Re: 4.5-rc0 regression: GPRS breaks after very short while
On Fri 2016-05-20 06:12:59, Sebastian Reichel wrote: > Hi, > > On Wed, May 18, 2016 at 10:12:51PM +0200, Pavel Machek wrote: > > > The last one only changes a comment, so not relevant, the other > > > two fix error handling during initialization and are most likely > > > also not the problem. Can you try to bisect the problem? > > > > It will not be fun, but I guess it is doable. But before I start it > > would be good to know -- does it work for you? > > I currently don't have a second SIM card with network access for my > dev phone, so I usually only test without gprs connection. As a > result I don't really know if longer data connection would work. > > Also I'm quite busy until next weekend with other stuff. If you did > not find anything until then, I will do some testing/debugging. Ok, it looks like problems got less reproducible for me. Best regards, Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Re: 4.5-rc0 regression: GPRS breaks after very short while
Hi, On Wed, May 18, 2016 at 10:12:51PM +0200, Pavel Machek wrote: > > The last one only changes a comment, so not relevant, the other > > two fix error handling during initialization and are most likely > > also not the problem. Can you try to bisect the problem? > > It will not be fun, but I guess it is doable. But before I start it > would be good to know -- does it work for you? I currently don't have a second SIM card with network access for my dev phone, so I usually only test without gprs connection. As a result I don't really know if longer data connection would work. Also I'm quite busy until next weekend with other stuff. If you did not find anything until then, I will do some testing/debugging. -- Sebastian signature.asc Description: PGP signature
Re: 4.5-rc0 regression: GPRS breaks after very short while
Hi! > On Wed, May 18, 2016 at 09:49:24AM +0200, Pavel Machek wrote: > > I was afraid another GSM chip desoldered itself in N900, but no, it > > seems to be software this time. > > > > In 4.4, GPRS is rock solid, can survive overnight ping. > > > > In 4.5-rc0 and v4.6, modem fails when I use gprs for like a > > minute. Phone is not usable. > > I only send three modem related changes for v4.5: > > https://git.kernel.org/cgit/linux/kernel/git/sre/linux-hsi.git/commit/?h=hsi-for-4.5&id=525e1abc6b5a7d8bd9006de1da6e99722305ea2b > HSI: omap_ssi: fix handling ida_simple_get result > > https://git.kernel.org/cgit/linux/kernel/git/sre/linux-hsi.git/commit/?h=hsi-for-4.5&id=6bf6ded3008a407ccab5ad72eb59ab71b97cc290 > HSI: omap_ssi_port: fix handling of_get_named_gpio result > > https://git.kernel.org/cgit/linux/kernel/git/sre/linux-hsi.git/commit/?h=hsi-for-4.5&id=403c5c0650816375527a6feecfb255d9b494dda3 > HSI: Remove struct hsi_client private fields from kernel-doc > > The last one only changes a comment, so not relevant, the other > two fix error handling during initialization and are most likely > also not the problem. Can you try to bisect the problem? It will not be fun, but I guess it is doable. But before I start it would be good to know -- does it work for you? Thanks and best regards, Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Re: 4.5-rc0 regression: GPRS breaks after very short while
Hi Pavel, On Wed, May 18, 2016 at 09:49:24AM +0200, Pavel Machek wrote: > I was afraid another GSM chip desoldered itself in N900, but no, it > seems to be software this time. > > In 4.4, GPRS is rock solid, can survive overnight ping. > > In 4.5-rc0 and v4.6, modem fails when I use gprs for like a > minute. Phone is not usable. I only send three modem related changes for v4.5: https://git.kernel.org/cgit/linux/kernel/git/sre/linux-hsi.git/commit/?h=hsi-for-4.5&id=525e1abc6b5a7d8bd9006de1da6e99722305ea2b HSI: omap_ssi: fix handling ida_simple_get result https://git.kernel.org/cgit/linux/kernel/git/sre/linux-hsi.git/commit/?h=hsi-for-4.5&id=6bf6ded3008a407ccab5ad72eb59ab71b97cc290 HSI: omap_ssi_port: fix handling of_get_named_gpio result https://git.kernel.org/cgit/linux/kernel/git/sre/linux-hsi.git/commit/?h=hsi-for-4.5&id=403c5c0650816375527a6feecfb255d9b494dda3 HSI: Remove struct hsi_client private fields from kernel-doc The last one only changes a comment, so not relevant, the other two fix error handling during initialization and are most likely also not the problem. Can you try to bisect the problem? -- Sebastian signature.asc Description: PGP signature
4.5-rc0 regression: GPRS breaks after very short while
Hi! I was afraid another GSM chip desoldered itself in N900, but no, it seems to be software this time. In 4.4, GPRS is rock solid, can survive overnight ping. In 4.5-rc0 and v4.6, modem fails when I use gprs for like a minute. Phone is not usable. Best regards, Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html