Re: [PATCH v3 1/3] riscv: mm: Use hint address in mmap if available

[Yangyu Chen <cyy@cyyself.name>]

On Wed, 2024-01-31 at 22:41 +0800, Yangyu Chen wrote:
> On Tue, 2024-01-30 at 17:07 -0800, Charlie Jenkins wrote:
> > On riscv it is guaranteed that the address returned by mmap is less
> > than
> > the hint address. Allow mmap to return an address all the way up to
> > addr, if provided, rather than just up to the lower address space.
> > 
> > This provides a performance benefit as well, allowing mmap to exit
> > after
> > checking that the address is in range rather than searching for a
> > valid
> > address.
> > 
> > It is possible to provide an address that uses at most the same
> > number
> > of bits, however it is significantly more computationally expensive
> > to
> > provide that number rather than setting the max to be the hint
> > address.
> > There is the instruction clz/clzw in Zbb that returns the highest
> > set
> > bit
> > which could be used to performantly implement this, but it would
> > still
> > be slower than the current implementation. At worst case, half of
> > the
> > address would not be able to be allocated when a hint address is
> > provided.
> > 
> > Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
> > ---
> >  arch/riscv/include/asm/processor.h | 27 +++++++++++---------------
> > -
> >  1 file changed, 11 insertions(+), 16 deletions(-)
> > 
> > diff --git a/arch/riscv/include/asm/processor.h
> > b/arch/riscv/include/asm/processor.h
> > index f19f861cda54..8ece7a8f0e18 100644
> > --- a/arch/riscv/include/asm/processor.h
> > +++ b/arch/riscv/include/asm/processor.h
> > @@ -14,22 +14,16 @@
> >  
> >  #include <asm/ptrace.h>
> >  
> > -#ifdef CONFIG_64BIT
> > -#define DEFAULT_MAP_WINDOW	(UL(1) << (MMAP_VA_BITS - 1))
> > -#define STACK_TOP_MAX		TASK_SIZE_64
> > -
> >  #define arch_get_mmap_end(addr, len, flags)			\
> >  ({								\
> >  	unsigned long
> > mmap_end;					\
> >  	typeof(addr) _addr = (addr);				\
> > -	if ((_addr) == 0 || (IS_ENABLED(CONFIG_COMPAT) &&
> > is_compat_task())) \
> > +	if ((_addr) == 0 ||					\
> > +	    (IS_ENABLED(CONFIG_COMPAT) && is_compat_task()) ||	\
> > +	    ((_addr + len) > BIT(VA_BITS -
> > 1)))			\
> >  		mmap_end = STACK_TOP_MAX;			\
> > -	else if ((_addr) >= VA_USER_SV57)			\
> > -		mmap_end = STACK_TOP_MAX;			\
> > -	else if ((((_addr) >= VA_USER_SV48)) && (VA_BITS >=
> > VA_BITS_SV48)) \
> > -		mmap_end = VA_USER_SV48;			\
> >  	else							\
> > -		mmap_end = VA_USER_SV39;			\
> > +		mmap_end = (_addr + len);			\
> >  	mmap_end;						\
> >  })
> >  
> > @@ -39,17 +33,18 @@
> >  	typeof(addr) _addr = (addr);				\
> >  	typeof(base) _base = (base);				\
> >  	unsigned long rnd_gap = DEFAULT_MAP_WINDOW - (_base);	\
> > -	if ((_addr) == 0 || (IS_ENABLED(CONFIG_COMPAT) &&
> > is_compat_task())) \
> > +	if ((_addr) == 0 ||					\
> > +	    (IS_ENABLED(CONFIG_COMPAT) && is_compat_task()) ||	\
> > +	    ((_addr + len) > BIT(VA_BITS -
> > 1)))			\
> >  		mmap_base = (_base);				\
> > -	else if (((_addr) >= VA_USER_SV57) && (VA_BITS >=
> > VA_BITS_SV57)) \
> > -		mmap_base = VA_USER_SV57 - rnd_gap;		\
> > -	else if ((((_addr) >= VA_USER_SV48)) && (VA_BITS >=
> > VA_BITS_SV48)) \
> > -		mmap_base = VA_USER_SV48 - rnd_gap;		\
> >  	else							\
> > -		mmap_base = VA_USER_SV39 - rnd_gap;		\
> > +		mmap_base = (_addr + len) - rnd_gap;		\
> >  	mmap_base;						\
> >  })
> >  
> > +#ifdef CONFIG_64BIT
> > +#define DEFAULT_MAP_WINDOW	(UL(1) << (MMAP_VA_BITS - 1))
> > +#define STACK_TOP_MAX		TASK_SIZE_64
> >  #else
> >  #define DEFAULT_MAP_WINDOW	TASK_SIZE
> >  #define STACK_TOP_MAX		TASK_SIZE
> > 
> 
> I have carefully tested your patch on qemu with sv57. A bug that
> needs
> to be solved is that mmap with the same hint address without
> MAP_FIXED
> set will fail the second time.
> 
> Userspace code to reproduce the bug:
> 
> #include <sys/mman.h>
> #include <stdio.h>
> #include <stdint.h>
> 
> void test(char *addr) {
>     char *res = mmap(addr, 4096, PROT_READ | PROT_WRITE,
> MAP_ANONYMOUS
> > MAP_PRIVATE, -1, 0);
>     printf("hint %p got %p.\n", addr, res);
> }
> 
> int main (void) {
>     test(1<<30);
>     test(1<<30);
>     test(1<<30);
>     return 0;
> }
> 
> output:
> 
> hint 0x40000000 got 0x40000000.
> hint 0x40000000 got 0xffffffffffffffff.
> hint 0x40000000 got 0xffffffffffffffff.
> 
> output on x86:
> 
> hint 0x40000000 got 0x40000000.
> hint 0x40000000 got 0x7f9171363000.
> hint 0x40000000 got 0x7f9171362000.
> 
> It may need to implement a special arch_get_unmapped_area and
> arch_get_unmapped_area_topdown function.
> 

This is because hint address < rnd_gap. I have tried to let mmap_base =
min((_addr + len), (base) + TASK_SIZE - DEFAULT_MAP_WINDOW). However it
does not work for bottom-up while ulimit -s is unlimited. You said this
behavior is expected from patch v2 review. However it brings a new
regression even on sv39 systems.

I still don't know the reason why use addr+len as the upper-bound. I
think solution like x86/arm64/powerpc provide two address space switch
based on whether hint address above the default map window is enough.