All of lore.kernel.org
 help / color / mirror / Atom feed
From: peterz@infradead.org (Peter Zijlstra)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 2/2] arm64: bpf: add BPF XADD instruction
Date: Wed, 11 Nov 2015 19:54:15 +0100	[thread overview]
Message-ID: <20151111185415.GI11639@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <20151111184427.GH11639@twins.programming.kicks-ass.net>

On Wed, Nov 11, 2015 at 07:44:27PM +0100, Peter Zijlstra wrote:
> On Wed, Nov 11, 2015 at 07:31:28PM +0100, Peter Zijlstra wrote:
> > > Add new one that does 'fetch_and_add' ? What is the real use case it
> > > will be used for?
> > 
> > Look at all the atomic_{add,dec}_return*() users in the kernel. A typical
> > example would be a reader-writer lock implementations. See
> > include/asm-generic/rwsem.h for examples.
> 
> Maybe a better example would be refcounting, where you free on 0.
> 
> 	if (!fetch_add(&obj->ref, -1))
> 		free(obj);

Urgh, too used to the atomic_add_return(), which returns post op. That
wants to be:

	if (fetch_add(&obj->ref, -1) == 1)
		free(obj);

Note that I would very much recommend _against_ encoding the post-op
thing in instructions. It works for reversible operations (like add) but
is pointless for irreversible operations (like or).

That is, given or_return(), you cannot reconstruct the state
prior to the operation, so or_return() provides less information than
fetch_or().

WARNING: multiple messages have this Message-ID (diff)
From: Peter Zijlstra <peterz@infradead.org>
To: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: David Miller <davem@davemloft.net>,
	will.deacon@arm.com, daniel@iogearbox.net, arnd@arndb.de,
	yang.shi@linaro.org, linaro-kernel@lists.linaro.org,
	eric.dumazet@gmail.com, zlim.lnx@gmail.com, ast@kernel.org,
	linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
	xi.wang@gmail.com, catalin.marinas@arm.com,
	linux-arm-kernel@lists.infradead.org, yhs@plumgrid.com,
	bblanco@plumgrid.com
Subject: Re: [PATCH 2/2] arm64: bpf: add BPF XADD instruction
Date: Wed, 11 Nov 2015 19:54:15 +0100	[thread overview]
Message-ID: <20151111185415.GI11639@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <20151111184427.GH11639@twins.programming.kicks-ass.net>

On Wed, Nov 11, 2015 at 07:44:27PM +0100, Peter Zijlstra wrote:
> On Wed, Nov 11, 2015 at 07:31:28PM +0100, Peter Zijlstra wrote:
> > > Add new one that does 'fetch_and_add' ? What is the real use case it
> > > will be used for?
> > 
> > Look at all the atomic_{add,dec}_return*() users in the kernel. A typical
> > example would be a reader-writer lock implementations. See
> > include/asm-generic/rwsem.h for examples.
> 
> Maybe a better example would be refcounting, where you free on 0.
> 
> 	if (!fetch_add(&obj->ref, -1))
> 		free(obj);

Urgh, too used to the atomic_add_return(), which returns post op. That
wants to be:

	if (fetch_add(&obj->ref, -1) == 1)
		free(obj);

Note that I would very much recommend _against_ encoding the post-op
thing in instructions. It works for reversible operations (like add) but
is pointless for irreversible operations (like or).

That is, given or_return(), you cannot reconstruct the state
prior to the operation, so or_return() provides less information than
fetch_or().



  reply	other threads:[~2015-11-11 18:54 UTC|newest]

