netdev - Re: [PATCH v9 07/27] sfc: use cxl api for regs setup and checking

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

Message-ID: <02473860-3fee-00f5-1c86-1242c0bf42e8@amd.com>
Date: Mon, 20 Jan 2025 15:44:43 +0000
From: Alejandro Lucero Palau <alucerop@....com>
To: Dan Williams <dan.j.williams@...el.com>, alejandro.lucero-palau@....com,
 linux-cxl@...r.kernel.org, netdev@...r.kernel.org, edward.cree@....com,
 davem@...emloft.net, kuba@...nel.org, pabeni@...hat.com,
 edumazet@...gle.com, dave.jiang@...el.com
Subject: Re: [PATCH v9 07/27] sfc: use cxl api for regs setup and checking


On 1/18/25 01:53, Dan Williams wrote:
> alejandro.lucero-palau@ wrote:
>> From: Alejandro Lucero <alucerop@....com>
>>
>> Use cxl code for registers discovery and mapping.
>>
>> Validate capabilities found based on those registers against expected
>> capabilities.
>>
>> Signed-off-by: Alejandro Lucero <alucerop@....com>
>> Reviewed-by: Martin Habets <habetsm.xilinx@...il.com>
>> Reviewed-by: Zhi Wang <zhi@...dia.com>
>> Acked-by: Edward Cree <ecree.xilinx@...il.com>
>> ---
>>   drivers/net/ethernet/sfc/efx_cxl.c | 21 +++++++++++++++++++++
>>   1 file changed, 21 insertions(+)
>>
>> diff --git a/drivers/net/ethernet/sfc/efx_cxl.c b/drivers/net/ethernet/sfc/efx_cxl.c
>> index 12c9d50cbb26..29368d010adc 100644
>> --- a/drivers/net/ethernet/sfc/efx_cxl.c
>> +++ b/drivers/net/ethernet/sfc/efx_cxl.c
>> @@ -22,6 +22,8 @@ int efx_cxl_init(struct efx_probe_data *probe_data)
>>   {
>>   	struct efx_nic *efx = &probe_data->efx;
>>   	struct pci_dev *pci_dev = efx->pci_dev;
>> +	DECLARE_BITMAP(expected, CXL_MAX_CAPS);
>> +	DECLARE_BITMAP(found, CXL_MAX_CAPS);
>>   	struct efx_cxl *cxl;
>>   	struct resource res;
>>   	u16 dvsec;
>> @@ -64,6 +66,25 @@ int efx_cxl_init(struct efx_probe_data *probe_data)
>>   		goto err_resource_set;
>>   	}
>>   
>> +	rc = cxl_pci_accel_setup_regs(pci_dev, cxl->cxlds);
>> +	if (rc) {
>> +		pci_err(pci_dev, "CXL accel setup regs failed");
>> +		goto err_resource_set;
>> +	}
>> +
>> +	bitmap_clear(expected, 0, CXL_MAX_CAPS);
>> +	set_bit(CXL_DEV_CAP_HDM, expected);
>> +	set_bit(CXL_DEV_CAP_HDM, expected);
>> +	set_bit(CXL_DEV_CAP_RAS, expected);
>> +
>> +	if (!cxl_pci_check_caps(cxl->cxlds, expected, found)) {
>> +		pci_err(pci_dev,
>> +			"CXL device capabilities found(%pb) not as expected(%pb)",
>> +			found, expected);
>> +		rc = -EIO;
>> +		goto err_resource_set;
>> +	}
>> +
> Walk the existing valid bits in the reg maps. If you want to do this
> with bitmaps you can convert reg_map valid bits into a bitmap locally,
> but that redundant infrastructure can be left out of the core.


This is the accel driver. Some way for doing this needs to be 
implemented, and although I'm not against solving the duplication 
between the reg_map valid bit and the new capability bitmap, an accel 
driver will need this piece of code or something equivalent.