Re: [linux-sunxi] Banana Pi, Kernel 5.9, packet loss and bad network connectivity

2020-11-12 Thread Marc Haber
On Sun, Nov 08, 2020 at 03:55:49PM +0100, Jernej Škrabec wrote:
> 1. Apply DT fix: https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/
> linux.git/commit/?h=sunxi/fixes-

I can confirm that this one-liner fixes the issue on my Banana Pis.

Greetings
Marc

-- 
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20201112110641.GD12779%40torres.zugschlus.de.


Re: [linux-sunxi] Banana Pi, Kernel 5.9, packet loss and bad network connectivity

2020-11-08 Thread Marc Haber
On Mon, Oct 19, 2020 at 12:14:29PM +0800, Chen-Yu Tsai wrote:
> The proper fix now is to check the schematics, and change the phy-mode 
> property
> to the correct type. It used to be that we used "rgmii" for all RGMII PHYs.
> 
> Turns out there are  four modes of RGMII, depending on whether the MAC or PHY
> enable TX / RX signal delays.
> 
> Most boards in sunxi are designed so that the PHY is providing both TX and RX
> delays, which is done by pulling up the TXDLY and RXDLY pins at power up.
> 
> The commit "net: phy: realtek: fix rtl8211e rx/tx delay config" fixes the PHY
> driver so that it will correctly "reconfigure" the PHY to the requested mode.
> Previously this was not happening due to incorrect register definitions.
> 
> To summarize, the PHY driver fix exposes errors in our DTs.

The Banana Pi still doesn't work properly in kernel 5.9.6. I am not in
a position to check schematics and to fix DTs.

Look at git log for arch/arm/boot/dts/sun7i-a20-bananapi.dts, that would
be you, Chen-Yu Tsai and Maxime Ripard.

Is there anything I can do to help debugging?

Greetings
Marc

-- 
-----
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20201108144454.GB12779%40torres.zugschlus.de.


Re: [linux-sunxi] Banana Pi, Kernel 5.9, packet loss and bad network connectivity

2020-10-16 Thread Marc Haber
Hi Jernej,

On Thu, Oct 15, 2020 at 09:23:37PM +0200, Jernej Škrabec wrote:
> Dne četrtek, 15. oktober 2020 ob 21:11:39 CEST je Marc Haber napisal(a):
> > when I boot my Banana Pi (with Debian buster installed) with kernel 5.9,
> > the network interface doesn't work too well. Packet loss of more than
> > 50 %, machine nearly unuseable. Going back to 5.8.13 fixes the issue,
> > going forward to 5.9 again makes the issue reappear.
> > 
> > Does anybody have a good guess what to try first before going the
> > "implant 5.8.13 stmmac driver to 5.9, maybe bisect" route?
> 
> Try to revert https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/
> linux.git/commit/?h=v5.9=bbc4d71d63549bcd003a430de18a72a742d8c91e
> 
> This commit was found to cause issue on Beelink GS1 (Allwinner H6) which has 
> RTL8211e PHY. Since Banana Pi have same PHY, it might cause issue there too.

Reverting this commit does actually help here. Will there be a fix in
Greg's kernel releases?

Thank you for helping!

Greetings
Marc

-- 
-----
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20201016184438.GR32306%40torres.zugschlus.de.


[linux-sunxi] Banana Pi, Kernel 5.9, packet loss and bad network connectivity

2020-10-15 Thread Marc Haber
Hi,

when I boot my Banana Pi (with Debian buster installed) with kernel 5.9,
the network interface doesn't work too well. Packet loss of more than
50 %, machine nearly unuseable. Going back to 5.8.13 fixes the issue,
going forward to 5.9 again makes the issue reappear.

Does anybody have a good guess what to try first before going the
"implant 5.8.13 stmmac driver to 5.9, maybe bisect" route?

Greetings
Marc

-- 
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20201015191139.GP32306%40torres.zugschlus.de.


