From: Borislav Petkov <borislav.petkov@amd.com>
To: <mingo@elte.hu>, <hpa@zytor.com>, <tglx@linutronix.de>,
<norsk5@yahoo.com>, <aris@redhat.com>
Cc: <linux-kernel@vger.kernel.org>, <x86@kernel.org>
Subject: [PATCH 15/20] EDAC, AMD: decode data cache MCEs
Date: Tue, 28 Jul 2009 17:06:15 +0200 [thread overview]
Message-ID: <1248793580-29899-16-git-send-email-borislav.petkov@amd.com> (raw)
In-Reply-To: <1248793580-29899-1-git-send-email-borislav.petkov@amd.com>
Those get reported in MC0_STATUS, see Table 92, F10h BKDG (31116, rev.
3.28) for more details.
Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
drivers/edac/edac_mce_amd.c | 56 +++++++++++++++++++++++++++++++++++++++++-
1 files changed, 54 insertions(+), 2 deletions(-)
diff --git a/drivers/edac/edac_mce_amd.c b/drivers/edac/edac_mce_amd.c
index e332da0..07875d6 100644
--- a/drivers/edac/edac_mce_amd.c
+++ b/drivers/edac/edac_mce_amd.c
@@ -118,6 +118,49 @@ const char *ext_msgs[] = {
};
EXPORT_SYMBOL_GPL(ext_msgs);
+static void amd_decode_dc_mce(u64 mc0_status)
+{
+ u32 ec = mc0_status & 0xffff;
+ u32 xec = (mc0_status >> 16) & 0xf;
+
+ pr_emerg(" Data Cache Error");
+
+ if (xec == 1 && TLB_ERROR(ec))
+ pr_cont(": %s TLB multimatch.\n", LL_MSG(ec));
+ else if (xec == 0) {
+ if (mc0_status & (1ULL << 40))
+ pr_cont(" during Data Scrub.\n");
+ else if (TLB_ERROR(ec))
+ pr_cont(": %s TLB parity error.\n", LL_MSG(ec));
+ else if (MEM_ERROR(ec)) {
+ u8 ll = ec & 0x3;
+ u8 tt = (ec >> 2) & 0x3;
+ u8 rrrr = (ec >> 4) & 0xf;
+
+ /* see F10h BKDG (31116), Table 92. */
+ if (ll == 0x1) {
+ if (tt != 0x1)
+ goto wrong_dc_mce;
+
+ pr_cont(": Data/Tag %s error.\n", RRRR_MSG(ec));
+
+ } else if (ll == 0x2 && rrrr == 0x3)
+ pr_cont(" during L1 linefill from L2.\n");
+ else
+ goto wrong_dc_mce;
+ } else if (BUS_ERROR(ec) && boot_cpu_data.x86 == 0xf)
+ pr_cont(" during system linefill.\n");
+ else
+ goto wrong_dc_mce;
+ } else
+ goto wrong_dc_mce;
+
+ return;
+
+wrong_dc_mce:
+ pr_warning("Corrupted DC MCE info?\n");
+}
+
void amd_decode_nb_mce(int node_id, struct err_regs *regs, int handle_errors)
{
u32 ec = ERROR_CODE(regs->nbsl);
@@ -201,9 +244,12 @@ void decode_mce(struct mce *m)
pr_cont("\n");
- amd_decode_err_code(m->status & 0xffff);
+ switch (m->bank) {
+ case 0:
+ amd_decode_dc_mce(m->status);
+ break;
- if (m->bank == 4) {
+ case 4:
regs.nbsl = (u32) m->status;
regs.nbsh = (u32)(m->status >> 32);
regs.nbeal = (u32) m->addr;
@@ -211,5 +257,11 @@ void decode_mce(struct mce *m)
node = cpu_data(m->extcpu).cpu_node_id;
amd_decode_nb_mce(node, ®s, 1);
+ break;
+
+ default:
+ break;
}
+
+ amd_decode_err_code(m->status & 0xffff);
}
--
1.6.3.3
next prev parent reply other threads:[~2009-07-28 15:06 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-07-28 15:06 [RFC PATCH 0/20 v2] marry mcheck to EDAC Borislav Petkov
2009-07-28 15:06 ` [PATCH 01/20] topology: introduce cpu_node information for multi-node processors Borislav Petkov
2009-07-28 15:06 ` [PATCH 02/20] x86: provide CPU topology " Borislav Petkov
2009-07-28 15:06 ` [PATCH 03/20] x86: add cpu_node topology detection for AMD Magny-Cours Borislav Petkov
2009-07-28 15:06 ` [PATCH 04/20] x86: cacheinfo: fixup L3 cache information " Borislav Petkov
2009-07-28 15:06 ` [PATCH 05/20] x86: mcheck: make use of cpu_node_mask instead of cpu_core_mask to support multi-node processors Borislav Petkov
2009-07-28 15:06 ` [PATCH 06/20] EDAC: move MCE error descriptions to EDAC core Borislav Petkov
2009-07-28 15:06 ` [PATCH 07/20] EDAC: beef up ErrorCodeExt error signatures Borislav Petkov
2009-07-28 15:06 ` [PATCH 08/20] amd64_edac: cleanup amd64_process_error_info Borislav Petkov
2009-07-28 15:06 ` [PATCH 09/20] amd64_edac: cleanup/complete NB MCE decoding Borislav Petkov
2009-07-28 15:06 ` [PATCH 10/20] amd64_edac: remove memory and GART TLB error decoders Borislav Petkov
2009-07-28 15:06 ` [PATCH 11/20] amd64_edac: cleanup amd64_decode_bus_error Borislav Petkov
2009-07-28 15:06 ` [PATCH 12/20] x86, mce: pass mce info to EDAC for decoding Borislav Petkov
2009-07-28 15:06 ` [PATCH 13/20] EDAC, AMD: carve out MCi_STATUS decoding Borislav Petkov
2009-07-28 15:06 ` [PATCH 14/20] EDAC, AMD: carve out decoding of MCi_STATUS ErrorCode Borislav Petkov
2009-07-28 15:06 ` Borislav Petkov [this message]
2009-07-28 15:06 ` [PATCH 16/20] EDAC, AMD: decode instruction cache MCEs Borislav Petkov
2009-07-28 15:06 ` [PATCH 17/20] EDAC, AMD: decode bus unit MCEs Borislav Petkov
2009-07-28 15:06 ` [PATCH 18/20] EDAC, AMD: decode load store MCEs Borislav Petkov
2009-07-28 15:06 ` [PATCH 19/20] EDAC, AMD: decode FR MCEs Borislav Petkov
2009-07-28 15:06 ` [PATCH 20/20] x86, mce: do not compile mcelog message on AMD Borislav Petkov
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=1248793580-29899-16-git-send-email-borislav.petkov@amd.com \
--to=borislav.petkov@amd.com \
--cc=aris@redhat.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=norsk5@yahoo.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/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
all inboxes | Powered by JetHome®