* [PATCH] gas: Fix some x86_64 testcases for Solaris not using R_X86_64_PLT32 [PR25732]
@ 2020-04-01 10:46 Rainer Orth
2020-04-01 10:58 ` H.J. Lu
0 siblings, 1 reply; 5+ messages in thread
From: Rainer Orth @ 2020-04-01 10:46 UTC (permalink / raw)
To: binutils
[-- Attachment #1: Type: text/plain, Size: 1639 bytes --]
As reported in PR gas/25732, some testcases currently FAIL on
Solaris/x86:
FAIL: x86-64 branch 2
FAIL: x86-64 branch 3
FAIL: x86-64 MPX branch
FAIL: x86-64 branch with BND prefix
FAIL: x86-64 jump
Since https://sourceware.org/ml/binutils/2019-03/msg00163.html, gas
doesn't emit R_X86_64_PLT32 as branch marker on Solaris. Since the
testsuite lacks a way to preprocess dump files, adjusted copies of the
affected dumps are now used on Solaris. Unfortunately, those dumps
weren't adapted when the original testcases were changed or other
testcases started to differ between non-Solaris and Solaris targets.
The following patch fixes that, re-syncing the affected dump files or
creating new Solaris-specific ones.
Tested on i386-pc-solaris2.11, x86_64-pc-solaris2.11,
x86_64-pc-linux-gnu, and i686-pc-linux-gnu.
Ok for master?
Rainer
--
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University
2020-03-31 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
PR gas/25732
* testsuite/gas/i386/solaris/x86-64-branch-2.d: New file.
* testsuite/gas/i386/solaris/x86-64-branch-3.d: New file.
* testsuite/gas/i386/solaris/x86-64-jump.d: Incorporate changes to
testsuite/gas/i386/x86-64-jump.d.
* gas/testsuite/gas/i386/solaris/x86-64-mpx-branch-1.d:
Incorporate changes to
gas/testsuite/gas/i386/x86-64-mpx-branch-1.d.
* testsuite/gas/i386/solaris/x86-64-mpx-branch-2.d : Incorporate
changes to testsuite/gas/i386/x86-64-mpx-branch-2.d.
* testsuite/gas/i386/x86-64-branch-2.d: Skip on *-*-solaris*.
* testsuite/gas/i386/x86-64-branch-3.d: Likewise.
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: gas-solx64-testsuite-pr25732.patch --]
[-- Type: text/x-patch, Size: 8833 bytes --]
# HG changeset patch
# Parent 6fd0216a33b99f01c183bfc56b992fe0316681e4
gas: Fix some x86_64 testcases for Solaris not using R_X86_64_PLT32 (PR gas/25732)
diff --git a/gas/testsuite/gas/i386/solaris/x86-64-branch-2.d b/gas/testsuite/gas/i386/solaris/x86-64-branch-2.d
new file mode 100644
--- /dev/null
+++ b/gas/testsuite/gas/i386/solaris/x86-64-branch-2.d
@@ -0,0 +1,20 @@
+#as: -J
+#objdump: -dwr
+#source: ../x86-64-branch-2.s
+#name: x86-64 branch 2
+
+.*: +file format .*
+
+Disassembly of section .text:
+
+0+ <bar-0xb>:
+[ ]*[a-f0-9]+: 66 e9 00 00 jmpw 4 <bar-0x7> 2: R_X86_64_PC16 foo-0x2
+[ ]*[a-f0-9]+: 66 48 e9 00 00 00 00 data16 jmpq b <bar> 7: R_X86_64_PC32 foo-0x4
+
+0+b <bar>:
+[ ]*[a-f0-9]+: 89 c3 mov %eax,%ebx
+[ ]*[a-f0-9]+: 66 e8 00 00 callw 11 <bar\+0x6> f: R_X86_64_PC16 foo-0x2
