[PATCH v2 00/15] Introduce generic headers

[PATCH v2 00/15] Introduce generic headers

[Oleksii Kurochko]

Some headers are common between several architectures, so the current patch series
provide them.

Another one reason to have them as generic is a simplification of adding support
necessary to make a complete Xen build as it was/is being done in the patch series [1]
and [2].

Also, instead of providing generic/stub headers, it was used
"#ifdef CONFIG_* #include <asm/*.h> #endif" instead of providing empty headers.

Some headers are still arguable if they should be in asm-generic, and it wasn't a response so
I decided to leave headers for now and made some updates for them.

Some patches related to delay.h, pci.h and xenoprof.h were sent as separate patches: [3], [4]
and [5].

[1] https://lore.kernel.org/xen-devel/cover.1694543103.git.sanastasio@raptorengineering.com/
[2] https://lore.kernel.org/xen-devel/cover.1692181079.git.oleksii.kurochko@gmail.com/
[3] https://lore.kernel.org/xen-devel/3d55bce44bd6ab9973cbe0ea2fc136cc44d35df2.1698759633.git.oleksii.kurochko@gmail.com/
[4] https://lore.kernel.org/xen-devel/314745757996935d8b2ae9919410c1abc0c86ce9.camel@gmail.com/T/#t
[5] https://gitlab.com/xen-project/xen/-/commit/ad193a79c25ae24f1de5f37900e1da74ab48e8bd

---
Changes in V2:
 - Update the commit message of the cover letter.
 - Drop the following patches because they are arch-specific or was sent as a separate patch:
   - xen/asm-generic: introduce stub header event.h
	 - xen/asm-generic: introduce stub header spinlock.h
	 - [PATCH v1 03/29] xen/asm-generic: introduce stub header cpufeature.h
	 - [PATCH v1 07/29] xen/asm-generic: introduce stub header guest_atomics.h
	 - [PATCH v1 10/29] xen/asm-generic: introduce stub header iommu.h
	 - [PATCH v1 12/29] xen/asm-generic: introduce stub header pci.h because separate patch was sent [5]
	 - [PATCH v1 14/29] xen/asm-generic: introduce stub header setup.h
	 - [PATCH v1 15/29] xen/asm-generic: introduce stub header xenoprof.h because of [3].
	 - [PATCH v1 16/29] xen/asm-generic: introduce stub header flushtlb.h
	 - [PATCH v1 22/29] xen/asm-generic: introduce stub header delay.h because of [3]
	 - [PATCH v1 23/29] xen/asm-generic: introduce stub header domain.h
	 - [PATCH v1 24/29] xen/asm-generic: introduce stub header guest_access.h
	 - [PATCH v1 25/29] xen/asm-generic: introduce stub header irq.h ( probably not so generic as I expected, I'll back to it if it will be necessary in the future )
	 - [PATCH v1 28/29] xen/asm-generic: introduce stub header p2m.h ( probably not so generic as I expected, I'll back to it if it will be necessary in the future )
 - For the rest of the patches please look at changes for each patch separately.

Oleksii Kurochko (15):
  xen/asm-generic: introduce stub header paging.h
  xen/asm-generic: introduce generic device.h
  xen: ifdef inclusion of <asm/grant_table.h> in <xen/grant_table.h>
  xen/asm-generic: introduce generic hypercall.h
  xen/asm-generic: introduce generic header iocap.h
  xen/asm-generic: ifdef inclusion of <asm/mem_access.h>
  xen/asm-generic: introduce stub header <asm/random.h>
  xen/asm-generic: introduce generic header percpu.h
  xen/asm-generic: introduce generic header smp.h
  xen/asm-generic: introduce generalized hardirq.h
  xen/asm-generic: introduce generic div64.h header
  xen/asm-generic: introduce generic header altp2m.h
  xen/asm-generic: introduce stub header monitor.h
  xen/asm-generic: introduce stub header numa.h
  xen/asm-generic: introduce stub header softirq.h

 xen/arch/arm/include/asm/Makefile   |   1 +
 xen/arch/arm/include/asm/paging.h   |  16 ----
 xen/arch/ppc/include/asm/Makefile   |   1 +
 xen/arch/ppc/include/asm/paging.h   |   7 --
 xen/include/asm-generic/altp2m.h    |  34 +++++++
 xen/include/asm-generic/device.h    | 140 ++++++++++++++++++++++++++++
 xen/include/asm-generic/div64.h     |  27 ++++++
 xen/include/asm-generic/hardirq.h   |  29 ++++++
 xen/include/asm-generic/hypercall.h |  18 ++++
 xen/include/asm-generic/iocap.h     |  17 ++++
 xen/include/asm-generic/monitor.h   |  62 ++++++++++++
 xen/include/asm-generic/numa.h      |  40 ++++++++
 xen/include/asm-generic/paging.h    |  19 ++++
 xen/include/asm-generic/percpu.h    |  35 +++++++
 xen/include/asm-generic/random.h    |  20 ++++
 xen/include/asm-generic/smp.h       |  28 ++++++
 xen/include/asm-generic/softirq.h   |  17 ++++
 xen/include/xen/grant_table.h       |   3 +
 xen/include/xen/mem_access.h        |   2 +
 19 files changed, 493 insertions(+), 23 deletions(-)
 delete mode 100644 xen/arch/arm/include/asm/paging.h
 delete mode 100644 xen/arch/ppc/include/asm/paging.h
 create mode 100644 xen/include/asm-generic/altp2m.h
 create mode 100644 xen/include/asm-generic/device.h
 create mode 100644 xen/include/asm-generic/div64.h
 create mode 100644 xen/include/asm-generic/hardirq.h
 create mode 100644 xen/include/asm-generic/hypercall.h
 create mode 100644 xen/include/asm-generic/iocap.h
 create mode 100644 xen/include/asm-generic/monitor.h
 create mode 100644 xen/include/asm-generic/numa.h
 create mode 100644 xen/include/asm-generic/paging.h
 create mode 100644 xen/include/asm-generic/percpu.h
 create mode 100644 xen/include/asm-generic/random.h
 create mode 100644 xen/include/asm-generic/smp.h
 create mode 100644 xen/include/asm-generic/softirq.h

-- 
2.41.0

[PATCH v2 01/15] xen/asm-generic: introduce stub header paging.h

[Oleksii Kurochko]

The patch introduces generic paging.h header for Arm, PPC and
RISC-V.

All mentioned above architectures use hardware virt extensions
and hardware pagetable extensions thereby it makes sense to set
paging_mode_translate and paging_mode_external by default.

Also in this patch Arm and PPC architectures are switched to
generic paging.h header.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
	- evaluate argument of macros
	- covert macros to true
	- use proper tabs
	- switch Arm & PPC to generic paging.h
	- update commit message
---
 xen/arch/arm/include/asm/Makefile |  1 +
 xen/arch/arm/include/asm/paging.h | 16 ----------------
 xen/arch/ppc/include/asm/Makefile |  1 +
 xen/arch/ppc/include/asm/paging.h |  7 -------
 xen/include/asm-generic/paging.h  | 19 +++++++++++++++++++
 5 files changed, 21 insertions(+), 23 deletions(-)
 delete mode 100644 xen/arch/arm/include/asm/paging.h
 delete mode 100644 xen/arch/ppc/include/asm/paging.h
 create mode 100644 xen/include/asm-generic/paging.h

diff --git a/xen/arch/arm/include/asm/Makefile b/xen/arch/arm/include/asm/Makefile
index 821addb0bf..2128bb015f 100644
--- a/xen/arch/arm/include/asm/Makefile
+++ b/xen/arch/arm/include/asm/Makefile
@@ -1,2 +1,3 @@
 # SPDX-License-Identifier: GPL-2.0-only
 generic-y += vm_event.h
+generic-y += paging.h
diff --git a/xen/arch/arm/include/asm/paging.h b/xen/arch/arm/include/asm/paging.h
deleted file mode 100644
index 6d1a000246..0000000000
--- a/xen/arch/arm/include/asm/paging.h
+++ /dev/null
@@ -1,16 +0,0 @@
-#ifndef _XEN_PAGING_H
-#define _XEN_PAGING_H
-
-#define paging_mode_translate(d)              (1)
-#define paging_mode_external(d)               (1)
-
-#endif /* XEN_PAGING_H */
-
-/*
- * Local variables:
- * mode: C
- * c-file-style: "BSD"
- * c-basic-offset: 4
- * indent-tabs-mode: nil
- * End:
- */
diff --git a/xen/arch/ppc/include/asm/Makefile b/xen/arch/ppc/include/asm/Makefile
index 821addb0bf..2128bb015f 100644
--- a/xen/arch/ppc/include/asm/Makefile
+++ b/xen/arch/ppc/include/asm/Makefile
@@ -1,2 +1,3 @@
 # SPDX-License-Identifier: GPL-2.0-only
 generic-y += vm_event.h
+generic-y += paging.h
diff --git a/xen/arch/ppc/include/asm/paging.h b/xen/arch/ppc/include/asm/paging.h
deleted file mode 100644
index eccacece29..0000000000
--- a/xen/arch/ppc/include/asm/paging.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef __ASM_PPC_PAGING_H__
-#define __ASM_PPC_PAGING_H__
-
-#define paging_mode_translate(d)              (1)
-#define paging_mode_external(d)               (1)
-
-#endif /* __ASM_PPC_PAGING_H__ */
diff --git a/xen/include/asm-generic/paging.h b/xen/include/asm-generic/paging.h
new file mode 100644
index 0000000000..8df534cfdc
--- /dev/null
+++ b/xen/include/asm-generic/paging.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ASM_GENERIC_PAGING_H__
+#define __ASM_GENERIC_PAGING_H__
+
+#include <xen/stdbool.h>
+
+#define paging_mode_translate(d)    ((void)(d), true)
+#define paging_mode_external(d)     ((void)(d), true)
+
+#endif /* __ASM_GENERIC_PAGING_H__ */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 02/15] xen/asm-generic: introduce generic device.h

[Oleksii Kurochko]

Arm, PPC and RISC-V use the same device.h thereby device.h
was moved to asm-generic. Arm's device.h was taken as a base with
the following changes:
 - #ifdef PCI related things.
 - #ifdef ACPI related things.
 - Rename #ifdef guards.
 - Add SPDX tag.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
It is still open question if device.h should be in asm-generic. Need more opinions.

