blob: 0787ef27d271500ad6f2b72ed06a536c3a63fe19 [file] [log] [blame]
From 0a163f60b4a316c4b6f1726a71c84755f3bd85e7 Mon Sep 17 00:00:00 2001
From: Martin Jansa <martin.jansa@lge.com>
Date: Wed, 16 Sep 2020 04:36:04 -0700
Subject: [PATCH] Revert "ld.hugetlbfs: fix -Ttext-segment argument on AArch64"
This reverts commit 852dcc963ce44861ed7c4e225aa92ff2b5b43579.
* works around
ERROR: libhugetlbfs-1_2.23-r0 do_populate_sysroot: Fatal errors occurred in subprocesses:
Command '['arm-oe-linux-gnueabi-strip', '--remove-section=.comment', '--remove-section=.note', 'libhugetlbfs/1_2.23-r0/sysroot-destdir/usr/lib/libhugetlbfs/tests/obj32/linkhuge_rw']' returned non-zero exit status 1.
Subprocess output:arm-oe-linux-gnueabi-strip:
libhugetlbfs/1_2.23-r0/sysroot-destdir/usr/lib/libhugetlbfs/tests/obj32/stmuFa58: error: PHDR segment not covered by LOAD segment
arm-oe-linux-gnueabi-strip: libhugetlbfs/1_2.23-r0/sysroot-destdir/usr/lib/libhugetlbfs/tests/obj32/stmuFa58[.interp]: file format not recognized
I don't see anything suspicious in the build (the same cmdline as in 2.22
version), but it uses
libhugetlbfs/1_2.22-r0-old/temp/log.do_compile:arm-oe-linux-gnueabi-gcc
-mthumb -mfpu=neon-vfpv4 -mfloat-abi=hard -mcpu=cortex-a7 -Wl,-O1
-Wl,--hash-style=gnu -Wl,--as-needed -Wl,-z,relro,-z,now
-fstack-protector-strong -D_FORTIFY_SOURCE=2 -Wformat -Wformat-security
-Werror=format-security -Werror=return-type
--sysroot=/jenkins/mjansa/build/ros/oe-melodic-gatesgarth/libhugetlbfs/1_2.22-r0/recipe-sysroot
-I.. -O2 -Wall -g -o obj32/linkhuge_rw.o -c linkhuge_rw.c
libhugetlbfs/1_2.22-r0-old/temp/log.do_compile:arm-oe-linux-gnueabi-gcc
-mthumb -mfpu=neon-vfpv4 -mfloat-abi=hard -mcpu=cortex-a7 -Wl,-O1
-Wl,--hash-style=gnu -Wl,--as-needed -Wl,-z,relro,-z,now
-fstack-protector-strong -D_FORTIFY_SOURCE=2 -Wformat -Wformat-security
-Werror=format-security -Werror=return-type
--sysroot=/jenkins/mjansa/build/ros/oe-melodic-gatesgarth/libhugetlbfs/1_2.22-r0/recipe-sysroot
-B./obj32 -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed -Wl,-z,relro,-z,now
-ldl -L../obj32 -o obj32/linkhuge_rw -Wl,--no-as-needed -lpthread -ldl
-lhugetlbfs_privutils -Wl,--hugetlbfs-align obj32/linkhuge_rw.o
obj32/testutils.o
libhugetlbfs/1_2.23-r0-new/temp/log.do_compile:arm-oe-linux-gnueabi-gcc
-mthumb -mfpu=neon-vfpv4 -mfloat-abi=hard -mcpu=cortex-a7 -Wl,-O1
-Wl,--hash-style=gnu -Wl,--as-needed -Wl,-z,relro,-z,now
-fstack-protector-strong -D_FORTIFY_SOURCE=2 -Wformat -Wformat-security
-Werror=format-security -Werror=return-type
--sysroot=/jenkins/mjansa/build/ros/oe-melodic-gatesgarth/libhugetlbfs/1_2.23-r0/recipe-sysroot
-I.. -O2 -Wall -g -o obj32/linkhuge_rw.o -c linkhuge_rw.c
libhugetlbfs/1_2.23-r0-new/temp/log.do_compile:arm-oe-linux-gnueabi-gcc
-mthumb -mfpu=neon-vfpv4 -mfloat-abi=hard -mcpu=cortex-a7 -Wl,-O1
-Wl,--hash-style=gnu -Wl,--as-needed -Wl,-z,relro,-z,now
-fstack-protector-strong -D_FORTIFY_SOURCE=2 -Wformat -Wformat-security
-Werror=format-security -Werror=return-type
--sysroot=/jenkins/mjansa/build/ros/oe-melodic-gatesgarth/libhugetlbfs/1_2.23-r0/recipe-sysroot
-B./obj32 -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed -Wl,-z,relro,-z,now
-ldl -L../obj32 -o obj32/linkhuge_rw -Wl,--no-as-needed -lpthread -ldl
-lhugetlbfs_privutils -Wl,--hugetlbfs-align obj32/linkhuge_rw.o
obj32/testutils.o
And the git log between 2.22 and 2.23 is also very short and looks
reasonable.
https://github.com/libhugetlbfs/libhugetlbfs/compare/2.22...2.23
When checking with readelf -l it also shows the error about PHDR segment:
arm-oe-linux-gnueabi-readelf -l
./1_2.22-r0-old/sysroot-destdir/usr/lib/libhugetlbfs/tests/obj32/linkhuge_rw
Elf file type is DYN (Shared object file)
Entry point 0x201105
There are 10 program headers, starting at offset 52
Program Headers:
Type Offset VirtAddr PhysAddr FileSiz MemSiz Flg Align
PHDR 0x000034 0x00200034 0x00200034 0x00140 0x00140 R 0x4
INTERP 0x000174 0x00200174 0x00200174 0x0001d 0x0001d R 0x1
[Requesting program interpreter: /usr/lib/ld-linux-armhf.so.3]
LOAD 0x000000 0x00200000 0x00200000 0x1222c 0x1222c R E 0x200000
LOAD 0x1ffdf0 0x005ffdf0 0x005ffdf0 0x102e0 0x202ec RW 0x200000
DYNAMIC 0x1ffdf8 0x005ffdf8 0x005ffdf8 0x00128 0x00128 RW 0x4
NOTE 0x000194 0x00200194 0x00200194 0x00044 0x00044 R 0x4
GNU_EH_FRAME 0x012224 0x00212224 0x00212224 0x00008 0x00008 R 0x4
GNU_STACK 0x000000 0x00000000 0x00000000 0x00000 0x00000 RW 0x10
EXIDX 0x001c5c 0x00201c5c 0x00201c5c 0x00008 0x00008 R 0x4
GNU_RELRO 0x1ffdf0 0x005ffdf0 0x005ffdf0 0x00210 0x00210 RW 0x4
Section to Segment mapping:
Segment Sections...
00
01 .interp
02 .interp .note.ABI-tag .note.gnu.build-id .dynsym .dynstr
.gnu.hash .gnu.version .gnu.version_r .rel.dyn .rel.plt .init .plt .text
.fini .ARM.extab .ARM.exidx .rodata .eh_frame .eh_frame_hdr
03 .fini_array .init_array .dynamic .got .data .bss
04 .dynamic
05 .note.ABI-tag .note.gnu.build-id
06 .eh_frame_hdr
07
08 .ARM.extab .ARM.exidx
09 .fini_array .init_array .dynamic .got
arm-oe-linux-gnueabi-readelf -l
./1_2.23-r0-new/sysroot-destdir/usr/lib/libhugetlbfs/tests/obj32/linkhuge_rw
Elf file type is DYN (Shared object file)
Entry point 0x31cd1
There are 10 program headers, starting at offset 52
Program Headers:
Type Offset VirtAddr PhysAddr FileSiz MemSiz Flg Align
PHDR 0x000000 0x00000000 0x00000000 0x00000 0x00000 R 0
readelf: Error: the PHDR segment is not covered by a LOAD segment
INTERP 0x030d40 0x00030d40 0x00030d40 0x0001d 0x0001d R 0x1
[Requesting program interpreter: /usr/lib/ld-linux-armhf.so.3]
LOAD 0x030d40 0x00030d40 0x00030d40 0x120b8 0x120b8 R E 0x200000
LOAD 0x1ffdf0 0x003ffdf0 0x003ffdf0 0x102e0 0x202ec RW 0x200000
DYNAMIC 0x1ffdf8 0x003ffdf8 0x003ffdf8 0x00128 0x00128 RW 0x4
NOTE 0x030d60 0x00030d60 0x00030d60 0x00044 0x00044 R 0x4
GNU_EH_FRAME 0x042df0 0x00042df0 0x00042df0 0x00008 0x00008 R 0x4
GNU_STACK 0x000000 0x00000000 0x00000000 0x00000 0x00000 RW 0x10
EXIDX 0x032828 0x00032828 0x00032828 0x00008 0x00008 R 0x4
GNU_RELRO 0x1ffdf0 0x003ffdf0 0x003ffdf0 0x00210 0x00210 RW 0x4
Section to Segment mapping:
Segment Sections...
