| Message ID | 20210515221320.1255291-14-arnd@kernel.org (mailing list archive) |
|---|---|
| State | RFC |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | drivers/net/Space.c cleanup | expand |
On Sun, 16 May 2021, Arnd Bergmann wrote: > From: Arnd Bergmann <arnd@arndb.de> > > This driver always warns about unused functions because it includes > an file that it doesn't actually need: > I don't think you can omit #include "lib8390.c" here without changing driver behaviour, because of the macros in effect. I think this change would need some actual testing unless you can show that the module binary does not change.
On Sun, May 16, 2021 at 6:24 AM Finn Thain <fthain@telegraphics.com.au> wrote: > On Sun, 16 May 2021, Arnd Bergmann wrote: > > From: Arnd Bergmann <arnd@arndb.de> > > > > This driver always warns about unused functions because it includes > > an file that it doesn't actually need: > > I don't think you can omit #include "lib8390.c" here without changing > driver behaviour, because of the macros in effect. > > I think this change would need some actual testing unless you can show > that the module binary does not change. Michael posted a similar but different patch a while ago, involving calling ax_NS8390_reinit(): https://lore.kernel.org/linux-m68k/1528604559-972-3-git-send-email-schmitzmic@gmail.com/ Gr{oetje,eeting}s, Geert
On Sun, May 16, 2021 at 11:04 AM Geert Uytterhoeven <geert@linux-m68k.org> wrote: > > On Sun, May 16, 2021 at 6:24 AM Finn Thain <fthain@telegraphics.com.au> wrote: > > On Sun, 16 May 2021, Arnd Bergmann wrote: > > > From: Arnd Bergmann <arnd@arndb.de> > > > > > > This driver always warns about unused functions because it includes > > > an file that it doesn't actually need: > > > > I don't think you can omit #include "lib8390.c" here without changing > > driver behaviour, because of the macros in effect. > > > > I think this change would need some actual testing unless you can show > > that the module binary does not change. > > Michael posted a similar but different patch a while ago, involving > calling ax_NS8390_reinit(): > https://lore.kernel.org/linux-m68k/1528604559-972-3-git-send-email-schmitzmic@gmail.com/ Ah nice. As far as I can tell, the two versions are functionally equivalent based on my reading of the port accessors, but he probably tested his version, so I'll drop mine from this series. Nothing else depends on this one, I just included it since it was an obvious thing to fix. Arnd
diff --git a/drivers/net/ethernet/8390/Makefile b/drivers/net/ethernet/8390/Makefile index 85c83c566ec6..304897d5f0f9 100644 --- a/drivers/net/ethernet/8390/Makefile +++ b/drivers/net/ethernet/8390/Makefile @@ -16,5 +16,5 @@ obj-$(CONFIG_PCMCIA_PCNET) += pcnet_cs.o 8390.o obj-$(CONFIG_STNIC) += stnic.o 8390.o obj-$(CONFIG_ULTRA) += smc-ultra.o 8390.o obj-$(CONFIG_WD80x3) += wd.o 8390.o -obj-$(CONFIG_XSURF100) += xsurf100.o +obj-$(CONFIG_XSURF100) += xsurf100.o 8390.o obj-$(CONFIG_ZORRO8390) += zorro8390.o diff --git a/drivers/net/ethernet/8390/xsurf100.c b/drivers/net/ethernet/8390/xsurf100.c index e2c963821ffe..11d5d43e7202 100644 --- a/drivers/net/ethernet/8390/xsurf100.c +++ b/drivers/net/ethernet/8390/xsurf100.c @@ -42,10 +42,7 @@ /* Ensure we have our RCR base value */ #define AX88796_PLATFORM -static unsigned char version[] = - "ax88796.c: Copyright 2005,2007 Simtec Electronics\n"; - -#include "lib8390.c" +#include "8390.h" /* from ne.c */ #define NE_CMD EI_SHIFT(0x00) @@ -232,7 +229,7 @@ static void xs100_block_output(struct net_device *dev, int count, if (jiffies - dma_start > 2 * HZ / 100) { /* 20ms */ netdev_warn(dev, "timeout waiting for Tx RDC.\n"); ei_local->reset_8390(dev); - ax_NS8390_init(dev, 1); + NS8390p_init(dev, 1); break; } }