From: Mauro Carvalho Chehab <mchehab@kernel.org>
To: Robert Richter <rrichter@marvell.com>
Cc: Borislav Petkov <bp@alien8.de>, Tony Luck <tony.luck@intel.com>,
James Morse <james.morse@arm.com>,
"linux-edac@vger.kernel.org" <linux-edac@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 02/19] EDAC: Remove EDAC_DIMM_OFF() macro
Date: Fri, 11 Oct 2019 07:09:03 -0300 [thread overview]
Message-ID: <20191011070903.546d865f@coco.lan> (raw)
In-Reply-To: <20191010202418.25098-3-rrichter@marvell.com>
Em Thu, 10 Oct 2019 20:25:07 +0000
Robert Richter <rrichter@marvell.com> escreveu:
> The EDAC_DIMM_OFF() macro takes 5 arguments to get the DIMM's index.
> This can be much simplified now as the offset is already stored in
> struct dimm_info. Use it directly and completely remove the
> EDAC_DIMM_OFF() macro.
Ah, now I understand why you added the dimm idx field. One more reason
why to split it on a separate patch (or perhaps merge it here).
>
> Another advantage is that edac_mc_alloc() could be used even if the
> exact size of the layers is unknown. Only the number of DIMMs would be
> needed.
>
> Signed-off-by: Robert Richter <rrichter@marvell.com>
> ---
> drivers/edac/edac_mc.c | 15 +++++--------
> drivers/edac/edac_mc_sysfs.c | 20 ++++--------------
> include/linux/edac.h | 41 ------------------------------------
> 3 files changed, 9 insertions(+), 67 deletions(-)
>
> diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c
> index 3d45adb5957f..72088d49b03b 100644
> --- a/drivers/edac/edac_mc.c
> +++ b/drivers/edac/edac_mc.c
> @@ -314,10 +314,10 @@ struct mem_ctl_info *edac_mc_alloc(unsigned int mc_num,
> struct dimm_info *dimm;
> u32 *ce_per_layer[EDAC_MAX_LAYERS], *ue_per_layer[EDAC_MAX_LAYERS];
> unsigned int pos[EDAC_MAX_LAYERS];
> - unsigned int size, tot_dimms = 1, count = 1;
> + unsigned int idx, size, tot_dimms = 1, count = 1;
> unsigned int tot_csrows = 1, tot_channels = 1, tot_errcount = 0;
> void *pvt, *p, *ptr = NULL;
> - int i, j, row, chn, n, len, off;
> + int i, j, row, chn, n, len;
> bool per_rank = false;
>
> BUG_ON(n_layers > EDAC_MAX_LAYERS || n_layers == 0);
> @@ -425,20 +425,15 @@ struct mem_ctl_info *edac_mc_alloc(unsigned int mc_num,
> memset(&pos, 0, sizeof(pos));
> row = 0;
> chn = 0;
> - for (i = 0; i < tot_dimms; i++) {
> + for (idx = 0; idx < tot_dimms; idx++) {
> chan = mci->csrows[row]->channels[chn];
> - off = EDAC_DIMM_OFF(layer, n_layers, pos[0], pos[1], pos[2]);
> - if (off < 0 || off >= tot_dimms) {
> - edac_mc_printk(mci, KERN_ERR, "EDAC core bug: EDAC_DIMM_OFF is trying to do an illegal data access\n");
> - goto error;
> - }
>
> dimm = kzalloc(sizeof(**mci->dimms), GFP_KERNEL);
> if (!dimm)
> goto error;
> - mci->dimms[off] = dimm;
> + mci->dimms[idx] = dimm;
> dimm->mci = mci;
> - dimm->idx = off;
> + dimm->idx = idx;
>
> /*
> * Copy DIMM location and initialize it.
> diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
> index 32d016f1ecd1..91e4c8f155af 100644
> --- a/drivers/edac/edac_mc_sysfs.c
> +++ b/drivers/edac/edac_mc_sysfs.c
> @@ -557,14 +557,8 @@ static ssize_t dimmdev_ce_count_show(struct device *dev,
> {
> struct dimm_info *dimm = to_dimm(dev);
> u32 count;
> - int off;
> -
> - off = EDAC_DIMM_OFF(dimm->mci->layers,
> - dimm->mci->n_layers,
> - dimm->location[0],
> - dimm->location[1],
> - dimm->location[2]);
> - count = dimm->mci->ce_per_layer[dimm->mci->n_layers-1][off];
> +
> + count = dimm->mci->ce_per_layer[dimm->mci->n_layers-1][dimm->idx];
> return sprintf(data, "%u\n", count);
> }
>
> @@ -574,14 +568,8 @@ static ssize_t dimmdev_ue_count_show(struct device *dev,
> {
> struct dimm_info *dimm = to_dimm(dev);
> u32 count;
> - int off;
> -
> - off = EDAC_DIMM_OFF(dimm->mci->layers,
> - dimm->mci->n_layers,
> - dimm->location[0],
> - dimm->location[1],
> - dimm->location[2]);
> - count = dimm->mci->ue_per_layer[dimm->mci->n_layers-1][off];
> +
> + count = dimm->mci->ue_per_layer[dimm->mci->n_layers-1][dimm->idx];
> return sprintf(data, "%u\n", count);
> }
>
> diff --git a/include/linux/edac.h b/include/linux/edac.h
> index 7f9804438589..79c5564ee314 100644
> --- a/include/linux/edac.h
> +++ b/include/linux/edac.h
> @@ -362,47 +362,6 @@ struct edac_mc_layer {
> */
> #define EDAC_MAX_LAYERS 3
>
> -/**
> - * EDAC_DIMM_OFF - Macro responsible to get a pointer offset inside a pointer
> - * array for the element given by [layer0,layer1,layer2]
> - * position
> - *
> - * @layers: a struct edac_mc_layer array, describing how many elements
> - * were allocated for each layer
> - * @nlayers: Number of layers at the @layers array
> - * @layer0: layer0 position
> - * @layer1: layer1 position. Unused if n_layers < 2
> - * @layer2: layer2 position. Unused if n_layers < 3
> - *
> - * For 1 layer, this macro returns "var[layer0] - var";
> - *
> - * For 2 layers, this macro is similar to allocate a bi-dimensional array
> - * and to return "var[layer0][layer1] - var";
> - *
> - * For 3 layers, this macro is similar to allocate a tri-dimensional array
> - * and to return "var[layer0][layer1][layer2] - var".
