From: Felipe Balbi <balbi-l0cyMroinI0@public.gmane.org>
To: Alexander Kochetkov <al.kochet-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-omap-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Tony Lindgren <tony-4v6yS6AI5VpBDgjK7y7TUQ@public.gmane.org>,
Wolfram Sang <wsa-z923LK4zBo2bacvFa/9K2g@public.gmane.org>,
Felipe Balbi <balbi-l0cyMroinI0@public.gmane.org>
Subject: Re: [PATCH 2/2] i2c: omap: fix "Too much work in one IRQ" irq handling
Date: Fri, 14 Nov 2014 19:48:59 -0600 [thread overview]
Message-ID: <20141115014859.GC808@saruman> (raw)
In-Reply-To: <1416014452-6712-2-git-send-email-al.kochet-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
[-- Attachment #1: Type: text/plain, Size: 1911 bytes --]
Hi,
On Sat, Nov 15, 2014 at 05:20:52AM +0400, Alexander Kochetkov wrote:
> commit 66b9298878742f08cb6e79b7c7d5632d782fd1e1 (i2c: omap: switch over
> to do {} while loop) changed the interrupt handler to abort transfers
> in case interrupt serviced 100 times but commit's comment states that
> "No functional changes otherwise.".
look at the patch again, the commit you describe is *not* the one giving
up on servicing interrupts after 100 times. That commit, really, *only*
switched over from while() {} to do {} while(), the only functional
change there is that the while loop will always execute at least once.
> Also, original commit could report status 0 (no error) on aborted transfers.
how ? This is an interesting bug which deserves further explanation.
> The patch restore original logic. In case interrupt serviced 100 times just
> quit interrupt handler and don't abort active transfer.
>
> Signed-off-by: Alexander Kochetkov <al.kochet-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
> ---
> drivers/i2c/busses/i2c-omap.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/i2c/busses/i2c-omap.c b/drivers/i2c/busses/i2c-omap.c
> index 9af7095..34b9011 100644
> --- a/drivers/i2c/busses/i2c-omap.c
> +++ b/drivers/i2c/busses/i2c-omap.c
> @@ -920,7 +920,7 @@ omap_i2c_isr_thread(int this_irq, void *dev_id)
> dev_dbg(dev->dev, "IRQ (ISR = 0x%04x)\n", stat);
> if (count++ == 100) {
> dev_warn(dev->dev, "Too much work in one IRQ\n");
> - break;
> + goto out;
quite frankly, this looks *very* wrong. It creates the possibility for
us never completing a command which would cause several timeouts.
How have you tested this and how have you figured this was the actual
bug ? Based on commit log not matching the patch body (which 'original
logic' are you talking about ?), I have to NAK this patch.
--
balbi
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
next prev parent reply other threads:[~2014-11-15 1:48 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-15 1:20 [PATCH 1/2] i2c: omap: fix NACK and Arbitration Lost irq handling Alexander Kochetkov
[not found] ` <1416014452-6712-1-git-send-email-al.kochet-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-11-15 1:20 ` [PATCH 2/2] i2c: omap: fix "Too much work in one IRQ" " Alexander Kochetkov
[not found] ` <1416014452-6712-2-git-send-email-al.kochet-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-11-15 1:48 ` Felipe Balbi [this message]
2014-11-15 2:37 ` Alexander Kochetkov
2014-11-15 3:47 ` Felipe Balbi
2014-11-15 3:53 ` Felipe Balbi
2014-11-15 5:42 ` Alexander Kochetkov
2014-11-16 15:45 ` Felipe Balbi
2014-11-17 14:41 ` Wolfram Sang
2014-11-18 16:00 ` Felipe Balbi
2014-11-18 16:12 ` Wolfram Sang
2014-11-20 16:38 ` Alexander Kochetkov
[not found] ` <2DED62C3-7C54-49E0-A39B-F68D5DAC66B1-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-11-20 16:43 ` Wolfram Sang
2014-11-18 16:31 ` Alexander Kochetkov
[not found] ` <5D39428D-F359-4F04-8ACC-D607011B88B9-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-11-18 16:38 ` Felipe Balbi
2014-11-15 1:43 ` [PATCH 1/2] i2c: omap: fix NACK and Arbitration Lost " Felipe Balbi
[not found] ` <2159E044-9130-410D-905B-B941408DCDCD@gmail.com>
[not found] ` <2159E044-9130-410D-905B-B941408DCDCD-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-11-15 2:48 ` Alexander Kochetkov
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=20141115014859.GC808@saruman \
--to=balbi-l0cymroini0@public.gmane.org \
--cc=al.kochet-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-omap-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=tony-4v6yS6AI5VpBDgjK7y7TUQ@public.gmane.org \
--cc=wsa-z923LK4zBo2bacvFa/9K2g@public.gmane.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).