linux-kernel - Re: [PATCH v12 2/2] hisi_acc_vfio_pci: adapt to new migration configuration

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <7d1bef79-2cd0-9784-6dd9-d7a995ec3238@huawei.com>
Date: Thu, 6 Nov 2025 10:26:24 +0800
From: liulongfang <liulongfang@...wei.com>
To: Alex Williamson <alex@...zbot.org>
CC: <alex.williamson@...hat.com>, <jgg@...dia.com>,
	<herbert@...dor.apana.org.au>, <shameerkolothum@...il.com>,
	<jonathan.cameron@...wei.com>, <linux-crypto@...r.kernel.org>,
	<kvm@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v12 2/2] hisi_acc_vfio_pci: adapt to new migration
 configuration

On 2025/11/6 6:32, Alex Williamson wrote:
> On Thu, 30 Oct 2025 09:57:44 +0800
> Longfang Liu <liulongfang@...wei.com> wrote:
>> diff --git a/drivers/vfio/pci/hisilicon/hisi_acc_vfio_pci.h b/drivers/vfio/pci/hisilicon/hisi_acc_vfio_pci.h
>> index 91002ceeebc1..419a378c3d1d 100644
>> --- a/drivers/vfio/pci/hisilicon/hisi_acc_vfio_pci.h
>> +++ b/drivers/vfio/pci/hisilicon/hisi_acc_vfio_pci.h
>> @@ -50,8 +50,10 @@
>>  #define QM_QUE_ISO_CFG_V	0x0030
>>  #define QM_PAGE_SIZE		0x0034
>>  
>> -#define QM_EQC_DW0		0X8000
>> -#define QM_AEQC_DW0		0X8020
>> +#define QM_EQC_VF_DW0		0X8000
>> +#define QM_AEQC_VF_DW0		0X8020
>> +#define QM_EQC_PF_DW0		0x1c00
>> +#define QM_AEQC_PF_DW0		0x1c20
>>  
>>  #define ACC_DRV_MAJOR_VER 1
>>  #define ACC_DRV_MINOR_VER 0
>> @@ -59,6 +61,22 @@
>>  #define ACC_DEV_MAGIC_V1	0XCDCDCDCDFEEDAACC
>>  #define ACC_DEV_MAGIC_V2	0xAACCFEEDDECADEDE
>>  
>> +#define QM_MIG_REGION_OFFSET		0x180000
>> +#define QM_MIG_REGION_SIZE		0x2000
>> +
>> +/**
>> + * On HW_ACC_MIG_VF_CTRL mode, the configuration domain supporting live
>> + * migration functionality is located in the latter 32KB of the VF's BAR2.
>> + * The Guest is only provided with the first 32KB of the VF's BAR2.
>> + * On HW_ACC_MIG_PF_CTRL mode, the configuration domain supporting live
>> + * migration functionality is located in the PF's BAR2, and the entire 64KB
>> + * of the VF's BAR2 is allocated to the Guest.
>> + */
>> +enum hw_drv_mode {
>> +	HW_ACC_MIG_VF_CTRL = 0,
>> +	HW_ACC_MIG_PF_CTRL,
>> +};
>> +
>>  struct acc_vf_data {
>>  #define QM_MATCH_SIZE offsetofend(struct acc_vf_data, qm_rsv_state)
>>  	/* QM match information */
>> @@ -125,6 +143,7 @@ struct hisi_acc_vf_core_device {
>>  	struct pci_dev *vf_dev;
>>  	struct hisi_qm *pf_qm;
>>  	struct hisi_qm vf_qm;
>> +	int drv_mode;
> 
> I can fix this on commit rather than send a new version, but is there
> any reason we wouldn't make use of the enum here:
> 
> 	enum hw_drv_mode drv_mode;
> 
> ?  Thanks,
>

I didn't use the enum type here due to my habit of declaring enum variables with
integer types.
If needed, please directly modify it to use the enum type definition.

Thanks.
Longfang.

> Alex
> 
>>  	/*
>>  	 * vf_qm_state represents the QM_VF_STATE register value.
>>  	 * It is set by Guest driver for the ACC VF dev indicating
> 
> .
>