xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Daniel Kiper <daniel.kiper@oracle.com>
To: xen-devel@lists.xenproject.org
Cc: jgross@suse.com, sstabellini@kernel.org,
	andrew.cooper3@citrix.com, cardoe@cardoe.com,
	pgnet.dev@gmail.com, ning.sun@intel.com, david.vrabel@citrix.com,
	jbeulich@suse.com, qiaowei.ren@intel.com,
	richard.l.maliszewski@intel.com, gang.wei@intel.com,
	fu.wei@linaro.org
Subject: [PATCH v4 03/19] x86/boot: create *.lnk files with linker script
Date: Sat,  6 Aug 2016 01:04:26 +0200	[thread overview]
Message-ID: <1470438282-4226-4-git-send-email-daniel.kiper@oracle.com> (raw)
In-Reply-To: <1470438282-4226-1-git-send-email-daniel.kiper@oracle.com>

Newer GCC (e.g. gcc version 5.1.1 20150618 (Red Hat 5.1.1-4) (GCC)) does
some code optimizations by creating data sections (e.g. jump addresses
for C switch/case are calculated using data in .rodata section). This
thing is not accepted by *.lnk build recipe which requires that only .text
section lives in output. Potentially we can inhibit this GCC behavior by
using special options, e.g. -fno-tree-switch-conversion. However, this
does not guarantee that in the future new similar optimizations or anything
else which creates not accepted sections will not break our build recipes
again. I do not mention that probably this is not good idea to just disable
random optimizations. So, take over full control on *.lnk linking process
by using linker script and merge required text and data sections into one
.text section.

Additionally, remove .got.plt section which is not used in our final code.

Signed-off-by: Daniel Kiper <daniel.kiper@oracle.com>
---
v4 - suggestions/fixes:
   - remove my name from copyright (Oracle requirement)
     (suggested by Konrad Rzeszutek Wilk),
   - improve comments,
     (suggested by Jan Beulich),
   - improve commit message
     (suggested by Jan Beulich).
---
 xen/arch/x86/boot/build32.lds |   51 +++++++++++++++++++++++++++++++++++++++++
 xen/arch/x86/boot/build32.mk  |   10 +++++---
 2 files changed, 58 insertions(+), 3 deletions(-)
 create mode 100644 xen/arch/x86/boot/build32.lds

diff --git a/xen/arch/x86/boot/build32.lds b/xen/arch/x86/boot/build32.lds
new file mode 100644
index 0000000..b14c7d5
--- /dev/null
+++ b/xen/arch/x86/boot/build32.lds
@@ -0,0 +1,51 @@
+/*
+ * Copyright (c) 2016 Oracle and/or its affiliates. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+ENTRY(_start)
+
+SECTIONS
+{
+  /* Merge code and data into one section. */
+  .text : {
+        *(.text)
+        *(.text.*)
+        *(.rodata)
+  }
+
+  /DISCARD/ : {
+        /*
+         * PIC/PIE executable contains .got.plt section even
+         * if it is not linked with dynamic libraries. In such
+         * case it is just placeholder for _GLOBAL_OFFSET_TABLE_
+         * symbol and .PLT0. .PLT0 is filled by dynamic linker
+         * and our code is not supposed to be loaded by dynamic
+         * linker. So, from our point of view .PLT0 is unused.
+         * This means that there is pretty good chance that
+         * we can safely drop .got.plt as a whole here. Sadly
+         * this is not true. _GLOBAL_OFFSET_TABLE_ is used as
+         * a reference for relative addressing (and only for
+         * that thing) and ld complains if we remove .got.plt
+         * section here because it cannot find required symbol.
+         * However, _GLOBAL_OFFSET_TABLE_ is no longer needed
+         * in final output. So, drop .got.plt section during
+         * conversion to plain binary format.
+         *
+         * Please check build32.mk for more details.
+         */
+        /* *(.got.plt) */
+  }
+}
diff --git a/xen/arch/x86/boot/build32.mk b/xen/arch/x86/boot/build32.mk
index 4a7d388..eb02b4b 100644
--- a/xen/arch/x86/boot/build32.mk
+++ b/xen/arch/x86/boot/build32.mk
@@ -12,20 +12,24 @@ CFLAGS := $(filter-out -flto,$(CFLAGS))
 	(od -v -t x $< | tr -s ' ' | awk 'NR > 1 {print s} {s=$$0}' | \
 	sed 's/ /,0x/g' | sed 's/,0x$$//' | sed 's/^[0-9]*,/ .long /') >$@
 
+#
+# Drop .got.plt during conversion to plain binary format.
+# Please check build32.lds for more details.
+#
 %.bin: %.lnk
-	$(OBJCOPY) -O binary $< $@
+	$(OBJCOPY) -O binary -R .got.plt $< $@
 
 %.lnk: %.o
 	$(OBJDUMP) -h $< | sed -n '/[0-9]/{s,00*,0,g;p;}' |\
 		while read idx name sz rest; do \
 			case "$$name" in \
-			.data|.data.*|.rodata|.rodata.*|.bss|.bss.*) \
+			.data|.data.*|.rodata.*|.bss|.bss.*) \
 				test $$sz != 0 || continue; \
 				echo "Error: non-empty $$name: 0x$$sz" >&2; \
 				exit $$(expr $$idx + 1);; \
 			esac; \
 		done
-	$(LD) $(LDFLAGS_DIRECT) -N -Ttext 0 -o $@ $<
+	$(LD) $(LDFLAGS_DIRECT) -N -T build32.lds -o $@ $<
 
