0ecf1..e005136 100644 
> > --- a/hw/i386/pc.c 
> > +++ b/hw/i386/pc.c 
> > @@ -1183,6 +1183,28 @@ void pc_acpi_init(const char *default_dsdt) 
> >      } 
> >  } 
> >   
> > +FWCfgState *xen_load_linux(const char *kernel_filename, 
> > +                           const char *kernel_cmdline, 
> > +                           const char *initrd_filename, 
> > +                           ram_addr_t below_4g_mem_size, 
> > +                           PcGuestInfo *guest_info) 
> > +{ 
> > +    int i; 
> > +    FWCfgState *fw_cfg; 
> > + 
> > +    assert(kernel_filename != NULL); 
> > + 
> > +    fw_cfg = fw_cfg_init(BIOS_CFG_IOPORT, BIOS_CFG_IOPORT + 1, 0, 0); 
>  
> Is it actually OK to initialize just BIOS_CFG_IOPORT and avoid 
> everything else currently done in bochs_bios_init? 
> Would it make sense to simply call bochs_bios_init? 
>  
>  
> > +    rom_set_fw(fw_cfg); 
> > + 
> > +    load_linux(fw_cfg, kernel_filename, initrd_filename, kernel_cmdline,  
> below_4g_mem_size); 
> > +    for (i = 0; i < nb_option_roms; i++) { 
> > +        rom_add_option(option_rom[i].name, option_rom[i].bootindex); 
> > +    } 
>  
> Wouldn't this have the unintended consequence of possibly loading other 
> option_roms into the guest memory? 

For xen, no.

Before this patch, indeed there is another option_rom "kvmvapic" in the list,
but since the option_rom list is never loaded in xen case, that won't harm. But
as expected, I think in xen case, it should not be in option_rom list at all, 
since
we won't expect to load it. In v1, I tried to check the option_rom name to
bypass it and load multiboot.bin/linuxboot.bin only.

In comments, Paolo Bonzini suggests a better way, that is to add following line
in xen_apic_realize():
+ s->vapic_control = 0;
In this way, "kvmvapic" won't be added to the option_rom list. I think that
should be the normal way for xen.

So, since V2, I updated in this way.
http://lists.gnu.org/archive/html/qemu-devel/2014-06/msg00704.html

>  
>  
> > +    guest_info->fw_cfg = fw_cfg; 
> > +    return fw_cfg; 
> > +} 
> > + 
> >  FWCfgState *pc_memory_init(MemoryRegion *system_memory, 
> >                             const char *kernel_filename, 
> >                             const char *kernel_cmdline, 
> > diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c 
> > index a48e263..b737868 100644 
> > --- a/hw/i386/pc_piix.c 
> > +++ b/hw/i386/pc_piix.c 
> > @@ -158,6 +158,13 @@ static void pc_init1(MachineState *machine, 
> >                         machine->initrd_filename, 
> >                         below_4g_mem_size, above_4g_mem_size, 
> >                         rom_memory, &ram_memory, guest_info); 
> > +    } else if (machine->kernel_filename != NULL) { 
> > +        /* For xen HVM direct kernel boot, load linux here */ 
> > +        fw_cfg = xen_load_linux(machine->kernel_filename, 
> > +                                machine->kernel_cmdline, 
> > +                                machine->initrd_filename, 
> > +                                below_4g_mem_size, 
> > +                                guest_info); 
> >      } 
> >   
> >      gsi_state = g_malloc0(sizeof(*gsi_state)); 
> > diff --git a/hw/i386/xen/xen_apic.c b/hw/i386/xen/xen_apic.c 
> > index 63bb7f7..f5acd6a 100644 
> > --- a/hw/i386/xen/xen_apic.c 
> > +++ b/hw/i386/xen/xen_apic.c 
> > @@ -40,6 +40,7 @@ static void xen_apic_realize(DeviceState *dev, Error  
> **errp) 
> >  { 
> >      APICCommonState *s = APIC_COMMON(dev); 
> >   
> > +    s->vapic_control = 0; 
> >      memory_region_init_io(&s->io_memory, OBJECT(s), &xen_apic_io_ops, s, 
> >                            "xen-apic-msi", APIC_SPACE_SIZE); 
>  
> Why this change? It is not mentioned in the commit message. 
>  
>  
> > diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h 
> > index ca7a0bd..171a597 100644 
> > --- a/include/hw/i386/pc.h 
> > +++ b/include/hw/i386/pc.h 
> > @@ -134,6 +134,11 @@ PcGuestInfo *pc_guest_info_init(ram_addr_t  
> below_4g_mem_size, 
> >  void pc_pci_as_mapping_init(Object *owner, MemoryRegion *system_memory, 
> >                              MemoryRegion *pci_address_space); 
> >   
> > +FWCfgState *xen_load_linux(const char *kernel_filename, 
> > +                           const char *kernel_cmdline, 
> > +                           const char *initrd_filename, 
> > +                           ram_addr_t below_4g_mem_size, 
> > +                           PcGuestInfo *guest_info); 
> >  FWCfgState *pc_memory_init(MemoryRegion *system_memory, 
> >                             const char *kernel_filename, 
> >                             const char *kernel_cmdline, 
> > --  
> > 1.8.4.5 
> >  
>  
> _______________________________________________ 
> Xen-devel mailing list 
> Xen-devel@xxxxxxxxxxxxx 
> http://lists.xen.org/xen-devel 
>  
>  


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel