Re: [PATCH v15 09/12] x86/smpboot: Support parallel startup of secondary CPUs

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Mar 16, 2023 at 6:21 PM Usama Arif <usama.arif@xxxxxxxxxxxxx> wrote:
>
> From: David Woodhouse <dwmw@xxxxxxxxxxxx>
>
> Rework the real-mode startup code to allow for APs to be brought up in
> parallel. This is in two parts:
>
> 1. Introduce a bit-spinlock to prevent them from all using the real
>    mode stack at the same time.
>
> 2. Avoid needing to use the global smpboot_control variable to pass
>    each AP its CPU#.
>
> To achieve the latter, export the cpuid_to_apicid[] array so that each
> AP can find its own CPU# by searching therein based on its APIC ID.
>
> Introduce flags in the top bits of smpboot_control which indicate methods
> by which an AP should find its CPU#. For a serialized bringup, the CPU#
> is explicitly passed in the low bits of smpboot_control as before. For
> parallel mode there are flags directing the AP to find its APIC ID in
> CPUID leaf 0x0b (for X2APIC mode) or CPUID leaf 0x01 where 8 bits are
> sufficient, then perform the cpuid_to_apicid[] lookup with that.
>
> Parallel startup may be disabled by a command line option, and also if:
>  • AMD SEV-ES is in use, since the AP may not use CPUID that early.
>  • X2APIC is enabled, but CPUID leaf 0xb is not present and correct.
>  • X2APIC is not enabled but not even CPUID leaf 0x01 exists.
>
> Aside from the fact that APs will now look up their CPU# via the
> newly-exported cpuid_to_apicid[] table, there is no behavioural change
> intended yet, since new parallel CPUHP states have not — yet — been
> added.
>
> [ tglx: Initial proof of concept patch with bitlock and APIC ID lookup ]
> [ dwmw2: Rework and testing, commit message, CPUID 0x1 and CPU0 support ]
> [ seanc: Fix stray override of initial_gs in common_cpu_up() ]
> [ Oleksandr Natalenko: reported suspend/resume issue fixed in
>   x86_acpi_suspend_lowlevel ]
> Co-developed-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> Co-developed-by: Brian Gerst <brgerst@xxxxxxxxx>
> Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> Signed-off-by: Brian Gerst <brgerst@xxxxxxxxx>
> Signed-off-by: David Woodhouse <dwmw@xxxxxxxxxxxx>
> Signed-off-by: Usama Arif <usama.arif@xxxxxxxxxxxxx>
> Tested-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
> Tested-by: Kim Phillips <kim.phillips@xxxxxxx>
> Tested-by: Oleksandr Natalenko <oleksandr@xxxxxxxxxxxxxx>
> Tested-by: Guilherme G. Piccoli <gpiccoli@xxxxxxxxxx>
> ---
>  .../admin-guide/kernel-parameters.txt         |  3 +
>  arch/x86/include/asm/cpu.h                    |  1 +
>  arch/x86/include/asm/realmode.h               |  3 +
>  arch/x86/include/asm/smp.h                    |  6 ++
>  arch/x86/kernel/acpi/sleep.c                  |  9 ++-
>  arch/x86/kernel/apic/apic.c                   |  2 +-
>  arch/x86/kernel/cpu/topology.c                |  3 +-
>  arch/x86/kernel/head_64.S                     | 65 +++++++++++++++++++
>  arch/x86/kernel/smpboot.c                     | 50 +++++++++++++-
>  arch/x86/realmode/init.c                      |  3 +
>  arch/x86/realmode/rm/trampoline_64.S          | 27 ++++++--
>  11 files changed, 163 insertions(+), 9 deletions(-)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index 6cfa6e3996cf..7bb7020f97e2 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -3819,6 +3819,9 @@
>
>         nomodule        Disable module load
>
> +       no_parallel_bringup
> +                       [X86,SMP] Disable parallel bring-up of secondary cores.
> +
>         nopat           [X86] Disable PAT (page attribute table extension of
>                         pagetables) support.