Changes in V2:
	- take ( as common ) device.h from Arm as PPC and RISC-V use it as a base.
	- #ifdef PCI related things.
	- #ifdef ACPI related things.
	- rename DEVICE_GIC to DEVIC_IC.
	- rename #ifdef guards.
	- switch Arm and PPC to generic device.h
	- add SPDX tag
	- update the commit message

---
 xen/include/asm-generic/device.h | 140 +++++++++++++++++++++++++++++++
 1 file changed, 140 insertions(+)
 create mode 100644 xen/include/asm-generic/device.h

diff --git a/xen/include/asm-generic/device.h b/xen/include/asm-generic/device.h
new file mode 100644
index 0000000000..5478fc11ca
--- /dev/null
+++ b/xen/include/asm-generic/device.h
@@ -0,0 +1,140 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ASM_GENERIC_DEVICE_H__
+#define __ASM_GENERIC_DEVICE_H__
+
+enum device_type
+{
+    DEV_DT,
+#ifdef HAS_PCI
+    DEV_PCI,
+#endif
+};
+
+struct dev_archdata {
+    void *iommu;    /* IOMMU private data */
+};
+
+/* struct device - The basic device structure */
+struct device
+{
+    enum device_type type;
+#ifdef CONFIG_HAS_DEVICE_TREE
+    struct dt_device_node *of_node; /* Used by drivers imported from Linux */
+#endif
+    struct dev_archdata archdata;
+    struct iommu_fwspec *iommu_fwspec; /* per-device IOMMU instance data */
+};
+
+typedef struct device device_t;
+
+#include <xen/device_tree.h>
+
+#ifdef HAS_PCI
+#define dev_is_pci(dev) ((dev)->type == DEV_PCI)
+#endif
+
+#define dev_is_dt(dev)  ((dev)->type == DEV_DT)
+
+enum device_class
+{
+    DEVICE_SERIAL,
+    DEVICE_IOMMU,
+    DEVICE_IC,
+#ifdef HAS_PCI
+    DEVICE_PCI_HOSTBRIDGE,
+#endif
+    /* Use for error */
+    DEVICE_UNKNOWN,
+};
+
+struct device_desc {
+    /* Device name */
+    const char *name;
+    /* Device class */
+    enum device_class class;
+    /* List of devices supported by this driver */
+    const struct dt_device_match *dt_match;
+    /*
+     * Device initialization.
+     *
+     * -EAGAIN is used to indicate that device probing is deferred.
+     */
+    int (*init)(struct dt_device_node *dev, const void *data);
+};
+
+#ifdef CONFIG_ACPI
+
+struct acpi_device_desc {
+    /* Device name */
+    const char *name;
+    /* Device class */
+    enum device_class class;
+    /* type of device supported by the driver */
+    const int class_type;
+    /* Device initialization */
+    int (*init)(const void *data);
+};
+
+/**
+ *  acpi_device_init - Initialize a device
+ *  @class: class of the device (serial, network...)
+ *  @data: specific data for initializing the device
+ *
+ *  Return 0 on success.
+ */
+int acpi_device_init(enum device_class class,
+                     const void *data, int class_type);
+
+#endif /* CONFIG_ACPI */
+
+/**
+ *  device_init - Initialize a device
+ *  @dev: device to initialize
+ *  @class: class of the device (serial, network...)
+ *  @data: specific data for initializing the device
+ *
+ *  Return 0 on success.
+ */
+int device_init(struct dt_device_node *dev, enum device_class class,
+                const void *data);
+
+/**
+ * device_get_type - Get the type of the device
+ * @dev: device to match
+ *
+ * Return the device type on success or DEVICE_ANY on failure
+ */
+enum device_class device_get_class(const struct dt_device_node *dev);
+
+#define DT_DEVICE_START(_name, _namestr, _class)                    \
+static const struct device_desc __dev_desc_##_name __used           \
+__section(".dev.info") = {                                          \
+    .name = _namestr,                                               \
+    .class = _class,                                                \
+
+#define DT_DEVICE_END                                               \
+};
+
+#ifdef CONFIG_ACPI
+
+#define ACPI_DEVICE_START(_name, _namestr, _class)                    \
+static const struct acpi_device_desc __dev_desc_##_name __used           \
+__section(".adev.info") = {                       \
+    .name = _namestr,                                               \
+    .class = _class,                                                \
+
+#define ACPI_DEVICE_END                                               \
+};
+
+#endif /* CONFIG_ACPI */
+
+#endif /* __ASM_GENERIC_DEVICE_H__ */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 03/15] xen: ifdef inclusion of <asm/grant_table.h> in <xen/grant_table.h>

[Oleksii Kurochko]

Ifdefing inclusion of <asm/grant_table.h> allows to avoid
generation of empty <asm/grant_table.h> for cases when
CONFIG_GRANT_TABLE is not enabled.

Suggested-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
	- ifdef inclusion of asm/grant_table.h in xen/grant_table.h to avoid
	  generation of empty headers for PPC and RISC-V archs.
	- update commit message
	- add Suggested-by: Jan Beulich <jbeulich@suse.com>
	- Remove provided before asm-generic/grant_table.h header.
---
 xen/include/xen/grant_table.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/xen/include/xen/grant_table.h b/xen/include/xen/grant_table.h
index 85fe6b7b5e..50edfecfb6 100644
--- a/xen/include/xen/grant_table.h
+++ b/xen/include/xen/grant_table.h
@@ -26,7 +26,10 @@
 #include <xen/mm-frame.h>
 #include <xen/rwlock.h>
 #include <public/grant_table.h>
+
+#ifdef CONFIG_GRANT_TABLE
 #include <asm/grant_table.h>
+#endif
 
 struct grant_table;
 
-- 
2.41.0

[PATCH v2 04/15] xen/asm-generic: introduce generic hypercall.h

[Oleksii Kurochko]

Introduce an empty generic hypercall.h for archs which don't
implement it.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
 - add check that <asm/hypercall.h> isn't included directly.
---
 xen/include/asm-generic/hypercall.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)
 create mode 100644 xen/include/asm-generic/hypercall.h

diff --git a/xen/include/asm-generic/hypercall.h b/xen/include/asm-generic/hypercall.h
new file mode 100644
index 0000000000..7743b35c0d
--- /dev/null
+++ b/xen/include/asm-generic/hypercall.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __XEN_HYPERCALL_H__
+#error "asm/hypercall.h should not be included directly - include xen/hypercall.h instead"
+#endif
+
+#ifndef __ASM_GENERIC_HYPERCALL_H__
+#define __ASM_GENERIC_HYPERCALL_H__
+
+#endif /* __ASM_GENERIC_HYPERCALL_H__ */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 05/15] xen/asm-generic: introduce generic header iocap.h

[Oleksii Kurochko]

iocap.h is common for Arm, PPC and RISC-V architectures thereby
it was moved to asm-generic.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
The same question as with device.h. Should it be in asm-generic?

Changes in V2:
 - update the commit message
---
 xen/include/asm-generic/iocap.h | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)
 create mode 100644 xen/include/asm-generic/iocap.h

diff --git a/xen/include/asm-generic/iocap.h b/xen/include/asm-generic/iocap.h
new file mode 100644
index 0000000000..dd7cb45488
--- /dev/null
+++ b/xen/include/asm-generic/iocap.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ASM_GENERIC_IOCAP_H__
+#define __ASM_GENERIC_IOCAP_H__
+
+#define cache_flush_permitted(d)                        \
+    (!rangeset_is_empty((d)->iomem_caps))
+
+#endif /* __ASM_GENERIC_IOCAP_H__ */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 06/15] xen/asm-generic: ifdef inclusion of <asm/mem_access.h>

[Oleksii Kurochko]

ifdefing inclusion of <asm/mem_access.h> in <xen/mem_access.h>
allows to avoid generation of empty <asm/mem_access.h> header
for the case when !CONFIG_MEM_ACCESS.

Suggested-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
	- add Suggested-by: Jan Beulich <jbeulich@suse.com>
	- update the commit message
	- remove <asm-generic/mem_access.h>
---
 xen/include/xen/mem_access.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/xen/include/xen/mem_access.h b/xen/include/xen/mem_access.h
index 4e4811680d..87d93b31f6 100644
--- a/xen/include/xen/mem_access.h
+++ b/xen/include/xen/mem_access.h
@@ -33,7 +33,9 @@
  */
 struct vm_event_st;
 
+#ifdef CONFIG_MEM_ACCESS
 #include <asm/mem_access.h>
+#endif
 
 /*
  * Additional access types, which are used to further restrict
-- 
2.41.0

[PATCH v2 07/15] xen/asm-generic: introduce stub header <asm/random.h>

[Oleksii Kurochko]

<asm/random.h> is common for Arm, PPC and RISC-V thereby it
is moved to asm-generic.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
 - update the commit messages
---
 xen/include/asm-generic/random.h | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)
 create mode 100644 xen/include/asm-generic/random.h

diff --git a/xen/include/asm-generic/random.h b/xen/include/asm-generic/random.h
new file mode 100644
index 0000000000..cd2307e70b
--- /dev/null
+++ b/xen/include/asm-generic/random.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ASM_GENERIC_RANDOM_H__
+#define __ASM_GENERIC_RANDOM_H__
+
+static inline unsigned int arch_get_random(void)
+{
+    return 0;
+}
+
+#endif /* __ASM_GENERIC_RANDOM_H__ */
+
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: BSD
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 08/15] xen/asm-generic: introduce generic header percpu.h

[Oleksii Kurochko]

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
	- use smp_processor_id() instead of get_processor_id().
 	- update commit message .
---
 xen/include/asm-generic/percpu.h | 35 ++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)
 create mode 100644 xen/include/asm-generic/percpu.h

diff --git a/xen/include/asm-generic/percpu.h b/xen/include/asm-generic/percpu.h
new file mode 100644
index 0000000000..85a3f3ef17
--- /dev/null
+++ b/xen/include/asm-generic/percpu.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ASM_GENERIC_PERCPU_H__
+#define __ASM_GENERIC_PERCPU_H__
+
+#ifndef __ASSEMBLY__
+
+#include <xen/types.h>
+
+extern char __per_cpu_start[], __per_cpu_data_end[];
+extern unsigned long __per_cpu_offset[NR_CPUS];
+void percpu_init_areas(void);
+
+#define per_cpu(var, cpu)  \
+    (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset[cpu]))
+
+#define this_cpu(var) \
+    (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset[smp_processor_id()]))
+
+#define per_cpu_ptr(var, cpu)  \
+    (*RELOC_HIDE(var, __per_cpu_offset[cpu]))
+#define this_cpu_ptr(var) \
+    (*RELOC_HIDE(var, smp_processor_id()))
+
+#endif
+
+#endif /* __ASM_GENERIC_PERCPU_H__ */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 09/15] xen/asm-generic: introduce generic header smp.h

[Oleksii Kurochko]

<asm/smp.h> is expcted to be generic between Arm, PPC and RISC-V
there by it is moved to asm-generic.

