* [PATCH 2/3] pata/at91: use new introduce smc accessor [not found] <20111208150348.GD23884@game.jcrosoft.org> @ 2011-12-08 15:23 ` Jean-Christophe PLAGNIOL-VILLARD 2011-12-08 22:34 ` Ryan Mallon 2011-12-09 10:59 ` Sergei Shtylyov 2011-12-08 15:23 ` [PATCH 3/3] ide/at91: " Jean-Christophe PLAGNIOL-VILLARD 1 sibling, 2 replies; 11+ messages in thread From: Jean-Christophe PLAGNIOL-VILLARD @ 2011-12-08 15:23 UTC (permalink / raw) To: linux-arm-kernel Cc: Nicolas Ferre, Jean-Christophe PLAGNIOL-VILLARD, linux-ide this will allow to use the pata_at91 on a single zImage Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com> Cc: Nicolas Ferre <nicolas.ferre@atmel.com> Cc: linux-ide@vger.kernel.org --- Hi, it's depends on other patch for AT91 can we apply via at91 Best Regards, J. drivers/ata/pata_at91.c | 43 +++++++++++++++++++++---------------------- 1 files changed, 21 insertions(+), 22 deletions(-) diff --git a/drivers/ata/pata_at91.c b/drivers/ata/pata_at91.c index 5249e6d..c8d1154 100644 --- a/drivers/ata/pata_at91.c +++ b/drivers/ata/pata_at91.c @@ -207,11 +207,11 @@ static void set_smc_timing(struct device *dev, struct ata_device *adev, { int ret = 0; int use_iordy; + struct sam9_smc_config smc; unsigned int t6z; /* data tristate time in ns */ unsigned int cycle; /* SMC Cycle width in MCK ticks */ unsigned int setup; /* SMC Setup width in MCK ticks */ unsigned int pulse; /* CFIOR and CFIOW pulse width in MCK ticks */ - unsigned int cs_setup = 0;/* CS4 or CS5 setup width in MCK ticks */ unsigned int cs_pulse; /* CS4 or CS5 pulse width in MCK ticks*/ unsigned int tdf_cycles; /* SMC TDF MCK ticks */ unsigned long mck_hz; /* MCK frequency in Hz */ @@ -244,26 +244,25 @@ static void set_smc_timing(struct device *dev, struct ata_device *adev, } dev_dbg(dev, "Use IORDY=%u, TDF Cycles=%u\n", use_iordy, tdf_cycles); - info->mode |= AT91_SMC_TDF_(tdf_cycles); /* write SMC Setup Register */ - at91_sys_write(AT91_SMC_SETUP(info->cs), - AT91_SMC_NWESETUP_(setup) | - AT91_SMC_NRDSETUP_(setup) | - AT91_SMC_NCS_WRSETUP_(cs_setup) | - AT91_SMC_NCS_RDSETUP_(cs_setup)); + smc.nrd_setup = setup; + smc.nwe_setup = smc.nrd_setup; + smc.ncs_read_setup = 0; + smc.ncs_write_setup = smc.ncs_read_setup; /* write SMC Pulse Register */ - at91_sys_write(AT91_SMC_PULSE(info->cs), - AT91_SMC_NWEPULSE_(pulse) | - AT91_SMC_NRDPULSE_(pulse) | - AT91_SMC_NCS_WRPULSE_(cs_pulse) | - AT91_SMC_NCS_RDPULSE_(cs_pulse)); + smc.nrd_pulse = pulse; + smc.nwe_pulse = smc.nrd_pulse; + smc.ncs_read_pulse =cs_pulse; + smc.ncs_write_pulse = smc.ncs_read_pulse; /* write SMC Cycle Register */ - at91_sys_write(AT91_SMC_CYCLE(info->cs), - AT91_SMC_NWECYCLE_(cycle) | - AT91_SMC_NRDCYCLE_(cycle)); + smc.read_cycle = cycle; + smc.write_cycle = smc.read_cycle; /* write SMC Mode Register*/ - at91_sys_write(AT91_SMC_MODE(info->cs), info->mode); + smc.tdf_cycles = tdf_cycles; + smc.mode = info->mode; + + sam9_smc_configure(0, info->cs, &smc); } static void pata_at91_set_piomode(struct ata_port *ap, struct ata_device *adev) @@ -288,20 +287,20 @@ static unsigned int pata_at91_data_xfer_noirq(struct ata_device *dev, struct at91_ide_info *info = dev->link->ap->host->private_data; unsigned int consumed; unsigned long flags; - unsigned int mode; + struct sam9_smc_config smc; local_irq_save(flags); - mode = at91_sys_read(AT91_SMC_MODE(info->cs)); + sam9_smc_read_mode(0, info->cs, &smc); /* set 16bit mode before writing data */ - at91_sys_write(AT91_SMC_MODE(info->cs), - (mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_16); + smc.mode = (smc.mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_16; + sam9_smc_write_mode(0, info->cs, &smc); consumed = ata_sff_data_xfer(dev, buf, buflen, rw); /* restore 8bit mode after data is written */ - at91_sys_write(AT91_SMC_MODE(info->cs), - (mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_8); + smc.mode = (smc.mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_8; + sam9_smc_write_mode(0, info->cs, &smc); local_irq_restore(flags); return consumed; -- 1.7.7 ^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [PATCH 2/3] pata/at91: use new introduce smc accessor 2011-12-08 15:23 ` [PATCH 2/3] pata/at91: use new introduce smc accessor Jean-Christophe PLAGNIOL-VILLARD @ 2011-12-08 22:34 ` Ryan Mallon 2011-12-09 6:24 ` Jean-Christophe PLAGNIOL-VILLARD 2011-12-09 10:59 ` Sergei Shtylyov 1 sibling, 1 reply; 11+ messages in thread From: Ryan Mallon @ 2011-12-08 22:34 UTC (permalink / raw) To: Jean-Christophe PLAGNIOL-VILLARD Cc: linux-arm-kernel, Nicolas Ferre, linux-ide On 09/12/11 02:23, Jean-Christophe PLAGNIOL-VILLARD wrote: > this will allow to use the pata_at91 on a single zImage > > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com> > Cc: Nicolas Ferre <nicolas.ferre@atmel.com> > Cc: linux-ide@vger.kernel.org Some comments below, ~Ryan > --- > Hi, > > it's depends on other patch for AT91 can we