From: Kevin O'Connor <kevin@koconnor.net>
To: Richard Henderson <rth@twiddle.net>
Cc: peter.maydell@linaro.org, alex.bennee@linaro.org,
qemu-devel@nongnu.org, aurelien@aurel32.net
Subject: Re: [Qemu-devel] [PATCH v3 05/25] tcg: Allow extra data to be attached to insn_start
Date: Wed, 23 Sep 2015 10:55:49 -0400 [thread overview]
Message-ID: <20150923145549.GA29755@morn.lan> (raw)
In-Reply-To: <1442953507-4074-6-git-send-email-rth@twiddle.net>
On Tue, Sep 22, 2015 at 01:24:47PM -0700, Richard Henderson wrote:
> With an eye toward having this data replace the gen_opc_* arrays
> that each target collects in order to enable restore_state_from_tb.
Hi Richard,
Instead of having each architecture front-end determine the constants
to be restored on an exception, have you considered having the tcg
liveness pass automatically detect them?
What I was thinking was if:
- each front-end stored each constant on every instruction using
regular "movi" ops
- tcg_liveness_analysis() tracked which global memory "sync" writes
are purely due to an op that can raise an exception
- then tcg_liveness_analysis() could remove "movi" instructions with
outputs that are needed only during an exception and place the
constant directly in the compressed table itself.
I'm curious if this was considered and if there is a reason it
wouldn't work well.
-Kevin
next prev parent reply other threads:[~2015-09-23 14:55 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-09-22 20:24 [Qemu-devel] [PATCH v3 00/25] Do away with TB retranslation Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 01/25] tcg: Rename debug_insn_start to insn_start Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 02/25] target-*: Unconditionally emit tcg_gen_insn_start Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 03/25] target-*: Increment num_insns immediately after tcg_gen_insn_start Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 04/25] target-*: Introduce and use cpu_breakpoint_test Richard Henderson
2015-09-23 19:19 ` Peter Maydell
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 05/25] tcg: Allow extra data to be attached to insn_start Richard Henderson
2015-09-23 14:55 ` Kevin O'Connor [this message]
2015-09-23 16:37 ` Richard Henderson
2015-09-23 16:38 ` Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 06/25] target-arm: Add condexec state " Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 07/25] target-i386: Add cc_op " Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 08/25] target-mips: Add delayed branch " Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 09/25] target-s390x: Add cc_op " Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 10/25] target-sh4: Add flags " Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 11/25] target-cris: Mirror gen_opc_pc into insn_start Richard Henderson
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 12/25] target-sparc: Tidy gen_branch_a interface Richard Henderson
2015-09-22 21:23 ` Aurelien Jarno
2015-09-24 19:42 ` Aurelien Jarno
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 13/25] target-sparc: Split out gen_branch_n Richard Henderson
2015-09-24 19:42 ` Aurelien Jarno
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 14/25] target-sparc: Remove gen_opc_jump_pc Richard Henderson
2015-09-24 19:42 ` Aurelien Jarno
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 15/25] target-sparc: Add npc state to insn_start Richard Henderson
2015-09-24 19:42 ` Aurelien Jarno
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 16/25] tcg: Merge cpu_gen_code into tb_gen_code Richard Henderson
2015-09-24 19:48 ` Aurelien Jarno
2015-09-22 20:24 ` [Qemu-devel] [PATCH v3 17/25] target-*: Drop cpu_gen_code define Richard Henderson
2015-09-24 19:49 ` Aurelien Jarno
2015-09-22 20:25 ` [Qemu-devel] [PATCH v3 18/25] tcg: Add TCG_MAX_INSNS Richard Henderson
2015-09-24 20:02 ` Aurelien Jarno
2015-09-24 20:43 ` Richard Henderson
2015-09-22 20:25 ` [Qemu-devel] [PATCH v3 19/25] tcg: Pass data argument to restore_state_to_opc Richard Henderson
2015-09-24 20:11 ` Aurelien Jarno
2015-09-22 20:25 ` [Qemu-devel] [PATCH v3 20/25] tcg: Save insn data and use it in cpu_restore_state_from_tb Richard Henderson
2015-09-23 19:20 ` Peter Maydell
2015-09-25 21:10 ` Aurelien Jarno
2015-09-25 23:05 ` Richard Henderson
2015-09-22 20:25 ` [Qemu-devel] [PATCH v3 21/25] tcg: Remove gen_intermediate_code_pc Richard Henderson
2015-09-25 21:11 ` Aurelien Jarno
2015-09-22 20:25 ` [Qemu-devel] [PATCH v3 22/25] tcg: Remove tcg_gen_code_search_pc Richard Henderson
2015-09-25 21:11 ` Aurelien Jarno
2015-09-22 20:25 ` [Qemu-devel] [PATCH v3 23/25] tcg: Emit prologue to the beginning of code_gen_buffer Richard Henderson
2015-09-23 19:28 ` Peter Maydell
2015-09-23 19:39 ` Richard Henderson
2015-09-22 20:25 ` [Qemu-devel] [PATCH v3 24/25] tcg: Allocate a guard page after code_gen_buffer Richard Henderson
2015-09-23 19:39 ` Peter Maydell
2015-09-23 20:00 ` Richard Henderson
2015-09-23 20:37 ` Peter Maydell
2015-09-23 22:12 ` Richard Henderson
2015-09-22 20:25 ` [Qemu-devel] [PATCH v3 25/25] tcg: Check for overflow via highwater mark Richard Henderson
2015-09-23 19:42 ` Peter Maydell
2015-09-23 20:01 ` Richard Henderson
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=20150923145549.GA29755@morn.lan \
--to=kevin@koconnor.net \
--cc=alex.bennee@linaro.org \
--cc=aurelien@aurel32.net \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
/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).