Le mardi 28 avril 2020 à 14:44 +0200, Greg KH a écrit : > On Sun, Apr 26, 2020 at 07:20:30PM +0200, Akira shimahara wrote: > > Le dimanche 26 avril 2020 à 19:09 +0200, Greg KH a écrit : > > > On Sun, Apr 26, 2020 at 02:36:52PM +0200, Akira shimahara wrote: > > > > Le dimanche 26 avril 2020 à 09:51 +0200, Greg KH a écrit : > > > > > On Sat, Apr 25, 2020 at 05:31:41PM +0200, Akira Shimahara > > > > > wrote: > > > > > > From: Akira SHIMAHARA <[email protected]> > > > > > > > > > > > > Patch for enhacement of w1_therm module. Added features : > > > > > > - Bulk read : send one command for all the slaves > > > > > > on the bus to trigger temperature conversion > > > > > > - Optimized conversion time regarding to device resolution > > > > > > - Dedicated sysfs entry for powering read, > > > > > > resolution set/get, eeprom save/restore > > > > > > - Alarms settings and reading > > > > > > - Code optimization to mitigate bus traffic > > > > > > (devices information are stored to avoid > > > > > > interrogating each device every-time) > > > > > > > > > > > > Following sysfs entry are added : > > > > > > - temperature (RO) : return the temperature in 1/1000° > > > > > > - ext_power (RO) : return the power status of the device > > > > > > - resolution (RW) : get or set the device resolution > > > > > > (supported > > > > > > devices) > > > > > > - eeprom (WO) :trigger a save or restore to/from device > > > > > > EEPROM > > > > > > - alarms (RW) : read or write TH and TL in the device RAM > > > > > > - therm_bulk_read (RW) : Attribute at master level to > > > > > > trigger > > > > > > bulk read and to survey the progress of devices > > > > > > conversions > > > > > > - w1_slave has been kept for compatibility > > > > > > > > > > > > Main motivation was to improve temperature reading speed, > > > > > > which > > > > > > depend > > > > > > on resolution settings of devices. The module store the > > > > > > powwer > > > > > > status and > > > > > > the resolution of each device so that during reading > > > > > > operation, > > > > > > no > > > > > > transaction is required on the bus, which improve speed. > > > > > > The hardware status is checked as soon as a new device is > > > > > > detected, > > > > > > when a user change occurred, or when the corresponding sys > > > > > > file > > > > > > is > > > > > > accessed by user. > > > > > > > > > > > > The bulk read allow to trigger convserion of all devices on > > > > > > the > > > > > > bus > > > > > > at > > > > > > the same time. It will apply a strong pull up on the line > > > > > > if at > > > > > > least > > > > > > one device required it. The duration of the pull up is the > > > > > > max > > > > > > time > > > > > > required by a device on the line. > > > > > > > > > > > > Please let me know any feedback you have on this patch. > > > > > > > > > > > > Thanks ahead, > > > > > > > > > > > > Signed-off-by: Akira Shimahara <[email protected]> > > > > > > --- > > > > > > Changes in v2: > > > > > > - Adding documentation in Documentatin/ABI/testing/sysfs- > > > > > > driver- > > > > > > w1_therm > > > > > > - Updating existing documentation in > > > > > > Documentation/w1/slaves/w1_therm.rst > > > > > > > > > > > > drivers/w1/slaves/w1_therm.c | 1406 > > > > > > ++++++++++++++++++++++++++-- > > > > > > ------ > > > > > > drivers/w1/slaves/w1_therm.h | 386 ++++++++++ > > > > > > 2 files changed, 1470 insertions(+), 322 deletions(-) > > > > > > create mode 100644 drivers/w1/slaves/w1_therm.h > > > > > > > > > > No documentation files are added here :( > > > > > > > > > > > > > It's in the PATCH 2/2 included in the previous mail. You want > > > > me to > > > > merge in one commit ?. I thought it was easier for you to keep > > > > a > > > > track > > > > of the v1. > > > > > > What previous mail? I don't see a patch 2/2 here, did you not > > > cc: > > > me? > > > > > > thanks, > > > > > > greg kh > > > > I'm so sorry Greg, I made a mistake. Please find hereby the patch > > 2/2. > > Let me know if you want me to merge into 1 commit. > > No, multiple patches are good. > > And the first patch really should be split up into smaller pieces > too. > Each patch should only do 1 thing, not lots of things all at once. > > Can you do that and send a patch series please? > > thanks, > > greg k-h
Well noted, I will do it tomorrow Regards Akira Shimahara
