gcc: Enable Ada testing

3 files changed, 92 insertions, 17 deletions

diff --git a/open_issues/gcc.mdwn b/open_issues/gcc.mdwn
index 3627179b..194a401a 100644
--- a/open_issues/gcc.mdwn
+++ b/open_issues/gcc.mdwn
@@ -479,9 +479,10 @@ Last reviewed up to the Git mirror's commit
             0 0x102c000
             0 0x102e000
 
-  * `gcc/ada` (not all reviewed)
+  * `gcc/ada`, `gcc/testsuite/ada`, `gcc/testsuite/gnat.dg`, `gnattools`,
+    `libada` (not all reviewed)
 
-      * [[Ada (GNAT)|gnat]] support is work in progress.
+    See [[Ada (GNAT)|gnat]].
 
   * `gcc/go`, `gcc/testsuite/go.test`, `libgo` (not reviewed)
 
@@ -992,13 +993,14 @@ Last reviewed up to the Git mirror's commit
 
 Here's a log of a GCC build run; this is from Git commit
 17f0f5acdcfb9181203ca9b6d8f8cbc64e5ce843 (2016-11-30),
+plus *GNU Hurd Ada fixes*,
 plus *libstdc++ RLIMIT changes* ([[!message-id
 "8760n66mky.fsf@euler.schwinge.homeip.net"]]), run on kepler.SCHWINGE
 and laplace.SCHWINGE.
 
     $ export LC_ALL=C
     $ (cd ../master/ && contrib/gcc_update --touch)
-    $ ../master/configure --prefix="$PWD".install SHELL=/bin/dash CC=gcc-6 CXX=g++-6 --enable-languages=all --disable-libcilkrts --disable-libsanitizer --disable-libvtv --disable-libmpx --with-native-system-header-dir=/usr/include --enable-multiarch 2>&1 | tee log_build
+    $ ../master/configure --prefix="$PWD".install SHELL=/bin/dash CC=gcc-6 CXX=g++-6 --enable-languages=all,ada --disable-libcilkrts --disable-libsanitizer --disable-libvtv --disable-libmpx --with-native-system-header-dir=/usr/include --enable-multiarch 2>&1 | tee log_build
     [...]
     $ make 2>&1 | tee log_build_
     [...]
@@ -1014,8 +1016,8 @@ until these is auto-detected/fixed/ported for GNU Hurd.
 `/include` for GNU Hurd.  Explicit `--enable-multiarch` because *auto check [is
 disabled if] configured with --native-system-header-dir*.
 
-This takes up around 3.9 GiB, and runs for [[3 h 57 min|performance#measure]]
-on kepler.SCHWINGE and [[4 h 37 min|performance#measure]] on laplace.SCHWINGE.
+This takes up around 4.6 GiB, and runs for [[5 h 16 min|performance#measure]]
+on kepler.SCHWINGE and [[5 h 14 min|performance#measure]] on laplace.SCHWINGE.
 
 <!--
 
@@ -1132,8 +1134,8 @@ on kepler.SCHWINGE and [[4 h 37 min|performance#measure]] on laplace.SCHWINGE.
     $ make install 2>&1 | tee log_install
     [...]
 
-This takes up around 0.9 GiB, and runs for [[1 min|performance#measure]] on
-kepler.SCHWINGE and [[3 min|performance#measure]] on laplace.SCHWINGE.
+This takes up around 1.3 GiB, and runs for [[1 min|performance#measure]] on
+kepler.SCHWINGE and [[5 min|performance#measure]] on laplace.SCHWINGE.
 
 
 ## Analysis
@@ -1152,8 +1154,8 @@ kepler.SCHWINGE and [[3 min|performance#measure]] on laplace.SCHWINGE.
     $ make -k check 2>&1 | tee log_test
     [...]
 
-This runs for [[5 h 43 min|performance#measure]] on kepler.SCHWINGE and [[12 h
-36 min|performance#measure]] on laplace.SCHWINGE.
+This runs for [[6 h 12 min|performance#measure]] on kepler.SCHWINGE and [[13 h
+34 min|performance#measure]] on laplace.SCHWINGE.
 
 
 ## Analysis
@@ -1447,9 +1449,6 @@ This runs for [[5 h 43 min|performance#measure]] on kepler.SCHWINGE and [[12 h
 
     2015-11-03
 
-    With work-in-progress [[Ada (GNAT)|gnat]] support (but that's not relevant
-    here).
-
         splitting /home/thomas/tmp/gcc/hurd/master.build/gcc/testsuite/ada/acats/tests/cb/cb1010a.ada into:
            cb1010a.adb
         BUILD cb1010a.adb
diff --git a/open_issues/gnat.mdwn b/open_issues/gnat.mdwn
index 1ab0866f..9c255ed5 100644
--- a/open_issues/gnat.mdwn
+++ b/open_issues/gnat.mdwn
@@ -125,6 +125,82 @@ GCC trunk r234252.
 Still incomplete?  (Changes to generic files?)
 
