From: Eduardo Habkost <ehabkost@redhat.com>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: qemu-devel@nongnu.org, "Paolo Bonzini" <pbonzini@redhat.com>,
"Andreas Färber" <afaerber@suse.de>,
"Richard Henderson" <rth@twiddle.net>
Subject: [Qemu-devel] [PULL 07/12] target-i386: add ABM to Haswell* and Broadwell* CPU models
Date: Mon, 5 Oct 2015 13:06:29 -0300 [thread overview]
Message-ID: <1444061194-32753-8-git-send-email-ehabkost@redhat.com> (raw)
In-Reply-To: <1444061194-32753-1-git-send-email-ehabkost@redhat.com>
From: Paolo Bonzini <pbonzini@redhat.com>
ABM is only implemented as a single instruction set by AMD; all AMD
processors support both instructions or neither. Intel considers POPCNT
as part of SSE4.2, and LZCNT as part of BMI1, but Intel also uses AMD's
ABM flag to indicate support for both POPCNT and LZCNT. It has to be
added to Haswell and Broadwell because Haswell, by adding LZCNT, has
completed the ABM.
Tested with "qemu-kvm -cpu Haswell-noTSX,enforce" (and also with older
machine types) on an Haswell-EP machine.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
---
include/hw/i386/pc.h | 22 +++++++++++++++++++++-
target-i386/cpu.c | 8 ++++----
2 files changed, 25 insertions(+), 5 deletions(-)
diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index c13e91d..8662a29 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -298,7 +298,27 @@ int e820_get_num_entries(void);
bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
#define PC_COMPAT_2_4 \
- HW_COMPAT_2_4
+ HW_COMPAT_2_4 \
+ {\
+ .driver = "Haswell-" TYPE_X86_CPU,\
+ .property = "abm",\
+ .value = "off",\
+ },\
+ {\
+ .driver = "Haswell-noTSX-" TYPE_X86_CPU,\
+ .property = "abm",\
+ .value = "off",\
+ },\
+ {\
+ .driver = "Broadwell-" TYPE_X86_CPU,\
+ .property = "abm",\
+ .value = "off",\
+ },\
+ {\
+ .driver = "Broadwell-noTSX-" TYPE_X86_CPU,\
+ .property = "abm",\
+ .value = "off",\
+ },
#define PC_COMPAT_2_3 \
PC_COMPAT_2_4 \
diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index 6cccc77..21bad59 100644
--- a/target-i386/cpu.c
+++ b/target-i386/cpu.c
@@ -1081,7 +1081,7 @@ static X86CPUDefinition builtin_x86_defs[] = {
CPUID_EXT2_LM | CPUID_EXT2_RDTSCP | CPUID_EXT2_NX |
CPUID_EXT2_SYSCALL,
.features[FEAT_8000_0001_ECX] =
- CPUID_EXT3_LAHF_LM,
+ CPUID_EXT3_ABM | CPUID_EXT3_LAHF_LM,
.features[FEAT_7_0_EBX] =
CPUID_7_0_EBX_FSGSBASE | CPUID_7_0_EBX_BMI1 |
CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP |
@@ -1116,7 +1116,7 @@ static X86CPUDefinition builtin_x86_defs[] = {
CPUID_EXT2_LM | CPUID_EXT2_RDTSCP | CPUID_EXT2_NX |
CPUID_EXT2_SYSCALL,
.features[FEAT_8000_0001_ECX] =
- CPUID_EXT3_LAHF_LM,
+ CPUID_EXT3_ABM | CPUID_EXT3_LAHF_LM,
.features[FEAT_7_0_EBX] =
CPUID_7_0_EBX_FSGSBASE | CPUID_7_0_EBX_BMI1 |
CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP |
@@ -1153,7 +1153,7 @@ static X86CPUDefinition builtin_x86_defs[] = {
CPUID_EXT2_LM | CPUID_EXT2_RDTSCP | CPUID_EXT2_NX |
CPUID_EXT2_SYSCALL,
.features[FEAT_8000_0001_ECX] =
- CPUID_EXT3_LAHF_LM | CPUID_EXT3_3DNOWPREFETCH,
+ CPUID_EXT3_ABM | CPUID_EXT3_LAHF_LM | CPUID_EXT3_3DNOWPREFETCH,
.features[FEAT_7_0_EBX] =
CPUID_7_0_EBX_FSGSBASE | CPUID_7_0_EBX_BMI1 |
CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP |
@@ -1191,7 +1191,7 @@ static X86CPUDefinition builtin_x86_defs[] = {
CPUID_EXT2_LM | CPUID_EXT2_RDTSCP | CPUID_EXT2_NX |
CPUID_EXT2_SYSCALL,
.features[FEAT_8000_0001_ECX] =
- CPUID_EXT3_LAHF_LM | CPUID_EXT3_3DNOWPREFETCH,
+ CPUID_EXT3_ABM | CPUID_EXT3_LAHF_LM | CPUID_EXT3_3DNOWPREFETCH,
.features[FEAT_7_0_EBX] =
CPUID_7_0_EBX_FSGSBASE | CPUID_7_0_EBX_BMI1 |
CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP |
--
2.1.0
next prev parent reply other threads:[~2015-10-05 16:07 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-05 16:06 [Qemu-devel] [PULL 00/12] X86 queue, 2015-10-05 Eduardo Habkost
2015-10-05 16:06 ` [Qemu-devel] [PULL 01/12] cpu: Introduce X86CPUTopoInfo structure for argument simplification Eduardo Habkost
2015-10-05 16:06 ` [Qemu-devel] [PULL 02/12] vl: Add another sanity check to smp_parse() function Eduardo Habkost
2015-10-05 16:06 ` [Qemu-devel] [PULL 03/12] target-i386: Convert kvm_default_*features to property/value pairs Eduardo Habkost
2015-10-05 16:06 ` [Qemu-devel] [PULL 04/12] target-i386: Move breakpoint related functions to new file Eduardo Habkost
2015-10-05 16:06 ` [Qemu-devel] [PULL 05/12] target-i386: Make check_hw_breakpoints static Eduardo Habkost
2015-10-05 16:06 ` [Qemu-devel] [PULL 06/12] target-i386: get/put MSR_TSC_AUX across reset and migration Eduardo Habkost
2015-10-05 16:06 ` Eduardo Habkost [this message]
2015-10-05 16:06 ` [Qemu-devel] [PULL 08/12] Correctly re-init EFER state during INIT IPI Eduardo Habkost
2015-10-05 16:06 ` [Qemu-devel] [PULL 09/12] apic: move APIC's MMIO region mapping into APIC Eduardo Habkost
2015-10-05 16:06 ` [Qemu-devel] [PULL 10/12] x86: use new method to correct reset sequence Eduardo Habkost
2015-10-05 16:06 ` [Qemu-devel] [PULL 11/12] cpu/apic: drop icc bus/bridge Eduardo Habkost
2015-10-05 16:06 ` [Qemu-devel] [PULL 12/12] icc_bus: drop the unused files Eduardo Habkost
2015-10-06 13:47 ` [Qemu-devel] [PULL 00/12] X86 queue, 2015-10-05 Peter Maydell
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=1444061194-32753-8-git-send-email-ehabkost@redhat.com \
--to=ehabkost@redhat.com \
--cc=afaerber@suse.de \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
/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;
as well as URLs for NNTP newsgroup(s).