x86, microcode, AMD: Fix early ucode loading

The original idea to use the microcode cache for the APs doesn't pan out
because we do memory allocation there very early and with IRQs disabled
and we don't want to involve GFP_ATOMIC allocations. Not if it can be
helped.

Thus, extend the caching of the BSP patch approach to the APs and
iterate over the ucode in the initrd instead of using the cache. We
still save the relevant patches to it but later, right before we
jettison the initrd.

While at it, fix early ucode loading on 32-bit too.

Signed-off-by: Borislav Petkov <bp@suse.de>
Tested-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@amd.com>

arch/x86/include/asm/microcode_amd.h

@@ -61,11 +61,10 @@ extern int __apply_microcode_amd(struct microcode_amd *mc_amd);
 extern int apply_microcode_amd(int cpu);
 extern enum ucode_state load_microcode_amd(u8 family, const u8 *data, size_t size);
 
+#define PATCH_MAX_SIZE PAGE_SIZE
+extern u8 amd_ucode_patch[PATCH_MAX_SIZE];
+
 #ifdef CONFIG_MICROCODE_AMD_EARLY
-#ifdef CONFIG_X86_32
-#define MPB_MAX_SIZE PAGE_SIZE
-extern u8 amd_bsp_mpb[MPB_MAX_SIZE];
-#endif
 extern void __init load_ucode_amd_bsp(void);
 extern void load_ucode_amd_ap(void);
 extern int __init save_microcode_in_initrd_amd(void);

arch/x86/kernel/microcode_amd.c

@@ -182,10 +182,10 @@ int __apply_microcode_amd(struct microcode_amd *mc_amd)
 {
 	u32 rev, dummy;
 
-	wrmsrl(MSR_AMD64_PATCH_LOADER, (u64)(long)&mc_amd->hdr.data_code);
+	native_wrmsrl(MSR_AMD64_PATCH_LOADER, (u64)(long)&mc_amd->hdr.data_code);
 
 	/* verify patch application was successful */
-	rdmsr(MSR_AMD64_PATCH_LEVEL, rev, dummy);
+	native_rdmsr(MSR_AMD64_PATCH_LEVEL, rev, dummy);
 	if (rev != mc_amd->hdr.patch_id)
 		return -1;
 
@@ -332,6 +332,9 @@ static int verify_and_add_patch(u8 family, u8 *fw, unsigned int leftover)
 	patch->patch_id  = mc_hdr->patch_id;
 	patch->equiv_cpu = proc_id;
 
+	pr_debug("%s: Added patch_id: 0x%08x, proc_id: 0x%04x\n",
+		 __func__, patch->patch_id, proc_id);
+
 	/* ... and add to cache. */
 	update_cache(patch);
 
@@ -390,9 +393,9 @@ enum ucode_state load_microcode_amd(u8 family, const u8 *data, size_t size)
 	if (cpu_data(smp_processor_id()).cpu_index == boot_cpu_data.cpu_index) {
 		struct ucode_patch *p = find_patch(smp_processor_id());
 		if (p) {
-			memset(amd_bsp_mpb, 0, MPB_MAX_SIZE);
-			memcpy(amd_bsp_mpb, p->data, min_t(u32, ksize(p->data),
-							   MPB_MAX_SIZE));
+			memset(amd_ucode_patch, 0, PATCH_MAX_SIZE);
+			memcpy(amd_ucode_patch, p->data, min_t(u32, ksize(p->data),
+							       PATCH_MAX_SIZE));
 		}
 	}
 #endif

arch/x86/kernel/microcode_amd_early.c

@@ -2,6 +2,7 @@
  * Copyright (C) 2013 Advanced Micro Devices, Inc.
  *
  * Author: Jacob Shin <jacob.shin@amd.com>
+ * Fixes: Borislav Petkov <bp@suse.de>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -15,10 +16,18 @@
 #include <asm/setup.h>
 #include <asm/microcode_amd.h>
 
-static bool ucode_loaded;
+/*
+ * This points to the current valid container of microcode patches which we will
+ * save from the initrd before jettisoning its contents.
+ */
+static u8 *container;
+static size_t container_size;
+
 static u32 ucode_new_rev;
-static unsigned long ucode_offset;
-static size_t ucode_size;
+u8 amd_ucode_patch[PATCH_MAX_SIZE];
+static u16 this_equiv_id;
+
+struct cpio_data ucode_cpio;
 
