Merge tag 'x86-apic-2023-04-24' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 APIC updates from Thomas Gleixner:

 - Fix the incorrect handling of atomic offset updates in
   reserve_eilvt_offset()

   The check for the return value of atomic_cmpxchg() is not compared
   against the old value, it is compared against the new value, which
   makes it two round on success.

   Convert it to atomic_try_cmpxchg() which does the right thing.

 - Handle IO/APIC less systems correctly

   When IO/APIC is not advertised by ACPI then the computation of the
   lower bound for dynamically allocated interrupts like MSI goes wrong.

   This lower bound is used to exclude the IO/APIC legacy GSI space as
   that must stay reserved for the legacy interrupts.

   In case that the system, e.g. VM, does not advertise an IO/APIC the
   lower bound stays at 0.

   0 is an invalid interrupt number except for the legacy timer
   interrupt on x86. The return value is unchecked in the core code, so
   it ends up to allocate interrupt number 0 which is subsequently
   considered to be invalid by the caller, e.g. the MSI allocation code.

   A similar problem was already cured for device tree based systems
   years ago, but that missed - or did not envision - the zero IO/APIC
   case.

   Consolidate the zero check and return the provided "from" argument to
   the core code call site, which is guaranteed to be greater than 0.

 - Simplify the X2APIC cluster CPU mask logic for CPU hotplug

   Per cluster CPU masks are required for X2APIC in cluster mode to
   determine the correct cluster for a target CPU when calculating the
   destination for IPIs

   These masks are established when CPUs are borught up. The first CPU
   in a cluster must allocate a new cluster CPU mask. As this happens
   during the early startup of a CPU, where memory allocations cannot be
   done, the mask has to be allocated by the control CPU.

   The current implementation allocates a clustermask just in case and
   if the to be brought up CPU is the first in a cluster the CPU takes
   over this allocation from a global pointer.

   This works nicely in the fully serialized CPU bringup scenario which
   is used today, but would fail completely for parallel bringup of
   CPUs.

   The cluster association of a CPU can be computed from the APIC ID
   which is enumerated by ACPI/MADT.

   So the cluster CPU masks can be preallocated and associated upfront
   and the upcoming CPUs just need to set their corresponding bit.

   Aside of preparing for parallel bringup this is a valuable
   simplification on its own.

 - Remove global variables which control the early startup of secondary
   CPUs on 64-bit

   The only information which is needed by a starting CPU is the Linux
   CPU number. The CPU number allows it to retrieve the rest of the
   required data from already existing per CPU storage.

   So instead of initial_stack, early_gdt_desciptor and initial_gs
   provide a new variable smpboot_control which contains the Linux CPU
   number for now. The starting CPU can retrieve and compute all
   required information for startup from there.

   Aside of being a cleanup, this is also preparing for parallel CPU
   bringup, where starting CPUs will look up their Linux CPU number via
   the APIC ID, when smpboot_control has the corresponding control bit
   set.

 - Make cc_vendor globally accesible

   Subsequent parallel bringup changes require access to cc_vendor
   because confidental computing platforms need special treatment in the
   early startup phase vs. CPUID and APCI ID readouts.

   The change makes cc_vendor global and provides stub accessors in case
   that CONFIG_ARCH_HAS_CC_PLATFORM is not set.

   This was merged from the x86/cc branch in anticipation of further
   parallel bringup commits which require access to cc_vendor. Due to
   late discoveries of fundamental issue with those patches these
   commits never happened.

   The merge commit is unfortunately in the middle of the APIC commits
   so unraveling it would have required a rebase or revert. As the
   parallel bringup seems to be well on its way for 6.5 this would be
   just pointless churn. As the commit does not contain any functional
   change it's not a risk to keep it.

