From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 15110 invoked by alias); 17 Feb 2004 15:42:18 -0000 Mailing-List: contact ecos-maintainers-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Post: List-Help: , Sender: ecos-maintainers-owner@sources.redhat.com Received: (qmail 15103 invoked from network); 17 Feb 2004 15:42:17 -0000 Subject: Re: FWD: JFFS2 : Node info mutex fix suggestion. From: Gary Thomas To: David Woodhouse Cc: Andrew Lunn , eCos Maintainers In-Reply-To: <1077030562.17692.55.camel@hermes> References: <20040216170742.GF29388@lunn.ch> <1077027103.17692.32.camel@hermes> <1077029910.18703.103.camel@hades.cambridge.redhat.com> <1077030562.17692.55.camel@hermes> Content-Type: multipart/mixed; boundary="=-jitzgZ8XCG86qpzSJNJ+" Organization: MLB Associates Message-Id: <1077032535.17692.65.camel@hermes> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.5 (1.4.5-7) Date: Tue, 17 Feb 2004 15:42:00 -0000 X-Spam-Level: X-Spam-Status: No, hits=0.1 required=5.0 tests=AWL autolearn=no version=2.60 X-Spam-Checker-Version: SpamAssassin 2.60 (1.212-2003-09-23-exp) on hermes.chez-thomas.org X-SW-Source: 2004-02/txt/msg00018.txt.bz2 --=-jitzgZ8XCG86qpzSJNJ+ Content-Type: text/plain Content-Transfer-Encoding: 7bit Content-length: 1125 On Tue, 2004-02-17 at 08:09, Gary Thomas wrote: > On Tue, 2004-02-17 at 07:58, David Woodhouse wrote: > > On Tue, 2004-02-17 at 07:11 -0700, Gary Thomas wrote: > > > On Mon, 2004-02-16 at 10:07, Andrew Lunn wrote: > > > > Hi Folks > > > > > > > > Does anybody know how locking should work in the jffs2 package? > > > > > > > > To me this patch looks the wrong solution. > > > > > > I agree. I did already pass this to David, with no reply :-( > > > > Hmmm. I think the answer is to remove the one in jffs2_do_new_inode() > > instead. > > > > That function should rely on the OS-specific code to pass it a struct > > jffs2_inode_info which is initialised correctly, and that includes > > having a locked semaphore in it. > > > > I've committed the appropriate change to my CVS tree. > > Thanks, I'll merge this to the eCos repository. There are also some > changes in our tree (CDL stuff mostly) that need to be pushed your > way. I'll take care of that as well. Done (diffs attached). David, what magic do you use to run "syncmail" when you make CVS changes? -- Gary Thomas MLB Associates --=-jitzgZ8XCG86qpzSJNJ+ Content-Disposition: attachment; filename=diffs Content-Type: text/plain; name=diffs; charset=iso-8859-1 Content-Transfer-Encoding: 7bit Content-length: 3220 ? jffs2-200302041443.epk ? jffs2-200311190423.epk ? jffs2-200312111628.epk ? jffs2-200402170804.epk Index: ChangeLog =================================================================== RCS file: /home/cvs/mtd/fs/jffs2/ecos/ChangeLog,v retrieving revision 1.3 diff -u -5 -p -r1.3 ChangeLog --- ChangeLog 26 Nov 2003 22:23:08 -0000 1.3 +++ ChangeLog 17 Feb 2004 15:39:06 -0000 @@ -1,5 +1,22 @@ +2004-02-17 David Woodhouse + + * src/fs-ecos.c: + Don't re-initialise the already-locked f->sem. It makes eCos unhappy. + +2004-01-27 David Woodhouse + + * src/write.c: + Fix bug noted by Howard Gray; dirents belong to, and should dirty, + the _parent_inode, not the child (which may be zero in the case + of an unlink). + +004-01-05 Thomas Koeller + + * cdl/jffs2.cdl: Re-added CYGPKG_FS_JFFS2_CFLAGS_REMOVE that had been + removed by previous change. + 2003-11-26 David Woodhouse JFFS2 cleanup and import of newer code. Remove last vestiges of Linuxisms such as 'struct inode' from the core code, leaving eCos with no excuse, and renaming the eCos 'struct inode' to make that Index: mkepk.sh =================================================================== RCS file: /home/cvs/mtd/fs/jffs2/ecos/mkepk.sh,v retrieving revision 1.6 diff -u -5 -p -r1.6 mkepk.sh --- mkepk.sh 3 Dec 2003 09:40:09 -0000 1.6 +++ mkepk.sh 17 Feb 2004 15:39:07 -0000 @@ -5,11 +5,11 @@ EPK_TMPDIR=/tmp/jffs2-epk-$$ MTDDIR=`dirname $0`/../../.. COREFILES="build.c compr_zlib.c LICENCE scan.c compr.c gc.c pushpull.h compr_rtime.c histo.h nodelist.c read.c write.c compr_rubin.c erase.c histo_mips.h nodelist.h readinode.c compr_rubin.h nodemgmt.c" ECOSFILES="os-ecos.h flashio.c gcthread.c dir-ecos.c fs-ecos.c malloc-ecos.c" INCFILES="jffs2.h jffs2_fs_i.h jffs2_fs_sb.h" -DOCFILES="README.Locking CONTRIBUTORS TODO ecos/doc/TODO.eCos ecos/doc/readme.txt" +DOCFILES="README.Locking TODO ecos/doc/TODO.eCos ecos/doc/readme.txt" mkdir $EPK_TMPDIR || exit 1 mkdir -p $EPK_TMPDIR/fs/jffs2/current/src mkdir -p $EPK_TMPDIR/fs/jffs2/current/cdl mkdir -p $EPK_TMPDIR/fs/jffs2/current/include/linux Index: cdl/jffs2.cdl =================================================================== RCS file: /home/cvs/mtd/fs/jffs2/ecos/cdl/jffs2.cdl,v retrieving revision 1.13 diff -u -5 -p -r1.13 jffs2.cdl --- cdl/jffs2.cdl 29 Nov 2003 00:30:30 -0000 1.13 +++ cdl/jffs2.cdl 17 Feb 2004 15:39:07 -0000 @@ -198,10 +198,21 @@ cdl_package CYGPKG_FS_JFFS2 { building the JFFS2 package. These flags are used in addition to the set of global flags." } + cdl_option CYGPKG_FS_JFFS2_CFLAGS_REMOVE { + display "Suppressed compiler flags" + flavor data + no_define + default_value { "" } + description " + This option modifies the set of compiler flags for + building the JFFS2 package. These flags are removed from + the set of global flags if present." + } + # ---------------------------------------------------------------- # Tests cdl_option CYGPKG_FS_JFFS2_TESTS { display "JFFS2 FS tests" --=-jitzgZ8XCG86qpzSJNJ+--