sysfs: cannot create duplicate filename .../system76_acpi::kbd_backlight/color
Subsystems (4)
Hi,
I notice a regression report on Bugzilla [1]. Quoting from it:
Quoted 42 lines
Loading the system76-acpi kernel module fails on linux 6.6-rc7. This does not seem to be an issue with system76-acpi itself, because reverting commit #5d36931f0fe51665c04f56c027613d22e6a03411, which is the only change made to this driver across the 6.6 development cycle, does not fix the issue. On 6.5.8, everything works fine. My hardware is a Clevo-based Alder Lake laptop running coreboot, roughly similar to the System76 Darter Pro 8.
backtrace:
[ 266.399036] sysfs: cannot create duplicate filename '/devices/LNXSYSTM:00/LNXSYBUS:00/17761776:00/leds/system76_acpi::kbd_backlight/color' [ 266.399045] CPU: 1 PID: 2896 Comm: modprobe Not tainted 6.6.0-rc7 #1 [ 266.399050] Hardware name: Notebook NS5x_NS7xPU/NS5x_NS7xPU, BIOS Dasharo (coreboot+UEFI) v1.6.0 03/30/2023 [ 266.399053] Call Trace: [ 266.399057] <TASK> [ 266.399063] dump_stack_lvl+0x36/0x50 [ 266.399080] sysfs_warn_dup+0x5a/0x70 [ 266.399088] sysfs_add_file_mode_ns+0x11a/0x130 [ 266.399094] internal_create_group+0x125/0x3b0 [ 266.399101] internal_create_groups+0x42/0xa0 [ 266.399107] device_add+0x5b1/0x8a0 [ 266.399113] ? kstrdup+0x4c/0x70 [ 266.399119] device_create_groups_vargs+0xce/0xf0 [ 266.399124] device_create_with_groups+0x4b/0x70 [ 266.399129] led_classdev_register_ext+0x1d2/0x470 [led_class] [ 266.399149] ? devm_led_classdev_register_ext+0x3a/0x90 [led_class] [ 266.399162] devm_led_classdev_register_ext+0x50/0x90 [led_class] [ 266.399173] system76_add+0x18b/0x460 [system76_acpi] [ 266.399186] acpi_device_probe+0x47/0x130 [ 266.399193] really_probe+0x19b/0x3e0 [ 266.399199] ? __pfx___driver_attach+0x10/0x10 [ 266.399205] __driver_probe_device+0x78/0x160 [ 266.399211] driver_probe_device+0x1f/0x90 [ 266.399217] __driver_attach+0xd2/0x1c0 [ 266.399222] bus_for_each_dev+0x85/0xd0 [ 266.399227] bus_add_driver+0x116/0x220 [ 266.399233] driver_register+0x59/0x100 [ 266.399242] ? __pfx_system76_driver_init+0x10/0x10 [system76_acpi] [ 266.399252] do_one_initcall+0x5a/0x300 [ 266.399260] do_init_module+0x60/0x240 [ 266.399267] init_module_from_file+0x86/0xc0 [ 266.399275] __x64_sys_finit_module+0x18a/0x350 [ 266.399282] do_syscall_64+0x5d/0x90 [ 266.399289] ? syscall_exit_to_user_mode+0x26/0x40 [ 266.399295] ? do_syscall_64+0x6c/0x90 [ 266.399300] ? do_syscall_64+0x6c/0x90 [ 266.399305] entry_SYSCALL_64_after_hwframe+0x6e/0xd8 [ 266.399314] RIP: 0033:0x7f5c11b38d7d [ 266.399360] Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 7b d0 0b 00 f7 d8 64 89 01 48
Quoted 8 lines
[ 266.399364] RSP: 002b:00007ffe30e15b88 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 [ 266.399370] RAX: ffffffffffffffda RBX: 000055a8d48d6c10 RCX: 00007f5c11b38d7d [ 266.399372] RDX: 0000000000000000 RSI: 000055a8d3077d8b RDI: 0000000000000003 [ 266.399375] RBP: 000055a8d3077d8b R08: 00007f5c11bf6b00 R09: 00007ffe30e15bd0 [ 266.399376] R10: 0000000000000050 R11: 0000000000000246 R12: 0000000000040000 [ 266.399378] R13: 000055a8d48d6c90 R14: 000055a8d48d6390 R15: 000055a8d48d7090 [ 266.399382] </TASK> [ 266.399410] System76 ACPI Driver: probe of 17761776:00 failed with error -17
See Bugzilla for the full thread and attached dmesg output.