Re: [linux-sunxi] Re: Linux 5.2 leaves Banana Pi networkless

2019-08-25 Thread Marc Haber
On Fri, Aug 23, 2019 at 12:21:18PM -0700, Diego Karbon wrote:
> On Thursday, 11 July 2019 22:32:31 UTC+2, Marc Haber wrote:
> > On Thu, Jul 11, 2019 at 11:54:12PM +0800, Chen-Yu Tsai wrote: 
> > > CONFIG_CLK_SUNXI=y 
> > > CONFIG_CLK_SUNXI_CLOCKS=y 
> > > 
> > > in your config? 
> > > 
> >
> > That nailed it. Set the two options, and the box runs again. Thanks! 
> >
> > Greetings 
> > Marc 
> >
> Guys, I have similar problem with OLinuXino Lime. Where should I set those 
> options? In config for recompiling the kernel or in the other place?

Yes, those configuration options go in the kernel config (make
menuconfig/nconfig et al) and recompile the kernel.

Greetings
Marc

-- 
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20190825103635.GD7019%40torres.zugschlus.de.


Re: [linux-sunxi] Re: Linux 5.2 leaves Banana Pi networkless

2019-07-11 Thread Marc Haber
On Thu, Jul 11, 2019 at 11:54:12PM +0800, Chen-Yu Tsai wrote:
> CONFIG_CLK_SUNXI=y
> CONFIG_CLK_SUNXI_CLOCKS=y
> 
> in your config?
> 
> The A20 GMAC clock driver has not been ported to the newer
> sunxi-ng / SUNXI_CCU driver bundle, so you need to enable
> the legacy clock drivers.

That nailed it. Set the two options, and the box runs again. Thanks!

Greetings
Marc

-- 
-----
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20190711203228.GP21539%40torres.zugschlus.de.
For more options, visit https://groups.google.com/d/optout.


[linux-sunxi] Re: Linux 5.2 leaves Banana Pi networkless

2019-07-11 Thread Marc Haber
On Wed, Jul 10, 2019 at 10:50:35PM +0200, Marc Haber wrote:
> and the running system has no network interface. Suspicious entries
> would be:
> 
> |Jul 10 22:39:15 entrada kernel: [6.510668] sun7i-dwmac 1c5.ethernet: 
> PTP uses main clock
> |Jul 10 22:39:15 entrada kernel: [6.516592] sun7i-dwmac 1c5.ethernet: 
> no reset control found
> |Jul 10 22:39:15 entrada kernel: [6.527634] sun7i-dwmac 1c5.ethernet: 
> could not get tx clock

This is the unsuccessful result of an afternoon of bisecting:

git bisect start
# bad: [0ecfebd2b52404ae0c54a878c872bb93363ada36] Linux 5.2
git bisect bad 0ecfebd2b52404ae0c54a878c872bb93363ada36
# good: [e93c9c99a629c61837d5a7fc2120cd2b6c70dbdd] Linux 5.1
git bisect good e93c9c99a629c61837d5a7fc2120cd2b6c70dbdd
# good: [e93c9c99a629c61837d5a7fc2120cd2b6c70dbdd] Linux 5.1
git bisect good e93c9c99a629c61837d5a7fc2120cd2b6c70dbdd
# good: [a2d635decbfa9c1e4ae15cb05b68b2559f7f827c] Merge tag
# 'drm-next-2019-05-09' of git://anongit.freedesktop.org/drm/drm
git bisect good a2d635decbfa9c1e4ae15cb05b68b2559f7f827c
# bad: [22c58fd70ca48a29505922b1563826593b08cc00] Merge tag 'armsoc-soc'
# of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
git bisect bad 22c58fd70ca48a29505922b1563826593b08cc00
# bad: [c367dc8d0dd2a1e1ed9fdc2dd831053bdfdf0968] Merge branch
# 'next-tomoyo2' of
# git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security
git bisect bad c367dc8d0dd2a1e1ed9fdc2dd831053bdfdf0968
# good: [45182e4e1f8ac04708ca7508c51d9103f07d81ab] Merge branch
# 'i2c/for-5.2' of
# git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
git bisect good 45182e4e1f8ac04708ca7508c51d9103f07d81ab
# bad: [9b6c9e96f941c5ab13dad7278a3622f58e5672fc] Merge
# git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc
git bisect bad 9b6c9e96f941c5ab13dad7278a3622f58e5672fc
# bad: [c1157f60d72e8b20efc670cef28883832f42406c] Merge branch
# 'clk-parent-rewrite-1' into clk-next
git bisect bad c1157f60d72e8b20efc670cef28883832f42406c

