From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tom Rini Date: Mon, 7 Jan 2013 09:20:38 -0500 Subject: [U-Boot] [PATCH v2 5/6] mx6qsabrelite: Add support to dynamically choose between ftd use or not In-Reply-To: References: <1356722226-23186-1-git-send-email-otavio@ossystems.com.br> <1356722226-23186-6-git-send-email-otavio@ossystems.com.br> <20130107140451.GA3402@bill-the-cat> Message-ID: <50EAD9B6.1050305@ti.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 01/07/2013 09:17 AM, Otavio Salvador wrote: > On Mon, Jan 7, 2013 at 12:04 PM, Tom Rini wrote: >> On Fri, Dec 28, 2012 at 05:17:05PM -0200, Otavio Salvador wrote: >> >>> Signed-off-by: Otavio Salvador --- >>> Changes in v2: - Allow use of dynamic/static ip - Allow force >>> use, or not, of fdt - Change 'auto' to 'try' >> [snip] >>> @@ -169,13 +173,43 @@ "loaduimage=fatload mmc >>> ${mmcdev}:${mmcpart} ${loadaddr} ${uimage}\0" \ "mmcboot=echo >>> Booting from mmc ...; " \ "run mmcargs; " \ - "bootm\0" \ + >>> "if test ${boot_fdt} = yes; then " \ + "if >>> fatload mmc ${mmcdev}:${mmcpart} ${ftd_addr} ${ftd_file}; then >>> " \ >> >> How about adding loadfdtfile after 'loaduimage' above? > > The command for netboot is different so we'd need to have two vars > for it. I think it will be confusing ... Yes, but there's already loaduimage that's fixed to FAT from MMC. So you're being consistent. >>> + "bootm ${loadaddr} - >>> ${ftd_addr}; " \ + "else " \ + "if test >>> ${boot_fdt} = try; then " \ + "bootm; " \ + "else " \ + "echo >>> ERROR: Cannot load the DT, aborting...; " \ >> >> Strings must not be broken, so this is fine. But, you aren't >> really aborting. If you had a loop of "try mmcboot. Fail? Try >> netboot" it would continue. So the error message should perhaps >> just be about cannot load DT from mmc? > > I don't have a strong opinion about this. I'd prefer to abort here > so user really knows it failed. In case user wish it to not be > fatal it can use the 'try' in boot_fdt. > > How I could abort here? Without firing up a board, maybe tossing in break? Or setting a variable (abortcmd) that you could test for in other clauses to make sure it's obeyed and we break out? - -- Tom -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with undefined - http://www.enigmail.net/ iQIcBAEBAgAGBQJQ6tm2AAoJENk4IS6UOR1WefAQALNWepGwYYkyUkoj2YfqQQf6 D23SyuZRAEUKxi7mkFZdBEDwyIHVvM8v4HEXkxMigKq6jd++xpGWqOvygUhu7NCK i4vfPXadgyt+3BLt5BGLLQDf1GNrsTpxjPuw+AEDsdt0hgo9fWSoOegpsO2q6kIj xSspwUpzBE6Dtzh9uomRw7J/o4uvDdRv7MC4krWB70SC9y8hN0p/syV4cYvjqLTp WWawT7qKfNwkfPH5tAmPb5b2V3JVGSMkvOMPuH3QoNSfhVTSE6EdKK1v3WWa933Z ucAGtJgDlPgRhTioIp78Nsi1JXwygHByA/ZYnxre9jimJZ5lxd8m5TpqwgnpkJN0 2sUtSg3WcSSqr+ckMXOswKPfT9wx0Vgy+elP/i4LlJ3d1uoDJufLt+v0nkKYHDR0 foC0udYrjSDZiA1kmX8u4w+sVGWAlAgawkiDSA3hYY1Y+GUeVzKGxwMNSvLrzrWq ga3aAWbr26KIZaBBjHRPlkPWzo8rvLDCWEWQkSBnF+o3Zccc5tebGIhYp/eMxt+T ewUCT0pW5HuXAeRvqNPwdiGDw2qrvFcrd6rnVpSNic/0rYyJRTUa4hmWv/FZm8PQ whCijnBCBVRnBc+M7K2YJTFcRlR4oC0GnLe6K7gDGT5fhHln11AbHa0san8DkrnE bzvjfkjxGtsaNKzR7wNV =Ew4G -----END PGP SIGNATURE-----