linux-nvdimm.lists.01.org archive mirror
 help / color / mirror / Atom feed
From: Dan Williams <dan.j.williams@intel.com>
To: "Kaneda, Erik" <erik.kaneda@intel.com>
Cc: Yi Zhang <yi.zhang@redhat.com>,
	"Moore, Robert" <robert.moore@intel.com>,
	linux-nvdimm <linux-nvdimm@lists.01.org>,
	"Wysocki, Rafael J" <rafael.j.wysocki@intel.com>,
	nvdimm@lists.linux.dev
Subject: Re: [bug report] system panic at nfit_get_smbios_id+0x6e/0xf0 [nfit] during boot
Date: Thu, 6 May 2021 14:17:37 -0700	[thread overview]
Message-ID: <CAPcyv4g8+JHem1dJy4Hnf0eu0LfCX59u9Zta9AEXXRSa835GyA@mail.gmail.com> (raw)
In-Reply-To: <MWHPR11MB1599C0D93E6535796F3E0F21F0589@MWHPR11MB1599.namprd11.prod.outlook.com>

[-- Attachment #1: Type: text/plain, Size: 2214 bytes --]

On Thu, May 6, 2021 at 10:28 AM Kaneda, Erik <erik.kaneda@intel.com> wrote:
>
>
>
> > -----Original Message-----
> > From: Yi Zhang <yi.zhang@redhat.com>
> > Sent: Wednesday, May 5, 2021 8:05 PM
> > To: Williams, Dan J <dan.j.williams@intel.com>; Moore, Robert
> > <robert.moore@intel.com>
> > Cc: linux-nvdimm <linux-nvdimm@lists.01.org>; Kaneda, Erik
> > <erik.kaneda@intel.com>; Wysocki, Rafael J <rafael.j.wysocki@intel.com>
> > Subject: Re: [bug report] system panic at nfit_get_smbios_id+0x6e/0xf0
> > [nfit] during boot
> >
> > On Sat, May 1, 2021 at 2:05 PM Dan Williams <dan.j.williams@intel.com>
> > wrote:
> > >
> > > On Fri, Apr 30, 2021 at 7:28 PM Yi Zhang <yi.zhang@redhat.com> wrote:
> > > >
> > > > Hi
> > > >
> > > > With the latest Linux tree, my DCPMM server boot failed with the
> > > > bellow panic log, pls help check it, let me know if you need any test
> > > > for it.
> > >
> > > So v5.12 is ok but v5.12+ is not?
> > >
> > > Might you be able to bisect?
> >
> > Hi Dan
> > This issue was introduced with this patch, let me know if you need more info.
> >
> > commit cf16b05c607bd716a0a5726dc8d577a89fdc1777
> > Author: Bob Moore <robert.moore@intel.com>
> > Date:   Tue Apr 6 14:30:15 2021 -0700
> >
> >     ACPICA: ACPI 6.4: NFIT: add Location Cookie field
> >
> >     Also, update struct size to reflect these changes in nfit core driver.
> >
> >     ACPICA commit af60199a9a1de9e6844929fd4cc22334522ed195
> >
> >     Link: https://github.com/acpica/acpica/commit/af60199a
> >     Cc: Dan Williams <dan.j.williams@intel.com>
> >     Signed-off-by: Bob Moore <robert.moore@intel.com>
> >     Signed-off-by: Erik Kaneda <erik.kaneda@intel.com>
> >     Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> >
>
> It's likely that this change forced the nfit driver's code to parse the ACPI table so that it assumes that the location cookie field is always enabled and the NFIT was parsed incorrectly. Does the NFIT table on this platform contain a valid cookie field?
>

This was my fault. When I saw the size change fly by, I should have
remembered to go update all the places that do "sizeof(struct
acpi_nfit_system_address)".

Yi Zhang, can you give the attached patch a try:

[-- Attachment #2: nfit-fix.patch --]
[-- Type: text/x-patch, Size: 9735 bytes --]

ACPI: NFIT: Fix support for variable 'SPA' structure size

From: Dan Williams <dan.j.williams@intel.com>

ACPI 6.4 introduced the "SpaLocationCookie" to the NFIT "System Physical
Address (SPA) Range Structure". The presence of that new field is
indicated by the ACPI_NFIT_LOCATION_COOKIE_VALID flag. Pre-ACPI-6.4
firmware implementations omit the flag and maintain the original size of
the structure.

Update the implementation to check that flag to determine the size
rather than the ACPI 6.4 compliant definition of 'struct
acpi_nfit_system_address' from the Linux ACPICA definitions.

Update the test infrastructure for the new expectations as well, i.e.
continue to emulate the ACPI 6.3 definition of that structure.

