* [PATCH] wrong code on m68k with -mlong-jump-table-offsets and -malign-int (PR target/112413)
@ 2023-12-11 12:51 Mikael Pettersson
2023-12-11 15:42 ` Jeff Law
0 siblings, 1 reply; 2+ messages in thread
From: Mikael Pettersson @ 2023-12-11 12:51 UTC (permalink / raw)
To: gcc-patches; +Cc: Mikael Pettersson
On m68k the compiler assumes that the PC-relative jump-via-jump-table
instruction and the jump table are adjacent with no padding in between.
When -mlong-jump-table-offsets is combined with -malign-int, a 2-byte
nop may be inserted before the jump table, causing the jump to add the
fetched offset to the wrong PC base and thus jump to the wrong address.
Fixed by referencing the jump table via its label. On the test case
in the PR the object code change is (the moveal at 16 is the nop):
a: 6536 bcss 42 <f+0x42>
c: e588 lsll #2,%d0
e: 203b 0808 movel %pc@(18 <f+0x18>,%d0:l),%d0
- 12: 4efb 0802 jmp %pc@(16 <f+0x16>,%d0:l)
+ 12: 4efb 0804 jmp %pc@(18 <f+0x18>,%d0:l)
16: 284c moveal %a4,%a4
18: 0000 0020 orib #32,%d0
1c: 0000 002c orib #44,%d0
Bootstrapped and tested on m68k-linux-gnu, no regressions.
Note: I don't have commit rights to I would need assistance applying this.
2023-12-11 Mikael Pettersson <mikpelinux@gmail.com>
PR target/112413
* config/m68k/linux.h (ASM_RETURN_CASE_JUMP): For
TARGET_LONG_JUMP_TABLE_OFFSETS, reference the jump table
via its label.
* config/m68k/m68kelf.h (ASM_RETURN_CASE_JUMP: Likewise.
* config/m68k/netbsd-elf.h (ASM_RETURN_CASE_JUMP): Likewise.
---
gcc/config/m68k/linux.h | 4 ++--
gcc/config/m68k/m68kelf.h | 4 ++--
gcc/config/m68k/netbsd-elf.h | 4 ++--
3 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/gcc/config/m68k/linux.h b/gcc/config/m68k/linux.h
index 2e1cb5498b8..37069c4d082 100644
--- a/gcc/config/m68k/linux.h
+++ b/gcc/config/m68k/linux.h
@@ -102,12 +102,12 @@ along with GCC; see the file COPYING3. If not see
if (ADDRESS_REG_P (operands[0])) \
return "jmp %%pc@(2,%0:l)"; \
else if (TARGET_LONG_JUMP_TABLE_OFFSETS) \
- return "jmp %%pc@(2,%0:l)"; \
+ return "jmp %%pc@(%l1,%0:l)"; \
else \
return "ext%.l %0\n\tjmp %%pc@(2,%0:l)"; \
} \
else if (TARGET_LONG_JUMP_TABLE_OFFSETS) \
- return "jmp %%pc@(2,%0:l)"; \
+ return "jmp %%pc@(%l1,%0:l)"; \
else \
return "jmp %%pc@(2,%0:w)"; \
} while (0)
diff --git a/gcc/config/m68k/m68kelf.h b/gcc/config/m68k/m68kelf.h
index 01ee724ef2b..f89c9b70455 100644
--- a/gcc/config/m68k/m68kelf.h
+++ b/gcc/config/m68k/m68kelf.h
@@ -59,12 +59,12 @@ along with GCC; see the file COPYING3. If not see
if (ADDRESS_REG_P (operands[0])) \
return "jmp %%pc@(2,%0:l)"; \
else if (TARGET_LONG_JUMP_TABLE_OFFSETS) \
- return "jmp %%pc@(2,%0:l)"; \
+ return "jmp %%pc@(%l1,%0:l)"; \
else \
return "ext%.l %0\n\tjmp %%pc@(2,%0:l)"; \
} \
else if (TARGET_LONG_JUMP_TABLE_OFFSETS) \
- return "jmp %%pc@(2,%0:l)"; \
+ return "jmp %%pc@(%l1,%0:l)"; \
else \
return "jmp %%pc@(2,%0:w)"; \
} while (0)
diff --git a/gcc/config/m68k/netbsd-elf.h b/gcc/config/m68k/netbsd-elf.h
index 4d4a6d71cc4..6ba581b7b18 100644
--- a/gcc/config/m68k/netbsd-elf.h
+++ b/gcc/config/m68k/netbsd-elf.h
@@ -137,12 +137,12 @@ while (0)
if (ADDRESS_REG_P (operands[0])) \
return "jmp %%pc@(2,%0:l)"; \
else if (TARGET_LONG_JUMP_TABLE_OFFSETS) \
- return "jmp %%pc@(2,%0:l)"; \
+ return "jmp %%pc@(%l1,%0:l)"; \
else \
return "ext%.l %0\n\tjmp %%pc@(2,%0:l)"; \
} \
else if (TARGET_LONG_JUMP_TABLE_OFFSETS) \
- return "jmp %%pc@(2,%0:l)"; \
+ return "jmp %%pc@(%l1,%0:l)"; \
else \
return "jmp %%pc@(2,%0:w)"; \
} while (0)
--
2.43.0
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH] wrong code on m68k with -mlong-jump-table-offsets and -malign-int (PR target/112413)
2023-12-11 12:51 [PATCH] wrong code on m68k with -mlong-jump-table-offsets and -malign-int (PR target/112413) Mikael Pettersson
@ 2023-12-11 15:42 ` Jeff Law
0 siblings, 0 replies; 2+ messages in thread
From: Jeff Law @ 2023-12-11 15:42 UTC (permalink / raw)
To: Mikael Pettersson, gcc-patches
On 12/11/23 05:51, Mikael Pettersson wrote:
> On m68k the compiler assumes that the PC-relative jump-via-jump-table
> instruction and the jump table are adjacent with no padding in between.
>
> When -mlong-jump-table-offsets is combined with -malign-int, a 2-byte
> nop may be inserted before the jump table, causing the jump to add the
> fetched offset to the wrong PC base and thus jump to the wrong address.
>
> Fixed by referencing the jump table via its label. On the test case
> in the PR the object code change is (the moveal at 16 is the nop):
>
> a: 6536 bcss 42 <f+0x42>
> c: e588 lsll #2,%d0
> e: 203b 0808 movel %pc@(18 <f+0x18>,%d0:l),%d0
> - 12: 4efb 0802 jmp %pc@(16 <f+0x16>,%d0:l)
> + 12: 4efb 0804 jmp %pc@(18 <f+0x18>,%d0:l)
> 16: 284c moveal %a4,%a4
> 18: 0000 0020 orib #32,%d0
> 1c: 0000 002c orib #44,%d0
>
> Bootstrapped and tested on m68k-linux-gnu, no regressions.
>
> Note: I don't have commit rights to I would need assistance applying this.
>
> 2023-12-11 Mikael Pettersson <mikpelinux@gmail.com>
>
> PR target/112413
> * config/m68k/linux.h (ASM_RETURN_CASE_JUMP): For
> TARGET_LONG_JUMP_TABLE_OFFSETS, reference the jump table
> via its label.
> * config/m68k/m68kelf.h (ASM_RETURN_CASE_JUMP: Likewise.
> * config/m68k/netbsd-elf.h (ASM_RETURN_CASE_JUMP): Likewise.
THanks. Installed.
jeff
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2023-12-11 15:42 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-12-11 12:51 [PATCH] wrong code on m68k with -mlong-jump-table-offsets and -malign-int (PR target/112413) Mikael Pettersson
2023-12-11 15:42 ` Jeff Law
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).