From: Steven Rostedt <rostedt@goodmis.org>
To: Borislav Petkov <bp@alien8.de>
Cc: "Luck, Tony" <tony.luck@intel.com>,
"Kani, Toshimitsu" <toshi.kani@hpe.com>,
"rjw@rjwysocki.net" <rjw@rjwysocki.net>,
"lenb@kernel.org" <lenb@kernel.org>,
"mchehab@kernel.org" <mchehab@kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-acpi@vger.kernel.org" <linux-acpi@vger.kernel.org>,
"linux-edac@vger.kernel.org" <linux-edac@vger.kernel.org>
Subject: Re: [PATCH v2 4/7] ghes_edac: avoid multiple calls to dmi_walk()
Date: Wed, 16 Aug 2017 09:59:01 -0400 [thread overview]
Message-ID: <20170816095901.33b0d4c2@gandalf.local.home> (raw)
In-Reply-To: <20170816082931.p6rpvtlxwt5nccxr@pd.tnic>
On Wed, 16 Aug 2017 10:29:31 +0200
Borislav Petkov <bp@alien8.de> wrote:
> ---
> diff --git a/drivers/edac/ghes_edac.c b/drivers/edac/ghes_edac.c
> index 6f80eb65c26c..a22fabef4791 100644
> --- a/drivers/edac/ghes_edac.c
> +++ b/drivers/edac/ghes_edac.c
> @@ -28,10 +28,14 @@ struct ghes_edac_pvt {
> char msg[80];
> };
>
> -static LIST_HEAD(ghes_reglist);
> -static DEFINE_MUTEX(ghes_edac_lock);
> -static int ghes_edac_mc_num;
> +static struct ghes_edac_pvt *ghes_pvt;
>
> +/*
> + * Sync with other, potentially concurrent callers of
> + * ghes_edac_report_mem_error(). We don't know what the
> + * "inventive" firmware would do.
> + */
> +static DEFINE_SPINLOCK(ghes_lock);
>
> /* Memory Device - Type 17 of SMBIOS spec */
> struct memdev_dmi_entry {
> @@ -169,14 +173,11 @@ void ghes_edac_report_mem_error(struct ghes *ghes, int sev,
> enum hw_event_mc_err_type type;
> struct edac_raw_error_desc *e;
> struct mem_ctl_info *mci;
> - struct ghes_edac_pvt *pvt = NULL;
> + struct ghes_edac_pvt *pvt = ghes_pvt;
> + unsigned long flags;
> char *p;
> u8 grain_bits;
>
> - list_for_each_entry(pvt, &ghes_reglist, list) {
> - if (ghes == pvt->ghes)
> - break;
> - }
> if (!pvt) {
> pr_err("Internal error: Can't find EDAC structure\n");
> return;
> @@ -398,8 +399,16 @@ void ghes_edac_report_mem_error(struct ghes *ghes, int sev,
> (e->page_frame_number << PAGE_SHIFT) | e->offset_in_page,
> grain_bits, e->syndrome, pvt->detail_location);
>
> - /* Report the error via EDAC API */
> + /*
> + * We can do the locking below because GHES defers error processing
> + * from NMI to IRQ context. Whenever that changes, we'd at least
> + * know.
> + */
> + WARN_ON_ONCE(in_nmi());
Should the above be:
if (WARN_ON_ONCE(in_nmi()))
return;
To prevent a deadlock? Or do we not care?
> +
> + spin_lock_irqsave(&ghes_lock, flags);
> edac_raw_mc_handle_error(type, mci, e);
> + spin_unlock_irqrestore(&ghes_lock, flags);
The above looks fine, as long as there's nothing before it that needs
synchronization.
> }
> EXPORT_SYMBOL_GPL(ghes_edac_report_mem_error);
>
> @@ -409,9 +418,14 @@ int ghes_edac_register(struct ghes *ghes, struct device *dev)
> int rc, num_dimm = 0;
> struct mem_ctl_info *mci;
> struct edac_mc_layer layers[1];
> - struct ghes_edac_pvt *pvt;
> struct ghes_edac_dimm_fill dimm_fill;
>
> + /*
> + * We have only one logical memory controller to which all DIMMs belong.
> + */
> + if (ghes_pvt)
> + return 0;
What's the likelihood of two calls to ghes_edac_register being done
simultaneously? Because two calls at the same time will get past this.