Without this fix the kernel fails to validate 'SPA' structures and this
leads to a crash in nfit_get_smbios_id() since that routine assumes that
SPAs are valid if it finds valid SMBIOS tables.

    BUG: unable to handle page fault for address: ffffffffffffffa8
    [..]
    Call Trace:
     skx_get_nvdimm_info+0x56/0x130 [skx_edac]
     skx_get_dimm_config+0x1f5/0x213 [skx_edac]
     skx_register_mci+0x132/0x1c0 [skx_edac]

Reported-by: Yi Zhang <yi.zhang@redhat.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/acpi/nfit/core.c         |   17 +++++++++++----
 tools/testing/nvdimm/test/nfit.c |   42 +++++++++++++++++++++++---------------
 2 files changed, 37 insertions(+), 22 deletions(-)

diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
index 958aaac869e8..bfecb79e8c82 100644
--- a/drivers/acpi/nfit/core.c
+++ b/drivers/acpi/nfit/core.c
@@ -686,29 +686,36 @@ int nfit_spa_type(struct acpi_nfit_system_address *spa)
 	return -1;
 }
 
-static bool add_spa(struct acpi_nfit_desc *acpi_desc,
+static size_t sizeof_spa(struct acpi_nfit_system_address *spa)
+{
+	if (spa->flags & ACPI_NFIT_LOCATION_COOKIE_VALID)
+		return sizeof(*spa);
+	return sizeof(*spa) - 8;
+}
+
+static bool noinline add_spa(struct acpi_nfit_desc *acpi_desc,
 		struct nfit_table_prev *prev,
 		struct acpi_nfit_system_address *spa)
 {
 	struct device *dev = acpi_desc->dev;
 	struct nfit_spa *nfit_spa;
 
-	if (spa->header.length != sizeof(*spa))
+	if (spa->header.length != sizeof_spa(spa))
 		return false;
 
 	list_for_each_entry(nfit_spa, &prev->spas, list) {
-		if (memcmp(nfit_spa->spa, spa, sizeof(*spa)) == 0) {
+		if (memcmp(nfit_spa->spa, spa, sizeof_spa(spa)) == 0) {
 			list_move_tail(&nfit_spa->list, &acpi_desc->spas);
 			return true;
 		}
 	}
 
-	nfit_spa = devm_kzalloc(dev, sizeof(*nfit_spa) + sizeof(*spa),
+	nfit_spa = devm_kzalloc(dev, sizeof(*nfit_spa) + sizeof_spa(spa),
 			GFP_KERNEL);
 	if (!nfit_spa)
 		return false;
 	INIT_LIST_HEAD(&nfit_spa->list);
-	memcpy(nfit_spa->spa, spa, sizeof(*spa));
+	memcpy(nfit_spa->spa, spa, sizeof_spa(spa));
 	list_add_tail(&nfit_spa->list, &acpi_desc->spas);
 	dev_dbg(dev, "spa index: %d type: %s\n",
 			spa->range_index,
diff --git a/tools/testing/nvdimm/test/nfit.c b/tools/testing/nvdimm/test/nfit.c
index 9b185bf82da8..54f367cbadae 100644
--- a/tools/testing/nvdimm/test/nfit.c
+++ b/tools/testing/nvdimm/test/nfit.c
@@ -1871,9 +1871,16 @@ static void smart_init(struct nfit_test *t)
 	}
 }
 
+static size_t sizeof_spa(struct acpi_nfit_system_address *spa)
+{
+	/* until spa location cookie support is added... */
+	return sizeof(*spa) - 8;
+}
+
 static int nfit_test0_alloc(struct nfit_test *t)
 {
-	size_t nfit_size = sizeof(struct acpi_nfit_system_address) * NUM_SPA
+	struct acpi_nfit_system_address *spa = NULL;
+	size_t nfit_size = sizeof_spa(spa) * NUM_SPA
 			+ sizeof(struct acpi_nfit_memory_map) * NUM_MEM
 			+ sizeof(struct acpi_nfit_control_region) * NUM_DCR
 			+ offsetof(struct acpi_nfit_control_region,
@@ -1937,7 +1944,8 @@ static int nfit_test0_alloc(struct nfit_test *t)
 