> - *
> - * A loop could be used here to make it more generic, but, as we only have
> - * 3 layers, this is a little faster.
> - *
> - * By design, layers can never be 0 or more than 3. If that ever happens,
> - * a NULL is returned, causing an OOPS during the memory allocation routine,
> - * with would point to the developer that he's doing something wrong.
> - */
> -#define EDAC_DIMM_OFF(layers, nlayers, layer0, layer1, layer2) ({ \
> - int __i; \
> - if ((nlayers) == 1) \
> - __i = layer0; \
> - else if ((nlayers) == 2) \
> - __i = (layer1) + ((layers[1]).size * (layer0)); \
> - else if ((nlayers) == 3) \
> - __i = (layer2) + ((layers[2]).size * ((layer1) + \
> - ((layers[1]).size * (layer0)))); \
> - else \
> - __i = -EINVAL; \
> - __i; \
> -})
> -
> struct dimm_info {
> struct device dev;
>
Thanks,
Mauro
next prev parent reply other threads:[~2019-10-11 10:09 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-10 20:25 [PATCH 00/19] EDAC: Rework edac_mc and ghes drivers Robert Richter
2019-10-10 20:25 ` [PATCH 01/19] EDAC: Replace EDAC_DIMM_PTR() macro with edac_get_dimm() function Robert Richter
2019-10-11 9:58 ` Mauro Carvalho Chehab
2019-10-11 11:38 ` Robert Richter
2019-10-10 20:25 ` [PATCH 02/19] EDAC: Remove EDAC_DIMM_OFF() macro Robert Richter
2019-10-11 10:09 ` Mauro Carvalho Chehab [this message]
2019-10-11 11:36 ` Robert Richter
2019-10-10 20:25 ` [PATCH 03/19] EDAC: Introduce mci_for_each_dimm() iterator Robert Richter
2019-10-11 10:14 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 04/19] EDAC, mc: Do not BUG_ON() in edac_mc_alloc() Robert Richter
2019-10-11 10:15 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 05/19] EDAC, mc: Reduce indentation level in edac_mc_handle_error() Robert Richter
2019-10-10 22:10 ` Joe Perches
2019-10-11 6:50 ` Robert Richter
2019-10-11 10:20 ` Mauro Carvalho Chehab
2019-10-11 10:50 ` Joe Perches
2019-10-11 12:08 ` Robert Richter
2019-10-11 14:49 ` Joe Perches
2019-10-11 10:17 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 06/19] EDAC, mc: Remove per layer counters Robert Richter
2019-10-11 10:40 ` Mauro Carvalho Chehab
2019-10-14 11:12 ` Robert Richter
2019-10-10 20:25 ` [PATCH 07/19] EDAC, mc: Rename iterator variable to idx Robert Richter
2019-10-11 10:41 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 08/19] EDAC, mc: Split edac_mc_alloc() into smaller functions Robert Richter
2019-10-11 10:43 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 09/19] EDAC, mc: Reorder functions edac_mc_alloc*() Robert Richter
2019-10-11 10:45 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 10/19] EDAC, mc: Rework edac_raw_mc_handle_error() to use struct dimm_info Robert Richter
2019-10-11 10:48 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 11/19] EDAC: Remove misleading comment in struct edac_raw_error_desc Robert Richter
2019-10-11 10:49 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 12/19] EDAC: Store error type " Robert Richter
2019-10-11 10:54 ` Mauro Carvalho Chehab
2019-10-14 11:47 ` Robert Richter
2019-10-10 20:25 ` [PATCH 13/19] EDAC, mc: Determine mci pointer from the error descriptor Robert Richter
2019-10-11 10:56 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 14/19] EDAC, mc: Create new function edac_inc_csrow() Robert Richter
2019-10-11 11:08 ` Mauro Carvalho Chehab
2019-10-14 11:58 ` Robert Richter
2019-10-10 20:25 ` [PATCH 15/19] EDAC, ghes: Use standard kernel macros for page calculations Robert Richter
2019-10-11 11:10 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 16/19] EDAC, ghes: Fix grain calculation Robert Richter
2019-10-11 11:22 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 17/19] EDAC, ghes: Remove intermediate buffer pvt->detail_location Robert Richter
2019-10-11 11:20 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 18/19] EDAC, ghes: Unify trace_mc_event() code with edac_mc driver Robert Richter
2019-10-11 11:23 ` Mauro Carvalho Chehab
2019-10-10 20:25 ` [PATCH 19/19] EDAC, Documentation: Describe CPER module definition and DIMM ranks Robert Richter
2019-10-11 11:29 ` Mauro Carvalho Chehab
2019-10-10 20:36 ` [PATCH 00/19] EDAC: Rework edac_mc and ghes drivers Robert Richter
2019-10-14 12:00 ` Robert Richter
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20191011070903.546d865f@coco.lan \
--to=mchehab@kernel.org \
--cc=bp@alien8.de \
--cc=james.morse@arm.com \
--cc=linux-edac@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=rrichter@marvell.com \
--cc=tony.luck@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.