Merge: x86/bugs: Extend VMware Retbleed workaround to Nehalem & earlier CPUs

MR: https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-9/-/merge_requests/4768

JIRA: https://issues.redhat.com/browse/RHEL-48649    
Upstream Status: RHEL only

To address the incorrect ARCH_CAP_RSBA bit set by VMware hypervisor,
a retbleed_whitelist[] was introduced in RHEL only commit 890c306
("x86/bugs: Workaround for incorrectly set X86_BUG_RETBLEED under
VMware") to work around incorrect activation of IBRS which could have
some noticeable performance impact. As it was a Westmere CPU that was
having this problem at the time, the whitelist included Westmere and
all the follow-on CPUs before Skylake.

Now it is reported that some Nehalem using customers also have the same
problem. Extend the retbleed_whitelist[] to include all the Intel FAM6
processors before Skylake as well to prevent them from being marked as
vulnerable to Retbleed.

Signed-off-by: Waiman Long <longman@redhat.com>

Approved-by: Steve Best <sbest@redhat.com>
Approved-by: Lenny Szubowicz <lszubowi@redhat.com>
Approved-by: Jay Shin <jaeshin@redhat.com>
Approved-by: CKI KWF Bot <cki-ci-bot+kwf-gitlab-com@redhat.com>

Merged-by: Lucas Zampieri <lzampier@redhat.com>

Author: zampierilucas

arch/x86/kernel/cpu/common.c

@@ -161,6 +161,15 @@ static void ppin_init(struct cpuinfo_x86 *c)
 static const bool cpu_in_retbleed_whitelist(struct cpuinfo_x86 *c)
 {
 	static const bool retbleed_whitelist[] = {
+		[INTEL_FAM6_CORE_YONAH]		= true,
+		[INTEL_FAM6_CORE2_MEROM]	= true,
+		[INTEL_FAM6_CORE2_MEROM_L]	= true,
+		[INTEL_FAM6_CORE2_PENRYN]	= true,
+		[INTEL_FAM6_CORE2_DUNNINGTON]	= true,
+		[INTEL_FAM6_NEHALEM]		= true,
+		[INTEL_FAM6_NEHALEM_G]		= true,
+		[INTEL_FAM6_NEHALEM_EP]		= true,
+		[INTEL_FAM6_NEHALEM_EX]		= true,
 		[INTEL_FAM6_WESTMERE]		= true,
 		[INTEL_FAM6_WESTMERE_EP]	= true,
 		[INTEL_FAM6_WESTMERE_EX]	= true,