apply via at91 > > Best Regards, > J. > drivers/ata/pata_at91.c | 43 +++++++++++++++++++++---------------------- > 1 files changed, 21 insertions(+), 22 deletions(-) > > diff --git a/drivers/ata/pata_at91.c b/drivers/ata/pata_at91.c > index 5249e6d..c8d1154 100644 > --- a/drivers/ata/pata_at91.c > +++ b/drivers/ata/pata_at91.c > @@ -207,11 +207,11 @@ static void set_smc_timing(struct device *dev, struct ata_device *adev, > { > int ret = 0; > int use_iordy; > + struct sam9_smc_config smc; > unsigned int t6z; /* data tristate time in ns */ > unsigned int cycle; /* SMC Cycle width in MCK ticks */ > unsigned int setup; /* SMC Setup width in MCK ticks */ > unsigned int pulse; /* CFIOR and CFIOW pulse width in MCK ticks */ > - unsigned int cs_setup = 0;/* CS4 or CS5 setup width in MCK ticks */ > unsigned int cs_pulse; /* CS4 or CS5 pulse width in MCK ticks*/ > unsigned int tdf_cycles; /* SMC TDF MCK ticks */ > unsigned long mck_hz; /* MCK frequency in Hz */ > @@ -244,26 +244,25 @@ static void set_smc_timing(struct device *dev, struct ata_device *adev, > } > > dev_dbg(dev, "Use IORDY=%u, TDF Cycles=%u\n", use_iordy, tdf_cycles); > - info->mode |= AT91_SMC_TDF_(tdf_cycles); > > /* write SMC Setup Register */ > - at91_sys_write(AT91_SMC_SETUP(info->cs), > - AT91_SMC_NWESETUP_(setup) | > - AT91_SMC_NRDSETUP_(setup) | > - AT91_SMC_NCS_WRSETUP_(cs_setup) | > - AT91_SMC_NCS_RDSETUP_(cs_setup)); > + smc.nrd_setup = setup; > + smc.nwe_setup = smc.nrd_setup; > + smc.ncs_read_setup = 0; > + smc.ncs_write_setup = smc.ncs_read_setup; > /* write SMC Pulse Register */ > - at91_sys_write(AT91_SMC_PULSE(info->cs), > - AT91_SMC_NWEPULSE_(pulse) | > - AT91_SMC_NRDPULSE_(pulse) | > - AT91_SMC_NCS_WRPULSE_(cs_pulse) | > - AT91_SMC_NCS_RDPULSE_(cs_pulse)); > + smc.nrd_pulse = pulse; > + smc.nwe_pulse = smc.nrd_pulse; > + smc.ncs_read_pulse =cs_pulse; Nitpick: Whitespace around the = operator. > + smc.ncs_write_pulse = smc.ncs_read_pulse; > /* write SMC Cycle Register */ > - at91_sys_write(AT91_SMC_CYCLE(info->cs), > - AT91_SMC_NWECYCLE_(cycle) | > - AT91_SMC_NRDCYCLE_(cycle)); > + smc.read_cycle = cycle; > + smc.write_cycle = smc.read_cycle; > /* write SMC Mode Register*/ > - at91_sys_write(AT91_SMC_MODE(info->cs), info->mode); > + smc.tdf_cycles = tdf_cycles; The "write SMC Mode Register" comment should be removed. > + smc.mode = info->mode; > + > + sam9_smc_configure(0, info->cs, &smc); This new function returns an int, should we be checking the return value here? > } > > static void pata_at91_set_piomode(struct ata_port *ap, struct ata_device *adev) > @@ -288,20 +287,20 @@ static unsigned int pata_at91_data_xfer_noirq(struct ata_device *dev, > struct at91_ide_info *info = dev->link->ap->host->private_data; > unsigned int consumed; > unsigned long flags; > - unsigned int mode; > + struct sam9_smc_config smc; > > local_irq_save(flags); > - mode = at91_sys_read(AT91_SMC_MODE(info->cs)); > + sam9_smc_read_mode(0, info->cs, &smc); > > /* set 16bit mode before writing data */ > - at91_sys_write(AT91_SMC_MODE(info->cs), > - (mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_16); > + smc.mode = (smc.mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_16; > + sam9_smc_write_mode(0, info->cs, &smc); Do sam9_smc_read/write_mode really need to pass the whole smc structure? The only fields used are mode and tdf_cycles. It might be clearer to pass those directly. Also the original code here doesn't write tdf_cycles as part of the mode. Perhaps it would be better to have sam9_smc_write_mode to be: int sam9_smc_write_mode(int id, int cs, unsigned mode); and in set_smc_timing above explicitly or in the tdf_cycles bits? > > consumed = ata_sff_data_xfer(dev, buf, buflen, rw); > > /* restore 8bit mode after data is written */ > - at91_sys_write(AT91_SMC_MODE(info->cs), > - (mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_8); > + smc.mode = (smc.mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_8; > + sam9_smc_write_mode(0, info->cs, &smc); > > local_irq_restore(flags); > return consumed; ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH 2/3] pata/at91: use new introduce smc accessor 2011-12-08 22:34 ` Ryan Mallon @ 2011-12-09 6:24 ` Jean-Christophe PLAGNIOL-VILLARD 0 siblings, 0 replies; 11+ messages in thread From: Jean-Christophe PLAGNIOL-VILLARD @ 2011-12-09 6:24 UTC (permalink / raw) To: Ryan Mallon; +Cc: linux-ide, Nicolas Ferre, linux-arm-kernel On 09:34 Fri 09 Dec , Ryan Mallon wrote: > On 09/12/11 02:23, Jean-Christophe PLAGNIOL-VILLARD wrote: > > > this will allow to use the pata_at91 on a single zImage > > > > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com> > > Cc: Nicolas Ferre <nicolas.ferre@atmel.com> > > Cc: linux-ide@vger.kernel.org > > > Some comments below, > > ~Ryan > > > --- > > Hi, > > > > it's depends on other patch for AT91 can we apply via at91 > > > > Best Regards, > > J. > > drivers/ata/pata_at91.c | 43 +++++++++++++++++++++---------------------- > > 1 files changed, 21 insertions(+), 22 deletions(-) > > > > diff --git a/drivers/ata/pata_at91.c b/drivers/ata/pata_at91.c > > index 5249e6d..c8d1154 100644 > > --- a/drivers/ata/pata_at91.c > > +++ b/drivers/ata/pata_at91.c > > @@ -207,11 +207,11 @@ static void set_smc_timing(struct device *dev, struct ata_device *adev, > > { > > int ret = 0; > > int use_iordy; > > + struct sam9_smc_config smc; > > unsigned int t6z; /* data tristate time in ns */ > > unsigned int cycle; /* SMC Cycle width in MCK ticks */ > > unsigned int setup; /* SMC Setup width in MCK ticks */ > > unsigned int pulse; /* CFIOR and CFIOW pulse width in MCK ticks */ > > - unsigned int cs_setup = 0;/* CS4 or CS5 setup width in MCK ticks */ > > unsigned int cs_pulse; /* CS4 or CS5 pulse width in MCK ticks*/ > > unsigned int tdf_cycles; /* SMC TDF MCK ticks */ > > unsigned long mck_hz; /* MCK frequency in Hz */ > > @@ -244,26 +244,25 @@ static void set_smc_timing(struct device *dev, struct ata_device *adev, > > } > > > > dev_dbg(dev, "Use IORDY=%u, TDF Cycles=%u\n", use_iordy, tdf_cycles); > > - info->mode |= AT91_SMC_TDF_(tdf_cycles); > > > > /* write SMC Setup Register */ > > - at91_sys_write(AT91_SMC_SETUP(info->cs), > > - AT91_SMC_NWESETUP_(setup) | > > - AT91_SMC_NRDSETUP_(setup) | > > - AT91_SMC_NCS_WRSETUP_(cs_setup) | > > - AT91_SMC_NCS_RDSETUP_(cs_setup)); > > + smc.nrd_setup = setup; > > + smc.nwe_setup = smc.nrd_setup; > > + smc.ncs_read_setup = 0; > > + smc.ncs_write_setup = smc.ncs_read_setup; > > /* write SMC Pulse Register */ > > - at91_sys_write(AT91_SMC_PULSE(info->cs), > > - AT91_SMC_NWEPULSE_(pulse) | > > - AT91_SMC_NRDPULSE_(pulse) | > > - AT91_SMC_NCS_WRPULSE_(cs_pulse) | > > - AT91_SMC_NCS_RDPULSE_(cs_pulse)); > > + smc.nrd_pulse = pulse; > > + smc.nwe_pulse = smc.nrd_pulse; > > + smc.ncs_read_pulse =cs_pulse; > > > Nitpick: Whitespace around the = operator. > > > + smc.ncs_write_pulse = smc.ncs_read_pulse; > > /* write SMC Cycle Register */ > > - at91_sys_write(AT91_SMC_CYCLE(info->cs), > > - AT91_SMC_NWECYCLE_(cycle) | > > - AT91_SMC_NRDCYCLE_(cycle)); > > + smc.read_cycle = cycle; > > + smc.write_cycle = smc.read_cycle; > > /* write SMC Mode Register*/ > > - at91_sys_write(AT91_SMC_MODE(info->cs), info->mode); > > + smc.tdf_cycles = tdf_cycles; > > > The "write SMC Mode Register" comment should be removed. > > > + smc.mode = info->mode; > > + > > + sam9_smc_configure(0, info->cs, &smc); > > > This new function returns an int, should we be checking the return value > here? > > > } > > > > static void pata_at91_set_piomode(struct ata_port *ap, struct ata_device *adev) > > @@ -288,20 +287,20 @@ static unsigned int pata_at91_data_xfer_noirq(struct ata_device *dev, > > struct at91_ide_info *info = dev->link->ap->host->private_data; > > unsigned int consumed; > > unsigned long flags; > > - unsigned int mode; > > + struct sam9_smc_config smc; > > > > local_irq_save(flags); > > - mode = at91_sys_read(AT91_SMC_MODE(info->cs)); > > + sam9_smc_read_mode(0, info->cs, &smc); > > > > /* set 16bit mode before writing data */ > > - at91_sys_write(AT91_SMC_MODE(info->cs), > > - (mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_16); > > + smc.mode = (smc.mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_16; > > + sam9_smc_write_mode(0, info->cs, &smc); > > > Do sam9_smc_read/write_mode really need to pass the whole smc structure? > The only fields used are mode and tdf_cycles. It might be clearer to > pass those directly. > > Also the original code here doesn't write tdf_cycles as part of the > mode. Perhaps it would be better to have sam9_smc_write_mode to be: > > int sam9_smc_write_mode(int id, int cs, unsigned mode); no as you will force the write to read to content before updating it which I do not want the the need to you update the mode > > and in set_smc_timing above explicitly or in the tdf_cycles bits? the mode and tdf_cycles are closely related so I choose to manipulate them together and as it's supposed to be register independant you always pass the struct and the implemetation manage the write Best Regards, J. ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH 2/3] pata/at91: use new introduce smc accessor 2011-12-08 15:23 ` [PATCH 2/3] pata/at91: use new introduce smc accessor Jean-Christophe PLAGNIOL-VILLARD 2011-12-08 22:34 ` Ryan Mallon @ 2011-12-09 10:59 ` Sergei Shtylyov 2011-12-09 18:42 ` Jean-Christophe PLAGNIOL-VILLARD 1 sibling, 1 reply; 11+ messages in thread From: Sergei Shtylyov @ 2011-12-09 10:59 UTC (permalink / raw) To: Jean-Christophe PLAGNIOL-VILLARD Cc: linux-arm-kernel, Nicolas Ferre, linux-ide Hello. On 08-12-2011 19:23, Jean-Christophe PLAGNIOL-VILLARD wrote: Your subject doesn't parse. Maybe you meant "use newly introduced smc accessor"? > this will allow to use the pata_at91 on a single zImage > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD<plagnioj@jcrosoft.com> > Cc: Nicolas Ferre<nicolas.ferre@atmel.com> > Cc: linux-ide@vger.kernel.org > --- > Hi, > it's depends on other patch for AT91 can we apply via at91 > Best Regards, > J. > drivers/ata/pata_at91.c | 43 +++++++++++++++++++++---------------------- > 1 files changed, 21 insertions(+), 22 deletions(-) > diff --git a/drivers/ata/pata_at91.c b/drivers/ata/pata_at91.c > index 5249e6d..c8d1154 100644 > --- a/drivers/ata/pata_at91.c > +++ b/drivers/ata/pata_at91.c > @@ -207,11 +207,11 @@ static void set_smc_timing(struct device *dev, struct ata_device *adev, > { > int ret = 0; > int use_iordy; > + struct sam9_smc_config smc; > unsigned int t6z; /* data tristate time in ns */ > unsigned int cycle; /* SMC Cycle width in MCK ticks */ > unsigned int setup; /* SMC Setup width in MCK ticks */ > unsigned int pulse; /* CFIOR and CFIOW pulse width in MCK ticks */ > - unsigned int cs_setup = 0;/* CS4 or CS5 setup width in MCK ticks */ > unsigned int cs_pulse; /* CS4 or CS5 pulse width in MCK ticks*/ > unsigned int tdf_cycles; /* SMC TDF MCK ticks */ > unsigned long mck_hz; /* MCK frequency in Hz */ > @@ -244,26 +244,25 @@ static void set_smc_timing(struct device *dev, struct ata_device *adev, > } > > dev_dbg(dev, "Use IORDY=%u, TDF Cycles=%u\n", use_iordy, tdf_cycles); > - info->mode |= AT91_SMC_TDF_(tdf_cycles); > > /* write SMC Setup Register */ > - at91_sys_write(AT91_SMC_SETUP(info->cs), > - AT91_SMC_NWESETUP_(setup) | > - AT91_SMC_NRDSETUP_(setup) | > - AT91_SMC_NCS_WRSETUP_(cs_setup) | > - AT91_SMC_NCS_RDSETUP_(cs_setup)); > + smc.nrd_setup = setup; > + smc.nwe_setup = smc.nrd_setup; > + smc.ncs_read_setup = 0; > + smc.ncs_write_setup = smc.ncs_read_setup; Why not: smc.nrd_setup = smc.nwe_setup = setup; smc.ncs_read_setup = smc.ncs_write_setup = 0; > /* write SMC Pulse Register */ > - at91_sys_write(AT91_SMC_PULSE(info->cs), > - AT91_SMC_NWEPULSE_(pulse) | > - AT91_SMC_NRDPULSE_(pulse) | > - AT91_SMC_NCS_WRPULSE_(cs_pulse) | > - AT91_SMC_NCS_RDPULSE_(cs_pulse)); > + smc.nrd_pulse = pulse; > + smc.nwe_pulse = smc.nrd_pulse; > + smc.ncs_read_pulse =cs_pulse; Add space after =, please. > + smc.ncs_write_pulse = smc.ncs_read_pulse; Why not: smc.nrd_pulse = smc.nwe_pulse = pulse; smc.ncs_read_pulse = smc.ncs_write_pulse = cs_pulse; > /* write SMC Cycle Register */ > - at91_sys_write(AT91_SMC_CYCLE(info->cs), > - AT91_SMC_NWECYCLE_(cycle) | > - AT91_SMC_NRDCYCLE_(cycle)); > + smc.read_cycle = cycle; > + smc.write_cycle = smc.read_cycle; Why not: smc.read_cycle = smc.write_cycle = cycle; WBR, Sergei ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH 2/3] pata/at91: use new introduce smc accessor 2011-12-09 10:59 ` Sergei Shtylyov @ 2011-12-09 18:42 ` Jean-Christophe PLAGNIOL-VILLARD 0 siblings, 0 replies; 11+ messages in thread From: Jean-Christophe PLAGNIOL-VILLARD @ 2011-12-09 18:42 UTC (permalink / raw) To: Sergei Shtylyov; +Cc: linux-arm-kernel, Nicolas Ferre, linux-ide On 14:59 Fri 09 Dec , Sergei Shtylyov wrote: > Hello. > > On 08-12-2011 19:23, Jean-Christophe PLAGNIOL-VILLARD wrote: > > Your subject doesn't parse. Maybe you meant "use newly introduced > smc accessor"? yeap > > >this will allow to use the pata_at91 on a single zImage > > >Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD<plagnioj@jcrosoft.com> > >Cc: Nicolas Ferre<nicolas.ferre@atmel.com> > >Cc: linux-ide@vger.kernel.org > >--- > >Hi, > > > it's depends on other patch for AT91 can we apply via at91 > > >Best Regards, > >J. > > drivers/ata/pata_at91.c | 43 +++++++++++++++++++++---------------------- > > 1 files changed, 21 insertions(+), 22 deletions(-) > > >diff --git a/drivers/ata/pata_at91.c b/drivers/ata/pata_at91.c > >index 5249e6d..c8d1154 100644 > >--- a/drivers/ata/pata_at91.c > >+++ b/drivers/ata/pata_at91.c > >@@ -207,11 +207,11 @@ static void set_smc_timing(struct device *dev, struct ata_device *adev, > > { > > int ret = 0; > > int use_iordy; > >+ struct sam9_smc_config smc; > > unsigned int t6z; /* data tristate time in ns */ > > unsigned int cycle; /* SMC Cycle width in MCK ticks */ > > unsigned int setup; /* SMC Setup width in MCK ticks */ > > unsigned int pulse; /* CFIOR and CFIOW pulse width in MCK ticks */ > >- unsigned int cs_setup = 0;/* CS4 or CS5 setup width in MCK ticks */ > > unsigned int cs_pulse; /* CS4 or CS5 pulse width in MCK ticks*/ > > unsigned int tdf_cycles; /* SMC TDF MCK ticks */ > > unsigned long mck_hz; /* MCK frequency in Hz */ > >@@ -244,26 +244,25 @@ static void set_smc_timing(struct device *dev, struct ata_device *adev, > > } > > > > dev_dbg(dev, "Use IORDY=%u, TDF Cycles=%u\n", use_iordy, tdf_cycles); > >- info->mode |= AT91_SMC_TDF_(tdf_cycles); > > > > /* write SMC Setup Register */ > >- at91_sys_write(AT91_SMC_SETUP(info->cs), > >- AT91_SMC_NWESETUP_(setup) | > >- AT91_SMC_NRDSETUP_(setup) | > >- AT91_SMC_NCS_WRSETUP_(cs_setup) | > >- AT91_SMC_NCS_RDSETUP_(cs_setup)); > >+ smc.nrd_setup = setup; > >+ smc.nwe_setup = smc.nrd_setup; > >+ smc.ncs_read_setup = 0; > >+ smc.ncs_write_setup = smc.ncs_read_setup; > > Why not: > > smc.nrd_setup = smc.nwe_setup = setup; > smc.ncs_read_setup = smc.ncs_write_setup = 0; why not Best Regards, J. ^ permalink raw reply [flat|nested] 11+ messages in thread
