From mboxrd@z Thu Jan 1 00:00:00 1970 From: Roger Subject: Re: [PATCH] mmc: rtsx: usb backend needs LED support Date: Thu, 8 May 2014 16:08:15 +0800 Message-ID: <536B3B6F.1010209@realtek.com> References: <7208467.nZIXMIfPji@wuerfel> <201404300228.s3U2S7WA026782@rtits1.realtek.com> <53606F49.1000401@realtek.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: driverdev-devel-bounces@linuxdriverproject.org Sender: driverdev-devel-bounces@linuxdriverproject.org To: Ulf Hansson , Lee Jones Cc: Samuel Ortiz , Arnd Bergmann , Greg Kroah-Hartman , linux-mmc , Chris Ball , "linux-kernel@vger.kernel.org" , Wei WANG , devel@linuxdriverproject.org, Dan Carpenter List-Id: linux-mmc@vger.kernel.org On 04/30/2014 04:19 PM, Ulf Hansson wrote: > On 30 April 2014 05:34, Roger wrote: >> >> On 04/29/2014 08:46 PM, Arnd Bergmann wrote: >>> >>> On Tuesday 29 April 2014 13:05:15 Ulf Hansson wrote: >>>> >>>> On 29 April 2014 11:45, Arnd Bergmann wrote: >>>>> >>>>> drivers/built-in.o: In function `rtsx_usb_sdmmc_drv_remove': >>>>> :(.text+0x806480): undefined reference to `led_classdev_unregister' >>>>> drivers/built-in.o: In function `rtsx_usb_sdmmc_drv_probe': >>>>> :(.text+0x806708): undefined reference to `led_classdev_register' >>>>> >>>>> This adds an explicit dependency in Kconfig >>>> >>>> >>>> I think the proper solution is to fix the dependency in the driver code >>>> instead. >>>> >>>> There are already some ifdefs hackery for making it optional to use >>>> leds, apparently that's not working properly. >>>> >>> >>> Ah, right, I misinterpreted the bug. Here is a new version: >>> >>> 8<--------- >>>> >>>> From 5b5588f8c9b8ded8b296fd32d87b2d118e548a29 Mon Sep 17 00:00:00 2001 >>> >>> From: Arnd Bergmann >>> Date: Tue, 29 Apr 2014 11:41:40 +0200 >>> Subject: [PATCH] mmc: rtsx: usb backend needs LED support >>> >>> Building the rtsx USB driver uses the LED classdev base >>> support if available, but that fails if the classdev >>> is a module and the MMC driver is built-in, leading to this >>> link error. >>> >>> drivers/built-in.o: In function `rtsx_usb_sdmmc_drv_remove': >>> :(.text+0x806480): undefined reference to `led_classdev_unregister' >>> drivers/built-in.o: In function `rtsx_usb_sdmmc_drv_probe': >>> :(.text+0x806708): undefined reference to `led_classdev_register' >>> >>> This adds an explicit dependency in Kconfig to ensure that >>> the MMC driver has to be a module if LEDS_CLASS is a module, >>> but still allows it to be built-in if LEDS_CLASS is either >>> disabled or built-in. >>> >>> Signed-off-by: Arnd Bergmann >>> >>> diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig >>> index 92d91fe..68da9b8 100644 >>> >>> --- a/drivers/mmc/host/Kconfig >>> +++ b/drivers/mmc/host/Kconfig >>> @@ -696,6 +696,7 @@ config MMC_REALTEK_PCI >>> config MMC_REALTEK_USB >>> tristate "Realtek USB SD/MMC Card Interface Driver" >>> depends on MFD_RTSX_USB >>> + depends on m || LEDS_CLASS!=m >>> >>> help >>> Say Y here to include driver code to support SD/MMC card >>> interface >>> of Realtek RTS5129/39 series card reader >>> >>> . >>> >> >> I think Ulf's idea is to fix the bug by modifying the .c files. >> I really found the problem of my ifdef hackery and it should do something >> similar in sdhci.c as: >> >> From: Roger Tseng >> Date: Wed, 30 Apr 2014 11:11:25 +0800 >> Subject: [PATCH] mmc: rtsx: fix possible linking error if built-in >> >> rtsx_usb_sdmmc module uses the LED classdev if available, but the code >> failed >> to consider the situation that it is built-in and the LED classdev is a >> module, >> leading to following linking error: >> >> LD init/built-in.o >> drivers/built-in.o: In function `rtsx_usb_sdmmc_drv_remove': >> rtsx_usb_sdmmc.c:(.text+0x2a018e): undefined reference to >> `led_classdev_unregister' >> drivers/built-in.o: In function `rtsx_usb_sdmmc_drv_probe': >> rtsx_usb_sdmmc.c:(.text+0x2a197e): undefined reference to >> `led_classdev_register' >> >> Fix by excluding such condition when defining macro RTSX_USB_USE_LEDS_CLASS. >> >> Signed-off-by: Roger Tseng > > Acked-by: Ulf Hansson > Would this patch be merged into linux-next or Lee's mfd.git ib-mfd-mmc-memstick-3.16 branch? By the way, should I resend my version? >> --- >> drivers/mmc/host/rtsx_usb_sdmmc.c | 3 ++- >> 1 file changed, 2 insertions(+), 1 deletion(-) >> >> diff --git a/drivers/mmc/host/rtsx_usb_sdmmc.c >> b/drivers/mmc/host/rtsx_usb_sdmmc.c >> index e11fafa..38bdda5 100644 >> --- a/drivers/mmc/host/rtsx_usb_sdmmc.c >> +++ b/drivers/mmc/host/rtsx_usb_sdmmc.c >> @@ -34,7 +34,8 @@ >> #include >> #include >> >> -#if defined(CONFIG_LEDS_CLASS) || defined(CONFIG_LEDS_CLASS_MODULE) >> +#if defined(CONFIG_LEDS_CLASS) || (defined(CONFIG_LEDS_CLASS_MODULE) && \ >> + defined(CONFIG_MMC_REALTEK_USB_MODULE)) >> #include >> #include >> #define RTSX_USB_USE_LEDS_CLASS >> Best regards, Roger Tseng