On Tue, Nov 5, 2024 at 5:50 PM Mayshao-oc <[email protected]> wrote:
>
>
> >
> >
> > On Tue, Nov 5, 2024 at 2:34 PM Liu, Hongtao <[email protected]> wrote:
> > >
> > >
> > >
> > > > -----Original Message-----
> > > > From: MayShao-oc <[email protected]>
> > > > Sent: Tuesday, November 5, 2024 11:20 AM
> > > > To: [email protected]; [email protected]; Liu, Hongtao
> > > > <[email protected]>; [email protected]
> > > > Cc: [email protected]; [email protected]; [email protected];
> > > > [email protected]
> > > > Subject: [PATCH] [x86_64] Add flag to control tight loops alignment opt
> > > >
> > > > Hi all:
> > > > This patch add -malign-tight-loops flag to control
> > > > pass_align_tight_loops.
> > > > The motivation is that pass_align_tight_loops may cause performance
> > > > regression in nested loops.
> > > >
> > > > The example code as follows:
> > > >
> > > > #define ITER 20000
> > > > #define ITER_O 10
> > > >
> > > > int i, j,k;
> > > > int array[ITER];
> > > >
> > > > void loop()
> > > > {
> > > > int i;
> > > > for(k = 0; k < ITER_O; k++)
> > > > for(j = 0; j < ITER; j++)
> > > > for(i = 0; i < ITER; i++)
> > > > {
> > > > array[i] += j;
> > > > array[i] += i;
> > > > array[i] += 2*j;
> > > > array[i] += 2*i;
> > > > }
> > > > }
> > > >
> > > > When I compile it with gcc -O1 loop.c, the output assembly as
> > > > follows.
> > > > It is not optimal, because of too many nops insert in the outer loop.
> > > >
> > > > 0000000000400540 <loop>:
> > > > 400540: 48 83 ec 08 sub $0x8,%rsp
> > > > 400544: bf 0a 00 00 00 mov $0xa,%edi
> > > > 400549: b9 00 00 00 00 mov $0x0,%ecx
> > > > 40054e: 8d 34 09 lea (%rcx,%rcx,1),%esi
> > > > 400551: b8 00 00 00 00 mov $0x0,%eax
> > > > 400556: 66 66 2e 0f 1f 84 00 data16 nopw %cs:0x0(%rax,%rax,1)
> > > > 40055d: 00 00 00 00
> > > > 400561: 66 66 2e 0f 1f 84 00 data16 nopw %cs:0x0(%rax,%rax,1)
> > > > 400568: 00 00 00 00
> > > > 40056c: 66 66 2e 0f 1f 84 00 data16 nopw %cs:0x0(%rax,%rax,1)
> > > > 400573: 00 00 00 00
> > > > 400577: 66 0f 1f 84 00 00 00 nopw 0x0(%rax,%rax,1)
> > > > 40057e: 00 00
> > > > 400580: 89 ca mov %ecx,%edx
> > > > 400582: 03 14 85 60 10 60 00 add 0x601060(,%rax,4),%edx
> > > > 400589: 01 c2 add %eax,%edx
> > > > 40058b: 01 f2 add %esi,%edx
> > > > 40058d: 8d 14 42 lea (%rdx,%rax,2),%edx
> > > > 400590: 89 14 85 60 10 60 00 mov %edx,0x601060(,%rax,4)
> > > > 400597: 48 83 c0 01 add $0x1,%rax
> > > > 40059b: 48 3d 20 4e 00 00 cmp $0x4e20,%rax
> > > > 4005a1: 75 dd jne 400580 <loop+0x40>
> > > >
> > > > I benchmark this program in the intel Xeon, and find the
> > > > optimization may
> > > > cause a 40% performance regression (6.6B cycles VS 9.3B cycles).
> > On SPR, align is 25% better than no_align case.
>
> I found no_align is 10% better in zhaoxin yongfeng, so I test this
> program in Xeon 4210R, and found a
> 40% performance regression.So I think this maybe a general regression, and
> need a flag to control.As you say,
> On SPR, align is better,so its not a general regression.
> Could you please benchmark this in Xeon 4210R, or a similar arch?
> I am not delve into intel 4210R arch, and I think a 40% drop is not
> explainable. Maybe I make a mistakes.
> I could confirm in zhaoxin yongfeng, no_align is 10% better.