 /*
  * Microcode patch container file is prepended to the initrd in cpio format.
@@ -32,9 +41,6 @@ static struct cpio_data __init find_ucode_in_initrd(void)
 	char *path;
 	void *start;
 	size_t size;
-	unsigned long *uoffset;
-	size_t *usize;
-	struct cpio_data cd;
 
 #ifdef CONFIG_X86_32
 	struct boot_params *p;
@@ -47,30 +53,50 @@ static struct cpio_data __init find_ucode_in_initrd(void)
 	path    = (char *)__pa_nodebug(ucode_path);
 	start   = (void *)p->hdr.ramdisk_image;
 	size    = p->hdr.ramdisk_size;
-	uoffset = (unsigned long *)__pa_nodebug(&ucode_offset);
-	usize   = (size_t *)__pa_nodebug(&ucode_size);
 #else
 	path    = ucode_path;
 	start   = (void *)(boot_params.hdr.ramdisk_image + PAGE_OFFSET);
 	size    = boot_params.hdr.ramdisk_size;
-	uoffset = &ucode_offset;
-	usize   = &ucode_size;
 #endif
 
-	cd = find_cpio_data(path, start, size, &offset);
-	if (!cd.data)
-		return cd;
+	return find_cpio_data(path, start, size, &offset);
+}
 
-	if (*(u32 *)cd.data != UCODE_MAGIC) {
-		cd.data = NULL;
-		cd.size = 0;
-		return cd;
+static size_t compute_container_size(u8 *data, u32 total_size)
+{
+	size_t size = 0;
+	u32 *header = (u32 *)data;
+
+	if (header[0] != UCODE_MAGIC ||
+	    header[1] != UCODE_EQUIV_CPU_TABLE_TYPE || /* type */
+	    header[2] == 0)                            /* size */
+		return size;
+
+	size = header[2] + CONTAINER_HDR_SZ;
+	total_size -= size;
+	data += size;
+
+	while (total_size) {
+		u16 patch_size;
+
+		header = (u32 *)data;
+
+		if (header[0] != UCODE_UCODE_TYPE)
+			break;
+
+		/*
+		 * Sanity-check patch size.
+		 */
+		patch_size = header[1];
+		if (patch_size > PATCH_MAX_SIZE)
+			break;
+
+		size	   += patch_size + SECTION_HDR_SIZE;
+		data	   += patch_size + SECTION_HDR_SIZE;
+		total_size -= patch_size + SECTION_HDR_SIZE;
 	}
 
-	*uoffset = (u8 *)cd.data - (u8 *)start;
-	*usize   = cd.size;
-
-	return cd;
+	return size;
 }
 
 /*
@@ -85,23 +111,22 @@ static struct cpio_data __init find_ucode_in_initrd(void)
 static void apply_ucode_in_initrd(void *ucode, size_t size)
 {
 	struct equiv_cpu_entry *eq;
+	size_t *cont_sz;
 	u32 *header;
-	u8  *data;
+	u8  *data, **cont;
 	u16 eq_id = 0;
 	int offset, left;
-	u32 rev, eax;
+	u32 rev, eax, ebx, ecx, edx;
 	u32 *new_rev;
-	unsigned long *uoffset;
-	size_t *usize;
 
 #ifdef CONFIG_X86_32
 	new_rev = (u32 *)__pa_nodebug(&ucode_new_rev);
-	uoffset = (unsigned long *)__pa_nodebug(&ucode_offset);
-	usize   = (size_t *)__pa_nodebug(&ucode_size);
+	cont_sz = (size_t *)__pa_nodebug(&container_size);
+	cont	= (u8 **)__pa_nodebug(&container);
 #else
 	new_rev = &ucode_new_rev;
-	uoffset = &ucode_offset;
-	usize   = &ucode_size;
+	cont_sz = &container_size;
+	cont	= &container;
 #endif
 
 	data   = ucode;
@@ -109,23 +134,37 @@ static void apply_ucode_in_initrd(void *ucode, size_t size)
 	header = (u32 *)data;
 
 	/* find equiv cpu table */
-
-	if (header[1] != UCODE_EQUIV_CPU_TABLE_TYPE || /* type */
+	if (header[0] != UCODE_MAGIC ||
+	    header[1] != UCODE_EQUIV_CPU_TABLE_TYPE || /* type */
 	    header[2] == 0)                            /* size */
 		return;
 
-	eax = cpuid_eax(0x00000001);
+	eax = 0x00000001;
+	ecx = 0;
+	native_cpuid(&eax, &ebx, &ecx, &edx);
 
 	while (left > 0) {
 		eq = (struct equiv_cpu_entry *)(data + CONTAINER_HDR_SZ);
 