 %.o: %.c
 	$(CC) $(CFLAGS) -c -fpic $< -o $@
-- 
1.7.10.4


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

  parent reply	other threads:[~2016-08-05 23:05 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-05 23:04 [PATCH v4 00/19] x86: multiboot2 protocol support Daniel Kiper
2016-08-05 23:04 ` [PATCH v4 01/19] x86: allow EFI reboot method neither on EFI platforms Daniel Kiper
2016-08-09 12:08   ` Jan Beulich
2016-08-05 23:04 ` [PATCH v4 02/19] x86/boot: remove multiboot1_header_end from symbol table Daniel Kiper
2016-08-09 13:24   ` Andrew Cooper
2016-08-09 13:52     ` Jan Beulich
2016-08-09 14:09       ` Andrew Cooper
2016-08-09 14:30         ` Jan Beulich
2016-08-05 23:04 ` Daniel Kiper [this message]
2016-08-11 13:40   ` [PATCH v4 03/19] x86/boot: create *.lnk files with linker script Jan Beulich
2016-08-05 23:04 ` [PATCH v4 04/19] x86/boot/reloc: reduce assembly usage as much as possible Daniel Kiper
2016-08-11 13:56   ` Jan Beulich
2016-08-05 23:04 ` [PATCH v4 05/19] x86/boot: call reloc() using stdcall calling convention Daniel Kiper
2016-08-11 13:59   ` Jan Beulich
2016-08-05 23:04 ` [PATCH v4 06/19] x86/boot/reloc: create generic alloc and copy functions Daniel Kiper
2016-08-11 14:12   ` Jan Beulich
2016-08-11 14:17     ` Jan Beulich
2016-08-18  8:53       ` Daniel Kiper
2016-08-18  9:41         ` Jan Beulich
2016-08-18 12:18           ` Daniel Kiper
2016-08-18 13:21             ` Jan Beulich
2016-08-05 23:04 ` [PATCH v4 07/19] x86/boot: use %ecx instead of %eax Daniel Kiper
2016-08-05 23:04 ` [PATCH v4 08/19] x86/boot/reloc: Rename some variables and rearrange code a bit Daniel Kiper
2016-08-11 14:16   ` Jan Beulich
2016-08-05 23:04 ` [PATCH v4 09/19] x86: add multiboot2 protocol support Daniel Kiper
2016-08-17 15:39   ` Jan Beulich
2016-08-18  9:23     ` Daniel Kiper
2016-08-18  9:43       ` Jan Beulich
2016-08-18 11:41         ` Daniel Kiper
2016-08-18 13:19           ` Jan Beulich
2016-08-05 23:04 ` [PATCH v4 10/19] efi: move efi struct initialization to xen/common/lib.c Daniel Kiper
2016-08-17 15:56   ` Jan Beulich
2016-08-18 10:17     ` Daniel Kiper
2016-08-18 11:17       ` Jan Beulich
2016-08-05 23:04 ` [PATCH v4 11/19] efi: create efi_enabled() Daniel Kiper
2016-08-17 16:08   ` Jan Beulich
2016-08-05 23:04 ` [PATCH v4 12/19] efi: introduce EFI_RS to ease control on runtime services usage Daniel Kiper
2016-08-17 16:12   ` Jan Beulich
2016-08-18 10:30     ` Daniel Kiper
2016-08-18 11:18       ` Jan Beulich
2016-08-18 11:49         ` Daniel Kiper
2016-08-05 23:04 ` [PATCH v4 13/19] efi: EFI_RS bit in efi.flags must be controlled by efi=[no-]rs command line argument Daniel Kiper
2016-08-05 23:04 ` [PATCH v4 14/19] efi: build xen.gz with EFI code Daniel Kiper
2016-08-19  9:24   ` Jan Beulich
2016-08-19 10:09     ` Daniel Kiper
2016-08-05 23:04 ` [PATCH v4 15/19] x86/efi: create new early memory allocator Daniel Kiper
2016-08-05 23:04 ` [PATCH v4 16/19] x86: add multiboot2 protocol support for EFI platforms Daniel Kiper
2016-08-05 23:04 ` [PATCH v4 17/19] x86/boot: implement early command line parser in C Daniel Kiper
2016-08-05 23:04 ` [PATCH v4 18/19] x86: make Xen early boot code relocatable Daniel Kiper
2016-08-05 23:04 ` [PATCH v4 19/19] x86: add multiboot2 protocol support for relocatable images Daniel Kiper

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=1470438282-4226-4-git-send-email-daniel.kiper@oracle.com \
    --to=daniel.kiper@oracle.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=cardoe@cardoe.com \
    --cc=david.vrabel@citrix.com \
    --cc=fu.wei@linaro.org \
    --cc=gang.wei@intel.com \
    --cc=jbeulich@suse.com \
    --cc=jgross@suse.com \
    --cc=ning.sun@intel.com \
    --cc=pgnet.dev@gmail.com \
    --cc=qiaowei.ren@intel.com \
    --cc=richard.l.maliszewski@intel.com \
    --cc=sstabellini@kernel.org \
    --cc=xen-devel@lists.xenproject.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).