* tag 'x86-apic-2023-04-24' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/ioapic: Don't return 0 from arch_dynirq_lower_bound()
  x86/apic: Fix atomic update of offset in reserve_eilvt_offset()
  x86/coco: Export cc_vendor
  x86/smpboot: Reference count on smpboot_setup_warm_reset_vector()
  x86/smpboot: Remove initial_gs
  x86/smpboot: Remove early_gdt_descr on 64-bit
  x86/smpboot: Remove initial_stack on 64-bit
  x86/apic/x2apic: Allow CPU cluster_mask to be populated in parallel

Author: torvalds

arch/x86/coco/core.c

@@ -13,7 +13,7 @@
 #include <asm/coco.h>
 #include <asm/processor.h>
 
-static enum cc_vendor vendor __ro_after_init;
+enum cc_vendor cc_vendor __ro_after_init;
 static u64 cc_mask __ro_after_init;
 
 static bool intel_cc_platform_has(enum cc_attr attr)
@@ -99,7 +99,7 @@ static bool amd_cc_platform_has(enum cc_attr attr)
 
 bool cc_platform_has(enum cc_attr attr)
 {
-	switch (vendor) {
+	switch (cc_vendor) {
 	case CC_VENDOR_AMD:
 		return amd_cc_platform_has(attr);
 	case CC_VENDOR_INTEL:
@@ -119,7 +119,7 @@ u64 cc_mkenc(u64 val)
 	 * - for AMD, bit *set* means the page is encrypted
 	 * - for AMD with vTOM and for Intel, *clear* means encrypted
 	 */
-	switch (vendor) {
+	switch (cc_vendor) {
 	case CC_VENDOR_AMD:
 		if (sev_status & MSR_AMD64_SNP_VTOM)
 			return val & ~cc_mask;
@@ -135,7 +135,7 @@ u64 cc_mkenc(u64 val)
 u64 cc_mkdec(u64 val)
 {
 	/* See comment in cc_mkenc() */
-	switch (vendor) {
+	switch (cc_vendor) {
 	case CC_VENDOR_AMD:
 		if (sev_status & MSR_AMD64_SNP_VTOM)
 			return val | cc_mask;
@@ -149,11 +149,6 @@ u64 cc_mkdec(u64 val)
 }
 EXPORT_SYMBOL_GPL(cc_mkdec);
 
-__init void cc_set_vendor(enum cc_vendor v)
-{
-	vendor = v;
-}
-
 __init void cc_set_mask(u64 mask)
 {
 	cc_mask = mask;

arch/x86/include/asm/coco.h

@@ -10,13 +10,30 @@ enum cc_vendor {
 	CC_VENDOR_INTEL,
 };
 
-void cc_set_vendor(enum cc_vendor v);
-void cc_set_mask(u64 mask);
-
 #ifdef CONFIG_ARCH_HAS_CC_PLATFORM
+extern enum cc_vendor cc_vendor;
+
+static inline enum cc_vendor cc_get_vendor(void)
+{
+	return cc_vendor;
+}
+
+static inline void cc_set_vendor(enum cc_vendor vendor)
+{
+	cc_vendor = vendor;
+}
+
+void cc_set_mask(u64 mask);
 u64 cc_mkenc(u64 val);
 u64 cc_mkdec(u64 val);
 #else
+static inline enum cc_vendor cc_get_vendor(void)
+{
+	return CC_VENDOR_NONE;
+}
+
+static inline void cc_set_vendor(enum cc_vendor vendor) { }
+
 static inline u64 cc_mkenc(u64 val)
 {
 	return val;

arch/x86/include/asm/processor.h

@@ -647,7 +647,11 @@ static inline void spin_lock_prefetch(const void *x)
 #define KSTK_ESP(task)		(task_pt_regs(task)->sp)
 
 #else
-#define INIT_THREAD { }
+extern unsigned long __end_init_task[];
+
+#define INIT_THREAD {							    \
+	.sp	= (unsigned long)&__end_init_task - sizeof(struct pt_regs), \
+}
 
 extern unsigned long KSTK_ESP(struct task_struct *task);
 

arch/x86/include/asm/realmode.h

@@ -59,7 +59,6 @@ extern struct real_mode_header *real_mode_header;
 extern unsigned char real_mode_blob_end[];
 
 extern unsigned long initial_code;
-extern unsigned long initial_gs;
 extern unsigned long initial_stack;
 #ifdef CONFIG_AMD_MEM_ENCRYPT
 extern unsigned long initial_vc_handler;

arch/x86/include/asm/smp.h

@@ -199,5 +199,8 @@ extern void nmi_selftest(void);
 #define nmi_selftest() do { } while (0)
 #endif
 