 static int nfit_test1_alloc(struct nfit_test *t)
 {
-	size_t nfit_size = sizeof(struct acpi_nfit_system_address) * 2
+	struct acpi_nfit_system_address *spa = NULL;
+	size_t nfit_size = sizeof_spa(spa) * 2
 		+ sizeof(struct acpi_nfit_memory_map) * 2
 		+ offsetof(struct acpi_nfit_control_region, window_size) * 2;
 	int i;
@@ -2000,7 +2008,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	 */
 	spa = nfit_buf;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_PM), 16);
 	spa->range_index = 0+1;
 	spa->address = t->spa_set_dma[0];
@@ -2014,7 +2022,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_PM), 16);
 	spa->range_index = 1+1;
 	spa->address = t->spa_set_dma[1];
@@ -2024,7 +2032,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa2 (dcr0) dimm0 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_DCR), 16);
 	spa->range_index = 2+1;
 	spa->address = t->dcr_dma[0];
@@ -2034,7 +2042,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa3 (dcr1) dimm1 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_DCR), 16);
 	spa->range_index = 3+1;
 	spa->address = t->dcr_dma[1];
@@ -2044,7 +2052,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa4 (dcr2) dimm2 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_DCR), 16);
 	spa->range_index = 4+1;
 	spa->address = t->dcr_dma[2];
@@ -2054,7 +2062,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa5 (dcr3) dimm3 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_DCR), 16);
 	spa->range_index = 5+1;
 	spa->address = t->dcr_dma[3];
@@ -2064,7 +2072,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa6 (bdw for dcr0) dimm0 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_BDW), 16);
 	spa->range_index = 6+1;
 	spa->address = t->dimm_dma[0];
@@ -2074,7 +2082,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa7 (bdw for dcr1) dimm1 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_BDW), 16);
 	spa->range_index = 7+1;
 	spa->address = t->dimm_dma[1];
@@ -2084,7 +2092,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa8 (bdw for dcr2) dimm2 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_BDW), 16);
 	spa->range_index = 8+1;
 	spa->address = t->dimm_dma[2];
@@ -2094,7 +2102,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa9 (bdw for dcr3) dimm3 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_BDW), 16);
 	spa->range_index = 9+1;
 	spa->address = t->dimm_dma[3];
@@ -2581,7 +2589,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 		/* spa10 (dcr4) dimm4 */
 		spa = nfit_buf + offset;
 		spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-		spa->header.length = sizeof(*spa);
+		spa->header.length = sizeof_spa(spa);
 		memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_DCR), 16);
 		spa->range_index = 10+1;
 		spa->address = t->dcr_dma[4];
@@ -2595,7 +2603,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 		 */
 		spa = nfit_buf + offset;
 		spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-		spa->header.length = sizeof(*spa);
+		spa->header.length = sizeof_spa(spa);
 		memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_PM), 16);
 		spa->range_index = 11+1;
 		spa->address = t->spa_set_dma[2];
@@ -2605,7 +2613,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 		/* spa12 (bdw for dcr4) dimm4 */
 		spa = nfit_buf + offset;
 		spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-		spa->header.length = sizeof(*spa);
+		spa->header.length = sizeof_spa(spa);
 		memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_BDW), 16);
 		spa->range_index = 12+1;
 		spa->address = t->dimm_dma[4];
@@ -2739,7 +2747,7 @@ static void nfit_test1_setup(struct nfit_test *t)
 	/* spa0 (flat range with no bdw aliasing) */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_PM), 16);
 	spa->range_index = 0+1;
 	spa->address = t->spa_set_dma[0];
@@ -2749,7 +2757,7 @@ static void nfit_test1_setup(struct nfit_test *t)
 	/* virtual cd region */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_VCD), 16);
 	spa->range_index = 0;
 	spa->address = t->spa_set_dma[1];

[-- Attachment #3: Type: text/plain, Size: 167 bytes --]

_______________________________________________
Linux-nvdimm mailing list -- linux-nvdimm@lists.01.org
To unsubscribe send an email to linux-nvdimm-leave@lists.01.org

  reply	other threads:[~2021-05-06 21:17 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-01  2:27 [bug report] system panic at nfit_get_smbios_id+0x6e/0xf0 [nfit] during boot Yi Zhang
2021-05-01  6:05 ` Dan Williams
2021-05-06  3:05   ` Yi Zhang
2021-05-06 17:27     ` Kaneda, Erik
2021-05-06 21:17       ` Dan Williams [this message]
2021-05-07  1:41         ` Yi Zhang

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=CAPcyv4g8+JHem1dJy4Hnf0eu0LfCX59u9Zta9AEXXRSa835GyA@mail.gmail.com \
    --to=dan.j.williams@intel.com \
    --cc=erik.kaneda@intel.com \
    --cc=linux-nvdimm@lists.01.org \
    --cc=nvdimm@lists.linux.dev \
    --cc=rafael.j.wysocki@intel.com \
    --cc=robert.moore@intel.com \
    --cc=yi.zhang@redhat.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).