Right now it is common only by PPC and RISC-V but during work on
support of the mentioned arhcs <asm/smp.h> is expected to be the
same.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
	- drop #ifded ASSEMBLY as this header isn't expected to be included in asm files.
	- update the commit message.
---
 xen/include/asm-generic/smp.h | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)
 create mode 100644 xen/include/asm-generic/smp.h

diff --git a/xen/include/asm-generic/smp.h b/xen/include/asm-generic/smp.h
new file mode 100644
index 0000000000..6740a2064c
--- /dev/null
+++ b/xen/include/asm-generic/smp.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ASM_GENERIC_SMP_H
+#define __ASM_GENERIC_SMP_H
+
+#include <xen/cpumask.h>
+#include <xen/percpu.h>
+
+DECLARE_PER_CPU(cpumask_var_t, cpu_sibling_mask);
+DECLARE_PER_CPU(cpumask_var_t, cpu_core_mask);
+
+#define cpu_is_offline(cpu) unlikely(!cpu_online(cpu))
+
+/*
+ * Do we, for platform reasons, need to actually keep CPUs online when we
+ * would otherwise prefer them to be off?
+ */
+#define park_offline_cpus false
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: BSD
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 10/15] xen/asm-generic: introduce generalized hardirq.h

[Oleksii Kurochko]

<asm/hardirq.h> is common through archs thereby it is moved
to asm-generic.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
	- add #include <xen/cache.h>.
	- update the commit message
---
 xen/include/asm-generic/hardirq.h | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)
 create mode 100644 xen/include/asm-generic/hardirq.h

diff --git a/xen/include/asm-generic/hardirq.h b/xen/include/asm-generic/hardirq.h
new file mode 100644
index 0000000000..ddccf460b9
--- /dev/null
+++ b/xen/include/asm-generic/hardirq.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ASM_GENERIC_HARDIRQ_H
+#define __ASM_GENERIC_HARDIRQ_H
+
+#include <xen/cache.h>
+#include <xen/smp.h>
+
+typedef struct {
+        unsigned long __softirq_pending;
+        unsigned int __local_irq_count;
+} __cacheline_aligned irq_cpustat_t;
+
+#include <xen/irq_cpustat.h>    /* Standard mappings for irq_cpustat_t above */
+
+#define in_irq() (local_irq_count(smp_processor_id()) != 0)
+
+#define irq_enter()     (local_irq_count(smp_processor_id())++)
+#define irq_exit()      (local_irq_count(smp_processor_id())--)
+
+#endif /* __ASM_GENERIC_HARDIRQ_H */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 11/15] xen/asm-generic: introduce generic div64.h header

[Oleksii Kurochko]

All archs have the do_div implementation for BITS_PER_LONG == 64
so do_div64.h is moved to asm-generic.

The patch introduces header stub necessry for full Xen build.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
	- rename base to divisor
	- add "#if BITS_PER_LONG == 64"
	- fix code style
---
 xen/include/asm-generic/div64.h | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)
 create mode 100644 xen/include/asm-generic/div64.h

diff --git a/xen/include/asm-generic/div64.h b/xen/include/asm-generic/div64.h
new file mode 100644
index 0000000000..068d8a11ad
--- /dev/null
+++ b/xen/include/asm-generic/div64.h
@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ASM_GENERIC_DIV64
+#define __ASM_GENERIC_DIV64
+
+#include <xen/types.h>
+
+#if BITS_PER_LONG == 64
+
+#define do_div(n, divisor) ({                   \
+    uint32_t divisor_ = (divisor);              \
+    uint32_t rem_ = (uint64_t)(n) % divisor_;   \
+    (n) = (uint64_t)(n) / divisor_;             \
+    rem_;                                       \
+})
+
+#endif /* BITS_PER_LONG */
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 12/15] xen/asm-generic: introduce generic header altp2m.h

[Oleksii Kurochko]

<asm/alt2pm.h> is common between archs so it is moved to
asm-generic.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
	- change uint16_t to unsigned int in declaration of altp2m_vcpu_idx
	- update the commit message
---
 xen/include/asm-generic/altp2m.h | 34 ++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)
 create mode 100644 xen/include/asm-generic/altp2m.h

diff --git a/xen/include/asm-generic/altp2m.h b/xen/include/asm-generic/altp2m.h
new file mode 100644
index 0000000000..39865a842a
--- /dev/null
+++ b/xen/include/asm-generic/altp2m.h
@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ASM_GENERIC_ALTP2M_H
+#define __ASM_GENERIC_ALTP2M_H
+
+#include <xen/bug.h>
+
+struct domain;
+struct vcpu;
+
+/* Alternate p2m on/off per domain */
+static inline bool altp2m_active(const struct domain *d)
+{
+    /* Not implemented on GENERIC. */
+    return false;
+}
+
+/* Alternate p2m VCPU */
+static inline unsigned int altp2m_vcpu_idx(const struct vcpu *v)
+{
+    /* Not implemented on GENERIC, should not be reached. */
+    BUG();
+    return 0;
+}
+
+#endif /* __ASM_GENERIC_ALTP2M_H */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: BSD
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 13/15] xen/asm-generic: introduce stub header monitor.h

[Oleksii Kurochko]

The header is shared between archs so it is moved to asm-generic.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
	- remove inclusion of "+#include <public/domctl.h>"
	- add "struct xen_domctl_monitor_op;"
	- remove one of SPDX tags.
---
 xen/include/asm-generic/monitor.h | 62 +++++++++++++++++++++++++++++++
 1 file changed, 62 insertions(+)
 create mode 100644 xen/include/asm-generic/monitor.h

diff --git a/xen/include/asm-generic/monitor.h b/xen/include/asm-generic/monitor.h
new file mode 100644
index 0000000000..4e46f2e91a
--- /dev/null
+++ b/xen/include/asm-generic/monitor.h
@@ -0,0 +1,62 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * include/asm-GENERIC/monitor.h
+ *
+ * Arch-specific monitor_op domctl handler.
+ *
+ * Copyright (c) 2015 Tamas K Lengyel (tamas@tklengyel.com)
+ * Copyright (c) 2016, Bitdefender S.R.L.
+ *
+ */
+
+#ifndef __ASM_GENERIC_MONITOR_H__
+#define __ASM_GENERIC_MONITOR_H__
+
+#include <xen/sched.h>
+
+struct xen_domctl_monitor_op;
+
+static inline
+void arch_monitor_allow_userspace(struct domain *d, bool allow_userspace)
+{
+}
+
+static inline
+int arch_monitor_domctl_op(struct domain *d, struct xen_domctl_monitor_op *mop)
+{
+    /* No arch-specific monitor ops on GENERIC. */
+    return -EOPNOTSUPP;
+}
+
+int arch_monitor_domctl_event(struct domain *d,
+                              struct xen_domctl_monitor_op *mop);
+
+static inline
+int arch_monitor_init_domain(struct domain *d)
+{
+    /* No arch-specific domain initialization on GENERIC. */
+    return 0;
+}
+
+static inline
+void arch_monitor_cleanup_domain(struct domain *d)
+{
+    /* No arch-specific domain cleanup on GENERIC. */
+}
+
+static inline uint32_t arch_monitor_get_capabilities(struct domain *d)
+{
+    return 0;
+}
+
+#endif /* __ASM_GENERIC_MONITOR_H__ */
+
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: BSD
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 14/15] xen/asm-generic: introduce stub header numa.h

[Oleksii Kurochko]

<asm/numa.h> is common through some archs so it is moved
to asm-generic.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
	- update the commit message.
	- change u8 to uint8_t.
	- add ifnded CONFIG_NUMA.
---
 xen/include/asm-generic/numa.h | 40 ++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)
 create mode 100644 xen/include/asm-generic/numa.h

diff --git a/xen/include/asm-generic/numa.h b/xen/include/asm-generic/numa.h
new file mode 100644
index 0000000000..353642c353
--- /dev/null
+++ b/xen/include/asm-generic/numa.h
@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ARCH_GENERIC_NUMA_H
+#define __ARCH_GENERIC_NUMA_H
+
+#include <xen/types.h>
+#include <xen/mm.h>
+
+typedef uint8_t nodeid_t;
+
+#ifndef CONFIG_NUMA
+
+/* Fake one node for now. See also node_online_map. */
+#define cpu_to_node(cpu) 0
+#define node_to_cpumask(node)   (cpu_online_map)
+
+/*
+ * TODO: make first_valid_mfn static when NUMA is supported on Arm, this
+ * is required because the dummy helpers are using it.
+ */
+extern mfn_t first_valid_mfn;
+
+/* XXX: implement NUMA support */
+#define node_spanned_pages(nid) (max_page - mfn_x(first_valid_mfn))
+#define node_start_pfn(nid) (mfn_x(first_valid_mfn))
+#define __node_distance(a, b) (20)
+
+#endif
+
+#define arch_want_default_dmazone() (false)
+
+#endif /* __ARCH_GENERIC_NUMA_H */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

[PATCH v2 15/15] xen/asm-generic: introduce stub header softirq.h

[Oleksii Kurochko]

<asm/softirq.h> is common between Arm, PPC and RISC-V so it is
moved to asm-generic.

Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
---
Changes in V2:
    - update the commit message.
---
 xen/include/asm-generic/softirq.h | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)
 create mode 100644 xen/include/asm-generic/softirq.h

diff --git a/xen/include/asm-generic/softirq.h b/xen/include/asm-generic/softirq.h
new file mode 100644
index 0000000000..83be855e50
--- /dev/null
+++ b/xen/include/asm-generic/softirq.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __ASM_GENERIC_SOFTIRQ_H__
+#define __ASM_GENERIC_SOFTIRQ_H__
+
+#define NR_ARCH_SOFTIRQS       0
+
+#define arch_skip_send_event_check(cpu) 0
+
+#endif /* __ASM_GENERIC_SOFTIRQ_H__ */
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.41.0

Re: [PATCH v2 06/15] xen/asm-generic: ifdef inclusion of <asm/mem_access.h>

[Oleksii]

This patch should be reworked as it fails Arm builds:
https://gitlab.com/xen-project/people/olkur/xen/-/pipelines/1068867920

It looks like it is not possible just to #ifdef CONFIG_MEM_ACCESS.

An empty asm-generic mem_access.h will be better solution.

Could you please share your opinion?

