KVM Archive on lore.kernel.org
 help / color / Atom feed
From: Jonathan Cameron <Jonathan.Cameron@Huawei.com>
To: Dongjiu Geng <gengdongjiu@huawei.com>
Cc: <pbonzini@redhat.com>, <mst@redhat.com>, <imammedo@redhat.com>,
	<shannon.zhaosl@gmail.com>, <peter.maydell@linaro.org>,
	<fam@euphon.net>, <rth@twiddle.net>, <ehabkost@redhat.com>,
	<mtosatti@redhat.com>, <xuwei5@huawei.com>, <james.morse@arm.com>,
	<qemu-devel@nongnu.org>, <kvm@vger.kernel.org>,
	<qemu-arm@nongnu.org>, <zhengxiang9@huawei.com>,
	<linuxarm@huawei.com>
Subject: Re: [PATCH v22 4/9] ACPI: Build Hardware Error Source Table
Date: Wed, 5 Feb 2020 16:43:28 +0000
Message-ID: <20200205164328.00006f1e@Huawei.com> (raw)
In-Reply-To: <1578483143-14905-5-git-send-email-gengdongjiu@huawei.com>

On Wed, 8 Jan 2020 19:32:18 +0800
Dongjiu Geng <gengdongjiu@huawei.com> wrote:

> This patch builds Hardware Error Source Table(HEST) via fw_cfg blobs.
> Now it only supports ARMv8 SEA, a type of Generic Hardware Error
> Source version 2(GHESv2) error source. Afterwards, we can extend
> the supported types if needed. For the CPER section, currently it
> is memory section because kernel mainly wants userspace to handle
> the memory errors.
> 
> This patch follows the spec ACPI 6.2 to build the Hardware Error
> Source table. For more detailed information, please refer to
> document: docs/specs/acpi_hest_ghes.rst
> 
> build_append_ghes_notify() will help to add Hardware Error Notification
> to ACPI tables without using packed C structures and avoid endianness
> issues as API doesn't need explicit conversion.
> 
> Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com>
> Signed-off-by: Xiang Zheng <zhengxiang9@huawei.com>
> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
> Acked-by: Xiang Zheng <zhengxiang9@huawei.com>

Hi. 

I was forwards porting my old series adding CCIX error injection support
and came across a place this could 'possibly' be improved.

I say possibly because it's really about enabling more flexibility
in how this code is reused than actually 'fixing' anything here.

If you don't make the change here, I'll just add a precursor patch to my
series.  Just seems nice to tidy it up at source.

The rest of the parts of this series I am using seems to work great.

Thanks!

Jonathan

> ---
>  hw/acpi/ghes.c           | 118 ++++++++++++++++++++++++++++++++++++++++++++++-
>  hw/arm/virt-acpi-build.c |   2 +
>  include/hw/acpi/ghes.h   |  40 ++++++++++++++++
>  3 files changed, 159 insertions(+), 1 deletion(-)
> 
> diff --git a/hw/acpi/ghes.c b/hw/acpi/ghes.c
> index b7fdbbb..9d37798 100644
> --- a/hw/acpi/ghes.c
> +++ b/hw/acpi/ghes.c
> @@ -34,9 +34,42 @@
> +
...
> +/* Build Generic Hardware Error Source version 2 (GHESv2) */
> +static void build_ghes_v2(GArray *table_data, int source_id, BIOSLinker *linker)
This function takes source ID, which uses the enum of all sources registered.
However, it doesn't use it to locate the actual physical addresses.

Currently the code effectively assumes the value is 0.

> +{
> +    uint64_t address_offset;
> +    /*
> +     * Type:
> +     * Generic Hardware Error Source version 2(GHESv2 - Type 10)
> +     */
> +    build_append_int_noprefix(table_data, ACPI_GHES_SOURCE_GENERIC_ERROR_V2, 2);
> +    /* Source Id */
> +    build_append_int_noprefix(table_data, source_id, 2);
> +    /* Related Source Id */
> +    build_append_int_noprefix(table_data, 0xffff, 2);
> +    /* Flags */
> +    build_append_int_noprefix(table_data, 0, 1);
> +    /* Enabled */
> +    build_append_int_noprefix(table_data, 1, 1);
> +
> +    /* Number of Records To Pre-allocate */
> +    build_append_int_noprefix(table_data, 1, 4);
> +    /* Max Sections Per Record */
> +    build_append_int_noprefix(table_data, 1, 4);
> +    /* Max Raw Data Length */
> +    build_append_int_noprefix(table_data, ACPI_GHES_MAX_RAW_DATA_LENGTH, 4);
> +
> +    address_offset = table_data->len;
> +    /* Error Status Address */
> +    build_append_gas(table_data, AML_AS_SYSTEM_MEMORY, 0x40, 0,
> +                     4 /* QWord access */, 0);
> +    bios_linker_loader_add_pointer(linker, ACPI_BUILD_TABLE_FILE,
> +        address_offset + GAS_ADDR_OFFSET,
> +        sizeof(uint64_t), ACPI_GHES_ERRORS_FW_CFG_FILE, 0);

The offset here would need to be source_id * sizeof(uint64_t) I think

> +
> +    /*
> +     * Notification Structure
> +     * Now only enable ARMv8 SEA notification type
> +     */
> +    build_ghes_hw_error_notification(table_data, ACPI_GHES_NOTIFY_SEA);
Perhaps a switch for this to allow for other options later.

	switch (source_id) {
	case ACPI_HEST_SRC_ID_SEA:
		...
		break;
	default:
	//print some error message.

