* [RESEND PATCH 0/3] Introduce the request_atomic() for the host
@ 2020-03-04 7:41 Baolin Wang
2020-03-04 7:42 ` [RESEND PATCH 1/3] mmc: host: " Baolin Wang
` (2 more replies)
0 siblings, 3 replies; 12+ messages in thread
From: Baolin Wang @ 2020-03-04 7:41 UTC (permalink / raw)
To: adrian.hunter, ulf.hansson
Cc: orsonzhai, zhang.lyra, baolin.wang7, arnd, linux-mmc, linux-kernel
This patch set introduces a new request_atomic() interface for the
MMC host controller, which is used to submit a request to host in
the atomic context, such as in the irq hard handler, to reduce the
request latency.
Any comments are welcome. Thanks.
Baolin Wang (3):
mmc: host: Introduce the request_atomic() for the host
mmc: host: sdhci-sprd: Implement the request_atomic() API
mmc: host: hsq: Support request_atomic() API
drivers/mmc/host/mmc_hsq.c | 5 ++++-
drivers/mmc/host/sdhci-sprd.c | 28 ++++++++++++++++++++++++++--
drivers/mmc/host/sdhci.c | 27 +++++++++++++++++++--------
drivers/mmc/host/sdhci.h | 1 +
include/linux/mmc/host.h | 3 +++
5 files changed, 53 insertions(+), 11 deletions(-)
--
1.9.1
^ permalink raw reply [flat|nested] 12+ messages in thread* [RESEND PATCH 1/3] mmc: host: Introduce the request_atomic() for the host 2020-03-04 7:41 [RESEND PATCH 0/3] Introduce the request_atomic() for the host Baolin Wang @ 2020-03-04 7:42 ` Baolin Wang 2020-03-16 11:47 ` Ulf Hansson 2020-03-16 13:08 ` Adrian Hunter 2020-03-04 7:42 ` [RESEND PATCH 2/3] mmc: host: sdhci-sprd: Implement the request_atomic() API Baolin Wang 2020-03-04 7:42 ` [RESEND PATCH 3/3] mmc: host: hsq: Support " Baolin Wang 2 siblings, 2 replies; 12+ messages in thread From: Baolin Wang @ 2020-03-04 7:42 UTC (permalink / raw) To: adrian.hunter, ulf.hansson Cc: orsonzhai, zhang.lyra, baolin.wang7, arnd, linux-mmc, linux-kernel The SD host controller can process one request in the atomic context if the card is nonremovable, which means we can submit next request in the irq hard handler when using the MMC software queue to reduce the latency. Thus this patch adds a new API request_atomic() for the host controller and implement it for the SD host controller. Suggested-by: Adrian Hunter <adrian.hunter@intel.com> Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> --- drivers/mmc/host/sdhci.c | 27 +++++++++++++++++++-------- drivers/mmc/host/sdhci.h | 1 + include/linux/mmc/host.h | 3 +++ 3 files changed, 23 insertions(+), 8 deletions(-) diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c index 9c37451..4febbcb 100644 --- a/drivers/mmc/host/sdhci.c +++ b/drivers/mmc/host/sdhci.c @@ -2016,17 +2016,12 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, * * \*****************************************************************************/ -void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) +static void sdhci_start_request(struct mmc_host *mmc, struct mmc_request *mrq, + int present) { - struct sdhci_host *host; - int present; + struct sdhci_host *host = mmc_priv(mmc); unsigned long flags; - host = mmc_priv(mmc); - - /* Firstly check card presence */ - present = mmc->ops->get_cd(mmc); - spin_lock_irqsave(&host->lock, flags); sdhci_led_activate(host); @@ -2043,6 +2038,22 @@ void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) spin_unlock_irqrestore(&host->lock, flags); } + +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq) +{ + sdhci_start_request(mmc, mrq, 1); +} +EXPORT_SYMBOL_GPL(sdhci_request_atomic); + +void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) +{ + int present; + + /* Firstly check card presence */ + present = mmc->ops->get_cd(mmc); + + sdhci_start_request(mmc, mrq, present); +} EXPORT_SYMBOL_GPL(sdhci_request); void sdhci_set_bus_width(struct sdhci_host *host, int width) diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h index cac2d97..5507a73 100644 --- a/drivers/mmc/host/sdhci.h +++ b/drivers/mmc/host/sdhci.h @@ -775,6 +775,7 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, void sdhci_set_power_noreg(struct sdhci_host *host, unsigned char mode, unsigned short vdd); void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq); +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq); void sdhci_set_bus_width(struct sdhci_host *host, int width); void sdhci_reset(struct sdhci_host *host, u8 mask); void sdhci_set_uhs_signaling(struct sdhci_host *host, unsigned timing); diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h index 562ed06..db5e59c 100644 --- a/include/linux/mmc/host.h +++ b/include/linux/mmc/host.h @@ -92,6 +92,9 @@ struct mmc_host_ops { int err); void (*pre_req)(struct mmc_host *host, struct mmc_request *req); void (*request)(struct mmc_host *host, struct mmc_request *req); + /* Submit one request to host in atomic context. */ + void (*request_atomic)(struct mmc_host *host, + struct mmc_request *req); /* * Avoid calling the next three functions too often or in a "fast -- 1.9.1 ^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: [RESEND PATCH 1/3] mmc: host: Introduce the request_atomic() for the host 2020-03-04 7:42 ` [RESEND PATCH 1/3] mmc: host: " Baolin Wang @ 2020-03-16 11:47 ` Ulf Hansson 2020-03-17 2:00 ` Baolin Wang 2020-03-16 13:08 ` Adrian Hunter 1 sibling, 1 reply; 12+ messages in thread From: Ulf Hansson @ 2020-03-16 11:47 UTC (permalink / raw) To: Baolin Wang Cc: Adrian Hunter, Orson Zhai, Chunyan Zhang, Arnd Bergmann, linux-mmc, Linux Kernel Mailing List On Wed, 4 Mar 2020 at 08:42, Baolin Wang <baolin.wang7@gmail.com> wrote: > > The SD host controller can process one request in the atomic context if > the card is nonremovable, which means we can submit next request in the > irq hard handler when using the MMC software queue to reduce the latency. > Thus this patch adds a new API request_atomic() for the host controller > and implement it for the SD host controller. > > Suggested-by: Adrian Hunter <adrian.hunter@intel.com> > Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> > --- > drivers/mmc/host/sdhci.c | 27 +++++++++++++++++++-------- > drivers/mmc/host/sdhci.h | 1 + > include/linux/mmc/host.h | 3 +++ > 3 files changed, 23 insertions(+), 8 deletions(-) I think the code split of the changes in the series can be improved a bit, so I suggest you move the code around in the series to reach this: 1. A patch that adds the new host ops callback, combined with the change you have in patch3. 2. The sdhci core specific changes, from $subject patch. 3. The sdhci-sprd changes, as in patch2. Other than that, I think the code looks good to me, besides a minor comment on patch2, see separate reply. Kind regards Uffe > > diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c > index 9c37451..4febbcb 100644 > --- a/drivers/mmc/host/sdhci.c > +++ b/drivers/mmc/host/sdhci.c > @@ -2016,17 +2016,12 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, > * * > \*****************************************************************************/ > > -void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > +static void sdhci_start_request(struct mmc_host *mmc, struct mmc_request *mrq, > + int present) > { > - struct sdhci_host *host; > - int present; > + struct sdhci_host *host = mmc_priv(mmc); > unsigned long flags; > > - host = mmc_priv(mmc); > - > - /* Firstly check card presence */ > - present = mmc->ops->get_cd(mmc); > - > spin_lock_irqsave(&host->lock, flags); > > sdhci_led_activate(host); > @@ -2043,6 +2038,22 @@ void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > > spin_unlock_irqrestore(&host->lock, flags); > } > + > +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq) > +{ > + sdhci_start_request(mmc, mrq, 1); > +} > +EXPORT_SYMBOL_GPL(sdhci_request_atomic); > + > +void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > +{ > + int present; > + > + /* Firstly check card presence */ > + present = mmc->ops->get_cd(mmc); > + > + sdhci_start_request(mmc, mrq, present); > +} > EXPORT_SYMBOL_GPL(sdhci_request); > > void sdhci_set_bus_width(struct sdhci_host *host, int width) > diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h > index cac2d97..5507a73 100644 > --- a/drivers/mmc/host/sdhci.h > +++ b/drivers/mmc/host/sdhci.h > @@ -775,6 +775,7 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, > void sdhci_set_power_noreg(struct sdhci_host *host, unsigned char mode, > unsigned short vdd); > void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq); > +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq); > void sdhci_set_bus_width(struct sdhci_host *host, int width); > void sdhci_reset(struct sdhci_host *host, u8 mask); > void sdhci_set_uhs_signaling(struct sdhci_host *host, unsigned timing); > diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h > index 562ed06..db5e59c 100644 > --- a/include/linux/mmc/host.h > +++ b/include/linux/mmc/host.h > @@ -92,6 +92,9 @@ struct mmc_host_ops { > int err); > void (*pre_req)(struct mmc_host *host, struct mmc_request *req); > void (*request)(struct mmc_host *host, struct mmc_request *req); > + /* Submit one request to host in atomic context. */ > + void (*request_atomic)(struct mmc_host *host, > + struct mmc_request *req); > > /* > * Avoid calling the next three functions too often or in a "fast > -- > 1.9.1 > ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [RESEND PATCH 1/3] mmc: host: Introduce the request_atomic() for the host 2020-03-16 11:47 ` Ulf Hansson @ 2020-03-17 2:00 ` Baolin Wang 0 siblings, 0 replies; 12+ messages in thread From: Baolin Wang @ 2020-03-17 2:00 UTC (permalink / raw) To: Ulf Hansson Cc: Adrian Hunter, Orson Zhai, Chunyan Zhang, Arnd Bergmann, linux-mmc, Linux Kernel Mailing List On Mon, Mar 16, 2020 at 7:48 PM Ulf Hansson <ulf.hansson@linaro.org> wrote: > > On Wed, 4 Mar 2020 at 08:42, Baolin Wang <baolin.wang7@gmail.com> wrote: > > > > The SD host controller can process one request in the atomic context if > > the card is nonremovable, which means we can submit next request in the > > irq hard handler when using the MMC software queue to reduce the latency. > > Thus this patch adds a new API request_atomic() for the host controller > > and implement it for the SD host controller. > > > > Suggested-by: Adrian Hunter <adrian.hunter@intel.com> > > Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> > > --- > > drivers/mmc/host/sdhci.c | 27 +++++++++++++++++++-------- > > drivers/mmc/host/sdhci.h | 1 + > > include/linux/mmc/host.h | 3 +++ > > 3 files changed, 23 insertions(+), 8 deletions(-) > > I think the code split of the changes in the series can be improved a > bit, so I suggest you move the code around in the series to reach > this: > > 1. A patch that adds the new host ops callback, combined with the > change you have in patch3. > 2. The sdhci core specific changes, from $subject patch. > 3. The sdhci-sprd changes, as in patch2. > > Other than that, I think the code looks good to me, besides a minor > comment on patch2, see separate reply. Sure. Will do as you suggested. Thanks. -- Baolin Wang ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [RESEND PATCH 1/3] mmc: host: Introduce the request_atomic() for the host 2020-03-04 7:42 ` [RESEND PATCH 1/3] mmc: host: " Baolin Wang 2020-03-16 11:47 ` Ulf Hansson @ 2020-03-16 13:08 ` Adrian Hunter 2020-03-17 3:36 ` Baolin Wang 1 sibling, 1 reply; 12+ messages in thread From: Adrian Hunter @ 2020-03-16 13:08 UTC (permalink / raw) To: Baolin Wang, ulf.hansson Cc: orsonzhai, zhang.lyra, arnd, linux-mmc, linux-kernel On 4/03/20 9:42 am, Baolin Wang wrote: > The SD host controller can process one request in the atomic context if > the card is nonremovable, which means we can submit next request in the > irq hard handler when using the MMC software queue to reduce the latency. > Thus this patch adds a new API request_atomic() for the host controller > and implement it for the SD host controller. > > Suggested-by: Adrian Hunter <adrian.hunter@intel.com> > Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> > --- > drivers/mmc/host/sdhci.c | 27 +++++++++++++++++++-------- > drivers/mmc/host/sdhci.h | 1 + > include/linux/mmc/host.h | 3 +++ > 3 files changed, 23 insertions(+), 8 deletions(-) > > diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c > index 9c37451..4febbcb 100644 > --- a/drivers/mmc/host/sdhci.c > +++ b/drivers/mmc/host/sdhci.c > @@ -2016,17 +2016,12 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, > * * > \*****************************************************************************/ > > -void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > +static void sdhci_start_request(struct mmc_host *mmc, struct mmc_request *mrq, > + int present) > { > - struct sdhci_host *host; > - int present; > + struct sdhci_host *host = mmc_priv(mmc); > unsigned long flags; > > - host = mmc_priv(mmc); > - > - /* Firstly check card presence */ > - present = mmc->ops->get_cd(mmc); > - > spin_lock_irqsave(&host->lock, flags); > > sdhci_led_activate(host); > @@ -2043,6 +2038,22 @@ void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > > spin_unlock_irqrestore(&host->lock, flags); > } > + > +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq) > +{ > + sdhci_start_request(mmc, mrq, 1); > +} > +EXPORT_SYMBOL_GPL(sdhci_request_atomic); > + > +void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > +{ > + int present; > + > + /* Firstly check card presence */ > + present = mmc->ops->get_cd(mmc); > + > + sdhci_start_request(mmc, mrq, present); > +} > EXPORT_SYMBOL_GPL(sdhci_request); > > void sdhci_set_bus_width(struct sdhci_host *host, int width) > diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h > index cac2d97..5507a73 100644 > --- a/drivers/mmc/host/sdhci.h > +++ b/drivers/mmc/host/sdhci.h > @@ -775,6 +775,7 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, > void sdhci_set_power_noreg(struct sdhci_host *host, unsigned char mode, > unsigned short vdd); > void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq); > +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq); > void sdhci_set_bus_width(struct sdhci_host *host, int width); > void sdhci_reset(struct sdhci_host *host, u8 mask); > void sdhci_set_uhs_signaling(struct sdhci_host *host, unsigned timing); > diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h > index 562ed06..db5e59c 100644 > --- a/include/linux/mmc/host.h > +++ b/include/linux/mmc/host.h > @@ -92,6 +92,9 @@ struct mmc_host_ops { > int err); > void (*pre_req)(struct mmc_host *host, struct mmc_request *req); > void (*request)(struct mmc_host *host, struct mmc_request *req); > + /* Submit one request to host in atomic context. */ > + void (*request_atomic)(struct mmc_host *host, > + struct mmc_request *req); This doesn't have the flexibility to return "busy". For example, sdhci_send_command() will potentially wait quite some time if the inhibit bits are set. That is not good in interrupt context. It would be better to return immediately in that case and have the caller fall back to a non-atomic context. Thoughts? > > /* > * Avoid calling the next three functions too often or in a "fast > ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [RESEND PATCH 1/3] mmc: host: Introduce the request_atomic() for the host 2020-03-16 13:08 ` Adrian Hunter @ 2020-03-17 3:36 ` Baolin Wang 2020-03-17 7:32 ` Adrian Hunter 0 siblings, 1 reply; 12+ messages in thread From: Baolin Wang @ 2020-03-17 3:36 UTC (permalink / raw) To: Adrian Hunter Cc: Ulf Hansson, Orson Zhai, Chunyan Zhang, Arnd Bergmann, linux-mmc, LKML On Mon, Mar 16, 2020 at 9:09 PM Adrian Hunter <adrian.hunter@intel.com> wrote: > > On 4/03/20 9:42 am, Baolin