public inbox for binutils@sourceware.org
 help / color / mirror / Atom feed
* Fw: Re: [PATCH] [RISC-V] Optimize GP-relative addressing for linker.
@ 2023-06-28  2:58 Die Li
  0 siblings, 0 replies; only message in thread
From: Die Li @ 2023-06-28  2:58 UTC (permalink / raw)
  To: palmer, binutils, nelson

[-- Attachment #1: Type: text/plain, Size: 10333 bytes --]

Hi,
On 2023-06-28 10:42,  Die Li wrote:
>
Hi,
On 2023-06-27 23:43,  Kito Cheng wrote:
>
>Testcase in your git commit has a base offset, but the testcase in the
>code only contains a base with no offset?

The "testcase in the code" is derived from test.c file in the commit message.
Due to the requirement of an `_start` symbol for `ld`, the `main` symbol
was changed to `_start` for input, while everything else remained unchanged.

with the gp-relax.s in the code:
Before linking, the information in the symbol table is as follows:

Symbol table '.symtab' contains 13 entries:
    Num:    Value                 Size  Type            Bind         Vis      Ndx Name
...
     9: 0000000000000000     4 OBJECT  GLOBAL DEFAULT    5 global_var
    10: 0000000000000000  3680 OBJECT  GLOBAL DEFAULT    4 global_array
    11: 0000000000000e60   428 OBJECT  GLOBAL DEFAULT    4 person
...
The symbols global_array and person occupy a significant amount of space,
with sizes of 3680 and 428, respectively.

before this patch:
after linking, we have:
00000000000100e8 <_start>:
   100e8:	000117b7          	lui	a5,0x11
   100ec:	10878793          	addi	a5,a5,264 # 11108 <global_array>
   100f0:	00012737          	lui	a4,0x12
   100f4:	f6870713          	addi	a4,a4,-152 # 11f68 <person>

Check the symbol table after linking:
   Num:    Value          Size Type    Bind   Vis      Ndx Name
     7: 00000000000118f4     0 NOTYPE  GLOBAL DEFAULT  ABS __global_pointer$
     8: 00000000000110f4     4 OBJECT  GLOBAL DEFAULT    2 global_var
    12: 0000000000011f58   428 OBJECT  GLOBAL DEFAULT    3 person
    17: 00000000000110f8  3680 OBJECT  GLOBAL DEFAULT    3 global_array

The symbol can be relaxed by "__global_pointer$" within the address range of
[0x110f4, 0x120f3], which is calculated as [0x118f4 - 2^11, 0x118f4 + 2^11 - 1].
Since the symbols global_array, global_var, and person all fall within this range,
they should all be subject to relaxation during linking. However, the current linker
takes into account the size of symbols, resulting in a failure to relax the addresses
during linking.


after this patch:
after linking, we have:
00000000000100e8 <_start>:
   100e8:	80418793          	addi	a5,gp,-2044 # 110f8 <global_array>
   100ec:	66418713          	addi	a4,gp,1636 # 11f58 <person>


with the test.c in the commit message:
Please check the image with the name of "difference with and
without patch".
When accessing global_array[365] in int global_array[920], the linker first obtains
the base address of global_array and then offsets it by 365 * 4 bytes to obtain the
address of the element global_array[365]. The same principle applies to accessing
structure members.

>
>I am a little concerned about the case with an offset like "lui
>a5,%hi(global_array+256) addi    a5,a5,%lo(global_array+256)" is still
>right for this optimization?

As long as the address of global_array + 256 falls within the range of
[0x118f4 - 2^11, 0x118f4 + 2^11 - 1] mentioned earlier, relaxation during linking
is possible. I conducted a local experiment, and the results are as follows:

before the patch:
00000000000100e8 <_start>:
   100e8:	000117b7          	lui	a5,0x11
   100ec:	10878793          	addi	a5,a5,264 # 11108 <global_array>
   100f0:	00012737          	lui	a4,0x12
   100f4:	f6870713          	addi	a4,a4,-152 # 11f68 <person>
   100f8:	000117b7          	lui	a5,0x11
   100fc:	20878793          	addi	a5,a5,520 # 11208 <global_array+0x100>

after the patch:
00000000000100e8 <_start>:
   100e8:	80418793          	addi	a5,gp,-2044 # 110f8 <global_array>
   100ec:	66418713          	addi	a4,gp,1636 # 11f58 <person>
   100f0:	90418793          	addi	a5,gp,-1788 # 111f8 <global_array+0x100>