-#endif /* __ASSEMBLY__ */
+extern unsigned int smpboot_control;
+
+#endif /* !__ASSEMBLY__ */
+
 #endif /* _ASM_X86_SMP_H */

arch/x86/kernel/acpi/sleep.c

@@ -111,13 +111,26 @@ int x86_acpi_suspend_lowlevel(void)
 	saved_magic = 0x12345678;
 #else /* CONFIG_64BIT */
 #ifdef CONFIG_SMP
-	initial_stack = (unsigned long)temp_stack + sizeof(temp_stack);
-	early_gdt_descr.address =
-			(unsigned long)get_cpu_gdt_rw(smp_processor_id());
-	initial_gs = per_cpu_offset(smp_processor_id());
+	/*
+	 * As each CPU starts up, it will find its own stack pointer
+	 * from its current_task->thread.sp. Typically that will be
+	 * the idle thread for a newly-started AP, or even the boot
+	 * CPU which will find it set to &init_task in the static
+	 * per-cpu data.
+	 *
+	 * Make the resuming CPU use the temporary stack at startup
+	 * by setting current->thread.sp to point to that. The true
+	 * %rsp will be restored with the rest of the CPU context,
+	 * by do_suspend_lowlevel(). And unwinders don't care about
+	 * the abuse of ->thread.sp because it's a dead variable
+	 * while the thread is running on the CPU anyway; the true
+	 * value is in the actual %rsp register.
+	 */
+	current->thread.sp = (unsigned long)temp_stack + sizeof(temp_stack);
+	smpboot_control = smp_processor_id();
 #endif
 	initial_code = (unsigned long)wakeup_long64;
-       saved_magic = 0x123456789abcdef0L;
+	saved_magic = 0x123456789abcdef0L;
 #endif /* CONFIG_64BIT */
 
 	/*

arch/x86/kernel/apic/apic.c

@@ -422,10 +422,9 @@ static unsigned int reserve_eilvt_offset(int offset, unsigned int new)
 		if (vector && !eilvt_entry_is_changeable(vector, new))
 			/* may not change if vectors are different */
 			return rsvd;
-		rsvd = atomic_cmpxchg(&eilvt_offsets[offset], rsvd, new);
-	} while (rsvd != new);
+	} while (!atomic_try_cmpxchg(&eilvt_offsets[offset], &rsvd, new));
 
-	rsvd &= ~APIC_EILVT_MASKED;
+	rsvd = new & ~APIC_EILVT_MASKED;
 	if (rsvd && rsvd != vector)
 		pr_info("LVT offset %d assigned for vector 0x%02x\n",
 			offset, rsvd);

arch/x86/kernel/apic/io_apic.c

@@ -2478,17 +2478,21 @@ static int io_apic_get_redir_entries(int ioapic)
 
 unsigned int arch_dynirq_lower_bound(unsigned int from)
 {
+	unsigned int ret;
+
 	/*
 	 * dmar_alloc_hwirq() may be called before setup_IO_APIC(), so use
 	 * gsi_top if ioapic_dynirq_base hasn't been initialized yet.
 	 */
-	if (!ioapic_initialized)
-		return gsi_top;
+	ret = ioapic_dynirq_base ? : gsi_top;
+
 	/*
-	 * For DT enabled machines ioapic_dynirq_base is irrelevant and not
-	 * updated. So simply return @from if ioapic_dynirq_base == 0.
+	 * For DT enabled machines ioapic_dynirq_base is irrelevant and
+	 * always 0. gsi_top can be 0 if there is no IO/APIC registered.
+	 * 0 is an invalid interrupt number for dynamic allocations. Return
+	 * @from instead.
 	 */
-	return ioapic_dynirq_base ? : from;
+	return ret ? : from;
 }
 
 #ifdef CONFIG_X86_32

arch/x86/kernel/apic/x2apic_cluster.c

@@ -9,11 +9,7 @@
 
 #include "local.h"
 
-struct cluster_mask {
-	unsigned int	clusterid;
-	int		node;
-	struct cpumask	mask;
-};
+#define apic_cluster(apicid) ((apicid) >> 4)
 
