From: Emilio Cota <cota@braap.org>
To: Richard Henderson <richard.henderson@linaro.org>
Cc: qemu-devel@nongnu.org, peter.maydell@linaro.org
Subject: Re: [PATCH 00/27] tcg: Simplify temporary usage
Date: Wed, 1 Mar 2023 07:11:48 -0500 [thread overview]
Message-ID: <Y/9BBPCJHNCQWTRL@cota-l14> (raw)
In-Reply-To: <abd9ad4a-d2b7-fd3e-1e67-db9d067bd0c0@linaro.org>
On Wed, Feb 15, 2023 at 20:15:37 -1000, Richard Henderson wrote:
> On 2/10/23 02:35, Emilio Cota wrote:
> > I ran yesterday linux-user SPEC06 benchmarks from your tcg-life branch.
> > I do see perf regressions for two workloads (sjeng and xalancbmk).
> > With perf(1) I see liveness_pass* are at 0.00%, so I wonder: is it
> > possible that the emitted code isn't quite the same?
>
> Everything that I checked by hand was the same, but it's possible.
> It's a tedious process. You'd definitely want to turn off ASR.
I've checked with -jitdump and perf whether there was any difference
in the generated code before vs. after for the most common TBs.
They were identical.
Benchmarking without ASR didn't make a difference, unfortunately.
> My current branch has __attribute__((noreturn)) added to all of the liveness
> passes, so that they don't get folded into tcg_gen_code. But I still would
> expect 0%.
I'll bisect the series in the next few days see exactly where
the perf regression begins so that at least we know where to look.
Thanks,
Emilio
next prev parent reply other threads:[~2023-03-01 12:12 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-30 20:59 [PATCH 00/27] tcg: Simplify temporary usage Richard Henderson
2023-01-30 20:59 ` [PATCH] target/arm: Fix physical address resolution for Stage2 Richard Henderson
2023-01-30 21:00 ` Richard Henderson
2023-01-30 20:59 ` [PATCH 01/27] tcg: Adjust TCGContext.temps_in_use check Richard Henderson
2023-01-30 23:30 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 02/27] accel/tcg: Pass max_insn to gen_intermediate_code by pointer Richard Henderson
2023-02-16 7:24 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 03/27] accel/tcg: Use more accurate max_insns for tb_overflow Richard Henderson
2023-01-30 20:59 ` [PATCH 04/27] tcg: Remove branch-to-next regardless of reference count Richard Henderson
2023-01-30 20:59 ` [PATCH 05/27] tcg: Rename TEMP_LOCAL to TEMP_TB Richard Henderson
2023-01-30 23:34 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 06/27] tcg: Add liveness_pass_0 Richard Henderson
2023-01-30 20:59 ` [PATCH 07/27] tcg: Remove TEMP_NORMAL Richard Henderson
2023-01-30 23:36 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 08/27] tcg: Pass TCGTempKind to tcg_temp_new_internal Richard Henderson
2023-01-30 23:36 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 09/27] tcg: Add tcg_temp_ebb_new_{i32,i64,ptr} Richard Henderson
2023-01-30 23:59 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 10/27] tcg: Add tcg_gen_movi_ptr Richard Henderson
2023-02-16 7:15 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 11/27] tcg: Use tcg_temp_ebb_new_* in tcg/ Richard Henderson
2023-01-30 23:59 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 12/27] accel/tcg/plugin: Use tcg_temp_ebb_* Richard Henderson
2023-02-16 7:22 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 13/27] accel/tcg/plugin: Tidy plugin_gen_disable_mem_helpers Richard Henderson
2023-01-30 23:39 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 14/27] tcg: Don't re-use TEMP_TB temporaries Richard Henderson
2023-01-30 20:59 ` [PATCH 15/27] tcg: Change default temp lifetime to TEMP_TB Richard Henderson
2023-01-30 23:52 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 16/27] target/arm: Drop copies in gen_sve_{ldr,str} Richard Henderson
2023-01-30 20:59 ` [PATCH 17/27] target/arm: Don't use tcg_temp_local_new_* Richard Henderson
2023-01-30 23:44 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 18/27] target/cris: Don't use tcg_temp_local_new Richard Henderson
2023-01-30 23:43 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 19/27] target/hexagon: Don't use tcg_temp_local_new_* Richard Henderson
2023-01-30 23:43 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 20/27] target/hppa: Don't use tcg_temp_local_new Richard Henderson
2023-01-30 23:41 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 21/27] target/i386: " Richard Henderson
2023-02-16 7:13 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 22/27] target/mips: " Richard Henderson
2023-01-30 23:42 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 23/27] target/ppc: " Richard Henderson
2023-01-30 23:43 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 24/27] target/xtensa: Don't use tcg_temp_local_new_* Richard Henderson
2023-01-30 23:44 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 25/27] exec/gen-icount: Don't use tcg_temp_local_new_i32 Richard Henderson
2023-01-30 23:45 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 26/27] tcg: Remove tcg_temp_local_new_*, tcg_const_local_* Richard Henderson
2023-01-30 23:45 ` Philippe Mathieu-Daudé
2023-01-30 20:59 ` [PATCH 27/27] tcg: Update docs/devel/tcg-ops.rst for temporary changes Richard Henderson
2023-02-10 12:35 ` [PATCH 00/27] tcg: Simplify temporary usage Emilio Cota
2023-02-16 6:15 ` Richard Henderson
2023-03-01 12:11 ` Emilio Cota [this message]
2023-02-16 6:11 ` 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=Y/9BBPCJHNCQWTRL@cota-l14 \
--to=cota@braap.org \
--cc=peter.maydell@linaro.org \
--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).