From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from ozlabs.org ([203.10.76.45]) by merlin.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1UGMoI-0002D2-Rg for linux-mtd@lists.infradead.org; Fri, 15 Mar 2013 05:04:08 +0000 From: Rusty Russell To: James Hogan Subject: Re: [RFC -next] linux/linkage.h: fix symbol prefix handling In-Reply-To: <5141AB3A.2060705@imgtec.com> References: <1362656642-2693-1-git-send-email-james.hogan@imgtec.com> <87sj46wyf2.fsf@rustcorp.com.au> <5139AC39.90805@imgtec.com> <874ngiwije.fsf@rustcorp.com.au> <20130311230731.0c610d86569d09f6e23b8e61@canb.auug.org.au> <87ppz5usts.fsf@rustcorp.com.au> <20130312233300.9d86d47bb5369923ad520a21@canb.auug.org.au> <87ehfkuq27.fsf@rustcorp.com.au> <20130313063157.GB19681@merkur.ravnborg.org> <87zjy6tytz.fsf@rustcorp.com.au> <5141AB3A.2060705@imgtec.com> Date: Fri, 15 Mar 2013 15:07:31 +1030 Message-ID: <87obelth10.fsf@rustcorp.com.au> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Michal Marek , Stephen Rothwell , Mike Frysinger , linux-kbuild@vger.kernel.org, linux-kernel@vger.kernel.org, Paul Gortmaker , linux-next@vger.kernel.org, linux-mtd@lists.infradead.org, Al Viro , Jean Delvare , uclinux-dist-devel@blackfin.uclinux.org, Andrew Morton , Sam Ravnborg , Guenter Roeck List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , James Hogan writes: > On 14/03/13 04:00, Rusty Russell wrote: >> From: Rusty Russell >> Subject: CONFIG_SYMBOL_PREFIX: cleanup. ... > Reviewed-by: James Hogan > Tested-by: James Hogan (metag) Thanks. > The only other special case of symbol prefixing I'm aware of is in > scripts/genksyms/genksyms.c. It makes the decision at runtime based > on the --arch=$ARCH argument, and is the only use of the arch argument: > >> if ((strcmp(arch, "h8300") == 0) || (strcmp(arch, "blackfin") == 0) || >> (strcmp(arch, "metag") == 0)) >> mod_prefix = "_"; > > Does the patch below look reasonable in addition to your patch? > (Note: I'm not sure if genksyms is only used internally or whether > it's API should be kept stable?). > > Thanks > James > > Subject: [PATCH] genksyms: pass symbol-prefix instead of arch Agreed, I've applied this. Thanks, Rusty.