Merge 5.10.194 into android12-5.10-lts

Changes in 5.10.194
	module: Expose module_init_layout_section()
	arm64: module-plts: inline linux/moduleloader.h
	arm64: module: Use module_init_layout_section() to spot init sections
	ARM: module: Use module_init_layout_section() to spot init sections
	mhi: pci_generic: Fix implicit conversion warning
	Revert "drm/amdgpu: install stub fence into potential unused fence pointers"
	Revert "MIPS: Alchemy: fix dbdma2"
	rcu: Prevent expedited GP from enabling tick on offline CPU
	rcu-tasks: Fix IPI failure handling in trc_wait_for_one_reader
	rcu-tasks: Wait for trc_read_check_handler() IPIs
	rcu-tasks: Add trc_inspect_reader() checks for exiting critical section
	Linux 5.10.194

Change-Id: I1e4230071f3ca3c9e811aeba31446875028d7b3d
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Makefile

@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0
 VERSION = 5
 PATCHLEVEL = 10
-SUBLEVEL = 193
+SUBLEVEL = 194
 EXTRAVERSION =
 NAME = Dare mighty things
 

arch/arm/kernel/module-plts.c

@@ -256,7 +256,7 @@ int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs,
 		/* sort by type and symbol index */
 		sort(rels, numrels, sizeof(Elf32_Rel), cmp_rel, NULL);
 
-		if (strncmp(secstrings + dstsec->sh_name, ".init", 5) != 0)
+		if (!module_init_layout_section(secstrings + dstsec->sh_name))
 			core_plts += count_plts(syms, dstsec->sh_addr, rels,
 						numrels, s->sh_info);
 		else

arch/arm64/kernel/module-plts.c

@@ -7,6 +7,7 @@
 #include <linux/ftrace.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/moduleloader.h>
 #include <linux/slab.h>
 #include <linux/sort.h>
 
@@ -376,7 +377,7 @@ int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs,
 		if (nents)
 			sort(rels, nents, sizeof(Elf64_Rela), cmp_rela, NULL);
 
-		if (!str_has_prefix(secstrings + dstsec->sh_name, ".init"))
+		if (!module_init_layout_section(secstrings + dstsec->sh_name))
 			core_plts += count_plts(syms, rels, numrels,
 						sechdrs[i].sh_info, dstsec);
 		else

arch/mips/alchemy/common/dbdma.c

@@ -30,7 +30,6 @@
  *
  */
 
-#include <linux/dma-map-ops.h> /* for dma_default_coherent */
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/slab.h>
@@ -624,18 +623,17 @@ u32 au1xxx_dbdma_put_source(u32 chanid, dma_addr_t buf, int nbytes, u32 flags)
 		dp->dscr_cmd0 &= ~DSCR_CMD0_IE;
 
 	/*
-	 * There is an erratum on certain Au1200/Au1550 revisions that could
+	 * There is an errata on the Au1200/Au1550 parts that could result
-	 * result in "stale" data being DMA'ed. It has to do with the snoop
+	 * in "stale" data being DMA'ed. It has to do with the snoop logic on
-	 * logic on the cache eviction buffer.  dma_default_coherent is set
+	 * the cache eviction buffer.  DMA_NONCOHERENT is on by default for
-	 * to false on these parts.
+	 * these parts. If it is fixed in the future, these dma_cache_inv will
+	 * just be nothing more than empty macros. See io.h.
 	 */
-	if (!dma_default_coherent)
+	dma_cache_wback_inv((unsigned long)buf, nbytes);
-		dma_cache_wback_inv(KSEG0ADDR(buf), nbytes);
 	dp->dscr_cmd0 |= DSCR_CMD0_V;	/* Let it rip */
 	wmb(); /* drain writebuffer */
 	dma_cache_wback_inv((unsigned long)dp, sizeof(*dp));
 	ctp->chan_ptr->ddma_dbell = 0;
-	wmb(); /* force doorbell write out to dma engine */
 
 	/* Get next descriptor pointer. */
 	ctp->put_ptr = phys_to_virt(DSCR_GET_NXTPTR(dp->dscr_nxtptr));
@@ -687,18 +685,17 @@ u32 au1xxx_dbdma_put_dest(u32 chanid, dma_addr_t buf, int nbytes, u32 flags)
 			  dp->dscr_source1, dp->dscr_dest0, dp->dscr_dest1);
 #endif
 	/*
-	 * There is an erratum on certain Au1200/Au1550 revisions that could
+	 * There is an errata on the Au1200/Au1550 parts that could result in
-	 * result in "stale" data being DMA'ed. It has to do with the snoop
+	 * "stale" data being DMA'ed. It has to do with the snoop logic on the
-	 * logic on the cache eviction buffer.  dma_default_coherent is set
+	 * cache eviction buffer.  DMA_NONCOHERENT is on by default for these
-	 * to false on these parts.
+	 * parts. If it is fixed in the future, these dma_cache_inv will just
+	 * be nothing more than empty macros. See io.h.
 	 */
-	if (!dma_default_coherent)
+	dma_cache_inv((unsigned long)buf, nbytes);
-		dma_cache_inv(KSEG0ADDR(buf), nbytes);
 	dp->dscr_cmd0 |= DSCR_CMD0_V;	/* Let it rip */
 	wmb(); /* drain writebuffer */
 	dma_cache_wback_inv((unsigned long)dp, sizeof(*dp));
 	ctp->chan_ptr->ddma_dbell = 0;
-	wmb(); /* force doorbell write out to dma engine */
 
 	/* Get next descriptor pointer. */
 	ctp->put_ptr = phys_to_virt(DSCR_GET_NXTPTR(dp->dscr_nxtptr));

drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c

@@ -2155,7 +2155,6 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 	amdgpu_vm_bo_base_init(&bo_va->base, vm, bo);
 
 	bo_va->ref_count = 1;
-	bo_va->last_pt_update = dma_fence_get_stub();
 	INIT_LIST_HEAD(&bo_va->valids);
 	INIT_LIST_HEAD(&bo_va->invalids);
 
@@ -2868,8 +2867,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		vm->update_funcs = &amdgpu_vm_cpu_funcs;
 	else
 		vm->update_funcs = &amdgpu_vm_sdma_funcs;
-
+	vm->last_update = NULL;
-	vm->last_update = dma_fence_get_stub();
 	vm->last_unlocked = dma_fence_get_stub();
 
 	mutex_init(&vm->eviction_lock);
@@ -3044,7 +3042,7 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		vm->update_funcs = &amdgpu_vm_sdma_funcs;
 	}
 	dma_fence_put(vm->last_update);
-	vm->last_update = dma_fence_get_stub();
+	vm->last_update = NULL;
 	vm->is_compute_context = true;
 
 	if (vm->pasid) {

include/linux/moduleloader.h

@@ -39,6 +39,11 @@ bool module_init_section(const char *name);
  */
 bool module_exit_section(const char *name);
 
+/* Describes whether within_module_init() will consider this an init section
+ * or not. This behaviour changes with CONFIG_MODULE_UNLOAD.
+ */
+bool module_init_layout_section(const char *sname);
+
 /*
  * Apply the given relocation to the (simplified) ELF.  Return -error
  * or 0.

kernel/module.c

@@ -2297,7 +2297,7 @@ void *__symbol_get(const char *symbol)
 }
 EXPORT_SYMBOL_GPL(__symbol_get);
 
-static bool module_init_layout_section(const char *sname)
+bool module_init_layout_section(const char *sname)
 {
 #ifndef CONFIG_MODULE_UNLOAD
 	if (module_exit_section(sname))

kernel/rcu/tasks.h

@@ -874,7 +874,7 @@ reset_ipi:
 static bool trc_inspect_reader(struct task_struct *t, void *arg)
 {
 	int cpu = task_cpu(t);
-	bool in_qs = false;
+	int nesting;
 	bool ofl = cpu_is_offline(cpu);
 
 	if (task_curr(t)) {
@@ -894,18 +894,18 @@ static bool trc_inspect_reader(struct task_struct *t, void *arg)
 		n_heavy_reader_updates++;
 		if (ofl)
 			n_heavy_reader_ofl_updates++;
-		in_qs = true;
+		nesting = 0;
 	} else {
 		// The task is not running, so C-language access is safe.
-		in_qs = likely(!t->trc_reader_nesting);
+		nesting = t->trc_reader_nesting;
 	}
 
-	// Mark as checked so that the grace-period kthread will
+	// If not exiting a read-side critical section, mark as checked
-	// remove it from the holdout list.
+	// so that the grace-period kthread will remove it from the
-	t->trc_reader_checked = true;
+	// holdout list.
-
+	t->trc_reader_checked = nesting >= 0;
-	if (in_qs)
+	if (nesting <= 0)
-		return true;  // Already in quiescent state, done!!!
+		return !nesting;  // If in QS, done, otherwise try again later.
 
 	// The task is in a read-side critical section, so set up its
 	// state so that it will awaken the grace-period kthread upon exit
@@ -958,9 +958,11 @@ static void trc_wait_for_one_reader(struct task_struct *t,
 		if (smp_call_function_single(cpu, trc_read_check_handler, t, 0)) {
 			// Just in case there is some other reason for
 			// failure than the target CPU being offline.
+			WARN_ONCE(1, "%s():  smp_call_function_single() failed for CPU: %d\n",
+				  __func__, cpu);
 			rcu_tasks_trace.n_ipis_fails++;
 			per_cpu(trc_ipi_to_cpu, cpu) = false;
-			t->trc_ipi_to_cpu = cpu;
+			t->trc_ipi_to_cpu = -1;
 		}
 	}
 }
@@ -1081,14 +1083,28 @@ static void check_all_holdout_tasks_trace(struct list_head *hop,
 	}
 }
 
+static void rcu_tasks_trace_empty_fn(void *unused)
+{
+}
+
 /* Wait for grace period to complete and provide ordering. */
 static void rcu_tasks_trace_postgp(struct rcu_tasks *rtp)
 {
+	int cpu;
 	bool firstreport;
 	struct task_struct *g, *t;
 	LIST_HEAD(holdouts);
 	long ret;
 
+	// Wait for any lingering IPI handlers to complete.  Note that
+	// if a CPU has gone offline or transitioned to userspace in the
+	// meantime, all IPI handlers should have been drained beforehand.
+	// Yes, this assumes that CPUs process IPIs in order.  If that ever
+	// changes, there will need to be a recheck and/or timed wait.
+	for_each_online_cpu(cpu)
+		if (smp_load_acquire(per_cpu_ptr(&trc_ipi_to_cpu, cpu)))
+			smp_call_function_single(cpu, rcu_tasks_trace_empty_fn, NULL, 1);
+
 	// Remove the safety count.
 	smp_mb__before_atomic();  // Order vs. earlier atomics
 	atomic_dec(&trc_n_readers_need_end);

kernel/rcu/tree_exp.h

@@ -507,7 +507,10 @@ static void synchronize_rcu_expedited_wait(void)
 				if (rdp->rcu_forced_tick_exp)
 					continue;
 				rdp->rcu_forced_tick_exp = true;
+				preempt_disable();
+				if (cpu_online(cpu))
 					tick_dep_set_cpu(cpu, TICK_DEP_BIT_RCU_EXP);
+				preempt_enable();
 			}
 		}
 		j = READ_ONCE(jiffies_till_first_fqs);