+		*cont = data;
+
+		/* Advance past the container header */
 		offset = header[2] + CONTAINER_HDR_SZ;
 		data  += offset;
 		left  -= offset;
 
 		eq_id = find_equiv_id(eq, eax);
-		if (eq_id)
+		if (eq_id) {
+			this_equiv_id = eq_id;
+			*cont_sz = compute_container_size(*cont, left + offset);
+
+			/*
+			 * truncate how much we need to iterate over in the
+			 * ucode update loop below
+			 */
+			left = *cont_sz - offset;
 			break;
+		}
 
 		/*
 		 * support multiple container files appended together. if this
@@ -145,19 +184,18 @@ static void apply_ucode_in_initrd(void *ucode, size_t size)
 
 		/* mark where the next microcode container file starts */
 		offset    = data - (u8 *)ucode;
-		*uoffset += offset;
-		*usize   -= offset;
 		ucode     = data;
 	}
 
 	if (!eq_id) {
-		*usize = 0;
+		*cont = NULL;
+		*cont_sz = 0;
 		return;
 	}
 
 	/* find ucode and update if needed */
 
-	rdmsr(MSR_AMD64_PATCH_LEVEL, rev, eax);
+	native_rdmsr(MSR_AMD64_PATCH_LEVEL, rev, eax);
 
 	while (left > 0) {
 		struct microcode_amd *mc;
@@ -168,73 +206,83 @@ static void apply_ucode_in_initrd(void *ucode, size_t size)
 			break;
 
 		mc = (struct microcode_amd *)(data + SECTION_HDR_SIZE);
-		if (eq_id == mc->hdr.processor_rev_id && rev < mc->hdr.patch_id)
-			if (__apply_microcode_amd(mc) == 0) {
+
+		if (eq_id == mc->hdr.processor_rev_id && rev < mc->hdr.patch_id) {
+
+			if (!__apply_microcode_amd(mc)) {
 				rev = mc->hdr.patch_id;
 				*new_rev = rev;
+
+				/* save ucode patch */
+				memcpy(amd_ucode_patch, mc,
+				       min_t(u32, header[1], PATCH_MAX_SIZE));
 			}
+		}
 
 		offset  = header[1] + SECTION_HDR_SIZE;
 		data   += offset;
 		left   -= offset;
 	}
-
-	/* mark where this microcode container file ends */
-	offset  = *usize - (data - (u8 *)ucode);
-	*usize -= offset;
-
-	if (!(*new_rev))
-		*usize = 0;
 }
 
 void __init load_ucode_amd_bsp(void)
 {
-	struct cpio_data cd = find_ucode_in_initrd();
-	if (!cd.data)
+	struct cpio_data cp;
+	void **data;
+	size_t *size;
+
+#ifdef CONFIG_X86_32
+	data =  (void **)__pa_nodebug(&ucode_cpio.data);
+	size = (size_t *)__pa_nodebug(&ucode_cpio.size);
+#else
+	data = &ucode_cpio.data;
+	size = &ucode_cpio.size;
+#endif
+
+	cp = find_ucode_in_initrd();
+	if (!cp.data)
 		return;
 
-	apply_ucode_in_initrd(cd.data, cd.size);
+	*data = cp.data;
+	*size = cp.size;
+
+	apply_ucode_in_initrd(cp.data, cp.size);
 }
 
 #ifdef CONFIG_X86_32
-u8 amd_bsp_mpb[MPB_MAX_SIZE];
-
 /*
  * On 32-bit, since AP's early load occurs before paging is turned on, we
  * cannot traverse cpu_equiv_table and pcache in kernel heap memory. So during
  * cold boot, AP will apply_ucode_in_initrd() just like the BSP. During
- * save_microcode_in_initrd_amd() BSP's patch is copied to amd_bsp_mpb, which
- * is used upon resume from suspend.
+ * save_microcode_in_initrd_amd() BSP's patch is copied to amd_ucode_patch,
+ * which is used upon resume from suspend.
  */
 void load_ucode_amd_ap(void)
 {
 	struct microcode_amd *mc;
-	unsigned long *initrd;
-	unsigned long *uoffset;
 	size_t *usize;
-	void *ucode;
+	void **ucode;
 
-	mc = (struct microcode_amd *)__pa(amd_bsp_mpb);
+	mc = (struct microcode_amd *)__pa(amd_ucode_patch);
 	if (mc->hdr.patch_id && mc->hdr.processor_rev_id) {
 		__apply_microcode_amd(mc);
 		return;
 	}
 