~ Oleksii

On Fri, 2023-11-10 at 18:30 +0200, Oleksii Kurochko wrote:
> ifdefing inclusion of <asm/mem_access.h> in <xen/mem_access.h>
> allows to avoid generation of empty <asm/mem_access.h> header
> for the case when !CONFIG_MEM_ACCESS.
> 
> Suggested-by: Jan Beulich <jbeulich@suse.com>
> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> ---
> Changes in V2:
>         - add Suggested-by: Jan Beulich <jbeulich@suse.com>
>         - update the commit message
>         - remove <asm-generic/mem_access.h>
> ---
>  xen/include/xen/mem_access.h | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/xen/include/xen/mem_access.h
> b/xen/include/xen/mem_access.h
> index 4e4811680d..87d93b31f6 100644
> --- a/xen/include/xen/mem_access.h
> +++ b/xen/include/xen/mem_access.h
> @@ -33,7 +33,9 @@
>   */
>  struct vm_event_st;
>  
> +#ifdef CONFIG_MEM_ACCESS
>  #include <asm/mem_access.h>
> +#endif
>  
>  /*
>   * Additional access types, which are used to further restrict

Re: [PATCH v2 03/15] xen: ifdef inclusion of <asm/grant_table.h> in <xen/grant_table.h>

[Oleksii]

I missed to check the patch properly.

The patch fails for Arm randconfigs:
https://gitlab.com/xen-project/people/olkur/xen/-/pipelines/1068865674

I need to do an additional investigation.

Does it make sense to re-send this patch separately out of this patch
series?

~ Oleksii

On Fri, 2023-11-10 at 18:30 +0200, Oleksii Kurochko wrote:
> Ifdefing inclusion of <asm/grant_table.h> allows to avoid
> generation of empty <asm/grant_table.h> for cases when
> CONFIG_GRANT_TABLE is not enabled.
> 
> Suggested-by: Jan Beulich <jbeulich@suse.com>
> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> ---
> Changes in V2:
>         - ifdef inclusion of asm/grant_table.h in xen/grant_table.h
> to avoid
>           generation of empty headers for PPC and RISC-V archs.
>         - update commit message
>         - add Suggested-by: Jan Beulich <jbeulich@suse.com>
>         - Remove provided before asm-generic/grant_table.h header.
> ---
>  xen/include/xen/grant_table.h | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/xen/include/xen/grant_table.h
> b/xen/include/xen/grant_table.h
> index 85fe6b7b5e..50edfecfb6 100644
> --- a/xen/include/xen/grant_table.h
> +++ b/xen/include/xen/grant_table.h
> @@ -26,7 +26,10 @@
>  #include <xen/mm-frame.h>
>  #include <xen/rwlock.h>
>  #include <public/grant_table.h>
> +
> +#ifdef CONFIG_GRANT_TABLE
>  #include <asm/grant_table.h>
> +#endif
>  
>  struct grant_table;
>  

Re: [PATCH v2 03/15] xen: ifdef inclusion of <asm/grant_table.h> in <xen/grant_table.h>

[Oleksii]

On Sat, 2023-11-11 at 12:25 +0200, Oleksii wrote:
> I missed to check the patch properly.
> 
> The patch fails for Arm randconfigs:
> https://gitlab.com/xen-project/people/olkur/xen/-/pipelines/1068865674
> 
> I need to do an additional investigation.
So the only one macro cause compile issue if move it to
xen/grant_table.h compilation will pass:

--- a/xen/include/xen/grant_table.h
+++ b/xen/include/xen/grant_table.h
@@ -23,10 +23,14 @@
 #ifndef __XEN_GRANT_TABLE_H__
 #define __XEN_GRANT_TABLE_H__
 
+#include <xen/kernel.h>
 #include <xen/mm-frame.h>
 #include <xen/rwlock.h>
 #include <public/grant_table.h>
+
+#ifdef CONFIG_GRANT_TABLE
 #include <asm/grant_table.h>
+#endif
 
 struct grant_table;
 
@@ -112,6 +116,16 @@ static inline int gnttab_acquire_resource(
     return -EINVAL;
 }
 
+/*
+ * The region used by Xen on the memory will never be mapped in DOM0
+ * memory layout. Therefore it can be used for the grant table.
+ *
+ * Only use the text section as it's always present and will contain
+ * enough space for a large grant table
+ */
+#define gnttab_dom0_frames()                                         
\
+    min_t(unsigned int, opt_max_grant_frames, PFN_DOWN(_etext -
_stext))
+
 #endif /* CONFIG_GRANT_TABLE */
 
 #endif /* __XEN_GRANT_TABLE_H__ */


But gnttab_dom0_frames() is used only for ARM, so probably moving it to
<xen/grant_table.h> is not a good idea.

Could you please tell me your opinion? Thanks in advance.

~ Oleksii

> 
> Does it make sense to re-send this patch separately out of this patch
> series?
> 
> ~ Oleksii
> 
> On Fri, 2023-11-10 at 18:30 +0200, Oleksii Kurochko wrote:
> > Ifdefing inclusion of <asm/grant_table.h> allows to avoid
> > generation of empty <asm/grant_table.h> for cases when
> > CONFIG_GRANT_TABLE is not enabled.
> > 
> > Suggested-by: Jan Beulich <jbeulich@suse.com>
> > Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> > ---
> > Changes in V2:
> >         - ifdef inclusion of asm/grant_table.h in xen/grant_table.h
> > to avoid
> >           generation of empty headers for PPC and RISC-V archs.
> >         - update commit message
> >         - add Suggested-by: Jan Beulich <jbeulich@suse.com>
> >         - Remove provided before asm-generic/grant_table.h header.
> > ---
> >  xen/include/xen/grant_table.h | 3 +++
> >  1 file changed, 3 insertions(+)
> > 
> > diff --git a/xen/include/xen/grant_table.h
> > b/xen/include/xen/grant_table.h
> > index 85fe6b7b5e..50edfecfb6 100644
> > --- a/xen/include/xen/grant_table.h
> > +++ b/xen/include/xen/grant_table.h
> > @@ -26,7 +26,10 @@
> >  #include <xen/mm-frame.h>
> >  #include <xen/rwlock.h>
> >  #include <public/grant_table.h>
> > +
> > +#ifdef CONFIG_GRANT_TABLE
> >  #include <asm/grant_table.h>
> > +#endif
> >  
> >  struct grant_table;
> >  
> 

Re: [PATCH v2 03/15] xen: ifdef inclusion of <asm/grant_table.h> in <xen/grant_table.h>

[Jan Beulich]

On 13.11.2023 14:13, Oleksii wrote:
> On Sat, 2023-11-11 at 12:25 +0200, Oleksii wrote:
>> I missed to check the patch properly.
>>
>> The patch fails for Arm randconfigs:
>> https://gitlab.com/xen-project/people/olkur/xen/-/pipelines/1068865674
>>
>> I need to do an additional investigation.
> So the only one macro cause compile issue if move it to
> xen/grant_table.h compilation will pass:
> 
> --- a/xen/include/xen/grant_table.h
> +++ b/xen/include/xen/grant_table.h
> @@ -23,10 +23,14 @@
>  #ifndef __XEN_GRANT_TABLE_H__
>  #define __XEN_GRANT_TABLE_H__
>  
> +#include <xen/kernel.h>
>  #include <xen/mm-frame.h>
>  #include <xen/rwlock.h>
>  #include <public/grant_table.h>
> +
> +#ifdef CONFIG_GRANT_TABLE
>  #include <asm/grant_table.h>
> +#endif
>  
>  struct grant_table;
>  
> @@ -112,6 +116,16 @@ static inline int gnttab_acquire_resource(
>      return -EINVAL;
>  }
>  
> +/*
> + * The region used by Xen on the memory will never be mapped in DOM0
> + * memory layout. Therefore it can be used for the grant table.
> + *
> + * Only use the text section as it's always present and will contain
> + * enough space for a large grant table
> + */
> +#define gnttab_dom0_frames()                                         
> \
> +    min_t(unsigned int, opt_max_grant_frames, PFN_DOWN(_etext -
> _stext))
> +
>  #endif /* CONFIG_GRANT_TABLE */
>  
>  #endif /* __XEN_GRANT_TABLE_H__ */
> 
> 
> But gnttab_dom0_frames() is used only for ARM, so probably moving it to
> <xen/grant_table.h> is not a good idea.

Indeed. But wouldn't dealing with this again be a matter of having
Arm's domain_build.c simply include asm/grant_table.h explicitly, if need
be alongside xen/grant_table.h?

Jan

Re: [PATCH v2 03/15] xen: ifdef inclusion of <asm/grant_table.h> in <xen/grant_table.h>

[Oleksii]

On Mon, 2023-11-13 at 14:29 +0100, Jan Beulich wrote:
> On 13.11.2023 14:13, Oleksii wrote:
> > On Sat, 2023-11-11 at 12:25 +0200, Oleksii wrote:
> > > I missed to check the patch properly.
> > > 
> > > The patch fails for Arm randconfigs:
> > > https://gitlab.com/xen-project/people/olkur/xen/-/pipelines/1068865674
> > > 
> > > I need to do an additional investigation.
> > So the only one macro cause compile issue if move it to
> > xen/grant_table.h compilation will pass:
> > 
> > --- a/xen/include/xen/grant_table.h
> > +++ b/xen/include/xen/grant_table.h
> > @@ -23,10 +23,14 @@
> >  #ifndef __XEN_GRANT_TABLE_H__
> >  #define __XEN_GRANT_TABLE_H__
> >  
> > +#include <xen/kernel.h>
> >  #include <xen/mm-frame.h>
> >  #include <xen/rwlock.h>
> >  #include <public/grant_table.h>
> > +
> > +#ifdef CONFIG_GRANT_TABLE
> >  #include <asm/grant_table.h>
> > +#endif
> >  
> >  struct grant_table;
> >  
> > @@ -112,6 +116,16 @@ static inline int gnttab_acquire_resource(
> >      return -EINVAL;
> >  }
> >  
> > +/*
> > + * The region used by Xen on the memory will never be mapped in
> > DOM0
> > + * memory layout. Therefore it can be used for the grant table.
> > + *
> > + * Only use the text section as it's always present and will
> > contain
> > + * enough space for a large grant table
> > + */
> > +#define
> > gnttab_dom0_frames()                                         
> > \
> > +    min_t(unsigned int, opt_max_grant_frames, PFN_DOWN(_etext -
> > _stext))
> > +
> >  #endif /* CONFIG_GRANT_TABLE */
> >  
> >  #endif /* __XEN_GRANT_TABLE_H__ */
> > 
> > 
> > But gnttab_dom0_frames() is used only for ARM, so probably moving
> > it to
> > <xen/grant_table.h> is not a good idea.
> 
> Indeed. But wouldn't dealing with this again be a matter of having
> Arm's domain_build.c simply include asm/grant_table.h explicitly, if
> need
> be alongside xen/grant_table.h?
It can be a solution. Then I'll send a separate patch.