* [PATCH 3/3] ide/at91: use new introduce smc accessor [not found] <20111208150348.GD23884@game.jcrosoft.org> 2011-12-08 15:23 ` [PATCH 2/3] pata/at91: use new introduce smc accessor Jean-Christophe PLAGNIOL-VILLARD @ 2011-12-08 15:23 ` Jean-Christophe PLAGNIOL-VILLARD 2011-12-08 22:38 ` Ryan Mallon ` (2 more replies) 1 sibling, 3 replies; 11+ messages in thread From: Jean-Christophe PLAGNIOL-VILLARD @ 2011-12-08 15:23 UTC (permalink / raw) To: linux-arm-kernel Cc: Nicolas Ferre, Jean-Christophe PLAGNIOL-VILLARD, linux-ide this will allow to use the pata_at91 on a single zImage Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com> Cc: Nicolas Ferre <nicolas.ferre@atmel.com> Cc: linux-ide@vger.kernel.org --- Hi, it's depends on other patch for AT91 can we apply via at91 Best Regards, J. drivers/ide/at91_ide.c | 65 +++++++++++++++++++++++++++-------------------- 1 files changed, 37 insertions(+), 28 deletions(-) diff --git a/drivers/ide/at91_ide.c b/drivers/ide/at91_ide.c index 41d4155..407595b 100644 --- a/drivers/ide/at91_ide.c +++ b/drivers/ide/at91_ide.c @@ -59,41 +59,50 @@ #define ALT_MODE 0x00e00000 #define REGS_SIZE 8 -#define enter_16bit(cs, mode) do { \ - mode = at91_sys_read(AT91_SMC_MODE(cs)); \ - at91_sys_write(AT91_SMC_MODE(cs), mode | AT91_SMC_DBW_16); \ -} while (0) +static inline void enter_16bit(int cs, struct sam9_smc_config *smc) +{ + sam9_smc_read_mode(0, cs, smc); + smc->mode = (smc->mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_16; + sam9_smc_write_mode(0, cs, smc); +} -#define leave_16bit(cs, mode) at91_sys_write(AT91_SMC_MODE(cs), mode); +static inline void leave_16bit(int cs, struct sam9_smc_config *smc) +{ + smc->mode = (smc->mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_8; + sam9_smc_write_mode(0, cs, smc); +} static void set_smc_timings(const u8 chipselect, const u16 cycle, const u16 setup, const u16 pulse, const u16 data_float, int use_iordy) { - unsigned long mode = AT91_SMC_READMODE | AT91_SMC_WRITEMODE | + struct sam9_smc_config smc; + + smc.mode = AT91_SMC_READMODE | AT91_SMC_WRITEMODE | AT91_SMC_BAT_SELECT; /* disable or enable waiting for IORDY signal */ if (use_iordy) - mode |= AT91_SMC_EXNWMODE_READY; + smc.mode |= AT91_SMC_EXNWMODE_READY; - /* add data float cycles if needed */ - if (data_float) - mode |= AT91_SMC_TDF_(data_float); + /* add data ofloat cycles if needed */ + smc.tdf_cycles = data_float; - at91_sys_write(AT91_SMC_MODE(chipselect), mode); + /* write SMC Setup Register */ + smc.nrd_setup = setup; + smc.nwe_setup = smc.nrd_setup; + smc.ncs_read_setup = 0; + smc.ncs_write_setup = smc.ncs_read_setup; + /* write SMC Pulse Register */ + smc.nrd_pulse = pulse; + smc.nwe_pulse = smc.nrd_pulse; + smc.ncs_read_pulse = cycle; + smc.ncs_write_pulse = smc.ncs_read_pulse; + /* write SMC Cycle Register */ + smc.read_cycle = cycle; + smc.write_cycle = smc.read_cycle; - /* setup timings in SMC */ - at91_sys_write(AT91_SMC_SETUP(chipselect), AT91_SMC_NWESETUP_(setup) | - AT91_SMC_NCS_WRSETUP_(0) | - AT91_SMC_NRDSETUP_(setup) | - AT91_SMC_NCS_RDSETUP_(0)); - at91_sys_write(AT91_SMC_PULSE(chipselect), AT91_SMC_NWEPULSE_(pulse) | - AT91_SMC_NCS_WRPULSE_(cycle) | - AT91_SMC_NRDPULSE_(pulse) | - AT91_SMC_NCS_RDPULSE_(cycle)); - at91_sys_write(AT91_SMC_CYCLE(chipselect), AT91_SMC_NWECYCLE_(cycle) | - AT91_SMC_NRDCYCLE_(cycle)); + sam9_smc_configure(0, chipselect, &smc); } static unsigned int calc_mck_cycles(unsigned int ns, unsigned int mck_hz) @@ -146,15 +155,15 @@ static void at91_ide_input_data(ide_drive_t *drive, struct ide_cmd *cmd, ide_hwif_t *hwif = drive->hwif; struct ide_io_ports *io_ports = &hwif->io_ports; u8 chipselect = hwif->select_data; - unsigned long mode; + struct sam9_smc_config smc; pdbg("cs %u buf %p len %d\n", chipselect, buf, len); len++; - enter_16bit(chipselect, mode); + enter_16bit(chipselect, &smc); readsw((void __iomem *)io_ports->data_addr, buf, len / 2); - leave_16bit(chipselect, mode); + leave_16bit(chipselect, &smc); } static void at91_ide_output_data(ide_drive_t *drive, struct ide_cmd *cmd, @@ -163,13 +172,13 @@ static void at91_ide_output_data(ide_drive_t *drive, struct ide_cmd *cmd, ide_hwif_t *hwif = drive->hwif; struct ide_io_ports *io_ports = &hwif->io_ports; u8 chipselect = hwif->select_data; - unsigned long mode; + struct sam9_smc_config smc; pdbg("cs %u buf %p len %d\n", chipselect, buf, len); - enter_16bit(chipselect, mode); + enter_16bit(chipselect, &smc); writesw((void __iomem *)io_ports->data_addr, buf, len / 