qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Andre Przywara <andre.przywara@amd.com>
To: anthony@codemonkey.ws
Cc: Andre Przywara <andre.przywara@amd.com>, qemu-devel@nongnu.org
Subject: [Qemu-devel] [PATCH 3/3] set CPUID bits to present cores and threads topology
Date: Wed, 19 Aug 2009 15:42:42 +0200	[thread overview]
Message-ID: <1250689362-11067-4-git-send-email-andre.przywara@amd.com> (raw)
In-Reply-To: <1250689362-11067-1-git-send-email-andre.przywara@amd.com>

Controlled by the enhanced -smp option set the CPUID bits to present the
guest the desired topology. This is vendor specific, but (with the exception
of the CMP_LEGACY bit) not conflicting, so we set all bits everytime.
There is no real multithreading support for AMD CPUs, so report cores
instead.

Signed-off-by: Andre Przywara <andre.przywara@amd.com>
---
 target-i386/helper.c |   28 +++++++++++++++++++++++++---
 1 files changed, 25 insertions(+), 3 deletions(-)

diff --git a/target-i386/helper.c b/target-i386/helper.c
index 7de4c07..d4dfcbd 100644
--- a/target-i386/helper.c
+++ b/target-i386/helper.c
@@ -1631,6 +1631,10 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         *ebx = (env->cpuid_apic_id << 24) | 8 << 8; /* CLFLUSH size in quad words, Linux wants it. */
         *ecx = env->cpuid_ext_features;
         *edx = env->cpuid_features;
+        if (env->nr_cores * env->nr_threads > 1) {
+            *ebx |= (env->nr_cores * env->nr_threads) << 16;
+            *edx |= 1 << 28;    /* HTT bit */
+        }
         break;
     case 2:
         /* cache info: needed for Pentium Pro compatibility */
@@ -1641,21 +1645,29 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         break;
     case 4:
         /* cache info: needed for Core compatibility */
+        if (env->nr_cores > 1) {
+        	*eax = (env->nr_cores - 1) << 26;
+        } else {
+        	*eax = 0;
+        }
         switch (count) {
             case 0: /* L1 dcache info */
-                *eax = 0x0000121;
+                *eax |= 0x0000121;
                 *ebx = 0x1c0003f;
                 *ecx = 0x000003f;
                 *edx = 0x0000001;
                 break;
             case 1: /* L1 icache info */
-                *eax = 0x0000122;
+                *eax |= 0x0000122;
                 *ebx = 0x1c0003f;
                 *ecx = 0x000003f;
                 *edx = 0x0000001;
                 break;
             case 2: /* L2 cache info */
-                *eax = 0x0000143;
+                *eax |= 0x0000143;
+                if (env->nr_threads > 1) {
+                    *eax |= (env->nr_threads - 1) << 14;
+                }
                 *ebx = 0x3c0003f;
                 *ecx = 0x0000fff;
                 *edx = 0x0000001;
@@ -1708,6 +1720,13 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         *ecx = env->cpuid_ext3_features;
         *edx = env->cpuid_ext2_features;
 
+        if (env->nr_cores * env->nr_threads > 1 &&
+            env->cpuid_vendor1 == CPUID_VENDOR_AMD_1 &&
+            env->cpuid_vendor2 == CPUID_VENDOR_AMD_2 &&
+            env->cpuid_vendor3 == CPUID_VENDOR_AMD_3) {
+            *ecx |= 1 << 1;    /* CmpLegacy bit */
+        }
+
         if (kvm_enabled()) {
             /* Nested SVM not yet supported in KVM */
             *ecx &= ~CPUID_EXT3_SVM;
@@ -1762,6 +1781,9 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         *ebx = 0;
         *ecx = 0;
         *edx = 0;
+        if (env->nr_cores * env->nr_threads > 1) {
+            *ecx |= (env->nr_cores * env->nr_threads) - 1;
+        }
         break;
     case 0x8000000A:
         *eax = 0x00000001; /* SVM Revision */
-- 
1.6.1.3

      parent reply	other threads:[~2009-08-19 13:47 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-08-19 13:42 [Qemu-devel] [PATCH 0/3]: Introduce multi-core and multi-thread support for guests Andre Przywara
2009-08-19 13:42 ` [Qemu-devel] [PATCH 1/3] extend -smp parsing to include cores= and threads= options Andre Przywara
2009-08-19 13:42 ` [Qemu-devel] [PATCH 2/3] push CPUID level to 4 to allow Intel multicore decoding Andre Przywara
2009-08-20 10:06   ` Avi Kivity
2009-08-20 10:36     ` Andre Przywara
2009-08-20 11:06       ` Avi Kivity
2009-08-20 19:03         ` [Qemu-devel] [PATCH] allow overriding of CPUID level on command line Andre Przywara
2009-08-25 12:21         ` [Qemu-devel] [PATCH 2/3] push CPUID level to 4 to allow Intel multicore decoding Andre Przywara
2009-08-20 19:30       ` Jamie Lokier
2009-08-20 21:35         ` Andre Przywara
2009-08-19 13:42 ` Andre Przywara [this message]

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=1250689362-11067-4-git-send-email-andre.przywara@amd.com \
    --to=andre.przywara@amd.com \
    --cc=anthony@codemonkey.ws \
    --cc=qemu-devel@nongnu.org \
    /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).