Already c1157f60d72e8b20efc670cef28883832f42406c wouldn't boot,
but one can see the error message "could not get tx clock" before both
CPU cores get stalled.

The next commit was f6111b9d7970df07f783d3891735003fef2bc37e, which also
stalls on both CPU cores, but without emitting the "could not get tx
clock" message. I don't know whether this kernel is good or bad.

git log 
45182e4e1f8ac04708ca7508c51d9103f07d81ab..c1157f60d72e8b20efc670cef28883832f42406c

lets me end up with 189 lines of log

A second unsuccessful result of bisecting just
drivers/net/ethernet/stmicro/stmmac:

git bisect start 'drivers/net/ethernet/stmicro/stmmac'
# bad: [0ecfebd2b52404ae0c54a878c872bb93363ada36] Linux 5.2
git bisect bad 0ecfebd2b52404ae0c54a878c872bb93363ada36
# good: [e93c9c99a629c61837d5a7fc2120cd2b6c70dbdd] Linux 5.1
git bisect good e93c9c99a629c61837d5a7fc2120cd2b6c70dbdd
# bad: [f75b6f303bd80249a56cce9028954b4f731270e3] Merge 
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
git bisect bad f75b6f303bd80249a56cce9028954b4f731270e3
# bad: [d4c26eb6e721683a0f93e346ce55bc8dc3cbb175] net: ethernet: stmmac: 
dwmac-sun8i: enable support of unicast filtering
git bisect bad d4c26eb6e721683a0f93e346ce55bc8dc3cbb175

e9989339063dd3a566b67caaa857e6077bfee2cb ends up in a CPU soft lockup as
well.

Am therefore out of ideas. Help?

Greetings
Marc


-- 
-----
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20190711154605.GM21539%40torres.zugschlus.de.
For more options, visit https://groups.google.com/d/optout.


Re: [linux-sunxi] Linux 5.2 leaves Banana Pi networkless

2019-07-11 Thread Marc Haber
On Thu, Jul 11, 2019 at 01:52:03PM +0200, Marc Haber wrote:
> 0ae1f46c55f87e864c14fc46bd6cf0496fd391d7,
> 179930a62f3182eb384ae1e4d1ad47e97f6f5e90 and
> 82b3ad7445968ba70edc42d97ba3510c4623679a
> 
> There is also a fourth one preventing a soft CPU lockup at boot that
> made it into Greg's releases, I'm currently searching for it again to be
> able to bisect.

That one would be f87391558acf816b48f325a493d81d45dec40da0

Greetings
Marc

-- 
-----
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/2019075604.GJ21539%40torres.zugschlus.de.
For more options, visit https://groups.google.com/d/optout.


Re: [linux-sunxi] Linux 5.2 leaves Banana Pi networkless

2019-07-11 Thread Marc Haber
On Thu, Jul 11, 2019 at 09:28:02AM +0200, Corentin Labbe wrote:
> On Wed, Jul 10, 2019 at 10:50:35PM +0200, Marc Haber wrote:
> > Hi,
> > 
> > 5.2 is another mainline kernel that doesn't work on the Banana Pi. On
> 
> 
> Hello, which banana pi ? there are so many.

