* [RFC PATCH 0/2] mmc: allow to rely on the DT to apply quirks @ 2021-10-14 14:30 Jerome Pouiller 2021-10-14 14:30 ` [RFC PATCH 1/2] mmc: rewrite mmc_fixup_device() Jerome Pouiller 2021-10-14 14:30 ` [RFC PATCH 2/2] mmc: allow to match the device tree to apply quirks Jerome Pouiller 0 siblings, 2 replies; 4+ messages in thread From: Jerome Pouiller @ 2021-10-14 14:30 UTC (permalink / raw) To: Ulf Hansson Cc: linux-mmc, linux-kernel, linux-wireless, Kalle Valo, Jérôme Pouiller From: Jérôme Pouiller <jerome.pouiller@silabs.com> Hi Ulf, I try to write a patch to problem discussed here[1]. This patch is not complete. Especially, it does not contains the necessary FIXUP macro in card.h. I would like to have your comment on the concept. [1] https://lore.kernel.org/netdev/CAPDyKFr62Kykg3=9WiXAV8UToqjw8gj4t6bbM7pGQ+iGGQRLmg@mail.gmail.com/ Jérôme Pouiller (2): mmc: rewrite mmc_fixup_device() mmc: allow to match the device tree to apply quirks drivers/mmc/core/card.h | 3 ++ drivers/mmc/core/quirks.h | 59 ++++++++++++++++++++++++++++----------- 2 files changed, 45 insertions(+), 17 deletions(-) -- 2.33.0 ^ permalink raw reply [flat|nested] 4+ messages in thread
* [RFC PATCH 1/2] mmc: rewrite mmc_fixup_device() 2021-10-14 14:30 [RFC PATCH 0/2] mmc: allow to rely on the DT to apply quirks Jerome Pouiller @ 2021-10-14 14:30 ` Jerome Pouiller 2021-10-14 14:30 ` [RFC PATCH 2/2] mmc: allow to match the device tree to apply quirks Jerome Pouiller 1 sibling, 0 replies; 4+ messages in thread From: Jerome Pouiller @ 2021-10-14 14:30 UTC (permalink / raw) To: Ulf Hansson Cc: linux-mmc, linux-kernel, linux-wireless, Kalle Valo, Jérôme Pouiller From: Jérôme Pouiller <jerome.pouiller@silabs.com> Currently, mmc_fixup_device() is a bit difficult to read because of particularly long condition. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> --- drivers/mmc/core/quirks.h | 41 +++++++++++++++++++++++---------------- 1 file changed, 24 insertions(+), 17 deletions(-) diff --git a/drivers/mmc/core/quirks.h b/drivers/mmc/core/quirks.h index d68e6e513a4f..c7ef2d14b359 100644 --- a/drivers/mmc/core/quirks.h +++ b/drivers/mmc/core/quirks.h @@ -152,22 +152,29 @@ static inline void mmc_fixup_device(struct mmc_card *card, u64 rev = cid_rev_card(card); for (f = table; f->vendor_fixup; f++) { - if ((f->manfid == CID_MANFID_ANY || - f->manfid == card->cid.manfid) && - (f->oemid == CID_OEMID_ANY || - f->oemid == card->cid.oemid) && - (f->name == CID_NAME_ANY || - !strncmp(f->name, card->cid.prod_name, - sizeof(card->cid.prod_name))) && - (f->cis_vendor == card->cis.vendor || - f->cis_vendor == (u16) SDIO_ANY_ID) && - (f->cis_device == card->cis.device || - f->cis_device == (u16) SDIO_ANY_ID) && - (f->ext_csd_rev == EXT_CSD_REV_ANY || - f->ext_csd_rev == card->ext_csd.rev) && - rev >= f->rev_start && rev <= f->rev_end) { - dev_dbg(&card->dev, "calling %ps\n", f->vendor_fixup); - f->vendor_fixup(card, f->data); - } + if (f->manfid != CID_MANFID_ANY && + f->manfid != card->cid.manfid) + continue; + if (f->oemid != CID_OEMID_ANY && + f->oemid != card->cid.oemid) + continue; + if (f->name != CID_NAME_ANY && + strncmp(f->name, card->cid.prod_name, + sizeof(card->cid.prod_name))) + continue; + if (f->cis_vendor != (u16)SDIO_ANY_ID && + f->cis_vendor != card->cis.vendor) + continue; + if (f->cis_device != (u16)SDIO_ANY_ID && + f->cis_device != card->cis.device) + continue; + if (f->ext_csd_rev != EXT_CSD_REV_ANY && + f->ext_csd_rev != card->ext_csd.rev) + continue; + if (rev < f->rev_start || rev > f->rev_end) + continue; + + dev_dbg(&card->dev, "calling %ps\n", f->vendor_fixup); + f->vendor_fixup(card, f->data); } } -- 2.33.0 ^ permalink raw reply related [flat|nested] 4+ messages in thread
