From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:52818) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VGt7r-0003gp-6i for qemu-devel@nongnu.org; Tue, 03 Sep 2013 12:06:48 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VGt7l-0006nh-Lw for qemu-devel@nongnu.org; Tue, 03 Sep 2013 12:06:43 -0400 Received: from cantor2.suse.de ([195.135.220.15]:43537 helo=mx2.suse.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VGt7l-0006nM-A0 for qemu-devel@nongnu.org; Tue, 03 Sep 2013 12:06:37 -0400 From: =?UTF-8?q?Andreas=20F=C3=A4rber?= Date: Tue, 3 Sep 2013 18:06:27 +0200 Message-Id: <1378224387-18760-5-git-send-email-afaerber@suse.de> In-Reply-To: <1378224387-18760-1-git-send-email-afaerber@suse.de> References: <1378224387-18760-1-git-send-email-afaerber@suse.de> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: [Qemu-devel] [PULL 4/4] target-i386: Use #defines instead of magic numbers for CPUID cache info List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Eduardo Habkost , =?UTF-8?q?Andreas=20F=C3=A4rber?= From: Eduardo Habkost This is an attempt to make the CPUID cache topology code clearer, by replacing the magic numbers in the code with #defines, and moving all the cache information to the same place in the file. I took care of comparing the assembly output of compiling target-i386/cpu.c before and after applying this change, to make sure not a single bit was changed on cpu_x86_cpuid() before and after applying this patch (unfortunately I had to manually check existing differences, because of __LINE__ expansions on object_class_dynamic_cast_assert() calls). This even keeps the code bug-compatible with the previous version: today the cache information returned on AMD cache information leaves (CPUID 0x80000005 & 0x80000006) do not match the information returned on CPUID leaves 2 and 4. The L2 cache information on CPUID leaf 2 also doesn't match the information on CPUID leaf 2. The new constants should make it easier to eventually fix those inconsistencies. All inconsistencies I have found are documented in code comments. Signed-off-by: Eduardo Habkost Reviewed-by: liguang Signed-off-by: Andreas F=C3=A4rber --- target-i386/cpu.c | 184 +++++++++++++++++++++++++++++++++++++++++++++++-= ------ 1 file changed, 162 insertions(+), 22 deletions(-) diff --git a/target-i386/cpu.c b/target-i386/cpu.c index 42c5de0..c36345e 100644 --- a/target-i386/cpu.c +++ b/target-i386/cpu.c @@ -48,6 +48,118 @@ #include "hw/i386/apic_internal.h" #endif =20 + +/* Cache topology CPUID constants: */ + +/* CPUID Leaf 2 Descriptors */ + +#define CPUID_2_L1D_32KB_8WAY_64B 0x2c +#define CPUID_2_L1I_32KB_8WAY_64B 0x30 +#define CPUID_2_L2_2MB_8WAY_64B 0x7d + + +/* CPUID Leaf 4 constants: */ + +/* EAX: */ +#define CPUID_4_TYPE_DCACHE 1 +#define CPUID_4_TYPE_ICACHE 2 +#define CPUID_4_TYPE_UNIFIED 3 + +#define CPUID_4_LEVEL(l) ((l) << 5) + +#define CPUID_4_SELF_INIT_LEVEL (1 << 8) +#define CPUID_4_FULLY_ASSOC (1 << 9) + +/* EDX: */ +#define CPUID_4_NO_INVD_SHARING (1 << 0) +#define CPUID_4_INCLUSIVE (1 << 1) +#define CPUID_4_COMPLEX_IDX (1 << 2) + +#define ASSOC_FULL 0xFF + +/* AMD associativity encoding used on CPUID Leaf 0x80000006: */ +#define AMD_ENC_ASSOC(a) (a <=3D 1 ? a : \ + a =3D=3D 2 ? 