+[ ]*[a-f0-9]+: 66 48 e8 00 00 00 00 data16 callq 18 <bar\+0xd> 14: R_X86_64_PC32 foo-0x4
+[ ]*[a-f0-9]+: 66 c3 retw *
+[ ]*[a-f0-9]+: 66 c2 08 00 retw \$0x8
+#pass
diff --git a/gas/testsuite/gas/i386/solaris/x86-64-branch-3.d b/gas/testsuite/gas/i386/solaris/x86-64-branch-3.d
new file mode 100644
--- /dev/null
+++ b/gas/testsuite/gas/i386/solaris/x86-64-branch-3.d
@@ -0,0 +1,22 @@
+#as: -J -mintel64
+#objdump: -dwr -Mintel64
+#source: ../x86-64-branch-3.s
+#name: x86-64 branch 3
+
+.*: +file format .*
+
+Disassembly of section .text:
+
+0+ <bar-0xd>:
+[ ]*[a-f0-9]+: 66 e9 00 00 00 00 data16 jmpq 6 <bar-0x7> 2: R_X86_64_PC32 foo-0x4
+[ ]*[a-f0-9]+: 66 48 e9 00 00 00 00 data16 rex\.W jmpq d <bar> 9: R_X86_64_PC32 foo-0x4
+
+0+d <bar>:
+[ ]*[a-f0-9]+: 89 c3 mov %eax,%ebx
+[ ]*[a-f0-9]+: 66 e8 00 00 00 00 data16 callq 15 <bar\+0x8> 11: R_X86_64_PC32 foo-0x4
+[ ]*[a-f0-9]+: 66 48 e8 00 00 00 00 data16 rex\.W callq 1c <bar\+0xf> 18: R_X86_64_PC32 foo-0x4
+[ ]*[a-f0-9]+: 66 c7 f8 00 00 xbeginw 21 <bar\+0x14> 1f: R_X86_64_PC16 foo-0x2
+[ ]*[a-f0-9]+: 66 48 c7 f8 00 00 00 00 data16 xbeginq 29 <bar\+0x1c> 25: R_X86_64_PC32 foo-0x4
+[ ]*[a-f0-9]+: 48 ff 18 lcallq \*\(%rax\)
+[ ]*[a-f0-9]+: 48 ff 29 ljmpq \*\(%rcx\)
+#pass
diff --git a/gas/testsuite/gas/i386/solaris/x86-64-jump.d b/gas/testsuite/gas/i386/solaris/x86-64-jump.d
--- a/gas/testsuite/gas/i386/solaris/x86-64-jump.d
+++ b/gas/testsuite/gas/i386/solaris/x86-64-jump.d
@@ -31,14 +31,22 @@ 0+ <.text>:
[ ]*[a-f0-9]+: e3 00 jrcxz 0x69 68: R_X86_64_PC8 \$\+0x1
[ ]*[a-f0-9]+: 90 nop
[ ]*[a-f0-9]+: 66 ff 13 callw \*\(%rbx\)
+[ ]*[a-f0-9]+: 66 ff 1b lcallw \*\(%rbx\)
+[ ]*[a-f0-9]+: ff 1b lcall \*\(%rbx\)
+[ ]*[a-f0-9]+: ff 13 callq \*\(%rbx\)
+[ ]*[a-f0-9]+: ff 13 callq \*\(%rbx\)
[ ]*[a-f0-9]+: ff 1b lcall \*\(%rbx\)
[ ]*[a-f0-9]+: 66 ff 23 jmpw \*\(%rbx\)
+[ ]*[a-f0-9]+: 66 ff 2b ljmpw \*\(%rbx\)
[ ]*[a-f0-9]+: ff 2b ljmp \*\(%rbx\)
-[ ]*[a-f0-9]+: eb 00 jmp 0x76
-[ ]*[a-f0-9]+: 90 nop
-[ ]*[a-f0-9]+: 67 e3 00 jecxz 0x7a
+[ ]*[a-f0-9]+: ff 23 jmpq \*\(%rbx\)
+[ ]*[a-f0-9]+: ff 23 jmpq \*\(%rbx\)
+[ ]*[a-f0-9]+: ff 2b ljmp \*\(%rbx\)
+[ ]*[a-f0-9]+: eb 00 jmp 0x[0-9a-f]*
[ ]*[a-f0-9]+: 90 nop
-[ ]*[a-f0-9]+: e3 00 jrcxz 0x7d
+[ ]*[a-f0-9]+: 67 e3 00 jecxz 0x[0-9a-f]*
[ ]*[a-f0-9]+: 90 nop
-[ ]*[a-f0-9]+: eb 00 jmp 0x80
+[ ]*[a-f0-9]+: e3 00 jrcxz 0x[0-9a-f]*
+[ ]*[a-f0-9]+: 90 nop
+[ ]*[a-f0-9]+: eb 00 jmp 0x[0-9a-f]*
#pass
diff --git a/gas/testsuite/gas/i386/solaris/x86-64-mpx-branch-1.d b/gas/testsuite/gas/i386/solaris/x86-64-mpx-branch-1.d
--- a/gas/testsuite/gas/i386/solaris/x86-64-mpx-branch-1.d
+++ b/gas/testsuite/gas/i386/solaris/x86-64-mpx-branch-1.d
@@ -8,22 +8,24 @@
Disassembly of section .text:
-0+ <foo1-0xc>:
-[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 6 <foo1-0x6> 2: R_X86_64_PC32 \*ABS\*\+0x10003c