How do I find out? The dtb that is being used is the
sun7i-a20-bananapi.dtb, it's an Cortex-A7 CPU stepping r0p4. I just
bought it as a "Banana Pi", it has SATA and Gigabit Ethernet.

> > boot, the syslog is spewed with hundreds of lines of
> > 
> > | sun4i-ss 1c15000.crypto-engine: ERROR: Invalid keylen 26
> > 
> 
> This is due to extended crypto tests and is removed in my latest version of 
> crypto engine.
> And this is unrelated to your network issue.

... so I can ignore it and the box will run just fine?

> > and the running system has no network interface. Suspicious entries
> > would be:
> > 
> > |Jul 10 22:39:15 entrada kernel: [6.510668] sun7i-dwmac 
> > 1c5.ethernet: PTP uses main clock
> > |Jul 10 22:39:15 entrada kernel: [6.516592] sun7i-dwmac 
> > 1c5.ethernet: no reset control found
> > |Jul 10 22:39:15 entrada kernel: [6.527634] sun7i-dwmac 
> > 1c5.ethernet: could not get tx clock
> > 
> > while a working kernel (5.1.16 with the three patches that are crucial
> > for the Banana Pi but never made it into one of Greg's releases) shows
> > this:
> 
> What are thoses 3 patches ?

0ae1f46c55f87e864c14fc46bd6cf0496fd391d7,
179930a62f3182eb384ae1e4d1ad47e97f6f5e90 and
82b3ad7445968ba70edc42d97ba3510c4623679a

There is also a fourth one preventing a soft CPU lockup at boot that
made it into Greg's releases, I'm currently searching for it again to be
able to bisect.

Greetings
Marc

-- 
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/2019075203.GI21539%40torres.zugschlus.de.
For more options, visit https://groups.google.com/d/optout.


[linux-sunxi] Linux 5.2 leaves Banana Pi networkless

2019-07-10 Thread Marc Haber
Hi,

5.2 is another mainline kernel that doesn't work on the Banana Pi. On
boot, the syslog is spewed with hundreds of lines of

| sun4i-ss 1c15000.crypto-engine: ERROR: Invalid keylen 26

and the running system has no network interface. Suspicious entries
would be:

|Jul 10 22:39:15 entrada kernel: [6.510668] sun7i-dwmac 1c5.ethernet: 
PTP uses main clock
|Jul 10 22:39:15 entrada kernel: [6.516592] sun7i-dwmac 1c5.ethernet: 
no reset control found
|Jul 10 22:39:15 entrada kernel: [6.527634] sun7i-dwmac 1c5.ethernet: 
could not get tx clock

while a working kernel (5.1.16 with the three patches that are crucial
for the Banana Pi but never made it into one of Greg's releases) shows
this:

|Jul 10 22:40:47 entrada kernel: [6.531594] sun7i-dwmac 1c5.ethernet: 
PTP uses main clock
|Jul 10 22:40:47 entrada kernel: [6.537596] sun7i-dwmac 1c5.ethernet: 
no reset control found
|Jul 10 22:40:47 entrada kernel: [6.659144] sun7i-dwmac 1c5.ethernet: 
Version ID not available
|Jul 10 22:40:47 entrada kernel: [6.665491] sun7i-dwmac 1c5.ethernet:   
 DWMAC1000
|Jul 10 22:40:47 entrada kernel: [6.670577] sun7i-dwmac 1c5.ethernet: 
DMA HW capability register supported
|Jul 10 22:40:47 entrada kernel: [6.677877] sun7i-dwmac 1c5.ethernet: 
Normal descriptors
|Jul 10 22:40:47 entrada kernel: [6.683664] sun7i-dwmac 1c5.ethernet: 
Ring mode enabled
|Jul 10 22:40:47 entrada kernel: [   12.814713] sun7i-dwmac 1c5.ethernet 
eth0: No Safety Features support found
|Jul 10 22:40:47 entrada kernel: [   12.814735] sun7i-dwmac 1c5.ethernet 
eth0: RX IPC Checksum Offload disabled
|Jul 10 22:40:47 entrada kernel: [   12.814750] sun7i-dwmac 1c5.ethernet 
eth0: No MAC Management Counters available
|Jul 10 22:40:47 entrada kernel: [   12.814763] sun7i-dwmac 1c5.ethernet 
eth0: PTP not supported by HW
|Jul 10 22:40:50 entrada kernel: [   17.928555] sun7i-dwmac 1c5.ethernet 
eth0: Link is Up - 1Gbps/Full - flow control off