	}
> +
> +    /* Error Status Block Length */
> +    build_append_int_noprefix(table_data, ACPI_GHES_MAX_RAW_DATA_LENGTH, 4);
> +
> +    /*
> +     * Read Ack Register
> +     * ACPI 6.1: 18.3.2.8 Generic Hardware Error Source
> +     * version 2 (GHESv2 - Type 10)
> +     */
> +    address_offset = table_data->len;
> +    build_append_gas(table_data, AML_AS_SYSTEM_MEMORY, 0x40, 0,
> +                     4 /* QWord access */, 0);
> +    bios_linker_loader_add_pointer(linker, ACPI_BUILD_TABLE_FILE,
> +        address_offset + GAS_ADDR_OFFSET,
> +        sizeof(uint64_t), ACPI_GHES_ERRORS_FW_CFG_FILE,
> +        ACPI_GHES_ERROR_SOURCE_COUNT * sizeof(uint64_t));

Offset of (ACPI_GHES_ERROR_SOURCE_COUNT + source_id) * sizeof(uint64_t)

> +
> +    /*
> +     * Read Ack Preserve
> +     * We only provide the first bit in Read Ack Register to OSPM to write
> +     * while the other bits are preserved.
> +     */
> +    build_append_int_noprefix(table_data, ~0x1ULL, 8);
> +    /* Read Ack Write */
> +    build_append_int_noprefix(table_data, 0x1, 8);
> +}




  parent reply index

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-08 11:32 [PATCH v22 0/9] Add ARMv8 RAS virtualization support in QEMU Dongjiu Geng
2020-01-08 11:32 ` [PATCH v22 1/9] hw/arm/virt: Introduce a RAS machine option Dongjiu Geng
2020-01-08 11:32 ` [PATCH v22 2/9] docs: APEI GHES generation and CPER record description Dongjiu Geng
2020-01-15 15:11   ` Igor Mammedov
2020-01-08 11:32 ` [PATCH v22 3/9] ACPI: Build related register address fields via hardware error fw_cfg blob Dongjiu Geng
2020-01-23 15:14   ` Igor Mammedov
2020-02-02 14:01     ` gengdongjiu
2020-01-08 11:32 ` [PATCH v22 4/9] ACPI: Build Hardware Error Source Table Dongjiu Geng
2020-01-23 15:48   ` Igor Mammedov
2020-02-02 14:21     ` gengdongjiu
2020-02-05 16:43   ` Jonathan Cameron [this message]
2020-02-10 11:18     ` gengdongjiu
2020-01-08 11:32 ` [PATCH v22 5/9] ACPI: Record the Generic Error Status Block address Dongjiu Geng
2020-01-16 16:44   ` Peter Maydell
2020-01-17 10:36     ` gengdongjiu
2020-02-13 15:28     ` gengdongjiu
2020-01-17  7:39   ` Philippe Mathieu-Daudé
2020-01-17 10:47     ` gengdongjiu
2020-01-17 11:20       ` Philippe Mathieu-Daudé
2020-01-28 14:41   ` Igor Mammedov
2020-01-28 16:19     ` Igor Mammedov
2020-02-02 12:44     ` gengdongjiu
2020-02-03  7:51       ` Igor Mammedov
2020-01-08 11:32 ` [PATCH v22 6/9] KVM: Move hwpoison page related functions into kvm-all.c Dongjiu Geng
2020-01-08 11:32 ` [PATCH v22 7/9] ACPI: Record Generic Error Status Block(GESB) table Dongjiu Geng
2020-01-28 15:29   ` Igor Mammedov
2020-02-02 13:42     ` gengdongjiu
2020-02-03  7:55       ` Igor Mammedov
2020-01-08 11:32 ` [PATCH v22 8/9] target-arm: kvm64: handle SIGBUS signal from kernel or KVM Dongjiu Geng
2020-01-16 16:28   ` Peter Maydell
2020-01-16 16:40     ` Peter Maydell
2020-01-17 10:04     ` gengdongjiu
2020-01-20 12:15       ` Peter Maydell
2020-01-22 15:30         ` gengdongjiu
2020-01-08 11:32 ` [PATCH v22 9/9] MAINTAINERS: Add ACPI/HEST/GHES entries Dongjiu Geng
2020-01-16 16:46   ` Peter Maydell
2020-01-17  7:22     ` Philippe Mathieu-Daudé
2020-01-17 10:16       ` gengdongjiu
2020-01-17 11:09       ` Peter Maydell
2020-01-17 11:22         ` Philippe Mathieu-Daudé
2020-01-19  8:19           ` gengdongjiu
2020-01-17 12:19         ` Michael S. Tsirkin
2020-01-09  3:38 ` [PATCH v22 0/9] Add ARMv8 RAS virtualization support in QEMU gengdongjiu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200205164328.00006f1e@Huawei.com \
    --to=jonathan.cameron@huawei.com \
    --cc=ehabkost@redhat.com \
    --cc=fam@euphon.net \
    --cc=gengdongjiu@huawei.com \
    --cc=imammedo@redhat.com \
    --cc=james.morse@arm.com \
    --cc=kvm@vger.kernel.org \
    --cc=linuxarm@huawei.com \
    --cc=mst@redhat.com \
    --cc=mtosatti@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    --cc=shannon.zhaosl@gmail.com \
    --cc=xuwei5@huawei.com \
    --cc=zhengxiang9@huawei.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

KVM Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/kvm/0 kvm/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 kvm kvm/ https://lore.kernel.org/kvm \
		kvm@vger.kernel.org
	public-inbox-index kvm

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.kvm


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git