From fd11670d2d35c5d939b03ba1ca80eb81c1f636b6 Mon Sep 17 00:00:00 2001 From: Shiju Jose Date: Mon, 12 Feb 2024 10:56:25 +0000 Subject: [PATCH] rasdaemon: ras-mc-ctl: Add support for CXL generic trace events Add support for CXL generic events to the ras-mc-ctl tool. Signed-off-by: Shiju Jose Signed-off-by: Mauro Carvalho Chehab --- util/ras-mc-ctl.in | 83 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 83 insertions(+) diff --git a/util/ras-mc-ctl.in b/util/ras-mc-ctl.in index 1bd0946..165f2ee 100755 --- a/util/ras-mc-ctl.in +++ b/util/ras-mc-ctl.in @@ -1280,6 +1280,34 @@ sub get_cxl_ce_error_status_text return join (", ", @out); } +use constant { + CXL_EVENT_RECORD_FLAG_PERMANENT => 0x0004, + CXL_EVENT_RECORD_FLAG_MAINT_NEEDED => 0x0008, + CXL_EVENT_RECORD_FLAG_PERF_DEGRADED => 0x0010, + CXL_EVENT_RECORD_FLAG_HW_REPLACE => 0x0020, +}; + +sub get_cxl_hdr_flags_text +{ + my $flags = $_[0]; + my @out; + + if ($flags & CXL_EVENT_RECORD_FLAG_PERMANENT) { + push @out, (sprintf "\'PERMANENT_CONDITION\' "); + } + if ($flags & CXL_EVENT_RECORD_FLAG_MAINT_NEEDED) { + push @out, (sprintf "\'MAINTENANCE_NEEDED\' "); + } + if ($flags & CXL_EVENT_RECORD_FLAG_PERF_DEGRADED) { + push @out, (sprintf "\'PERFORMANCE_DEGRADED\' "); + } + if ($flags & CXL_EVENT_RECORD_FLAG_HW_REPLACE) { + push @out, (sprintf "\'HARDWARE_REPLACEMENT_NEEDED\' "); + } + + return join (", ", @out); +} + sub summary { require DBI; @@ -1408,6 +1436,22 @@ sub summary print "No CXL poison errors.\n\n"; } $query_handle->finish; + + # CXL generic errors + $query = "select memdev, count(*) from cxl_generic_event$conf{opt}{since} group by memdev"; + $query_handle = $dbh->prepare($query); + $query_handle->execute(); + $query_handle->bind_columns(\($memdev, $count)); + $out = ""; + while($query_handle->fetch()) { + $out .= "\t$memdev errors: $count\n"; + } + if ($out ne "") { + print "CXL generic events summary:\n$out\n"; + } else { + print "No CXL generic errors.\n\n"; + } + $query_handle->finish; } # extlog errors @@ -1518,6 +1562,7 @@ sub errors my ($memdev, $host, $serial, $error_status, $first_error, $header_log); my ($log_type, $first_ts, $last_ts); my ($trace_type, $region, $region_uuid, $hpa, $dpa, $dpa_length, $source, $flags, $overflow_ts); + my ($hdr_uuid, $hdr_flags, $hdr_handle, $hdr_related_handle, $hdr_ts, $hdr_length, $hdr_maint_op_class, $data); my $dbh = DBI->connect("dbi:SQLite:dbname=$dbname", "", "", {}); @@ -1691,6 +1736,44 @@ sub errors } else { print "No CXL poison errors.\n\n"; } + + # CXL generic errors + use constant CXL_EVENT_RECORD_DATA_LENGTH => 0x50; + $query = "select id, timestamp, memdev, host, serial, log_type, hdr_uuid, hdr_flags, hdr_handle, hdr_related_handle, hdr_ts, hdr_length, hdr_maint_op_class, data from cxl_generic_event$conf{opt}{since} order by id"; + $query_handle = $dbh->prepare($query); + $query_handle->execute(); + $query_handle->bind_columns(\($id, $timestamp, $memdev, $host, $serial, $log_type, $hdr_uuid, $hdr_flags, $hdr_handle, $hdr_related_handle, $hdr_ts, $hdr_length, $hdr_maint_op_class, $data)); + $out = ""; + while($query_handle->fetch()) { + $out .= "$id $timestamp error: "; + $out .= "memdev=$memdev, " if (defined $memdev && length $memdev); + $out .= "host=$host, " if (defined $host && length $host); + $out .= sprintf "serial=0x%llx, ", $serial if (defined $serial && length $serial); + $out .= "log=$log_type, " if (defined $log_type && length $log_type); + $out .= "hdr_uuid=$hdr_uuid, " if (defined $hdr_uuid && length $hdr_uuid); + $out .= sprintf "hdr_flags=0x%llx %s, ", $hdr_flags, get_cxl_hdr_flags_text($hdr_flags) if (defined $hdr_flags && length $hdr_flags); + $out .= sprintf "hdr_handle=0x%x, ", $hdr_handle if (defined $hdr_handle && length $hdr_handle); + $out .= sprintf "hdr_related_handle=0x%x, ", $hdr_related_handle if (defined $hdr_related_handle && length $hdr_related_handle); + $out .= "hdr_timestamp=$hdr_ts, " if (defined $hdr_ts && length $hdr_ts); + $out .= sprintf "hdr_length=%u, ", $hdr_length if (defined $hdr_length && length $hdr_length); + $out .= sprintf "hdr_maint_op_class=%u, ", $hdr_maint_op_class if (defined $hdr_maint_op_class && length $hdr_maint_op_class); + if (defined $data && length $data) { + $out .= sprintf "data:\n"; + my @bytes = unpack "C*", $data; + for (my $i = 0; $i < CXL_EVENT_RECORD_DATA_LENGTH; $i++) { + if (($i > 0) && (($i % 16) == 0)) { + $out .= sprintf "\n %08x: ", $i; + } + $out .= sprintf "%02x%02x%02x%02x ", $bytes[$i], $bytes[$i + 1], $bytes[$i + 2], $bytes[$i + 3]; + } + } + $out .= "\n"; + } + if ($out ne "") { + print "CXL generic events:\n$out\n"; + } else { + print "No CXL generic errors.\n\n"; + } } # Extlog errors -- 2.50.1