 /*
  * __x2apic_send_IPI_mask() possibly needs to read
@@ -23,8 +19,7 @@ struct cluster_mask {
 static u32 *x86_cpu_to_logical_apicid __read_mostly;
 
 static DEFINE_PER_CPU(cpumask_var_t, ipi_mask);
-static DEFINE_PER_CPU_READ_MOSTLY(struct cluster_mask *, cluster_masks);
-static struct cluster_mask *cluster_hotplug_mask;
+static DEFINE_PER_CPU_READ_MOSTLY(struct cpumask *, cluster_masks);
 
 static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 {
@@ -60,18 +55,18 @@ __x2apic_send_IPI_mask(const struct cpumask *mask, int vector, int apic_dest)
 
 	/* Collapse cpus in a cluster so a single IPI per cluster is sent */
 	for_each_cpu(cpu, tmpmsk) {
-		struct cluster_mask *cmsk = per_cpu(cluster_masks, cpu);
+		struct cpumask *cmsk = per_cpu(cluster_masks, cpu);
 
 		dest = 0;
-		for_each_cpu_and(clustercpu, tmpmsk, &cmsk->mask)
+		for_each_cpu_and(clustercpu, tmpmsk, cmsk)
 			dest |= x86_cpu_to_logical_apicid[clustercpu];
 
 		if (!dest)
 			continue;
 
 		__x2apic_send_IPI_dest(dest, vector, APIC_DEST_LOGICAL);
 		/* Remove cluster CPUs from tmpmask */
-		cpumask_andnot(tmpmsk, tmpmsk, &cmsk->mask);
+		cpumask_andnot(tmpmsk, tmpmsk, cmsk);
 	}
 
 	local_irq_restore(flags);
@@ -105,55 +100,98 @@ static u32 x2apic_calc_apicid(unsigned int cpu)
 
 static void init_x2apic_ldr(void)
 {
-	struct cluster_mask *cmsk = this_cpu_read(cluster_masks);
-	u32 cluster, apicid = apic_read(APIC_LDR);
-	unsigned int cpu;
+	struct cpumask *cmsk = this_cpu_read(cluster_masks);
 
-	x86_cpu_to_logical_apicid[smp_processor_id()] = apicid;
+	BUG_ON(!cmsk);
 
-	if (cmsk)
-		goto update;
-
-	cluster = apicid >> 16;
-	for_each_online_cpu(cpu) {
-		cmsk = per_cpu(cluster_masks, cpu);
-		/* Matching cluster found. Link and update it. */
-		if (cmsk && cmsk->clusterid == cluster)
-			goto update;
+	cpumask_set_cpu(smp_processor_id(), cmsk);
+}
+
+/*
+ * As an optimisation during boot, set the cluster_mask for all present
+ * CPUs at once, to prevent each of them having to iterate over the others
+ * to find the existing cluster_mask.
+ */
+static void prefill_clustermask(struct cpumask *cmsk, unsigned int cpu, u32 cluster)
+{
+	int cpu_i;
+
+	for_each_present_cpu(cpu_i) {
+		struct cpumask **cpu_cmsk = &per_cpu(cluster_masks, cpu_i);
+		u32 apicid = apic->cpu_present_to_apicid(cpu_i);
+
+		if (apicid == BAD_APICID || cpu_i == cpu || apic_cluster(apicid) != cluster)
+			continue;
+
+		if (WARN_ON_ONCE(*cpu_cmsk == cmsk))
+			continue;
+
+		BUG_ON(*cpu_cmsk);
+		*cpu_cmsk = cmsk;
 	}
-	cmsk = cluster_hotplug_mask;
-	cmsk->clusterid = cluster;
-	cluster_hotplug_mask = NULL;
-update:
-	this_cpu_write(cluster_masks, cmsk);
-	cpumask_set_cpu(smp_processor_id(), &cmsk->mask);
 }
 
