All of lore.kernel.org
 help / color / mirror / Atom feed
From: Yinghai Lu <yhlu.kernel@gmail.com>
To: Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, Yinghai Lu <yhlu.kernel@gmail.com>
Subject: [PATCH 6/9] x86: make amd_64 have 32 bit code
Date: Sun,  7 Sep 2008 17:58:55 -0700	[thread overview]
Message-ID: <1220835538-18280-6-git-send-email-yhlu.kernel@gmail.com> (raw)
In-Reply-To: <1220835538-18280-1-git-send-email-yhlu.kernel@gmail.com>

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
---
 arch/x86/kernel/cpu/amd_64.c |  259 ++++++++++++++++++++++++++++++++++++++++--
 1 files changed, 247 insertions(+), 12 deletions(-)

diff --git a/arch/x86/kernel/cpu/amd_64.c b/arch/x86/kernel/cpu/amd_64.c
index 1d83601..32e7352 100644
--- a/arch/x86/kernel/cpu/amd_64.c
+++ b/arch/x86/kernel/cpu/amd_64.c
@@ -16,7 +16,171 @@
 
 #include "cpu.h"
 
-#ifdef CONFIG_NUMA
+#ifdef CONFIG_X86_32
+/*
+ *	B step AMD K6 before B 9730xxxx have hardware bugs that can cause
+ *	misexecution of code under Linux. Owners of such processors should
+ *	contact AMD for precise details and a CPU swap.
+ *
+ *	See	http://www.multimania.com/poulot/k6bug.html
+ *		http://www.amd.com/K6/k6docs/revgd.html
+ *
+ *	The following test is erm.. interesting. AMD neglected to up
+ *	the chip setting when fixing the bug but they also tweaked some
+ *	performance at the same time..
+ */
+
+extern void vide(void);
+__asm__(".align 4\nvide: ret");
+
+static void __cpuinit init_amd_k5(struct cpuinfo_x86 *c)
+{
+/*
+ * General Systems BIOSen alias the cpu frequency registers
+ * of the Elan at 0x000df000. Unfortuantly, one of the Linux
+ * drivers subsequently pokes it, and changes the CPU speed.
+ * Workaround : Remove the unneeded alias.
+ */
+#define CBAR		(0xfffc) /* Configuration Base Address  (32-bit) */
+#define CBAR_ENB	(0x80000000)
+#define CBAR_KEY	(0X000000CB)
+	if (c->x86_model == 9 || c->x86_model == 10) {
+		if (inl (CBAR) & CBAR_ENB)
+			outl (0 | CBAR_KEY, CBAR);
+	}
+}
+
+
+static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c)
+{
+	u32 l, h;
+	int mbytes = num_physpages >> (20-PAGE_SHIFT);
+
+	if (c->x86_model < 6) {
+		/* Based on AMD doc 20734R - June 2000 */
+		if (c->x86_model == 0) {
+			clear_cpu_cap(c, X86_FEATURE_APIC);
+			set_cpu_cap(c, X86_FEATURE_PGE);
+		}
+		return;
+	}
+
+	if (c->x86_model == 6 && c->x86_mask == 1) {
+		const int K6_BUG_LOOP = 1000000;
+		int n;
+		void (*f_vide)(void);
+		unsigned long d, d2;
+
+		printk(KERN_INFO "AMD K6 stepping B detected - ");
+
+		/*
+		 * It looks like AMD fixed the 2.6.2 bug and improved indirect
+		 * calls at the same time.
+		 */
+
+		n = K6_BUG_LOOP;
+		f_vide = vide;
+		rdtscl(d);
+		while (n--)
+			f_vide();
+		rdtscl(d2);
+		d = d2-d;
+
+		if (d > 20*K6_BUG_LOOP)
+			printk("system stability may be impaired when more than 32 MB are used.\n");
+		else
+			printk("probably OK (after B9730xxxx).\n");
+		printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n");
+	}
+
+	/* K6 with old style WHCR */
+	if (c->x86_model < 8 ||
+	   (c->x86_model == 8 && c->x86_mask < 8)) {
+		/* We can only write allocate on the low 508Mb */
+		if (mbytes > 508)
+			mbytes = 508;
+
+		rdmsr(MSR_K6_WHCR, l, h);
+		if ((l&0x0000FFFF) == 0) {
+			unsigned long flags;
+			l = (1<<0)|((mbytes/4)<<1);
+			local_irq_save(flags);
+			wbinvd();
+			wrmsr(MSR_K6_WHCR, l, h);
+			local_irq_restore(flags);
+			printk(KERN_INFO "Enabling old style K6 write allocation for %d Mb\n",
+				mbytes);
+		}
+		return;
+	}
+
+	if ((c->x86_model == 8 && c->x86_mask > 7) ||
+	     c->x86_model == 9 || c->x86_model == 13) {
+		/* The more serious chips .. */
+
+		if (mbytes > 4092)
+			mbytes = 4092;
+
+		rdmsr(MSR_K6_WHCR, l, h);
+		if ((l&0xFFFF0000) == 0) {
+			unsigned long flags;
+			l = ((mbytes>>2)<<22)|(1<<16);
+			local_irq_save(flags);
+			wbinvd();
+			wrmsr(MSR_K6_WHCR, l, h);
+			local_irq_restore(flags);
+			printk(KERN_INFO "Enabling new style K6 write allocation for %d Mb\n",
+				mbytes);
+		}
+
+		return;
+	}
+
+	if (c->x86_model == 10) {
+		/* AMD Geode LX is model 10 */
+		/* placeholder for any needed mods */
+		return;
+	}
+}
+
+static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c)
+{
+	u32 l, h;
+
+	/*
+	 * Bit 15 of Athlon specific MSR 15, needs to be 0
+	 * to enable SSE on Palomino/Morgan/Barton CPU's.
+	 * If the BIOS didn't enable it already, enable it here.
+	 */
+	if (c->x86_model >= 6 && c->x86_model <= 10) {
+		if (!cpu_has(c, X86_FEATURE_XMM)) {
+			printk(KERN_INFO "Enabling disabled K7/SSE Support.\n");
+			rdmsr(MSR_K7_HWCR, l, h);
+			l &= ~0x00008000;
+			wrmsr(MSR_K7_HWCR, l, h);
+			set_cpu_cap(c, X86_FEATURE_XMM);
+		}
+	}
+
+	/*
+	 * It's been determined by AMD that Athlons since model 8 stepping 1
+	 * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx
+	 * As per AMD technical note 27212 0.2
+	 */
+	if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) {
+		rdmsr(MSR_K7_CLK_CTL, l, h);
+		if ((l & 0xfff00000) != 0x20000000) {
+			printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l,
+				((l & 0x000fffff)|0x20000000));
+			wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
+		}
+	}
+
+	set_cpu_cap(c, X86_FEATURE_K7);
+}
+#endif
+
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
 static int __cpuinit nearby_node(int apicid)
 {
 	int i, node;
@@ -41,7 +205,7 @@ static int __cpuinit nearby_node(int apicid)
  */
 static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
 {
-#ifdef CONFIG_SMP
+#ifdef CONFIG_X86_HT
 	unsigned bits;
 
