Message ID | 4FC49407.6040700@canonical.com |
---|---|
State | New |
Headers | show |
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 On 05/29/2012 03:16 AM, Paolo Pisati wrote: > The following changes since commit > 0e97c1de13c74ac21c291d7f4e27977b4f144465: > > UBUNTU: Ubuntu-3.4.0-201.2 (2012-05-22 08:39:58 -0600) > > are available in the git repository at: > > git://kernel.ubuntu.com/ppisati/ubuntu-quantal.git > tags/Ubuntu-3.4.0-201.3 > > for you to fetch changes up to > b801977686ce19726221690538ff877f76f0cc3f: > > UBUNTU: Ubuntu-3.4.0-201.3 (2012-05-28 12:03:43 +0200) > > ---------------------------------------------------------------- > Ubuntu-3.4.0-201.3 > > ---------------------------------------------------------------- > Mark Brown (1): ASoC: sgtl5000: Convert mic bias to a supply > widget > > Paolo Pisati (8): UBUNTU: [Config] restore SND_SOC and SND_OMAP_SOC > support UBUNTU: SAUCE: omap-pm: export missing symbol > (omap_pm_get_dev_context_loss_count) UBUNTU: SAUCE: > snd_soc_omap_hdmi: add a dummy symbol to enforce dependency on > snd_soc_omap UBUNTU: SAUCE: create aess device at boot time even > when snd_soc_abe=m UBUNTU: SAUCE: omap4_abe: reuse #included > firmware UBUNTU: SAUCE: snd_soc_abe_twl6040: add a dummy symbol to > enforce dependency on snd_soc_omap and snd_soc_mcpdm UBUNTU: Start > new release UBUNTU: Ubuntu-3.4.0-201.3 > > arch/arm/mach-omap2/devices.c | 1 + > arch/arm/plat-omap/omap-pm-noop.c | 2 + > debian.ti-omap4/abi/3.4.0-200.1/abiname | 1 - > .../abi/3.4.0-200.1/armhf/ignore.modules | 1 - > debian.ti-omap4/abi/3.4.0-200.1/armhf/omap4 |10299 > -------------------- .../abi/3.4.0-200.1/armhf/omap4.compiler > | 1 - debian.ti-omap4/abi/3.4.0-201.2/abiname | 1 > + debian.ti-omap4/abi/3.4.0-201.2/armhf/omap4 | 9861 > +++++++++++++++++++ .../abi/3.4.0-201.2/armhf/omap4.compiler > | 1 + .../armhf/omap4.modules | 225 > +- debian.ti-omap4/changelog | 20 + > debian.ti-omap4/config/config.common.ubuntu | 106 +- > sound/soc/codecs/sgtl5000.c | 6 +- > sound/soc/omap/omap-abe-core.c | 4 +- > sound/soc/omap/omap-abe-twl6040.c | 9 + > sound/soc/omap/omap-hdmi.c | 5 + > sound/soc/omap/omap-mcpdm.c | 6 + > sound/soc/omap/omap-pcm.c | 6 + 18 > files changed, 10036 insertions(+), 10519 deletions(-) delete mode > 100644 debian.ti-omap4/abi/3.4.0-200.1/abiname delete mode 100644 > debian.ti-omap4/abi/3.4.0-200.1/armhf/ignore.modules delete mode > 100644 debian.ti-omap4/abi/3.4.0-200.1/armhf/omap4 delete mode > 100644 debian.ti-omap4/abi/3.4.0-200.1/armhf/omap4.compiler create > mode 100644 debian.ti-omap4/abi/3.4.0-201.2/abiname create mode > 100644 debian.ti-omap4/abi/3.4.0-201.2/armhf/omap4 create mode > 100644 debian.ti-omap4/abi/3.4.0-201.2/armhf/omap4.compiler rename > debian.ti-omap4/abi/{3.4.0-200.1 => > 3.4.0-201.2}/armhf/omap4.modules (89%) > > Some info about the pull: > > "ASoC: sgtl5000: Convert mic bias to a supply widget" is an > upstream cherrypick (that fixes sgtl5000 compilation when =m). > > "snd_soc_omap_hdmi: add a dummy symbol" and "snd_soc_abe_twl6040: > add a dummy symbol" aren't striclty necessary but they add a > dependecy on some compulsory modules - withouth these we get a: > > "snd_soc_register failed: -517" > > and no clue of what's wrong - adding DEBUG to sound/soc/soc-core.c > let us know which DAIs&c are missing but we still have to figure > out which module implements what - adding a dependency make it > autoload at boot time, way simpler. > > "omap4_abe: reuse #included firmware" reuses a firmware file that > is already #included: while there are comments in the code about > loading the firmware dinamically if snd_soc_abe*=m, i don't have > this file and it adds another dependency (think of initrd or > another package to be shipped/installed) while the above include is > always present, explot it. > > "create aess device at boot time even when snd_soc_abe=m" is > needed because snd_abe* attaches to "platform:aess". > > The rest should be self explanatory. > > All of them were already forwarded to agreen/TI, but until they > show up in TI/tilt, i want to carry them as SAUCE so i can continue > my work. > Rather then these SAUCE patches to force load modules, why not just build them into the kernel ? Won't they load correctly then ? rtg - -- Tim Gardner tim.gardner@canonical.