qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: Pierrick Bouvier <pierrick.bouvier@linaro.org>, qemu-devel@nongnu.org
Cc: alex.bennee@linaro.org
Subject: Re: [PATCH 07/22] plugins: Use emit_before_op for PLUGIN_GEN_AFTER_INSN
Date: Tue, 19 Mar 2024 09:56:33 -1000	[thread overview]
Message-ID: <1eb36792-22e4-4f0c-b7b5-968b78e3a1bc@linaro.org> (raw)
In-Reply-To: <5b9ff51d-439f-4e65-acb5-909bdfaf350c@linaro.org>

On 3/19/24 03:32, Pierrick Bouvier wrote:
>>   static void plugin_gen_inject(struct qemu_plugin_tb *plugin_tb)
>>   {
>> -    TCGOp *op;
>> +    TCGOp *op, *next;
>>       int insn_idx = -1;
>>       pr_ops();
>> -    QTAILQ_FOREACH(op, &tcg_ctx->ops, link) {
>> +    /*
>> +     * While injecting code, we cannot afford to reuse any ebb temps
>> +     * that might be live within the existing opcode stream.
>> +     * The simplest solution is to release them all and create new.
>> +     */
>> +    memset(tcg_ctx->free_temps, 0, sizeof(tcg_ctx->free_temps));
>> +
> 
> Not an expert at this, but wouldn't that break an existing TB that already has some ops on 
> those temps?

No, this only affects allocation of new temps -- if free_temps is empty, a new temp will 
be allocated from tcg_ctx->nb_temps++.

Zeroing free_temps here ensures that we *do not* reuse a temp that might already be live 
across any plugin insertion point.  Between insertion points, we will free plugin temps 
and only reuse those.


r~


  reply	other threads:[~2024-03-19 19:56 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-16  1:56 [PATCH 00/22] plugins: Rewrite plugin code generation Richard Henderson
2024-03-16  1:56 ` [PATCH 01/22] tcg: Add TCGContext.emit_before_op Richard Henderson
2024-03-19 10:55   ` Pierrick Bouvier
2024-03-19 14:04   ` Alex Bennée
2024-03-19 21:23     ` Richard Henderson
2024-03-16  1:57 ` [PATCH 02/22] tcg: Make tcg/helper-info.h self-contained Richard Henderson
2024-03-18 10:01   ` Alex Bennée
2024-03-18 10:03   ` Alex Bennée
2024-03-16  1:57 ` [PATCH 03/22] tcg: Pass function pointer to tcg_gen_call* Richard Henderson
2024-03-18 10:02   ` Alex Bennée
2024-03-16  1:57 ` [PATCH 04/22] plugins: Zero new qemu_plugin_dyn_cb entries Richard Henderson
2024-03-18 10:03   ` Alex Bennée
2024-03-16  1:57 ` [PATCH 05/22] plugins: Move function pointer in qemu_plugin_dyn_cb Richard Henderson
2024-03-18 10:04   ` Alex Bennée
2024-03-19 13:18   ` Pierrick Bouvier
2024-03-19 21:30     ` Richard Henderson
2024-03-20  5:31       ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 06/22] plugins: Create TCGHelperInfo for all out-of-line callbacks Richard Henderson
2024-03-19 13:12   ` Pierrick Bouvier
2024-03-19 19:51     ` Richard Henderson
2024-03-20  5:22       ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 07/22] plugins: Use emit_before_op for PLUGIN_GEN_AFTER_INSN Richard Henderson
2024-03-19 13:32   ` Pierrick Bouvier
2024-03-19 19:56     ` Richard Henderson [this message]
2024-03-20  5:36       ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 08/22] plugins: Use emit_before_op for PLUGIN_GEN_FROM_TB Richard Henderson
2024-03-19 13:22   ` Pierrick Bouvier
2024-03-19 19:57     ` Richard Henderson
2024-03-16  1:57 ` [PATCH 09/22] plugins: Add PLUGIN_GEN_AFTER_TB Richard Henderson
2024-03-19 13:33   ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 10/22] plugins: Use emit_before_op for PLUGIN_GEN_FROM_INSN Richard Henderson
2024-03-19 13:34   ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 11/22] plugins: Use emit_before_op for PLUGIN_GEN_FROM_MEM Richard Henderson
2024-03-19 13:35   ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 12/22] plugins: Remove plugin helpers Richard Henderson
2024-03-18 16:38   ` Alex Bennée
2024-03-16  1:57 ` [PATCH 13/22] tcg: Remove TCG_CALL_PLUGIN Richard Henderson
2024-03-19 13:06   ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 14/22] tcg: Remove INDEX_op_plugin_cb_{start,end} Richard Henderson
2024-03-19 13:04   ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 15/22] plugins: Simplify callback queues Richard Henderson
2024-03-19 13:28   ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 16/22] plugins: Introduce PLUGIN_CB_MEM_REGULAR Richard Henderson
2024-03-19 13:28   ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 17/22] plugins: Replace pr_ops with a proper debug dump flag Richard Henderson
2024-03-19 12:59   ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 18/22] plugins: Split out common cb expanders Richard Henderson
2024-03-19 13:29   ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 19/22] plugins: Merge qemu_plugin_tb_insn_get to plugin-gen.c Richard Henderson
2024-03-19 13:30   ` Pierrick Bouvier
2024-03-16  1:57 ` [PATCH 20/22] plugins: Move qemu_plugin_insn_cleanup_fn to tcg.c Richard Henderson
2024-03-18 17:44   ` Alex Bennée
2024-03-16  1:57 ` [PATCH 21/22] plugins: Inline plugin_gen_empty_callback Richard Henderson
2024-03-18 19:09   ` Alex Bennée
2024-03-16  1:57 ` [PATCH 22/22] plugins: Update the documentation block for plugin-gen.c Richard Henderson
2024-03-19 12:56   ` Pierrick Bouvier
2024-03-19 13:38 ` [PATCH 00/22] plugins: Rewrite plugin code generation Pierrick Bouvier

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=1eb36792-22e4-4f0c-b7b5-968b78e3a1bc@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=alex.bennee@linaro.org \
    --cc=pierrick.bouvier@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).