netdev - Re: [PATCH v10 06/26] 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 for Android: free password hash cracker in your pocket

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

Message-ID: <67a3d8cfaafd_2ee275294cb@iweiny-mobl.notmuch>
Date: Wed, 5 Feb 2025 15:31:59 -0600
From: Ira Weiny <ira.weiny@...el.com>
To: <alucerop@....com>, <linux-cxl@...r.kernel.org>, <netdev@...r.kernel.org>,
	<dan.j.williams@...el.com>, <edward.cree@....com>, <davem@...emloft.net>,
	<kuba@...nel.org>, <pabeni@...hat.com>, <edumazet@...gle.com>,
	<dave.jiang@...el.com>
CC: Alejandro Lucero <alucerop@....com>
Subject: Re: [PATCH v10 06/26] sfc: use cxl api for regs setup and checking

alucerop@ wrote:
> From: Alejandro Lucero <alucerop@....com>
> 

I'm still going through the series to better understand the over all arch
you need.  But I did find a couple of minor issues so I'll make those
comments straight off.

[snip]

> @@ -46,9 +50,37 @@ int efx_cxl_init(struct efx_probe_data *probe_data)
>  		return PTR_ERR(cxl->cxlmds);
>  	}
>  
> +	bitmap_clear(expected, 0, CXL_MAX_CAPS);
> +	set_bit(CXL_DEV_CAP_HDM, expected);
> +	set_bit(CXL_DEV_CAP_HDM, expected);

Why set HDM x2?

> +	set_bit(CXL_DEV_CAP_RAS, expected);
> +
> +	rc = cxl_pci_accel_setup_regs(pci_dev, cxl->cxlmds, found);
> +	if (rc) {
> +		pci_err(pci_dev, "CXL accel setup regs failed");
> +		goto err_regs;
> +	}
> +
> +	/*
> +	 * Checking mandatory caps are there as, at least, a subset of those
> +	 * found.
> +	 */
> +	if (!bitmap_subset(expected, found, CXL_MAX_CAPS)) {
> +		pci_err(pci_dev,
> +			"CXL device capabilities found(%pb) not as expected(%pb)",
> +			found, expected);
> +		rc = -EIO;
> +		goto err_regs;
> +	}
> +
>  	probe_data->cxl = cxl;
>  
>  	return 0;
> +
> +err_regs:
> +	kfree(probe_data->cxl);

Is this freeing what you want here?  AFAICS probe_data->cxl is not set
until after the checks work.

I think this is best handled by using __free() on cxl and no_free_ptr()
when setting probe_data?

Ira

> +	return rc;
> +
>  }
>  
>  void efx_cxl_exit(struct efx_probe_data *probe_data)
> -- 
> 2.17.1
> 
>