Am I the only Banana Pi user left, or why do so many mainline kernel
releases have crippling bugs for the BPi?

Greetings
Marc

-- 
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20190710205035.GA21539%40torres.zugschlus.de.
For more options, visit https://groups.google.com/d/optout.


[linux-sunxi] Re: Linux 5.1 not booting on banana Pi

2019-05-14 Thread Marc Haber
Hi Eric,

On Mon, May 13, 2019 at 10:51:41AM -0700, Eric Biggers wrote:
>   crypto: sun4i-ss - Handle better absence/presence of IV

82b3ad7445968ba70edc42d97ba3510c4623679a

>   crypto: sun4i-ss - remove ivsize from ECB

179930a62f3182eb384ae1e4d1ad47e97f6f5e90

>   crypto: sun4i-ss - Fix invalid calculation of hash end

f87391558acf816b48f325a493d81d45dec40da0

>   crypto: sun4i-ss - fallback when length is not multiple of blocksize

0ae1f46c55f87e864c14fc46bd6cf0496fd391d7

> Corentin, if these actually fix this problem, I suggest sending a request to
> sta...@vger.kernel.org to get these fixes picked up for 5.1.y.

I can confirm that applying those four patches to a 5.1.1 tree results
in a bootable and working kernel on the Banana Pi even with the extra
crypto tests enabled.

Thanks for helping!

Greetings
Marc

-- 
-----
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20190514105621.GW3835%40torres.zugschlus.de.
For more options, visit https://groups.google.com/d/optout.


[linux-sunxi] Re: Linux 5.1 not booting on banana Pi

2019-05-13 Thread Marc Haber
On Mon, May 13, 2019 at 02:32:55PM +0800, Herbert Xu wrote:
> On Fri, May 10, 2019 at 11:38:04AM +0200, Marc Haber wrote:
> > Bisecting the kernel identified 4e7babba30d820c4195b1d58cf51dce3c22ecf2b
> > which is unfortunately too complex to validly roll back (adding Eric and
> > Herbert, Authors and Off-Signers to Cc).
> > 
> > As a short-term workaround, setting CONFIG_CRYPTO_MANAGER_DISABLE_TESTS
> > seems to help.
> 
> You can disable the new tests using CONFIG_CRYPTO_MANAGER_EXTRA_TESTS.

Hang also happens with Linux 5.1.1 with
CONFIG_CRYPTO_MANAGER_EXTRA_TESTS disabled.

> Please also give the latest kernel a try as it has a number of fixes
> to the sun4i driver.

What would the "latest kernel" be? 5.1.1?

Greetings
Marc

-- 
-----
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20190513070644.GS3835%40torres.zugschlus.de.
For more options, visit https://groups.google.com/d/optout.


[linux-sunxi] Re: Linux 5.1 not booting on banana Pi

2019-05-10 Thread Marc Haber
Bisecting the kernel identified 4e7babba30d820c4195b1d58cf51dce3c22ecf2b
which is unfortunately too complex to validly roll back (adding Eric and
Herbert, Authors and Off-Signers to Cc).