-- Steve
> +
> /* Get the number of DIMMs */
> dmi_walk(ghes_edac_count_dimms, &num_dimm);
>
> @@ -425,26 +439,17 @@ int ghes_edac_register(struct ghes *ghes, struct device *dev)
> layers[0].size = num_dimm;
> layers[0].is_virt_csrow = true;
>
> - /*
> - * We need to serialize edac_mc_alloc() and edac_mc_add_mc(),
> - * to avoid duplicated memory controller numbers
> - */
> - mutex_lock(&ghes_edac_lock);
> - mci = edac_mc_alloc(ghes_edac_mc_num, ARRAY_SIZE(layers), layers,
> - sizeof(*pvt));
> + mci = edac_mc_alloc(1, ARRAY_SIZE(layers), layers, sizeof(struct ghes_edac_pvt));
> if (!mci) {
> pr_info("Can't allocate memory for EDAC data\n");
> - mutex_unlock(&ghes_edac_lock);
> return -ENOMEM;
> }
>
> - pvt = mci->pvt_info;
> - memset(pvt, 0, sizeof(*pvt));
> - list_add_tail(&pvt->list, &ghes_reglist);
> - pvt->ghes = ghes;
> - pvt->mci = mci;
> - mci->pdev = dev;
> + ghes_pvt = mci->pvt_info;
> + ghes_pvt->ghes = ghes;
> + ghes_pvt->mci = mci;
>
> + mci->pdev = dev;
> mci->mtype_cap = MEM_FLAG_EMPTY;
> mci->edac_ctl_cap = EDAC_FLAG_NONE;
> mci->edac_cap = EDAC_FLAG_NONE;
> @@ -452,36 +457,23 @@ int ghes_edac_register(struct ghes *ghes, struct device *dev)
> mci->ctl_name = "ghes_edac";
> mci->dev_name = "ghes";
>
> - if (!ghes_edac_mc_num) {
> - if (!fake) {
> - pr_info("This EDAC driver relies on BIOS to enumerate memory and get error reports.\n");
> - pr_info("Unfortunately, not all BIOSes reflect the memory layout correctly.\n");
> - pr_info("So, the end result of using this driver varies from vendor to vendor.\n");
> - pr_info("If you find incorrect reports, please contact your hardware vendor\n");
> - pr_info("to correct its BIOS.\n");
> - pr_info("This system has %d DIMM sockets.\n",
> - num_dimm);
> - } else {
> - pr_info("This system has a very crappy BIOS: It doesn't even list the DIMMS.\n");
> - pr_info("Its SMBIOS info is wrong. It is doubtful that the error report would\n");
> - pr_info("work on such system. Use this driver with caution\n");
> - }
> + if (!fake) {
> + pr_info("This EDAC driver relies on BIOS to enumerate memory and get error reports.\n");
> + pr_info("Unfortunately, not all BIOSes reflect the memory layout correctly.\n");
> + pr_info("So, the end result of using this driver varies from vendor to vendor.\n");
> + pr_info("If you find incorrect reports, please contact your hardware vendor\n");
> + pr_info("to correct its BIOS.\n");
> + pr_info("This system has %d DIMM sockets.\n", num_dimm);
> + } else {
> + pr_info("This system has a very crappy BIOS: It doesn't even list the DIMMS.\n");
> + pr_info("Its SMBIOS info is wrong. It is doubtful that the error report would\n");
> + pr_info("work on such system. Use this driver with caution\n");
> }
>
> if (!fake) {
> - /*
> - * Fill DIMM info from DMI for the memory controller #0
> - *
> - * Keep it in blank for the other memory controllers, as
> - * there's no reliable way to properly credit each DIMM to
> - * the memory controller, as different BIOSes fill the
> - * DMI bank location fields on different ways
> - */
> - if (!ghes_edac_mc_num) {
> - dimm_fill.count = 0;
> - dimm_fill.mci = mci;
> - dmi_walk(ghes_edac_dmidecode, &dimm_fill);
> - }
> + dimm_fill.count = 0;
> + dimm_fill.mci = mci;
> + dmi_walk(ghes_edac_dmidecode, &dimm_fill);
> } else {
> struct dimm_info *dimm = EDAC_DIMM_PTR(mci->layers, mci->dimms,
> mci->n_layers, 0, 0, 0);
> @@ -497,12 +489,8 @@ int ghes_edac_register(struct ghes *ghes, struct device *dev)
> if (rc < 0) {
> pr_info("Can't register at EDAC core\n");
> edac_mc_free(mci);
> - mutex_unlock(&ghes_edac_lock);
> return -ENODEV;
> }
> -
> - ghes_edac_mc_num++;
> - mutex_unlock(&ghes_edac_lock);
> return 0;
> }
> EXPORT_SYMBOL_GPL(ghes_edac_register);
> @@ -510,15 +498,9 @@ EXPORT_SYMBOL_GPL(ghes_edac_register);
