From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail.bootlin.com ([62.4.15.54]) by bombadil.infradead.org with esmtp (Exim 4.90_1 #2 (Red Hat Linux)) id 1f6W3s-0005S2-94 for linux-mtd@lists.infradead.org; Thu, 12 Apr 2018 06:50:36 +0000 Date: Thu, 12 Apr 2018 08:49:19 +0200 From: Miquel Raynal To: Abhishek Sahu Cc: Boris Brezillon , Archit Taneja , Richard Weinberger , linux-arm-msm@vger.kernel.org, linux-kernel@vger.kernel.org, Marek Vasut , linux-mtd@lists.infradead.org, Cyrille Pitchen , Andy Gross , Brian Norris , David Woodhouse Subject: Re: [PATCH 3/9] mtd: nand: qcom: erased page detection for uncorrectable errors only Message-ID: <20180412084919.1ca7991d@xps13> In-Reply-To: <2c93157a2982365ceaa8af17d5e3b97a@codeaurora.org> References: <1522845745-6624-1-git-send-email-absahu@codeaurora.org> <1522845745-6624-4-git-send-email-absahu@codeaurora.org> <20180410105945.65f2cade@xps13> <2c93157a2982365ceaa8af17d5e3b97a@codeaurora.org> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Hi Abhishek, On Thu, 12 Apr 2018 12:03:58 +0530, Abhishek Sahu wrote: > On 2018-04-10 14:29, Miquel Raynal wrote: > > Hi Abhishek, =20 > > > On Wed, 4 Apr 2018 18:12:19 +0530, Abhishek Sahu =20 > > wrote: =20 > > >> The NAND flash controller generates ECC uncorrectable error =20 > >> first in case of completely erased page. Currently driver > >> applies the erased page detection logic for other operation > >> errors also so fix this and return EIO for other operational > >> errors. > > > I am sorry I don't understand very well what is the purpose of this = =20 > > patch, could you please explain it again? =20 > > > Do you mean that you want to avoid having rising ECC errors when you = =20 > > read erased pages? > > Thanks Miquel for your review. =20 >=20 > QCOM NAND flash controller has in built erased page > detection HW. > Following is the flow in the HW if controller tries > to read erased page >=20 > 1. First ECC uncorrectable error will be generated from > ECC engine since ECC engine first calculates the ECC with > all 0xff and match the calculated ECC with ECC code in OOB > (which is again all 0xff). > 2. After getting ECC error, erased CW detection HW checks if > all the bytes in page are 0xff and then it updates the > status in separate register NAND_ERASED_CW_DETECT_STATUS >=20 > So the erased CW detect status should be checked only if > ECC engine generated the uncorrectable error. >=20 > Currently for all other operational errors also (like TIMEOUT, > MPU errors etc), the erased CW detect register was being > checked. This is very clear, thanks. I don't know very much this controller so I think you can add this information in the commit message for future reference. >=20 > >> >> Signed-off-by: Abhishek Sahu =20 > >> --- > >> drivers/mtd/nand/qcom_nandc.c | 8 +++++++- > >> 1 file changed, 7 insertions(+), 1 deletion(-) =20 > >> >> diff --git a/drivers/mtd/nand/qcom_nandc.c >> b/drivers/mtd/nand/qc= om_nandc.c =20 > >> index 17321fc..57c16a6 100644 > >> --- a/drivers/mtd/nand/qcom_nandc.c > >> +++ b/drivers/mtd/nand/qcom_nandc.c > >> @@ -1578,6 +1578,7 @@ static int parse_read_errors(struct >> qcom_nand= _host *host, u8 *data_buf, > >> struct nand_ecc_ctrl *ecc =3D &chip->ecc; > >> unsigned int max_bitflips =3D 0; > >> struct read_stats *buf; > >> + bool flash_op_err =3D false; > >> int i; =20 > >> >> buf =3D (struct read_stats *)nandc->reg_read_buf; =20 > >> @@ -1599,7 +1600,7 @@ static int parse_read_errors(struct >> qcom_nand= _host *host, u8 *data_buf, > >> buffer =3D le32_to_cpu(buf->buffer); > >> erased_cw =3D le32_to_cpu(buf->erased_cw); =20 > >> >> - if (flash & (FS_OP_ERR | FS_MPU_ERR)) { =20 > >> + if ((flash & FS_OP_ERR) && (buffer & BS_UNCORRECTABLE_BIT)) { > > > And later you have another "if (buffer & BS_UNCORRECTABLE_BIT)" which= =20 > > is then redundant, unless that is not what you actually want to do? =20 >=20 > Yes. That check seems to be redundant. I will fix that. >=20 > > > Maybe you can add comments before the if ()/ else if () to explain in= =20 > > which case you enter each branch. =20 >=20 > Sure. That would be better. Will add the same in next patch set. >=20 > > >> bool erased; =20 > >> >> /* ignore erased codeword errors */ =20 > >> @@ -1641,6 +1642,8 @@ static int parse_read_errors(struct >> qcom_nand= _host *host, u8 *data_buf, > >> max_t(unsigned int, max_bitflips, ret); > >> } > >> } > >> + } else if (flash & (FS_OP_ERR | FS_MPU_ERR)) { > >> + flash_op_err =3D true; > >> } else { > >> unsigned int stat; =20 > >> >> @@ -1654,6 +1657,9 @@ static int parse_read_errors(struct >> qcom_n= and_host *host, u8 *data_buf, =20 > >> oob_buf +=3D oob_len + ecc->bytes; > >> } =20 > >> >> + if (flash_op_err) =20 > >> + return -EIO; > >> + > > > In you are propagating an error related to the controller, this is =20 > > fine, but I think you just want to raise the fact that a NAND > > uncorrectable error occurred, in this case you should just increment > > mtd->ecc_stats.failed and return 0 (returning max_bitflips here would >= be > > fine too has it would be 0 too). =20 >=20 > The flash_op_err will be for other operational errors only (like timeo= ut, > MPU error, device failure etc). For correctable errors, >=20 > ret =3D nand_check_erased_ecc_chunk(data_buf, > data_len, eccbuf, ecclen, oob_buf, > extraooblen, ecc->strength); Why do you need nand_check_erased_ecc_chunk() if the blank page check is done in hw? Thanks, Miqu=C3=A8l > if (ret < 0) { > mtd->ecc_stats.failed++; > } else { > mtd->ecc_stats.corrected +=3D ret; >=20 > Already, it is incrementing mtd->ecc_stats.failed >=20 > Thanks, > Abhishek --=20 Miquel Raynal, Bootlin (formerly Free Electrons) Embedded Linux and Kernel engineering https://bootlin.com