xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Julien Grall <julien.grall@arm.com>
To: Shannon Zhao <zhaoshenglong@huawei.com>, xen-devel@lists.xen.org
Cc: stefano.stabellini@citrix.com, shannon.zhao@linaro.org,
	peter.huangpeng@huawei.com
Subject: Re: [PATCH v6 01/22] arm/acpi: Estimate memory required for acpi/efi tables
Date: Fri, 18 Mar 2016 18:44:57 +0000	[thread overview]
Message-ID: <56EC4CA9.4090309@arm.com> (raw)
In-Reply-To: <1458207668-12012-2-git-send-email-zhaoshenglong@huawei.com>

Hi Shannon,

On 17/03/16 09:40, Shannon Zhao wrote:
> diff --git a/xen/arch/arm/domain_build.c b/xen/arch/arm/domain_build.c
> index 83676e4..4e20499 100644
> --- a/xen/arch/arm/domain_build.c
> +++ b/xen/arch/arm/domain_build.c
> @@ -12,6 +12,8 @@
>   #include <xen/libfdt/libfdt.h>
>   #include <xen/guest_access.h>
>   #include <xen/iocap.h>
> +#include <xen/acpi.h>
> +#include <acpi/actables.h>
>   #include <asm/device.h>
>   #include <asm/setup.h>
>   #include <asm/platform.h>
> @@ -1354,6 +1356,79 @@ static int prepare_dtb(struct domain *d, struct kernel_info *kinfo)
>       return -EINVAL;
>   }
>
> +#ifdef CONFIG_ACPI
> +static int estimate_acpi_efi_size(struct domain *d, struct kernel_info *kinfo)
> +{
> +    size_t efi_size, acpi_size, madt_size;
> +    u64 addr;
> +    struct acpi_table_rsdp *rsdp_tbl;
> +    struct acpi_table_header *table;
> +
> +    efi_size = estimate_efi_size(kinfo->mem.nr_banks);
> +
> +    acpi_size = ROUNDUP(sizeof(struct acpi_table_fadt), 8);
> +    acpi_size += ROUNDUP(sizeof(struct acpi_table_stao), 8);
> +
> +    madt_size = sizeof(struct acpi_table_madt)
> +                + sizeof(struct acpi_madt_generic_interrupt) * d->max_vcpus
> +                + sizeof(struct acpi_madt_generic_distributor);
> +    if ( d->arch.vgic.version == GIC_V3 )
> +        madt_size += sizeof(struct acpi_madt_generic_redistributor)
> +                     * d->arch.vgic.nr_regions;
> +    acpi_size += ROUNDUP(madt_size, 8);
> +
> +    addr = acpi_os_get_root_pointer();
> +    if ( !addr )
> +    {
> +        printk("Unable to get acpi root pointer\n");
> +        return -EINVAL;
> +    }
> +    rsdp_tbl = acpi_os_map_memory(addr, sizeof(struct acpi_table_rsdp));

acpi_os_map_memory can fail. If you are sure that in this specific case 
it will never, add a BUG_ON(rsdp_tbl != NULL). Otherwise, please add 
proper check.

> +    table = acpi_os_map_memory(rsdp_tbl->xsdt_physical_address,
> +                               sizeof(struct acpi_table_header));

Ditto.

> +    /* Add place for STAO table in XSDT table */
> +    acpi_size += ROUNDUP(table->length + sizeof(u64), 8);
> +    acpi_os_unmap_memory(table, sizeof(struct acpi_table_header));
> +    acpi_os_unmap_memory(rsdp_tbl, sizeof(struct acpi_table_rsdp));
> +
> +    acpi_size += ROUNDUP(sizeof(struct acpi_table_rsdp), 8);
> +    d->arch.efi_acpi_len = ROUNDUP(efi_size, 8) + ROUNDUP(acpi_size, 8);
> +
> +    return 0;
> +}
> +
> +static int prepare_acpi(struct domain *d, struct kernel_info *kinfo)
> +{
> +    int rc = 0;
> +    int order;
> +
> +    rc = estimate_acpi_efi_size(d, kinfo);
> +    if ( rc != 0 )
> +        return rc;
> +
> +    order = get_order_from_bytes(d->arch.efi_acpi_len);
> +    d->arch.efi_acpi_table = alloc_xenheap_pages(order, 0);
> +    if ( d->arch.efi_acpi_table == NULL )
> +    {
> +        printk("unable to allocate memory!\n");
> +        return -ENOMEM;
> +    }
> +    memset(d->arch.efi_acpi_table, 0, d->arch.efi_acpi_len);
> +
> +    /* For ACPI, Dom0 doesn't use kinfo->gnttab_start to get the grant table
> +     * region. So we use it as the ACPI table mapped address. */