Thread overview: 86+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-10 22:41 [PATCH 0/2] arm64: bpf: add BPF_ST and BPF_XADD instructions support Yang Shi
2015-11-10 22:41 ` Yang Shi
2015-11-10 22:41 ` Yang Shi
2015-11-10 22:41 ` [PATCH 1/2] arm64: bpf: add 'store immediate' instruction Yang Shi
2015-11-10 22:41   ` Yang Shi
2015-11-11  2:45   ` Z Lim
2015-11-11  2:45     ` Z Lim
2015-11-11 12:12     ` Will Deacon
2015-11-11 12:12       ` Will Deacon
2015-11-11 12:39       ` Will Deacon
2015-11-11 12:39         ` Will Deacon
2015-11-12 19:33         ` Shi, Yang
2015-11-12 19:33           ` Shi, Yang
2015-11-13  3:45           ` Z Lim
2015-11-13  3:45             ` Z Lim
2015-11-23 19:34             ` Shi, Yang
2015-11-23 19:34               ` Shi, Yang
2015-11-10 22:41 ` [PATCH 2/2] arm64: bpf: add BPF XADD instruction Yang Shi
2015-11-10 22:41   ` Yang Shi
2015-11-11  0:08   ` Eric Dumazet
2015-11-11  0:08     ` Eric Dumazet
2015-11-11  0:26     ` Shi, Yang
2015-11-11  0:26       ` Shi, Yang
2015-11-11  0:42       ` Alexei Starovoitov
2015-11-11  0:42         ` Alexei Starovoitov
2015-11-11  2:52         ` Z Lim
2015-11-11  2:52           ` Z Lim
2015-11-11  8:49           ` Arnd Bergmann
2015-11-11  8:49             ` Arnd Bergmann
2015-11-11 10:24             ` Will Deacon
2015-11-11 10:24               ` Will Deacon
2015-11-11 10:42               ` Daniel Borkmann
2015-11-11 10:42                 ` Daniel Borkmann
2015-11-11 11:58                 ` Will Deacon
2015-11-11 11:58                   ` Will Deacon
2015-11-11 12:21                   ` Daniel Borkmann
2015-11-11 12:21                     ` Daniel Borkmann
2015-11-11 12:38                     ` Will Deacon
2015-11-11 12:38                       ` Will Deacon
2015-11-11 12:58                       ` Peter Zijlstra
2015-11-11 12:58                         ` Peter Zijlstra
2015-11-11 15:52                         ` Daniel Borkmann
2015-11-11 15:52                           ` Daniel Borkmann
2015-11-11 16:23                           ` Will Deacon
2015-11-11 16:23                             ` Will Deacon
2015-11-11 17:27                             ` Alexei Starovoitov
2015-11-11 17:27                               ` Alexei Starovoitov
2015-11-11 17:35                               ` David Miller
2015-11-11 17:35                                 ` David Miller
2015-11-11 17:44                                 ` Will Deacon
2015-11-11 17:44                                   ` Will Deacon
2015-11-11 19:01                                   ` David Miller
2015-11-11 19:01                                     ` David Miller
2015-11-11 17:57                                 ` Peter Zijlstra
2015-11-11 17:57                                   ` Peter Zijlstra
2015-11-11 18:11                                   ` Alexei Starovoitov
2015-11-11 18:11                                     ` Alexei Starovoitov
2015-11-11 18:31                                     ` Peter Zijlstra
2015-11-11 18:31                                       ` Peter Zijlstra
2015-11-11 18:41                                       ` Peter Zijlstra
2015-11-11 18:41                                         ` Peter Zijlstra
2015-11-11 18:44                                       ` Peter Zijlstra
2015-11-11 18:44                                         ` Peter Zijlstra
2015-11-11 18:54                                         ` Peter Zijlstra [this message]
2015-11-11 18:54                                           ` Peter Zijlstra
2015-11-11 19:55                                           ` Alexei Starovoitov
2015-11-11 19:55                                             ` Alexei Starovoitov
2015-11-11 22:21                                             ` Peter Zijlstra
2015-11-11 22:21                                               ` Peter Zijlstra
2015-11-11 23:40                                               ` Alexei Starovoitov
2015-11-11 23:40                                                 ` Alexei Starovoitov
2015-11-11 23:40                                                 ` Alexei Starovoitov
2015-11-12  8:57                                                 ` Peter Zijlstra
2015-11-12  8:57                                                   ` Peter Zijlstra
2015-11-11 18:50                                       ` Daniel Borkmann
2015-11-11 18:50                                         ` Daniel Borkmann
2015-11-11 19:04                                         ` David Miller
2015-11-11 19:04                                           ` David Miller
2015-11-11 19:23                                         ` Peter Zijlstra
2015-11-11 19:23                                           ` Peter Zijlstra
2015-11-11 19:41                                           ` Daniel Borkmann
2015-11-11 19:41                                             ` Daniel Borkmann
2015-11-11 18:46                                     ` Will Deacon
2015-11-11 18:46                                       ` Will Deacon
2015-11-11 19:01                                     ` David Miller
2015-11-11 19:01                                       ` David Miller

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=20151111185415.GI11639@twins.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.