From: Wake Liu via ltp <ltp@lists.linux.it>
To: Cyril Hrubis <chrubis@suse.cz>
Cc: Wake Liu <wakel@google.com>, ltp@lists.linux.it
Subject: Re: [LTP] [PATCH] sys_rtc01: Rewrite the RTC check
Date: Thu, 19 Nov 2026 15:06:19 +0000 [thread overview]
Message-ID: <20261119150619.449861-1-wakel@google.com> (raw)
In-Reply-To: <20261119103711.3002153-1-chrubis@suse.cz>
Hi Cyril,
On Thu, Nov 19, 2026 at 11:37:11AM +0100, Cyril Hrubis wrote:
> Instead of comparing system time against RTC, which proved to be
> unreliable we now sample the RTC since_epoch and compare it with
> CLOCK_MONOTONIC_RAW timer.
Thanks for reworking this! Dropping the system time comparison and timing
since_epoch transitions against CLOCK_MONOTONIC_RAW is much cleaner and
avoids all the STA_UNSYNC / CONFIG_RTC_SYSTOHC corner cases on our
systems.
Two small things I noticed while reviewing and testing the patch:
> +static void poll_rtc(struct rtc_dev *rtc)
> +{
> + struct timespec start, end;
> + struct rtc_edge edge;
> + long long val;
> +
> + SAFE_CLOCK_GETTIME(CLOCK_MONOTONIC_RAW, &start);
> + val = TST_SYSFS_READ_LI(PATH_CLASS_RTC "/%s/since_epoch", rtc->name);
> + SAFE_CLOCK_GETTIME(CLOCK_MONOTONIC_RAW, &end);
The kernel exports since_epoch as time64_t ("%lld") and val is already
long long, so this should use TST_SYSFS_READ_LLI() (and likewise in
read_rtc_snapshot() with a long long since_epoch). Otherwise on 32-bit
userspace an uninitialized or secondary RTC set past 2038 clamps "%ld"
to LONG_MAX on every read, making poll_rtc() see 0 transitions (and
mismatching 64-bit timegm() when _TIME_BITS=64).
> static struct tst_test test = {
> + .setup = setup,
> .test_all = do_test,
> + .runtime = 5,
> };
Since the test requires at least 2 transitions, could we also set
.min_runtime = 3 (or 4)? Without it, running with e.g.
LTP_RUNTIME_MUL=0.2 clamps runtime to 1s and fails with:
sys_rtc01.c:261: TFAIL: rtc0: since_epoch did not advance (1 transitions seen)
With those two fixed, feel free to add:
Reviewed-by: Wake Liu <wakel@google.com>
Best regards,
Wake
--
Mailing list info: https://lists.linux.it/listinfo/ltp
next prev parent reply other threads:[~2026-11-19 15:06 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-19 10:37 [LTP] [PATCH] sys_rtc01: Rewrite the RTC check Cyril Hrubis
2026-11-19 12:13 ` [LTP] " linuxtestproject.agent
2026-11-19 15:06 ` Wake Liu via ltp [this message]
2026-11-20 9:39 ` [LTP] [PATCH] " Cyril Hrubis
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=20261119150619.449861-1-wakel@google.com \
--to=ltp@lists.linux.it \
--cc=chrubis@suse.cz \
--cc=wakel@google.com \
/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.