From mboxrd@z Thu Jan 1 00:00:00 1970 From: Lukasz Majewski Date: Sat, 15 Jun 2019 17:28:49 +0200 Subject: [U-Boot] [PATCH v2 2/2] fastboot: Fix slot names reported by getvar In-Reply-To: <20190614155056.16749-3-semen.protsenko@linaro.org> References: <20190614155056.16749-1-semen.protsenko@linaro.org> <20190614155056.16749-3-semen.protsenko@linaro.org> Message-ID: <20190615172849.77c5915e@jawa> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Hi Sam, > Fastboot tool recently underwent changes w.r.t. A/B slot format: > 1. In commit [1] the new slot format was introduced, according to > new A/B specification [2]. New slot format is "a", and old format "_a" > is now considered legacy. > 2. In commit [3] the legacy format "_a" was fixed and fastboot tool > should support both "a" and "_a" slot formats now. > 3. Finally, commit [4] drops the legacy slot format ("_a") > completely. That makes the latest fastboot tool incompatible with > "_a" format. > > Last change leads to next error, when running "fastboot flash" with > current U-Boot: > > $ fastboot flash boot boot.img > Sending 'boot__a' (11301 KB) OKAY [ 0.451s] > Writing 'boot__a' FAILED (remote: 'cannot find > partition') fastboot: error: Command failed > > To overcome this issue we should report slot names in "a" format > instead of "_a". Of course, this change breaks U-Boot compatibility > with older fastboot tools, but that shouldn't be a problem as A/B is > not implemented in U-Boot yet, and there are not users for slotted > partitions out there anyway. This fact can be checked like this: > > $ grep -Ir \b'boot_a\b' * > > Let's use new slot format in order to fix "fastboot flash" with > slotted partitions and to be in sync with AOSP master. > > With this patch, U-Boot depends on most recent fastboot tool (patch > [1] or later), for working with slotted partitions. Is there any "version" number scheme for the fastboot protocol specification? I do remember that in the past there were some mismatches for some "fastboot" specification depending on vendor. Now it turns out that there is a problem between AOSP versions ... (My point is that for example even lthor had version number for specification update. Is something similar for fastboot?). > > [1] > https://android.googlesource.com/platform/system/core/+/8091947847d5e5130b09d2ac0a4bdc900f3b77c5 > [2] > https://source.android.com/devices/tech/ota/ab/ab_implement#partitions > [3] > https://android.googlesource.com/platform/system/core/+/04396f62da6150b94e02d50e5302fd980048833d > [4] > https://android.googlesource.com/platform/system/core/+/42b18a518bac85c3eea14206f6cbafbd1e98a31f > > Signed-off-by: Sam Protsenko > --- > Changes in v2: > - don't change slot format in "slot-suffixes" variable (it's now > dropped in [PATCH 1/2]) > - improve commit message > > drivers/fastboot/fb_getvar.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/fastboot/fb_getvar.c > b/drivers/fastboot/fb_getvar.c index f89c7f62e6..9ee5054485 100644 > --- a/drivers/fastboot/fb_getvar.c > +++ b/drivers/fastboot/fb_getvar.c > @@ -174,8 +174,8 @@ static void getvar_platform(char *var_parameter, > char *response) > static void getvar_current_slot(char *var_parameter, char *response) > { > - /* A/B not implemented, for now always return _a */ > - fastboot_okay("_a", response); > + /* A/B not implemented, for now always return "a" */ > + fastboot_okay("a", response); > } > > #if CONFIG_IS_ENABLED(FASTBOOT_FLASH) Best regards, Lukasz Majewski -- DENX Software Engineering GmbH, Managing Director: Wolfgang Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email: lukma at denx.de -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 488 bytes Desc: OpenPGP digital signature URL: