qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: mttcg@greensocs.com, "QEMU Developers" <qemu-devel@nongnu.org>,
	"KONRAD Frédéric" <fred.konrad@greensocs.com>
Subject: Re: [Qemu-devel] [PATCH 05/10] cpu-exec: elide more icount code if CONFIG_USER_ONLY
Date: Sat, 29 Aug 2015 09:07:58 +0200	[thread overview]
Message-ID: <55E15A4E.6080009@redhat.com> (raw)
In-Reply-To: <CAFEAcA8ob8A+Sa4ytM=xOadQ26ezr6EhzEy=cGkP9U32MSYBZg@mail.gmail.com>



On 28/08/2015 16:56, Peter Maydell wrote:
> What's the rationale for this? Mostly we prefer not to
> add ifdefs in code if we can get away with compiling it for
> both cases, even if the resulting code isn't used.

True.  The rationale is three-fold:

1) It makes sense to abort if TB_EXIT_ICOUNT_EXPIRED is returned from
user-mode emulation TCG code.  Then you need to either leave the
unreachable code after abort() or #ifdef cpu_exec_nocache out.

2) Dually, cpu_exec_nocache's future locking requirements (take tb_lock,
mostly) are not yet respected as of this patch.  Marking the function as
#ifdef CONFIG_SOFTMMU explains why.  Once you do that, you have to
abort() either in cpu_exec_nocache or in the caller.


Reason (2) is stronger for me.  You mentioned a few times your hope that
MTTCG would lead to more correct and more documented thread support in
user-mode emulation, and I think this patch does improve the
documentation slightly.

Paolo

  reply	other threads:[~2015-08-29  7:08 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-12 16:40 [Qemu-devel] [PATCH 00/10] translate-all.c thread-safety Paolo Bonzini
2015-08-12 16:40 ` [Qemu-devel] [PATCH 01/10] cpus: protect work list with work_mutex Paolo Bonzini
2015-08-28 14:33   ` Peter Maydell
2015-08-12 16:40 ` [Qemu-devel] [PATCH 02/10] cpus: remove tcg_halt_cond global variable Paolo Bonzini
2015-08-13 13:05   ` Frederic Konrad
2015-08-13 13:08     ` Paolo Bonzini
2015-08-13 13:19       ` Frederic Konrad
2015-08-28 14:36   ` Peter Maydell
2015-08-29  6:52     ` Paolo Bonzini
2015-08-12 16:40 ` [Qemu-devel] [PATCH 03/10] replace spinlock by QemuMutex Paolo Bonzini
2015-08-13 12:17   ` Frederic Konrad
2015-08-13 13:12     ` Paolo Bonzini
2015-08-13 13:21       ` Frederic Konrad
2015-08-28 14:49   ` Peter Maydell
2015-08-28 14:53     ` Frederic Konrad
2015-08-29  6:51       ` Paolo Bonzini
2015-08-12 16:40 ` [Qemu-devel] [PATCH 04/10] exec-all: remove non-TCG stuff from exec-all.h header Paolo Bonzini
2015-08-28 14:53   ` Peter Maydell
2015-08-29  6:55     ` Paolo Bonzini
2015-08-12 16:40 ` [Qemu-devel] [PATCH 05/10] cpu-exec: elide more icount code if CONFIG_USER_ONLY Paolo Bonzini
2015-08-13 13:14   ` Frederic Konrad
2015-08-13 14:06     ` Paolo Bonzini
2015-08-28 14:56   ` Peter Maydell
2015-08-29  7:07     ` Paolo Bonzini [this message]
2015-08-12 16:40 ` [Qemu-devel] [PATCH 06/10] tcg: code_bitmap is not used by user-mode emulation Paolo Bonzini
2015-08-28 14:57   ` Peter Maydell
2015-08-29  7:13     ` Paolo Bonzini
2015-08-12 16:40 ` [Qemu-devel] [PATCH 07/10] tcg: comment on which functions have to be called with mmap_lock held Paolo Bonzini
2015-08-28 15:33   ` Peter Maydell
2015-08-29  6:57     ` Paolo Bonzini
2015-08-12 16:41 ` [Qemu-devel] [PATCH 08/10] tcg: add memory barriers in page_find_alloc accesses Paolo Bonzini
2015-08-12 20:37   ` Emilio G. Cota
2015-08-13  8:13     ` Paolo Bonzini
2015-08-13 19:50       ` Emilio G. Cota
2015-08-28 15:40   ` Peter Maydell
2015-08-29  6:58     ` Paolo Bonzini
2015-08-12 16:41 ` [Qemu-devel] [PATCH 09/10] exec: make mmap_lock/mmap_unlock globally available Paolo Bonzini
2015-08-28 15:42   ` Peter Maydell
2015-08-12 16:41 ` [Qemu-devel] [PATCH 10/10] cpu-exec: fix lock hierarchy for user-mode emulation Paolo Bonzini
2015-08-28 15:59   ` Peter Maydell
2015-08-12 16:41 ` [Qemu-devel] [PATCH 11/10] tcg: comment on which functions have to be called with tb_lock held Paolo Bonzini
2015-08-13 12:51   ` Frederic Konrad
2015-08-13 12:59     ` Paolo Bonzini
2015-08-13 13:32       ` Frederic Konrad
2015-08-13 14:39         ` Paolo Bonzini
2015-08-13 15:32           ` Peter Maydell
2015-08-13 16:20             ` Paolo Bonzini
2015-08-12 16:41 ` [Qemu-devel] [PATCH 12/10] tcg: protect TBContext with tb_lock Paolo Bonzini
2015-08-13 12:57   ` Frederic Konrad
2015-08-13 13:01     ` Paolo Bonzini
2015-08-13 13:04       ` Frederic Konrad
2015-08-14  9:26 ` [Qemu-devel] [PATCH 00/10] translate-all.c thread-safety Frederic Konrad

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=55E15A4E.6080009@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=fred.konrad@greensocs.com \
    --cc=mttcg@greensocs.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.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).