| Message ID | Pine.NEB.4.64.1602191918380.15324@faeroes.freeshell.org (mailing list archive) |
|---|---|
| State | Accepted, archived |
| Headers | show |
On Fri, Feb 19, 2016 at 07:34:18PM -0600, John Dahlstrom wrote: > Some Lenovo ideapad models lack a physical rfkill switch. > On Lenovo models ideapad Y700 Touch-15ISK and ideapad Y700-15ISK, > ideapad-laptop would wrongly report all radios as blocked by > hardware which caused wireless network connections to fail. > > Add these models without an rfkill switch to the no_hw_rfkill list. > > Signed-off-by: John Dahlstrom <jodarom@sdf.org> > Cc: <stable@vger.kernel.org> # 4.5.x > Cc: <stable@vger.kernel.org> # 4.4.x > Cc: <stable@vger.kernel.org> # 4.3.x > Cc: <stable@vger.kernel.org> # 4.2.x > Cc: <stable@vger.kernel.org> # 4.1.x > Cc: <stable@vger.kernel.org> # 4.0.x > Cc: <stable@vger.kernel.org> # 3.19.x > Cc: <stable@vger.kernel.org> # 3.18.x > Cc: <stable@vger.kernel.org> # 3.17.x > Cc: <stable@vger.kernel.org> # 3.16.x Unfortunately, backporting this to stable is not quite so simple. First, 3.16 doesn't really work as between 3.16 and 3.17 the following patch landed: ce363c2 ideapad-laptop: Change Lenovo Yoga 2 series rfkill handling Which changes the name of the dmi_system_id struct from rfkill_blacklist to no_hw_rfkill_list. Following that, there were several additions to the list which should be applied before this patch to each stable kernel for which they haven't been picked up in order for this one to apply cleanly. Several of those are included below: $ git l v3.17.. drivers/platform/x86/ideapad-laptop.c b7287f0 ideapad-laptop: Add ideapad Y700 (15) to the no_hw_rfkill DMI list 6b31de3 ideapad-laptop: Add Lenovo Yoga 700 to no_hw_rfkill dmi list edde316 ideapad-laptop: Add Lenovo ideapad Y700-17ISK to no_hw_rfkill dmi list f71c882 ideapad-laptop: Add Lenovo Yoga 900 to no_hw_rfkill dmi list c789fff ideapad-laptop: include Yoga 3 1170 in add rfkill whitelist 74caab9 ideapad-laptop: add support for Yoga 3 ESC key 6d212b8 ideapad-laptop: add alternative representation for Yoga 2 to DMI table fa92a31 ideapad-laptop: Add Lenovo Yoga 3 14 to no_hw_rfkill dmi list 05fde26 Merge tag 'platform-drivers-x86-v4.2-1' of git://git.infradead.org/users/dvhart/linux-platform-drivers-x86 26bff5f ideapad-laptop: Port to new backlight interface selection API 4b200b4 ideapad: fix software rfkill setting 4fa9dab ideapad_laptop: Lenovo G50-30 fix rfkill reports wireless blocked 9b071a4 ideapad_laptop: Add Lenovo G40-30 to devices without radio switch 385336e Merge tag 'platform-drivers-x86-v3.19-1' of git://git.infradead.org/users/dvhart/linux-platform-drivers-x86 e6b5be2 Merge tag 'driver-core-3.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core 0098181 platform: x86: Deletion of checks before backlight_device_unregister() 725c7f6 quirk for Lenovo Yoga 3: no rfkill switch 3493f41 platform: x86: drop owner assignment from platform_drivers If you are going to specify a kernel version, you should also include the commits above necessary to make the patch apply cleanly. That would be a long list as you would need many of these for each version. What I'm going to do is include a single Cc to stable line without a kernel version. The maintainers will pull that back as far as they can using their own judgement. If you want this to go back earlier than they do on their own, you should submit it to linux-stable directly and include the Cc lines for the dependencies for each kernel you care to see this backported to. See the stable-kernel-rules for the specific formatting to accomplish this. > --- > Test configuration > Hardware: Lenovo ideapad Y700 Touch-15ISK > Kernel version: 4.4.2 > > Patch changelog > v2 split patch between Touch and non-Touch devices > v3 undo patch split and limit summary to 72 characters > v4 include comprehensive list of patchable kernels in Cc > > drivers/platform/x86/ideapad-laptop.c | 14 ++++++++++++++ > 1 file changed, 14 insertions(+) > > --- a/drivers/platform/x86/ideapad-laptop.c 