Anyway, I'm adding this regression to regzbot:
#regzbot introduced: v6.5..v6.6-rc7 https://bugzilla.kernel.org/show_bug.cgi?id=218045
Thanks.
[1]: https://bugzilla.kernel.org/show_bug.cgi?id=218045
Signature · 1 line
An old man doll... just what I always wanted! - Clara
Quoted 55 lines · Bagas Sanjaya
Hi,
I notice a regression report on Bugzilla [1]. Quoting from it:
Quoted 42 lines
Loading the system76-acpi kernel module fails on linux 6.6-rc7. This does not seem to be an issue with system76-acpi itself, because reverting commit #5d36931f0fe51665c04f56c027613d22e6a03411, which is the only change made to this driver across the 6.6 development cycle, does not fix the issue. On 6.5.8, everything works fine. My hardware is a Clevo-based Alder Lake laptop running coreboot, roughly similar to the System76 Darter Pro 8.
backtrace:
[ 266.399036] sysfs: cannot create duplicate filename '/devices/LNXSYSTM:00/LNXSYBUS:00/17761776:00/leds/system76_acpi::kbd_backlight/color' [ 266.399045] CPU: 1 PID: 2896 Comm: modprobe Not tainted 6.6.0-rc7 #1 [ 266.399050] Hardware name: Notebook NS5x_NS7xPU/NS5x_NS7xPU, BIOS Dasharo (coreboot+UEFI) v1.6.0 03/30/2023 [ 266.399053] Call Trace: [ 266.399057] <TASK> [ 266.399063] dump_stack_lvl+0x36/0x50 [ 266.399080] sysfs_warn_dup+0x5a/0x70 [ 266.399088] sysfs_add_file_mode_ns+0x11a/0x130 [ 266.399094] internal_create_group+0x125/0x3b0 [ 266.399101] internal_create_groups+0x42/0xa0 [ 266.399107] device_add+0x5b1/0x8a0 [ 266.399113] ? kstrdup+0x4c/0x70 [ 266.399119] device_create_groups_vargs+0xce/0xf0 [ 266.399124] device_create_with_groups+0x4b/0x70 [ 266.399129] led_classdev_register_ext+0x1d2/0x470 [led_class] [ 266.399149] ? devm_led_classdev_register_ext+0x3a/0x90 [led_class] [ 266.399162] devm_led_classdev_register_ext+0x50/0x90 [led_class] [ 266.399173] system76_add+0x18b/0x460 [system76_acpi] [ 266.399186] acpi_device_probe+0x47/0x130 [ 266.399193] really_probe+0x19b/0x3e0 [ 266.399199] ? __pfx___driver_attach+0x10/0x10 [ 266.399205] __driver_probe_device+0x78/0x160 [ 266.399211] driver_probe_device+0x1f/0x90 [ 266.399217] __driver_attach+0xd2/0x1c0 [ 266.399222] bus_for_each_dev+0x85/0xd0 [ 266.399227] bus_add_driver+0x116/0x220 [ 266.399233] driver_register+0x59/0x100 [ 266.399242] ? __pfx_system76_driver_init+0x10/0x10 [system76_acpi] [ 266.399252] do_one_initcall+0x5a/0x300 [ 266.399260] do_init_module+0x60/0x240 [ 266.399267] init_module_from_file+0x86/0xc0 [ 266.399275] __x64_sys_finit_module+0x18a/0x350 [ 266.399282] do_syscall_64+0x5d/0x90 [ 266.399289] ? syscall_exit_to_user_mode+0x26/0x40 [ 266.399295] ? do_syscall_64+0x6c/0x90 [ 266.399300] ? do_syscall_64+0x6c/0x90 [ 266.399305] entry_SYSCALL_64_after_hwframe+0x6e/0xd8 [ 266.399314] RIP: 0033:0x7f5c11b38d7d [ 266.399360] Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 7b d0 0b 00 f7 d8 64 89 01 48Quoted 8 lines
[ 266.399364] RSP: 002b:00007ffe30e15b88 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 [ 266.399370] RAX: ffffffffffffffda RBX: 000055a8d48d6c10 RCX: 00007f5c11b38d7d [ 266.399372] RDX: 0000000000000000 RSI: 000055a8d3077d8b RDI: 0000000000000003 [ 266.399375] RBP: 000055a8d3077d8b R08: 00007f5c11bf6b00 R09: 00007ffe30e15bd0 [ 266.399376] R10: 0000000000000050 R11: 0000000000000246 R12: 0000000000040000 [ 266.399378] R13: 000055a8d48d6c90 R14: 000055a8d48d6390 R15: 000055a8d48d7090 [ 266.399382] </TASK> [ 266.399410] System76 ACPI Driver: probe of 17761776:00 failed with error -17
Quoted 6 lines
See Bugzilla for the full thread and attached dmesg output.
