| Message ID | 20180329124745.24694-1-cmc@babblebit.net (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Thu, Mar 29 2018, Chris Coffey wrote: > This fixes the following sparse warning: > > drivers/staging/mt7621-eth/ethtool.c:213:6: warning: symbol > 'mtk_set_ethtool_ops' was not declared. Should it be static? > > Signed-off-by: Chris Coffey <cmc@babblebit.net> Reviewed-by: NeilBrown <neil@brown.name> Thanks, NeilBrown > --- > Changes in v2: > - Per GregKH's feedback (thanks!), don't add unnecessary new .h file > dependencies. This patch version reverts those changes and fixes the > problem directly in ethtool.c (which is that it didn't include > ethtool.h anywhere -- mtk_set_ethtool_ops is not static). > > drivers/staging/mt7621-eth/ethtool.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/staging/mt7621-eth/ethtool.c b/drivers/staging/mt7621-eth/ethtool.c > index 38ba0c040a..5268c5ca09 100644 > --- a/drivers/staging/mt7621-eth/ethtool.c > +++ b/drivers/staging/mt7621-eth/ethtool.c > @@ -13,6 +13,7 @@ > */ > > #include "mtk_eth_soc.h" > +#include "ethtool.h" > > static const char mtk_gdma_str[][ETH_GSTRING_LEN] = { > #define _FE(x...) # x, > -- > 2.11.0
On Mon, 2018-04-02 at 09:34 +1000, NeilBrown wrote: > On Thu, Mar 29 2018, Chris Coffey wrote: > > > This fixes the following sparse warning: > > > > drivers/staging/mt7621-eth/ethtool.c:213:6: warning: symbol > > 'mtk_set_ethtool_ops' was not declared. Should it be static? > > > > Signed-off-by: Chris Coffey <cmc@babblebit.net> > > Reviewed-by: NeilBrown <neil@brown.name> > > Thanks, > NeilBrown > Hi, Neil Forgive me I cannot find the cover letter in the original series in my mailbox to make a reply, so I rudely made here just letting you know something good to the growth of mt7621 support in upstream. do you have maintained an out-of-tree branch to boot the mt7621 machine with those staging patches? If so, it would become a bit easier for me that maybe I could give a hand for migrating these staging driver for mt7621 to mainline. I thought mmc, pci, ethernet, gsw and hsdma all could probably reuse the current mainline code. Sean > > > --- > > Changes in v2: > > - Per GregKH's feedback (thanks!), don't add unnecessary new .h file > > dependencies. This patch version reverts those changes and fixes the > > problem directly in ethtool.c (which is that it didn't include > > ethtool.h anywhere -- mtk_set_ethtool_ops is not static). > > > > drivers/staging/mt7621-eth/ethtool.c | 1 + > > 1 file changed, 1 insertion(+) > > > > diff --git a/drivers/staging/mt7621-eth/ethtool.c b/drivers/staging/mt7621-eth/ethtool.c > > index 38ba0c040a..5268c5ca09 100644 > > --- a/drivers/staging/mt7621-eth/ethtool.c > > +++ b/drivers/staging/mt7621-eth/ethtool.c > > @@ -13,6 +13,7 @@ > > */ > > > > #include "mtk_eth_soc.h" > > +#include "ethtool.h" > > > > static const char mtk_gdma_str[][ETH_GSTRING_LEN] = { > > #define _FE(x...) # x, > > -- > > 2.11.0 > _______________________________________________ > Linux-mediatek mailing list > Linux-mediatek@lists.infradead.org > http://lists.infradead.org/mailman/listinfo/linux-mediatek
On Mon, Apr 02 2018, Sean Wang wrote: > On Mon, 2018-04-02 at 09:34 +1000, NeilBrown wrote: >> On Thu, Mar 29 2018, Chris Coffey wrote: >> >> > This fixes the following sparse warning: >> > >> > drivers/staging/mt7621-eth/ethtool.c:213:6: warning: symbol >> > 'mtk_set_ethtool_ops' was not declared. Should it be static? >> > >> > Signed-off-by: Chris Coffey <cmc@babblebit.net> >> >> Reviewed-by: NeilBrown <neil@brown.name> >> >> Thanks, >> NeilBrown >> > Hi, Neil > > Forgive me I cannot find the cover letter in the original series in my > mailbox to make a reply, so I rudely made here just letting you know > something good to the growth of mt7621 support in upstream. > > do you have maintained an out-of-tree branch to boot the mt7621 machine > with those staging patches? I can boot v4.16 plus staging-next on my mt7621 with no other patches. An important fix (a63d706ea71919) landed in 4.16-rc7. It do have https://github.com/neilbrown/linux/commits/gnubee/v4.15 which was part of preparation for this, but now that very nearly all I need is in mainline or on its way, I won't be doing much more in that tree. > > If so, it would become a bit easier for me that maybe I could give a > hand for migrating these staging driver for mt7621 to mainline. I > thought mmc, pci, ethernet, gsw and hsdma all could probably reuse the > current mainline code. I agree that it is quite likely that several of these drivers could and should reuse current mainline code. I would love to have some help sorting this out. I won't have much time myself to dig into it for several weeks, but I'll make time to review and test any code that is contributed. Thanks, NeilBrown
