Leave a page alone when a shadow write-back cannot hold the write it reports.

With shadow storage a write-back at a fence copies a page that has no base copy whole from the shadow into guest
memory. Every write the replay records gets base copies, so such a page means the pool the write was noted on was
finalized before the fence and another was made at its address: the new record's shadow is a copy of guest memory
from its own start, holds no byte of that write, and went over everything the CPU had written to the page since.
The page is now left alone when the record is younger than the submission whose fence it is, or when the CPU has
written the page since the shadow's last copy-in; the shadow line counts such pages.
--probe-vkmem has the case: with NX2NAVI_SHADOW=1 it fails without this change (0x11 over the CPU's 0x22, with
imports and with mirrors) and passes with it, as does the run without shadow storage.
Whether a game does finalize and remake a pool inside that window was not observed in play.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
This commit is contained in:
chrg6
2026-11-11 04:58:57 -05:00
committed by tangerineturtle1
parent 2e28ba12df
commit ae70219570
4 changed files with 59 additions and 3 deletions

View File

@@ -147,6 +147,14 @@ static volatile int32_t cb_late; // the callback being run belongs
int vk_sync_effect_late(void) { return __atomic_load_n(&cb_late, __ATOMIC_RELAXED); }
// The value whose side effect the calling thread is running, 0 when it is not the completion thread inside one.
static volatile uint64_t cb_value;
static volatile uint32_t cb_tid;
uint64_t vk_sync_cb_value(void) {
return __atomic_load_n(&cb_tid, __ATOMIC_ACQUIRE) == plat_thread_id() ? __atomic_load_n(&cb_value, __ATOMIC_RELAXED) : 0;
}
void vk_sync_exec_stats(uint64_t *waits, uint64_t *early, uint64_t *lag_ns, uint64_t *lag_max_ns) {
*waits = __atomic_exchange_n(&exw_n, 0, __ATOMIC_RELAXED), *early = __atomic_exchange_n(&exw_early, 0, __ATOMIC_RELAXED);
*lag_ns = __atomic_exchange_n(&exw_lag_ns, 0, __ATOMIC_RELAXED);
@@ -482,6 +490,7 @@ WEAK int vk_exec_record(VkCommandBuffer cb, const uint64_t *handles, int n, uint
static void completion_main(void *arg) {
VkCtx *c = arg;
__atomic_store_n(&cb_tid, plat_thread_id(), __ATOMIC_RELEASE);
plat_mutex_lock(&st_lock);
for (;;) {
while (!stop && done == issued && !(qn && queue[qhead].fence <= done))
@@ -528,7 +537,9 @@ static void completion_main(void *arg) {
plat_mutex_unlock(&st_lock);
__atomic_store_n(&cb_late, p.fence <= __atomic_load_n(&exw_value, __ATOMIC_ACQUIRE), __ATOMIC_RELAXED);
uint64_t c0 = (rt_trace & RT_TRACE_PERF) ? kern_now_ns() : 0;
__atomic_store_n(&cb_value, p.fence, __ATOMIC_RELAXED);
p.fn(p.arg);
__atomic_store_n(&cb_value, 0, __ATOMIC_RELAXED);
uint64_t c1 = c0 ? kern_now_ns() : 0;
__atomic_store_n(&cb_late, 0, __ATOMIC_RELAXED);
plat_mutex_lock(&st_lock);