From: Dan Carpenter <dan.carpenter@linaro.org>
To: swboyd@chromium.org
Cc: linux-arm-msm@vger.kernel.org
Subject: [bug report] soc: qcom: rpmh-rsc: Sleep waiting for tcs slots to be free
Date: Wed, 8 May 2024 17:49:34 +0300 [thread overview]
Message-ID: <911181ed-c430-4592-ad26-4dc948834e08@moroto.mountain> (raw)
Hello Stephen Boyd,
Commit 2bc20f3c8487 ("soc: qcom: rpmh-rsc: Sleep waiting for tcs
slots to be free") from Jul 24, 2020 (linux-next), leads to the
following Smatch static checker warning:
drivers/soc/qcom/rpmh-rsc.c:658 rpmh_rsc_send_data()
warn: mixing irqsave and irq
drivers/soc/qcom/rpmh-rsc.c
645 int rpmh_rsc_send_data(struct rsc_drv *drv, const struct tcs_request *msg)
646 {
647 struct tcs_group *tcs;
648 int tcs_id;
649 unsigned long flags;
650
651 tcs = get_tcs_for_msg(drv, msg);
652 if (IS_ERR(tcs))
653 return PTR_ERR(tcs);
654
655 spin_lock_irqsave(&drv->lock, flags);
flags saves if this is called with IRQs disabled. I don't think it is.
656
657 /* Wait forever for a free tcs. It better be there eventually! */
--> 658 wait_event_lock_irq(drv->tcs_wait,
659 (tcs_id = claim_tcs_for_req(drv, tcs, msg)) >= 0,
660 drv->lock);
This will enable IRQs and then disable them again. If this were called
with IRQs disabled then this would probably be bad. (But again, I don't
think it is).
661
662 tcs->req[tcs_id - tcs->offset] = msg;
663 set_bit(tcs_id, drv->tcs_in_use);
664 if (msg->state == RPMH_ACTIVE_ONLY_STATE && tcs->type != ACTIVE_TCS) {
665 /*
666 * Clear previously programmed WAKE commands in selected
667 * repurposed TCS to avoid triggering them. tcs->slots will be
668 * cleaned from rpmh_flush() by invoking rpmh_rsc_invalidate()
669 */
670 write_tcs_reg_sync(drv, drv->regs[RSC_DRV_CMD_ENABLE], tcs_id, 0);
671 enable_tcs_irq(drv, tcs_id, true);
672 }
673 spin_unlock_irqrestore(&drv->lock, flags);
And then it sets it back to whatever it was when it was called. So
that's fine.
674
675 /*
676 * These two can be done after the lock is released because:
677 * - We marked "tcs_in_use" under lock.
678 * - Once "tcs_in_use" has been marked nobody else could be writing
679 * to these registers until the interrupt goes off.
680 * - The interrupt can't go off until we trigger w/ the last line
681 * of __tcs_set_trigger() below.
682 */
683 __tcs_buffer_write(drv, tcs_id, 0, msg);
684 __tcs_set_trigger(drv, tcs_id, true);
685
686 return 0;
687 }
regards,
dan carpenter
next reply other threads:[~2024-05-08 14:49 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-08 14:49 Dan Carpenter [this message]
2024-05-08 21:01 ` [bug report] soc: qcom: rpmh-rsc: Sleep waiting for tcs slots to be free Stephen Boyd
2024-05-09 6:41 ` Dan Carpenter
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=911181ed-c430-4592-ad26-4dc948834e08@moroto.mountain \
--to=dan.carpenter@linaro.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=swboyd@chromium.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.