>
> diff --git a/arch/x86/include/asm/cpu.h b/arch/x86/include/asm/cpu.h
> index 78796b98a544..ef8ba318dca1 100644
> --- a/arch/x86/include/asm/cpu.h
> +++ b/arch/x86/include/asm/cpu.h
> @@ -97,5 +97,6 @@ static inline bool intel_cpu_signatures_match(unsigned int s1, unsigned int p1,
>  extern u64 x86_read_arch_cap_msr(void);
>  int intel_find_matching_signature(void *mc, unsigned int csig, int cpf);
>  int intel_microcode_sanity_check(void *mc, bool print_err, int hdr_type);
> +int check_extended_topology_leaf(int leaf);
>
>  #endif /* _ASM_X86_CPU_H */
> diff --git a/arch/x86/include/asm/realmode.h b/arch/x86/include/asm/realmode.h
> index f6a1737c77be..87e5482acd0d 100644
> --- a/arch/x86/include/asm/realmode.h
> +++ b/arch/x86/include/asm/realmode.h
> @@ -52,6 +52,7 @@ struct trampoline_header {
>         u64 efer;
>         u32 cr4;
>         u32 flags;
> +       u32 lock;
>  #endif
>  };
>
> @@ -64,6 +65,8 @@ extern unsigned long initial_stack;
>  extern unsigned long initial_vc_handler;
>  #endif
>
> +extern u32 *trampoline_lock;
> +
>  extern unsigned char real_mode_blob[];
>  extern unsigned char real_mode_relocs[];
>
> diff --git a/arch/x86/include/asm/smp.h b/arch/x86/include/asm/smp.h
> index bf2c51df9e0b..1cf4f1e57570 100644
> --- a/arch/x86/include/asm/smp.h
> +++ b/arch/x86/include/asm/smp.h
> @@ -203,4 +203,10 @@ extern unsigned int smpboot_control;
>
>  #endif /* !__ASSEMBLY__ */
>
> +/* Control bits for startup_64 */
> +#define STARTUP_APICID_CPUID_0B        0x80000000
> +#define STARTUP_APICID_CPUID_01        0x40000000
> +
> +#define STARTUP_PARALLEL_MASK (STARTUP_APICID_CPUID_01 | STARTUP_APICID_CPUID_0B)
> +
>  #endif /* _ASM_X86_SMP_H */
> diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
> index 1328c221af30..6dfecb27b846 100644
> --- a/arch/x86/kernel/acpi/sleep.c
> +++ b/arch/x86/kernel/acpi/sleep.c
> @@ -16,6 +16,7 @@
>  #include <asm/cacheflush.h>
>  #include <asm/realmode.h>
>  #include <asm/hypervisor.h>
> +#include <asm/smp.h>
>
>  #include <linux/ftrace.h>
>  #include "../../realmode/rm/wakeup.h"
> @@ -127,7 +128,13 @@ int x86_acpi_suspend_lowlevel(void)
>          * value is in the actual %rsp register.
>          */
>         current->thread.sp = (unsigned long)temp_stack + sizeof(temp_stack);
> -       smpboot_control = smp_processor_id();
> +       /*
> +        * Ensure the CPU knows which one it is when it comes back, if
> +        * it isn't in parallel mode and expected to work that out for
> +        * itself.
> +        */
> +       if (!(smpboot_control & STARTUP_PARALLEL_MASK))
> +               smpboot_control = smp_processor_id();
>  #endif
>         initial_code = (unsigned long)wakeup_long64;
>         saved_magic = 0x123456789abcdef0L;
> diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
> index 20d9a604da7c..ac1d7e5da1f2 100644
> --- a/arch/x86/kernel/apic/apic.c
> +++ b/arch/x86/kernel/apic/apic.c
> @@ -2377,7 +2377,7 @@ static int nr_logical_cpuids = 1;
>  /*
>   * Used to store mapping between logical CPU IDs and APIC IDs.