-[ ]*[a-f0-9]+: f2 e9 00 00 00 00 bnd jmpq c <foo1> 8: R_X86_64_PC32 \*ABS\*\+0x10003c
+0+ <foo1-0x1c>:
+[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 6 <foo1-0x16> 2: R_X86_64_PC32 \*ABS\*\+0x10003c
+[ ]*[a-f0-9]+: f2 e9 00 00 00 00 bnd jmpq c <foo1-0x10> 8: R_X86_64_PC32 \*ABS\*\+0x10003c
+[ ]*[a-f0-9]+: 66 f2 48 e8 00 00 00 00 data16 bnd callq 14 <foo1-0x8> 10: R_X86_64_PC32 \*ABS\*\+0x10003c
+[ ]*[a-f0-9]+: 66 f2 48 e9 00 00 00 00 data16 bnd jmpq 1c <foo1> 18: R_X86_64_PC32 \*ABS\*\+0x10003c
-0+c <foo1>:
-[ ]*[a-f0-9]+: f2 eb fd bnd jmp c <foo1>
-[ ]*[a-f0-9]+: f2 72 fa bnd jb c <foo1>
-[ ]*[a-f0-9]+: f2 e8 f4 ff ff ff bnd callq c <foo1>
-[ ]*[a-f0-9]+: f2 eb 09 bnd jmp 24 <foo2>
-[ ]*[a-f0-9]+: f2 72 06 bnd jb 24 <foo2>
-[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 24 <foo2>
+0+1c <foo1>:
+[ ]*[a-f0-9]+: f2 eb fd bnd jmp 1c <foo1>
+[ ]*[a-f0-9]+: f2 72 fa bnd jb 1c <foo1>
+[ ]*[a-f0-9]+: f2 e8 f4 ff ff ff bnd callq 1c <foo1>
+[ ]*[a-f0-9]+: f2 eb 09 bnd jmp 34 <foo2>
+[ ]*[a-f0-9]+: f2 72 06 bnd jb 34 <foo2>
+[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 34 <foo2>
-0+24 <foo2>:
-[ ]*[a-f0-9]+: f2 e9 00 00 00 00 bnd jmpq 2a <foo2\+0x6> 26: R_X86_64_PC32 foo-0x4
-[ ]*[a-f0-9]+: f2 0f 82 00 00 00 00 bnd jb 31 <foo2\+0xd> 2d: R_X86_64_PC32 foo-0x4
-[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 37 <foo2\+0x13> 33: R_X86_64_PC32 foo-0x4
-[ ]*[a-f0-9]+: f2 e9 00 00 00 00 bnd jmpq 3d <foo2\+0x19> 39: R_X86_64_PLT32 foo-0x4
-[ ]*[a-f0-9]+: f2 0f 82 00 00 00 00 bnd jb 44 <foo2\+0x20> 40: R_X86_64_PLT32 foo-0x4
-[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 4a <foo2\+0x26> 46: R_X86_64_PLT32 foo-0x4
+0+34 <foo2>:
+[ ]*[a-f0-9]+: f2 e9 00 00 00 00 bnd jmpq 3a <foo2\+0x6> 36: R_X86_64_PC32 foo-0x4
+[ ]*[a-f0-9]+: f2 0f 82 00 00 00 00 bnd jb 41 <foo2\+0xd> 3d: R_X86_64_PC32 foo-0x4
+[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 47 <foo2\+0x13> 43: R_X86_64_PC32 foo-0x4
+[ ]*[a-f0-9]+: f2 e9 00 00 00 00 bnd jmpq 4d <foo2\+0x19> 49: R_X86_64_PLT32 foo-0x4
+[ ]*[a-f0-9]+: f2 0f 82 00 00 00 00 bnd jb 54 <foo2\+0x20> 50: R_X86_64_PLT32 foo-0x4
+[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 5a <foo2\+0x26> 56: R_X86_64_PLT32 foo-0x4
diff --git a/gas/testsuite/gas/i386/solaris/x86-64-mpx-branch-2.d b/gas/testsuite/gas/i386/solaris/x86-64-mpx-branch-2.d
--- a/gas/testsuite/gas/i386/solaris/x86-64-mpx-branch-2.d
+++ b/gas/testsuite/gas/i386/solaris/x86-64-mpx-branch-2.d
@@ -1,29 +1,5 @@
#as: -J -madd-bnd-prefix
-#source: ../x86-64-mpx-branch-2.s
+#source: ../x86-64-mpx-branch-1.s
#objdump: -dwr
#name: x86-64 branch with BND prefix
-
-.*: +file format .*
-
-
-Disassembly of section .text:
-
-0+ <foo1-0xc>:
-[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 6 <foo1-0x6> 2: R_X86_64_PC32 \*ABS\*\+0x10003c
-[ ]*[a-f0-9]+: f2 e9 00 00 00 00 bnd jmpq c <foo1> 8: R_X86_64_PC32 \*ABS\*\+0x10003c
-
-0+c <foo1>:
-[ ]*[a-f0-9]+: f2 eb fd bnd jmp c <foo1>
-[ ]*[a-f0-9]+: f2 72 fa bnd jb c <foo1>
-[ ]*[a-f0-9]+: f2 e8 f4 ff ff ff bnd callq c <foo1>
-[ ]*[a-f0-9]+: f2 eb 09 bnd jmp 24 <foo2>
-[ ]*[a-f0-9]+: f2 72 06 bnd jb 24 <foo2>
-[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 24 <foo2>
-
-0+24 <foo2>:
-[ ]*[a-f0-9]+: f2 e9 00 00 00 00 bnd jmpq 2a <foo2\+0x6> 26: R_X86_64_PC32 foo-0x4
-[ ]*[a-f0-9]+: f2 0f 82 00 00 00 00 bnd jb 31 <foo2\+0xd> 2d: R_X86_64_PC32 foo-0x4
-[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 37 <foo2\+0x13> 33: R_X86_64_PC32 foo-0x4
-[ ]*[a-f0-9]+: f2 e9 00 00 00 00 bnd jmpq 3d <foo2\+0x19> 39: R_X86_64_PLT32 foo-0x4
-[ ]*[a-f0-9]+: f2 0f 82 00 00 00 00 bnd jb 44 <foo2\+0x20> 40: R_X86_64_PLT32 foo-0x4
-[ ]*[a-f0-9]+: f2 e8 00 00 00 00 bnd callq 4a <foo2\+0x26> 46: R_X86_64_PLT32 foo-0x4
+#dump: x86-64-mpx-branch-1.d
diff --git a/gas/testsuite/gas/i386/x86-64-branch-2.d b/gas/testsuite/gas/i386/x86-64-branch-2.d
--- a/gas/testsuite/gas/i386/x86-64-branch-2.d
+++ b/gas/testsuite/gas/i386/x86-64-branch-2.d
@@ -1,6 +1,7 @@
#as: -J
#objdump: -dwr
#name: x86-64 branch 2
+#notarget: *-*-solaris*
.*: +file format .*
diff --git a/gas/testsuite/gas/i386/x86-64-branch-3.d b/gas/testsuite/gas/i386/x86-64-branch-3.d
--- a/gas/testsuite/gas/i386/x86-64-branch-3.d
+++ b/gas/testsuite/gas/i386/x86-64-branch-3.d
@@ -1,6 +1,7 @@
#as: -J -mintel64
#objdump: -dwr -Mintel64
#name: x86-64 branch 3
+#notarget: *-*-solaris*
.*: +file format .*
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] gas: Fix some x86_64 testcases for Solaris not using R_X86_64_PLT32 [PR25732]
2020-04-01 10:46 [PATCH] gas: Fix some x86_64 testcases for Solaris not using R_X86_64_PLT32 [PR25732] Rainer Orth
@ 2020-04-01 10:58 ` H.J. Lu
2020-04-02 6:19 ` Fangrui Song
0 siblings, 1 reply; 5+ messages in thread
From: H.J. Lu @ 2020-04-01 10:58 UTC (permalink / raw)
To: Rainer Orth; +Cc: Binutils
On Wed, Apr 1, 2020 at 3:47 AM Rainer Orth <ro@cebitec.uni-bielefeld.de> wrote:
>
> As reported in PR gas/25732, some testcases currently FAIL on
> Solaris/x86:
>
> FAIL: x86-64 branch 2
> FAIL: x86-64 branch 3
> FAIL: x86-64 MPX branch
> FAIL: x86-64 branch with BND prefix
> FAIL: x86-64 jump
>
> Since https://sourceware.org/ml/binutils/2019-03/msg00163.html, gas
> doesn't emit R_X86_64_PLT32 as branch marker on Solaris. Since the
> testsuite lacks a way to preprocess dump files, adjusted copies of the
> affected dumps are now used on Solaris. Unfortunately, those dumps
> weren't adapted when the original testcases were changed or other
> testcases started to differ between non-Solaris and Solaris targets.