2016-02-14 15:05:20.000000000 -0600 > +++ b/drivers/platform/x86/ideapad-laptop.c 2016-02-16 03:54:48.484423725 -0600 > @@ -864,4 +864,18 @@ static const struct dmi_system_id no_hw_rfkill_list[] = { > DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo G50-30"), > }, > + }, > + { > + .ident = "Lenovo ideapad Y700-15ISK", > + .matches = { > + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), > + DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo ideapad Y700-15ISK"), > + }, > + }, > + { > + .ident = "Lenovo ideapad Y700 Touch-15ISK", > + .matches = { > + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), > + DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo ideapad Y700 Touch-15ISK"), > + }, > }, > { >
On Mon, 22 Feb 2016, Darren Hart wrote: > Unfortunately, backporting this to stable is not quite so simple. > > First, 3.16 doesn't really work as between 3.16 and 3.17 the following patch > landed: > > ce363c2 ideapad-laptop: Change Lenovo Yoga 2 series rfkill handling > > Which changes the name of the dmi_system_id struct from rfkill_blacklist to > no_hw_rfkill_list. > > Following that, there were several additions to the list which should be applied > before this patch to each stable kernel for which they haven't been picked up in > order for this one to apply cleanly. Several of those are included below: > Despite the change in the no_hw_rfkill_list, GNU patch still yields the correct output but with fuzz. I interpret apply cleanly to mean that the patch must also apply with zero fuzz, such as with "patch -F 0". In the case where the context has changed (including changes to an enclosing struct or function), I gather that a patch exactly modified for an older kernel is to be sent to stable@vger.kernel.org after the unmodified patch is accepted upstream with a commit ID. > $ git l v3.17.. drivers/platform/x86/ideapad-laptop.c [...] > If you are going to specify a kernel version, you should also include the > commits above necessary to make the patch apply cleanly. That would be a long > list as you would need many of these for each version. > Thank you for that information. One commit is sufficient to apply the patch to all kernel versions without fuzz: 4fa9dab ideapad_laptop: Lenovo G50-30 fix rfkill reports wireless blocked > What I'm going to do is include a single Cc to stable line without a kernel > version. The maintainers will pull that back as far as they can using their own > judgement. If you want this to go back earlier than they do on their own, you > should submit it to linux-stable directly and include the Cc lines for the > dependencies for each kernel you care to see this backported to. See the > stable-kernel-rules for the specific formatting to accomplish this. > I've submitted v5 of the patch with a single Cc line with the prerequisite kernel(s) and commit ID specified exactly. Kind regards, John -- To unsubscribe from this list: send the line "unsubscribe platform-driver-x86" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Mon, Feb 22, 2016 at 10:29:48PM -0600, John Dahlstrom wrote: > On Mon, 22 Feb 2016, Darren Hart wrote: > > >Unfortunately, backporting this to stable is not quite so simple. > > > >First, 3.16 doesn't really work as between 3.16 and 3.17 the following patch > >landed: > > > >ce363c2 ideapad-laptop: Change Lenovo Yoga 2 series rfkill handling > > > >Which changes the name of the dmi_system_id struct from rfkill_blacklist to > >no_hw_rfkill_list. > > > >Following that, there were several additions to the list which should be applied > >before this patch to each stable kernel for which they haven't been picked up in > >order for this one to apply cleanly. Several of those are included below: > > > > Despite the change in the no_hw_rfkill_list, GNU patch still yields the > correct output but with fuzz. I interpret apply cleanly to mean that > the patch must also apply with zero fuzz, such as with "patch -F 0". > > In the case where the context has changed (including changes to an > enclosing struct