Wang wrote: > > The SD host controller can process one request in the atomic context if > > the card is nonremovable, which means we can submit next request in the > > irq hard handler when using the MMC software queue to reduce the latency. > > Thus this patch adds a new API request_atomic() for the host controller > > and implement it for the SD host controller. > > > > Suggested-by: Adrian Hunter <adrian.hunter@intel.com> > > Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> > > --- > > drivers/mmc/host/sdhci.c | 27 +++++++++++++++++++-------- > > drivers/mmc/host/sdhci.h | 1 + > > include/linux/mmc/host.h | 3 +++ > > 3 files changed, 23 insertions(+), 8 deletions(-) > > > > diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c > > index 9c37451..4febbcb 100644 > > --- a/drivers/mmc/host/sdhci.c > > +++ b/drivers/mmc/host/sdhci.c > > @@ -2016,17 +2016,12 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, > > * * > > \*****************************************************************************/ > > > > -void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > > +static void sdhci_start_request(struct mmc_host *mmc, struct mmc_request *mrq, > > + int present) > > { > > - struct sdhci_host *host; > > - int present; > > + struct sdhci_host *host = mmc_priv(mmc); > > unsigned long flags; > > > > - host = mmc_priv(mmc); > > - > > - /* Firstly check card presence */ > > - present = mmc->ops->get_cd(mmc); > > - > > spin_lock_irqsave(&host->lock, flags); > > > > sdhci_led_activate(host); > > @@ -2043,6 +2038,22 @@ void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > > > > spin_unlock_irqrestore(&host->lock, flags); > > } > > + > > +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq) > > +{ > > + sdhci_start_request(mmc, mrq, 1); > > +} > > +EXPORT_SYMBOL_GPL(sdhci_request_atomic); > > + > > +void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > > +{ > > + int present; > > + > > + /* Firstly check card presence */ > > + present = mmc->ops->get_cd(mmc); > > + > > + sdhci_start_request(mmc, mrq, present); > > +} > > EXPORT_SYMBOL_GPL(sdhci_request); > > > > void sdhci_set_bus_width(struct sdhci_host *host, int width) > > diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h > > index cac2d97..5507a73 100644 > > --- a/drivers/mmc/host/sdhci.h > > +++ b/drivers/mmc/host/sdhci.h > > @@ -775,6 +775,7 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, > > void sdhci_set_power_noreg(struct sdhci_host *host, unsigned char mode, > > unsigned short vdd); > > void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq); > > +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq); > > void sdhci_set_bus_width(struct sdhci_host *host, int width); > > void sdhci_reset(struct sdhci_host *host, u8 mask); > > void sdhci_set_uhs_signaling(struct sdhci_host *host, unsigned timing); > > diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h > > index 562ed06..db5e59c 100644 > > --- a/include/linux/mmc/host.h > > +++ b/include/linux/mmc/host.h > > @@ -92,6 +92,9 @@ struct mmc_host_ops { > > int err); > > void (*pre_req)(struct mmc_host *host, struct mmc_request *req); > > void (*request)(struct mmc_host *host, struct mmc_request *req); > > + /* Submit one request to host in atomic context. */ > > + void (*request_atomic)(struct mmc_host *host, > > + struct mmc_request *req); > > This doesn't have the flexibility to return "busy". For example, > sdhci_send_command() will potentially wait quite some time if the inhibit > bits are set. That is not good in interrupt context. It would be better to > return immediately in that case and have the caller fall back to a > non-atomic context. Thoughts? Yes, I unserstood your concern. But the sdhci_send_command() is already under the spin_lock_irqsave() protection, which will also disable the interrupt for some time if the inhibit bits are set. That is same with moving it in interrupt context. Moreover, if the previous command complete interrupt and transfer complete interrupt are normal, we should not meet this issue of polling inhibit bits (I have not met this issue on my platform). So I think we can remove the polling here? If the inhibit bits are set, I think the command complete interrupt or the transfer complete interrupt have been abnormal, so we can just return the error here. What do you think? Thanks. -- Baolin Wang ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [RESEND PATCH 1/3] mmc: host: Introduce the request_atomic() for the host 2020-03-17 3:36 ` Baolin Wang @ 2020-03-17 7:32 ` Adrian Hunter 2020-03-17 9:17 ` Baolin Wang 0 siblings, 1 reply; 12+ messages in thread From: Adrian Hunter @ 2020-03-17 7:32 UTC (permalink / raw) To: Baolin Wang Cc: Ulf Hansson, Orson Zhai, Chunyan Zhang, Arnd Bergmann, linux-mmc, LKML On 17/03/20 5:36 am, Baolin Wang wrote: > On Mon, Mar 16, 2020 at 9:09 PM Adrian Hunter <adrian.hunter@intel.com> wrote: >> >> On 4/03/20 9:42 am, Baolin Wang wrote: >>> The SD host controller can process one request in the atomic context if >>> the card is nonremovable, which means we can submit next request in the >>> irq hard handler when using the MMC software queue to reduce the latency. >>> Thus