bpf.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Brendan Jackman <jackmanb@google.com>
To: Yonghong Song <yhs@fb.com>
Cc: bpf@vger.kernel.org, Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	KP Singh <kpsingh@chromium.org>,
	Florent Revest <revest@chromium.org>,
	linux-kernel@vger.kernel.org, Jann Horn <jannh@google.com>
Subject: Re: [PATCH bpf-next v4 06/11] bpf: Add BPF_FETCH field / create atomic_fetch_add instruction
Date: Tue, 8 Dec 2020 09:31:18 +0000	[thread overview]
Message-ID: <X89H5i6OaYoNaUhN@google.com> (raw)
In-Reply-To: <4163e34b-754a-5607-c28a-4c575a2cc6e5@fb.com>

On Mon, Dec 07, 2020 at 05:41:05PM -0800, Yonghong Song wrote:
> 
> 
> On 12/7/20 8:07 AM, Brendan Jackman wrote:
> > The BPF_FETCH field can be set in bpf_insn.imm, for BPF_ATOMIC
> > instructions, in order to have the previous value of the
> > atomically-modified memory location loaded into the src register
> > after an atomic op is carried out.
> > 
> > Suggested-by: Yonghong Song <yhs@fb.com>
> > Signed-off-by: Brendan Jackman <jackmanb@google.com>
> > ---
> >   arch/x86/net/bpf_jit_comp.c    |  4 ++++
> >   include/linux/filter.h         |  1 +
> >   include/uapi/linux/bpf.h       |  3 +++
> >   kernel/bpf/core.c              | 13 +++++++++++++
> >   kernel/bpf/disasm.c            |  7 +++++++
> >   kernel/bpf/verifier.c          | 33 ++++++++++++++++++++++++---------
> >   tools/include/linux/filter.h   | 11 +++++++++++
> >   tools/include/uapi/linux/bpf.h |  3 +++
> >   8 files changed, 66 insertions(+), 9 deletions(-)
> > 
> > diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
> [...]
> 
> > index f345f12c1ff8..4e0100ba52c2 100644
> > --- a/tools/include/linux/filter.h
> > +++ b/tools/include/linux/filter.h
> > @@ -173,6 +173,7 @@
> >    * Atomic operations:
> >    *
> >    *   BPF_ADD                  *(uint *) (dst_reg + off16) += src_reg
> > + *   BPF_ADD | BPF_FETCH      src_reg = atomic_fetch_add(dst_reg + off16, src_reg);
> >    */
> >   #define BPF_ATOMIC64(OP, DST, SRC, OFF)				\
> > @@ -201,6 +202,16 @@
> >   		.off   = OFF,					\
> >   		.imm   = BPF_ADD })
> > +/* Atomic memory add with fetch, src_reg = atomic_fetch_add(dst_reg + off, src_reg); */
> > +
> > +#define BPF_ATOMIC_FETCH_ADD(SIZE, DST, SRC, OFF)		\
> > +	((struct bpf_insn) {					\
> > +		.code  = BPF_STX | BPF_SIZE(SIZE) | BPF_ATOMIC,	\
> > +		.dst_reg = DST,					\
> > +		.src_reg = SRC,					\
> > +		.off   = OFF,					\
> > +		.imm   = BPF_ADD | BPF_FETCH })
> 
> Not sure whether it is a good idea or not to fold this into BPF_ATOMIC
> macro. At least you can define BPF_ATOMIC macro and
> #define BPF_ATOMIC_FETCH_ADD(SIZE, DST, SRC, OFF)		\
>     BPF_ATOMIC(SIZE, DST, SRC, OFF, BPF_ADD | BPF_FETCH)
> 
> to avoid too many code duplications?

Oops.. I intended to totally get rid these and folded them into
BPF_ATOMIC{64,32}! OK, let's combine all of them into a single macro.
It will have to be called something slightly awkward like
BPF_ATOMIC_INSN because BPF_ATOMIC is the name of the BPF_OP.