or function), I gather that a patch exactly modified > for an older kernel is to be sent to stable@vger.kernel.org after the > unmodified patch is accepted upstream with a commit ID. > > >$ git l v3.17.. drivers/platform/x86/ideapad-laptop.c > [...] > > >If you are going to specify a kernel version, you should also include the > >commits above necessary to make the patch apply cleanly. That would be a long > >list as you would need many of these for each version. > > > > Thank you for that information. One commit is sufficient to apply the patch > to all kernel versions without fuzz: > > 4fa9dab ideapad_laptop: Lenovo G50-30 fix rfkill reports wireless blocked This does not yield correct results for me on 3.17 (or 3.17.8). Do you get different results from the following? dvhart@dvhart-mobl5:~/source/linux/linux-pdx86 [testing] $ git co -b 3.17 v3.17 Checking out files: 100% (36214/36214), done. Switched to a new branch '3.17' dvhart@dvhart-mobl5:~/source/linux/linux-pdx86 [3.17] $ git cp 4fa9dab [3.17 d2cdd70] ideapad_laptop: Lenovo G50-30 fix rfkill reports wireless blocked Author: Dmitry Tunin <hanipouspilot@gmail.com> Date: Sun Jan 18 15:44:40 2015 +0300 1 file changed, 7 insertions(+) dvhart@dvhart-mobl5:~/source/linux/linux-pdx86 [3.17] $ git am ~/incoming/PATCH_v5_1-1_ideapad-laptop_Add_ideapad_Y700_15_to_the_no_hw_rfkill_DMI_list.mbox Applying: ideapad-laptop: Add ideapad Y700 (15) to the no_hw_rfkill DMI list error: patch failed: drivers/platform/x86/ideapad-laptop.c:864 error: drivers/platform/x86/ideapad-laptop.c: patch does not apply Patch failed at 0001 ideapad-laptop: Add ideapad Y700 (15) to the no_hw_rfkill DMI list The copy of the patch that failed is found in: /home/dvhart/source/linux/linux-pdx86/.git/rebase-apply/patch When you have resolved this problem, run "git am --continue". If you prefer to skip this patch, run "git am --skip" instead. To restore the original branch and stop patching, run "git am --abort". dvhart@dvhart-mobl5:~/source/linux/linux-pdx86 [3.17|AM 1/1] $ patch -p1 < ~/incoming/PATCH_v5_1-1_ideapad-laptop_Add_ideapad_Y700_15_to_the_no_hw_rfkill_DMI_list.mbox patching file drivers/platform/x86/ideapad-laptop.c Hunk #1 succeeded at 864 with fuzz 2. $ cat drivers/platform/x86/ideapad-laptop.c | head -n 880 | tail -n 32 static int ideapad_acpi_add(struct platform_device *pdev) { int ret, i; int cfg; struct ideapad_private *priv; struct acpi_device *adev; ret = acpi_bus_get_device(ACPI_HANDLE(&pdev->dev), &adev); if (ret) return -ENODEV; if (read_method_int(adev->handle, "_CFG", &cfg)) return -ENODEV; priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL); if (!priv) }, { .ident = "Lenovo ideapad Y700-15ISK", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo ideapad Y700-15ISK"), }, }, { .ident = "Lenovo ideapad Y700 Touch-15ISK", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo ideapad Y700 Touch-15ISK"), }, return -ENOMEM; Note that the DMI match blocks were stuffed in the middle of the ideapad_acpi_add() function instead of the no_hw_rfkill. > > >What I'm going to do is include a single Cc to stable line without a kernel > >version. The maintainers will pull that back as far as they can using their own > >judgement. If you want this to go back earlier than they do on their own, you > >should submit it to linux-stable directly and include the Cc lines for the > >dependencies for each kernel you care to see this backported to. See the > >stable-kernel-rules for the specific formatting to accomplish this. > > > > I've submitted v5 of the patch with a single Cc line with the prerequisite > kernel(s) and commit ID specified exactly. > > Kind regards, > > John >
