Date: Tue, 4 Aug 2020 20:25:41 +0200 From: Emmanuel Vadot <manu@bidouilliste.com> To: Alan Cox <alc@rice.edu> Cc: Emmanuel Vadot <manu@FreeBSD.org>, src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: Re: svn commit: r363842 - head/sys/compat/linuxkpi/common/include/linux Message-ID: <20200804202541.0b6b3eb6ce26522ee12a796c@bidouilliste.com> In-Reply-To: <1c90911b-d5f6-f8b1-8f4d-97a88fa028bd@rice.edu> References: <202008041525.074FPNLh043749@repo.freebsd.org> <1c90911b-d5f6-f8b1-8f4d-97a88fa028bd@rice.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 4 Aug 2020 13:11:02 -0500 Alan Cox <alc@rice.edu> wrote: >=20 > On 8/4/20 10:25 AM, Emmanuel Vadot wrote: > > Author: manu > > Date: Tue Aug 4 15:25:22 2020 > > New Revision: 363842 > > URL: https://svnweb.freebsd.org/changeset/base/363842 > > > > Log: > > linuxkpi: Add clear_bit_unlock > > =20 > > This calls clear_bit and adds a memory barrier. > > =20 > > Sponsored by: The FreeBSD Foundation > > =20 > > Reviewed by: hselasky > > MFC after: 1 week > > Differential Revision: https://reviews.freebsd.org/D25943 > > > > Modified: > > head/sys/compat/linuxkpi/common/include/linux/bitops.h > > > > Modified: head/sys/compat/linuxkpi/common/include/linux/bitops.h > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D > > --- head/sys/compat/linuxkpi/common/include/linux/bitops.h Tue Aug 4 1= 5:00:02 2020 (r363841) > > +++ head/sys/compat/linuxkpi/common/include/linux/bitops.h Tue Aug 4 1= 5:25:22 2020 (r363842) > > @@ -275,6 +275,13 @@ find_next_zero_bit(const unsigned long *addr, unsi= gned > > #define test_bit(i, a) \ > > !!(READ_ONCE(((volatile const unsigned long *)(a))[BIT_WORD(i)]) = & BIT_MASK(i)) > > =20 > > +static inline void > > +clear_bit_unlock(long bit, volatile unsigned long *var) > > +{ > > + clear_bit(bit, var); > > + wmb(); >=20 >=20 > For an unlock operation, the memory barrier should come before the=20 > clear_bit() call, not after.=A0 See, for example, the alpha implementatio= n=20 > in Linux.=A0 Also, the correct "spelling" for this memory barrier in=20 > FreeBSD would be atomic_thread_fence_rel(). See, for example, the=20 > comment at the top of sys/amd64/include/atomic.h. Ah yes, thanks. I probably got lost looking for the linux implem but that does make sense, I'll fix that probably tomorow. Thanks. >=20 > > +} > > + > > static inline int > > test_and_clear_bit(long bit, volatile unsigned long *var) > > { --=20 Emmanuel Vadot <manu@bidouilliste.com> <manu@freebsd.org>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20200804202541.0b6b3eb6ce26522ee12a796c>