this patch adds a new API request_atomic() for the host controller >>> and implement it for the SD host controller. >>> >>> Suggested-by: Adrian Hunter <adrian.hunter@intel.com> >>> Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> >>> --- >>> drivers/mmc/host/sdhci.c | 27 +++++++++++++++++++-------- >>> drivers/mmc/host/sdhci.h | 1 + >>> include/linux/mmc/host.h | 3 +++ >>> 3 files changed, 23 insertions(+), 8 deletions(-) >>> >>> diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c >>> index 9c37451..4febbcb 100644 >>> --- a/drivers/mmc/host/sdhci.c >>> +++ b/drivers/mmc/host/sdhci.c >>> @@ -2016,17 +2016,12 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, >>> * * >>> \*****************************************************************************/ >>> >>> -void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) >>> +static void sdhci_start_request(struct mmc_host *mmc, struct mmc_request *mrq, >>> + int present) >>> { >>> - struct sdhci_host *host; >>> - int present; >>> + struct sdhci_host *host = mmc_priv(mmc); >>> unsigned long flags; >>> >>> - host = mmc_priv(mmc); >>> - >>> - /* Firstly check card presence */ >>> - present = mmc->ops->get_cd(mmc); >>> - >>> spin_lock_irqsave(&host->lock, flags); >>> >>> sdhci_led_activate(host); >>> @@ -2043,6 +2038,22 @@ void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) >>> >>> spin_unlock_irqrestore(&host->lock, flags); >>> } >>> + >>> +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq) >>> +{ >>> + sdhci_start_request(mmc, mrq, 1); >>> +} >>> +EXPORT_SYMBOL_GPL(sdhci_request_atomic); >>> + >>> +void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) >>> +{ >>> + int present; >>> + >>> + /* Firstly check card presence */ >>> + present = mmc->ops->get_cd(mmc); >>> + >>> + sdhci_start_request(mmc, mrq, present); >>> +} >>> EXPORT_SYMBOL_GPL(sdhci_request); >>> >>> void sdhci_set_bus_width(struct sdhci_host *host, int width) >>> diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h >>> index cac2d97..5507a73 100644 >>> --- a/drivers/mmc/host/sdhci.h >>> +++ b/drivers/mmc/host/sdhci.h >>> @@ -775,6 +775,7 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, >>> void sdhci_set_power_noreg(struct sdhci_host *host, unsigned char mode, >>> unsigned short vdd); >>> void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq); >>> +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq); >>> void sdhci_set_bus_width(struct sdhci_host *host, int width); >>> void sdhci_reset(struct sdhci_host *host, u8 mask); >>> void sdhci_set_uhs_signaling(struct sdhci_host *host, unsigned timing); >>> diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h >>> index 562ed06..db5e59c 100644 >>> --- a/include/linux/mmc/host.h >>> +++ b/include/linux/mmc/host.h >>> @@ -92,6 +92,9 @@ struct mmc_host_ops { >>> int err); >>> void (*pre_req)(struct mmc_host *host, struct mmc_request *req); >>> void (*request)(struct mmc_host *host, struct mmc_request *req); >>> + /* Submit one request to host in atomic context. */ >>> + void (*request_atomic)(struct mmc_host *host, >>> + struct mmc_request *req); >> >> This doesn't have the flexibility to return "busy". For example, >> sdhci_send_command() will potentially wait quite some time if the inhibit >> bits are set. That is not good in interrupt context. It would be better to >> return immediately in that case and have the caller fall back to a >> non-atomic context. Thoughts? > > Yes, I unserstood your concern. But the sdhci_send_command() is > already under the spin_lock_irqsave() protection, which will also > disable the interrupt for some time if the inhibit bits are set. That > is same with moving it in interrupt context. It is, but I would like to fix that too. > > Moreover, if the previous command complete interrupt and transfer > complete interrupt are normal, we should not meet this issue of > polling inhibit bits (I have not met this issue on my platform). So I > think we can remove the polling here? If the inhibit bits are set, I > think the command complete interrupt or the transfer complete > interrupt have been abnormal, so we can just return the error here. > What do you think? Thanks. > I suspect the inhibit polling might be needed for some host controllers in some situations. ie. taking it out would likely break things. ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [RESEND PATCH 1/3] mmc: host: Introduce the request_atomic() for the host 2020-03-17 7:32 ` Adrian Hunter @ 2020-03-17 9:17 ` Baolin Wang 0 siblings, 0 replies; 12+ messages in thread From: Baolin Wang @ 2020-03-17 9:17 UTC (permalink / raw) To: Adrian Hunter Cc: Ulf Hansson, Orson Zhai, Chunyan Zhang, Arnd Bergmann, linux-mmc, LKML On Tue, Mar 17, 2020 at 3:32 PM Adrian Hunter <adrian.hunter@intel.com> wrote: > > On 17/03/20 5:36 am, Baolin Wang wrote: > > On Mon, Mar 16, 2020 at 9:09 PM Adrian Hunter <adrian.hunter@intel.com> wrote: > >> > >> On 4/03/20 9:42 am, Baolin Wang wrote: > >>> The SD host controller can process one request in the atomic context if > >>> the card is nonremovable, which means we can submit next request in the > >>> irq hard handler when using the MMC software queue to