>
>On Tue, Jun 27, 2023 at 8:47 PM Die Li <lidie@eswincomputing.com> wrote:
>>
>> Consider the following test:
>>
>> //test.c file
>>
>> struct Person {
>>     char name[20];
>>     int age;
>>     int nation;
>>     int hobby[100];
>> };
>>
>> int global_var;
>> int global_array[920];
>> struct Person person;
>>
>> int main() {
>>     global_var = 3;
>>     global_array[365] = 16;
>>     sprintf(person.name, "Lee");
>>     person.age = 27;
>>     person.nation = 77;
>>     return 0;
>> }
>>
>> Cflags:
>> -Xlinker --relax
>>
>> Link relaxation can be turned on by the above option, and in fact
>> is turned on by default. After compiling, linking, and disassembling
>> the test files, there are the following results:
>>
>> Before this patch:
>> Disassembly of section \.text:
>> 0+[0-9a-f]+ <main>:
>> .*:[    ]+[0-9a-f]+[    ]+sw[   ]+[0-9a-f]+,gp,\-[0-9]+ # [0-9a-f]+ <global_var>
>> .*:[    ]+[0-9a-f]+[    ]+c\.lui[       ]+.*
>> .*:[    ]+[0-9a-f]+[    ]+addi[         ]+[0-9a-f]+,[0-9a-f]+,[0-9]+ # [0-9a-f]+ <global_array>
>> .*:[    ]+[0-9a-f]+[    ]+c\.lui[       ]+.*
>> .*:[    ]+[0-9a-f]+[    ]+addi[         ]+[0-9a-f]+,[0-9a-f]+,[0-9]+ # [0-9a-f]+ <person>
>>
>> After this patch:
>> Disassembly of section \.text:
>> 0+[0-9a-f]+ <main>:
>> .*:[    ]+[0-9a-f]+[    ]+sw[   ]+[0-9a-f]+,gp,\-[0-9]+ # [0-9a-f]+ <global_var>
>> .*:[    ]+[0-9a-f]+[    ]+addi[         ]+[0-9a-f]+,gp,\-[0-9]+ # [0-9a-f]+ <global_array>
>> .*:[    ]+[0-9a-f]+[    ]+addi[         ]+[0-9a-f]+,gp,\-[0-9]+ # [0-9a-f]+ <person>
>>
>> After applying this patch, both array element and structure member
>> accesses have been optimized. In fact, for both array elements and
>> structure members, access is based on the data and the base address
>> of the structure, which is the address of the first member of the
>> array or structure. However, current linker takes into account the
>> size of arrays and structures when performing GP-relative addressing.
>> As a result, some array and structure bases within the GP-relative
>> addressing range cannot benefit from relaxation optimization. This
>> patch resolves this issue.
>>
>> Signed-off-by: Die Li <lidie@eswincomputing.com>
>>
>> ChangeLog:
>>
>>         * bfd/elfnn-riscv.c (_bfd_riscv_relax_section): Update reserve_size.
>>         * ld/testsuite/ld-riscv-elf/ld-riscv-elf.exp: New test entry.
>>         * ld/testsuite/ld-riscv-elf/gp-relax.d: New test.
>>         * ld/testsuite/ld-riscv-elf/gp-relax.s: New test.
>> ---
>>  bfd/elfnn-riscv.c                          |  7 +++++
>>  ld/testsuite/ld-riscv-elf/gp-relax.d       | 12 +++++++++
>>  ld/testsuite/ld-riscv-elf/gp-relax.s       | 31 ++++++++++++++++++++++
>>  ld/testsuite/ld-riscv-elf/ld-riscv-elf.exp |  1 +
>>  4 files changed, 51 insertions(+)
>>  create mode 100644 ld/testsuite/ld-riscv-elf/gp-relax.d
>>  create mode 100644 ld/testsuite/ld-riscv-elf/gp-relax.s
>>
>> diff --git a/bfd/elfnn-riscv.c b/bfd/elfnn-riscv.c
>> index 09aa7be225e..29fc5484e0f 100644
>> --- a/bfd/elfnn-riscv.c
>> +++ b/bfd/elfnn-riscv.c
>> @@ -5169,6 +5169,13 @@ _bfd_riscv_relax_section (bfd *abfd, asection *sec,