The coding style for multi-lines comment is:

/*
  * For ACPI ...
  * ....
  */

> +    d->arch.efi_acpi_gpa = kinfo->gnttab_start;
> +
> +    return 0;
> +}
> +#else
> +static int prepare_acpi(struct domain *d, struct kernel_info *kinfo)
> +{
> +    /* Only booting with ACPI will hit here */
> +    BUG_ON(1);

This is equivalent of BUG()

> +    return -EINVAL;
> +}
> +#endif
>   static void dtb_load(struct kernel_info *kinfo)
>   {
>       void * __user dtb_virt = (void * __user)(register_t)kinfo->dtb_paddr;
> @@ -1540,7 +1615,11 @@ int construct_dom0(struct domain *d)
>       allocate_memory(d, &kinfo);
>       find_gnttab_region(d, &kinfo);
>
> -    rc = prepare_dtb(d, &kinfo);
> +    if ( acpi_disabled )
> +        rc = prepare_dtb(d, &kinfo);
> +    else
> +        rc = prepare_acpi(d, &kinfo);
> +
>       if ( rc < 0 )
>           return rc;
>

[...]

> diff --git a/xen/arch/arm/efi/efi-dom0.c b/xen/arch/arm/efi/efi-dom0.c
> new file mode 100644
> index 0000000..90a7699
> --- /dev/null
> +++ b/xen/arch/arm/efi/efi-dom0.c
> @@ -0,0 +1,50 @@
> +/*
> + *  efi-dom0.c - Domain0 EFI Boot Support
> + *
> + *  Copyright (C) 2016 Shannon Zhao <shannon.zhao@linaro.org>
> + *
> + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> + *
> + *  This program is free software; you can redistribute it and/or modify
> + *  it under the terms of the GNU General Public License as published by
> + *  the Free Software Foundation; either version 2 of the License, or
> + *  (at your option) any later version.
> + *
> + *  This program is distributed in the hope that it will be useful,
> + *  but WITHOUT ANY WARRANTY; without even the implied warranty of
> + *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + *  GNU General Public License for more details.
> + *
> + *  You should have received a copy of the GNU General Public License
> + *  along with this program; If not, see <http://www.gnu.org/licenses/>.
> + *
> + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> + */
> +
> +#include "efi.h"
> +#include "efi-dom0.h"
> +#include <asm/setup.h>
> +#include <asm/acpi.h>
> +
> +struct meminfo __initdata acpi_mem;

I'm worry about sharing data between the EFI stub and Xen. On ARM, the 
EFI stub is completely independent. Once it has finished to execute, it 
will call xen as if it runs without UEFI.

This means that BSS will be zeroed. I would have expect to see acpi_mem 
residing in BSS. So I'm wondering how this can work.

> +/* Constant to indicate "Xen" in unicode u16 format */
> +static const CHAR16 xen_efi_fw_vendor[] = {0x0058, 0x0065, 0x006E, 0x0000};
> +
> +size_t __init estimate_efi_size(int mem_nr_banks)
> +{
> +    size_t size;
> +    size_t est_size = sizeof(EFI_SYSTEM_TABLE);
> +    size_t ect_size = sizeof(EFI_CONFIGURATION_TABLE);
> +    size_t emd_size = sizeof(EFI_MEMORY_DESCRIPTOR);
> +    size_t fw_vendor_size = sizeof(xen_efi_fw_vendor);
> +    int acpi_mem_nr_banks = 0;
> +
> +    if ( !acpi_disabled )
> +        acpi_mem_nr_banks = acpi_mem.nr_banks;
> +
> +    size = ROUNDUP(est_size + ect_size + fw_vendor_size, 8);
> +    /* plus 1 for new created tables */
> +    size += ROUNDUP(emd_size * (mem_nr_banks + acpi_mem_nr_banks + 1), 8);
> +
> +    return size;
> +}

