Hello,
On 13/08/17 22:30, mja...@caviumnetworks.com wrote:
From: Manish Jaggi <mja...@cavium.com>
Added gicv3_its_acpi_init to update host_its_list from MADT table.
For ACPI, host_its sturcture stores dt_node as NULL.
s/sturcture/structure/
Future TOD0:
Cleanup :(1) Remove from host_its dt_node as it is required only for ACPI
Enhancement :(2) Provide a method to access translation_id and
other fields of madt generic translator.
I don't get those TODOs. This is not related to this patch and does not
really hence the commit message.
Signed-off-by: Manish Jaggi <mja...@cavium.com>
---
xen/arch/arm/gic-v3-its.c | 14 ++++++++++++++
xen/arch/arm/gic-v3.c | 8 ++++++++
xen/include/asm-arm/gic_v3_its.h | 13 +++++++++++++
3 files changed, 35 insertions(+)
diff --git a/xen/arch/arm/gic-v3-its.c b/xen/arch/arm/gic-v3-its.c
index f844a0d..c4f1288 100644
--- a/xen/arch/arm/gic-v3-its.c
+++ b/xen/arch/arm/gic-v3-its.c
@@ -32,6 +32,7 @@
#include <asm/page.h>
#define ITS_CMD_QUEUE_SZ SZ_1M
+#define ACPI_GICV3_ITS_MEM_SIZE (SZ_64K)
The (...) are not necessary.
/*
* No lock here, as this list gets only populated upon boot while scanning
@@ -1020,6 +1021,19 @@ void gicv3_its_dt_init(const struct dt_device_node *node)
}
}
+#ifdef CONFIG_ACPI
+int gicv3_its_acpi_init(struct acpi_subtable_header *header,
+ const unsigned long end)
I don't much like the idea of providing a callback that will be called
by gic-v3.c. I would much prefer to follow the same pattern as the DT
where gic-v3.c will call a function in the gic-v3-its.c that will
iterate on all possible ITS.
This will make a more sane interface.
Also, it would make sense to call it gicv3_its_acpi_probe.
+{
+ struct acpi_madt_generic_translator *its;
+
+ its = (struct acpi_madt_generic_translator *)header;
You probably want to add check BAD_MADT_ENTRY(its, end).
+
+ return add_to_host_its_list(its->base_address,
+ ACPI_GICV3_ITS_MEM_SIZE, NULL);
If you follow my suggestion in patch #1 regarding the return of
add_to_host_its_list, then you would need to check true/false and return
a correct errno.
Even if you don't follow it, please return an appropriate errno rather
than -1.
+}
+#endif
+
/*
* Local variables:
* mode: C
diff --git a/xen/arch/arm/gic-v3.c b/xen/arch/arm/gic-v3.c
index f990eae..0be8942 100644
--- a/xen/arch/arm/gic-v3.c
+++ b/xen/arch/arm/gic-v3.c
@@ -1567,6 +1567,14 @@ static void __init gicv3_acpi_init(void)
gicv3.rdist_stride = 0;
+ /* Parse ITS information */
+ count = acpi_table_parse_madt(ACPI_MADT_TYPE_GENERIC_TRANSLATOR,
+ gicv3_its_acpi_init, 0);
See my comment above.
+
+ if ( count <= 0 )
Hardware without ITS support will return 0 and therefore panic. You
don't want this to happen.
+ panic("GICv3: Can't get ITS entry");
+
+
/*
* In ACPI, 0 is considered as the invalid address. However the rest
* of the initialization rely on the invalid address to be
diff --git a/xen/include/asm-arm/gic_v3_its.h b/xen/include/asm-arm/gic_v3_its.h
index 1fac1c7..2b7493d 100644
--- a/xen/include/asm-arm/gic_v3_its.h
+++ b/xen/include/asm-arm/gic_v3_its.h
@@ -105,6 +105,7 @@
#include <xen/device_tree.h>
#include <xen/rbtree.h>
+#include <xen/acpi.h>
With the suggestion suggested above, you will not need to include
<xen/acpi.h> in gic_v3_its.h.
#define HOST_ITS_FLUSH_CMD_QUEUE (1U << 0)
#define HOST_ITS_USES_PTA (1U << 1)
@@ -135,6 +136,10 @@ extern struct list_head host_its_list;
/* Parse the host DT and pick up all host ITSes. */
void gicv3_its_dt_init(const struct dt_device_node *node);
+#ifdef CONFIG_ACPI
+int gicv3_its_acpi_init(struct acpi_subtable_header *header,
+ const unsigned long end);
+#endif
Newline here please.
bool gicv3_its_host_has_its(void);
unsigned int vgic_v3_its_count(const struct domain *d);
@@ -196,6 +201,14 @@ static inline void gicv3_its_dt_init(const struct
dt_device_node *node)
{
}
+#ifdef CONFIG_ACPI
+static inline int gicv3_its_acpi_init(struct acpi_subtable_header *header,
+ const unsigned long end)
+{
+ return false;
gicv3_its_acpi_init return an int not a bool. Please modify this
accordingly.
+}
+#endif
+
static inline bool gicv3_its_host_has_its(void)
{
return false;
Cheers,
--
Julien Grall
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