-static int alloc_clustermask(unsigned int cpu, int node)
+static int alloc_clustermask(unsigned int cpu, u32 cluster, int node)
 {
+	struct cpumask *cmsk = NULL;
+	unsigned int cpu_i;
+
+	/*
+	 * At boot time, the CPU present mask is stable. The cluster mask is
+	 * allocated for the first CPU in the cluster and propagated to all
+	 * present siblings in the cluster. If the cluster mask is already set
+	 * on entry to this function for a given CPU, there is nothing to do.
+	 */
 	if (per_cpu(cluster_masks, cpu))
 		return 0;
+
+	if (system_state < SYSTEM_RUNNING)
+		goto alloc;
+
 	/*
-	 * If a hotplug spare mask exists, check whether it's on the right
-	 * node. If not, free it and allocate a new one.
+	 * On post boot hotplug for a CPU which was not present at boot time,
+	 * iterate over all possible CPUs (even those which are not present
+	 * any more) to find any existing cluster mask.
 	 */
-	if (cluster_hotplug_mask) {
-		if (cluster_hotplug_mask->node == node)
-			return 0;
-		kfree(cluster_hotplug_mask);
+	for_each_possible_cpu(cpu_i) {
+		u32 apicid = apic->cpu_present_to_apicid(cpu_i);
+
+		if (apicid != BAD_APICID && apic_cluster(apicid) == cluster) {
+			cmsk = per_cpu(cluster_masks, cpu_i);
+			/*
+			 * If the cluster is already initialized, just store
+			 * the mask and return. There's no need to propagate.
+			 */
+			if (cmsk) {
+				per_cpu(cluster_masks, cpu) = cmsk;
+				return 0;
+			}
+		}
 	}
-
-	cluster_hotplug_mask = kzalloc_node(sizeof(*cluster_hotplug_mask),
-					    GFP_KERNEL, node);
-	if (!cluster_hotplug_mask)
+	/*
+	 * No CPU in the cluster has ever been initialized, so fall through to
+	 * the boot time code which will also populate the cluster mask for any
+	 * other CPU in the cluster which is (now) present.
+	 */
+alloc:
+	cmsk = kzalloc_node(sizeof(*cmsk), GFP_KERNEL, node);
+	if (!cmsk)
 		return -ENOMEM;
-	cluster_hotplug_mask->node = node;
+	per_cpu(cluster_masks, cpu) = cmsk;
+	prefill_clustermask(cmsk, cpu, cluster);
+
 	return 0;
 }
 
 static int x2apic_prepare_cpu(unsigned int cpu)
 {
-	if (alloc_clustermask(cpu, cpu_to_node(cpu)) < 0)
+	u32 phys_apicid = apic->cpu_present_to_apicid(cpu);
+	u32 cluster = apic_cluster(phys_apicid);
+	u32 logical_apicid = (cluster << 16) | (1 << (phys_apicid & 0xf));
+
+	x86_cpu_to_logical_apicid[cpu] = logical_apicid;
+
+	if (alloc_clustermask(cpu, cluster, cpu_to_node(cpu)) < 0)
 		return -ENOMEM;
 	if (!zalloc_cpumask_var(&per_cpu(ipi_mask, cpu), GFP_KERNEL))
 		return -ENOMEM;
@@ -162,10 +200,10 @@ static int x2apic_prepare_cpu(unsigned int cpu)
 
 static int x2apic_dead_cpu(unsigned int dead_cpu)
 {
-	struct cluster_mask *cmsk = per_cpu(cluster_masks, dead_cpu);
+	struct cpumask *cmsk = per_cpu(cluster_masks, dead_cpu);
 
 	if (cmsk)
-		cpumask_clear_cpu(dead_cpu, &cmsk->mask);
+		cpumask_clear_cpu(dead_cpu, cmsk);
 	free_cpumask_var(per_cpu(ipi_mask, dead_cpu));
 	return 0;
 }

arch/x86/kernel/asm-offsets.c

@@ -115,6 +115,7 @@ static void __used common(void)
 	OFFSET(TSS_sp1, tss_struct, x86_tss.sp1);
 	OFFSET(TSS_sp2, tss_struct, x86_tss.sp2);
 	OFFSET(X86_top_of_stack, pcpu_hot, top_of_stack);
+	OFFSET(X86_current_task, pcpu_hot, current_task);
 #ifdef CONFIG_CALL_DEPTH_TRACKING
 	OFFSET(X86_call_depth, pcpu_hot, call_depth);
 #endif