 	bits = c->x86_coreid_bits;
@@ -55,7 +219,7 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
 
 static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c)
 {
-#ifdef CONFIG_NUMA
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
 	int cpu = smp_processor_id();
 	int node;
 	unsigned apicid = hard_smp_processor_id();
@@ -91,7 +255,7 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c)
 
 static void __cpuinit early_init_amd_mc(struct cpuinfo_x86 *c)
 {
-#ifdef CONFIG_SMP
+#ifdef CONFIG_X86_HT
 	unsigned bits, ecx;
 
 	/* Multi core CPU? */
@@ -112,7 +276,6 @@ static void __cpuinit early_init_amd_mc(struct cpuinfo_x86 *c)
 	}
 
 	c->x86_coreid_bits = bits;
-
 #endif
 }
 
@@ -124,15 +287,21 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
 	if (c->x86_power & (1<<8))
 		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
 
+#ifdef CONFIG_X86_64
 	set_cpu_cap(c, X86_FEATURE_SYSCALL32);
+#else
+	/*  Set MTRR capability flag if appropriate */
+	if (c->x86 == 5)
+		if (c->x86_model == 13 || c->x86_model == 9 ||
+		    (c->x86_model == 8 && c->x86_mask >= 8))
+			set_cpu_cap(c, X86_FEATURE_K6_MTRR);
+#endif
 }
 
 static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 {
-	unsigned level;
-
 #ifdef CONFIG_SMP
-	unsigned long value;
+	unsigned long long value;
 
 	/*
 	 * Disable TLB flush filter by setting HWCR.FFDIS on K8
@@ -142,26 +311,55 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 	 * Errata 122 for all steppings (F+ have it disabled by default)
 	 */
 	if (c->x86 == 0xf) {
-		rdmsrl(MSR_K8_HWCR, value);
+		rdmsrl(MSR_K7_HWCR, value);
 		value |= 1 << 6;
-		wrmsrl(MSR_K8_HWCR, value);
+		wrmsrl(MSR_K7_HWCR, value);
 	}
 #endif
 
 	early_init_amd(c);
 
-	/* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
-	   3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
+	/*
+	 * Bit 31 in normal CPUID used for nonstandard 3DNow ID;
+	 * 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway
+	 */
 	clear_cpu_cap(c, 0*32+31);
 
