gh0s7/mckernel
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Revert "memory_range_lock: Enable interrupt when trylock fails"

Browse Source 
This reverts commit 0d3ef65092.

Reason for revert: This fix causes circular dependency with memory_range manipulation and TLB flush. See #1394.

Change-Id: I4774e81ff300c199629e283e538c0a30ad0eeaae
This commit is contained in:
Masamichi Takagi
2019-09-20 07:18:15 +00:00
committed by Ken Sato
parent 41d37bcd30
commit edd3ea0103
12 changed files with 102 additions and 393 deletions
Download Patch File Download Diff File Expand all files Collapse all files

36
kernel/process.c
View File

@@ -244,7 +244,7 @@ static int
init_process_vm(struct process *owner, struct address_space *asp, struct process_vm *vm)
{
int i;
ihk_mc_rwlock_init(&vm->memory_range_lock);
ihk_mc_spinlock_init(&vm->memory_range_lock);
ihk_mc_spinlock_init(&vm->page_table_lock);
ihk_atomic_set(&vm->refcount, 1);
@@ -785,9 +785,8 @@ static int copy_user_ranges(struct process_vm *vm, struct process_vm *orgvm)
struct vm_range *range;
struct vm_range *last_insert;
struct copy_args args;
unsigned long irqflags;
memory_range_write_lock(orgvm, &irqflags);
ihk_mc_spinlock_lock_noirq(&orgvm->memory_range_lock);
/* Iterate original process' vm_range list and take a copy one-by-one */
last_insert = NULL;
@@ -849,7 +848,7 @@ static int copy_user_ranges(struct process_vm *vm, struct process_vm *orgvm)
// memory_stat_rss_add() is called in child-node, i.e. copy_user_pte()
}
memory_range_write_unlock(orgvm, &irqflags);
ihk_mc_spinlock_unlock_noirq(&orgvm->memory_range_lock);
return 0;
@@ -869,14 +868,14 @@ err_rollback:
if (dest_range) {
free_process_memory_range(vm, dest_range);
}
if (src_range == last_insert) {
break;
}
src_range = next_process_memory_range(orgvm, src_range);
}
}
memory_range_write_unlock(orgvm, &irqflags);
ihk_mc_spinlock_unlock_noirq(&orgvm->memory_range_lock);
return -1;
}
@@ -2165,7 +2164,6 @@ static int do_page_fault_process_vm(struct process_vm *vm, void *fault_addr0, ui
struct vm_range *range;
struct thread *thread = cpu_local_var(current);
int locked = 0;
unsigned long irqflags;
dkprintf("[%d]do_page_fault_process_vm(%p,%lx,%lx)\n",
ihk_mc_get_processor_id(), vm, fault_addr0, reason);
@@ -2176,12 +2174,10 @@ static int do_page_fault_process_vm(struct process_vm *vm, void *fault_addr0, ui
if (thread->vm->is_memory_range_lock_taken) {
goto skip;
}
irqflags = cpu_disable_interrupt_save();
if (ihk_mc_read_trylock(&vm->memory_range_lock)) {
if (ihk_mc_spinlock_trylock_noirq(&vm->memory_range_lock)) {
locked = 1;
break;
}
cpu_restore_interrupt(irqflags);
}
} else {
skip:;
@@ -2297,8 +2293,7 @@ skip:;
error = 0;
out:
if (locked) {
ihk_mc_read_unlock(&vm->memory_range_lock);
cpu_restore_interrupt(irqflags);
ihk_mc_spinlock_unlock_noirq(&vm->memory_range_lock);
}
dkprintf("[%d]do_page_fault_process_vm(%p,%lx,%lx): %d\n",
ihk_mc_get_processor_id(), vm, fault_addr0,
@@ -2599,10 +2594,9 @@ void flush_process_memory(struct process_vm *vm)
struct vm_range *range;
struct rb_node *node, *next = rb_first(&vm->vm_range_tree);
int error;
unsigned long irqflags;
dkprintf("flush_process_memory(%p)\n", vm);
memory_range_write_lock(vm, &irqflags);
ihk_mc_spinlock_lock_noirq(&vm->memory_range_lock);
/* Let concurrent page faults know the VM will be gone */
vm->exiting = 1;
while ((node = next)) {
@@ -2620,7 +2614,7 @@ void flush_process_memory(struct process_vm *vm)
}
}
}
memory_range_write_unlock(vm, &irqflags);
ihk_mc_spinlock_unlock_noirq(&vm->memory_range_lock);
dkprintf("flush_process_memory(%p):\n", vm);
return;
}
@@ -2630,13 +2624,12 @@ void free_process_memory_ranges(struct process_vm *vm)
int error;
struct vm_range *range;
struct rb_node *node, *next = rb_first(&vm->vm_range_tree);
unsigned long irqflags;
if (vm == NULL) {
return;
}
memory_range_write_lock(vm, &irqflags);
ihk_mc_spinlock_lock_noirq(&vm->memory_range_lock);
while ((node = next)) {
range = rb_entry(node, struct vm_range, vm_rb_node);
next = rb_next(node);
@@ -2649,7 +2642,7 @@ void free_process_memory_ranges(struct process_vm *vm)
/* through */
}
}
memory_range_write_unlock(vm, &irqflags);
ihk_mc_spinlock_unlock_noirq(&vm->memory_range_lock);
}
static void free_thread_pages(struct thread *thread)
@@ -2738,9 +2731,8 @@ free_all_process_memory_range(struct process_vm *vm)
struct vm_range *range;
struct rb_node *node, *next = rb_first(&vm->vm_range_tree);
int error;
unsigned long irqflags;
memory_range_write_lock(vm, &irqflags);
ihk_mc_spinlock_lock_noirq(&vm->memory_range_lock);
while ((node = next)) {
range = rb_entry(node, struct vm_range, vm_rb_node);
next = rb_next(node);
@@ -2753,7 +2745,7 @@ free_all_process_memory_range(struct process_vm *vm)
/* through */
}
}
memory_range_write_unlock(vm, &irqflags);
ihk_mc_spinlock_unlock_noirq(&vm->memory_range_lock);
}
void
@@ -3173,7 +3165,7 @@ void sched_init(void)
&idle_thread->proc->children_list);
ihk_mc_init_context(&idle_thread->ctx, NULL, idle);
ihk_mc_rwlock_init(&idle_thread->vm->memory_range_lock);
ihk_mc_spinlock_init(&idle_thread->vm->memory_range_lock);
idle_thread->vm->vm_range_tree = RB_ROOT;
idle_thread->vm->vm_range_numa_policy_tree = RB_ROOT;
idle_thread->proc->pid = 0;