All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrei Borzenkov <arvidjaar@gmail.com>
To: Mark H Weaver <mhw@netris.org>
Cc: The development of GNU GRUB <grub-devel@gnu.org>
Subject: Re: [PATCH] Ensure that MIPS target code is compiled for the O32 ABI.
Date: Sun, 13 Sep 2015 17:10:37 +0300	[thread overview]
Message-ID: <55F583DD.2020805@gmail.com> (raw)
In-Reply-To: <87a8srjb1h.fsf@netris.org>

[-- Attachment #1: Type: text/plain, Size: 2979 bytes --]

13.09.2015 05:41, Mark H Weaver пишет:
> Andrei Borzenkov <arvidjaar@gmail.com> writes:
>
>>> ld: /tmp/conftest-efc530.o: compiled for a big endian system and target is little endian
>>> ld: /tmp/conftest-efc530.o: endianness incompatible with that of the selected emulation
>>
>> This is the same problem we had on powerpc. Your binutils are set to
>> use little-endian output by default; we need to explicitly enable
>> either big or little endian in this case as well. Could you try
>> attached patch to verify?
>
> I'm glad to report that with your new patch, GRUB successfully builds on
> GuixSD mips64el with TARGET_CC=clang.  I haven't yet tried booting it
> though.
>

Did you have chance to test whether compiled grub actually work?

>>> Your patch works for me with gcc, although config.log shows this warning
>>> when building the test program in your new code with the correct flags
>>> for gcc (-mabi=32):
>>>
>>>     ld: warning: cannot find entry symbol __start; defaulting to 00000000004000d0
>>>
>>
>> Well, I just reused existing checks in this case; we worry about it later.
>
> Okay.  I guess the name of the start symbol depends on the platform and
> linker.  E.g. boot/mips/loongson/fwstart.S has this:
>
> --8<---------------cut here---------------start------------->8---
> 	.global start,_start,__start
> start:
> _start:
> __start:	
> --8<---------------cut here---------------end--------------->8---
>
> Other platforms have only '_start', or both 'start' and '_start'.
>
>>> I also worry that linking an O32 executable might be fragile on a pure
>>> non-multilib N32 system such as GNU Guix, where we lack even the gcc
>>> runtime library (libgcc_s) for O32.  For now it seems to work, but I
>>> wonder if it might be more robust to do the test compilation with -c and
>>> -freestanding?
>>>
>>
>> It is using -nostdlib; is it not enough?
>
> The warning above makes me doubt the portability and future-proofness of
> the link step of this test.  This problem could be avoided entirely by
> omitting the link step.  The link step serves no purpose in this test,
> and is a potential source of portability problems.
>

This step helped us to catch endianness mismatch; we want to ensure in 
this case that full toolchain works to produce final binary.

> Having said that, if you prefer to keep the link step, I guess it will
> probably be okay in practice to define 'start', '_start', and '__start'
> as above.  Do as you think best.
>


Please test two attached patches. They separate o32 test from endianness 
test. I compile-tested them using cross-binutils for mips and 
powerpc64le to compile for mipsel and powerpc. The first patch also adds 
additional _start and __start symbols to endianness tests to avoid 
compiler warnings.

I am interested in both native and cross compilation - i.e. please test 
for both --target=mipsel and --target=mips both with gcc and clang. 
config.log from all runs would be interesting even for successful cases.

[-- Attachment #2: 0001-configure-try-to-find-options-to-force-endian-on-MIP.patch --]
[-- Type: text/x-patch, Size: 3108 bytes --]

From ca80d768a43685e3e67fe9092dc10d9c574ee596 Mon Sep 17 00:00:00 2001
From: Andrei Borzenkov <arvidjaar@gmail.com>
Date: Sun, 13 Sep 2015 16:51:13 +0300
Subject: [PATCH 1/2] configure: try to find options to force endian on MIPS

---
 configure.ac | 43 ++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 40 insertions(+), 3 deletions(-)

diff --git a/configure.ac b/configure.ac
index c864311..4760fe8 100644
--- a/configure.ac
+++ b/configure.ac
@@ -562,11 +562,11 @@ AC_COMPILE_IFELSE(
 ]])],
 [grub_cv_cc_target_clang=no], [grub_cv_cc_target_clang=yes])])
 
