From mboxrd@z Thu Jan 1 00:00:00 1970 From: Denys Vlasenko Subject: Re: bnx2 dirver's firmware images Date: Wed, 19 Sep 2007 21:29:22 +0100 Message-ID: <200709192129.23034.vda.linux@googlemail.com> References: <200709181823.26429.vda.linux@googlemail.com> <200709190930.39882.vda.linux@googlemail.com> <1190235642.9540.261.camel@dell> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-15" Content-Transfer-Encoding: 7bit Cc: "David Miller" , linux-kernel@vger.kernel.org, "netdev" To: "Michael Chan" Return-path: In-Reply-To: <1190235642.9540.261.camel@dell> Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Wednesday 19 September 2007 22:00, Michael Chan wrote: > On Wed, 2007-09-19 at 09:30 +0100, Denys Vlasenko wrote: > + /* gzip header (1f,8b,08... 10 bytes total + possible asciz filename) > + * is stripped, 32-bit unpacked size (LE) is prepended instead */ > + sz = *zbuf++; > + sz = (sz << 8) + *zbuf++; > + sz = (sz << 8) + *zbuf++; > + sz = (sz << 8) + *zbuf++; > > I don't have a problem with removing the gzip header. It doesn't > contain very useful information other than a valid header for sanity > check. But I don't think we need to arbitrarily add the unpacked size > in front of the gzipped data. The driver knows the size (e.g. the size > of RAM on the chip) and should pass it to the call. The driver should > also allocate the memory for the unpacked data instead of allocating the > memory inside the call and freeing it by the caller. For example, the > driver may need to use pci_alloc_consistent() if the firmware is to be > DMA'ed to the chip. > > Other than that, everything else looks fine. Thanks. Are you saying that you successfully run-tested it? -- vda