>   */
> -static int cpuid_to_apicid[] = {
> +int cpuid_to_apicid[] = {
>         [0 ... NR_CPUS - 1] = -1,
>  };
>
> diff --git a/arch/x86/kernel/cpu/topology.c b/arch/x86/kernel/cpu/topology.c
> index 5e868b62a7c4..4373442e500a 100644
> --- a/arch/x86/kernel/cpu/topology.c
> +++ b/arch/x86/kernel/cpu/topology.c
> @@ -9,6 +9,7 @@
>  #include <asm/apic.h>
>  #include <asm/memtype.h>
>  #include <asm/processor.h>
> +#include <asm/cpu.h>
>
>  #include "cpu.h"
>
> @@ -32,7 +33,7 @@ EXPORT_SYMBOL(__max_die_per_package);
>  /*
>   * Check if given CPUID extended topology "leaf" is implemented
>   */
> -static int check_extended_topology_leaf(int leaf)
> +int check_extended_topology_leaf(int leaf)
>  {
>         unsigned int eax, ebx, ecx, edx;
>
> diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
> index 6a8238702eab..65bca47d84a1 100644
> --- a/arch/x86/kernel/head_64.S
> +++ b/arch/x86/kernel/head_64.S
> @@ -25,6 +25,7 @@
>  #include <asm/export.h>
>  #include <asm/nospec-branch.h>
>  #include <asm/fixmap.h>
> +#include <asm/smp.h>
>
>  /*
>   * We are not able to switch in one step to the final KERNEL ADDRESS SPACE
> @@ -234,8 +235,61 @@ SYM_INNER_LABEL(secondary_startup_64_no_verify, SYM_L_GLOBAL)
>         ANNOTATE_NOENDBR // above
>
>  #ifdef CONFIG_SMP
> +       /*
> +        * For parallel boot, the APIC ID is retrieved from CPUID, and then
> +        * used to look up the CPU number.  For booting a single CPU, the
> +        * CPU number is encoded in smpboot_control.
> +        *
> +        * Bit 31       STARTUP_APICID_CPUID_0B flag (use CPUID 0x0b)
> +        * Bit 30       STARTUP_APICID_CPUID_01 flag (use CPUID 0x01)
> +        * Bit 0-24     CPU# if STARTUP_APICID_CPUID_xx flags are not set
> +        */
>         movl    smpboot_control(%rip), %ecx
> +       testl   $STARTUP_APICID_CPUID_0B, %ecx
> +       jnz     .Luse_cpuid_0b
> +       testl   $STARTUP_APICID_CPUID_01, %ecx
> +       jnz     .Luse_cpuid_01
> +       andl    $0x0FFFFFFF, %ecx
> +       jmp     .Lsetup_cpu
> +
> +.Luse_cpuid_01:
> +       mov     $0x01, %eax
> +       cpuid
> +       mov     %ebx, %edx
> +       shr     $24, %edx
> +       jmp     .Lsetup_AP
>
> +.Luse_cpuid_0b:
> +       mov     $0x0B, %eax
> +       xorl    %ecx, %ecx
> +       cpuid
> +
> +.Lsetup_AP:
> +       /* EDX contains the APIC ID of the current CPU */
> +       xorq    %rcx, %rcx
> +       leaq    cpuid_to_apicid(%rip), %rbx
> +
> +.Lfind_cpunr:
> +       cmpl    (%rbx,%rcx,4), %edx
> +       jz      .Lsetup_cpu
> +       inc     %ecx
> +#ifdef CONFIG_FORCE_NR_CPUS
> +       cmpl    $NR_CPUS, %ecx
> +#else
> +       cmpl    nr_cpu_ids(%rip), %ecx
> +#endif
> +       jb      .Lfind_cpunr
> +
> +       /*  APIC ID not found in the table. Drop the trampoline lock and bail. */
> +       movq    trampoline_lock(%rip), %rax
> +       lock
> +       btrl    $0, (%rax)
> +
> +1:     cli
> +       hlt
> +       jmp     1b
> +
> +.Lsetup_cpu:
>         /* Get the per cpu offset for the given CPU# which is in ECX */
>         movq    __per_cpu_offset(,%rcx,8), %rdx
>  #else
> @@ -264,6 +318,14 @@ SYM_INNER_LABEL(secondary_startup_64_no_verify, SYM_L_GLOBAL)
>         lgdt    (%rsp)
>         addq    $16, %rsp
>
> +       /* Drop the realmode protection. For the boot CPU the pointer is NULL! */
> +       movq    trampoline_lock(%rip), %rax
> +       testq   %rax, %rax
> +       jz      .Lsetup_data_segments
> +       lock
> +       btrl    $0, (%rax)
> +
> +.Lsetup_data_segments:
>         /* set up data segments */
>         xorl %eax,%eax
>         movl %eax,%ds

This can still go earlier, right after "movq TASK_threadsp(%rax),
%rsp".  The GDT descriptor is placed on the idle thread stack, so it's
safe to drop the lock before it.

--
Brian Gerst




[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux