[PATCH V2 0/3] riscv: atomic: Optimize AMO instructions usage
Guo Ren
guoren at kernel.org
Thu Apr 21 02:39:09 PDT 2022
Hi Dan,
On Thu, Apr 21, 2022 at 1:03 AM Dan Lustig <dlustig at nvidia.com> wrote:
>
> On 4/20/2022 1:33 AM, Guo Ren wrote:
> > Thx Dan,
> >
> > On Wed, Apr 20, 2022 at 1:12 AM Dan Lustig <dlustig at nvidia.com> wrote:
> >>
> >> On 4/17/2022 12:51 AM, Guo Ren wrote:
> >>> Hi Boqun & Andrea,
> >>>
> >>> On Sun, Apr 17, 2022 at 10:26 AM Boqun Feng <boqun.feng at gmail.com> wrote:
> >>>>
> >>>> On Sun, Apr 17, 2022 at 12:49:44AM +0800, Guo Ren wrote:
> >>>> [...]
> >>>>>
> >>>>> If both the aq and rl bits are set, the atomic memory operation is
> >>>>> sequentially consistent and cannot be observed to happen before any
> >>>>> earlier memory operations or after any later memory operations in the
> >>>>> same RISC-V hart and to the same address domain.
> >>>>> "0: lr.w %[p], %[c]\n"
> >>>>> " sub %[rc], %[p], %[o]\n"
> >>>>> " bltz %[rc], 1f\n".
> >>>>> - " sc.w.rl %[rc], %[rc], %[c]\n"
> >>>>> + " sc.w.aqrl %[rc], %[rc], %[c]\n"
> >>>>> " bnez %[rc], 0b\n"
> >>>>> - " fence rw, rw\n"
> >>>>> "1:\n"
> >>>>> So .rl + fence rw, rw is over constraints, only using sc.w.aqrl is more proper.
> >>>>>
> >>>>
> >>>> Can .aqrl order memory accesses before and after it (not against itself,
> >>>> against each other), i.e. act as a full memory barrier? For example, can
> >>> From the RVWMO spec description, the .aqrl annotation appends the same
> >>> effect with "fence rw, rw" to the AMO instruction, so it's RCsc.
> >>>
> >>> Not only .aqrl, and I think the below also could be an RCsc when
> >>> sc.w.aq is executed:
> >>> A: Pre-Access
> >>> B: lr.w.rl ADDR-0
> >>> ...
> >>> C: sc.w.aq ADDR-0
> >>> D: Post-Acess
> >>> Because sc.w.aq has overlap address & data dependency on lr.w.rl, the
> >>> global memory order should be A->B->C->D when sc.w.aq is executed. For
> >>> the amoswap
> >>
> >> These opcodes aren't actually meaningful, unfortunately.
> >>
> >> Quoting the ISA manual chapter 10.2: "Software should not set the rl bit
> >> on an LR instruction unless the aq bit is also set, nor should software
> >> set the aq bit on an SC instruction unless the rl bit is also set."
> > 1. Oh, I've missed the behind half of the ISA manual. But why can't we
> > utilize lr.rl & sc.aq in software programming to guarantee the
> > sequence?
>
> lr.aq and sc.rl map more naturally to hardware than lr.rl and sc.aq.
> Plus, they just aren't common operations to begin with, e.g., there
> is no smp_store_acquire() or smp_load_release(), nor are there
> equivalents in C/C++ atomics.
First, thx for pointing out that my patch violates the rules defined
in the ISA manual. I've abandoned these parts in v3.
It's easy to let hw support lr.rl & sc.aq (eg: our hardware supports
them). I agree there are no equivalents in C/C++ atomics. But they are
useful for LR/SC pairs to implement atomic_acqurie/release semantics.
Compare below:
A): fence rw, r; lr
B): lr.rl
The A has another "fence ,r" effect in semantics, it's over commit
from a software design view.
ps: Current definition has problems:
#define RISCV_ACQUIRE_BARRIER "\tfence r , rw\n"
#define RISCV_RELEASE_BARRIER "\tfence rw, w\n"
#define __cmpxchg_release(ptr, old, new, size) \
...
__asm__ __volatile__ ( \
RISCV_RELEASE_BARRIER \
"0: lr.w %0, %2\n" \
That means "fence rw, w" can't prevent lr.w beyond the fence, we need
a "fence.rw. r" here. Here is the Fixup patch which I'm preparing:
>From 14c93aca0c3b10cf134791cf491b459972a36ec4 Mon Sep 17 00:00:00 2001
From: Guo Ren <guoren at linux.alibaba.com>
Date: Thu, 21 Apr 2022 16:44:48 +0800
Subject: [PATCH] riscv: atomic: Fixup wrong __atomic_acquire/release_fence
implementation
Current RISCV_ACQUIRE/RELEASE_BARRIER is for spin_lock not atomic.
__cmpxchg_release(ptr, old, new, size)
...
__asm__ __volatile__ (
RISCV_RELEASE_BARRIER
"0: lr.w %0, %2\n"
The "fence rw, w -> lr.w" is invalid and lr would beyond fence, so
we need "fence rw, r -> lr.w" here. Atomic acquire is the same.
Fixes: 0123f4d76ca6 ("riscv/spinlock: Strengthen implementations with fences")
Signed-off-by: Guo Ren <guoren at linux.alibaba.com>
Signed-off-by: Guo Ren <guoren at kernel.org>
Cc: Palmer Dabbelt <palmer at dabbelt.com>
Cc: Mark Rutland <mark.rutland at arm.com>
Cc: Andrea Parri <parri.andrea at gmail.com>
Cc: Dan Lustig <dlustig at nvidia.com>
Cc: stable at vger.kernel.org
---
arch/riscv/include/asm/atomic.h | 4 ++--
arch/riscv/include/asm/cmpxchg.h | 8 ++++----
arch/riscv/include/asm/fence.h | 4 ++++
3 files changed, 10 insertions(+), 6 deletions(-)
diff --git a/arch/riscv/include/asm/atomic.h b/arch/riscv/include/asm/atomic.h
index aef8aa9ac4f4..7cd66eba6ec3 100644
--- a/arch/riscv/include/asm/atomic.h
+++ b/arch/riscv/include/asm/atomic.h
@@ -20,10 +20,10 @@
#include <asm/barrier.h>
#define __atomic_acquire_fence() \
- __asm__ __volatile__(RISCV_ACQUIRE_BARRIER "" ::: "memory")
+ __asm__ __volatile__(RISCV_ATOMIC_ACQUIRE_BARRIER "":::"memory")
#define __atomic_release_fence() \
- __asm__ __volatile__(RISCV_RELEASE_BARRIER "" ::: "memory");
+ __asm__ __volatile__(RISCV_ATOMIC_RELEASE_BARRIER"" ::: "memory");
static __always_inline int arch_atomic_read(const atomic_t *v)
{
diff --git a/arch/riscv/include/asm/cmpxchg.h b/arch/riscv/include/asm/cmpxchg.h
index 9269fceb86e0..605edc2fca3b 100644
--- a/arch/riscv/include/asm/cmpxchg.h
+++ b/arch/riscv/include/asm/cmpxchg.h
@@ -217,7 +217,7 @@
" bne %0, %z3, 1f\n" \
" sc.w %1, %z4, %2\n" \
" bnez %1, 0b\n" \
- RISCV_ACQUIRE_BARRIER \
+ RISCV_ATOMIC_ACQUIRE_BARRIER \
"1:\n" \
: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr) \
: "rJ" ((long)__old), "rJ" (__new) \
@@ -229,7 +229,7 @@
" bne %0, %z3, 1f\n" \
" sc.d %1, %z4, %2\n" \
" bnez %1, 0b\n" \
- RISCV_ACQUIRE_BARRIER \
+ RISCV_ATOMIC_ACQUIRE_BARRIER \
"1:\n" \
: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr) \
: "rJ" (__old), "rJ" (__new) \
@@ -259,7 +259,7 @@
switch (size) { \
case 4: \
__asm__ __volatile__ ( \
- RISCV_RELEASE_BARRIER \
+ RISCV_ATOMIC_RELEASE_BARRIER \
"0: lr.w %0, %2\n" \
" bne %0, %z3, 1f\n" \
" sc.w %1, %z4, %2\n" \
@@ -271,7 +271,7 @@
break; \
case 8: \
__asm__ __volatile__ ( \
- RISCV_RELEASE_BARRIER \
+ RISCV_ATOMIC_RELEASE_BARRIER \
"0: lr.d %0, %2\n" \
" bne %0, %z3, 1f\n" \
" sc.d %1, %z4, %2\n" \
diff --git a/arch/riscv/include/asm/fence.h b/arch/riscv/include/asm/fence.h
index 2b443a3a487f..4e446d64f04f 100644
--- a/arch/riscv/include/asm/fence.h
+++ b/arch/riscv/include/asm/fence.h
@@ -4,9 +4,13 @@
#ifdef CONFIG_SMP
#define RISCV_ACQUIRE_BARRIER "\tfence r , rw\n"
#define RISCV_RELEASE_BARRIER "\tfence rw, w\n"
+#define RISCV_ATOMIC_ACQUIRE_BARRIER "\tfence w , rw\n"
+#define RISCV_ATOMIC_RELEASE_BARRIER "\tfence rw, r\n"
#else
#define RISCV_ACQUIRE_BARRIER
#define RISCV_RELEASE_BARRIER
+#define RISCV_ATOMIC_ACQUIRE_BARRIER
+#define RISCV_ATOMIC_RELEASE_BARRIER
#endif
#endif /* _ASM_RISCV_FENCE_H */
>
> > 2. Using .aqrl to replace the fence rw, rw is okay to ISA manual,
> > right? And reducing a fence instruction to gain better performance:
> > "0: lr.w %[p], %[c]\n"
> > " sub %[rc], %[p], %[o]\n"
> > " bltz %[rc], 1f\n".
> > - " sc.w.rl %[rc], %[rc], %[c]\n"
> > + " sc.w.aqrl %[rc], %[rc], %[c]\n"
> > " bnez %[rc], 0b\n"
> > - " fence rw, rw\n"
>
> Yes, using .aqrl is valid.
Thx and I think the below is also valid, right?
- RISCV_RELEASE_BARRIER \
- " amoswap.w %0, %2, %1\n" \
+ " amoswap.w.rl %0, %2, %1\n" \
- " amoswap.d %0, %2, %1\n" \
- RISCV_ACQUIRE_BARRIER \
+ " amoswap.d.aq %0, %2, %1\n" \
>
> Dan
>
> >>
> >> Dan
> >>
> >>> The purpose of the whole patchset is to reduce the usage of
> >>> independent fence rw, rw instructions, and maximize the usage of the
> >>> .aq/.rl/.aqrl aonntation of RISC-V.
> >>>
> >>> __asm__ __volatile__ ( \
> >>> "0: lr.w %0, %2\n" \
> >>> " bne %0, %z3, 1f\n" \
> >>> " sc.w.rl %1, %z4, %2\n" \
> >>> " bnez %1, 0b\n" \
> >>> " fence rw, rw\n" \
> >>> "1:\n" \
> >>>
> >>>> we end up with u == 1, v == 1, r1 on P0 is 0 and r1 on P1 is 0, for the
> >>>> following litmus test?
> >>>>
> >>>> C lr-sc-aqrl-pair-vs-full-barrier
> >>>>
> >>>> {}
> >>>>
> >>>> P0(int *x, int *y, atomic_t *u)
> >>>> {
> >>>> int r0;
> >>>> int r1;
> >>>>
> >>>> WRITE_ONCE(*x, 1);
> >>>> r0 = atomic_cmpxchg(u, 0, 1);
> >>>> r1 = READ_ONCE(*y);
> >>>> }
> >>>>
> >>>> P1(int *x, int *y, atomic_t *v)
> >>>> {
> >>>> int r0;
> >>>> int r1;
> >>>>
> >>>> WRITE_ONCE(*y, 1);
> >>>> r0 = atomic_cmpxchg(v, 0, 1);
> >>>> r1 = READ_ONCE(*x);
> >>>> }
> >>>>
> >>>> exists (u=1 /\ v=1 /\ 0:r1=0 /\ 1:r1=0)
> >>> I think my patchset won't affect the above sequence guarantee. Current
> >>> RISC-V implementation only gives RCsc when the original value is the
> >>> same at least once. So I prefer RISC-V cmpxchg should be:
> >>>
> >>>
> >>> - "0: lr.w %0, %2\n" \
> >>> + "0: lr.w.rl %0, %2\n" \
> >>> " bne %0, %z3, 1f\n" \
> >>> " sc.w.rl %1, %z4, %2\n" \
> >>> " bnez %1, 0b\n" \
> >>> - " fence rw, rw\n" \
> >>> "1:\n" \
> >>> + " fence w, rw\n" \
> >>>
> >>> To give an unconditional RSsc for atomic_cmpxchg.
> >>>
> >>>>
> >>>> Regards,
> >>>> Boqun
> >>>
> >>>
> >>>
> >
> >
> >
--
Best Regards
Guo Ren
ML: https://lore.kernel.org/linux-csky/
More information about the linux-riscv
mailing list