Missing the classic emacs comment:

/*
  * Local variables:
  * mode: C
  * c-file-style: "BSD"
  * c-basic-offset: 4
  * indent-tabs-mode: nil
  * End:
  */

> diff --git a/xen/arch/arm/efi/efi-dom0.h b/xen/arch/arm/efi/efi-dom0.h
> new file mode 100644
> index 0000000..a509cdd
> --- /dev/null
> +++ b/xen/arch/arm/efi/efi-dom0.h

Why does this header live in arch/arm/efi instead of include/asm-arm?

> @@ -0,0 +1,3 @@

The header should be properly guarded.

> +#include <asm/setup.h>
> +
> +extern struct meminfo acpi_mem;
> diff --git a/xen/include/asm-arm/setup.h b/xen/include/asm-arm/setup.h
> index 30ac53b..7f233a1 100644
> --- a/xen/include/asm-arm/setup.h
> +++ b/xen/include/asm-arm/setup.h
> @@ -51,6 +51,8 @@ void arch_init_memory(void);
>
>   void copy_from_paddr(void *dst, paddr_t paddr, unsigned long len);
>
> +size_t estimate_efi_size(int mem_nr_banks);
> +
>   int construct_dom0(struct domain *d);
>
>   void discard_initial_modules(void);
>

Regards,

-- 
Julien Grall

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

  reply	other threads:[~2016-03-18 18:44 UTC|newest]

