[Linux-parport] [PATCHv5 2/2] parport: parport_pc: Implement architecture and device check to cut off false-positives

[matwey at sai.msu.ru]

From: "Matwey V. Kornilov" <matwey at sai.msu.ru>

We definitely know that only x86 (32-bit) architecture is affected by the issue, so implement a stub instead of the actual check for other architectures.

We also know that motherboard LPT chipset is affected, so the port is either come from
  parport_pc_init (when `io' module param is used) or
  parport_pc_find_isa_ports (when default LPT ports are probbed: 0x378, 0x278, 0x3bc).
In both cases the port considered as 'legacy' and `dev' member of struct parport is NULL. See also comments for `struct parport' in parport.h

Signed-off-by: Matwey V. Kornilov <matwey at sai.msu.ru>
---
 drivers/parport/parport_pc.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/parport/parport_pc.c b/drivers/parport/parport_pc.c
index fedc06b..5b27747 100644
--- a/drivers/parport/parport_pc.c
+++ b/drivers/parport/parport_pc.c
@@ -1702,7 +1702,8 @@ static int parport_ECP_supported(struct parport *pb)
 }
 #endif
 
-static int intel_bug_present(struct parport *pb)
+#ifdef CONFIG_X86_32
+static int intel_bug_present_check_epp(struct parport *pb)
 {
 	const struct parport_pc_private *priv = pb->private_data;
 	int bug_present = 0;
@@ -1725,6 +1726,21 @@ static int intel_bug_present(struct parport *pb)
 
 	return bug_present;
 }
+static int intel_bug_present(struct parport *pb)
+{
+/* Check whether the device is legacy, not PCI or PCMCIA. Only legacy is known to be affected. */
+	if (pb->dev != NULL) {
+		return 0;
+	}
+
+	return intel_bug_present_check_epp(pb);
+}
+#else
+static int intel_bug_present(struct parport *pb)
+{
+	return 0;
+}
+#endif /* CONFIG_X86_32 */
 
 static int parport_ECPPS2_supported(struct parport *pb)
 {
-- 
1.8.1.4