Mesa (master): gallivm: allow large numbers of temporaries

[Zack Rusin]

Module: Mesa
Branch: master
Commit: 8507afc97fa3323c89ee4cd1359d2fa61015bcd0
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=8507afc97fa3323c89ee4cd1359d2fa61015bcd0

Author: Zack Rusin <zackr at vmware.com>
Date:   Mon Feb  3 21:40:24 2014 -0500

gallivm: allow large numbers of temporaries

The number of allowed temporaries increases almost with every
iteration of an api. We used to support 128, then we started
increasing and the newer api's support 4096+. So if we notice
that the number of temporaries is larger than our statically
allocated storage would allow we just treat them as indexable
temporaries and allocate them as an array from the start.

Signed-off-by: Zack Rusin <zackr at vmware.com>
Reviewed-by: Jose Fonseca <jfonseca at vmware.com>
Reviewed-by: Brian Paul <brianp at vmware.com>
Reviewed-by: Roland Scheidegger <sroland at vmware.com>

---

 src/gallium/auxiliary/gallivm/lp_bld_limits.h   |    8 +++++++-
 src/gallium/auxiliary/gallivm/lp_bld_tgsi.h     |    4 ++--
 src/gallium/auxiliary/gallivm/lp_bld_tgsi_aos.c |    2 +-
 src/gallium/auxiliary/gallivm/lp_bld_tgsi_soa.c |   11 ++++++++++-
 4 files changed, 20 insertions(+), 5 deletions(-)

diff --git a/src/gallium/auxiliary/gallivm/lp_bld_limits.h b/src/gallium/auxiliary/gallivm/lp_bld_limits.h
index 521b45b..e03bac6 100644
--- a/src/gallium/auxiliary/gallivm/lp_bld_limits.h
+++ b/src/gallium/auxiliary/gallivm/lp_bld_limits.h
@@ -43,7 +43,7 @@
  * the state trackers.
  */
 
-#define LP_MAX_TGSI_TEMPS 256
+#define LP_MAX_TGSI_TEMPS 4096
 
 #define LP_MAX_TGSI_ADDRS 16
 
@@ -53,6 +53,12 @@
 
 #define LP_MAX_TGSI_CONST_BUFFERS 16
 
+/*
+ * For quick access we cache temps in a statically
+ * allocated array. This defines the maximum size
+ * of that array.
+ */
+#define LP_MAX_INLINED_TEMPS 256
 
 /**
  * Maximum control flow nesting
diff --git a/src/gallium/auxiliary/gallivm/lp_bld_tgsi.h b/src/gallium/auxiliary/gallivm/lp_bld_tgsi.h
index 1a93951..e0a7c5d 100644
--- a/src/gallium/auxiliary/gallivm/lp_bld_tgsi.h
+++ b/src/gallium/auxiliary/gallivm/lp_bld_tgsi.h
@@ -445,7 +445,7 @@ struct lp_build_tgsi_soa_context
    struct tgsi_declaration_sampler_view sv[PIPE_MAX_SHADER_SAMPLER_VIEWS];
 
    LLVMValueRef immediates[LP_MAX_TGSI_IMMEDIATES][TGSI_NUM_CHANNELS];
-   LLVMValueRef temps[LP_MAX_TGSI_TEMPS][TGSI_NUM_CHANNELS];
+   LLVMValueRef temps[LP_MAX_INLINED_TEMPS][TGSI_NUM_CHANNELS];
    LLVMValueRef addr[LP_MAX_TGSI_ADDRS][TGSI_NUM_CHANNELS];
    LLVMValueRef preds[LP_MAX_TGSI_PREDS][TGSI_NUM_CHANNELS];
 
@@ -537,7 +537,7 @@ struct lp_build_tgsi_aos_context
    struct lp_build_sampler_aos *sampler;
 
    LLVMValueRef immediates[LP_MAX_TGSI_IMMEDIATES];
-   LLVMValueRef temps[LP_MAX_TGSI_TEMPS];
+   LLVMValueRef temps[LP_MAX_INLINED_TEMPS];
    LLVMValueRef addr[LP_MAX_TGSI_ADDRS];
    LLVMValueRef preds[LP_MAX_TGSI_PREDS];
 
diff --git a/src/gallium/auxiliary/gallivm/lp_bld_tgsi_aos.c b/src/gallium/auxiliary/gallivm/lp_bld_tgsi_aos.c
index c51fde0..fd5df0e 100644
--- a/src/gallium/auxiliary/gallivm/lp_bld_tgsi_aos.c
+++ b/src/gallium/auxiliary/gallivm/lp_bld_tgsi_aos.c
@@ -406,7 +406,7 @@ lp_emit_declaration_aos(
    for (idx = first; idx <= last; ++idx) {
       switch (decl->Declaration.File) {
       case TGSI_FILE_TEMPORARY:
-         assert(idx < LP_MAX_TGSI_TEMPS);
+         assert(idx < LP_MAX_INLINED_TEMPS);
          if (bld->indirect_files & (1 << TGSI_FILE_TEMPORARY)) {
             LLVMValueRef array_size = lp_build_const_int32(gallivm, last + 1);
             bld->temps_array = lp_build_array_alloca(bld->bld_base.base.gallivm,
diff --git a/src/gallium/auxiliary/gallivm/lp_bld_tgsi_soa.c b/src/gallium/auxiliary/gallivm/lp_bld_tgsi_soa.c
index 9db41a9..3ba2031 100644
--- a/src/gallium/auxiliary/gallivm/lp_bld_tgsi_soa.c
+++ b/src/gallium/auxiliary/gallivm/lp_bld_tgsi_soa.c
@@ -2672,8 +2672,8 @@ lp_emit_declaration_soa(
       assert(last <= bld->bld_base.info->file_max[decl->Declaration.File]);
       switch (decl->Declaration.File) {
       case TGSI_FILE_TEMPORARY:
-         assert(idx < LP_MAX_TGSI_TEMPS);
          if (!(bld->indirect_files & (1 << TGSI_FILE_TEMPORARY))) {
+            assert(idx < LP_MAX_INLINED_TEMPS);
             for (i = 0; i < TGSI_NUM_CHANNELS; i++)
                bld->temps[idx][i] = lp_build_alloca(gallivm, vec_type, "temp");
          }
@@ -3621,6 +3621,15 @@ lp_build_tgsi_soa(struct gallivm_state *gallivm,
    bld.bld_base.info = info;
    bld.indirect_files = info->indirect_files;
 
+   /*
+    * If the number of temporaries is rather large then we just
+    * allocate them as an array right from the start and treat
+    * like indirect temporaries.
+    */
+   if (info->file_max[TGSI_FILE_TEMPORARY] >= LP_MAX_INLINED_TEMPS) {
+      bld.indirect_files |= (1 << TGSI_FILE_TEMPORARY);
+   }
+
    bld.bld_base.soa = TRUE;
    bld.bld_base.emit_debug = emit_debug;
    bld.bld_base.emit_fetch_funcs[TGSI_FILE_CONSTANT] = emit_fetch_constant;