reduce the latency. > >>> Thus this patch adds a new API request_atomic() for the host controller > >>> and implement it for the SD host controller. > >>> > >>> Suggested-by: Adrian Hunter <adrian.hunter@intel.com> > >>> Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> > >>> --- > >>> drivers/mmc/host/sdhci.c | 27 +++++++++++++++++++-------- > >>> drivers/mmc/host/sdhci.h | 1 + > >>> include/linux/mmc/host.h | 3 +++ > >>> 3 files changed, 23 insertions(+), 8 deletions(-) > >>> > >>> diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c > >>> index 9c37451..4febbcb 100644 > >>> --- a/drivers/mmc/host/sdhci.c > >>> +++ b/drivers/mmc/host/sdhci.c > >>> @@ -2016,17 +2016,12 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, > >>> * * > >>> \*****************************************************************************/ > >>> > >>> -void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > >>> +static void sdhci_start_request(struct mmc_host *mmc, struct mmc_request *mrq, > >>> + int present) > >>> { > >>> - struct sdhci_host *host; > >>> - int present; > >>> + struct sdhci_host *host = mmc_priv(mmc); > >>> unsigned long flags; > >>> > >>> - host = mmc_priv(mmc); > >>> - > >>> - /* Firstly check card presence */ > >>> - present = mmc->ops->get_cd(mmc); > >>> - > >>> spin_lock_irqsave(&host->lock, flags); > >>> > >>> sdhci_led_activate(host); > >>> @@ -2043,6 +2038,22 @@ void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > >>> > >>> spin_unlock_irqrestore(&host->lock, flags); > >>> } > >>> + > >>> +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq) > >>> +{ > >>> + sdhci_start_request(mmc, mrq, 1); > >>> +} > >>> +EXPORT_SYMBOL_GPL(sdhci_request_atomic); > >>> + > >>> +void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq) > >>> +{ > >>> + int present; > >>> + > >>> + /* Firstly check card presence */ > >>> + present = mmc->ops->get_cd(mmc); > >>> + > >>> + sdhci_start_request(mmc, mrq, present); > >>> +} > >>> EXPORT_SYMBOL_GPL(sdhci_request); > >>> > >>> void sdhci_set_bus_width(struct sdhci_host *host, int width) > >>> diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h > >>> index cac2d97..5507a73 100644 > >>> --- a/drivers/mmc/host/sdhci.h > >>> +++ b/drivers/mmc/host/sdhci.h > >>> @@ -775,6 +775,7 @@ void sdhci_set_power(struct sdhci_host *host, unsigned char mode, > >>> void sdhci_set_power_noreg(struct sdhci_host *host, unsigned char mode, > >>> unsigned short vdd); > >>> void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq); > >>> +void sdhci_request_atomic(struct mmc_host *mmc, struct mmc_request *mrq); > >>> void sdhci_set_bus_width(struct sdhci_host *host, int width); > >>> void sdhci_reset(struct sdhci_host *host, u8 mask); > >>> void sdhci_set_uhs_signaling(struct sdhci_host *host, unsigned timing); > >>> diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h > >>> index 562ed06..db5e59c 100644 > >>> --- a/include/linux/mmc/host.h > >>> +++ b/include/linux/mmc/host.h > >>> @@ -92,6 +92,9 @@ struct mmc_host_ops { > >>> int err); > >>> void (*pre_req)(struct mmc_host *host, struct mmc_request *req); > >>> void (*request)(struct mmc_host *host, struct mmc_request *req); > >>> + /* Submit one request to host in atomic context. */ > >>> + void (*request_atomic)(struct mmc_host *host, > >>> + struct mmc_request *req); > >> > >> This doesn't have the flexibility to return "busy". For example, > >> sdhci_send_command() will potentially wait quite some time if the inhibit > >> bits are set. That is not good in interrupt context. It would be better to > >> return immediately in that case and have the caller fall back to a > >> non-atomic context. Thoughts? > > > > Yes, I unserstood your concern. But the sdhci_send_command() is > > already under the spin_lock_irqsave() protection, which will also > > disable the interrupt for some time if the inhibit bits are set. That > > is same with moving it in interrupt context. > > It is, but I would like to fix that too. OK. Like you suggested, cause we've aleady decided how to complete a request by sdhci_defer_done(), I need think about how to change to a non-atomic context for this unusual case. And since the original sdhci_send_command() has the same problem as I said, I perfer to create another patch set to fix this issue. > > Moreover, if the previous command complete interrupt and transfer > > complete interrupt are normal, we should not meet this issue of > > polling inhibit bits (I have not met this issue on my platform). So I > > think we can remove the polling here? If the inhibit bits are set, I > > think the command complete interrupt or the transfer complete > > interrupt have been abnormal, so we can just return the error here. > > What do you think? Thanks. > > > > I suspect the inhibit polling might be needed for some host controllers in > some situations. ie. taking it out would likely break things. Make sense. -- Baolin Wang ^ permalink raw reply [flat|nested] 12+ messages in thread