-	initrd  = (unsigned long *)__pa(&initrd_start);
-	uoffset = (unsigned long *)__pa(&ucode_offset);
-	usize   = (size_t *)__pa(&ucode_size);
+	ucode = (void *)__pa_nodebug(&container);
+	usize = (size_t *)__pa_nodebug(&container_size);
 
-	if (!*usize || !*initrd)
+	if (!*ucode || !*usize)
 		return;
 
-	ucode = (void *)((unsigned long)__pa(*initrd) + *uoffset);
-	apply_ucode_in_initrd(ucode, *usize);
+	apply_ucode_in_initrd(*ucode, *usize);
 }
 
 static void __init collect_cpu_sig_on_bsp(void *arg)
 {
 	unsigned int cpu = smp_processor_id();
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
 	uci->cpu_sig.sig = cpuid_eax(0x00000001);
 }
 #else
@@ -242,36 +290,54 @@ void load_ucode_amd_ap(void)
 {
 	unsigned int cpu = smp_processor_id();
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+	struct equiv_cpu_entry *eq;
+	struct microcode_amd *mc;
 	u32 rev, eax;
+	u16 eq_id;
+
+	/* Exit if called on the BSP. */
+	if (!cpu)
+		return;
+
+	if (!container)
+		return;
 
 	rdmsr(MSR_AMD64_PATCH_LEVEL, rev, eax);
-	eax = cpuid_eax(0x00000001);
 
 	uci->cpu_sig.rev = rev;
 	uci->cpu_sig.sig = eax;
 
-	if (cpu && !ucode_loaded) {
-		void *ucode;
+	eax = cpuid_eax(0x00000001);
+	eq  = (struct equiv_cpu_entry *)(container + CONTAINER_HDR_SZ);
 
-		if (!ucode_size || !initrd_start)
+	eq_id = find_equiv_id(eq, eax);
+	if (!eq_id)
+		return;
+
+	if (eq_id == this_equiv_id) {
+		mc = (struct microcode_amd *)amd_ucode_patch;
+
+		if (mc && rev < mc->hdr.patch_id) {
+			if (!__apply_microcode_amd(mc))
+				ucode_new_rev = mc->hdr.patch_id;
+		}
+
+	} else {
+		if (!ucode_cpio.data)
 			return;
 
-		ucode = (void *)(initrd_start + ucode_offset);
-		eax   = ((eax >> 8) & 0xf) + ((eax >> 20) & 0xff);
-		if (load_microcode_amd(eax, ucode, ucode_size) != UCODE_OK)
-			return;
-
-		ucode_loaded = true;
+		/*
+		 * AP has a different equivalence ID than BSP, looks like
+		 * mixed-steppings silicon so go through the ucode blob anew.
+		 */
+		apply_ucode_in_initrd(ucode_cpio.data, ucode_cpio.size);
 	}
-
-	apply_microcode_amd(cpu);
 }
 #endif
 
 int __init save_microcode_in_initrd_amd(void)
 {
 	enum ucode_state ret;
-	void *ucode;
 	u32 eax;
 
 #ifdef CONFIG_X86_32
@@ -280,22 +346,35 @@ int __init save_microcode_in_initrd_amd(void)
 
 	if (!uci->cpu_sig.sig)
 		smp_call_function_single(bsp, collect_cpu_sig_on_bsp, NULL, 1);
+
+	/*
+	 * Take into account the fact that the ramdisk might get relocated
+	 * and therefore we need to recompute the container's position in
+	 * virtual memory space.
+	 */
+	container = (u8 *)(__va((u32)relocated_ramdisk) +
+			   ((u32)container - boot_params.hdr.ramdisk_image));
 #endif
 	if (ucode_new_rev)
 		pr_info("microcode: updated early to new patch_level=0x%08x\n",
 			ucode_new_rev);
 
-	if (ucode_loaded || !ucode_size || !initrd_start)
-		return 0;
+	if (!container)
+		return -EINVAL;
 
-	ucode = (void *)(initrd_start + ucode_offset);
 	eax   = cpuid_eax(0x00000001);
 	eax   = ((eax >> 8) & 0xf) + ((eax >> 20) & 0xff);
 
-	ret = load_microcode_amd(eax, ucode, ucode_size);
+	ret = load_microcode_amd(eax, container, container_size);
 	if (ret != UCODE_OK)
 		return -EINVAL;
 
-	ucode_loaded = true;
+	/*
+	 * This will be freed any msec now, stash patches for the current
+	 * family and switch to patch cache for cpu hotplug, etc later.
+	 */
+	container = NULL;
+	container_size = 0;
+
 	return 0;
 }