[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH-for-10.1 v3 7/9] hw/arm/virt-acpi: Always build IORT table (e
From: |
Eric Auger |
Subject: |
Re: [PATCH-for-10.1 v3 7/9] hw/arm/virt-acpi: Always build IORT table (even with GIC ITS disabled) |
Date: |
Mon, 7 Apr 2025 15:30:07 +0200 |
User-agent: |
Mozilla Thunderbird |
Hi Philippe,
On 4/3/25 10:40 PM, Philippe Mathieu-Daudé wrote:
> Citing Gustavo [*]:
>
> Gating IORT table generation entirely based on the presence
> of ITS looks wrong because IORT table has data beyond GIC ITS,
> like for SMMUv3 etc..
> [*]
> bae6e29a-7290-47d2-8caf-14702ee091f0@linaro.org/">https://lore.kernel.org/qemu-devel/bae6e29a-7290-47d2-8caf-14702ee091f0@linaro.org/
>
> Reported-by: Gustavo Romero <gustavo.romero@linaro.org>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> ---
> hw/arm/virt-acpi-build.c | 38 +++++++++++++++++++++-----------------
> 1 file changed, 21 insertions(+), 17 deletions(-)
>
> diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c
> index 1c389ef5cdb..e7e27951cb9 100644
> --- a/hw/arm/virt-acpi-build.c
> +++ b/hw/arm/virt-acpi-build.c
> @@ -287,6 +287,7 @@ build_iort(GArray *table_data, BIOSLinker *linker,
> VirtMachineState *vms)
> uint32_t id = 0;
> GArray *smmu_idmaps = g_array_new(false, true,
> sizeof(AcpiIortIdMapping));
> GArray *its_idmaps = g_array_new(false, true, sizeof(AcpiIortIdMapping));
> + bool has_its = its_enabled(vms);
>
> AcpiTable table = { .sig = "IORT", .rev = 3, .oem_id = vms->oem_id,
> .oem_table_id = vms->oem_table_id };
> @@ -325,9 +326,12 @@ build_iort(GArray *table_data, BIOSLinker *linker,
> VirtMachineState *vms)
>
> nb_nodes = 3; /* RC, ITS, SMMUv3 */
> rc_mapping_count = smmu_idmaps->len + its_idmaps->len;
> - } else {
> + } else if (has_its) {
> nb_nodes = 2; /* RC, ITS */
> rc_mapping_count = 1;
> + } else {
> + nb_nodes = 1; /* RC */
> + rc_mapping_count = 1;
> }
> /* Number of IORT Nodes */
> build_append_int_noprefix(table_data, nb_nodes, 4);
> @@ -336,17 +340,19 @@ build_iort(GArray *table_data, BIOSLinker *linker,
> VirtMachineState *vms)
> build_append_int_noprefix(table_data, IORT_NODE_OFFSET, 4);
> build_append_int_noprefix(table_data, 0, 4); /* Reserved */
>
> - /* Table 12 ITS Group Format */
> - build_append_int_noprefix(table_data, 0 /* ITS Group */, 1); /* Type */
> - node_size = 20 /* fixed header size */ + 4 /* 1 GIC ITS Identifier */;
> - build_append_int_noprefix(table_data, node_size, 2); /* Length */
> - build_append_int_noprefix(table_data, 1, 1); /* Revision */
> - build_append_int_noprefix(table_data, id++, 4); /* Identifier */
> - build_append_int_noprefix(table_data, 0, 4); /* Number of ID mappings */
> - build_append_int_noprefix(table_data, 0, 4); /* Reference to ID Array */
> - build_append_int_noprefix(table_data, 1, 4); /* Number of ITSs */
> - /* GIC ITS Identifier Array */
> - build_append_int_noprefix(table_data, 0 /* MADT translation_id */, 4);
> + if (has_its) {
> + /* Table 12 ITS Group Format */
> + build_append_int_noprefix(table_data, 0 /* ITS Group */, 1); /* Type
> */ //
> + node_size = 20 /* fixed header size */ + 4 /* 1 GIC ITS Identifier
> */;
> + build_append_int_noprefix(table_data, node_size, 2); /* Length */
> + build_append_int_noprefix(table_data, 1, 1); /* Revision */
> + build_append_int_noprefix(table_data, id++, 4); /* Identifier */
> + build_append_int_noprefix(table_data, 0, 4); /* Number of ID
> mappings */
> + build_append_int_noprefix(table_data, 0, 4); /* Reference to ID
> Array */
> + build_append_int_noprefix(table_data, 1, 4); /* Number of ITSs */
> + /* GIC ITS Identifier Array */
> + build_append_int_noprefix(table_data, 0 /* MADT translation_id */,
> 4);
> + }
Do we allow configs where we have SMMU and no ITS?