As a short-term workaround, setting CONFIG_CRYPTO_MANAGER_DISABLE_TESTS
seems to help.

Greetings
Marc


On Wed, May 08, 2019 at 10:23:40PM +0200, Marc Haber wrote:
> From: Marc Haber 
> Subject: Linux 5.1 not booting on banana Pi
> To: linux-sunxi 
> Date: Wed, 8 May 2019 22:23:40 +0200
> User-Agent: Mutt/1.10.1 (2018-07-13)
> 
> Hi,
> 
> when I try booting Linux 5.1 on the Banana Pi, it stops with locked up
> CPUs:
> 
> [  OK  ] Started udev Coldplug all Devices.
> [  OK  ] Started Create Volatile Files and Directories.
> [  OK  ] Reached target System Time Synchronized.
>  Starting Update UTMP about System Boot/Shutdown...
> [  OK  ] Started Update UTMP about System Boot/Shutdown.
> [  OK  ] Started udev Kernel Device Manager.
> [   34.271840] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
> [   34.277793] rcu: 1-...!: (1 GPs behind) idle=fde/1/0x4000 
> softirq=1817/1820 fqs=23 
> [   34.285872] rcu: (detected by 0, t=5252 jiffies, g=1857, q=817)
> [   44.294216] rcu: rcu_sched kthread starved for 5206 jiffies! g1857 f0x0 
> RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=0
> [   44.304202] rcu: RCU grace-period kthread stack dump:
> [   68.019833] watchdog: BUG: soft lockup - CPU#0 stuck for 22s! 
> [systemd-udevd:260]
> [   96.019831] watchdog: BUG: soft lockup - CPU#0 stuck for 22s! 
> [systemd-udevd:260]
> [   97.291832] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
> [   97.297760] rcu: 1-...!: (1 GPs behind) idle=fde/1/0x4000 
> softirq=1817/1820 fqs=23 
> [   97.305838] rcu: (detected by 0, t=21007 jiffies, g=1857, q=817)
> [  107.314261] rcu: rcu_sched kthread starved for 20961 jiffies! g1857 f0x2 
> RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=0
> [  107.324333] rcu: RCU grace-period kthread stack dump:
> [  132.019830] watchdog: BUG: soft lockup - CPU#0 stuck for 23s! 
> [systemd-udevd:260]
> [  160.019832] watchdog: BUG: soft lockup - CPU#0 stuck for 23s! 
> [systemd-udevd:260]
> [  160.311830] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
> [  160.317758] rcu: 1-...!: (1 GPs behind) idle=fde/1/0x4000 
> softirq=1817/1820 fqs=23 
> [  160.325836] rcu: (detected by 0, t=36762 jiffies, g=1857, q=817)
> [  170.334257] rcu: rcu_sched kthread starved for 36716 jiffies! g1857 f0x2 
> RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=0
> [  170.344330] rcu: RCU grace-period kthread stack dump:
> 
> Any ideas what the problem could be, and how it could be solved?
> 
> Greetings
> Marc
> 
> -- 
> -
> Marc Haber | "I don't trust Computers. They | Mailadresse im Header
> Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
> Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20190510093804.GH26031%40torres.zugschlus.de.
For more options, visit https://groups.google.com/d/optout.


[linux-sunxi] Linux 5.1 not booting on banana Pi

2019-05-08 Thread Marc Haber
Hi,

when I try booting Linux 5.1 on the Banana Pi, it stops with locked up
CPUs:

[  OK  ] Started udev Coldplug all Devices.
[  OK  ] Started Create Volatile Files and Directories.
[  OK  ] Reached target System Time Synchronized.
 Starting Update UTMP about System Boot/Shutdown...
[  OK  ] Started Update UTMP about System Boot/Shutdown.
[  OK  ] Started udev Kernel Device Manager.
[   34.271840] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
[   34.277793] rcu: 1-...!: (1 GPs behind) idle=fde/1/0x4000 
softirq=1817/1820 fqs=23 
[   34.285872] rcu: (detected by 0, t=5252 jiffies, g=1857, q=817)
[   44.294216] rcu: rcu_sched kthread starved for 5206 jiffies! g1857 f0x0 
RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=0
[   44.304202] rcu: RCU grace-period kthread stack dump:
[   68.019833] watchdog: BUG: soft lockup - CPU#0 stuck for 22s! 
[systemd-udevd:260]
[   96.019831] watchdog: BUG: soft lockup - CPU#0 stuck for 22s! 
[systemd-udevd:260]
[   97.291832] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
[   97.297760] rcu: 1-...!: (1 GPs behind) idle=fde/1/0x4000 
softirq=1817/1820 fqs=23 
[   97.305838] rcu: (detected by 0, t=21007 jiffies, g=1857, q=817)
[  107.314261] rcu: rcu_sched kthread starved for 20961 jiffies! g1857 f0x2 
RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=0
[  107.324333] rcu: RCU grace-period kthread stack dump:
[  132.019830] watchdog: BUG: soft lockup - CPU#0 stuck for 23s! 
[systemd-udevd:260]
[  160.019832] watchdog: BUG: soft lockup - CPU#0 stuck for 23s! 
[systemd-udevd:260]
[  160.311830] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
[  160.317758] rcu: 1-...!: (1 GPs behind) idle=fde/1/0x4000 
softirq=1817/1820 fqs=23 
[  160.325836] rcu: (detected by 0, t=36762 jiffies, g=1857, q=817)
[  170.334257] rcu: rcu_sched kthread starved for 36716 jiffies! g1857 f0x2 
RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=0
[  170.344330] rcu: RCU grace-period kthread stack dump:

Any ideas what the problem could be, and how it could be solved?

Greetings
Marc

-- 
-----
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/linux-sunxi/20190508202340.GD26031%40torres.zugschlus.de.
For more options, visit https://groups.google.com/d/optout.


[linux-sunxi] No USB with kernel 4.12

2017-07-23 Thread Marc Haber
Hi,

I am using Debian Linux on a row of Banana Pi's with a self-built kernel
from the mainline archive.

Since 4.12, my USB is not working any more. Devices are not seen, lsusb
returns silently with a return code of 1. I guess that the offending
part of an strace lsusb is:

socket(AF_NETLINK, SOCK_RAW|SOCK_CLOEXEC|SOCK_NONBLOCK, NETLINK_KOBJECT_UEVENT) 
= 3
setsockopt(3, SOL_SOCKET, SO_ATTACH_FILTER, "\n\0\0\0T0\345\276", 8) = 0
bind(3, {sa_family=AF_NETLINK, nl_pid=0, nl_groups=0x02}, 12) = 0
getsockname(3, {sa_family=AF_NETLINK, nl_pid=26668, nl_groups=0x02}, [12]) 
= 0
setsockopt(3, SOL_SOCKET, SO_PASSCRED, [1], 4) = 0 fcntl64(3, F_GETFL)  
   = 0x802 (flags O_RDWR|O_NONBLOCK)
fcntl64(3, F_SETFL, O_RDWR|O_NONBLOCK)  = 0
[pid 26668] recvmsg(3, {msg_namelen=128}, 0) = -1 EAGAIN (Resource temporarily 
unavailable)

The modules do all load, but somehow the subsystem is not initialized
correctly.

I tried bisecting the kernel between v4.11 and v4.12, and between v4.11
and v4.12-rc1, and ended up twice with the first commit being

# first bad commit: [df6f37225d62316f8505f10ef1099ee40fa3978b] net: realtek: 
8139cp: use new api ethtool_{get|set}_link_ksettings

which is a commit that has nothing to do with USB. In fact, the 8139
driver is not even enabled in the kernel I build.

I cannot make any sense from this. Can you?

Greetings
Marc

-- 
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.