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBCgAGBQJPxL+nAAoJED12yEX6FEfKhz0P/2w7KNKHsp7JUbiC6QAIXNp4 TpaQmjAFuTlo4rh14qgsKgJMxcuaqgOOcFBxFHt0rWF3vz86w3+OWCgdX1l/9SbV tNfzsp1vZ1DLh8OWaNKVKGffIB4EIc4PCk/shuJZ8Cugj4hDVW1EOPj02/EwxIm5 HCAMH2SDGyckWwI1HKRXOc0y5LdeJBruAUAG5wGwe0pCMooTfnV2WJXSVT6JmFNJ XRlg1lHJSxeCTHE2ykgsN6JL1jT90fc+35V7ZaMtQEKko2+bDzsOE/zrhPBcMPm3 C97W3LdgEjJPG+uWyUjCMIyKBTa6rhn5t5y91ZezWlNwDEi83VjNwhRW0XA0lwcn GQzlHU2/8bLwnNtmB787945wIHnl69Y+qYSeeMopAJXs4oyXl5NFFUohORKJThbq Ln2DoOddi2WWPF3OFBijFlAO7AFoEkns07SF8DmYzD0sfGpuOFSwzvzmq7JgwsCB Df/DNDWSfgqLXtNo7rxm8oF6o5QWt74sIxzQbg2zzblvdao/7rF5V4OTfDgddp6D ss/xznT8Hu/mBx0Lqj3ZHyMAJeWOC/PhGNXx+RRfbN5LplGTR+lF8sDa/PTDhGTj gHH+wQ1EBGbKDZQLR8p1W6KpspclsK7obvtZd4Z0g87bIwPj80Ms+/m3KeeyWos2 xFeg9PMCBGE7HbX504t5 =sHRF -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 05/29/2012 02:23 PM, Tim Gardner wrote: > Rather then these SAUCE patches to force load modules, why not > just build them into the kernel ? Won't they load correctly then ? yes, but didn't we want to have everything as =m? - -- bye, p. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBAgAGBQJPxMU9AAoJEJdscTmfuQbQnWoP/3YHEnZHg7pO+rwYtlsEpBEZ wqyOMOK+DS+tIbjdE1B6W0W3HR5NHX3Nakjctepi8KRd+alIvEdYOh8RLGexqSy2 U8HAehdbbAXBatO/vFU+8FRQFBFdTCqDoobO1BQ372V7X3n/5F7nuwo8Y672BFNR 0Ay7l4vOJ3wmfCN9XQ3IdYRukrEy8D7U54IT0MFhL2WlgvRfDxg7DgGkc5t2NiGI Vesr74ESjzJV7USPTm7F8htUwDRvNRrpfUZgX91U8ibwUhT02y1yKFGdk2IXG73U IT2Ue7hXSEmwTnB4Q2fVm8nOylu2cp4ZCJOXXilbpstbkXlQa+yffWuuCokEo3X2 1OUy6WZ8hzqBneYRWVsKJqEsIiFv/DK7cD4Ofd8ohhEHyd3dJ8uiqivvxuIMnq6J ab9qYoAC/GUXhhDsI8B8ufgf0/RKh/V85Q20iOnkqcfHogteQTVVj9aCmG8r8Sgb 6Yeno0CY4+ZID2VYYS1lfbU5bx0hu1MvEIzKaoLDPek6/Fhh+Mu59PWx0y8uYIuK qimZP15af6reLiEd/sWlbqOhTgq8IVrLxnalmP/4UTkEg1jD9kxnN1mcl1hiofIZ GKcHjn1xvxUO0s2uGCNIh6bek4bZdQCeGZ+jzRyq0zIc6KH6yzceCexJiAGOvk62 Gmu1GSCp/Ryf9qtt36W/ =kxuj -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 On 05/29/2012 06:46 AM, Paolo Pisati wrote: > On 05/29/2012 02:23 PM, Tim Gardner wrote: > >> Rather then these SAUCE patches to force load modules, why not >> just build them into the kernel ? Won't they load correctly then >> ? > > yes, but didn't we want to have everything as =m? Maybe for the mainline distro kernel, but this is targeted to a specific platform. I'd just do the right thing with as few hacks as possible. rtg - -- Tim Gardner tim.gardner@canonical.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBCgAGBQJPxMjHAAoJED12yEX6FEfKrsEP/1QOfwZRsKSfqj7E7dNZhpwG pgW6VzD0monMZVM/eKs4O6ZRnrIKBTVTI/C2cRP/75DsTqWTEuY10johGMIkNmmB tSuNmu8nC6t0Zyo+iVpPD0tJdS+nwPWkRbicvVYAcehINa2UFd12of7CpaZEsP2n 16n+LOOeiMKBLSm/NxbrvVIRYq/UKeM7p8gTqAZNlS2tO6kFHIcsSAnUeQSsBaUt zlJbRPsSJWywONdeslNOIhwxTlZUQuINNJPjypLNTZLQ4ikcxq9LrRPha2mq8GmR fGqOWvrc51YXkJdjy9IYvi07BjZ82R3hZqFq8L38NmHk8cd60+ZgRs6AhuvQrE0w By7rc6UBuXnNSpcXb7gr+IZKm3bwRJZGp7cJo0N34E3LVO2UEzKzBWNTLisjqSE8 3IFJI6PUtT0aYBxENlOanGFRKs9B1DP3CkjY8R6W76EQ8jA2GzlAnvGAjg0yq1Ow I3OfuSq703lTYMwntshQq8QAfxzC81sd43CpTx+a6XR8GYyLlDc08OoOOOJgjmYV Fzsxfpdb8CwXGPy5FTPQuKF9ScFbzCEnJmrTi1t6KNwmXnFKLbN8erO3+IV5bO3V 7+yjn8MApG2RNET5Yj6swzjhLAiTf/BzDWEtlvHzDbS25jAXDB+D2YzmZphTVMRX 21Db3h6U3bqXfRzDHKDR =l3bP -----END PGP SIGNATURE-----