>>           if (h->type != STT_FUNC)
>>             reserve_size =
>>               (h->size - rel->r_addend) > h->size ? 0 : h->size - rel->r_addend;
>> +
>> +          /* For global pointer relative addressing, it is sufficient to ensure
>> +             that the symbol's base address falls within the range of global
>> +             pointer relative addressing.  */
>> +          if (h->type == STT_OBJECT)
>> +            reserve_size = 0;
>> +
>>           symtype = h->type;
>>         }
>>
>> diff --git a/ld/testsuite/ld-riscv-elf/gp-relax.d b/ld/testsuite/ld-riscv-elf/gp-relax.d
>> new file mode 100644
>> index 00000000000..ec2f59b1b19
>> --- /dev/null
>> +++ b/ld/testsuite/ld-riscv-elf/gp-relax.d
>> @@ -0,0 +1,12 @@
>> +#source: gp-relax.s
>> +#ld: --relax
>> +#objdump: -d -Mno-aliases
>> +
>> +.*:[   ]+file format .*
>> +
>> +
>> +Disassembly of section \.text:
>> +
>> +0+[0-9a-f]+ <_start>:
>> +.*:[   ]+[0-9a-f]+[    ]+addi[         ]+a5,gp,\-[0-9]+ # [0-9a-f]+ <global_array>
>> +.*:[   ]+[0-9a-f]+[    ]+addi[         ]+a4,gp,[0-9]+ # [0-9a-f]+ <person>
>> diff --git a/ld/testsuite/ld-riscv-elf/gp-relax.s b/ld/testsuite/ld-riscv-elf/gp-relax.s
>> new file mode 100644
>> index 00000000000..05548888ebf
>> --- /dev/null
>> +++ b/ld/testsuite/ld-riscv-elf/gp-relax.s
>> @@ -0,0 +1,31 @@
>> +       .text
>> +       .globl  global_var
>> +       .section        .sbss,"aw",@nobits
>> +       .align  2
>> +       .type   global_var, @object
>> +       .size   global_var, 4
>> +global_var:
>> +       .zero   4
>> +       .globl  global_array
>> +       .bss
>> +       .align  3
>> +       .type   global_array, @object
>> +       .size   global_array, 3680
>> +global_array:
>> +       .zero   3680
>> +       .globl  person
>> +       .align  3
>> +       .type   person, @object
>> +       .size   person, 428
>> +person:
>> +       .zero   428
>> +       .text
>> +       .align  1
>> +       .globl  _start
>> +       .type   _start, @function
>> +_start:
>> +       lui     a5,%hi(global_array)
>> +       addi    a5,a5,%lo(global_array)
>> +
>> +       lui     a4,%hi(person)
>> +       addi    a4,a4,%lo(person)
>> diff --git a/ld/testsuite/ld-riscv-elf/ld-riscv-elf.exp b/ld/testsuite/ld-riscv-elf/ld-riscv-elf.exp
>> index 947a266ba72..6a04955b23b 100644
>> --- a/ld/testsuite/ld-riscv-elf/ld-riscv-elf.exp
>> +++ b/ld/testsuite/ld-riscv-elf/ld-riscv-elf.exp
>> @@ -124,6 +124,7 @@ if [istarget "riscv*-*-*"] {
>>      run_dump_test "pcgp-relax-01"
>>      run_dump_test "pcgp-relax-01-norelaxgp"
>>      run_dump_test "pcgp-relax-02"
>> +    run_dump_test "gp-relax"
>>      run_dump_test "c-lui"
>>      run_dump_test "c-lui-2"
>>      run_dump_test "disas-jalr"
>> --
>> 2.17.1
>>

[-- Attachment #2: diffrence_with_and_without_the_patch.png --]
[-- Type: application/octet-stream, Size: 145884 bytes --]

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2023-06-28  2:58 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-06-28  2:58 Fw: Re: [PATCH] [RISC-V] Optimize GP-relative addressing for linker Die Li

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).