2); - leave_16bit(chipselect, mode); + leave_16bit(chipselect, &smc); } static void at91_ide_set_pio_mode(ide_hwif_t *hwif, ide_drive_t *drive) -- 1.7.7 ^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [PATCH 3/3] ide/at91: use new introduce smc accessor 2011-12-08 15:23 ` [PATCH 3/3] ide/at91: " Jean-Christophe PLAGNIOL-VILLARD @ 2011-12-08 22:38 ` Ryan Mallon 2011-12-09 6:19 ` Jean-Christophe PLAGNIOL-VILLARD 2011-12-09 11:06 ` Sergei Shtylyov 2011-12-09 13:59 ` Arnd Bergmann 2 siblings, 1 reply; 11+ messages in thread From: Ryan Mallon @ 2011-12-08 22:38 UTC (permalink / raw) To: Jean-Christophe PLAGNIOL-VILLARD Cc: linux-arm-kernel, Nicolas Ferre, linux-ide On 09/12/11 02:23, Jean-Christophe PLAGNIOL-VILLARD wrote: > this will allow to use the pata_at91 on a single zImage > > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com> > Cc: Nicolas Ferre <nicolas.ferre@atmel.com> > Cc: linux-ide@vger.kernel.org Couple of minor comments below, ~Ryan > --- > Hi, > > it's depends on other patch for AT91 can we apply via at91 > > Best Regards, > J. > drivers/ide/at91_ide.c | 65 +++++++++++++++++++++++++++-------------------- > 1 files changed, 37 insertions(+), 28 deletions(-) > > diff --git a/drivers/ide/at91_ide.c b/drivers/ide/at91_ide.c > index 41d4155..407595b 100644 > --- a/drivers/ide/at91_ide.c > +++ b/drivers/ide/at91_ide.c > @@ -59,41 +59,50 @@ > #define ALT_MODE 0x00e00000 > #define REGS_SIZE 8 > > -#define enter_16bit(cs, mode) do { \ > - mode = at91_sys_read(AT91_SMC_MODE(cs)); \ > - at91_sys_write(AT91_SMC_MODE(cs), mode | AT91_SMC_DBW_16); \ > -} while (0) > +static inline void enter_16bit(int cs, struct sam9_smc_config *smc) > +{ > + sam9_smc_read_mode(0, cs, smc); > + smc->mode = (smc->mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_16; > + sam9_smc_write_mode(0, cs, smc); > +} > > -#define leave_16bit(cs, mode) at91_sys_write(AT91_SMC_MODE(cs), mode); > +static inline void leave_16bit(int cs, struct sam9_smc_config *smc) > +{ > + smc->mode = (smc->mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_8; > + sam9_smc_write_mode(0, cs, smc); > +} > > static void set_smc_timings(const u8 chipselect, const u16 cycle, > const u16 setup, const u16 pulse, > const u16 data_float, int use_iordy) > { > - unsigned long mode = AT91_SMC_READMODE | AT91_SMC_WRITEMODE | > + struct sam9_smc_config smc; > + > + smc.mode = AT91_SMC_READMODE | AT91_SMC_WRITEMODE | > AT91_SMC_BAT_SELECT; > > /* disable or enable waiting for IORDY signal */ > if (use_iordy) > - mode |= AT91_SMC_EXNWMODE_READY; > + smc.mode |= AT91_SMC_EXNWMODE_READY; > > - /* add data float cycles if needed */ > - if (data_float) > - mode |= AT91_SMC_TDF_(data_float); > + /* add data ofloat cycles if needed */ Typo: ofloat -> float? > + smc.tdf_cycles = data_float; > > - at91_sys_write(AT91_SMC_MODE(chipselect), mode); > + /* write SMC Setup Register */ This comment is no longer correct. > + smc.nrd_setup = setup; > + smc.nwe_setup = smc.nrd_setup; > + smc.ncs_read_setup = 0; > + smc.ncs_write_setup = smc.ncs_read_setup; > + /* write SMC Pulse Register */ Nor is this one. > + smc.nrd_pulse = pulse; > + smc.nwe_pulse = smc.nrd_pulse; > + smc.ncs_read_pulse = cycle; > + smc.ncs_write_pulse = smc.ncs_read_pulse; > + /* write SMC Cycle Register */ or this one. > + smc.read_cycle = cycle; > + smc.write_cycle = smc.read_cycle; > > - /* setup timings in SMC */ > - at91_sys_write(AT91_SMC_SETUP(chipselect), AT91_SMC_NWESETUP_(setup) | > - AT91_SMC_NCS_WRSETUP_(0) | > - AT91_SMC_NRDSETUP_(setup) | > - AT91_SMC_NCS_RDSETUP_(0)); > - at91_sys_write(AT91_SMC_PULSE(chipselect), AT91_SMC_NWEPULSE_(pulse) | > - AT91_SMC_NCS_WRPULSE_(cycle) | > - AT91_SMC_NRDPULSE_(pulse) | > - AT91_SMC_NCS_RDPULSE_(cycle)); > - at91_sys_write(AT91_SMC_CYCLE(chipselect), AT91_SMC_NWECYCLE_(cycle) | > - AT91_SMC_NRDCYCLE_(cycle)); > + sam9_smc_configure(0, chipselect, &smc); Check return value? > } > > static unsigned int calc_mck_cycles(unsigned int ns, unsigned int mck_hz) > @@ -146,15 +155,15 @@ static void at91_ide_input_data(ide_drive_t *drive, struct ide_cmd *cmd, > ide_hwif_t *hwif = drive->hwif; > struct ide_io_ports *io_ports = &hwif->io_ports; > u8 chipselect = hwif->select_data; > - unsigned long mode; > + struct sam9_smc_config smc; > > pdbg("cs %u buf %p len %d\n", chipselect, buf, len); > > len++; > > - enter_16bit(chipselect, mode); > + enter_16bit(chipselect, &smc); > readsw((void __iomem *)io_ports->data_addr, buf, len / 2); > - leave_16bit(chipselect, mode); > + leave_16bit(chipselect, &smc); > } > > static void at91_ide_output_data(ide_drive_t *drive, struct ide_cmd *cmd, > @@ -163,13 +172,13 @@ static void at91_ide_output_data(ide_drive_t *drive, struct ide_cmd *cmd, > ide_hwif_t *hwif = drive->hwif; > struct ide_io_ports *io_ports = &hwif->io_ports; > u8 chipselect = hwif->select_data; > - unsigned long mode; > + struct sam9_smc_config smc; > > pdbg("cs %u buf %p len %d\n", chipselect, buf, len); > > - enter_16bit(chipselect, mode); > + enter_16bit(chipselect, &smc); > writesw((void __iomem *)io_ports->data_addr, buf, len / 2); > - leave_16bit(chipselect, mode); > + leave_16bit(chipselect, &smc); > } > > static void at91_ide_set_pio_mode(ide_hwif_t *hwif, ide_drive_t *drive) ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH 3/3] ide/at91: use new introduce smc accessor 2011-12-08 22:38 ` Ryan Mallon @ 2011-12-09 6:19 ` Jean-Christophe PLAGNIOL-VILLARD 0 siblings, 0 replies; 11+ messages in thread From: Jean-Christophe PLAGNIOL-VILLARD @ 2011-12-09 6:19 UTC (permalink / raw) To: Ryan Mallon; +Cc: linux-ide, Nicolas Ferre, linux-arm-kernel On 09:38 Fri 09 Dec , Ryan Mallon wrote: > On 09/12/11 02:23, Jean-Christophe PLAGNIOL-VILLARD wrote: > > > this will allow to use the pata_at91 on a single zImage > > > > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com> > > Cc: Nicolas Ferre <nicolas.ferre@atmel.com> > > Cc: linux-ide@vger.kernel.org > > > Couple of minor comments below, > > ~Ryan > > > --- > > Hi, > > > > it's depends on other patch for AT91 can we apply via at91 > > > > Best Regards, > > J. > > drivers/ide/at91_ide.c | 65 +++++++++++++++++++++++++++-------------------- > > 1 files changed, 37 insertions(+), 28 deletions(-) > > > > diff --git a/drivers/ide/at91_ide.c b/drivers/ide/at91_ide.c > > index 41d4155..407595b 100644 > > --- a/drivers/ide/at91_ide.c > > +++ b/drivers/ide/at91_ide.c > > @@ -59,41 +59,50 @@ > > #define ALT_MODE 0x00e00000 > > #define REGS_SIZE 8 > > > > -#define enter_16bit(cs, mode) do { \ > > - mode = at91_sys_read(AT91_SMC_MODE(cs)); \ > > - at91_sys_write(AT91_SMC_MODE(cs), mode | AT91_SMC_DBW_16); \ > > -} while (0) > > +static inline void enter_16bit(int cs, struct sam9_smc_config *smc) > > +{ > > + sam9_smc_read_mode(0, cs, smc); > > + smc->mode = (smc->mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_16; > > + sam9_smc_write_mode(0, cs, smc); > > +} > > > > -#define leave_16bit(cs, mode) at91_sys_write(AT91_SMC_MODE(cs), mode); > > +static inline void leave_16bit(int cs, struct sam9_smc_config *smc) > > +{ > > + smc->mode = (smc->mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_8; > > + sam9_smc_write_mode(0, cs, smc); > > +} > > > > static void set_smc_timings(const u8 chipselect, const u16 cycle, > > const u16 setup, const u16 pulse, > > const u16 data_float, int use_iordy) > > { > > - unsigned long mode = AT91_SMC_READMODE | AT91_SMC_WRITEMODE | > > + struct sam9_smc_config smc; > > + > > + smc.mode = AT91_SMC_READMODE | AT91_SMC_WRITEMODE | > > AT91_SMC_BAT_SELECT; > > > > /* disable or enable waiting for IORDY signal */ > > if (use_iordy) > > - mode |= AT91_SMC_EXNWMODE_READY; > > + smc.mode |= AT91_SMC_EXNWMODE_READY; > > > > - /* add data float cycles if needed */ > > - if (data_float) > > - mode |= AT91_SMC_TDF_(data_float); > > + /* add data ofloat cycles if needed */ > > > Typo: ofloat -> float? > > > + smc.tdf_cycles = data_float; > > > > - at91_sys_write(AT91_SMC_MODE(chipselect), mode); > > + /* write SMC Setup Register */ > > > This comment is no longer correct. > > > + smc.nrd_setup = setup; > > + smc.nwe_setup = smc.nrd_setup; > > + smc.ncs_read_setup = 0; > > + smc.ncs_write_setup = smc.ncs_read_setup; > > + /* write SMC Pulse Register */ > > > Nor is this one. > > > + smc.nrd_pulse = pulse; > > + smc.nwe_pulse = smc.nrd_pulse; > > + smc.ncs_read_pulse = cycle; > > + smc.ncs_write_pulse = smc.ncs_read_pulse; > > + /* write SMC Cycle Register */ > > > or this one. > > > + smc.read_cycle = cycle; > > + smc.write_cycle = smc.read_cycle; > > > > - /* setup timings in SMC */ > > - at91_sys_write(AT91_SMC_SETUP(chipselect), AT91_SMC_NWESETUP_(setup) | > > - AT91_SMC_NCS_WRSETUP_(0) | > > - AT91_SMC_NRDSETUP_(setup) | > > - AT91_SMC_NCS_RDSETUP_(0)); > > - at91_sys_write(AT91_SMC_PULSE(chipselect), AT91_SMC_NWEPULSE_(pulse) | > > - AT91_SMC_NCS_WRPULSE_(cycle) | > > - AT91_SMC_NRDPULSE_(pulse) | > > - AT91_SMC_NCS_RDPULSE_(cycle)); > > - at91_sys_write(AT91_SMC_CYCLE(chipselect), AT91_SMC_NWECYCLE_(cycle) | > > - AT91_SMC_NRDCYCLE_(cycle)); > > + sam9_smc_configure(0, chipselect, &smc); > > > Check return value? set_timings don't care about it so no need Best Regards, J. ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH 3/3] ide/at91: use new introduce smc accessor 2011-12-08 15:23 ` [PATCH 3/3] ide/at91: " Jean-Christophe PLAGNIOL-VILLARD 2011-12-08 22:38 ` Ryan Mallon @ 2011-12-09 11:06 ` Sergei Shtylyov 2011-12-09 18:40 ` Jean-Christophe PLAGNIOL-VILLARD 2011-12-09 13:59 ` Arnd Bergmann 2 siblings, 1 reply; 11+ messages in thread From: Sergei Shtylyov @ 2011-12-09 11:06 UTC (permalink / raw) To: Jean-Christophe PLAGNIOL-VILLARD Cc: linux-arm-kernel, Nicolas Ferre, linux-ide Hello. On 08-12-2011 19:23, Jean-Christophe PLAGNIOL-VILLARD wrote: Same comment about the subject. > this will allow to use the pata_at91 on a single zImage s/pata_at91/at91_ide/ > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD<plagnioj@jcrosoft.com> > Cc: Nicolas Ferre<nicolas.ferre@atmel.com> > Cc: linux-ide@vger.kernel.org > --- > Hi, > it's depends on other patch for AT91 can we apply via at91 I'm afraid David won't accept this, as the IDE tree is frozen for anything but the fixes. > diff --git a/drivers/ide/at91_ide.c b/drivers/ide/at91_ide.c > index 41d4155..407595b 100644 > --- a/drivers/ide/at91_ide.c > +++ b/drivers/ide/at91_ide.c > @@ -59,41 +59,50 @@ [...] > static void set_smc_timings(const u8 chipselect, const u16 cycle, > const u16 setup, const u16 pulse, > const u16 data_float, int use_iordy) > { > - unsigned long mode = AT91_SMC_READMODE | AT91_SMC_WRITEMODE | > + struct sam9_smc_config smc; > + > + smc.mode = AT91_SMC_READMODE | AT91_SMC_WRITEMODE | > AT91_SMC_BAT_SELECT; > > /* disable or enable waiting for IORDY signal */ > if (use_iordy) > - mode |= AT91_SMC_EXNWMODE_READY; > + smc.mode |= AT91_SMC_EXNWMODE_READY; > > - /* add data float cycles if needed */ > - if (data_float) > - mode |= AT91_SMC_TDF_(data_float); > + /* add data ofloat cycles if needed */ > + smc.tdf_cycles = data_float; > > - at91_sys_write(AT91_SMC_MODE(chipselect), mode); > + /* write SMC Setup Register */ > + smc.nrd_setup = setup; > + smc.nwe_setup = smc.nrd_setup; > + smc.ncs_read_setup = 0; > + smc.ncs_write_setup = smc.ncs_read_setup; smc.nrd_setup = smc.nwe_setup = setup; smc.ncs_read_setup = smc.ncs_write_setup = 0; > + /* write SMC Pulse Register */ > + smc.nrd_pulse = pulse; > + smc.nwe_pulse = smc.nrd_pulse; > + smc.ncs_read_pulse = cycle; > + smc.ncs_write_pulse = smc.ncs_read_pulse; smc.nrd_pulse = smc.nwe_pulse = pulse; > + smc.ncs_read_pulse = smc.ncs_write_pulse = cycle; > + /* write SMC Cycle Register */ > + smc.read_cycle = cycle; > + smc.write_cycle = smc.read_cycle; smc.read_cycle = smc.write_cycle = cycle; WBR, Sergei ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH 3/3] ide/at91: use new introduce smc accessor 2011-12-09 11:06 ` Sergei Shtylyov @ 2011-12-09 18:40 ` Jean-Christophe PLAGNIOL-VILLARD 0 siblings, 0 replies; 11+ messages in thread From: Jean-Christophe PLAGNIOL-VILLARD @ 2011-12-09 18:40 UTC (permalink / raw) To: Sergei Shtylyov; +Cc: linux-ide, Nicolas Ferre, linux-arm-kernel On 15:06 Fri 09 Dec , Sergei Shtylyov wrote: > Hello. > > On 08-12-2011 19:23, Jean-Christophe PLAGNIOL-VILLARD wrote: > > Same comment about the subject. > > >this will allow to use the pata_at91 on a single zImage > > s/pata_at91/at91_ide/ > > >Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD<plagnioj@jcrosoft.com> > >Cc: Nicolas Ferre<nicolas.ferre@atmel.com> > >Cc: linux-ide@vger.kernel.org > >--- > >Hi, > > > it's depends on other patch for AT91 can we apply via at91 > > I'm afraid David won't accept this, as the IDE tree is frozen for > anything but the fixes. basicaly it's a fix on at91 not ide. we can not compile all at91 together because of the at91_sys_read/write so I need to fix it on at91 this one of the place I need to touch. I'll not touch it otherwise Best Regards, J. ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH 3/3] ide/at91: use new introduce smc accessor 2011-12-08 15:23 ` [PATCH 3/3] ide/at91: " Jean-Christophe PLAGNIOL-VILLARD 2011-12-08 22:38 ` Ryan Mallon 2011-12-09 11:06 ` Sergei Shtylyov @ 2011-12-09 13:59 ` Arnd Bergmann 2 siblings, 0 replies; 11+ messages in thread From: Arnd Bergmann @ 2011-12-09 13:59 UTC (permalink / raw) To: linux-arm-kernel Cc: Jean-Christophe PLAGNIOL-VILLARD, Nicolas Ferre, linux-ide On Thursday 08 December 2011, Jean-Christophe PLAGNIOL-VILLARD wrote: > this will allow to use the pata_at91 on a single zImage > > J. > drivers/ide/at91_ide.c | 65 +++++++++++++++++++++++++++-------------------- > 1 files changed, 37 insertions(+), 28 deletions(-) The description doesn't match the file you are changing. Is there still a strong reason to keep using the at91_ide driver instead of migrating everybody to pata_at91 now? Arnd ^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2011-12-09 18:46 UTC | newest]
Thread overview: 11+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20111208150348.GD23884@game.jcrosoft.org>
2011-12-08 15:23 ` [PATCH 2/3] pata/at91: use new introduce smc accessor Jean-Christophe PLAGNIOL-VILLARD
2011-12-08 22:34 ` Ryan Mallon
2011-12-09 6:24 ` Jean-Christophe PLAGNIOL-VILLARD
2011-12-09 10:59 ` Sergei Shtylyov
2011-12-09 18:42 ` Jean-Christophe PLAGNIOL-VILLARD
2011-12-08 15:23 ` [PATCH 3/3] ide/at91: " Jean-Christophe PLAGNIOL-VILLARD
2011-12-08 22:38 ` Ryan Mallon
2011-12-09 6:19 ` Jean-Christophe PLAGNIOL-VILLARD
2011-12-09 11:06 ` Sergei Shtylyov
2011-12-09 18:40 ` Jean-Christophe PLAGNIOL-VILLARD
2011-12-09 13:59 ` Arnd Bergmann
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).