linux-kernel - [PATCH] mmc: core: apply SD quirks earlier during probe

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20240918145133.18480-1-obilaniu@gmail.com>
Date: Wed, 18 Sep 2024 10:51:33 -0400
From: Olexa Bilaniuk <obilaniu@...il.com>
To: keita.aihara@...y.com
Cc: Masaya.Takahashi@...y.com,
	Shingo.Takeuchi@...y.com,
	Tim.Bird@...y.com,
	jonathan@...pberrypi.com,
	linux-kernel@...r.kernel.org,
	linux-mmc@...r.kernel.org,
	ulf.hansson@...aro.org
Subject: [PATCH] mmc: core: apply SD quirks earlier during probe

> diff --git a/drivers/mmc/core/quirks.h b/drivers/mmc/core/quirks.h
> index cca71867bc4a..92905fc46436 100644
> --- a/drivers/mmc/core/quirks.h
> +++ b/drivers/mmc/core/quirks.h
> @@ -15,6 +15,19 @@
> 
>  #include "card.h"
> 
> +static const struct mmc_fixup __maybe_unused mmc_sd_fixups[] = {
> +	/*
> +	 * Kingston Canvas Go! Plus microSD cards never finish SD cache flush.
> +	 * This has so far only been observed on cards from 11/2019, while new
> +	 * cards from 2023/05 do not exhibit this behavior.
> +	 */
> +	_FIXUP_EXT("SD64G", CID_MANFID_KINGSTON_SD, 0x5449, 2019, 11,
> +		   0, -1ull, SDIO_ANY_ID, SDIO_ANY_ID, add_quirk_sd,
> +		   MMC_QUIRK_BROKEN_SD_CACHE, EXT_CSD_REV_ANY),
> +
> +	END_FIXUP
> +};
> +

By happenstance I have just such a Kingston Canvas Go! Plus card, but
in standard SD form-factor and not microSD, and dating from *08/2021*.
I had heretofore-inexplicable filesystem corruption problems with this
card that may well be attributable to a failure to flush the cache.
Despite diligence in issuing an explicit "sync" before ejecting, the
filesystem would reliably appear corrupt after reconnecting it, and
would require a "fsck", but no bad blocks could be identified.

The vitals of my card are:

  type:   SD
  name:   SD64G
  date:   08/2021
  manfid: 0x00009f (Kingston)
  oemid:  0x5449
  fwrev:  0x1
  hwrev:  0x6
  csd:    400e00325b590001cf9f7f800a400001
  ocr:    0x00300000

I think the quirk here can be safely broadened to all variants of
this card and to the date range 11/2019 - 08/2021 inclusive.

Sincerely,
Olexa Bilaniuk