> void ghes_edac_unregister(struct ghes *ghes)
> {
> struct mem_ctl_info *mci;
> - struct ghes_edac_pvt *pvt, *tmp;
> -
> - list_for_each_entry_safe(pvt, tmp, &ghes_reglist, list) {
> - if (ghes == pvt->ghes) {
> - mci = pvt->mci;
> - edac_mc_del_mc(mci->pdev);
> - edac_mc_free(mci);
> - list_del(&pvt->list);
> - }
> - }
> +
> + mci = ghes_pvt->mci;
> + edac_mc_del_mc(mci->pdev);
> + edac_mc_free(mci);
> }
> EXPORT_SYMBOL_GPL(ghes_edac_unregister);
>
next prev parent reply other threads:[~2017-08-16 13:59 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-03 21:57 [PATCH v2 0/7] enable ghes_edac on selected platforms Toshi Kani
2017-08-03 21:57 ` [PATCH v2 1/7] ACPI / blacklist: add acpi_match_oemlist() interface Toshi Kani
2017-08-04 3:42 ` Borislav Petkov
2017-08-04 20:39 ` Kani, Toshimitsu
2017-08-05 5:12 ` Borislav Petkov
2017-08-07 14:49 ` Kani, Toshimitsu
2017-08-03 21:57 ` [PATCH v2 2/7] intel_pstate: convert to use acpi_match_oemlist() Toshi Kani
2017-08-03 21:57 ` [PATCH v2 3/7] ACPI / APEI: add OSC APEI bit check for ghes_edac Toshi Kani
2017-08-04 3:44 ` Borislav Petkov
2017-08-04 20:49 ` Kani, Toshimitsu
2017-08-05 5:14 ` Borislav Petkov
2017-08-07 14:50 ` Kani, Toshimitsu
2017-08-03 21:57 ` [PATCH v2 4/7] ghes_edac: avoid multiple calls to dmi_walk() Toshi Kani
2017-08-04 4:05 ` Borislav Petkov
2017-08-04 21:02 ` Kani, Toshimitsu
2017-08-05 5:16 ` Borislav Petkov
2017-08-07 17:59 ` Kani, Toshimitsu
2017-08-11 9:04 ` Borislav Petkov
2017-08-14 15:57 ` Kani, Toshimitsu
2017-08-14 16:24 ` Borislav Petkov
2017-08-14 16:48 ` Kani, Toshimitsu
2017-08-14 17:05 ` Borislav Petkov
2017-08-14 17:52 ` Kani, Toshimitsu
2017-08-14 18:05 ` Borislav Petkov
2017-08-14 18:17 ` Kani, Toshimitsu
2017-08-14 18:35 ` Borislav Petkov
2017-08-14 19:02 ` Kani, Toshimitsu
2017-08-14 19:34 ` Borislav Petkov
2017-08-14 20:17 ` Kani, Toshimitsu
2017-08-14 20:39 ` Borislav Petkov
2017-08-15 15:35 ` Kani, Toshimitsu
2017-08-15 15:48 ` Luck, Tony
2017-08-15 15:53 ` Kani, Toshimitsu
2017-08-16 8:29 ` Borislav Petkov
2017-08-16 11:29 ` Borislav Petkov
2017-08-16 13:59 ` Steven Rostedt [this message]
2017-08-16 14:03 ` Borislav Petkov
2017-08-16 14:22 ` Steven Rostedt
2017-08-16 17:31 ` Borislav Petkov
2017-08-16 15:26 ` Kani, Toshimitsu
2017-08-16 16:42 ` Borislav Petkov
2017-08-16 17:28 ` Kani, Toshimitsu
2017-08-16 17:40 ` Borislav Petkov
2017-08-16 18:01 ` Kani, Toshimitsu
2017-08-17 21:08 ` Kani, Toshimitsu
2017-08-21 9:29 ` Borislav Petkov
2017-08-15 15:50 ` Borislav Petkov
2017-08-15 16:19 ` Kani, Toshimitsu
2017-08-03 21:57 ` [PATCH v2 5/7] ghes_edac: add platform check to enable ghes_edac Toshi Kani
2017-08-04 8:31 ` Borislav Petkov
2017-08-04 21:06 ` Kani, Toshimitsu
2017-08-05 5:37 ` Borislav Petkov
2017-08-07 14:54 ` Kani, Toshimitsu
2017-08-03 21:57 ` [PATCH v2 6/7] EDAC: add edac_check_mc_owner() to check MC owner Toshi Kani
2017-08-04 8:30 ` Borislav Petkov
2017-08-04 21:35 ` Kani, Toshimitsu
2017-08-05 5:44 ` Borislav Petkov
2017-08-07 14:55 ` Kani, Toshimitsu
2017-08-04 13:06 ` kbuild test robot
2017-08-04 15:21 ` Kani, Toshimitsu
2017-08-03 21:57 ` [PATCH v2 7/7] edac drivers: add MC owner check in init Toshi Kani
2017-08-04 8:39 ` Borislav Petkov
2017-08-04 21:48 ` Kani, Toshimitsu
2017-08-05 5:49 ` Borislav Petkov
2017-08-07 14:57 ` Kani, Toshimitsu
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=20170816095901.33b0d4c2@gandalf.local.home \
--to=rostedt@goodmis.org \
--cc=bp@alien8.de \
--cc=lenb@kernel.org \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-edac@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mchehab@kernel.org \
--cc=rjw@rjwysocki.net \
--cc=tony.luck@intel.com \
--cc=toshi.kani@hpe.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).