> I attach the Xeon 4210R cpuinfo, and the test binary for your
> reference. Thanks.
I reproduce with 30% regression on CLX, there's more frontend-bound
with aligned case, it's uarch specific, will make it a uarch tune.
> >
> > > > So I propose to add -malign-tight-loops flag to control tight loop
> > > > optimization to avoid this, we could disalbe this optimization by
> > > > default.
> > > > Bootstrapped X86_64.
> > > > Ok for trunk?
> > > >
> > > > BR
> > > > Mayshao
> > > >
> > > > gcc/ChangeLog:
> > > >
> > > > * config/i386/i386-features.cc (ix86_align_tight_loops): New flag.
> > > > * config/i386/i386.opt (malign-tight-loops): New option.
> > > > * doc/invoke.texi (-malign-tight-loops): Document.
> > > > ---
> > > > gcc/config/i386/i386-features.cc | 4 +++-
> > > > gcc/config/i386/i386.opt | 4 ++++
> > > > gcc/doc/invoke.texi | 7 ++++++-
> > > > 3 files changed, 13 insertions(+), 2 deletions(-)
> > > >
> > > > diff --git a/gcc/config/i386/i386-features.cc b/gcc/config/i386/i386-
> > > > features.cc
> > > > index e2e85212a4f..f9546e00b07 100644
> > > > --- a/gcc/config/i386/i386-features.cc
> > > > +++ b/gcc/config/i386/i386-features.cc
> > > > @@ -3620,7 +3620,9 @@ public:
> > > > /* opt_pass methods: */
> > > > bool gate (function *) final override
> > > > {
> > > > - return optimize && optimize_function_for_speed_p (cfun);
> > > > + return ix86_align_tight_loops
> > > > + && optimize
> > > > + && optimize_function_for_speed_p (cfun);
> > > > }
> > > >
> > > > unsigned int execute (function *) final override diff --git
> > > > a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt index
> > > > 64c295d344c..ec41de192bc 100644
> > > > --- a/gcc/config/i386/i386.opt
> > > > +++ b/gcc/config/i386/i386.opt
> > > > @@ -1266,6 +1266,10 @@ mlam=
> > > > Target RejectNegative Joined Enum(lam_type) Var(ix86_lam_type)
> > > > Init(lam_none) -mlam=[none|u48|u57] Instrument meta data position in
> > > > user data pointers.
> > > >
> > > > +malign-tight-loops
> > > > +Target Var(ix86_align_tight_loops) Init(0) Optimization Enable align
> > > > +tight loops.
> > >
> > > I'd like it to be on by default, so Init (1)?
> > >
> > > > +
> > > > Enum
> > > > Name(lam_type) Type(enum lam_type) UnknownError(unknown lam
> > > > type %qs)
> > > >
> > > > diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi index
> > > > 07920e07b4d..9ec1e1f0095 100644
> > > > --- a/gcc/doc/invoke.texi
> > > > +++ b/gcc/doc/invoke.texi
> > > > @@ -1510,7 +1510,7 @@ See RS/6000 and PowerPC Options.
> > > > -mindirect-branch=@var{choice} -mfunction-return=@var{choice} -
> > > > mindirect-branch-register -mharden-sls=@var{choice}
> > > > -mindirect-branch-cs-
> > > > prefix -mneeded -mno-direct-extern-access --munroll-only-small-loops -
> > > > mlam=@var{choice}}
> > > > +-munroll-only-small-loops -mlam=@var{choice} -malign-tight-loops}
> > > >
> > > > @emph{x86 Windows Options}
> > > >
> > > > @@ -36530,6 +36530,11 @@ LAM(linear-address masking) allows special
> > > > bits in the pointer to be used for metadata. The default is
> > > > @samp{none}. With
> > > > @samp{u48}, pointer bits in positions 62:48 can be used for metadata;
> > > > With
> > > > @samp{u57}, pointer bits in positions 62:57 can be used for metadata.
> > > > +
> > > > +@opindex malign-tight-loops
> > > > +@opindex mno-align-tight-loops
> > > > +@item -malign-tight-loops
> > > > +Controls tight loop alignment optimization.
> > > > @end table
> > > >
> > > > @node x86 Windows Options
> > > > --
> > > > 2.27.0
> > >
> >
> >
> > --
> > BR,
> > Hongtao
> BR
> Mayshao
--
BR,
Hongtao
