| Message ID | 20171218000021.16655-1-wsa+renesas@sang-engineering.com (mailing list archive) |
|---|---|
| State | Accepted |
| Delegated to: | Geert Uytterhoeven |
| Headers | show |
On 18/12/17 00:00, Wolfram Sang wrote: > Not all archs define reads* and writes*. Switch to ioread*_rep and > friends which is defined everywhere, so we can enable COMPILE_TEST after > that. > > Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com> > --- > > So, I pushed this to buildbot on top of Yamada-san's patch series and there > were no complaints, even with COMPILE_TEST enabled. I also did some tests on > HW and checksuming huge files on SD cards still works. > > However, I am not sure about this mixture of read* and ioread* functions. Shall > we convert maybe all of those? Might as well. They've been in Linux since what 2.6.9? > drivers/mmc/host/tmio_mmc.h | 8 ++++---- > 1 file changed, 4 insertions(+), 4 deletions(-) > > diff --git a/drivers/mmc/host/tmio_mmc.h b/drivers/mmc/host/tmio_mmc.h > index 76094345cbacf3..03519c4ca0aa1a 100644 > --- a/drivers/mmc/host/tmio_mmc.h > +++ b/drivers/mmc/host/tmio_mmc.h > @@ -233,7 +233,7 @@ static inline u16 sd_ctrl_read16(struct tmio_mmc_host *host, int addr) > static inline void sd_ctrl_read16_rep(struct tmio_mmc_host *host, int addr, > u16 *buf, int count) > { > - readsw(host->ctl + (addr << host->bus_shift), buf, count); > + ioread16_rep(host->ctl + (addr << host->bus_shift), buf, count); > } > > static inline u32 sd_ctrl_read16_and_16_as_32(struct tmio_mmc_host *host, > @@ -246,7 +246,7 @@ static inline u32 sd_ctrl_read16_and_16_as_32(struct tmio_mmc_host *host, > static inline void sd_ctrl_read32_rep(struct tmio_mmc_host *host, int addr, > u32 *buf, int count) > { > - readsl(host->ctl + (addr << host->bus_shift), buf, count); > + ioread32_rep(host->ctl + (addr << host->bus_shift), buf, count); > } > > static inline void sd_ctrl_write16(struct tmio_mmc_host *host, int addr, > @@ -263,7 +263,7 @@ static inline void sd_ctrl_write16(struct tmio_mmc_host *host, int addr, > static inline void sd_ctrl_write16_rep(struct tmio_mmc_host *host, int addr, > u16 *buf, int count) > { > - writesw(host->ctl + (addr << host->bus_shift), buf, count); > + iowrite16_rep(host->ctl + (addr << host->bus_shift), buf, count); > } > > static inline void sd_ctrl_write32_as_16_and_16(struct tmio_mmc_host *host, > @@ -276,7 +276,7 @@ static inline void sd_ctrl_write32_as_16_and_16(struct tmio_mmc_host *host, > static inline void sd_ctrl_write32_rep(struct tmio_mmc_host *host, int addr, > const u32 *buf, int count) > { > - writesl(host->ctl + (addr << host->bus_shift), buf, count); > + iowrite32_rep(host->ctl + (addr << host->bus_shift), buf, count); > } > > #endif >
On 18 December 2017 at 01:00, Wolfram Sang <wsa+renesas@sang-engineering.com> wrote: > Not all archs define reads* and writes*. Switch to ioread*_rep and > friends which is defined everywhere, so we can enable COMPILE_TEST after > that. > > Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com> Thanks, applied for next! I put the patch in front of Yamada-san's patch that made it possible to use COMPILE_TEST. Thus avoiding breaking bisecting. > --- > > So, I pushed this to buildbot on top of Yamada-san's patch series and there > were no complaints, even with COMPILE_TEST enabled. I also did some tests on > HW and checksuming huge files on SD cards still works. > > However, I am not sure about this mixture of read* and ioread* functions. Shall > we convert maybe all of those? Yes, seems like a good idea. [...] Kind regards Uffe
2017-12-18 22:01 GMT+09:00 Ulf Hansson <ulf.hansson@linaro.org>: > On 18 December 2017 at 01:00, Wolfram Sang > <wsa+renesas@sang-engineering.com> wrote: >> Not all archs define reads* and writes*. Switch to ioread*_rep and >> friends which is defined everywhere, so we can enable COMPILE_TEST after >> that. >> >> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com> > > Thanks, applied for next! > > I put the patch in front of Yamada-san's patch that made it possible > to use COMPILE_TEST. Thus avoiding breaking bisecting. > >> --- >> >> So, I pushed this to buildbot on top of Yamada-san's patch series and there >> were no complaints, even with COMPILE_TEST enabled. I also did some tests on >> HW and checksuming huge files on SD cards still works. >> >> However, I am not sure about this mixture of read* and ioread* functions. Shall >> we convert maybe all of those? > > Yes, seems like a good idea. > > [...] > BTW, is ioread* preferred to read*? We need to eliminate the root cause. Other drivers using reads* cannot enable COMPILE_TEST for the same reason. Wolfram, Can you send a patch for sparc64?
