vgsp: bounded newbind/rebuild diagnostic tokens (r234 discrimination)

r233: ctaid_probe attempt-1 completed (newbind fix works) but the NEXT
client (P2, fresh process after P1's CLEAN teardown+flush) stalled with
zero tokens. witness-r233.log shows why: after the last FREE
(180966.07) and rebuild #40 (180967.07) the guest sends ZERO traffic
for 9+ s (only 'completion hint rc=-2'); no channel binding ever exists,
so no entry-level recovery (pending-map/chanswitch/stale-entry) can
apply. Pattern across boots: a CLEAN client exit is followed by a
pre-registration stall; a KILLED client is followed by a working client.

Diagnostics added (bounded 16/session each):
- 'GR-STALL-newbind h=... put=... get=... action=window
   reason=stale-entry-newbind' when a binding's freshness window is
   armed (proves the reconcile runs per binding, not just once).
- 'GR-STALL-rebuild consumed=1 by=svc_thread ...' when the service loop
   consumes svc.rebuild_now (proves the forced rebuild is taken; rules
   out the stale-flag hypothesis).

wd_chanswitch/wd_newbind_until are per-channel/per-binding (armed on
every !same_gen registration) - no global-state stomping between
clients. make kernel green.

(cherry picked from commit 2e6921f4fe7d2a0dcd478430981367a6e354465c)

1 files changed, 24 insertions, 0 deletions

diff --git a/kernel/gsp_sema_service.c b/kernel/gsp_sema_service.c
index b7290d4..e7c690b 100644
--- a/kernel/gsp_sema_service.c
+++ b/kernel/gsp_sema_service.c
@@ -1029,6 +1029,10 @@ struct sema_svc {
 	bool			rebuild_now;
 	u32			gr_pending_map;	/* GR-STALL-pending-map lines
 						 * emitted (bounded proof) */
+	u32			gr_newbind_log;	/* GR-STALL-newbind window-armed
+						 * lines emitted (bounded proof) */
+	u32			gr_rebuild_log;	/* GR-STALL-rebuild consumption
+						 * lines emitted (bounded proof) */
 	u64			fwd_gr_hold_exp;	/* GRFIX FIX1: data-stalled holds
 						 * that reached the hard bound and
 						 * were consumed with the software
@@ -17254,6 +17258,17 @@ static void svc_pump_channels(struct nv_vgpu_device *vgpu)
 			svc_opdef_drop_channel(vgpu, ch->h_object, jiffies);
 			ch->wd_newbind_until = jiffies +
 				msecs_to_jiffies(SVC_GR_NEWBIND_MS);
+			/*
+			 * Bounded proof (r234 asks): the window was armed for
+			 * THIS binding; if a later stall shows no such line,
+			 * its binding never reached the pump (pre-registration
+			 * class) and this mechanism cannot apply.
+			 */
+			if (svc.gr_newbind_log < 16) {
+				svc.gr_newbind_log++;
+				pr_info("vgsp-gr: GR-STALL-newbind h=0x%x put=%u get=%u action=window reason=stale-entry-newbind\n",
+					ch->h_object, put, get);
+			}
 			if (put > 1) {
 				svc.gr_stale_wedge++;
 				pr_warn_ratelimited("vgsp-gr: GR-STALL-stale-wedge h=0x%x va=0x%llx last=%u put=%u get=%u sub=%llu waited_ms=0 action=recover reason=stale-entry-chanswitch reset=0 rewind=0 entry=0 cpa=0x%llx rpa=0x%llx\n",
@@ -19989,6 +20004,15 @@ static int svc_thread(void *arg)
 			 */
 			if (svc.rebuild_now) {
 				svc.rebuild_now = false;
+				/*
+				 * Bounded proof (r234 asks): a pump-requested
+				 * rebuild was actually consumed by the service
+				 * thread here (the only consumer).
+				 */
+				if (svc.gr_rebuild_log < 16) {
+					svc.gr_rebuild_log++;
+					pr_info("vgsp-gr: GR-STALL-rebuild consumed=1 by=svc_thread action=rebuild reason=stale-entry-newbind\n");
+				}
 				svc_rebuild(vgpu);
 				if (sess)
 					sess->last_build = svc.last_build;