 
+### 2016-12-14
+
+Based on commit 17f0f5acdcfb9181203ca9b6d8f8cbc64e5ce843, [[tschwinge]] tried
+to build with `--enable-languages=all,ada`, but both with `--disable-bootstrap`
+and `--enable-bootstrap` we run into:
+
+    [...]
+    make[3]: Entering directory '/media/erich/home/thomas/tmp/gcc/hurd/master.build/gcc/ada/tools'
+    ../../gnatbind -I- -I../rts -I. -I/home/thomas/tmp/gcc/hurd/master/gcc/ada -o b_gnatm.adb gnatmake.ali
+    error: "system.ads" has restriction No_Implicit_Dynamic_Code
+    error: but the following files violate this restriction:
+    error:   "make.adb"
+    error:   "makeutl.adb"
+    error:   "prj.adb"
+    error:   "prj-env.adb"
+    error:   "prj-conf.adb"
+    error:   "prj-nmsc.adb"
+    ../gcc-interface/Makefile:2987: recipe for target 'b_gnatm.adb' failed
+    make[3]: *** [b_gnatm.adb] Error 5
+    make[3]: Leaving directory '/media/erich/home/thomas/tmp/gcc/hurd/master.build/gcc/ada/tools'
+    Makefile:194: recipe for target 'gnattools-native' failed
+    make[2]: *** [gnattools-native] Error 2
+    make[2]: Leaving directory '/media/erich/home/thomas/tmp/gcc/hurd/master.build/gnattools'
+    Makefile:9650: recipe for target 'all-gnattools' failed
+    make[1]: *** [all-gnattools] Error 2
+    make[1]: Leaving directory '/media/erich/home/thomas/tmp/gcc/hurd/master.build'
+    Makefile:895: recipe for target 'all' failed
+    make: *** [all] Error 2
+
+(Partial) review of differences:
+
+`gcc/ada/gcc-interface/Makefile.in`, Linux vs. Hurd
+
+    LIBGNAT_TARGET_PAIRS = \
+      a-intnam.ads<a-intnam-linux.ads vs. a-intnam.ads<a-intnam-freebsd.ads
+      a-synbar.adb<a-synbar-posix.adb vs. nil
+      a-synbar.ads<a-synbar-posix.ads vs. nil
+      s-inmaop.adb<s-inmaop-posix.adb
+      s-intman.adb<s-intman-posix.adb
+      s-tpopsp.adb<s-tpopsp-tls.adb vs. s-tpopsp.adb<s-tpopsp-posix-foreign.adb
+      g-sercom.adb<g-sercom-linux.adb vs. nil
+      a-exetim.adb<a-exetim-posix.adb vs. nil
+      a-exetim.ads<a-exetim-default.ads vs. nil
+      s-linux.ads<s-linux.ads vs. nil
+      s-osinte.adb<s-osinte-posix.adb vs. s-osinte.adb<s-osinte-gnu.adb
+      $(ATOMICS_TARGET_PAIRS)
+      system.ads<system-linux-x86.ads vs. system.ads<system-freebsd-x86.ads
+      $(X86_TARGET_PAIRS)
+      s-mudido.adb<s-mudido-affinity.adb vs. nil
+      s-osinte.ads<s-osinte-linux.ads vs. s-osinte.ads<s-osinte-gnu.ads
+      s-osprim.adb<s-osprim-posix.adb
+      s-taprop.adb<s-taprop-linux.adb vs. s-taprop.adb<s-taprop-posix.adb
+      s-tasinf.ads<s-tasinf-linux.ads vs. nil
+      s-tasinf.adb<s-tasinf-linux.adb vs. nil
+      s-taspri.ads<s-taspri-posix.ads
+    
+    EH_MECHANISM=-gcc
+    THREADSLIB = -lpthread -lrt vs. THREADSLIB = -lpthread
+    EXTRA_GNATRTL_NONTASKING_OBJS += g-sse.o g-ssvety.o
+    EXTRA_GNATRTL_TASKING_OBJS=s-linux.o a-exetim.o vs. nil
+    
+    TOOLS_TARGET_PAIRS = \
+      mlib-tgt-specific.adb<mlib-tgt-specific-linux.adb
+      indepsw.adb<indepsw-gnu.adb
+    
+    GNATLIB_SHARED = gnatlib-shared-dual
+    MISCLIB = -ldl vs. nil
+    GMEM_LIB = gmemlib
+    LIBRARY_VERSION := $(LIB_VERSION)
+    
+    TOOLS1_LIBS=-ldl vs. nil
+
+Eventually [[tschwinge]] found <http://gcc.gnu.org/PR57188>, and similarly
+fixed the issue.
+
+
 ## `getcontext`/`makecontext`/`setcontext`/`swapcontext` usage analysis
 
 In context of [[glibc/t/tls-threadvar]].  Looking at GCC trunk commit
diff --git a/toolchain/logs b/toolchain/logs
-Subproject 1532082e156ed4229d8f7002b733deddb7da34a
+Subproject b9d5ce62343d50df7800d02ae97790ea9e530bd