-if test x$target_cpu = xpowerpc; then
+if test x$target_cpu = xpowerpc -o x$target_cpu = xmips; then
   AC_CACHE_CHECK([for options to get big-endian compilation], grub_cv_target_cc_big_endian, [
     grub_cv_target_cc_big_endian=no
-    for cand in "-target powerpc -Wl,-EB" "-target powerpc" \
-		"-target powerpc-linux-gnu -Wl,-EB" "-target powerpc-linux-gnu" \
+    for cand in "-target $target_cpu -Wl,-EB" "-target $target_cpu" \
+		"-target $target_cpu-linux-gnu -Wl,-EB" "-target $target_cpu-linux-gnu" \
 		"-mbig-endian"; do
       if test x"$grub_cv_target_cc_big_endian" != xno ; then
         break
@@ -577,6 +577,8 @@ if test x$target_cpu = xpowerpc; then
 #error still little endian
 #endif
 asm (".globl start; start:");
+asm (".globl _start; _start:");
+asm (".globl __start; __start:");
 void __main (void);
 void __main (void) {}
 int main (void);
@@ -595,6 +597,41 @@ int main (void);
   TARGET_CPPFLAGS="$TARGET_CPPFLAGS $skip_linkflags"
   TARGET_CCASFLAGS="$TARGET_CCASFLAGS $skip_linkflags"
   TARGET_LDFLAGS="$TARGET_LDFLAGS $grub_cv_target_cc_big_endian"
+elif test x$target_cpu = xmipsel; then
+  AC_CACHE_CHECK([for options to get little-endian compilation], grub_cv_target_cc_little_endian, [
+    grub_cv_target_cc_little_endian=no
+    for cand in "-target $target_cpu -Wl,-EL" "-target $target_cpu" \
+		"-target $target_cpu-linux-gnu -Wl,-EL" "-target $target_cpu-linux-gnu" \
+		"-mlittle-endian"; do
+      if test x"$grub_cv_target_cc_little_endian" != xno ; then
+        break
+      fi
+      CFLAGS="$TARGET_CFLAGS $cand -Werror"
+      AC_LINK_IFELSE([AC_LANG_PROGRAM([[
+#if defined(__BYTE_ORDER__) && defined(__ORDER_BIG_ENDIAN__) && (__ORDER_BIG_ENDIAN__ == __BYTE_ORDER__)
+#error still big endian
+#endif
+asm (".globl start; start:");
+asm (".globl _start; _start:");
+asm (".globl __start; __start:");
+void __main (void);
+void __main (void) {}
+int main (void);
+]], [[]])],
+		        [grub_cv_target_cc_little_endian="$cand"], [])
+    done
+  ])
+
+  if test x"$grub_cv_target_cc_little_endian" = xno ; then
+    AC_MSG_ERROR([could not force little-endian])
+  fi
+
+  skip_linkflags="$(echo "$grub_cv_target_cc_little_endian"|sed 's@-Wl,-EL@@')"
+
+  TARGET_CFLAGS="$TARGET_CFLAGS $skip_linkflags"
+  TARGET_CPPFLAGS="$TARGET_CPPFLAGS $skip_linkflags"
+  TARGET_CCASFLAGS="$TARGET_CCASFLAGS $skip_linkflags"
+  TARGET_LDFLAGS="$TARGET_LDFLAGS $grub_cv_target_cc_little_endian"
 fi
 
 AC_CACHE_CHECK([for options to compile assembly], [grub_cv_cc_target_asm_compile], [
-- 
1.9.1


[-- Attachment #3: 0002-configure-try-to-force-o32-ABI-on-MIPS.patch --]
[-- Type: text/x-patch, Size: 2195 bytes --]

From b69cb301dca85c60b25f0118e5532a2b12f7bae6 Mon Sep 17 00:00:00 2001
From: Andrei Borzenkov <arvidjaar@gmail.com>
Date: Sun, 13 Sep 2015 16:51:13 +0300
Subject: [PATCH 2/2] configure: try to force o32 ABI on MIPS

GRUB expects o32 ABI, in particular used assembly is valid only in this mode.
Some systems (e.g. GNU Guix) default to using newer n64 or n32 ABI. Try to
find suitable options to force o32.

For GCC this is simply -mabi=32. While clang supports this option as well,
o32 ABI is valid for MIPS target and n32/64 ABI are valid for MIPS64 target
only, so use "-target mips/mipsel -mabi=32".