Thanks.

~ Oleksii

Re: [PATCH v2 01/15] xen/asm-generic: introduce stub header paging.h

[Jan Beulich]

On 10.11.2023 17:30, Oleksii Kurochko wrote:
> The patch introduces generic paging.h header for Arm, PPC and
> RISC-V.
> 
> All mentioned above architectures use hardware virt extensions
> and hardware pagetable extensions thereby it makes sense to set
> paging_mode_translate and paging_mode_external by default.
> 
> Also in this patch Arm and PPC architectures are switched to
> generic paging.h header.
> 
> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>

Reviewed-by: Jan Beulich <jbeulich@suse.com>
with one nit (applicable twice):

> --- a/xen/arch/arm/include/asm/Makefile
> +++ b/xen/arch/arm/include/asm/Makefile
> @@ -1,2 +1,3 @@
>  # SPDX-License-Identifier: GPL-2.0-only
>  generic-y += vm_event.h
> +generic-y += paging.h

Like in obj-y lists it would be nice for the items here and ...

> --- a/xen/arch/ppc/include/asm/Makefile
> +++ b/xen/arch/ppc/include/asm/Makefile
> @@ -1,2 +1,3 @@
>  # SPDX-License-Identifier: GPL-2.0-only
>  generic-y += vm_event.h
> +generic-y += paging.h

... here to be sorted alphabetically from the beginning. Right here I'd
be happy to make the change while committing, but I'm not going to promise
making the same offer on subsequent changes (should the same issue exist
there).

Jan

Re: [PATCH v2 02/15] xen/asm-generic: introduce generic device.h

[Jan Beulich]

On 10.11.2023 17:30, Oleksii Kurochko wrote:
> --- /dev/null
> +++ b/xen/include/asm-generic/device.h
> @@ -0,0 +1,140 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +#ifndef __ASM_GENERIC_DEVICE_H__
> +#define __ASM_GENERIC_DEVICE_H__
> +
> +enum device_type
> +{
> +    DEV_DT,
> +#ifdef HAS_PCI
> +    DEV_PCI,
> +#endif
> +};
> +
> +struct dev_archdata {
> +    void *iommu;    /* IOMMU private data */
> +};
> +
> +/* struct device - The basic device structure */
> +struct device
> +{
> +    enum device_type type;
> +#ifdef CONFIG_HAS_DEVICE_TREE
> +    struct dt_device_node *of_node; /* Used by drivers imported from Linux */
> +#endif

There's just this instance where HAS_DEVICE_TREE is checked. Why not elsewhere?
Imo, if you really want this header in asm-generic/, then it wants to be truly
generic (i.e. not implying DT just like you're not implying PCI or ACPI).

Jan

Re: [PATCH v2 04/15] xen/asm-generic: introduce generic hypercall.h

[Jan Beulich]

On 10.11.2023 17:30, Oleksii Kurochko wrote:
> Introduce an empty generic hypercall.h for archs which don't
> implement it.
> 
> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>

Since - judging from Arm's present header - it looks technically possible
for an arch to get away with this:
Acked-by: Jan Beulich <jbeulich@suse.com>

Jan

Re: [PATCH v2 05/15] xen/asm-generic: introduce generic header iocap.h

[Jan Beulich]

On 10.11.2023 17:30, Oleksii Kurochko wrote:
> iocap.h is common for Arm, PPC and RISC-V architectures thereby
> it was moved to asm-generic.
> 
> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> ---
> The same question as with device.h. Should it be in asm-generic?

I think it's okay(ish) here, considering that ...

> --- /dev/null
> +++ b/xen/include/asm-generic/iocap.h
> @@ -0,0 +1,17 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +#ifndef __ASM_GENERIC_IOCAP_H__
> +#define __ASM_GENERIC_IOCAP_H__
> +
> +#define cache_flush_permitted(d)                        \
> +    (!rangeset_is_empty((d)->iomem_caps))

... the only thing referenced (iomem_caps) is a common entity. And
again Arm demonstrates that an arch can get away with just this.

Jan

Re: [PATCH v2 04/15] xen/asm-generic: introduce generic hypercall.h

[Jan Beulich]

On 13.11.2023 17:45, Jan Beulich wrote:
> On 10.11.2023 17:30, Oleksii Kurochko wrote:
>> Introduce an empty generic hypercall.h for archs which don't
>> implement it.
>>
>> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> 
> Since - judging from Arm's present header - it looks technically possible
> for an arch to get away with this:
> Acked-by: Jan Beulich <jbeulich@suse.com>

Actually - why does this patch not drop PPC's now redundant header right
away, along the lines of what you do in patch 1? This is also why I
refrained from offering A-b on patch 5.

Jan

Re: [PATCH v2 06/15] xen/asm-generic: ifdef inclusion of <asm/mem_access.h>

[Jan Beulich]

On 11.11.2023 11:24, Oleksii wrote:
> This patch should be reworked as it fails Arm builds:
> https://gitlab.com/xen-project/people/olkur/xen/-/pipelines/1068867920

Took me a while to actually find the error. Would have been nice if you
had explained what's going wrong, supplying the link only as extra info.

> It looks like it is not possible just to #ifdef CONFIG_MEM_ACCESS.
> 
> An empty asm-generic mem_access.h will be better solution.

I remain unconvinced. The issue looks to be with p2m_mem_access_check().
One solution would be to move that declaration into the common header.
But there's no common code referencing it, so Arm's and x86's version
might as well diverge at some point. Hence from my pov it again boils
down to Arm's traps.c including asm/mem_access.h explicitly, to bring
the declaration in scope. None of the common files really have a need
to have a dependency on the arch-specific header when MEM_ACCESS=n.

Jan

Re: [PATCH v2 06/15] xen/asm-generic: ifdef inclusion of <asm/mem_access.h>

[Oleksii]

On Mon, 2023-11-13 at 18:01 +0100, Jan Beulich wrote:
> On 11.11.2023 11:24, Oleksii wrote:
> > This patch should be reworked as it fails Arm builds:
> > https://gitlab.com/xen-project/people/olkur/xen/-/pipelines/1068867920
> 
> Took me a while to actually find the error. Would have been nice if
> you
> had explained what's going wrong, supplying the link only as extra
> info.
Sorry about that. I'll be more precise with links next time! But the
issue is exactly what you wrote below ( with p2m_mem_access_check and
some others from <asm/mem_access.h> ) .

> 
> > It looks like it is not possible just to #ifdef CONFIG_MEM_ACCESS.
> > 
> > An empty asm-generic mem_access.h will be better solution.
> 
> I remain unconvinced. The issue looks to be with
> p2m_mem_access_check().
> One solution would be to move that declaration into the common
> header.
> But there's no common code referencing it, so Arm's and x86's version
> might as well diverge at some point. Hence from my pov it again boils
> down to Arm's traps.c including asm/mem_access.h explicitly, to bring
> the declaration in scope. None of the common files really have a need
> to have a dependency on the arch-specific header when MEM_ACCESS=n.
I started to do that in that way you mentioned.

It should be included in 2 files, at least: p2m.c and traps.c.
I am finishind the testing if it is enough.
If it is enough, I will send a separate patch.

~ Oleksii

Re: [PATCH v2 04/15] xen/asm-generic: introduce generic hypercall.h

[Oleksii]

On Mon, 2023-11-13 at 17:50 +0100, Jan Beulich wrote:
> On 13.11.2023 17:45, Jan Beulich wrote:
> > On 10.11.2023 17:30, Oleksii Kurochko wrote:
> > > Introduce an empty generic hypercall.h for archs which don't
> > > implement it.
> > > 
> > > Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> > 
> > Since - judging from Arm's present header - it looks technically
> > possible
> > for an arch to get away with this:
> > Acked-by: Jan Beulich <jbeulich@suse.com>
> 
> Actually - why does this patch not drop PPC's now redundant header
> right
> away, along the lines of what you do in patch 1? This is also why I
> refrained from offering A-b on patch 5.
Missed that. I'll send a new patch series with dropped PPC's redundant
header.

Thanks.

~ Oleksii

Re: [PATCH v2 05/15] xen/asm-generic: introduce generic header iocap.h

[Oleksii]

On Mon, 2023-11-13 at 17:48 +0100, Jan Beulich wrote:
> On 10.11.2023 17:30, Oleksii Kurochko wrote:
> > iocap.h is common for Arm, PPC and RISC-V architectures thereby
> > it was moved to asm-generic.
> > 
> > Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> > ---
> > The same question as with device.h. Should it be in asm-generic?
> 
> I think it's okay(ish) here, considering that ...
> 
> > --- /dev/null
> > +++ b/xen/include/asm-generic/iocap.h
> > @@ -0,0 +1,17 @@
> > +/* SPDX-License-Identifier: GPL-2.0-only */
> > +#ifndef __ASM_GENERIC_IOCAP_H__
> > +#define __ASM_GENERIC_IOCAP_H__
> > +
> > +#define cache_flush_permitted(d)                        \
> > +    (!rangeset_is_empty((d)->iomem_caps))
> 
> ... the only thing referenced (iomem_caps) is a common entity. And
> again Arm demonstrates that an arch can get away with just this.
Thanks.

Then I remove redundant headers from Arm and PPC and send new patch
serires.

~ Oleksii

Re: [PATCH v2 02/15] xen/asm-generic: introduce generic device.h

[Oleksii]

On Mon, 2023-11-13 at 17:43 +0100, Jan Beulich wrote:
> On 10.11.2023 17:30, Oleksii Kurochko wrote:
> > --- /dev/null
> > +++ b/xen/include/asm-generic/device.h
> > @@ -0,0 +1,140 @@
> > +/* SPDX-License-Identifier: GPL-2.0-only */
> > +#ifndef __ASM_GENERIC_DEVICE_H__
> > +#define __ASM_GENERIC_DEVICE_H__
> > +
> > +enum device_type
> > +{
> > +    DEV_DT,
> > +#ifdef HAS_PCI
> > +    DEV_PCI,
> > +#endif
> > +};
> > +
> > +struct dev_archdata {
> > +    void *iommu;    /* IOMMU private data */
> > +};
> > +
> > +/* struct device - The basic device structure */
> > +struct device
> > +{
> > +    enum device_type type;
> > +#ifdef CONFIG_HAS_DEVICE_TREE
> > +    struct dt_device_node *of_node; /* Used by drivers imported
> > from Linux */
> > +#endif
> 
> There's just this instance where HAS_DEVICE_TREE is checked. Why not
> elsewhere?
> Imo, if you really want this header in asm-generic/, then it wants to
> be truly
> generic (i.e. not implying DT just like you're not implying PCI or
> ACPI).
Mostly copied this file from Arm and only that one part was #ifdef-ed
with HAS_DEVICE_TREE.
But it makes sense to #ifdef DEV_DT, inclusion of <xen/device_tree.h>
and definiotn of dev_is_dt() macros.

