Date:   Fri, 7 Feb 2020 09:41:21 +0100
From:   Miquel Raynal <mraynal@...nel.org>
To:     liaoweixiong <liaoweixiong@...winnertech.com>
Cc:     Kees Cook <keescook@...omium.org>,
        Anton Vorontsov <anton@...msg.org>,
        Colin Cross <ccross@...roid.com>,
        Tony Luck <tony.luck@...el.com>,
        Jonathan Corbet <corbet@....net>,
        Richard Weinberger <richard@....at>,
        Vignesh Raghavendra <vigneshr@...com>,
        Mauro Carvalho Chehab <mchehab+samsung@...nel.org>,
        "David S. Miller" <davem@...emloft.net>,
        Rob Herring <robh@...nel.org>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Jonathan Cameron <Jonathan.Cameron@...wei.com>,
        linux-doc@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-mtd@...ts.infradead.org
Subject: Re: [PATCH v1 11/11] mtd: new support oops logger based on
 pstore/blk

Hi Liao,

liaoweixiong <liaoweixiong@...winnertech.com> wrote on Fri, 7 Feb 2020
12:13:08 +0800:

> hi Miquel Raynal,
> 
> On 2020/2/6 PM 11:45, Miquel Raynal wrote:
> > Hi liao,
> > 
> > liaoweixiong <liaoweixiong@...winnertech.com> wrote on Thu, 6 Feb 2020
> > 21:10:47 +0800:
> >   
> >> hi Miquel Raynal,
> >>
> >> On 2020/1/23 AM 1:41, Miquel Raynal wrote:  
> >>> Hello,
> >>>  
> >>>    >>>>>>>> +/*  
> >>>>>>>> + * All zones will be read as pstore/blk will read zone one by one when do
> >>>>>>>> + * recover.
> >>>>>>>> + */
> >>>>>>>> +static ssize_t mtdpstore_read(char *buf, size_t size, loff_t off)
> >>>>>>>> +{
> >>>>>>>> +	struct mtdpstore_context *cxt = &oops_cxt;
> >>>>>>>> +	size_t retlen;
> >>>>>>>> +	int ret;
> >>>>>>>> +
> >>>>>>>> +	if (mtdpstore_block_isbad(cxt, off))
> >>>>>>>> +		return -ENEXT;
> >>>>>>>> +
> >>>>>>>> +	pr_debug("try to read off 0x%llx size %zu\n", off, size);
> >>>>>>>> +	ret = mtd_read(cxt->mtd, off, size, &retlen, (u_char *)buf);
> >>>>>>>> +	if ((ret < 0 && !mtd_is_bitflip(ret)) || size != retlen)  {  
> >>>>>>>
> >>>>>>> IIRC size != retlen does not mean it failed, but that you should
> >>>>>>> continue reading after retlen bytes, no?  
> >>>>>>>      >>  
> >>>>>> Yes, you are right. I will fix it. Thanks.  
> >>>>>>    >>>>> Also, mtd_is_bitflip() does not mean that you are reading a false  
> >>>>>>> buffer, but that the data has been corrected as it contained bitflips.
> >>>>>>> mtd_is_eccerr() however, would be meaningful.  
> >>>>>>>      >>  
> >>>>>> Sure I know mtd_is_bitflip() does not mean failure, but I do not think
> >>>>>> mtd_is_eccerr() should be here since the codes are ret < 0 and NOT
> >>>>>> mtd_is_bitflip().  
> >>>>>
> >>>>> Yes, just drop this check, only keep ret < 0.  
> >>>>>     >>  
> >>>> If I don't get it wrong, it should not	 be dropped here. Like your words,
> >>>> "mtd_is_bitflip() does not mean that you are reading a false buffer,
> >>>> but that the data has been corrected as it contained bitflips.", the
> >>>> data I get are valid even if mtd_is_bitflip() return true. It's correct
> >>>> data and it's no need to go to handle error. To me, the codes
> >>>> should be:
> >>>> 	if (ret < 0 && !mit_is_bitflip())
> >>>> 		[error handling]  
> >>>
> >>> Please check the implementation of mtd_is_bitflip(). You'll probably
> >>> figure out what I am saying.
> >>>
> >>> https://elixir.bootlin.com/linux/latest/source/include/linux/mtd/mtd.h#L585  
> >>>    >>  
> >> How about the codes as follows:
> >>
> >> for (done = 0, retlen = 0; done < size; done += retlen) {
> >> 	ret = mtd_read(..., &retlen, ...);
> >> 	if (!ret)
> >> 		continue;
> >> 	/*
> >> 	 * do nothing if bitflip and ecc error occurs because whether
> >> 	 * it's bitflip or ECC error, just a small number of bits flip
> >> 	 * and the impact on log data is so small. The mtdpstore just
> >> 	 * hands over what it gets and user can judge whether the data
> >> 	 * is valid or not.
> >> 	 */
> >> 	if (mtd_is_bitflip(ret)) {
> >> 		dev_warn("bitflip at....");
> >> 		continue;  
> 
> > I don't understand why do you check for bitflips. Bitflips have been
> > corrected at this stage, you just get the information that there
> > has been bitflips, but the data integrity is fine.
> >   
> 
> Both of bitflip and eccerror are not real wrong in this
> case. So we must check them.
> 
> > I am not against ignoring ECC errors in this case though. I would
> > propose:
> > 
> > 	for (...) {
> > 		if (ret < 0) {
> > 			complain;
> > 			return;
> > 		}
> >   
> 
> -117 (-EUCLEAN) means bitflip but be corrected.
> -74 (-EBADMSG) means ecc error that uncorrectable
> All of them are negative number that smaller than 0. If it just keeps
> "ret < 0", it can never make a difference between bitflip/eccerror
> and others.

I forgot about these, your're right, so what about:

	static bool mtdpstore_is_io_error(int ret)
	{
		return ret < 0 && !mtd_is_eccerr(ret)> && !mtd_is_bitflip(ret);
	}

	for (...) {
		if (mtdpstore_is_io_error(ret))
			return ret;

		/*
		 * Comment explaining why we still return valid data
		 * despite ECC errors.
		 */
		if (mtd_is_eccerr(ret))
			just-complain();
	}

This snippet makes a difference between any "controller/bus
timeout/error : data could not be retrieved" and "ECC error, maybe we
can still read it and try to understand (risky, must be warned)".

> 
> > 		if (mtd_is_eccerr())
> > 			complain;
> > 	}
> > 		  
> >> 	} else if (mtd_is_eccerr(ret)) {
> >> 		dev_warn("eccerr at....");
> >> 		retlen = retlen == 0 ? size : retlen;
> >> 		continue;
> >> 	} else {
> >> 		dev_err("read failure at...");
> >> 		/* this zone is broken, try next one */
> >> 		return -ENEXT;
> >> 	}
> >> }
> >>  
> > 
> > 
> > Thanks,
> > Miquèl
> >   

Thanks,
Miquèl

Powered by blists - more mailing lists