The IORT spec says "
ID mappings that are defined in an SMMU IORT node can only have an ITS
Group node as an output reference,
or no IDs in the case of a system that does not have ITS units. All
other object types are explicitly forbidden.
"
So in case we do not have any ITS we also need to take care of ID
mappings of SMMUs
Eric
>
> if (vms->iommu == VIRT_IOMMU_SMMUV3) {
> int irq = vms->irqmap[VIRT_SMMU] + ARM_SPI_BASE;
> @@ -429,7 +435,7 @@ build_iort(GArray *table_data, BIOSLinker *linker,
> VirtMachineState *vms)
> build_iort_id_mapping(table_data, range->input_base,
> range->id_count, IORT_NODE_OFFSET);
> }
> - } else {
> + } else if (has_its) {
> /* output IORT node is the ITS group node (the first node) */
> build_iort_id_mapping(table_data, 0, 0x10000, IORT_NODE_OFFSET);
> }
> @@ -979,10 +985,8 @@ void virt_acpi_build(VirtMachineState *vms,
> AcpiBuildTables *tables)
> vms->oem_table_id);
> }
>
> - if (its_enabled(vms)) {
> - acpi_add_table(table_offsets, tables_blob);
> - build_iort(tables_blob, tables->linker, vms);
> - }
> + acpi_add_table(table_offsets, tables_blob);
> + build_iort(tables_blob, tables->linker, vms);
>
> #ifdef CONFIG_TPM
> if (tpm_get_version(tpm_find()) == TPM_VERSION_2_0) {
- [PATCH-for-10.1 v3 6/9] qtest/bios-tables-test: Whitelist aarch64/virt 'its_off' variant blobs, (continued)
- [PATCH-for-10.1 v3 6/9] qtest/bios-tables-test: Whitelist aarch64/virt 'its_off' variant blobs, Philippe Mathieu-Daudé, 2025/04/03
- Re: [PATCH-for-10.1 v3 6/9] qtest/bios-tables-test: Whitelist aarch64/virt 'its_off' variant blobs, Gustavo Romero, 2025/04/03
- Re: [PATCH-for-10.1 v3 6/9] qtest/bios-tables-test: Whitelist aarch64/virt 'its_off' variant blobs, Igor Mammedov, 2025/04/09
- Re: [PATCH-for-10.1 v3 6/9] qtest/bios-tables-test: Whitelist aarch64/virt 'its_off' variant blobs, Gustavo Romero, 2025/04/09
- Re: [PATCH-for-10.1 v3 6/9] qtest/bios-tables-test: Whitelist aarch64/virt 'its_off' variant blobs, Igor Mammedov, 2025/04/10
- Re: [PATCH-for-10.1 v3 6/9] qtest/bios-tables-test: Whitelist aarch64/virt 'its_off' variant blobs, Gustavo Romero, 2025/04/10
- Re: [PATCH-for-10.1 v3 6/9] qtest/bios-tables-test: Whitelist aarch64/virt 'its_off' variant blobs, Gustavo Romero, 2025/04/17
- Re: [PATCH-for-10.1 v3 6/9] qtest/bios-tables-test: Whitelist aarch64/virt 'its_off' variant blobs, Michael S. Tsirkin, 2025/04/17
[PATCH-for-10.1 v3 7/9] hw/arm/virt-acpi: Always build IORT table (even with GIC ITS disabled), Philippe Mathieu-Daudé, 2025/04/03
[PATCH-for-10.1 v3 8/9] hw/arm/virt-acpi: Do not advertise disabled GIC ITS, Philippe Mathieu-Daudé, 2025/04/03
[PATCH-for-10.1 v3 9/9] qtest/bios-tables-test: Update aarch64/virt 'its_off' variant blobs, Philippe Mathieu-Daudé, 2025/04/03
Re: [PATCH-for-10.1 v3 0/9] hw/arm: GIC ITS=off ACPI tables fixes, Philippe Mathieu-Daudé, 2025/04/15