0x2 : \ + a =3D=3D 4 ? 0x4 : \ + a =3D=3D 8 ? 0x6 : \ + a =3D=3D 16 ? 0x8 : \ + a =3D=3D 32 ? 0xA : \ + a =3D=3D 48 ? 0xB : \ + a =3D=3D 64 ? 0xC : \ + a =3D=3D 96 ? 0xD : \ + a =3D=3D 128 ? 0xE : \ + a =3D=3D ASSOC_FULL ? 0xF : \ + 0 /* invalid value */) + + +/* Definitions of the hardcoded cache entries we expose: */ + +/* L1 data cache: */ +#define L1D_LINE_SIZE 64 +#define L1D_ASSOCIATIVITY 8 +#define L1D_SETS 64 +#define L1D_PARTITIONS 1 +/* Size =3D LINE_SIZE*ASSOCIATIVITY*SETS*PARTITIONS =3D 32KiB */ +#define L1D_DESCRIPTOR CPUID_2_L1D_32KB_8WAY_64B +/*FIXME: CPUID leaf 0x80000005 is inconsistent with leaves 2 & 4 */ +#define L1D_LINES_PER_TAG 1 +#define L1D_SIZE_KB_AMD 64 +#define L1D_ASSOCIATIVITY_AMD 2 + +/* L1 instruction cache: */ +#define L1I_LINE_SIZE 64 +#define L1I_ASSOCIATIVITY 8 +#define L1I_SETS 64 +#define L1I_PARTITIONS 1 +/* Size =3D LINE_SIZE*ASSOCIATIVITY*SETS*PARTITIONS =3D 32KiB */ +#define L1I_DESCRIPTOR CPUID_2_L1I_32KB_8WAY_64B +/*FIXME: CPUID leaf 0x80000005 is inconsistent with leaves 2 & 4 */ +#define L1I_LINES_PER_TAG 1 +#define L1I_SIZE_KB_AMD 64 +#define L1I_ASSOCIATIVITY_AMD 2 + +/* Level 2 unified cache: */ +#define L2_LINE_SIZE 64 +#define L2_ASSOCIATIVITY 16 +#define L2_SETS 4096 +#define L2_PARTITIONS 1 +/* Size =3D LINE_SIZE*ASSOCIATIVITY*SETS*PARTITIONS =3D 4MiB */ +/*FIXME: CPUID leaf 2 descriptor is inconsistent with CPUID leaf 4 */ +#define L2_DESCRIPTOR CPUID_2_L2_2MB_8WAY_64B +/*FIXME: CPUID leaf 0x80000006 is inconsistent with leaves 2 & 4 */ +#define L2_LINES_PER_TAG 1 +#define L2_SIZE_KB_AMD 512 + +/* No L3 cache: */ +#define L3_SIZE_KB 0 /* disabled */ +#define L3_ASSOCIATIVITY 0 /* disabled */ +#define L3_LINES_PER_TAG 0 /* disabled */ +#define L3_LINE_SIZE 0 /* disabled */ + +/* TLB definitions: */ + +#define L1_DTLB_2M_ASSOC 1 +#define L1_DTLB_2M_ENTRIES 255 +#define L1_DTLB_4K_ASSOC 1 +#define L1_DTLB_4K_ENTRIES 255 + +#define L1_ITLB_2M_ASSOC 1 +#define L1_ITLB_2M_ENTRIES 255 +#define L1_ITLB_4K_ASSOC 1 +#define L1_ITLB_4K_ENTRIES 255 + +#define L2_DTLB_2M_ASSOC 0 /* disabled */ +#define L2_DTLB_2M_ENTRIES 0 /* disabled */ +#define L2_DTLB_4K_ASSOC 4 +#define L2_DTLB_4K_ENTRIES 512 + +#define L2_ITLB_2M_ASSOC 0 /* disabled */ +#define L2_ITLB_2M_ENTRIES 0 /* disabled */ +#define L2_ITLB_4K_ASSOC 4 +#define L2_ITLB_4K_ENTRIES 512 + + + static void x86_cpu_vendor_words2str(char *dst, uint32_t vendor1, uint32_t vendor2, uint32_t vendor3) { @@ -1950,10 +2062,12 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t ind= ex, uint32_t count, break; case 2: /* cache info: needed for Pentium Pro compatibility */ - *eax =3D 1; + *eax =3D 1; /* Number of CPUID[EAX=3D2] calls required */ *ebx =3D 0; *ecx =3D 0; - *edx =3D 0x2c307d; + *edx =3D (L1D_DESCRIPTOR << 16) | \ + (L1I_DESCRIPTOR << 8) | \ + (L2_DESCRIPTOR); break; case 4: /* cache info: needed for Core compatibility */ @@ -1964,25 +2078,37 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t ind= ex, uint32_t count, } switch (count) { case 0: /* L1 dcache info */ - *eax |=3D 0x0000121; - *ebx =3D 0x1c0003f; - *ecx =3D 0x000003f; - *edx =3D 0x0000001; + *eax |=3D CPUID_4_TYPE_DCACHE | \ + CPUID_4_LEVEL(1) | \ + CPUID_4_SELF_INIT_LEVEL; + *ebx =3D (L1D_LINE_SIZE - 1) | \ + ((L1D_PARTITIONS - 1) << 12) | \ + ((L1D_ASSOCIATIVITY - 1) << 22); + *ecx =3D L1D_SETS - 1; + *edx =3D CPUID_4_NO_INVD_SHARING; break; case 1: /* L1 icache info */ - *eax |=3D 0x0000122; - *ebx =3D 0x1c0003f; - *ecx =3D 0x000003f; - *edx =3D 0x0000001; + *eax |=3D CPUID_4_TYPE_ICACHE | \ + CPUID_4_LEVEL(1) | \ + CPUID_4_SELF_INIT_LEVEL; + *ebx =3D (L1I_LINE_SIZE - 1) | \ + ((L1I_PARTITIONS - 1) << 12) | \ + ((L1I_ASSOCIATIVITY - 1) << 22); + *ecx =3D L1I_SETS - 1; + *edx =3D CPUID_4_NO_INVD_SHARING; break; case 2: /* L2 cache info */ - *eax |=3D 0x0000143; + *eax |=3D CPUID_4_TYPE_UNIFIED | \ + CPUID_4_LEVEL(2) | \ + CPUID_4_SELF_INIT_LEVEL; if (cs->nr_threads > 1) { *eax |=3D (cs->nr_threads - 1) << 14; } - *ebx =3D 0x3c0003f; - *ecx =3D 0x0000fff; - *edx =3D 0x0000001; + *ebx =3D (L2_LINE_SIZE - 1) | \ + ((L2_PARTITIONS - 1) << 12) | \ + ((L2_ASSOCIATIVITY - 1) << 22); + *ecx =3D L2_SETS - 1; + *edx =3D CPUID_4_NO_INVD_SHARING; break; default: /* end of info */ *eax =3D 0; @@ -2102,17 +2228,31 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t ind= ex, uint32_t count, break; case 0x80000005: /* cache info (L1 cache) */ - *eax =3D 0x01ff01ff; - *ebx =3D 0x01ff01ff; - *ecx =3D 0x40020140; - *edx =3D 0x40020140; + *eax =3D (L1_DTLB_2M_ASSOC << 24) | (L1_DTLB_2M_ENTRIES << 16) |= \ + (L1_ITLB_2M_ASSOC << 8) | (L1_ITLB_2M_ENTRIES); + *ebx =3D (L1_DTLB_4K_ASSOC << 24) | (L1_DTLB_4K_ENTRIES << 16) |= \ + (L1_ITLB_4K_ASSOC << 8) | (L1_ITLB_4K_ENTRIES); + *ecx =3D (L1D_SIZE_KB_AMD << 24) | (L1D_ASSOCIATIVITY_AMD << 16)= | \ + (L1D_LINES_PER_TAG << 8) | (L1D_LINE_SIZE); + *edx =3D (L1I_SIZE_KB_AMD << 24) | (L1I_ASSOCIATIVITY_AMD << 16)= | \ + (L1I_LINES_PER_TAG << 8) | (L1I_LINE_SIZE); break; case 0x80000006: /* cache info (L2 cache) */ - *eax =3D 0; - *ebx =3D 0x42004200; - *ecx =3D 0x02008140; - *edx =3D 0; + *eax =3D (AMD_ENC_ASSOC(L2_DTLB_2M_ASSOC) << 28) | \ + (L2_DTLB_2M_ENTRIES << 16) | \ + (AMD_ENC_ASSOC(L2_ITLB_2M_ASSOC) << 12) | \ + (L2_ITLB_2M_ENTRIES); + *ebx =3D (AMD_ENC_ASSOC(L2_DTLB_4K_ASSOC) << 28) | \ + (L2_DTLB_4K_ENTRIES << 16) | \ + (AMD_ENC_ASSOC(L2_ITLB_4K_ASSOC) << 12) | \ + (L2_ITLB_4K_ENTRIES); + *ecx =3D (L2_SIZE_KB_AMD << 16) | \ + (AMD_ENC_ASSOC(L2_ASSOCIATIVITY) << 12) | \ + (L2_LINES_PER_TAG << 8) | (L2_LINE_SIZE); + *edx =3D ((L3_SIZE_KB/512) << 18) | \ + (AMD_ENC_ASSOC(L3_ASSOCIATIVITY) << 12) | \ + (L3_LINES_PER_TAG << 8) | (L3_LINE_SIZE); break; case 0x80000008: /* virtual & phys address size in low 2 bytes. */ --=20 1.8.1.4