[PATCH 4.17 50/67] xen: setup pv irq ops vector earlier

From: Greg Kroah-Hartman
Date: Mon Jul 16 2018 - 03:39:18 EST


4.17-stable review patch. If anyone has any objections, please let me know.

------------------

From: Juergen Gross <jgross@xxxxxxxx>

commit 0ce0bba4e5e0eb9b753bb821785de5d23c494392 upstream.

Setting pv_irq_ops for Xen PV domains should be done as early as
possible in order to support e.g. very early printk() usage.

The same applies to xen_vcpu_info_reset(0), as it is needed for the
pv irq ops.

Move the call of xen_setup_machphys_mapping() after initializing the
pv functions as it contains a WARN_ON(), too.

Remove the no longer necessary conditional in xen_init_irq_ops()
from PVH V1 times to make clear this is a PV only function.

Cc: <stable@xxxxxxxxxxxxxxx> # 4.14
Signed-off-by: Juergen Gross <jgross@xxxxxxxx>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx>
Signed-off-by: Juergen Gross <jgross@xxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
arch/x86/xen/enlighten_pv.c | 24 +++++++++++-------------
arch/x86/xen/irq.c | 4 +---
2 files changed, 12 insertions(+), 16 deletions(-)

--- a/arch/x86/xen/enlighten_pv.c
+++ b/arch/x86/xen/enlighten_pv.c
@@ -1206,12 +1206,20 @@ asmlinkage __visible void __init xen_sta

xen_setup_features();

- xen_setup_machphys_mapping();
-
/* Install Xen paravirt ops */
pv_info = xen_info;
pv_init_ops.patch = paravirt_patch_default;
pv_cpu_ops = xen_cpu_ops;
+ xen_init_irq_ops();
+
+ /*
+ * Setup xen_vcpu early because it is needed for
+ * local_irq_disable(), irqs_disabled(), e.g. in printk().
+ *
+ * Don't do the full vcpu_info placement stuff until we have
+ * the cpu_possible_mask and a non-dummy shared_info.
+ */
+ xen_vcpu_info_reset(0);

x86_platform.get_nmi_reason = xen_get_nmi_reason;

@@ -1224,6 +1232,7 @@ asmlinkage __visible void __init xen_sta
* Set up some pagetable state before starting to set any ptes.
*/

+ xen_setup_machphys_mapping();
xen_init_mmu_ops();

/* Prevent unwanted bits from being set in PTEs. */
@@ -1249,20 +1258,9 @@ asmlinkage __visible void __init xen_sta
get_cpu_cap(&boot_cpu_data);
x86_configure_nx();

- xen_init_irq_ops();
-
/* Let's presume PV guests always boot on vCPU with id 0. */
per_cpu(xen_vcpu_id, 0) = 0;

- /*
- * Setup xen_vcpu early because idt_setup_early_handler needs it for
- * local_irq_disable(), irqs_disabled().
- *
- * Don't do the full vcpu_info placement stuff until we have
- * the cpu_possible_mask and a non-dummy shared_info.
- */
- xen_vcpu_info_reset(0);
-
idt_setup_early_handler();

xen_init_capabilities();
--- a/arch/x86/xen/irq.c
+++ b/arch/x86/xen/irq.c
@@ -128,8 +128,6 @@ static const struct pv_irq_ops xen_irq_o

void __init xen_init_irq_ops(void)
{
- /* For PVH we use default pv_irq_ops settings. */
- if (!xen_feature(XENFEAT_hvm_callback_vector))
- pv_irq_ops = xen_irq_ops;
+ pv_irq_ops = xen_irq_ops;
x86_init.irqs.intr_init = xen_init_IRQ;
}