Yamada-san, > BTW, is ioread* preferred to read*? Technically, yes. In reality, mostly not. > We need to eliminate the root cause. > Other drivers using reads* cannot enable COMPILE_TEST for the same reason. Sure, I understand this. There are not many left, however. I'd assume most switched to io*_rep. The sparc64 issue is known for at least two years. > Wolfram, > Can you send a patch for sparc64? As I wrote here already [1], this is too much of a side task for me. But if you like, please have a go... But I'll send the consistency patch for tmio in a second. Kind regards, Wolfram [1] https://lkml.org/lkml/2017/12/15/816
diff --git a/drivers/mmc/host/tmio_mmc.h b/drivers/mmc/host/tmio_mmc.h index 76094345cbacf3..03519c4ca0aa1a 100644 --- a/drivers/mmc/host/tmio_mmc.h +++ b/drivers/mmc/host/tmio_mmc.h @@ -233,7 +233,7 @@ static inline u16 sd_ctrl_read16(struct tmio_mmc_host *host, int addr) static inline void sd_ctrl_read16_rep(struct tmio_mmc_host *host, int addr, u16 *buf, int count) { - readsw(host->ctl + (addr << host->bus_shift), buf, count); + ioread16_rep(host->ctl + (addr << host->bus_shift), buf, count); } static inline u32 sd_ctrl_read16_and_16_as_32(struct tmio_mmc_host *host, @@ -246,7 +246,7 @@ static inline u32 sd_ctrl_read16_and_16_as_32(struct tmio_mmc_host *host, static inline void sd_ctrl_read32_rep(struct tmio_mmc_host *host, int addr, u32 *buf, int count) { - readsl(host->ctl + (addr << host->bus_shift), buf, count); + ioread32_rep(host->ctl + (addr << host->bus_shift), buf, count); } static inline void sd_ctrl_write16(struct tmio_mmc_host *host, int addr, @@ -263,7 +263,7 @@ static inline void sd_ctrl_write16(struct tmio_mmc_host *host, int addr, static inline void sd_ctrl_write16_rep(struct tmio_mmc_host *host, int addr, u16 *buf, int count) { - writesw(host->ctl + (addr << host->bus_shift), buf, count); + iowrite16_rep(host->ctl + (addr << host->bus_shift), buf, count); } static inline void sd_ctrl_write32_as_16_and_16(struct tmio_mmc_host *host, @@ -276,7 +276,7 @@ static inline void sd_ctrl_write32_as_16_and_16(struct tmio_mmc_host *host, static inline void sd_ctrl_write32_rep(struct tmio_mmc_host *host, int addr, const u32 *buf, int count) { - writesl(host->ctl + (addr << host->bus_shift), buf, count); + iowrite32_rep(host->ctl + (addr << host->bus_shift), buf, count); } #endif
Not all archs define reads* and writes*. Switch to ioread*_rep and friends which is defined everywhere, so we can enable COMPILE_TEST after that. Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com> --- So, I pushed this to buildbot on top of Yamada-san's patch series and there were no complaints, even with COMPILE_TEST enabled. I also did some tests on HW and checksuming huge files on SD cards still works. However, I am not sure about this mixture of read* and ioread* functions. Shall we convert maybe all of those? drivers/mmc/host/tmio_mmc.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)