From: "Alex Bennée" <alex.bennee@linaro.org>
To: "Emilio G. Cota" <cota@braap.org>
Cc: qemu-devel@nongnu.org,
Richard Henderson <richard.henderson@linaro.org>,
Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v2 14/17] translate-all: protect TB jumps with a per-destination-TB lock
Date: Fri, 20 Apr 2018 17:13:03 +0100 [thread overview]
Message-ID: <87k1t1amb4.fsf@linaro.org> (raw)
In-Reply-To: <1522980788-1252-15-git-send-email-cota@braap.org>
Emilio G. Cota <cota@braap.org> writes:
> This applies to both user-mode and !user-mode emulation.
>
> Instead of relying on a global lock, protect the list of incoming
> jumps with tb->jmp_lock. This lock also protects tb->cflags,
> so update all tb->cflags readers outside tb->jmp_lock to use
> atomic reads via tb_cflags().
>
> In order to find the destination TB (and therefore its jmp_lock)
> from the origin TB, we introduce tb->jmp_dest[].
>
> I considered not using a linked list of jumps, which simplifies
> code and makes the struct smaller. However, it unnecessarily increases
> memory usage, which results in a performance decrease. See for
> instance these numbers booting+shutting down debian-arm:
> Time (s) Rel. err (%) Abs. err (s) Rel. slowdown (%)
> ------------------------------------------------------------------------------
> before 20.88 0.74 0.154512 0.
> after 20.81 0.38 0.079078 -0.33524904
> GTree 21.02 0.28 0.058856 0.67049808
> GHashTable + xxhash 21.63 1.08 0.233604 3.5919540
>
> Using a hash table or a binary tree to keep track of the jumps
> doesn't really pay off, not only due to the increased memory usage,
> but also because most TBs have only 0 or 1 jumps to them. The maximum
> number of jumps when booting debian-arm that I measured is 35, but
> as we can see in the histogram below a TB with that many incoming jumps
> is extremely rare; the average TB has 0.80 incoming jumps.
>
> n_jumps: 379208; avg jumps/tb: 0.801099
> dist: [0.0,1.0)|▄█▁▁▁▁▁▁▁▁▁▁▁ ▁▁▁▁▁▁ ▁▁▁ ▁▁▁ ▁|[34.0,35.0]
>
> Signed-off-by: Emilio G. Cota <cota@braap.org>
This now conflicts with master
--
Alex Bennée
next prev parent reply other threads:[~2018-04-20 16:13 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-04-06 2:12 [Qemu-devel] [PATCH v2 00/17] tcg: tb_lock_removal redux v2 Emilio G. Cota
2018-04-06 2:12 ` [Qemu-devel] [PATCH v2 01/17] qht: require a default comparison function Emilio G. Cota
2018-04-06 2:12 ` [Qemu-devel] [PATCH v2 02/17] qht: return existing entry when qht_insert fails Emilio G. Cota
2018-04-06 2:12 ` [Qemu-devel] [PATCH v2 03/17] tcg: track TBs with per-region BST's Emilio G. Cota
2018-04-06 2:12 ` [Qemu-devel] [PATCH v2 04/17] tcg: move tb_ctx.tb_phys_invalidate_count to tcg_ctx Emilio G. Cota
2018-04-06 2:12 ` [Qemu-devel] [PATCH v2 05/17] translate-all: iterate over TBs in a page with PAGE_FOR_EACH_TB Emilio G. Cota
2018-04-06 2:12 ` [Qemu-devel] [PATCH v2 06/17] translate-all: make l1_map lockless Emilio G. Cota
2018-04-06 2:12 ` [Qemu-devel] [PATCH v2 07/17] translate-all: remove hole in PageDesc Emilio G. Cota
2018-04-06 2:12 ` [Qemu-devel] [PATCH v2 08/17] translate-all: work page-by-page in tb_invalidate_phys_range_1 Emilio G. Cota
2018-04-06 2:13 ` [Qemu-devel] [PATCH v2 09/17] translate-all: move tb_invalidate_phys_page_range up in the file Emilio G. Cota
2018-04-06 2:13 ` [Qemu-devel] [PATCH v2 10/17] translate-all: use per-page locking in !user-mode Emilio G. Cota
2018-04-14 3:29 ` Richard Henderson
2018-04-24 0:18 ` Emilio G. Cota
2018-05-08 16:30 ` Emilio G. Cota
2018-04-24 0:22 ` Emilio G. Cota
2018-04-06 2:13 ` [Qemu-devel] [PATCH v2 11/17] translate-all: add page_locked assertions Emilio G. Cota
2018-04-14 3:31 ` Richard Henderson
2018-04-24 0:27 ` Emilio G. Cota
2018-05-10 21:36 ` Emilio G. Cota
2018-04-06 2:13 ` [Qemu-devel] [PATCH v2 12/17] translate-all: add page_collection assertions Emilio G. Cota
2018-04-14 3:42 ` Richard Henderson
2018-04-24 0:31 ` Emilio G. Cota
2018-05-10 21:37 ` Emilio G. Cota
2018-04-06 2:13 ` [Qemu-devel] [PATCH v2 13/17] translate-all: discard TB when tb_link_page returns an existing matching TB Emilio G. Cota
2018-04-14 3:47 ` Richard Henderson
2018-04-06 2:13 ` [Qemu-devel] [PATCH v2 14/17] translate-all: protect TB jumps with a per-destination-TB lock Emilio G. Cota
2018-04-20 16:13 ` Alex Bennée [this message]
2018-04-06 2:13 ` [Qemu-devel] [PATCH v2 15/17] cputlb: remove tb_lock from tlb_flush functions Emilio G. Cota
2018-04-14 4:05 ` Richard Henderson
2018-04-06 2:13 ` [Qemu-devel] [PATCH v2 16/17] translate-all: remove tb_lock mention from cpu_restore_state_from_tb Emilio G. Cota
2018-04-14 4:05 ` Richard Henderson
2018-04-06 2:13 ` [Qemu-devel] [PATCH v2 17/17] tcg: remove tb_lock Emilio G. Cota
2018-04-06 2:32 ` [Qemu-devel] [PATCH v2 00/17] tcg: tb_lock_removal redux v2 no-reply
2018-04-20 16:17 ` Alex Bennée
2018-04-20 16:50 ` Emilio G. Cota
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=87k1t1amb4.fsf@linaro.org \
--to=alex.bennee@linaro.org \
--cc=cota@braap.org \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=richard.henderson@linaro.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 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).