[ /me promises to execise more Cut-N-Paste ] [radeon_drv.c] ... #define KMS_DRIVER_MAJOR 2 #define KMS_DRIVER_MINOR 3 #define KMS_DRIVER_PATCHLEVEL 0 +#define KMS_DRIVER_DATE YYYYMMDD ... .name = DRIVER_NAME, .desc = DRIVER_DESC, - .date = DRIVER_DATE, + .date = KMS_DRIVER_DATE, .major = KMS_DRIVER_MAJOR, .minor = KMS_DRIVER_MINOR, .patchlevel = KMS_DRIVER_PATCHLEVEL, - Sedat - On Mon, Apr 19, 2010 at 2:15 PM, Sedat Dilek <sedat.dilek at googlemail.com> wrote: > [ Changed CC to new ML-address @ fd.o ] > > On Mon, Apr 19, 2010 at 12:44 PM, Dave Airlie <airlied at redhat.com> wrote: >> On Mon, 2010-04-19 at 11:30 +0200, Sedat Dilek wrote: > [...] >>> Do changes in radeon_drv.c (KMS-wrapper) require also a version-bump >>> in the header-file? >>> I think yes. >> >> >> No they don't. KMS and UMS drivers are separate. >> >> I referred to bumping the date as lazy, we rarely bothered doing it in >> the past. >> > > Only to clarify: > > radeon_drv.h contains all version-informations (version, date, > changelog) for the radeon UserModeSetting driver and radeon_drv.c the > same for KernelModeSetting part? > > If that is the case, it would be good to maintain a KMS_DRIVER_DATE > define and change accordingly in ".date = DRIVER_DATE" line(s) - not > sure if both lines or only one of them. > > Positive side-effect could be people don't forget to bump the driver-date. > > [radeon_drv.c] > ... > #define KMS_DRIVER_MAJOR ? ? ? ?2 > #define KMS_DRIVER_MINOR ? ? ? ?3 > #define KMS_DRIVER_PATCHLEVEL ? 0 > #define KMS_DRIVER_DATE YYYYMMDD > ... > ? ? ? .name = DRIVER_NAME, > ? ? ? .desc = DRIVER_DESC, > ? ? ? .date = KMS_DRIVER_DATE, > ? ? ? .major = DRIVER_MAJOR, > ? ? ? .minor = DRIVER_MINOR, > ? ? ? .patchlevel = DRIVER_PATCHLEVEL, > ... > > > Kind Regards, > - Sedat - > > > #define KMS_DRIVER_MAJOR ? ? ? ?2 > #define KMS_DRIVER_MINOR ? ? ? ?3 > #define KMS_DRIVER_PATCHLEVEL ? 0 > #define KMS_DRIVER_DATE YYYYMMDD > > - Sedat - >