* [RFC PATCH 2/2] mmc: allow to match the device tree to apply quirks 2021-10-14 14:30 [RFC PATCH 0/2] mmc: allow to rely on the DT to apply quirks Jerome Pouiller 2021-10-14 14:30 ` [RFC PATCH 1/2] mmc: rewrite mmc_fixup_device() Jerome Pouiller @ 2021-10-14 14:30 ` Jerome Pouiller 2021-10-20 14:29 ` Ulf Hansson 1 sibling, 1 reply; 4+ messages in thread From: Jerome Pouiller @ 2021-10-14 14:30 UTC (permalink / raw) To: Ulf Hansson Cc: linux-mmc, linux-kernel, linux-wireless, Kalle Valo, Jérôme Pouiller From: Jérôme Pouiller <jerome.pouiller@silabs.com> MMC subsystem provides a way to apply quirks when a device match some properties (VID, PID, etc...) Unfortunately, some SDIO devices does not comply with the SDIO specification and does not provide reliable VID/PID (eg. Silabs WF200). So, the drivers for these devices rely on device tree to identify the device. This patch allows the MMC to also rely on the device tree to apply a quirk. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> --- drivers/mmc/core/card.h | 3 +++ drivers/mmc/core/quirks.h | 18 ++++++++++++++++++ 2 files changed, 21 insertions(+) diff --git a/drivers/mmc/core/card.h b/drivers/mmc/core/card.h index 7bd392d55cfa..2f73f8567e14 100644 --- a/drivers/mmc/core/card.h +++ b/drivers/mmc/core/card.h @@ -59,6 +59,9 @@ struct mmc_fixup { /* for MMC cards */ unsigned int ext_csd_rev; + /* Match against functions declared in device tree */ + const char *const *of_compatible; + void (*vendor_fixup)(struct mmc_card *card, int data); int data; }; diff --git a/drivers/mmc/core/quirks.h b/drivers/mmc/core/quirks.h index c7ef2d14b359..ef368386e711 100644 --- a/drivers/mmc/core/quirks.h +++ b/drivers/mmc/core/quirks.h @@ -10,6 +10,7 @@ * */ +#include <linux/of.h> #include <linux/mmc/sdio_ids.h> #include "card.h" @@ -145,6 +146,20 @@ static const struct mmc_fixup __maybe_unused sdio_fixup_methods[] = { END_FIXUP }; +static inline bool mmc_fixup_of_compatible_match(struct mmc_card *card, + const char *const *compat_list) +{ + struct device_node *of_node; + int i; + + for (i = 0; i < 7; i++) { + of_node = mmc_of_find_child_device(card->host, i); + if (of_node && of_device_compatible_match(of_node, compat_list)) + return true; + } + return false; +} + static inline void mmc_fixup_device(struct mmc_card *card, const struct mmc_fixup *table) { @@ -173,6 +188,9 @@ static inline void mmc_fixup_device(struct mmc_card *card, continue; if (rev < f->rev_start || rev > f->rev_end) continue; + if (f->of_compatible != NULL && + !mmc_fixup_of_compatible_match(card, f->of_compatible)) + continue; dev_dbg(&card->dev, "calling %ps\n", f->vendor_fixup); f->vendor_fixup(card, f->data); -- 2.33.0 ^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [RFC PATCH 2/2] mmc: allow to match the device tree to apply quirks 2021-10-14 14:30 ` [RFC PATCH 2/2] mmc: allow to match the device tree to apply quirks Jerome Pouiller @ 2021-10-20 14:29 ` Ulf Hansson 0 siblings, 0 replies; 4+ messages in thread From: Ulf Hansson @ 2021-10-20 14:29 UTC (permalink / raw) To: Jerome Pouiller Cc: linux-mmc, Linux Kernel Mailing List, linux-wireless, Kalle Valo On Thu, 14 Oct 2021 at 16:30, Jerome Pouiller <Jerome.Pouiller@silabs.com> wrote: > > From: Jérôme Pouiller <jerome.pouiller@silabs.com> > > MMC subsystem provides a way to apply quirks when a device match some > properties (VID, PID, etc...) Unfortunately, some SDIO devices does not > comply with the SDIO specification and does not provide reliable VID/PID > (eg. Silabs WF200). > > So, the drivers for these devices rely on device tree to identify the > device. > > This patch allows the MMC to also rely on the device tree to apply a > quirk. > > Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> > --- > drivers/mmc/core/card.h | 3 +++ > drivers/mmc/core/quirks.h | 18 ++++++++++++++++++ > 2 files changed, 21 insertions(+) > > diff --git a/drivers/mmc/core/card.h b/drivers/mmc/core/card.h > index 7bd392d55cfa..2f73f8567e14 100644 > --- a/drivers/mmc/core/card.h > +++ b/drivers/mmc/core/card.h > @@ -59,6 +59,9 @@ struct mmc_fixup { > /* for MMC cards */ > unsigned int ext_csd_rev; > > + /* Match against functions declared in device tree */ > + const char *const *of_compatible; const char is probably sufficient. > + > void (*vendor_fixup)(struct mmc_card *card, int data); > int data; > }; > diff --git a/drivers/mmc/core/quirks.h b/drivers/mmc/core/quirks.h > index c7ef2d14b359..ef368386e711 100644 > --- a/drivers/mmc/core/quirks.h > +++ b/drivers/mmc/core/quirks.h > @@ -10,6 +10,7 @@ > * > */ > > +#include <linux/of.h> > #include <linux/mmc/sdio_ids.h> > > #include "card.h" > @@ -145,6 +146,20 @@ static const struct mmc_fixup __maybe_unused sdio_fixup_methods[] = { > END_FIXUP > }; > > +static inline bool mmc_fixup_of_compatible_match(struct mmc_card *card, > + const char *const *compat_list) > +{ > + struct device_node *of_node; > + int i; > + > + for (i = 0; i < 7; i++) { We need to iterate up '8', to cover all functional devices. The function numbers are from 1->7, while we use '0' for the "mmc-card" compatible. > + of_node = mmc_of_find_child_device(card->host, i); > + if (of_node && of_device_compatible_match(of_node, compat_list)) > + return true; > + } Another option, which would avoid doing the iteration over and over again in mmc_of_find_child_device(), would be to simply do the parsing of the compatible in the child nodes, right here. In other words, just use for_each_child_of_node(). > + return false; > +} > + > static inline void mmc_fixup_device(struct mmc_card *card, > const struct mmc_fixup *table) > { > @@ -173,6 +188,9 @@ static inline void mmc_fixup_device(struct mmc_card *card, > continue; > if (rev < f->rev_start || rev > f->rev_end) > continue; > + if (f->of_compatible != NULL && > + !mmc_fixup_of_compatible_match(card, f->of_compatible)) > + continue; > > dev_dbg(&card->dev, "calling %ps\n", f->vendor_fixup); > f->vendor_fixup(card, f->data); Kind regards Uffe ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-10-20 14:30 UTC | newest] Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2021-10-14 14:30 [RFC PATCH 0/2] mmc: allow to rely on the DT to apply quirks Jerome Pouiller 2021-10-14 14:30 ` [RFC PATCH 1/2] mmc: rewrite mmc_fixup_device() Jerome Pouiller 2021-10-14 14:30 ` [RFC PATCH 2/2] mmc: allow to match the device tree to apply quirks Jerome Pouiller 2021-10-20 14:29 ` Ulf Hansson
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).