From mboxrd@z Thu Jan 1 00:00:00 1970 From: Auke Kok Subject: Re: [PATCH] ixgb: Delete IXGB_DBG() macro and call pr_debug() directly. Date: Tue, 10 Oct 2006 09:50:36 -0700 Message-ID: <452BCF5C.2090407@intel.com> References: <452BC6C9.3050902@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Linux kernel mailing list , trivial@kernel.org, NetDev , Jesse Brandeburg Return-path: Received: from mga01.intel.com ([192.55.52.88]:58125 "EHLO mga01.intel.com") by vger.kernel.org with ESMTP id S1030210AbWJJQwr (ORCPT ); Tue, 10 Oct 2006 12:52:47 -0400 To: "Robert P. J. Day" In-Reply-To: Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Robert P. J. Day wrote: > On Tue, 10 Oct 2006, Auke Kok wrote: > >> Robert P. J. Day wrote: >>> Delete the minimally-useful IXGB_DBG() macro and call pr_debug() >>> directly from the main routine. >>> >>> Signed-off-by: Robert P. J. Day >>> --- >>> diff --git a/drivers/net/ixgb/ixgb.h b/drivers/net/ixgb/ixgb.h >>> index 50ffe90..fb9fde5 100644 >>> --- a/drivers/net/ixgb/ixgb.h >>> +++ b/drivers/net/ixgb/ixgb.h >>> @@ -77,12 +77,6 @@ #include "ixgb_hw.h" >>> #include "ixgb_ee.h" >>> #include "ixgb_ids.h" >>> >>> -#ifdef _DEBUG_DRIVER_ >>> -#define IXGB_DBG(args...) printk(KERN_DEBUG "ixgb: " args) >>> -#else >>> -#define IXGB_DBG(args...) >>> -#endif >>> - >>> #define PFX "ixgb: " >>> #define DPRINTK(nlevel, klevel, fmt, args...) \ >>> (void)((NETIF_MSG_##nlevel & adapter->msg_enable) && \ >>> diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c >>> index e09f575..eada685 100644 >>> --- a/drivers/net/ixgb/ixgb_main.c >>> +++ b/drivers/net/ixgb/ixgb_main.c >>> @@ -1948,7 +1948,7 @@ #endif >>> >>> /* All receives must fit into a single buffer */ >>> >>> - IXGB_DBG("Receive packet consumed multiple buffers " >>> + pr_debug("ixgb: Receive packet consumed multiple >>> buffers " >>> "length<%x>\n", length); >>> >>> dev_kfree_skb_irq(skb); >>> >>> -- >>> >>> all right ... what did i mess up *this* time? :-) it's good >>> practice. that's my story and i'm sticking to it. >> >> We should really use dev_dbg() instead, as it retains the 'ethX:' >> annotation afaics. > > i actually tried to use that first, but it wasn't clear to me what i > would use as that first argument to dev_dbg(), given the definitions > in include/linux/device.h: > > #define dev_dbg(dev, format, arg...) \ > dev_printk(KERN_DEBUG , dev , format , ## arg) > > #define dev_printk(level, dev, format, arg...) \ > printk(level "%s %s: " format , dev_driver_string(dev) , > (dev)->bus_id , ## arg) > > if someone wants to tell me what, in the context of ixgb_main.c, i > would use as that "dev" argument, i'm all for that. (CC netdev since it's a network driver topic). all our macro's (e100, e1000, ixgb) use adapter->netdev->name inserted through the DPRINTK macro. if you'd really want to clean it all up, you'd have to replace all DPRINTK() calls with dev_dbg(adapter->netdev->name, ....) which would just make it more lengthy and uncomfortable to read. which puts this in a bigger perspective. I suppose the nicest way to do program these is to do something like this: #define ixgb_dbg(args...) dev_dbg(adapter->netdev->name, args) #define ixgb_err(args...) dev_err(adapter->netdev->name, args) #define ixgb_info(args...) dev_info(adapter->netdev->name, args) and use those consistently throughout the driver, ditto for e100/e1000. I'll look into it and see what I can do. Cheers, Auke