00
01 .interp
02 .interp .note.ABI-tag .note.gnu.build-id .dynsym .dynstr
.gnu.hash .gnu.version .gnu.version_r .rel.dyn .rel.plt .init .plt .text
.fini .ARM.extab .ARM.exidx .rodata .eh_frame .eh_frame_hdr
03 .fini_array .init_array .dynamic .got .data .bss
04 .dynamic
05 .note.ABI-tag .note.gnu.build-id
06 .eh_frame_hdr
07
08 .ARM.extab .ARM.exidx
09 .fini_array .init_array .dynamic .got
And the diff between these 2:
1_2.22-r0-old/sysroot-destdir/usr/lib/libhugetlbfs/tests/obj32/linkhuge_rw.readelf
1_2.23-r0-new/sysroot-destdir/usr/lib/libhugetlbfs/tests/obj32/linkhuge_rw.readelf
Elf file type is DYN (Shared object file)
-Entry point 0x201105
+Entry point 0x31cd1
There are 10 program headers, starting at offset 52
Program Headers:
Type Offset VirtAddr PhysAddr FileSiz MemSiz Flg Align
- PHDR 0x000034 0x00200034 0x00200034 0x00140 0x00140 R 0x4
- INTERP 0x000174 0x00200174 0x00200174 0x0001d 0x0001d R 0x1
+ PHDR 0x000000 0x00000000 0x00000000 0x00000 0x00000 R 0
+readelf: Error: the PHDR segment is not covered by a LOAD segment
+ INTERP 0x030d40 0x00030d40 0x00030d40 0x0001d 0x0001d R 0x1
[Requesting program interpreter: /usr/lib/ld-linux-armhf.so.3]
- LOAD 0x000000 0x00200000 0x00200000 0x1222c 0x1222c R E 0x200000
- LOAD 0x1ffdf0 0x005ffdf0 0x005ffdf0 0x102e0 0x202ec RW 0x200000
- DYNAMIC 0x1ffdf8 0x005ffdf8 0x005ffdf8 0x00128 0x00128 RW 0x4
- NOTE 0x000194 0x00200194 0x00200194 0x00044 0x00044 R 0x4
- GNU_EH_FRAME 0x012224 0x00212224 0x00212224 0x00008 0x00008 R 0x4
+ LOAD 0x030d40 0x00030d40 0x00030d40 0x120b8 0x120b8 R E 0x200000
+ LOAD 0x1ffdf0 0x003ffdf0 0x003ffdf0 0x102e0 0x202ec RW 0x200000
+ DYNAMIC 0x1ffdf8 0x003ffdf8 0x003ffdf8 0x00128 0x00128 RW 0x4
+ NOTE 0x030d60 0x00030d60 0x00030d60 0x00044 0x00044 R 0x4
+ GNU_EH_FRAME 0x042df0 0x00042df0 0x00042df0 0x00008 0x00008 R 0x4
GNU_STACK 0x000000 0x00000000 0x00000000 0x00000 0x00000 RW 0x10
- EXIDX 0x001c5c 0x00201c5c 0x00201c5c 0x00008 0x00008 R 0x4
- GNU_RELRO 0x1ffdf0 0x005ffdf0 0x005ffdf0 0x00210 0x00210 RW 0x4
+ EXIDX 0x032828 0x00032828 0x00032828 0x00008 0x00008 R 0x4
+ GNU_RELRO 0x1ffdf0 0x003ffdf0 0x003ffdf0 0x00210 0x00210 RW 0x4
Section to Segment mapping:
Revert fixes this build issue, but I still don't see why it fails this way.
Upstream-Status: Pending
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
# Please enter the commit message for your changes. Lines starting
# with '#' will be ignored, and an empty message aborts the commit.
#
# Date: Wed Sep 16 13:43:09 2020 +0200
#
# On branch jansa/master
# Changes to be committed:
# new file: meta-oe/recipes-benchmark/libhugetlbfs/files/0001-Revert-ld.hugetlbfs-fix-Ttext-segment-argument-on-AA.patch
# modified: meta-oe/recipes-benchmark/libhugetlbfs/libhugetlbfs_git.bb
#
# Untracked files:
# counts.txt
# diff
# log.svn
# log.svn2
# wip/
#
---
ld.hugetlbfs | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/ld.hugetlbfs b/ld.hugetlbfs
index 5e4e497..6ee8238 100755
--- a/ld.hugetlbfs
+++ b/ld.hugetlbfs
@@ -130,9 +130,7 @@ if [ "$HTLB_ALIGN" == "slice" ]; then
# targeting the ARM platform one needs to explicitly set the text segment offset
# otherwise it will be NULL.
case "$EMU" in
- armelf*_linux_eabi|aarch64elf*|aarch64linux*)
- printf -v TEXTADDR "%x" "$SLICE_SIZE"
- HTLBOPTS="$HTLBOPTS -Ttext-segment=$TEXTADDR" ;;
+ armelf*_linux_eabi|aarch64elf*|aarch64linux*) HTLBOPTS="$HTLBOPTS -Ttext-segment=$SLICE_SIZE" ;;
elf_i386) HTLBOPTS="$HTLBOPTS -Ttext-segment=0x08000000" ;;
elf64ppc|elf64lppc)
if [ "$MMU_TYPE" == "Hash" ] ; then
--
2.17.1