vgsp-gr: B-2 planted 64 MiB data-bounce region for large GR operands

Replace the 205-page / 68-pages-per-pointer GR operand window with a
boot-planted 16384-page (64 MiB) region at VA 0x0a000000 (PD0 80..111):
one dual PD0 entry per 2 MB window backed by one fresh 4 KB PT page,
every leaf PTE written at stage 3 before any engine fetch (the GMMU has
no PTE invalidate, so no leaf is rewritten at runtime). The 208-page
window, mail slot 206 and marker 207 stay as-is (mail injection and
copy diagnostics unchanged) but carry no operand data when the grown
region is on.

Sema service: reloc base and page cap derive from the shared header
constants; per-pointer span becomes a 4096-page runaway guard; a total
cap hit is fail-closed (launch stalled for the bounded hold, loud
CAP-HIT log, no silent truncation); the launch's own param/bank pages
are skipped as data candidates (budget hygiene). Copy-in, copy-out,
retirement wait and the window diagnostic index the active region
through one helper/chunk selector; checksum and dedup scratch move off
the kernel stack.

gsp_phys_gr_data_region=0 (insmod, boot-fixed) flips the whole path
back to the legacy 205-page window for A/B trialing.

1 files changed, 22 insertions, 0 deletions

diff --git a/kernel/gsp_phys_gpu.c b/kernel/gsp_phys_gpu.c
index 9b7235b..7932268 100644
--- a/kernel/gsp_phys_gpu.c
+++ b/kernel/gsp_phys_gpu.c
@@ -1075,6 +1075,28 @@ int gsp_phys_rmcall_gr_data_prewin_value(void)
 EXPORT_SYMBOL_GPL(gsp_phys_rmcall_gr_data_prewin_value);
 
 /*
+ * B-2: grown boot-fixed data-bounce region (gsp_phys_gr_data_region; default
+ * 1). When set, gsp_phys_rmcall_pdb_build plants the 16384-page (64 MiB)
+ * region at GSP_PHYS_RMCALL_GR_DATA_BOUNCE_VA0 (PD0 80..111) and the SVC
+ * relocates operand pages into it; the booter's copy-in/copy-out index the
+ * grown chunk. When 0, everything falls back to the legacy 205-page /
+ * 68-pages-per-pointer window below the mail page, for one-flip A/B
+ * trialing. Boot-fixed: it is read by the stage-3 plant, so set it at
+ * insmod, not at runtime.
+ */
+static int gsp_phys_gr_data_region_param = 1;
+module_param_named(gsp_phys_gr_data_region,
+		   gsp_phys_gr_data_region_param, int, 0644);
+MODULE_PARM_DESC(gsp_phys_gr_data_region,
+		 "Use the grown 64 MiB data-bounce region for GR operands (default 1; 0 = legacy 205-page window)");
+
+int gsp_phys_gr_data_region_value(void)
+{
+	return gsp_phys_gr_data_region_param;
+}
+EXPORT_SYMBOL_GPL(gsp_phys_gr_data_region_value);
+
+/*
  * m113: after planting the launch's data mappings, re-register the VAS page
  * directory (UNSET+SET) to force the server walker to re-snapshot the tree;
  * all evidence says the SM walker resolves only the commit-time tree, and