From: "Alex Bennée" <alex.bennee@linaro.org>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: sergey.fedorov@linaro.org,
Peter Crosthwaite <crosthwaite.peter@gmail.com>,
Stefan Weil <sw@weilnetz.de>,
Claudio Fontana <claudio.fontana@huawei.com>,
qemu-devel@nongnu.org, Alexander Graf <agraf@suse.de>,
Blue Swirl <blauwirbel@gmail.com>,
qemu-arm@nongnu.org, "Vassili Karpov (malc)" <av1474@comtv.ru>,
Sergey Fedorov <serge.fdrv@gmail.com>,
Aurelien Jarno <aurelien@aurel32.net>,
Richard Henderson <rth@twiddle.net>
Subject: Re: [Qemu-devel] [PATCH 1/8] tcg: Clean up direct block chaining data fields
Date: Thu, 24 Mar 2016 15:23:11 +0000 [thread overview]
Message-ID: <87io0brjc0.fsf@linaro.org> (raw)
In-Reply-To: <56F4039A.5050907@redhat.com>
Paolo Bonzini <pbonzini@redhat.com> writes:
> On 24/03/2016 16:01, Alex Bennée wrote:
>>>> >> OK I found that tricky to follow. Where does the value of the pointer
>>>> >> come from that sets these bottom bits? The TB jumping to this TB sets it?
>>
>> Where I get confused it what is the point of jmp_list_first? If these
>> are two circular lists do we care which the first in the list is? The
>> exit condition when coming out of searching seems when ntb with index =
>> orig tb with index.
>
> Say you have a list for blocks that jump to TB. The next pointer is in
> jmp_list_next[0] for blocks whose first jump is to TB. It is in
> jmp_list_next[1] for blocks whose second jump is to TB.
>
> However, because it is a circular list, you also need TB itself to be a
> part of the list. For TB, the next pointer is in jmp_list_first.
>
> Because TB probably doesn't jump to itself, the first link of the list
> of blocks that jumps to TB is not in jmp_list_next[]. Thus QEMU places
> it in tb->jmp_list_first.
>
> Say you have three tbs. TB1's first jump and TB2's second jump lead to
> TB0. Then the list starting at tb0->jmp_list_first goes like this:
>
> tb0->jmp_list_first = tb1 | 0;
> .--------------------' |
> | .--------'
> tb1->jmp_list_next[0] = tb2 | 1;
> .--------------------' |
> | .---------'
> tb2->jmp_list_next[1] = tb0 | 2;
>
> There is also a case where a TB jumps to itself; it then appears twice
> in the list with different values in the low bits, such as this:
>
> tb->jmp_list_first = tb | 0;
> .--------------------' |
> | .-------'
> tb->jmp_list_next[0] = tb | 2;
>
> Other blocks jumping to TB would appear in the same list, too, either
> before or after the tb|0 link.
Right I follow now. Extra ascii art always helps ;-)
--
Alex Bennée
next prev parent reply other threads:[~2016-03-24 15:23 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-24 10:39 [Qemu-devel] [PATCH 0/8] tcg: Direct block chaining clean-up sergey.fedorov
2016-03-24 10:39 ` [Qemu-devel] [PATCH 1/8] tcg: Clean up direct block chaining data fields sergey.fedorov
2016-03-24 13:42 ` Alex Bennée
2016-03-24 14:02 ` Sergey Fedorov
2016-03-24 15:01 ` Alex Bennée
2016-03-24 15:10 ` Sergey Fedorov
2016-03-24 15:11 ` Paolo Bonzini
2016-03-24 15:23 ` Alex Bennée [this message]
2016-03-28 22:12 ` Richard Henderson
2016-03-29 8:14 ` Paolo Bonzini
2016-03-29 8:51 ` Paolo Bonzini
2016-03-29 8:31 ` Sergey Fedorov
2016-03-29 15:37 ` Richard Henderson
2016-03-29 16:26 ` [Qemu-devel] [Qemu-arm] " Peter Maydell
2016-03-29 17:58 ` Sergey Fedorov
2016-03-24 10:39 ` [Qemu-devel] [PATCH 2/8] tcg: Use uintptr_t type for jmp_list_{next|first} fields of TB sergey.fedorov
2016-03-24 14:17 ` Sergey Fedorov
2016-03-24 14:58 ` Alex Bennée
2016-03-24 15:15 ` Sergey Fedorov
2016-03-24 10:39 ` [Qemu-devel] [PATCH 3/8] tcg: Rearrange tb_link_page() to avoid forward declaration sergey.fedorov
2016-03-24 15:04 ` Alex Bennée
2016-03-24 10:39 ` [Qemu-devel] [PATCH 4/8] tcg: Init TB's direct jumps before making it visible sergey.fedorov
2016-03-24 15:11 ` Alex Bennée
2016-03-24 15:31 ` Sergey Fedorov
2016-03-24 15:40 ` Alex Bennée
2016-03-24 15:58 ` Sergey Fedorov
2016-03-24 10:39 ` [Qemu-devel] [PATCH 5/8] tcg: Clarify "thread safaty" check in tb_add_jump() sergey.fedorov
2016-03-24 11:31 ` Paolo Bonzini
2016-03-24 12:41 ` Sergey Fedorov
2016-03-24 12:23 ` Artyom Tarasenko
2016-03-24 12:28 ` Sergey Fedorov
2016-03-24 10:39 ` [Qemu-devel] [PATCH 6/8] tcg: Rename tb_jmp_remove() to tb_remove_from_jmp_list() sergey.fedorov
2016-03-24 15:24 ` Alex Bennée
2016-03-24 10:39 ` [Qemu-devel] [PATCH 7/8] tcg: Extract removing of jumps to TB from tb_phys_invalidate() sergey.fedorov
2016-03-24 15:26 ` Alex Bennée
2016-03-24 10:39 ` [Qemu-devel] [PATCH 8/8] tcg: Clean up tb_jmp_unlink() sergey.fedorov
2016-03-24 15:36 ` Alex Bennée
2016-03-24 15:42 ` Sergey Fedorov
2016-03-24 15:52 ` Sergey Fedorov
2016-03-24 11:33 ` [Qemu-devel] [PATCH 0/8] tcg: Direct block chaining clean-up Paolo Bonzini
2016-03-24 12:21 ` Alex Bennée
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=87io0brjc0.fsf@linaro.org \
--to=alex.bennee@linaro.org \
--cc=agraf@suse.de \
--cc=aurelien@aurel32.net \
--cc=av1474@comtv.ru \
--cc=blauwirbel@gmail.com \
--cc=claudio.fontana@huawei.com \
--cc=crosthwaite.peter@gmail.com \
--cc=pbonzini@redhat.com \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
--cc=serge.fdrv@gmail.com \
--cc=sergey.fedorov@linaro.org \
--cc=sw@weilnetz.de \
/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).