I'll made the changes in next version of patch series.

Thanks for comments.

~ Oleksii

Re: [PATCH v2 07/15] xen/asm-generic: introduce stub header <asm/random.h>

[Jan Beulich]

On 10.11.2023 17:30, Oleksii Kurochko wrote:
> <asm/random.h> is common for Arm, PPC and RISC-V thereby it
> is moved to asm-generic.

When you say "moved", ...

> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> ---
> Changes in V2:
>  - update the commit messages
> ---
>  xen/include/asm-generic/random.h | 20 ++++++++++++++++++++
>  1 file changed, 20 insertions(+)
>  create mode 100644 xen/include/asm-generic/random.h

... you also want to actually move things.

Since the above comment matches ones on earlier patches, yet otoh in your
submissions of two individual patches you mentioned you sent them
separately because this series wasn't fully reviewed yet, would you mind
clarifying whether further going through this version of the series is
actually going to be a good use of time?

Jan

Re: [PATCH v2 13/15] xen/asm-generic: introduce stub header monitor.h

[Jan Beulich]

On 10.11.2023 17:30, Oleksii Kurochko wrote:
> --- /dev/null
> +++ b/xen/include/asm-generic/monitor.h
> @@ -0,0 +1,62 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/*
> + * include/asm-GENERIC/monitor.h
> + *
> + * Arch-specific monitor_op domctl handler.
> + *
> + * Copyright (c) 2015 Tamas K Lengyel (tamas@tklengyel.com)
> + * Copyright (c) 2016, Bitdefender S.R.L.
> + *
> + */
> +
> +#ifndef __ASM_GENERIC_MONITOR_H__
> +#define __ASM_GENERIC_MONITOR_H__
> +
> +#include <xen/sched.h>

What is this needed for? I expect ...

> +struct xen_domctl_monitor_op;
> +
> +static inline
> +void arch_monitor_allow_userspace(struct domain *d, bool allow_userspace)

... struct domain, but since you never de-reference any such pointer, forward-
declaring that (just like struct xen_domctl_monitor_op) would do here. Which
would leave you with needing at most xen/types.h, but maybe as little as
xen/stdbool.h and xen/stdint.h.

Jan

> +{
> +}
> +
> +static inline
> +int arch_monitor_domctl_op(struct domain *d, struct xen_domctl_monitor_op *mop)
> +{
> +    /* No arch-specific monitor ops on GENERIC. */
> +    return -EOPNOTSUPP;
> +}
> +
> +int arch_monitor_domctl_event(struct domain *d,
> +                              struct xen_domctl_monitor_op *mop);
> +
> +static inline
> +int arch_monitor_init_domain(struct domain *d)
> +{
> +    /* No arch-specific domain initialization on GENERIC. */
> +    return 0;
> +}
> +
> +static inline
> +void arch_monitor_cleanup_domain(struct domain *d)
> +{
> +    /* No arch-specific domain cleanup on GENERIC. */
> +}
> +
> +static inline uint32_t arch_monitor_get_capabilities(struct domain *d)
> +{
> +    return 0;
> +}
> +
> +#endif /* __ASM_GENERIC_MONITOR_H__ */
> +
> +
> +/*
> + * Local variables:
> + * mode: C
> + * c-file-style: BSD
> + * c-basic-offset: 4
> + * indent-tabs-mode: nil
> + * End:
> + */

Re: [PATCH v2 14/15] xen/asm-generic: introduce stub header numa.h

[Jan Beulich]

On 10.11.2023 17:30, Oleksii Kurochko wrote:
> --- /dev/null
> +++ b/xen/include/asm-generic/numa.h
> @@ -0,0 +1,40 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +#ifndef __ARCH_GENERIC_NUMA_H
> +#define __ARCH_GENERIC_NUMA_H
> +
> +#include <xen/types.h>
> +#include <xen/mm.h>

I'm afraid I can't spot what you need these for here. You clearly need
xen/stdint.h, and you need xen/mm-frame.h for mfn_t. Yes, max_page is
declared in xen/mm.h, but its questionable whether the header needs
including here for that reason, as all uses are in macros. (We aren't
anywhere near consistent in this regard). Plus you don't also include
xen/cpumask.h to pull in cpu_online_map (which another macro
references).

> +typedef uint8_t nodeid_t;
> +
> +#ifndef CONFIG_NUMA

Isn't it an error to include this header when NUMA=y?

> +/* Fake one node for now. See also node_online_map. */
> +#define cpu_to_node(cpu) 0
> +#define node_to_cpumask(node)   (cpu_online_map)
> +
> +/*
> + * TODO: make first_valid_mfn static when NUMA is supported on Arm, this
> + * is required because the dummy helpers are using it.
> + */
> +extern mfn_t first_valid_mfn;
> +
> +/* XXX: implement NUMA support */
> +#define node_spanned_pages(nid) (max_page - mfn_x(first_valid_mfn))
> +#define node_start_pfn(nid) (mfn_x(first_valid_mfn))
> +#define __node_distance(a, b) (20)
> +
> +#endif
> +
> +#define arch_want_default_dmazone() (false)
> +
> +#endif /* __ARCH_GENERIC_NUMA_H */
> +
> +/*
> + * Local variables:
> + * mode: C
> + * c-file-style: "BSD"
> + * c-basic-offset: 4
> + * indent-tabs-mode: nil
> + * End:
> + */

Re: [PATCH v2 07/15] xen/asm-generic: introduce stub header <asm/random.h>

[Oleksii]

On Wed, 2023-11-15 at 10:56 +0100, Jan Beulich wrote:
> On 10.11.2023 17:30, Oleksii Kurochko wrote:
> > <asm/random.h> is common for Arm, PPC and RISC-V thereby it
> > is moved to asm-generic.
> 
> When you say "moved", ...
> 
> > Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> > ---
> > Changes in V2:
> >  - update the commit messages
> > ---
> >  xen/include/asm-generic/random.h | 20 ++++++++++++++++++++
> >  1 file changed, 20 insertions(+)
> >  create mode 100644 xen/include/asm-generic/random.h
> 
> ... you also want to actually move things.
Sure, I'll delete Arm and PPC's random.h in the next patch series
version.

> 
> Since the above comment matches ones on earlier patches, yet otoh in
> your
> submissions of two individual patches you mentioned you sent them
> separately because this series wasn't fully reviewed yet, would you
> mind
> clarifying whether further going through this version of the series
> is
> actually going to be a good use of time?
I think it still makes sense to review this series.

I probably have to stop sending patches from this series separately. I
thought merging almost-ready patches would be a little faster if they
moved outside the patch series.

If it is possible to merge approved patches separately without getting
approved for the whole patch series, then what I did before doesn't
make sense, and I am sorry for this inconvenience.

I can return the patches I sent separately to this patch series.

~ Oleksii

Re: [PATCH v2 13/15] xen/asm-generic: introduce stub header monitor.h

[Oleksii]

On Wed, 2023-11-15 at 11:00 +0100, Jan Beulich wrote:
> On 10.11.2023 17:30, Oleksii Kurochko wrote:
> > --- /dev/null
> > +++ b/xen/include/asm-generic/monitor.h
> > @@ -0,0 +1,62 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +/*
> > + * include/asm-GENERIC/monitor.h
> > + *
> > + * Arch-specific monitor_op domctl handler.
> > + *
> > + * Copyright (c) 2015 Tamas K Lengyel (tamas@tklengyel.com)
> > + * Copyright (c) 2016, Bitdefender S.R.L.
> > + *
> > + */
> > +
> > +#ifndef __ASM_GENERIC_MONITOR_H__
> > +#define __ASM_GENERIC_MONITOR_H__
> > +
> > +#include <xen/sched.h>
> 
> What is this needed for? I expect ...
> 
> > +struct xen_domctl_monitor_op;
> > +
> > +static inline
> > +void arch_monitor_allow_userspace(struct domain *d, bool
> > allow_userspace)
> 
> ... struct domain, but since you never de-reference any such pointer,
> forward-
> declaring that (just like struct xen_domctl_monitor_op) would do
> here. Which
> would leave you with needing at most xen/types.h, but maybe as little
> as
> xen/stdbool.h and xen/stdint.h.
Yes, the reason for " #include <xen/sched.h> " was ' struct domain '.
Let's switch to forward-declaring.

Shouldn't it be included <xen/compiler.h> too for inline?

~ Oleksii

> > +{
> > +}
> > +
> > +static inline
> > +int arch_monitor_domctl_op(struct domain *d, struct
> > xen_domctl_monitor_op *mop)
> > +{
> > +    /* No arch-specific monitor ops on GENERIC. */
> > +    return -EOPNOTSUPP;
> > +}
> > +
> > +int arch_monitor_domctl_event(struct domain *d,
> > +                              struct xen_domctl_monitor_op *mop);
> > +
> > +static inline
> > +int arch_monitor_init_domain(struct domain *d)
> > +{
> > +    /* No arch-specific domain initialization on GENERIC. */
> > +    return 0;
> > +}
> > +
> > +static inline
> > +void arch_monitor_cleanup_domain(struct domain *d)
> > +{
> > +    /* No arch-specific domain cleanup on GENERIC. */
> > +}
> > +
> > +static inline uint32_t arch_monitor_get_capabilities(struct domain
> > *d)
> > +{
> > +    return 0;
> > +}
> > +
> > +#endif /* __ASM_GENERIC_MONITOR_H__ */
> > +
> > +
> > +/*
> > + * Local variables:
> > + * mode: C
> > + * c-file-style: BSD
> > + * c-basic-offset: 4
> > + * indent-tabs-mode: nil
> > + * End:
> > + */
> 

Re: [PATCH v2 14/15] xen/asm-generic: introduce stub header numa.h

[Oleksii]

On Wed, 2023-11-15 at 11:07 +0100, Jan Beulich wrote:
> On 10.11.2023 17:30, Oleksii Kurochko wrote:
> > --- /dev/null
> > +++ b/xen/include/asm-generic/numa.h
> > @@ -0,0 +1,40 @@
> > +/* SPDX-License-Identifier: GPL-2.0-only */
> > +#ifndef __ARCH_GENERIC_NUMA_H
> > +#define __ARCH_GENERIC_NUMA_H
> > +
> > +#include <xen/types.h>
> > +#include <xen/mm.h>
> 
> I'm afraid I can't spot what you need these for here. You clearly
> need
> xen/stdint.h, and you need xen/mm-frame.h for mfn_t. Yes, max_page is
> declared in xen/mm.h, but its questionable whether the header needs
> including here for that reason, as all uses are in macros. (We aren't
> anywhere near consistent in this regard). Plus you don't also include
> xen/cpumask.h to pull in cpu_online_map (which another macro
> references).
I agree with almost you wrote but should <xen/cpumas.h> be included
then? It looks like it is the same situation as with max_page and
<xen/mm.h>.

