linux-kernel - Re: [PATCH v2 14/24] EDAC, ghes: Rework memory hierarchy detection

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]

Date:   Tue, 20 Aug 2019 10:56:47 +0200
From:   Borislav Petkov <bp@...en8.de>
To:     Robert Richter <rrichter@...vell.com>
Cc:     James Morse <james.morse@....com>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        "linux-edac@...r.kernel.org" <linux-edac@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2 14/24] EDAC, ghes: Rework memory hierarchy detection

On Mon, Jun 24, 2019 at 03:09:24PM +0000, Robert Richter wrote:
> In a later patch we want to add more information about the memory
> hierarchy (NUMA topology, DIMM label information). Rework memory
> hierarchy detection to make the code extendable for this.
> 
> The general approach is roughly like:
> 
> 	mem_info_setup();
> 	for_each_node(nid) {
> 		mci = edac_mc_alloc(nid);
> 		mem_info_prepare_mci(mci);
> 		edac_mc_add_mc(mci);
> 	};
> 
> This patch introduces mem_info_setup() and mem_info_prepare_mci().

Avoid having "This patch" or "This commit" in the commit message. It is
tautologically useless.

> All data of the memory hierarchy is collected in a local struct
> ghes_mem_info.
> 
> Note: Per (NUMA) node registration will be implemented in a later
> patch.

That sentence is not needed in the commit message.

> Signed-off-by: Robert Richter <rrichter@...vell.com>
> ---
>  drivers/edac/ghes_edac.c | 166 ++++++++++++++++++++++++++++++---------
>  1 file changed, 127 insertions(+), 39 deletions(-)
> 
> diff --git a/drivers/edac/ghes_edac.c b/drivers/edac/ghes_edac.c
> index 8063996a311d..44bfb499b147 100644
> --- a/drivers/edac/ghes_edac.c
> +++ b/drivers/edac/ghes_edac.c
> @@ -65,17 +65,53 @@ struct memdev_dmi_entry {
>  	u16 conf_mem_clk_speed;
>  } __attribute__((__packed__));
>  
> -struct ghes_edac_dimm_fill {
> -	struct mem_ctl_info *mci;
> -	unsigned count;

All those "dimm" and "info" words everywhere are making my head spin.
Let's make it more readable:

> +struct ghes_dimm_info {
> +	struct dimm_info dimm_info;
> +	int		idx;
> +};
> +
> +struct ghes_mem_info {
> +	int num_dimm;
> +	struct ghes_dimm_info *dimms;
>  };
>  
> +static struct ghes_mem_info mem_info;

/* A DIMM */
struct ghes_dimm {
        struct dimm_info dimm;
        int idx;
};

/* The memory layout of the system */
struct ghes_memory {
        struct ghes_dimm *dimms;
        int num_dimms;
};

static struct ghes_memory mem;

> +
> +#define for_each_dimm(dimm)				\
> +	for (dimm = mem_info.dimms;			\
> +	     dimm < mem_info.dimms + mem_info.num_dimm;	\
> +	     dimm++)
> +
>  static void ghes_edac_count_dimms(const struct dmi_header *dh, void *arg)
>  {
> -	int *num_dimm = arg;
> +	int *num = arg;
>  
>  	if (dh->type == DMI_ENTRY_MEM_DEVICE)
> -		(*num_dimm)++;
> +		(*num)++;
> +}
> +
> +static int ghes_dimm_info_init(int num)

ghes_dimm_init()

... you get the idea - let's drop the _info crap.

> +{
> +	struct ghes_dimm_info *dimm;
> +	int idx = 0;
> +
> +	memset(&mem_info, 0, sizeof(mem_info));
> +
> +	if (num <= 0)
> +		return -EINVAL;

Move that check into the caller mem_info_setup() so that you don't do
the memset unnecessarily.

> +
> +	mem_info.dimms = kcalloc(num, sizeof(*mem_info.dimms), GFP_KERNEL);
> +	if (!mem_info.dimms)
> +		return -ENOMEM;
> +
> +	mem_info.num_dimm = num;
> +
> +	for_each_dimm(dimm) {
> +		dimm->idx	= idx;
> +		idx++;
> +	}

or simply

	for_each_dimm(dimm)
		dimm->idx = idx++;

> +
> +	return 0;
>  }
>  
>  static int get_dimm_smbios_index(u16 handle)

-- 
Regards/Gruss,
    Boris.

Good mailing practices for 400: avoid top-posting and trim the reply.