>
> The following patch fixes that, re-syncing the affected dump files or
> creating new Solaris-specific ones.
>
> Tested on i386-pc-solaris2.11, x86_64-pc-solaris2.11,
> x86_64-pc-linux-gnu, and i686-pc-linux-gnu.
>
> Ok for master?
>
OK.
BTW, please watch out for future adjustments similar to this.
Thanks.
--
H.J.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] gas: Fix some x86_64 testcases for Solaris not using R_X86_64_PLT32 [PR25732]
2020-04-01 10:58 ` H.J. Lu
@ 2020-04-02 6:19 ` Fangrui Song
2020-04-02 13:07 ` Rainer Orth
0 siblings, 1 reply; 5+ messages in thread
From: Fangrui Song @ 2020-04-02 6:19 UTC (permalink / raw)
To: Rainer Orth; +Cc: Binutils
On 2020-04-01, H.J. Lu via Binutils wrote:
>On Wed, Apr 1, 2020 at 3:47 AM Rainer Orth <ro@cebitec.uni-bielefeld.de> wrote:
>>
>> As reported in PR gas/25732, some testcases currently FAIL on
>> Solaris/x86:
>>
>> FAIL: x86-64 branch 2
>> FAIL: x86-64 branch 3
>> FAIL: x86-64 MPX branch
>> FAIL: x86-64 branch with BND prefix
>> FAIL: x86-64 jump
>>
>> Since https://sourceware.org/ml/binutils/2019-03/msg00163.html, gas
>> doesn't emit R_X86_64_PLT32 as branch marker on Solaris. Since the
>> testsuite lacks a way to preprocess dump files, adjusted copies of the
>> affected dumps are now used on Solaris. Unfortunately, those dumps
>> weren't adapted when the original testcases were changed or other
>> testcases started to differ between non-Solaris and Solaris targets.
>>
>> The following patch fixes that, re-syncing the affected dump files or
>> creating new Solaris-specific ones.
>>
>> Tested on i386-pc-solaris2.11, x86_64-pc-solaris2.11,
>> x86_64-pc-linux-gnu, and i686-pc-linux-gnu.
>>
>> Ok for master?
>>
>
>OK.
>
>BTW, please watch out for future adjustments similar to this.
>
>Thanks.
Why doesn't Solaris x86 use R_X86_64_PLT32?
Disclosure: I know close to zero about Solaris.
If a PC32 references a preemptible STT_FUNC, a canonical PLT entry (st_value>0, st_shndx=0) will be created.
This is a hack in ld.so and can slow down symbol resolution a bit.
My interpretation of PLT32 is that the address is not significant.