> 
> > +typedef uint8_t nodeid_t;
> > +
> > +#ifndef CONFIG_NUMA
> 
> Isn't it an error to include this header when NUMA=y?
It's still need to define arch_want_default_dmazone() macros which is
used by common code.
All other code is under #ifndef CONFIG_NUMA so it won't conflict with
defintions in <xen/numa.h>.

> 
> > +/* Fake one node for now. See also node_online_map. */
> > +#define cpu_to_node(cpu) 0
> > +#define node_to_cpumask(node)   (cpu_online_map)
> > +
> > +/*
> > + * TODO: make first_valid_mfn static when NUMA is supported on
> > Arm, this
> > + * is required because the dummy helpers are using it.
> > + */
> > +extern mfn_t first_valid_mfn;
> > +
> > +/* XXX: implement NUMA support */
> > +#define node_spanned_pages(nid) (max_page -
> > mfn_x(first_valid_mfn))
> > +#define node_start_pfn(nid) (mfn_x(first_valid_mfn))
> > +#define __node_distance(a, b) (20)
> > +
> > +#endif
> > +
> > +#define arch_want_default_dmazone() (false)
> > +
> > +#endif /* __ARCH_GENERIC_NUMA_H */
> > +
> > +/*
> > + * Local variables:
> > + * mode: C
> > + * c-file-style: "BSD"
> > + * c-basic-offset: 4
> > + * indent-tabs-mode: nil
> > + * End:
> > + */
> 

~ Oleksii

Re: [PATCH v2 13/15] xen/asm-generic: introduce stub header monitor.h

[Oleksii]

On Wed, 2023-11-15 at 14:54 +0200, Oleksii wrote:
> On Wed, 2023-11-15 at 11:00 +0100, Jan Beulich wrote:
> > On 10.11.2023 17:30, Oleksii Kurochko wrote:
> > > --- /dev/null
> > > +++ b/xen/include/asm-generic/monitor.h
> > > @@ -0,0 +1,62 @@
> > > +/* SPDX-License-Identifier: GPL-2.0 */
> > > +/*
> > > + * include/asm-GENERIC/monitor.h
> > > + *
> > > + * Arch-specific monitor_op domctl handler.
> > > + *
> > > + * Copyright (c) 2015 Tamas K Lengyel (tamas@tklengyel.com)
> > > + * Copyright (c) 2016, Bitdefender S.R.L.
> > > + *
> > > + */
> > > +
> > > +#ifndef __ASM_GENERIC_MONITOR_H__
> > > +#define __ASM_GENERIC_MONITOR_H__
> > > +
> > > +#include <xen/sched.h>
> > 
> > What is this needed for? I expect ...
> > 
> > > +struct xen_domctl_monitor_op;
> > > +
> > > +static inline
> > > +void arch_monitor_allow_userspace(struct domain *d, bool
> > > allow_userspace)
> > 
> > ... struct domain, but since you never de-reference any such
> > pointer,
> > forward-
> > declaring that (just like struct xen_domctl_monitor_op) would do
> > here. Which
> > would leave you with needing at most xen/types.h, but maybe as
> > little
> > as
> > xen/stdbool.h and xen/stdint.h.
> Yes, the reason for " #include <xen/sched.h> " was ' struct domain '.
> Let's switch to forward-declaring.
> 
> Shouldn't it be included <xen/compiler.h> too for inline?
It should be added "#include <xen/errno.h>" because EOPNOTSUPP is used
in arch_monitor_domctl_op.
> 
> ~ Oleksii
> 
> > > +{
> > > +}
> > > +
> > > +static inline
> > > +int arch_monitor_domctl_op(struct domain *d, struct
> > > xen_domctl_monitor_op *mop)
> > > +{
> > > +    /* No arch-specific monitor ops on GENERIC. */
> > > +    return -EOPNOTSUPP;
> > > +}
> > > +
> > > +int arch_monitor_domctl_event(struct domain *d,
> > > +                              struct xen_domctl_monitor_op
> > > *mop);
> > > +
> > > +static inline
> > > +int arch_monitor_init_domain(struct domain *d)
> > > +{
> > > +    /* No arch-specific domain initialization on GENERIC. */
> > > +    return 0;
> > > +}
> > > +
> > > +static inline
> > > +void arch_monitor_cleanup_domain(struct domain *d)
> > > +{
> > > +    /* No arch-specific domain cleanup on GENERIC. */
> > > +}
> > > +
> > > +static inline uint32_t arch_monitor_get_capabilities(struct
> > > domain
> > > *d)
> > > +{
> > > +    return 0;
> > > +}
> > > +
> > > +#endif /* __ASM_GENERIC_MONITOR_H__ */
> > > +
> > > +
> > > +/*
> > > + * Local variables:
> > > + * mode: C
> > > + * c-file-style: BSD
> > > + * c-basic-offset: 4
> > > + * indent-tabs-mode: nil
> > > + * End:
> > > + */
> > 
> 

Re: [PATCH v2 08/15] xen/asm-generic: introduce generic header percpu.h

[Oleksii]

I have to drop arch specific headers and switch to asm-generic version.
I'll do the same for patches 09-12 in the next patch version.

~ Oleksii
On Fri, 2023-11-10 at 18:30 +0200, Oleksii Kurochko wrote:
> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> ---
> Changes in V2:
>         - use smp_processor_id() instead of get_processor_id().
>         - update commit message .
> ---
>  xen/include/asm-generic/percpu.h | 35
> ++++++++++++++++++++++++++++++++
>  1 file changed, 35 insertions(+)
>  create mode 100644 xen/include/asm-generic/percpu.h
> 
> diff --git a/xen/include/asm-generic/percpu.h b/xen/include/asm-
> generic/percpu.h
> new file mode 100644
> index 0000000000..85a3f3ef17
> --- /dev/null
> +++ b/xen/include/asm-generic/percpu.h
> @@ -0,0 +1,35 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +#ifndef __ASM_GENERIC_PERCPU_H__
> +#define __ASM_GENERIC_PERCPU_H__
> +
> +#ifndef __ASSEMBLY__
> +
> +#include <xen/types.h>
> +
> +extern char __per_cpu_start[], __per_cpu_data_end[];
> +extern unsigned long __per_cpu_offset[NR_CPUS];
> +void percpu_init_areas(void);
> +
> +#define per_cpu(var, cpu)  \
> +    (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset[cpu]))
> +
> +#define this_cpu(var) \
> +    (*RELOC_HIDE(&per_cpu__##var,
> __per_cpu_offset[smp_processor_id()]))
> +
> +#define per_cpu_ptr(var, cpu)  \
> +    (*RELOC_HIDE(var, __per_cpu_offset[cpu]))
> +#define this_cpu_ptr(var) \
> +    (*RELOC_HIDE(var, smp_processor_id()))
> +
> +#endif
> +
> +#endif /* __ASM_GENERIC_PERCPU_H__ */
> +
> +/*
> + * Local variables:
> + * mode: C
> + * c-file-style: "BSD"
> + * c-basic-offset: 4
> + * indent-tabs-mode: nil
> + * End:
> + */

Re: [PATCH v2 07/15] xen/asm-generic: introduce stub header <asm/random.h>

[Jan Beulich]

On 15.11.2023 13:39, Oleksii wrote:
> On Wed, 2023-11-15 at 10:56 +0100, Jan Beulich wrote:
>> On 10.11.2023 17:30, Oleksii Kurochko wrote:
>>> <asm/random.h> is common for Arm, PPC and RISC-V thereby it
>>> is moved to asm-generic.
>>
>> When you say "moved", ...
>>
>>> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
>>> ---
>>> Changes in V2:
>>>  - update the commit messages
>>> ---
>>>  xen/include/asm-generic/random.h | 20 ++++++++++++++++++++
>>>  1 file changed, 20 insertions(+)
>>>  create mode 100644 xen/include/asm-generic/random.h
>>
>> ... you also want to actually move things.
> Sure, I'll delete Arm and PPC's random.h in the next patch series
> version.
> 
>>
>> Since the above comment matches ones on earlier patches, yet otoh in
>> your
>> submissions of two individual patches you mentioned you sent them
>> separately because this series wasn't fully reviewed yet, would you
>> mind
>> clarifying whether further going through this version of the series
>> is
>> actually going to be a good use of time?
> I think it still makes sense to review this series.
> 
> I probably have to stop sending patches from this series separately. I
> thought merging almost-ready patches would be a little faster if they
> moved outside the patch series.
> 
> If it is possible to merge approved patches separately without getting
> approved for the whole patch series,

We do this quite frequently, as long as it's clear that later patches
in a series (which are approved and hence can go in) don't depend on
earlier ones. Ones at the beginning of a series can go individually
anyway; the only time that I can think of right away where this would
not be desirable is if they introduced then-dead code.

Jan

> then what I did before doesn't
> make sense, and I am sorry for this inconvenience.
> 
> I can return the patches I sent separately to this patch series.
> 
> ~ Oleksii
> 

Re: [PATCH v2 13/15] xen/asm-generic: introduce stub header monitor.h

[Jan Beulich]

On 15.11.2023 13:54, Oleksii wrote:
> On Wed, 2023-11-15 at 11:00 +0100, Jan Beulich wrote:
>> On 10.11.2023 17:30, Oleksii Kurochko wrote:
>>> --- /dev/null
>>> +++ b/xen/include/asm-generic/monitor.h
>>> @@ -0,0 +1,62 @@
>>> +/* SPDX-License-Identifier: GPL-2.0 */
>>> +/*
>>> + * include/asm-GENERIC/monitor.h
>>> + *
>>> + * Arch-specific monitor_op domctl handler.
>>> + *
>>> + * Copyright (c) 2015 Tamas K Lengyel (tamas@tklengyel.com)
>>> + * Copyright (c) 2016, Bitdefender S.R.L.
>>> + *
>>> + */
>>> +
>>> +#ifndef __ASM_GENERIC_MONITOR_H__
>>> +#define __ASM_GENERIC_MONITOR_H__
>>> +
>>> +#include <xen/sched.h>
>>
>> What is this needed for? I expect ...
>>
>>> +struct xen_domctl_monitor_op;
>>> +
>>> +static inline
>>> +void arch_monitor_allow_userspace(struct domain *d, bool
>>> allow_userspace)
>>
>> ... struct domain, but since you never de-reference any such pointer,
>> forward-
>> declaring that (just like struct xen_domctl_monitor_op) would do
>> here. Which
>> would leave you with needing at most xen/types.h, but maybe as little
>> as
>> xen/stdbool.h and xen/stdint.h.
> Yes, the reason for " #include <xen/sched.h> " was ' struct domain '.
> Let's switch to forward-declaring.
> 
> Shouldn't it be included <xen/compiler.h> too for inline?

