| Message ID | 1422190968-14422-1-git-send-email-zajec5@gmail.com (mailing list archive) |
|---|---|
| State | Accepted |
| Delegated to: | Kalle Valo |
| Headers | show |
Rafa? Mi?ecki <zajec5@gmail.com> writes: > This allows us to drop some #ifdef magic (mess). > > Signed-off-by: Rafa? Mi?ecki <zajec5@gmail.com> > --- > V2: Return false in bcma_core_pci_is_in_hostmode > Don't (accidentally) modify bcma_host_soc_register_driver It would be far more reliable if you resend the whole patchset instead of resending invidiviual patches within the set. Otherwise the chances are that I apply the wrong version.
On 26 January 2015 at 07:58, Kalle Valo <kvalo@codeaurora.org> wrote: > Rafa? Mi?ecki <zajec5@gmail.com> writes: > >> This allows us to drop some #ifdef magic (mess). >> >> Signed-off-by: Rafa? Mi?ecki <zajec5@gmail.com> >> --- >> V2: Return false in bcma_core_pci_is_in_hostmode >> Don't (accidentally) modify bcma_host_soc_register_driver > > It would be far more reliable if you resend the whole patchset instead > of resending invidiviual patches within the set. Otherwise the chances > are that I apply the wrong version. Oops. I always take care of removing old versions from patchwork and using --in-reply-to, I was hoping it's OK. How would you like whole patches to be re-send? Should I resend them independently? Or should every patch from the patchset include In-Reply-To pointing to its previous version? Is this just an advise for the future, or would you like me to resend this patchset too?
Rafa? Mi?ecki <zajec5@gmail.com> writes: > On 26 January 2015 at 07:58, Kalle Valo <kvalo@codeaurora.org> wrote: >> Rafa? Mi?ecki <zajec5@gmail.com> writes: >> >>> This allows us to drop some #ifdef magic (mess). >>> >>> Signed-off-by: Rafa? Mi?ecki <zajec5@gmail.com> >>> --- >>> V2: Return false in bcma_core_pci_is_in_hostmode >>> Don't (accidentally) modify bcma_host_soc_register_driver >> >> It would be far more reliable if you resend the whole patchset instead >> of resending invidiviual patches within the set. Otherwise the chances >> are that I apply the wrong version. > > Oops. I always take care of removing old versions from patchwork Yeah, I noticed that. That's really helpful, thanks for that. > and using --in-reply-to, I was hoping it's OK. But still ordering is different which might introduce problems while I apply them. And like in your case, when I have to take the patches from email due to UTF-8 problems, it won't work at all. > How would you like whole patches to be re-send? Should I resend them > independently? Or should every patch from the patchset include > In-Reply-To pointing to its previous version? Let's say you have a ten patch patchset and you have to change something in patch 3. I would prefer that you resend the whole patchset (all 10 patches) and each patch in the patchset has "v2". So the version is actually version of the patchset, not of the individual patch. I assumed this was standard practice everywhere in the kernel, but I guess I was wrong. > Is this just an advise for the future, or would you like me to resend > this patchset too? For the future.
On 26 January 2015 at 13:31, Kalle Valo <kvalo@codeaurora.org> wrote: > Rafa? Mi?ecki <zajec5@gmail.com> writes: > >> On 26 January 2015 at 07:58, Kalle Valo <kvalo@codeaurora.org> wrote: >>> Rafa? Mi?ecki <zajec5@gmail.com> writes: >>> >>>> This allows us to drop some #ifdef magic (mess). >>>> >>>> Signed-off-by: Rafa? Mi?ecki <zajec5@gmail.com> >>>> --- >>>> V2: Return false in bcma_core_pci_is_in_hostmode >>>> Don't (accidentally) modify bcma_host_soc_register_driver >>> >>> It would be far more reliable if you resend the whole patchset instead >>> of resending invidiviual patches within the set. Otherwise the chances >>> are that I apply the wrong version. >> >> Oops. I always take care of removing old versions from patchwork > > Yeah, I noticed that. That's really helpful, thanks for that. > >> and using --in-reply-to, I was hoping it's OK. > > But still ordering is different which might introduce problems while I > apply them. And like in your case, when I have to take the patches from > email due to UTF-8 problems, it won't work at all. > >> How would you like whole patches to be re-send? Should I resend them >> independently? Or should every patch from the patchset include >> In-Reply-To pointing to its previous version? > > Let's say you have a ten patch patchset and you have to change something > in patch 3. I would prefer that you resend the whole patchset (all 10 > patches) and each patch in the patchset has "v2". So the version is > actually version of the patchset, not of the individual patch. > > I assumed this was standard practice everywhere in the kernel, but I > guess I was wrong. It could be just me, maybe I just didn't meet anyone complaining yet. Thanks a lot for your help!
diff --git a/drivers/bcma/bcma_private.h b/drivers/bcma/bcma_private.h index 3f314c9..ac6c5fc 100644 --- a/drivers/bcma/bcma_private.h +++ b/drivers/bcma/bcma_private.h @@ -107,6 +107,14 @@ extern int bcma_chipco_watchdog_register(struct bcma_drv_cc *cc); #ifdef CONFIG_BCMA_DRIVER_PCI_HOSTMODE bool bcma_core_pci_is_in_hostmode(struct bcma_drv_pci *pc); void bcma_core_pci_hostmode_init(struct bcma_drv_pci *pc); +#else +static inline bool bcma_core_pci_is_in_hostmode(struct bcma_drv_pci *pc) +{ + return false; +} +static inline void bcma_core_pci_hostmode_init(struct bcma_drv_pci *pc) +{ +} #endif /* CONFIG_BCMA_DRIVER_PCI_HOSTMODE */ #ifdef CONFIG_BCMA_DRIVER_GPIO diff --git a/drivers/bcma/driver_pci.c b/drivers/bcma/driver_pci.c index 50329d1..b85a505 100644 --- a/drivers/bcma/driver_pci.c +++ b/drivers/bcma/driver_pci.c @@ -226,11 +226,9 @@ void bcma_core_pci_init(struct bcma_drv_pci *pc) if (pc->setup_done) return; -#ifdef CONFIG_BCMA_DRIVER_PCI_HOSTMODE pc->hostmode = bcma_core_pci_is_in_hostmode(pc); if (pc->hostmode) bcma_core_pci_hostmode_init(pc); -#endif /* CONFIG_BCMA_DRIVER_PCI_HOSTMODE */ if (!pc->hostmode) bcma_core_pci_clientmode_init(pc);
This allows us to drop some #ifdef magic (mess). Signed-off-by: Rafa? Mi?ecki <zajec5@gmail.com> --- V2: Return false in bcma_core_pci_is_in_hostmode Don't (accidentally) modify bcma_host_soc_register_driver --- drivers/bcma/bcma_private.h | 8 ++++++++ drivers/bcma/driver_pci.c | 2 -- 2 files changed, 8 insertions(+), 2 deletions(-)