PC32 means the address is taken. Generally it is good for emitted code
sequences to have lower expectation.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] gas: Fix some x86_64 testcases for Solaris not using R_X86_64_PLT32 [PR25732]
2020-04-02 6:19 ` Fangrui Song
@ 2020-04-02 13:07 ` Rainer Orth
2020-04-03 0:59 ` Fangrui Song
0 siblings, 1 reply; 5+ messages in thread
From: Rainer Orth @ 2020-04-02 13:07 UTC (permalink / raw)
To: Fangrui Song; +Cc: Binutils
Hi Fangrui,
> On 2020-04-01, H.J. Lu via Binutils wrote:
>>On Wed, Apr 1, 2020 at 3:47 AM Rainer Orth <ro@cebitec.uni-bielefeld.de> wrote:
>>>
>>> As reported in PR gas/25732, some testcases currently FAIL on
>>> Solaris/x86:
>>>
>>> FAIL: x86-64 branch 2
>>> FAIL: x86-64 branch 3
>>> FAIL: x86-64 MPX branch
>>> FAIL: x86-64 branch with BND prefix
>>> FAIL: x86-64 jump
>>>
>>> Since https://sourceware.org/ml/binutils/2019-03/msg00163.html, gas
>>> doesn't emit R_X86_64_PLT32 as branch marker on Solaris. Since the
>>> testsuite lacks a way to preprocess dump files, adjusted copies of the
>>> affected dumps are now used on Solaris. Unfortunately, those dumps
>>> weren't adapted when the original testcases were changed or other
>>> testcases started to differ between non-Solaris and Solaris targets.
>>>
>>> The following patch fixes that, re-syncing the affected dump files or
>>> creating new Solaris-specific ones.
>>>
>>> Tested on i386-pc-solaris2.11, x86_64-pc-solaris2.11,
>>> x86_64-pc-linux-gnu, and i686-pc-linux-gnu.
>>>
>>> Ok for master?
>>>
>>
>>OK.
>>
>>BTW, please watch out for future adjustments similar to this.
>>
>>Thanks.
>
> Why doesn't Solaris x86 use R_X86_64_PLT32?
it does, it just doesn't want to implement support for HJ's branch
marker optimization referred to in the patch linked above. Some of that
is explained in my patch submission, buf if you want to know all the
gory details, ask Ali ;-)
> Disclosure: I know close to zero about Solaris.
That can be changed :-) For one, there's the Linkers and Libraries Guide
https://docs.oracle.com/cd/E37838_01/html/E36783/index.html
written by the linker engineers themselves. And for even more
background, you can check out their blog
http://www.linker-aliens.org/
Both make for quite interesting reading for anyone interested in ELF in
general and Solaris innovations in particular.
Rainer
--
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] gas: Fix some x86_64 testcases for Solaris not using R_X86_64_PLT32 [PR25732]
2020-04-02 13:07 ` Rainer Orth
@ 2020-04-03 0:59 ` Fangrui Song
0 siblings, 0 replies; 5+ messages in thread
From: Fangrui Song @ 2020-04-03 0:59 UTC (permalink / raw)
To: Rainer Orth; +Cc: Binutils
On 2020-04-02, Rainer Orth wrote:
>Hi Fangrui,
>
>> On 2020-04-01, H.J. Lu via Binutils wrote:
>>>On Wed, Apr 1, 2020 at 3:47 AM Rainer Orth <ro@cebitec.uni-bielefeld.de> wrote:
>>>>
>>>> As reported in PR gas/25732, some testcases currently FAIL on
>>>> Solaris/x86:
>>>>
>>>> FAIL: x86-64 branch 2
>>>> FAIL: x86-64 branch 3
>>>> FAIL: x86-64 MPX branch
>>>> FAIL: x86-64 branch with BND prefix
>>>> FAIL: x86-64 jump
>>>>
>>>> Since https://sourceware.org/ml/binutils/2019-03/msg00163.html, gas
>>>> doesn't emit R_X86_64_PLT32 as branch marker on Solaris. Since the
>>>> testsuite lacks a way to preprocess dump files, adjusted copies of the
>>>> affected dumps are now used on Solaris. Unfortunately, those dumps
>>>> weren't adapted when the original testcases were changed or other
>>>> testcases started to differ between non-Solaris and Solaris targets.
>>>>
>>>> The following patch fixes that, re-syncing the affected dump files or
>>>> creating new Solaris-specific ones.
>>>>
>>>> Tested on i386-pc-solaris2.11, x86_64-pc-solaris2.11,
>>>> x86_64-pc-linux-gnu, and i686-pc-linux-gnu.
>>>>
>>>> Ok for master?
>>>>
>>>
>>>OK.