I'm actually not sure why we (still?) override "inline" there. It's a
normal keyword in C99. IOW I don't think xen/compiler.h would be
needed here (just for that).

Jan

Re: [PATCH v2 14/15] xen/asm-generic: introduce stub header numa.h

[Jan Beulich]

On 15.11.2023 15:11, Oleksii wrote:
> On Wed, 2023-11-15 at 11:07 +0100, Jan Beulich wrote:
>> On 10.11.2023 17:30, Oleksii Kurochko wrote:
>>> --- /dev/null
>>> +++ b/xen/include/asm-generic/numa.h
>>> @@ -0,0 +1,40 @@
>>> +/* SPDX-License-Identifier: GPL-2.0-only */
>>> +#ifndef __ARCH_GENERIC_NUMA_H
>>> +#define __ARCH_GENERIC_NUMA_H
>>> +
>>> +#include <xen/types.h>
>>> +#include <xen/mm.h>
>>
>> I'm afraid I can't spot what you need these for here. You clearly
>> need
>> xen/stdint.h, and you need xen/mm-frame.h for mfn_t. Yes, max_page is
>> declared in xen/mm.h, but its questionable whether the header needs
>> including here for that reason, as all uses are in macros. (We aren't
>> anywhere near consistent in this regard). Plus you don't also include
>> xen/cpumask.h to pull in cpu_online_map (which another macro
>> references).
> I agree with almost you wrote but should <xen/cpumas.h> be included
> then? It looks like it is the same situation as with max_page and
> <xen/mm.h>.

Well, yes and no: Indeed the minimal requirement is to be consistent
(either include both or include neither). Personally I'd prefer if
headers would be included only if they are needed to successfully
compiler the header on its own. That limits needless dependencies on
the (otherwise) included headers. But I can easily see that others
might take the other sensible perspective.

>>> +typedef uint8_t nodeid_t;
>>> +
>>> +#ifndef CONFIG_NUMA
>>
>> Isn't it an error to include this header when NUMA=y?
> It's still need to define arch_want_default_dmazone() macros which is
> used by common code.

Oh, yes. I somehow managed to overlook that. Some of the #include-s then
want to move inside the #ifndef, though. (Ultimately I question the
placement of this #define in this header, but we can deal with that
separately.)

Jan

Re: [PATCH v2 08/15] xen/asm-generic: introduce generic header percpu.h

[Jan Beulich]

On 10.11.2023 17:30, Oleksii Kurochko wrote:
> --- /dev/null
> +++ b/xen/include/asm-generic/percpu.h
> @@ -0,0 +1,35 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +#ifndef __ASM_GENERIC_PERCPU_H__
> +#define __ASM_GENERIC_PERCPU_H__
> +
> +#ifndef __ASSEMBLY__
> +
> +#include <xen/types.h>
> +
> +extern char __per_cpu_start[], __per_cpu_data_end[];

Can we go one tiny step beyond what Arm presently has and make the
latter of the two const?

Jan

Re: [PATCH v2 08/15] xen/asm-generic: introduce generic header percpu.h

[Oleksii]

On Thu, 2023-11-16 at 08:36 +0100, Jan Beulich wrote:
> On 10.11.2023 17:30, Oleksii Kurochko wrote:
> > --- /dev/null
> > +++ b/xen/include/asm-generic/percpu.h
> > @@ -0,0 +1,35 @@
> > +/* SPDX-License-Identifier: GPL-2.0-only */
> > +#ifndef __ASM_GENERIC_PERCPU_H__
> > +#define __ASM_GENERIC_PERCPU_H__
> > +
> > +#ifndef __ASSEMBLY__
> > +
> > +#include <xen/types.h>
> > +
> > +extern char __per_cpu_start[], __per_cpu_data_end[];
> 
> Can we go one tiny step beyond what Arm presently has and make the
> latter of the two const?
I am not sure I think we will have compilation issue with the following
code in Arm and x86 because of [-Werror=discarded-qualifiers]:

static void cf_check _free_percpu_area(struct rcu_head *head)
{
    struct free_info *info = container_of(head, struct free_info, rcu);
    unsigned int cpu = info->cpu;
    char *p = __per_cpu_start + __per_cpu_offset[cpu];

    free_xenheap_pages(p, PERCPU_ORDER);
    __per_cpu_offset[cpu] = INVALID_PERCPU_AREA;
}

I guess cast can help.


~ Oleksii

Re: [PATCH v2 08/15] xen/asm-generic: introduce generic header percpu.h

[Jan Beulich]

On 16.11.2023 13:04, Oleksii wrote:
> On Thu, 2023-11-16 at 08:36 +0100, Jan Beulich wrote:
>> On 10.11.2023 17:30, Oleksii Kurochko wrote:
>>> --- /dev/null
>>> +++ b/xen/include/asm-generic/percpu.h
>>> @@ -0,0 +1,35 @@
>>> +/* SPDX-License-Identifier: GPL-2.0-only */
>>> +#ifndef __ASM_GENERIC_PERCPU_H__
>>> +#define __ASM_GENERIC_PERCPU_H__
>>> +
>>> +#ifndef __ASSEMBLY__
>>> +
>>> +#include <xen/types.h>
>>> +
>>> +extern char __per_cpu_start[], __per_cpu_data_end[];
>>
>> Can we go one tiny step beyond what Arm presently has and make the
>> latter of the two const?
> I am not sure I think we will have compilation issue with the following
> code in Arm and x86 because of [-Werror=discarded-qualifiers]:
> 
> static void cf_check _free_percpu_area(struct rcu_head *head)
> {
>     struct free_info *info = container_of(head, struct free_info, rcu);
>     unsigned int cpu = info->cpu;
>     char *p = __per_cpu_start + __per_cpu_offset[cpu];
> 
>     free_xenheap_pages(p, PERCPU_ORDER);
>     __per_cpu_offset[cpu] = INVALID_PERCPU_AREA;
> }

There's no use of __per_cpu_data_end here; I specifically didn't ask for both
of the declarations to have const added.

> I guess cast can help.

There may not be casts casting away constness, except maybe in very delicate
situations.

Jan

Re: [PATCH v2 08/15] xen/asm-generic: introduce generic header percpu.h

[Oleksii]

On Thu, 2023-11-16 at 13:28 +0100, Jan Beulich wrote:
> On 16.11.2023 13:04, Oleksii wrote:
> > On Thu, 2023-11-16 at 08:36 +0100, Jan Beulich wrote:
> > > On 10.11.2023 17:30, Oleksii Kurochko wrote:
> > > > --- /dev/null
> > > > +++ b/xen/include/asm-generic/percpu.h
> > > > @@ -0,0 +1,35 @@
> > > > +/* SPDX-License-Identifier: GPL-2.0-only */
> > > > +#ifndef __ASM_GENERIC_PERCPU_H__
> > > > +#define __ASM_GENERIC_PERCPU_H__
> > > > +
> > > > +#ifndef __ASSEMBLY__
> > > > +
> > > > +#include <xen/types.h>
> > > > +
> > > > +extern char __per_cpu_start[], __per_cpu_data_end[];
> > > 
> > > Can we go one tiny step beyond what Arm presently has and make
> > > the
> > > latter of the two const?
> > I am not sure I think we will have compilation issue with the
> > following
> > code in Arm and x86 because of [-Werror=discarded-qualifiers]:
> > 
> > static void cf_check _free_percpu_area(struct rcu_head *head)
> > {
> >     struct free_info *info = container_of(head, struct free_info,
> > rcu);
> >     unsigned int cpu = info->cpu;
> >     char *p = __per_cpu_start + __per_cpu_offset[cpu];
> > 
> >     free_xenheap_pages(p, PERCPU_ORDER);
> >     __per_cpu_offset[cpu] = INVALID_PERCPU_AREA;
> > }
> 
> There's no use of __per_cpu_data_end here; I specifically didn't ask
> for both
> of the declarations to have const added.
Yes, I misread your comment.

> 
> > I guess cast can help.
> 
> There may not be casts casting away constness, except maybe in very
> delicate
> situations.
> 

~ Oleksii

Re: [PATCH v2 09/15] xen/asm-generic: introduce generic header smp.h

[Oleksii]

I will drop this patch as it will be hard to make it generic for Arm,
PPC and RISC-V.

~ Oleksii

On Fri, 2023-11-10 at 18:30 +0200, Oleksii Kurochko wrote:
> <asm/smp.h> is expcted to be generic between Arm, PPC and RISC-V
> there by it is moved to asm-generic.
> 
> Right now it is common only by PPC and RISC-V but during work on
> support of the mentioned arhcs <asm/smp.h> is expected to be the
> same.
> 
> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
> ---
> Changes in V2:
>         - drop #ifded ASSEMBLY as this header isn't expected to be
> included in asm files.
>         - update the commit message.
> ---
>  xen/include/asm-generic/smp.h | 28 ++++++++++++++++++++++++++++
>  1 file changed, 28 insertions(+)
>  create mode 100644 xen/include/asm-generic/smp.h
> 
> diff --git a/xen/include/asm-generic/smp.h b/xen/include/asm-
> generic/smp.h
> new file mode 100644
> index 0000000000..6740a2064c
> --- /dev/null
> +++ b/xen/include/asm-generic/smp.h
> @@ -0,0 +1,28 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +#ifndef __ASM_GENERIC_SMP_H
> +#define __ASM_GENERIC_SMP_H
> +
> +#include <xen/cpumask.h>
> +#include <xen/percpu.h>
> +
> +DECLARE_PER_CPU(cpumask_var_t, cpu_sibling_mask);
> +DECLARE_PER_CPU(cpumask_var_t, cpu_core_mask);
> +
> +#define cpu_is_offline(cpu) unlikely(!cpu_online(cpu))
> +
> +/*
> + * Do we, for platform reasons, need to actually keep CPUs online
> when we
> + * would otherwise prefer them to be off?
> + */
> +#define park_offline_cpus false
> +
> +#endif
> +
> +/*
> + * Local variables:
> + * mode: C
> + * c-file-style: BSD
> + * c-basic-offset: 4
> + * indent-tabs-mode: nil
> + * End:
> + */