On Fri, 26 Feb 2016, Darren Hart wrote: > On Mon, Feb 22, 2016 at 10:29:48PM -0600, John Dahlstrom wrote: [...] >> Thank you for that information. One commit is sufficient to apply the patch >> to all kernel versions without fuzz: >> >> 4fa9dab ideapad_laptop: Lenovo G50-30 fix rfkill reports wireless blocked > > This does not yield correct results for me on 3.17 (or 3.17.8). Do you get > different results from the following? > [...] > Note that the DMI match blocks were stuffed in the middle of the > ideapad_acpi_add() function instead of the no_hw_rfkill. The v3, v4, and v5 patch submissions had been corrupted by spurious whitespace inserted by my email client. I have sent a corrected version, [PATCH v6 1/1] ideapad-laptop: Add ideapad Y700 (15) to the no_hw_rfkill DMI list After submission, I verified the successful application of the patch as received by the mailing list. I also ran git clone, "git checkout -b 3.17 v3.17", and "git cherry-pick 4fa9dab", followed by patch with success. Kind regards, John -- To unsubscribe from this list: send the line "unsubscribe platform-driver-x86" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Sat, Feb 27, 2016 at 12:49:03AM -0600, John Dahlstrom wrote: > On Fri, 26 Feb 2016, Darren Hart wrote: > >On Mon, Feb 22, 2016 at 10:29:48PM -0600, John Dahlstrom wrote: > [...] > >>Thank you for that information. One commit is sufficient to apply the patch > >>to all kernel versions without fuzz: > >> > >>4fa9dab ideapad_laptop: Lenovo G50-30 fix rfkill reports wireless blocked > > > >This does not yield correct results for me on 3.17 (or 3.17.8). Do you get > >different results from the following? > > > [...] > >Note that the DMI match blocks were stuffed in the middle of the > >ideapad_acpi_add() function instead of the no_hw_rfkill. > > The v3, v4, and v5 patch submissions had been corrupted by spurious > whitespace inserted by my email client. I have sent a corrected version, > [PATCH v6 1/1] ideapad-laptop: Add ideapad Y700 (15) to the no_hw_rfkill DMI list > > After submission, I verified the successful application of the patch > as received by the mailing list. > > I also ran git clone, "git checkout -b 3.17 v3.17", and > "git cherry-pick 4fa9dab", followed by patch with success. OK, I've replaced the one in for-next with this one. I'm not keen on backporting only certain dmi match patches back to stable, but I don't want to discourage people from working to get the hardware they care about supported in stable either. Thanks for sticking with it and seeing this through to completion.
--- a/drivers/platform/x86/ideapad-laptop.c 2016-02-14 15:05:20.000000000 -0600 +++ b/drivers/platform/x86/ideapad-laptop.c 2016-02-16 03:54:48.484423725 -0600 @@ -864,4 +864,18 @@ static const struct dmi_system_id no_hw_rfkill_list[] = { DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo G50-30"), }, + }, + { + .ident = "Lenovo ideapad Y700-15ISK", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo ideapad Y700-15ISK"), + }, + }, + { + .ident = "Lenovo ideapad Y700 Touch-15ISK", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo ideapad Y700 Touch-15ISK"), + }, }, {
Some Lenovo ideapad models lack a physical rfkill switch. On Lenovo models ideapad Y700 Touch-15ISK and ideapad Y700-15ISK, ideapad-laptop would wrongly report all radios as blocked by hardware which caused wireless network connections to fail. Add these models without an rfkill switch to the no_hw_rfkill list. Signed-off-by: John Dahlstrom <jodarom@sdf.org> Cc: <stable@vger.kernel.org> # 4.5.x Cc: <stable@vger.kernel.org> # 4.4.x Cc: <stable@vger.kernel.org> # 4.3.x Cc: <stable@vger.kernel.org> # 4.2.x Cc: <stable@vger.kernel.org> # 4.1.x Cc: <stable@vger.kernel.org> # 4.0.x Cc: <stable@vger.kernel.org> # 3.19.x Cc: <stable@vger.kernel.org> # 3.18.x Cc: <stable@vger.kernel.org> # 3.17.x Cc: <stable@vger.kernel.org> # 3.16.x --- Test configuration Hardware: Lenovo ideapad Y700 Touch-15ISK Kernel version: 4.4.2 Patch changelog v2 split patch between Touch and non-Touch devices v3 undo patch split and limit summary to 72 characters v4 include comprehensive list of patchable kernels in Cc drivers/platform/x86/ideapad-laptop.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) -- To unsubscribe from this list: send the line "unsubscribe platform-driver-x86" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html