>>>
>>>BTW, please watch out for future adjustments similar to this.
>>>
>>>Thanks.
>>
>> Why doesn't Solaris x86 use R_X86_64_PLT32?
>
>it does, it just doesn't want to implement support for HJ's branch
>marker optimization referred to in the patch linked above. Some of that
>is explained in my patch submission, buf if you want to know all the
>gory details, ask Ali ;-)
Looking at the tests, if I guess correctly, a @plt modifier is required
to emit R_X86_64_PLT32. I have put many thoughts in this area. HJ's
branch marker optimization is a great thing to me.
R_X86_64_PLT32 is categorized as R_PLT_PC (a PC-relative relocation
which may create a PLT entry). It can be relaxed to R_PC (a PC-relative
relocation) if the symbol is non-preemptible (logical AND (non-ifunc or
FreeBSD extension -z ifunc-noplt); I know Solaris will never support ifunc).
A PLT32 has less requirement than a PC32. A PLT32 means the address is
not significant. The linker can use whatever approach appropriate to
arrange for a call to the symbol. It can be a direct jump, a jump
through a PLT entry, or other novel approach if it prefers.
A PC32, on the other hand, is indistinguishable with an address taken
operation. This causes us copy relocations and canonical PLT entries (I
should note that I've seen a few people (at least Rafael Espindola and
Cary Coutant) using the term "canonical PLT entry" but it is not widely
used).
When emitting a jump/call instruction, it is unnecessary for the
compiler to emit extra info: R_X86_64_PLT32 vs R_X86_64_PC32;
R_PPC_LOCAL24PC vs R_PPC_PLTREL24 vs R_PPC_REL24. It is great that PPC64
dropped some relocation types in contrast to PPC32...
The non-preemptible property is a mergeable property of the symbol, not
an unmergeable property of the relocation.
I made a complaint about RISC-V's call relocation types while working on
the RISC-V port for lld
https://github.com/riscv/riscv-elf-psabi-doc/issues/98 R_RISCV_CALL vs R_RISCV_CALL_PLT
From the viewpoint of implementation, with proper abstraction,
R_X86_64_PLT32 -> R_X86_64_PC32 can be implemented with very little
code:
https://github.com/llvm/llvm-project/blob/master/lld/ELF/Relocations.cpp#L1329
plus the definition of fromPlt
if (!sym.isPreemptible && (!sym.isGnuIFunc() || config->zIfuncNoplt)) {
if (expr == R_GOT_PC && !isAbsoluteValue(sym)) {
/// For the curious, this is the recent https://sourceware.org/bugzilla/show_bug.cgi?id=25754
expr = target->adjustRelaxExpr(type, relocatedAddr, expr);
} else {
// The 0x8000 bit of r_addend of R_PPC_PLTREL24 is used to choose call
// stub type. It should be ignored if optimized to R_PC.
if (config->emachine == EM_PPC && expr == R_PPC32_PLTREL)
addend &= ~0x8000;
expr = fromPlt(expr);
}
}
>> Disclosure: I know close to zero about Solaris.
>
>That can be changed :-) For one, there's the Linkers and Libraries Guide
>
> https://docs.oracle.com/cd/E37838_01/html/E36783/index.html
>
>written by the linker engineers themselves. And for even more
>background, you can check out their blog
>
> http://www.linker-aliens.org/
>
>Both make for quite interesting reading for anyone interested in ELF in
>general and Solaris innovations in particular.
>
> Rainer
>--
>-----------------------------------------------------------------------------
>Rainer Orth, Center for Biotechnology, Bielefeld University
I take this as a very friendly invitation to ask me to read:) Thanks.
I have read some when I started to work on lld. They are phrased very
well and have taught me a lot. As a linker and loader lover, I should
finish reading the Linkers and Libraries Guide...
It is refreshing to get some non-GNU perspective.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2020-04-03 0:59 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-04-01 10:46 [PATCH] gas: Fix some x86_64 testcases for Solaris not using R_X86_64_PLT32 [PR25732] Rainer Orth
2020-04-01 10:58 ` H.J. Lu
2020-04-02 6:19 ` Fangrui Song
2020-04-02 13:07 ` Rainer Orth
2020-04-03 0:59 ` Fangrui Song
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 read-only IMAP folder(s) and NNTP newsgroup(s).