[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v3 15/16] ipmi: Add ACPI table entries
From: |
Igor Mammedov |
Subject: |
Re: [Qemu-devel] [PATCH v3 15/16] ipmi: Add ACPI table entries |
Date: |
Tue, 7 Jul 2015 11:47:44 +0200 |
On Mon, 8 Jun 2015 20:12:10 -0500
address@hidden wrote:
> From: Corey Minyard <address@hidden>
>
> Use the new ACPI table construction tools to create an ACPI
> entry for IPMI.
Above doesn't tell what is purpose of patch.
It would be nice for commit message to describe what the patch
actually does with references to relevant specs.
> Signed-off-by: Corey Minyard <address@hidden>
> Acked-by: Michael S. Tsirkin <address@hidden>
> ---
> hw/ipmi/Makefile.objs | 1 +
> hw/ipmi/ipmi_acpi.c | 122
> ++++++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 123 insertions(+)
> create mode 100644 hw/ipmi/ipmi_acpi.c
>
> diff --git a/hw/ipmi/Makefile.objs b/hw/ipmi/Makefile.objs
> index 81fb8e7..a5ba7d5 100644
> --- a/hw/ipmi/Makefile.objs
> +++ b/hw/ipmi/Makefile.objs
> @@ -4,3 +4,4 @@ common-obj-$(CONFIG_IPMI_LOCAL) += ipmi_bmc_extern.o
> common-obj-$(CONFIG_ISA_IPMI_KCS) += isa_ipmi_kcs.o
> common-obj-$(CONFIG_ISA_IPMI_BT) += isa_ipmi_bt.o
> common-obj-$(call land,$(CONFIG_IPMI),$(CONFIG_SMBIOS)) += ipmi_smbios.o
> +common-obj-$(call land,$(CONFIG_IPMI),$(CONFIG_ACPI)) += ipmi_acpi.o
if device is planned to be used only with PC machine I'd suggest to
put below functions into hw/i386/acpi-build.c
but if there is plans to make it work with ARM target as well
then a more generic hw/acpi/ipmi_acpi.c would be more suitable.
> diff --git a/hw/ipmi/ipmi_acpi.c b/hw/ipmi/ipmi_acpi.c
> new file mode 100644
> index 0000000..28ddbe4
> --- /dev/null
> +++ b/hw/ipmi/ipmi_acpi.c
> @@ -0,0 +1,122 @@
> +/*
> + * IPMI ACPI firmware handling
> + *
> + * Copyright (c) 2015 Corey Minyard, MontaVista Software, LLC
> + *
> + * Permission is hereby granted, free of charge, to any person obtaining a
> copy
> + * of this software and associated documentation files (the "Software"), to
> deal
> + * in the Software without restriction, including without limitation the
> rights
> + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
> + * copies of the Software, and to permit persons to whom the Software is
> + * furnished to do so, subject to the following conditions:
> + *
> + * The above copyright notice and this permission notice shall be included in
> + * all copies or substantial portions of the Software.
> + *
> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
> + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
> FROM,
> + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
> + * THE SOFTWARE.
> + */
> +
> +#include "hw/ipmi/ipmi.h"
> +#include "hw/acpi/aml-build.h"
> +#include "hw/acpi/acpi.h"
> +#include "hw/acpi/acpi-dev-tables.h"
> +#include <qemu/error-report.h>
> +
> +static Aml *aml_ipmi_crs(IPMIFwInfo *info)
> +{
> + Aml *crs = aml_resource_template();
> + uint8_t regspacing = info->register_spacing;
> +
> + if (regspacing == 1) {
> + regspacing = 0;
> + }
what's is purpose of ^^^ ???
Is it used but other code than ACPI, if not then just drop it and
hardcode it here naming it according to field it's used for (i.e. alignment)
> +
> + switch (info->memspace) {
> + case IPMI_MEMSPACE_IO:
> + aml_append(crs, aml_io(aml_decode16, info->base_address,
> + info->base_address + info->register_length - 1
> + regspacing, info->register_length));
> + break;
> + case IPMI_MEMSPACE_MEM32:
> + aml_append(crs,
> + aml_dword_memory(aml_pos_decode,
> + aml_min_fixed, aml_max_fixed,
> + aml_non_cacheable, aml_ReadWrite,
> + 0xffffffff,
> + info->base_address,
> + info->base_address + info->register_length - 1,
> + regspacing, info->register_length));
> + break;
> + case IPMI_MEMSPACE_MEM64:
> + aml_append(crs,
> + aml_qword_memory(aml_pos_decode,
> + aml_min_fixed, aml_max_fixed,
> + aml_non_cacheable, aml_ReadWrite,
> + 0xffffffffffffffffULL,
> + info->base_address,
> + info->base_address + info->register_length - 1,
> + regspacing, info->register_length));
> + break;
> + case IPMI_MEMSPACE_SMBUS:
> + aml_append(crs, aml_return(aml_int(info->base_address)));
> + break;
> + }
does device support reprogramming of "base_address" or is it fixed?
+ case for default with abort()
> +
> + if (info->interrupt_number) {
> + aml_append(crs, aml_irq_no_flags(info->interrupt_number));
> + }
> +
> + return crs;
> +}
> +
> +static void
> +ipmi_encode_one_acpi(IPMIFwInfo *info, void *opaque)
> +{
> + Aml *ssdt, *scope, *dev, *method;
> + char *name;
> + int version = ((info->ipmi_spec_major_revision << 8)
s/int/uint16_t/
> + | (info->ipmi_spec_minor_revision << 4));
> +
> + if (!info->acpi_parent) {
point of attaching should be defined here and not by device, hence I'd drop
this hunk
> + ipmi_debug("device %s not compatible with ACPI, no parent given.",
> + info->interface_name);
> + return;
> + }
> +
> + ssdt = init_aml_allocator();
drop this and just call this function directly from build_ssdt()
> +
> + scope = aml_scope("%s", info->acpi_parent);
> + name = g_strdup_printf("ipmi_%s", info->interface_name);
I don't see 'name' being reused neither freed,
just do below
..."_STR", aml_string("ipmi_%s", info->interface_name)...
> +
> + dev = aml_device("MI0");
> + aml_append(dev, aml_name_decl("_HID", aml_eisaid("IPI0001")));
> + aml_append(dev, aml_name_decl("_STR", aml_string("%s", name)));
> + aml_append(dev, aml_name_decl("_UID", aml_int(0)));
since it's the only device there is no need to provide UUID,
if there could be more than one BMCs than this part is wrong.
> + aml_append(dev, aml_name_decl("_CRS", aml_ipmi_crs(info)));
> + method = aml_method("_IFT", 0);
> + aml_append(method, aml_return(aml_int(info->interface_type)));
> + aml_append(dev, method);
> + method = aml_method("_SRV", 0);
> + aml_append(method, aml_return(aml_int(version)));
Is there a specific need to use methods for _IFT/SRV objects,
since they just return a value and don't do anything else, I'd suggest to
use named variables (i.e. aml_name_decl())
> + aml_append(dev, method);
> +
> + aml_append(scope, dev);
> +
> + aml_append(ssdt, scope);
> +
> + add_acpi_dev_table(ssdt->buf->data, ssdt->buf->len, "SSDT", 1);
> + free_aml_allocator();
> +}
> +
> +static void ipmi_acpi_init(void)
> +{
> + ipmi_register_fwinfo_handler(ipmi_encode_one_acpi, NULL);
> +}
> +
> +type_init(ipmi_acpi_init);
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- Re: [Qemu-devel] [PATCH v3 15/16] ipmi: Add ACPI table entries,
Igor Mammedov <=