> 
> > +
> >   /* Memory store, *(uint *) (dst_reg + off16) = imm32 */
> >   #define BPF_ST_MEM(SIZE, DST, OFF, IMM)				\
> > diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
> > index 98161e2d389f..d5389119291e 100644
> > --- a/tools/include/uapi/linux/bpf.h
> > +++ b/tools/include/uapi/linux/bpf.h
> > @@ -44,6 +44,9 @@
> >   #define BPF_CALL	0x80	/* function call */
> >   #define BPF_EXIT	0x90	/* function return */
> > +/* atomic op type fields (stored in immediate) */
> > +#define BPF_FETCH	0x01	/* fetch previous value into src reg */
> > +
> >   /* Register numbers */
> >   enum {
> >   	BPF_REG_0 = 0,
> > 

  reply	other threads:[~2020-12-08  9:32 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-07 16:07 [PATCH bpf-next v4 00/11] Atomics for eBPF Brendan Jackman
2020-12-07 16:07 ` [PATCH bpf-next v4 01/11] bpf: x86: Factor out emission of ModR/M for *(reg + off) Brendan Jackman
2020-12-07 21:04   ` John Fastabend
2020-12-07 16:07 ` [PATCH bpf-next v4 02/11] bpf: x86: Factor out emission of REX byte Brendan Jackman
2020-12-07 21:07   ` John Fastabend
2020-12-07 16:07 ` [PATCH bpf-next v4 03/11] bpf: x86: Factor out a lookup table for some ALU opcodes Brendan Jackman
2020-12-07 21:08   ` John Fastabend
2020-12-07 16:07 ` [PATCH bpf-next v4 04/11] bpf: Rename BPF_XADD and prepare to encode other atomics in .imm Brendan Jackman
2020-12-07 21:56   ` John Fastabend
2020-12-08  9:26     ` Brendan Jackman
2020-12-09  5:40       ` John Fastabend
2020-12-07 16:07 ` [PATCH bpf-next v4 05/11] bpf: Move BPF_STX reserved field check into BPF_STX verifier code Brendan Jackman
2020-12-08  1:35   ` Yonghong Song
2020-12-08  5:13   ` John Fastabend
2020-12-07 16:07 ` [PATCH bpf-next v4 06/11] bpf: Add BPF_FETCH field / create atomic_fetch_add instruction Brendan Jackman
2020-12-08  1:41   ` Yonghong Song
2020-12-08  9:31     ` Brendan Jackman [this message]
2020-12-08  5:31   ` John Fastabend
2020-12-08  9:59     ` Brendan Jackman
2020-12-07 16:07 ` [PATCH bpf-next v4 07/11] bpf: Add instructions for atomic_[cmp]xchg Brendan Jackman
2020-12-08  1:44   ` Yonghong Song
2020-12-08  6:37   ` John Fastabend
2020-12-14 15:39     ` Brendan Jackman
2020-12-08  6:42   ` John Fastabend
2020-12-07 16:07 ` [PATCH bpf-next v4 08/11] bpf: Pull out a macro for interpreting atomic ALU operations Brendan Jackman
2020-12-07 16:07 ` [PATCH bpf-next v4 09/11] bpf: Add bitwise atomic instructions Brendan Jackman
2020-12-08  1:47   ` Yonghong Song
2020-12-10  0:22   ` kernel test robot
2020-12-07 16:07 ` [PATCH bpf-next v4 10/11] bpf: Add tests for new BPF atomic operations Brendan Jackman
2020-12-08  3:18   ` Yonghong Song
2020-12-08 12:41     ` Brendan Jackman
2020-12-08 16:38       ` Yonghong Song
2020-12-08 16:59         ` Brendan Jackman
2020-12-08 18:15           ` Yonghong Song
2020-12-15 11:12             ` Brendan Jackman
2020-12-16  7:18               ` Yonghong Song
2020-12-16 11:51                 ` Brendan Jackman
2020-12-16 20:00                   ` Yonghong Song
2020-12-07 16:07 ` [PATCH bpf-next v4 11/11] bpf: Document new atomic instructions Brendan Jackman
2020-12-08  3:25   ` Yonghong Song

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=X89H5i6OaYoNaUhN@google.com \
    --to=jackmanb@google.com \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=jannh@google.com \
    --cc=kpsingh@chromium.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=revest@chromium.org \
    --cc=yhs@fb.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).