* [RESEND PATCH 2/3] mmc: host: sdhci-sprd: Implement the request_atomic() API 2020-03-04 7:41 [RESEND PATCH 0/3] Introduce the request_atomic() for the host Baolin Wang 2020-03-04 7:42 ` [RESEND PATCH 1/3] mmc: host: " Baolin Wang @ 2020-03-04 7:42 ` Baolin Wang 2020-03-16 11:47 ` Ulf Hansson 2020-03-04 7:42 ` [RESEND PATCH 3/3] mmc: host: hsq: Support " Baolin Wang 2 siblings, 1 reply; 12+ messages in thread From: Baolin Wang @ 2020-03-04 7:42 UTC (permalink / raw) To: adrian.hunter, ulf.hansson Cc: orsonzhai, zhang.lyra, baolin.wang7, arnd, linux-mmc, linux-kernel Implement the request_atomic() API for nonremovable cards, that means we can submit next request in the irq hard handler context to reduce latency. Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> --- drivers/mmc/host/sdhci-sprd.c | 28 ++++++++++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-) diff --git a/drivers/mmc/host/sdhci-sprd.c b/drivers/mmc/host/sdhci-sprd.c index 2ab42c5..bddf0f3 100644 --- a/drivers/mmc/host/sdhci-sprd.c +++ b/drivers/mmc/host/sdhci-sprd.c @@ -426,6 +426,27 @@ static void sdhci_sprd_request(struct mmc_host *mmc, struct mmc_request *mrq) sdhci_request(mmc, mrq); } +static void sdhci_sprd_request_atomic(struct mmc_host *mmc, + struct mmc_request *mrq) +{ + struct sdhci_host *host = mmc_priv(mmc); + struct sdhci_sprd_host *sprd_host = TO_SPRD_HOST(host); + + host->flags |= sprd_host->flags & SDHCI_AUTO_CMD23; + + /* + * From version 4.10 onward, ARGUMENT2 register is also as 32-bit + * block count register which doesn't support stuff bits of + * CMD23 argument on Spreadtrum's sd host controller. + */ + if (host->version >= SDHCI_SPEC_410 && + mrq->sbc && (mrq->sbc->arg & SDHCI_SPRD_ARG2_STUFF) && + (host->flags & SDHCI_AUTO_CMD23)) + host->flags &= ~SDHCI_AUTO_CMD23; + + sdhci_request_atomic(mmc, mrq); +} + static int sdhci_sprd_voltage_switch(struct mmc_host *mmc, struct mmc_ios *ios) { struct sdhci_host *host = mmc_priv(mmc); @@ -561,6 +582,11 @@ static int sdhci_sprd_probe(struct platform_device *pdev) if (ret) goto pltfm_free; + if (!mmc_card_is_removable(host->mmc)) + host->mmc_host_ops.request_atomic = sdhci_sprd_request_atomic; + else + host->always_defer_done = true; + sprd_host = TO_SPRD_HOST(host); sdhci_sprd_phy_param_parse(sprd_host, pdev->dev.of_node); @@ -654,8 +680,6 @@ static int sdhci_sprd_probe(struct platform_device *pdev) if (ret) goto err_cleanup_host; - host->always_defer_done = true; - ret = __sdhci_add_host(host); if (ret) goto err_cleanup_host; -- 1.9.1 ^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: [RESEND PATCH 2/3] mmc: host: sdhci-sprd: Implement the request_atomic() API 2020-03-04 7:42 ` [RESEND PATCH 2/3] mmc: host: sdhci-sprd: Implement the request_atomic() API Baolin Wang @ 2020-03-16 11:47 ` Ulf Hansson 2020-03-17 2:02 ` Baolin Wang 0 siblings, 1 reply; 12+ messages in thread From: Ulf Hansson @ 2020-03-16 11:47 UTC (permalink / raw) To: Baolin Wang Cc: Adrian Hunter, Orson Zhai, Chunyan Zhang, Arnd Bergmann, linux-mmc, Linux Kernel Mailing List On Wed, 4 Mar 2020 at 08:42, Baolin Wang <baolin.wang7@gmail.com> wrote: > > Implement the request_atomic() API for nonremovable cards, that means > we can submit next request in the irq hard handler context to reduce > latency. > > Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> > --- > drivers/mmc/host/sdhci-sprd.c | 28 ++++++++++++++++++++++++++-- > 1 file changed, 26 insertions(+), 2 deletions(-) > > diff --git a/drivers/mmc/host/sdhci-sprd.c b/drivers/mmc/host/sdhci-sprd.c > index 2ab42c5..bddf0f3 100644 > --- a/drivers/mmc/host/sdhci-sprd.c > +++ b/drivers/mmc/host/sdhci-sprd.c > @@ -426,6 +426,27 @@ static void sdhci_sprd_request(struct mmc_host *mmc, struct mmc_request *mrq) > sdhci_request(mmc, mrq); > } > > +static void sdhci_sprd_request_atomic(struct mmc_host *mmc, > + struct mmc_request *mrq) > +{ > + struct sdhci_host *host = mmc_priv(mmc); > + struct sdhci_sprd_host *sprd_host = TO_SPRD_HOST(host); > + > + host->flags |= sprd_host->flags & SDHCI_AUTO_CMD23; > + > + /* > + * From version 4.10 onward, ARGUMENT2 register is also as 32-bit > + * block count register which doesn't support stuff bits of > + * CMD23 argument on Spreadtrum's sd host controller. > + */ > + if (host->version >= SDHCI_SPEC_410 && > + mrq->sbc && (mrq->sbc->arg & SDHCI_SPRD_ARG2_STUFF) && > + (host->flags & SDHCI_AUTO_CMD23)) > + host->flags &= ~SDHCI_AUTO_CMD23; Looks like the above code is a copy of what is done in sdhci_sprd_request(). Perhaps add a helper function that deals with this to avoid open coding? > + > + sdhci_request_atomic(mmc, mrq); > +} > + > static int sdhci_sprd_voltage_switch(struct mmc_host *mmc, struct mmc_ios *ios) > { > struct sdhci_host *host = mmc_priv(mmc); > @@ -561,6 +582,11 @@ static int sdhci_sprd_probe(struct platform_device *pdev) > if (ret) > goto pltfm_free; > > + if (!mmc_card_is_removable(host->mmc)) > + host->mmc_host_ops.request_atomic = sdhci_sprd_request_atomic; > + else > + host->always_defer_done = true; > + > sprd_host = TO_SPRD_HOST(host); > sdhci_sprd_phy_param_parse(sprd_host, pdev->dev.of_node); > > @@ -654,8 +680,6 @@ static int sdhci_sprd_probe(struct platform_device *pdev) > if (ret) > goto err_cleanup_host; > > - host->always_defer_done = true; > - > ret = __sdhci_add_host(host); > if (ret) > goto err_cleanup_host; > -- > 1.9.1 > Kind regards Uffe ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [RESEND PATCH 2/3] mmc: host: sdhci-sprd: Implement the request_atomic() API 2020-03-16 11:47 ` Ulf Hansson @ 2020-03-17 2:02 ` Baolin Wang 0 siblings, 0 replies; 12+ messages in thread From: Baolin Wang @ 2020-03-17 2:02 UTC (permalink / raw) To: Ulf Hansson Cc: Adrian Hunter, Orson Zhai, Chunyan Zhang, Arnd Bergmann, linux-mmc, Linux Kernel Mailing List On Mon, Mar 16, 2020 at 7:47 PM Ulf Hansson <ulf.hansson@linaro.org> wrote: > > On Wed, 4 Mar 2020 at 08:42, Baolin Wang <baolin.wang7@gmail.com> wrote: > > > > Implement the request_atomic() API for nonremovable cards, that means > > we can submit next request in the irq hard handler context to reduce > > latency. > > > > Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> > > --- > > drivers/mmc/host/sdhci-sprd.c | 28 ++++++++++++++++++++++++++-- > > 1 file changed, 26 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/mmc/host/sdhci-sprd.c b/drivers/mmc/host/sdhci-sprd.c > > index 2ab42c5..bddf0f3 100644 > > --- a/drivers/mmc/host/sdhci-sprd.c > > +++ b/drivers/mmc/host/sdhci-sprd.c > > @@ -426,6 +426,27 @@ static void sdhci_sprd_request(struct mmc_host *mmc, struct mmc_request *mrq) > > sdhci_request(mmc, mrq); > > } > > > > +static void sdhci_sprd_request_atomic(struct mmc_host *mmc, > > + struct mmc_request *mrq) > > +{ > > + struct sdhci_host *host = mmc_priv(mmc); > > + struct sdhci_sprd_host *sprd_host = TO_SPRD_HOST(host); > > + > > + host->flags |= sprd_host->flags & SDHCI_AUTO_CMD23; > > + > > + /* > > + * From version 4.10 onward, ARGUMENT2 register is also as 32-bit > > + * block count register which doesn't support stuff bits of > > + * CMD23 argument on Spreadtrum's sd host controller. > > + */ > > + if (host->version >= SDHCI_SPEC_410 && > > + mrq->sbc && (mrq->sbc->arg & SDHCI_SPRD_ARG2_STUFF) && > > + (host->flags & SDHCI_AUTO_CMD23)) > > + host->flags &= ~SDHCI_AUTO_CMD23; > > Looks like the above code is a copy of what is done in > sdhci_sprd_request(). Perhaps add a helper function that deals with > this to avoid open coding? Sure. Will do in next version. Thanks. > > > + > > + sdhci_request_atomic(mmc, mrq); > > +} > > + > > static int sdhci_sprd_voltage_switch(struct mmc_host *mmc, struct mmc_ios *ios) > > { > > struct sdhci_host *host = mmc_priv(mmc); > > @@ -561,6 +582,11 @@ static int sdhci_sprd_probe(struct platform_device *pdev) > > if (ret) > > goto pltfm_free; > > > > + if (!mmc_card_is_removable(host->mmc)) > > + host->mmc_host_ops.request_atomic = sdhci_sprd_request_atomic; > > + else > > + host->always_defer_done = true; > > + > > sprd_host = TO_SPRD_HOST(host); > > sdhci_sprd_phy_param_parse(sprd_host, pdev->dev.of_node); > > > > @@ -654,8 +680,6 @@ static int sdhci_sprd_probe(struct platform_device *pdev) > > if (ret) > > goto err_cleanup_host; > > > > - host->always_defer_done = true; > > - > > ret = __sdhci_add_host(host); > > if (ret) > > goto err_cleanup_host; > > -- > > 1.9.1 > > > > Kind regards > Uffe -- Baolin Wang ^ permalink raw reply [flat|nested] 12+ messages in thread
* [RESEND PATCH 3/3] mmc: host: hsq: Support request_atomic() API 2020-03-04 7:41 [RESEND PATCH 0/3] Introduce the request_atomic() for the host Baolin Wang 2020-03-04 7:42 ` [RESEND PATCH 1/3] mmc: host: " Baolin Wang 2020-03-04 7:42 ` [RESEND PATCH 2/3] mmc: host: sdhci-sprd: Implement the request_atomic() API Baolin Wang @ 2020-03-04 7:42 ` Baolin Wang 2 siblings, 0 replies; 12+ messages in thread From: Baolin Wang @ 2020-03-04 7:42 UTC (permalink / raw) To: adrian.hunter, ulf.hansson Cc: orsonzhai, zhang.lyra, baolin.wang7, arnd, linux-mmc, linux-kernel Add support to submit a request by request_atomic() API. Signed-off-by: Baolin Wang <baolin.wang7@gmail.com> --- drivers/mmc/host/mmc_hsq.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/mmc/host/mmc_hsq.c b/drivers/mmc/host/mmc_hsq.c index 59d2776..fdbaa98 100644 --- a/drivers/mmc/host/mmc_hsq.c +++ b/drivers/mmc/host/mmc_hsq.c @@ -41,7 +41,10 @@ static void mmc_hsq_pump_requests(struct mmc_hsq *hsq) spin_unlock_irqrestore(&hsq->lock, flags); - mmc->ops->request(mmc, hsq->mrq); + if (mmc->ops->request_atomic) + mmc->ops->request_atomic(mmc, hsq->mrq); + else + mmc->ops->request(mmc, hsq->mrq); } static void mmc_hsq_update_next_tag(struct mmc_hsq *hsq, int remains) -- 1.9.1 ^ permalink raw reply related [flat|nested] 12+ messages in thread
end of thread, other threads:[~2020-03-17 9:17 UTC | newest] Thread overview: 12+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2020-03-04 7:41 [RESEND PATCH 0/3] Introduce the request_atomic() for the host Baolin Wang 2020-03-04 7:42 ` [RESEND PATCH 1/3] mmc: host: " Baolin Wang 2020-03-16 11:47 ` Ulf Hansson 2020-03-17 2:00 ` Baolin Wang 2020-03-16 13:08 ` Adrian Hunter 2020-03-17 3:36 ` Baolin Wang 2020-03-17 7:32 ` Adrian Hunter 2020-03-17 9:17 ` Baolin Wang 2020-03-04 7:42 ` [RESEND PATCH 2/3] mmc: host: sdhci-sprd: Implement the request_atomic() API Baolin Wang 2020-03-16 11:47 ` Ulf Hansson 2020-03-17 2:02 ` Baolin Wang 2020-03-04 7:42 ` [RESEND PATCH 3/3] mmc: host: hsq: Support " Baolin Wang
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).