public inbox for linux-i2c@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH v2 1/2 i2c-tools] decode-dimms: Complete check for out-of-bounds vendor ID
@ 2015-06-01  7:26 Jean Delvare
       [not found] ` <20150601092614.539ebc75-R0o5gVi9kd7kN2dkZ6Wm7A@public.gmane.org>
  0 siblings, 1 reply; 2+ messages in thread
From: Jean Delvare @ 2015-06-01  7:26 UTC (permalink / raw)
  To: Linux I2C; +Cc: Lubomir Rintel

We also need to discard vendor ID 0 and to report "Unknown" if the
vendor is in a page we don't support yet.

Additionally, check that data is present at all for DDR3 modules as we
already do for pre-DDR3 modules.
---
Changes since v1:
 * Added check for data presence.

 eeprom/decode-dimms |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

--- i2c-tools.orig/eeprom/decode-dimms	2015-06-01 08:19:59.292190666 +0200
+++ i2c-tools/eeprom/decode-dimms	2015-06-01 09:14:30.899898763 +0200
@@ -344,8 +344,12 @@ sub manufacturer_ddr3($$)
 	my ($count, $code) = @_;
 	my $manufacturer;
 
-	return "Invalid" if parity($code) != 1;
-	return "Unknown" if ($code & 0x7F) - 1 >= @{$vendors[$count & 0x7F]};
+	return "Undefined" unless spd_written($count, $code);
+
+	return "Invalid" if parity($code) != 1
+			 or ($code & 0x7F) == 0;
+	return "Unknown" if ($count & 0x7F) >= @vendors
+			 or ($code & 0x7F) - 1 >= @{$vendors[$count & 0x7F]};
 	$manufacturer = $vendors[$count & 0x7F][($code & 0x7F) - 1];
 	$manufacturer =~ s/ \(former .*\)$// if $opt_side_by_side;
 	$manufacturer .= "? (Invalid parity)" if parity($count) != 1;


-- 
Jean Delvare
SUSE L3 Support

^ permalink raw reply	[flat|nested] 2+ messages in thread

* [PATCH 2/2 i2c-tools] sensors-detect: Refactor manufacturer decoding
       [not found] ` <20150601092614.539ebc75-R0o5gVi9kd7kN2dkZ6Wm7A@public.gmane.org>
@ 2015-06-01  7:27   ` Jean Delvare
  0 siblings, 0 replies; 2+ messages in thread
From: Jean Delvare @ 2015-06-01  7:27 UTC (permalink / raw)
  To: Linux I2C; +Cc: Lubomir Rintel

Instead of reencoding pre-DDR3 manufacturer code to DDR3 manufacturer
code, move the common decoding to a separate function.
---
 eeprom/decode-dimms |   28 +++++++++++++++++-----------
 1 file changed, 17 insertions(+), 11 deletions(-)

--- i2c-tools.orig/eeprom/decode-dimms	2015-06-01 09:14:30.899898763 +0200
+++ i2c-tools/eeprom/decode-dimms	2015-06-01 09:19:01.931673815 +0200
@@ -336,6 +336,21 @@ sub parity($)
 	return ($parity & 1);
 }
 
+# The code byte includes parity, the count byte does not.
+sub manufacturer_common($$)
+{
+	my ($count, $code) = @_;
+	my $manufacturer;
+
+	return "Invalid" if parity($code) != 1
+			 or ($code &= 0x7F) == 0;
+	return "Unknown" if $count >= @vendors
+			 or $code - 1 >= @{$vendors[$count]};
+	$manufacturer = $vendors[$count][$code - 1];
+	$manufacturer =~ s/ \(former .*\)$// if $opt_side_by_side;
+	return $manufacturer;
+}
+
 # New encoding format (as of DDR3) for manufacturer just has a count of
 # leading 0x7F rather than all the individual bytes.  The count bytes includes
 # parity!
@@ -346,12 +361,7 @@ sub manufacturer_ddr3($$)
 
 	return "Undefined" unless spd_written($count, $code);
 
-	return "Invalid" if parity($code) != 1
-			 or ($code & 0x7F) == 0;
-	return "Unknown" if ($count & 0x7F) >= @vendors
-			 or ($code & 0x7F) - 1 >= @{$vendors[$count & 0x7F]};
-	$manufacturer = $vendors[$count & 0x7F][($code & 0x7F) - 1];
-	$manufacturer =~ s/ \(former .*\)$// if $opt_side_by_side;
+	$manufacturer = manufacturer_common($count & 0x7F, $code);
 	$manufacturer .= "? (Invalid parity)" if parity($count) != 1;
 	return $manufacturer;
 }
@@ -369,11 +379,7 @@ sub manufacturer(@)
 	}
 
 	return ("Invalid", []) unless defined $first;
-	return ("Invalid", [$first, @bytes]) if parity($first) != 1;
-	if (parity($ai) == 0) {
-		$ai |= 0x80;
-	}
-	return (manufacturer_ddr3($ai, $first), \@bytes);
+	return (manufacturer_common($ai, $first), \@bytes);
 }
 
 sub manufacturer_data(@)


-- 
Jean Delvare
SUSE L3 Support

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2015-06-01  7:27 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-06-01  7:26 [PATCH v2 1/2 i2c-tools] decode-dimms: Complete check for out-of-bounds vendor ID Jean Delvare
     [not found] ` <20150601092614.539ebc75-R0o5gVi9kd7kN2dkZ6Wm7A@public.gmane.org>
2015-06-01  7:27   ` [PATCH 2/2 i2c-tools] sensors-detect: Refactor manufacturer decoding Jean Delvare

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox