sched_ext: Maintain an online cid mask in the scheduler arenagrafted

Schedulers on the default cid mapping treat [0, nr_online_cids) as the
online set and restart on hotplug. Schedulers that install their own mapping
with scx_bpf_cid_override() have no way to learn which cids are online: the
count no longer identifies members and the CPU-form cpumask is unusable from
cid programs. This is an obvious hole in the cid API.

Add scx_bpf_online_cmask(), a kernel-maintained cmask in the scheduler's
arena, allocated alongside the per-CPU scratch masks and populated after the
cid mapping is finalized and before ops.init(), for child schedulers too.
The pointer stays valid through ops.exit() with no reference to take. It is
the arena offset as a void pointer, the same form struct_ops arena arguments
arrive in. The verifier types the void return as a scalar for the program's
arena cast.

The mask follows the SCX hotplug notifications: seeded from cpu_active_mask
and updated before ops.cid_online/offline() runs, so it lags cpu_online_mask
only inside a hotplug transition. Updates walk the scheduler list under the
lock that also serializes unlinking. Reads are live, not atomic snapshots.
Root initialization excludes hotplug.

v2: Reworded the getter kerneldoc (Andrea Righi).

Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>

1 files changed, 103 insertions, 0 deletions

diff --git a/arch/alpha/lib/stacktrace.c b/arch/alpha/lib/stacktrace.c
new file mode 100644
index 000000000..2b1176dd5
--- /dev/null
+++ b/arch/alpha/lib/stacktrace.c
@@ -0,0 +1,103 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/kernel.h>
+
+typedef unsigned int instr;
+
+#define MAJOR_OP	0xfc000000
+#define LDA_OP		0x20000000
+#define STQ_OP		0xb4000000
+#define BR_OP		0xc0000000
+
+#define STK_ALLOC_1	0x23de8000 /* lda $30,-X($30) */
+#define STK_ALLOC_1M	0xffff8000
+#define STK_ALLOC_2	0x43c0153e /* subq $30,X,$30 */
+#define STK_ALLOC_2M	0xffe01fff
+
+#define MEM_REG		0x03e00000
+#define MEM_BASE	0x001f0000
+#define MEM_OFF		0x0000ffff
+#define MEM_OFF_SIGN	0x00008000
+#define	BASE_SP		0x001e0000
+
+#define STK_ALLOC_MATCH(INSTR)			\
+  (((INSTR) & STK_ALLOC_1M) == STK_ALLOC_1	\
+   || ((INSTR) & STK_ALLOC_2M) == STK_ALLOC_2)
+#define STK_PUSH_MATCH(INSTR) \
+  (((INSTR) & (MAJOR_OP | MEM_BASE | MEM_OFF_SIGN)) == (STQ_OP | BASE_SP))
+#define MEM_OP_OFFSET(INSTR) \
+  (((long)((INSTR) & MEM_OFF) << 48) >> 48)
+#define MEM_OP_REG(INSTR) \
+  (((INSTR) & MEM_REG) >> 22)
+
+/* Branches, jumps, PAL calls, and illegal opcodes end a basic block. */
+#define BB_END(INSTR)						\
+  (((instr)(INSTR) >= BR_OP) | ((instr)(INSTR) < LDA_OP) |	\
+   ((((instr)(INSTR) ^ 0x60000000) < 0x20000000) &		\
+    (((instr)(INSTR) & 0x0c000000) != 0)))
+
+#define IS_KERNEL_TEXT(PC) ((unsigned long)(PC) > START_ADDR)
+
+static char reg_name[][4] = {
+	"v0 ", "t0 ", "t1 ", "t2 ", "t3 ", "t4 ", "t5 ", "t6 ", "t7 ",
+	"s0 ", "s1 ", "s2 ", "s3 ", "s4 ", "s5 ", "s6 ", "a0 ", "a1 ",
+	"a2 ", "a3 ", "a4 ", "a5 ", "t8 ", "t9 ", "t10", "t11", "ra ",
+	"pv ", "at ", "gp ", "sp ", "0"
+};
+
+
+static instr *
+display_stored_regs(instr * pro_pc, unsigned char * sp)
+{
+	instr * ret_pc = 0;
+	int reg;
+	unsigned long value;
+
+	printk("Prologue [<%p>], Frame %p:\n", pro_pc, sp);
+	while (!BB_END(*pro_pc))
+		if (STK_PUSH_MATCH(*pro_pc)) {
+			reg = (*pro_pc & MEM_REG) >> 21;
+			value = *(unsigned long *)(sp + (*pro_pc & MEM_OFF));
+			if (reg == 26)
+				ret_pc = (instr *)value;
+			printk("\t\t%s / 0x%016lx\n", reg_name[reg], value);
+		}
+	return ret_pc;
+}
+
+static instr *
+seek_prologue(instr * pc)
+{
+	while (!STK_ALLOC_MATCH(*pc))
+		--pc;
+	while (!BB_END(*(pc - 1)))
+		--pc;
+	return pc;
+}
+
+static long
+stack_increment(instr * prologue_pc)
+{
+	while (!STK_ALLOC_MATCH(*prologue_pc))
+		++prologue_pc;
+
+	/* Count the bytes allocated. */
+	if ((*prologue_pc & STK_ALLOC_1M) == STK_ALLOC_1M)
+		return -(((long)(*prologue_pc) << 48) >> 48);
+	else
+		return (*prologue_pc >> 13) & 0xff;
+}
+
+void
+stacktrace(void)
+{
+	instr * ret_pc;
+	instr * prologue = (instr *)stacktrace;
+	unsigned char *sp = (unsigned char *)current_stack_pointer;
+
+	printk("\tstack trace:\n");
+	do {
+		ret_pc = display_stored_regs(prologue, sp);
+		sp += stack_increment(prologue);
+		prologue = seek_prologue(ret_pc);
+	} while (IS_KERNEL_TEXT(ret_pc));
+}