Thread overview: 79+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-03-17  9:40 [PATCH v6 00/22] Prepare UEFI and ACPI tables for Dom0 on ARM64 Shannon Zhao
2016-03-17  9:40 ` [PATCH v6 01/22] arm/acpi: Estimate memory required for acpi/efi tables Shannon Zhao
2016-03-18 18:44   ` Julien Grall [this message]
2016-03-21 13:01     ` Jan Beulich
2016-03-22 13:01     ` Shannon Zhao
2016-03-22  0:28   ` Julien Grall
2016-03-24 12:36   ` Stefano Stabellini
2016-03-17  9:40 ` [PATCH v6 02/22] arm/acpi: Add a helper function to get the acpi table offset Shannon Zhao
2016-03-21 12:16   ` Julien Grall
2016-03-24 12:32   ` Stefano Stabellini
2016-03-17  9:40 ` [PATCH v6 03/22] arm/acpi: Prepare FADT table for Dom0 Shannon Zhao
2016-03-21 12:25   ` Julien Grall
2016-03-17  9:40 ` [PATCH v6 04/22] arm/gic: Add a new callback for creating MADT " Shannon Zhao
2016-03-21 15:18   ` Julien Grall
2016-03-17  9:40 ` [PATCH v6 05/22] arm/acpi: Prepare " Shannon Zhao
2016-03-21 15:26   ` Julien Grall
2016-03-21 15:51     ` Jan Beulich
2016-03-17  9:40 ` [PATCH v6 06/22] arm/acpi: Prepare STAO " Shannon Zhao
2016-03-21 15:31   ` Julien Grall
2016-03-17  9:40 ` [PATCH v6 07/22] arm/acpi: Prepare XSDT " Shannon Zhao
2016-03-21 15:43   ` Julien Grall
2016-03-17  9:40 ` [PATCH v6 08/22] arm/acpi: Prepare RSDP " Shannon Zhao
2016-03-21 15:47   ` Julien Grall
2016-03-17  9:40 ` [PATCH v6 09/22] arm/p2m: Add helper functions to map memory regions Shannon Zhao
2016-03-21 15:52   ` Julien Grall
2016-03-22 13:05     ` Shannon Zhao
2016-03-22 15:59       ` Julien Grall
2016-03-17  9:40 ` [PATCH v6 10/22] arm/acpi: Map all other tables for Dom0 Shannon Zhao
2016-03-21 15:57   ` Julien Grall
2016-03-17  9:40 ` [PATCH v6 11/22] arm/acpi: Prepare EFI system table " Shannon Zhao
2016-03-21 16:10   ` Julien Grall
2016-03-22 13:06     ` Shannon Zhao
2016-03-22 13:24       ` Jan Beulich
2016-03-17  9:40 ` [PATCH v6 12/22] arm/acpi: Prepare EFI memory descriptor " Shannon Zhao
2016-03-21 16:51   ` Julien Grall
2016-03-22 13:16     ` Shannon Zhao
2016-03-22 16:04       ` Julien Grall
2016-03-24 15:06         ` Shannon Zhao
2016-03-24 15:23           ` Julien Grall
2016-03-24 12:39   ` Stefano Stabellini
2016-03-17  9:40 ` [PATCH v6 13/22] arm/acpi: Map the new created EFI and ACPI tables to Dom0 Shannon Zhao
2016-03-22  0:42   ` Julien Grall
2016-03-22 13:18     ` Shannon Zhao
2016-03-22 16:16       ` Julien Grall
2016-03-24 14:59         ` Shannon Zhao
2016-03-24 15:35           ` Julien Grall
2016-03-17  9:41 ` [PATCH v6 14/22] arm/acpi: Create min DT stub for Dom0 Shannon Zhao
2016-03-22  1:07   ` Julien Grall
2016-03-22 13:23     ` Shannon Zhao
2016-03-17  9:41 ` [PATCH v6 15/22] arm/acpi: Permit access all Xen unused SPIs " Shannon Zhao
2016-03-22 18:18   ` Julien Grall
2016-03-24 15:01     ` Shannon Zhao
2016-03-24 15:37       ` Julien Grall
2016-03-25  7:13         ` Shannon Zhao
2016-03-17  9:41 ` [PATCH v6 16/22] arm/acpi: Configure SPI interrupt type and route to Dom0 dynamically Shannon Zhao
2016-03-22 19:17   ` Julien Grall
2016-03-24 12:41   ` Stefano Stabellini
2016-03-17  9:41 ` [PATCH v6 17/22] arm/gic: Add a new callback to deny Dom0 access to GIC regions Shannon Zhao
2016-03-22 19:33   ` Julien Grall
2016-03-24 12:45     ` Stefano Stabellini
2016-03-24 15:03       ` Shannon Zhao
2016-03-24 15:39         ` Julien Grall
2016-03-17  9:41 ` [PATCH v6 18/22] arm/acpi: Permit MMIO access of Xen unused devices for Dom0 Shannon Zhao
2016-03-23 11:50   ` Julien Grall
2016-03-17  9:41 ` [PATCH v6 19/22] hvm/params: Add a new delivery type for event-channel in HVM_PARAM_CALLBACK_IRQ Shannon Zhao
2016-03-17 10:42   ` Jan Beulich
2016-03-17 11:04     ` Shannon Zhao
2016-03-17 11:29       ` Jan Beulich
2016-03-17 11:56         ` Shannon Zhao
2016-03-17 12:42           ` Jan Beulich
2016-03-17  9:41 ` [PATCH v6 20/22] xen/acpi: Fix event-channel interrupt when booting with ACPI Shannon Zhao
2016-03-23 12:15   ` Julien Grall
2016-03-17  9:41 ` [PATCH v6 21/22] xen/arm: Add a hypercall for device mmio mapping Shannon Zhao
2016-03-17  9:41 ` [PATCH v6 22/22] xen/arm64: Add ACPI support Shannon Zhao
2016-03-17 10:52   ` Jan Beulich
2016-03-17 11:03     ` Shannon Zhao
2016-03-17 11:31       ` Jan Beulich
2016-03-17 13:10         ` Shannon Zhao
2016-03-17 13:24           ` Jan Beulich

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=56EC4CA9.4090309@arm.com \
    --to=julien.grall@arm.com \
    --cc=peter.huangpeng@huawei.com \
    --cc=shannon.zhao@linaro.org \
    --cc=stefano.stabellini@citrix.com \
    --cc=xen-devel@lists.xen.org \
    --cc=zhaoshenglong@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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).