On Mon, Apr 02, 2018 at 01:41:33PM +1000, NeilBrown wrote: > On Mon, Apr 02 2018, Sean Wang wrote: > > Hi, Neil > > > > Forgive me I cannot find the cover letter in the original series in my > > mailbox to make a reply, so I rudely made here just letting you know > > something good to the growth of mt7621 support in upstream. > > > > If so, it would become a bit easier for me that maybe I could give a > > hand for migrating these staging driver for mt7621 to mainline. I > > thought mmc, pci, ethernet, gsw and hsdma all could probably reuse the > > current mainline code. > > I agree that it is quite likely that several of these drivers could and > should reuse current mainline code. I would love to have some help > sorting this out. I won't have much time myself to dig into it for > several weeks, but I'll make time to review and test any code that is > contributed. > > Thanks, > NeilBrown Hi, I have started a patch series for the mmc driver that is currently in staging, that makes some of the first cleanups (whitespace, indent,..) and also starts to clean up the code, to make the similarities and differences to the current mainline code clearer. The series is far from being complete, but I think I'm going to send it out today, so you can look and comment. I would suggest that we continue cleaning the code that is currently in staging to remove the dead parts and make the diff to the mainline code obvious. Do you agree with that proposal or do you suggest a different plan? Christian
On Mon, Apr 02 2018, Christian Lütke-Stetzkamp wrote: > On Mon, Apr 02, 2018 at 01:41:33PM +1000, NeilBrown wrote: >> On Mon, Apr 02 2018, Sean Wang wrote: >> > Hi, Neil >> > >> > Forgive me I cannot find the cover letter in the original series in my >> > mailbox to make a reply, so I rudely made here just letting you know >> > something good to the growth of mt7621 support in upstream. >> > >> > If so, it would become a bit easier for me that maybe I could give a >> > hand for migrating these staging driver for mt7621 to mainline. I >> > thought mmc, pci, ethernet, gsw and hsdma all could probably reuse the >> > current mainline code. >> >> I agree that it is quite likely that several of these drivers could and >> should reuse current mainline code. I would love to have some help >> sorting this out. I won't have much time myself to dig into it for >> several weeks, but I'll make time to review and test any code that is >> contributed. >> >> Thanks, >> NeilBrown > > Hi, > > I have started a patch series for the mmc driver that is currently in > staging, that makes some of the first cleanups (whitespace, indent,..) > and also starts to clean up the code, to make the similarities and > differences to the current mainline code clearer. The series is far > from being complete, but I think I'm going to send it out today, so > you can look and comment. > > I would suggest that we continue cleaning the code that is currently > in staging to remove the dead parts and make the diff to the mainline > code obvious. Do you agree with that proposal or do you suggest a > different plan? I think this is an excellent plan - thanks for doing it. I'll hopefully look over your patches today and send any comments. Thanks, NeilBrown
diff --git a/drivers/staging/mt7621-eth/ethtool.c b/drivers/staging/mt7621-eth/ethtool.c index 38ba0c040a..5268c5ca09 100644 --- a/drivers/staging/mt7621-eth/ethtool.c +++ b/drivers/staging/mt7621-eth/ethtool.c @@ -13,6 +13,7 @@ */ #include "mtk_eth_soc.h" +#include "ethtool.h" static const char mtk_gdma_str[][ETH_GSTRING_LEN] = { #define _FE(x...) # x,
This fixes the following sparse warning: drivers/staging/mt7621-eth/ethtool.c:213:6: warning: symbol 'mtk_set_ethtool_ops' was not declared. Should it be static? Signed-off-by: Chris Coffey <cmc@babblebit.net> --- Changes in v2: - Per GregKH's feedback (thanks!), don't add unnecessary new .h file dependencies. This patch version reverts those changes and fixes the problem directly in ethtool.c (which is that it didn't include ethtool.h anywhere -- mtk_set_ethtool_ops is not static). drivers/staging/mt7621-eth/ethtool.c | 1 + 1 file changed, 1 insertion(+)