riscv: Fix breakage caused by linker relaxation
PATCH 1/1 riscv: Fix breakage caused by linker relaxation
Due to the two-instruction sequence needed to access arbitrary memory locations, the RISC-V linker aggressively optimises memory accesses and jumps at link-time. This is called "linker relaxation," and is discussed in this SiFive article
<https://www.sifive.com/blog/all-aboard-part-3-linker-relaxation-in-riscv-toolchain>.
One of the optimizations in place is to assume that the __global_pointer symbol is placed in the gp register. To quote the article:
"...The magic __global_pointer$ symbol is defined to point 0x800 bytes past the start of the .sdata section. The 0x800 magic number allows signed 12-bit offsets from __global_pointer$ to address symbols at the start of the .sdata section. The linker assumes that if this symbol is defined, then the gp register contains that value, which it can then use to relax accesses to global symbols within that 12-bit range. The compiler treats the gp register as a constant so it doesn't need to be saved or restored, which means it is generally only written by _start, the ELF entry point."
However, U-Boot instead keeps the global data pointer in gp. This causes memory accesses and jumps optimized to use the gp pointer to fail. To fix this problem, we undefine the __global_pointer symbol.
--- arch/riscv/cpu/u-boot.lds | 1 - 1 file changed, 1 deletion(-)
arch/riscv/cpu/u-boot.lds +0 -1
@@ -32,7 +32,6 @@ SECTIONS . = ALIGN(4); .data : { - __global_pointer$ = . + 0x800; *(.data*) } . = ALIGN(4);
Signature · 1 line
2.23.0
PATCH 1/1 riscv: Fix breakage caused by linker relaxation
Quoted 33 lines · Sean Anderson
Due to the two-instruction sequence needed to access arbitrary memory locations, the RISC-V linker aggressively optimises memory accesses and jumps at link-time. This is called "linker relaxation," and is discussed in this SiFive article
<https://www.sifive.com/blog/all-aboard-part-3-linker-relaxation-in-riscv-toolchain>.
One of the optimizations in place is to assume that the__global_pointersymbol is placed in the gp register. To quote the article:"...The magic __global_pointer$ symbol is defined to point
0x800bytes past the start of the .sdata section. The0x800magic number allows signed 12-bit offsets from __global_pointer$ to address symbols at the start of the .sdata section. The linker assumes that if this symbol is defined, then the gp register contains that value, which it can then use to relax accesses to global symbols within that 12-bit range. The compiler treats the gp register as a constant so it doesn't need to be saved or restored, which means it is generally only written by_start, the ELF entry point."However, U-Boot instead keeps the global data pointer in gp. This causes memory accesses and jumps optimized to use the gp pointer to fail. To fix this problem, we undefine the
__global_pointersymbol.Signed-off-by Sean Anderson <[email protected]>--- arch/riscv/cpu/u-boot.lds | 1 - 1 file changed, 1 deletion(-)arch/riscv/cpu/u-boot.lds +0
-0
arch/riscv/cpu/u-boot.lds SECTIONS
Quoted 8 lines
arch/riscv/cpu/u-boot.lds +0
-2@@ -32,7 +32,6 @@ SECTIONS . = ALIGN(4); .data : { - __global_pointer$ = . + 0x800; *(.data*) } . = ALIGN(4); --
Good catch!
But I wonder how U-Boot managed to work on RISC-V till today?
Regards,
Bin
PATCH 1/1 riscv: Fix breakage caused by linker relaxation
Quoted 1 line
But I wonder how U-Boot managed to work on RISC-V till today?
Hm. I think that linker relaxations may also have been disabled by linking with -pie. When I discovered this bug, I had disabled the flag since I was having a hard time getting my compiler to build with --enable-shared. When recompiling with -pie and with __global_pointer defined, the problem assembly code I noticed now uses the two-instruction auipc/lw sequence.
--Sean
A message that is not in the archive (sent off-list or not yet received): [email protected]
PATCH 1/1 riscv: Fix breakage caused by linker relaxation
Quoted 38 lines
Quoted 38 lines
From: Sean Anderson [mailto:seanga2 at gmail.com]
Sent: Wednesday, December 18, 2019 10:36 AM
To: u-boot at lists.denx.de; Rick Jian-Zhi Chen(陳建志)
Subject: [PATCH] riscv: Fix breakage caused by linker relaxationDue to the two-instruction sequence needed to access arbitrary memory locations, the RISC-V linker aggressively optimises memory accesses and jumps at link-time. This is called "linker relaxation," and is discussed in this SiFive article
<https://www.sifive.com/blog/all-aboard-part-3-linker-relaxation-in-riscv-tool
chain>.
One of the optimizations in place is to assume that the__global_pointersymbol is placed in the gp register. To quote the article:"...The magic __global_pointer$ symbol is defined to point
0x800bytes past the start of the .sdata section. The0x800magic number allows signed 12-bit offsets from __global_pointer$ to address symbols at the start of the .sdata section. The linker assumes that if this symbol is defined, then the gp register contains that value, which it can then use to relax accesses to global symbols within that 12-bit range. The compiler treats the gp register as a constant so it doesn't need to be saved or restored, which means it is generally only written by_start, the ELF entry point."However, U-Boot instead keeps the global data pointer in gp. This causes memory accesses and jumps optimized to use the gp pointer to fail. To fix this problem, we undefine the
__global_pointersymbol.Signed-off-by Sean Anderson <[email protected]>--- arch/riscv/cpu/u-boot.lds | 1 - 1 file changed, 1 deletion(-)arch/riscv/cpu/u-boot.lds index renamed from arch/riscv/cpu/u-boot.lds +0
-0
838a844399..c00d17c736100644arch/riscv/cpu/u-boot.lds +0
-0@@ -32,7 +32,6 @@ SECTIONS
Quoted 6 lines
arch/riscv/cpu/u-boot.lds SECTIONS · line 37
Quoted 6 lines
arch/riscv/cpu/u-boot.lds +0
-1. = ALIGN(4); .data : { - __global_pointer$ = . + 0x800; *(.data*) } . = ALIGN(4);