+#ifdef CONFIG_X86_64
 	/* On C+ stepping K8 rep microcode works well for copy/memset */
 	if (c->x86 == 0xf) {
+		u32 level;
+
 		level = cpuid_eax(1);
 		if((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
 			set_cpu_cap(c, X86_FEATURE_REP_GOOD);
 	}
 	if (c->x86 == 0x10 || c->x86 == 0x11)
 		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
+#else
+
+	/*
+	 *	FIXME: We should handle the K5 here. Set up the write
+	 *	range and also turn on MSR 83 bits 4 and 31 (write alloc,
+	 *	no bus pipeline)
+	 */
+
+	switch (c->x86) {
+	case 4:
+		init_amd_k5(c);
+		break;
+	case 5:
+		init_amd_k6(c);
+		break;
+	case 6: /* An Athlon/Duron */
+		init_amd_k7(c);
+		break;
+	}
+
+	/* K6s reports MCEs but don't actually have all the MSRs */
+	if (c->x86 < 6)
+		clear_cpu_cap(c, X86_FEATURE_MCE);
+#endif
 
 	/* Enable workaround for FXSAVE leak */
 	if (c->x86 >= 6)
@@ -176,6 +374,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 			break;
 		}
 	}
+
 	display_cacheinfo(c);
 
 	/* Multi core CPU? */
@@ -184,6 +383,10 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 		srat_detect_node(c);
 	}
 
+#ifdef CONFIG_X86_32
+	detect_ht(c);
+#endif
+
 	if (c->extended_cpuid_level >= 0x80000006) {
 		if ((c->x86 >= 0x0f) && (cpuid_edx(0x80000006) & 0xf000))
 			num_cache_leaves = 4;
@@ -199,6 +402,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 		set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC);
 	}
 
+#ifdef CONFIG_X86_64
 	if (c->x86 == 0x10) {
 		/* do this for boot cpu */
 		if (c == &boot_cpu_data)
@@ -225,11 +429,42 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 			set_memory_4k((unsigned long)__va(tseg), 1);
 		}
 	}
+#endif
+}
+
+#ifdef CONFIG_X86_32
+static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int size)
+{
+	/* AMD errata T13 (order #21922) */
+	if ((c->x86 == 6)) {
+		if (c->x86_model == 3 && c->x86_mask == 0)	/* Duron Rev A0 */
+			size = 64;
+		if (c->x86_model == 4 &&
+		    (c->x86_mask == 0 || c->x86_mask == 1))	/* Tbird rev A1/A2 */
+			size = 256;
+	}
+	return size;
 }
+#endif
 
 static struct cpu_dev amd_cpu_dev __cpuinitdata = {
 	.c_vendor	= "AMD",
 	.c_ident	= { "AuthenticAMD" },
+#ifdef CONFIG_X86_32
+	.c_models = {
+		{ .vendor = X86_VENDOR_AMD, .family = 4, .model_names =
+		  {
+			  [3] = "486 DX/2",
+			  [7] = "486 DX/2-WB",
+			  [8] = "486 DX/4",
+			  [9] = "486 DX/4-WB",
+			  [14] = "Am5x86-WT",
+			  [15] = "Am5x86-WB"
+		  }
+		},
+	},
+	.c_size_cache	= amd_size_cache,
+#endif
 	.c_early_init   = early_init_amd,
 	.c_init		= init_amd,
 	.c_x86_vendor	= X86_VENDOR_AMD,
-- 
1.5.4.5


  parent reply	other threads:[~2008-09-08  1:00 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-09-08  0:58 [PATCH 1/9] x86: cpu make amd.c more like amd_64.c v2 Yinghai Lu
2008-09-08  0:58 ` [PATCH 2/9] x86: remove duplicated force_mwait Yinghai Lu
2008-09-08  0:58 ` [PATCH 3/9] x86: add srat_detect_node for amd64 Yinghai Lu
2008-09-08  0:58 ` [PATCH 4/9] x86: merge header in amd/64.c Yinghai Lu
2008-09-08  0:58 ` [PATCH 5/9] x86: make amd.c have 64bit support code Yinghai Lu
2008-09-08  0:58 ` Yinghai Lu [this message]
2008-09-08  0:58 ` [PATCH 7/9] x86: make 64 bit to use amd.c Yinghai Lu
2008-09-08 13:33   ` Ingo Molnar
2008-09-08  0:58 ` [PATCH 8/9] x86: lapic address print out like io apic addr Yinghai Lu
2008-09-08 13:45   ` Ingo Molnar
2008-09-08  0:58 ` [PATCH 9/9] x86: little clean up intel.c/intel_64.c Yinghai Lu
2008-09-08 13:46   ` Ingo Molnar

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=1220835538-18280-6-git-send-email-yhlu.kernel@gmail.com \
    --to=yhlu.kernel@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=tglx@linutronix.de \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.