MoonchildProductions#1251 - Part 23: Allow AMD64 build to work.

https://developer.mozilla.org/en-US/docs/Mozilla/Developer_guide/Build_Instructions/Compiling_32-bit_Firefox_on_a_Linux_64-bit_OS

Setting this up turned out to be easier than I thought it would be. All I had to do was apply these instructions in reverse and add the following to my .mozconfig file:

CC="gcc -m64"
CXX="g++ -m64"
AS="gas --64"
ac_add_options --target=x86_64-pc-solaris2.11
export PKG_CONFIG_PATH=/usr/lib/amd64/pkgconfig
ac_add_options --libdir=/usr/lib/amd64
ac_add_options --x-libraries=/usr/lib/amd64

Most of these changes were fairly trivial, just requiring me to make a few of the changes I made earlier conditional on a 32-bit build. The biggest challenge was figuring out why the JavaScript engine triggered a segfault everytime it tried to allocate memory. But this patch fixes it:

https://github.com/OpenIndiana/oi-userland/blob/oi/hipster/components/web/firefox/patches/patch-js_src_gc_Memory.cpp.patch

Turns out that Solaris on AMD64 handles memory management in a fairly unusual way with a segmented memory model, but it's not that different from what we see on other 64-bit processors. In fact, I saw a SPARC crash for a similar reason, and noticed that it looked just like mine except the numbers in the first segment were reversed. Having played around with hex editors before, I had a feeling I might be dealing with a little-endian version of a big-endian problem, but I didn't expect that knowledge to actually yield an easy solution.

https://bugzilla.mozilla.org/show_bug.cgi?id=577056

https://www.oracle.com/technetwork/server-storage/solaris10/solaris-memory-135224.html

As far as I can tell, this was the last barrier to an AMD64 Solaris build of Pale Moon.

2 files changed, 76 insertions, 1 deletions

diff --git a/xpcom/reflect/xptcall/md/unix/moz.build b/xpcom/reflect/xptcall/md/unix/moz.build
index a005880284..8ee7b31817 100644
--- a/xpcom/reflect/xptcall/md/unix/moz.build
+++ b/xpcom/reflect/xptcall/md/unix/moz.build
@@ -53,7 +53,7 @@ if CONFIG['OS_ARCH'] == 'SunOS':
     if CONFIG['OS_TEST'] == 'x86_64':
             SOURCES += [
                 'xptcinvoke_asm_x86_64_unix.S',
-                'xptcinvoke_x86_64_unix.cpp',
+                'xptcinvoke_x86_64_solaris.cpp',
                 'xptcstubs_x86_64_linux.cpp'
             ]
     elif '86' in CONFIG['OS_TEST']:
diff --git a/xpcom/reflect/xptcall/md/unix/xptcinvoke_x86_64_solaris.cpp b/xpcom/reflect/xptcall/md/unix/xptcinvoke_x86_64_solaris.cpp
new file mode 100644
index 0000000000..a9db2a693a
--- /dev/null
+++ b/xpcom/reflect/xptcall/md/unix/xptcinvoke_x86_64_solaris.cpp
@@ -0,0 +1,75 @@
+/* -*- Mode: C; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 4 -*-
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
+
+// Platform specific code to invoke XPCOM methods on native objects
+
+#include "xptcprivate.h"
+
+// 6 integral parameters are passed in registers, but 1 is |this| which isn't
+// considered here.
+const uint32_t GPR_COUNT = 5;
+
+// 8 floating point parameters are passed in SSE registers
+const uint32_t FPR_COUNT = 8;
+
+extern "C" void
+InvokeCopyToStack(uint64_t * gpregs, double * fpregs,
+                     uint32_t paramCount, nsXPTCVariant * s,
+                     uint64_t* d)
+{
+    uint32_t nr_gpr = 0u; // skip one GP register for 'that'
+    uint32_t nr_fpr = 0u;
+    uint64_t value = 0u;
+
+    for (uint32_t i = 0; i < paramCount; i++, s++) {
+        if (s->IsPtrData())
+            value = (uint64_t) s->ptr;
+        else {
+            switch (s->type) {
+            case nsXPTType::T_FLOAT:                                break;
+            case nsXPTType::T_DOUBLE:                               break;
+            case nsXPTType::T_I8:     value = s->val.i8;            break;
+            case nsXPTType::T_I16:    value = s->val.i16;           break;
+            case nsXPTType::T_I32:    value = s->val.i32;           break;
+            case nsXPTType::T_I64:    value = s->val.i64;           break;
+            case nsXPTType::T_U8:     value = s->val.u8;            break;
+            case nsXPTType::T_U16:    value = s->val.u16;           break;
+            case nsXPTType::T_U32:    value = s->val.u32;           break;
+            case nsXPTType::T_U64:    value = s->val.u64;           break;
+            case nsXPTType::T_BOOL:   value = s->val.b;             break;
+            case nsXPTType::T_CHAR:   value = s->val.c;             break;
+            case nsXPTType::T_WCHAR:  value = s->val.wc;            break;
+            default:                  value = (uint64_t) s->val.p;  break;
+            }
+        }
+
+        if (!s->IsPtrData() && s->type == nsXPTType::T_DOUBLE) {
+            if (nr_fpr < FPR_COUNT)
+                fpregs[nr_fpr++] = s->val.d;
+            else {
+                *((double *)d) = s->val.d;
+                d++;
+            }
+        }
+        else if (!s->IsPtrData() && s->type == nsXPTType::T_FLOAT) {
+            if (nr_fpr < FPR_COUNT)
+                // The value in %xmm register is already prepared to
+                // be retrieved as a float. Therefore, we pass the
+                // value verbatim, as a double without conversion.
+                fpregs[nr_fpr++] = s->val.d;
+            else {
+                *((float *)d) = s->val.f;
+                d++;
+            }
+        }
+        else {
+            if (nr_gpr < GPR_COUNT)
+                gpregs[nr_gpr++] = value;
+            else
+                *d++ = value;
+        }
+    }
+}