vgsp: X/GL frame-pacer mirror (sema_xw_mirror) - unblocks the 10s/frame stall

r861 evidence: the X server's present path spins in a 10s-timeout wait
((EE) NVIDIA(GPU-0): WAIT (2, 8, 0x8000, <stuck 0xd4..>, <driver counter
+0x70/frame>)) comparing (progress_word & 0x3ffff) << 2 against its own
submitted-byte counter; the progress word (lone u32 0x140235 in an
otherwise-zero guest page) never advances because the emulation does not
execute the 2D method streams -> glxgears 0.09 FPS. A gdb write of the
word to 0x140000 | (counter >> 2) made the wait complete in 3 ms
(WAIT (0, ... <equal>, <equal>) immediately after the timeout print) and
silenced all further timeouts. This adds a 3 ms kthread that mirrors the
driver's counter word into the progress word; both guest PAs are injected
per boot via the sema_xw_mirror parameter (0,0 stops). Demo-path tuning
aid; no product logic depends on it.

1 files changed, 32 insertions, 0 deletions

diff --git a/kernel/main.c b/kernel/main.c
index a034b44..0e9452b 100644
--- a/kernel/main.c
+++ b/kernel/main.c
@@ -3097,6 +3097,38 @@ module_param_cb(sema_vidmem_dump, &nv_vgpu_sema_vidmem_dump_ops, NULL, 0200);
 MODULE_PARM_DESC(sema_vidmem_dump, "Dump emulated vidmem (addr,len) to the kernel log");
 
 /*
+ * X/GL frame-pacer mirror (SESSION 148 perf): mirror the X server's
+ * submitted-byte counter (guest PA) into its GPU-progress word (guest PA)
+ * so the driver's per-frame 10 s wait completes in milliseconds.
+ *   echo 0x112345678,0x112345000 > .../parameters/sema_xw_mirror
+ *   echo 0,0                     > .../parameters/sema_xw_mirror  (stop)
+ */
+static int nv_vgpu_sema_xw_mirror_set(const char *val,
+				      const struct kernel_param *kp)
+{
+	char buf[64];
+	char *comma;
+	u64 cnt = 0, mem = 0;
+
+	strscpy(buf, val, sizeof(buf));
+	comma = strchr(buf, ',');
+	if (!comma)
+		return -EINVAL;
+	*comma = '\0';
+	if (kstrtoull(buf, 0, &cnt) || kstrtoull(comma + 1, 0, &mem))
+		return -EINVAL;
+
+	return gsp_sema_service_xw_mirror(cnt, mem);
+}
+
+static const struct kernel_param_ops nv_vgpu_sema_xw_mirror_ops = {
+	.set = nv_vgpu_sema_xw_mirror_set,
+};
+
+module_param_cb(sema_xw_mirror, &nv_vgpu_sema_xw_mirror_ops, NULL, 0200);
+MODULE_PARM_DESC(sema_xw_mirror, "Mirror X frame counter (cnt_pa,mem_pa; 0,0=stop)");
+
+/*
  * Debug: search the whole emulated vidmem for an 8-byte pattern:
  *   echo 0x0000000400000001,16 > .../parameters/sema_vidmem_find
  */