Reported-By: Mark H Weaver <mhw@netris.org>
Also-By: Mark H Weaver <mhw@netris.org>
---
 configure.ac | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/configure.ac b/configure.ac
index 4760fe8..7b4e5bc 100644
--- a/configure.ac
+++ b/configure.ac
@@ -634,6 +634,37 @@ int main (void);
   TARGET_LDFLAGS="$TARGET_LDFLAGS $grub_cv_target_cc_little_endian"
 fi
 
+if test "x$target_cpu" = xmips || test "x$target_cpu" = xmipsel ; then
+  AC_CACHE_CHECK([for options to force MIPS o32 ABI], grub_cv_target_cc_mips_o32_abi, [
+    grub_cv_target_cc_mips_o32_abi=no
+    for arg in "" "-mabi=32" "-target $target_cpu -mabi=32" ; do
+      if test x"$grub_cv_target_cc_mips_o32_abi" != xno ; then
+        break
+      fi
+      CFLAGS="$TARGET_CFLAGS $arg -Werror"
+      AC_LINK_IFELSE([AC_LANG_PROGRAM([[
+#if !defined(_ABIO32) || !defined(_MIPS_SIM) || (_MIPS_SIM != _ABIO32)
+#error not o32 ABI
+#endif
+asm (".globl start; start:");
+asm (".globl _start; _start:");
+asm (".globl __start; __start:");
+void __main (void);
+void __main (void) {}
+int main (void);
+]], [[]])],
+		        [grub_cv_target_cc_mips_o32_abi="$arg"], [])
+    done
+  ])
+
+  if test x"$grub_cv_target_cc_mips_o32_abi" = xno ; then
+    AC_MSG_ERROR([could not force MIPS o32 ABI])
+  fi
+
+  TARGET_CFLAGS="$TARGET_CFLAGS $grub_cv_target_cc_mips_o32_abi"
+  TARGET_CCASFLAGS="$TARGET_CCASFLAGS $grub_cv_target_cc_mips_o32_abi"
+fi
+
 AC_CACHE_CHECK([for options to compile assembly], [grub_cv_cc_target_asm_compile], [
 test_program=
 case "x$target_cpu-$platform" in
-- 
1.9.1


  parent reply	other threads:[~2015-09-13 14:10 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-19 18:35 [PATCH] Ensure that MIPS target code is compiled for the O32 ABI Mark H Weaver
2015-08-21  4:05 ` Andrei Borzenkov
2015-08-22 18:04   ` Mark H Weaver
2015-08-23  9:51     ` Andrei Borzenkov
2015-08-23 17:27       ` Mark H Weaver
2015-08-24  6:50       ` [PATCH v2] " Mark H Weaver
2015-09-08 17:11         ` Vladimir 'φ-coder/phcoder' Serbinenko
2015-09-09  1:37           ` Mark H Weaver
2015-09-13  6:32           ` Andrei Borzenkov
2015-10-09 21:14             ` Vladimir 'φ-coder/phcoder' Serbinenko
2015-10-10  0:28               ` Vladimir 'φ-coder/phcoder' Serbinenko
2015-10-14  4:49                 ` Andrei Borzenkov
2015-10-14 12:16                   ` Vladimir 'phcoder' Serbinenko
2015-10-14 17:57                     ` Andrei Borzenkov
2015-08-24  7:19     ` [PATCH] " Andrei Borzenkov
2015-08-24  7:57       ` Mark H Weaver
2015-08-24  8:12         ` Andrei Borzenkov
2015-08-24  9:27           ` Andrei Borzenkov
2015-08-25 18:18             ` Mark H Weaver
2015-09-11 15:06               ` Andrei Borzenkov
2015-09-13  2:41                 ` Mark H Weaver
2015-09-13  3:46                   ` Mark H Weaver
2015-09-13 14:10                   ` Andrei Borzenkov [this message]
2015-09-13 23:37                     ` Mark H Weaver
2015-09-20  4:44                       ` Andrei Borzenkov
2015-09-21 16:09                         ` Mark H Weaver
2015-09-21 17:56                           ` Andrei Borzenkov
2015-09-22 15:03                             ` Mark H Weaver
2015-09-22 17:26                               ` Andrei Borzenkov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=55F583DD.2020805@gmail.com \
    --to=arvidjaar@gmail.com \
    --cc=grub-devel@gnu.org \
    --cc=mhw@netris.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.