Anyway, I'm adding this regression to regzbot:
#regzbot introduced: v6.5..v6.6-rc7 https://bugzilla.kernel.org/show_bug.cgi?id=218045
The reporter had narrowed down the culprit. He said on Bugzilla:
Quoted 1 line
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.
Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
Anyway, telling regzbot:
#regzbot introduced: c7d80059b086c4
Thanks.
Signature · 1 line
An old man doll... just what I always wanted! - Clara
Quoted 63 lines · Bagas Sanjaya
Quoted 55 lines · Bagas Sanjaya
Hi,
I notice a regression report on Bugzilla [1]. Quoting from it:
Quoted 42 lines
Loading the system76-acpi kernel module fails on linux 6.6-rc7. This does not seem to be an issue with system76-acpi itself, because reverting commit #5d36931f0fe51665c04f56c027613d22e6a03411, which is the only change made to this driver across the 6.6 development cycle, does not fix the issue. On 6.5.8, everything works fine. My hardware is a Clevo-based Alder Lake laptop running coreboot, roughly similar to the System76 Darter Pro 8.
backtrace:
[ 266.399036] sysfs: cannot create duplicate filename '/devices/LNXSYSTM:00/LNXSYBUS:00/17761776:00/leds/system76_acpi::kbd_backlight/color' [ 266.399045] CPU: 1 PID: 2896 Comm: modprobe Not tainted 6.6.0-rc7 #1 [ 266.399050] Hardware name: Notebook NS5x_NS7xPU/NS5x_NS7xPU, BIOS Dasharo (coreboot+UEFI) v1.6.0 03/30/2023 [ 266.399053] Call Trace: [ 266.399057] <TASK> [ 266.399063] dump_stack_lvl+0x36/0x50 [ 266.399080] sysfs_warn_dup+0x5a/0x70 [ 266.399088] sysfs_add_file_mode_ns+0x11a/0x130 [ 266.399094] internal_create_group+0x125/0x3b0 [ 266.399101] internal_create_groups+0x42/0xa0 [ 266.399107] device_add+0x5b1/0x8a0 [ 266.399113] ? kstrdup+0x4c/0x70 [ 266.399119] device_create_groups_vargs+0xce/0xf0 [ 266.399124] device_create_with_groups+0x4b/0x70 [ 266.399129] led_classdev_register_ext+0x1d2/0x470 [led_class] [ 266.399149] ? devm_led_classdev_register_ext+0x3a/0x90 [led_class] [ 266.399162] devm_led_classdev_register_ext+0x50/0x90 [led_class] [ 266.399173] system76_add+0x18b/0x460 [system76_acpi] [ 266.399186] acpi_device_probe+0x47/0x130 [ 266.399193] really_probe+0x19b/0x3e0 [ 266.399199] ? __pfx___driver_attach+0x10/0x10 [ 266.399205] __driver_probe_device+0x78/0x160 [ 266.399211] driver_probe_device+0x1f/0x90 [ 266.399217] __driver_attach+0xd2/0x1c0 [ 266.399222] bus_for_each_dev+0x85/0xd0 [ 266.399227] bus_add_driver+0x116/0x220 [ 266.399233] driver_register+0x59/0x100 [ 266.399242] ? __pfx_system76_driver_init+0x10/0x10 [system76_acpi] [ 266.399252] do_one_initcall+0x5a/0x300 [ 266.399260] do_init_module+0x60/0x240 [ 266.399267] init_module_from_file+0x86/0xc0 [ 266.399275] __x64_sys_finit_module+0x18a/0x350 [ 266.399282] do_syscall_64+0x5d/0x90 [ 266.399289] ? syscall_exit_to_user_mode+0x26/0x40 [ 266.399295] ? do_syscall_64+0x6c/0x90 [ 266.399300] ? do_syscall_64+0x6c/0x90 [ 266.399305] entry_SYSCALL_64_after_hwframe+0x6e/0xd8 [ 266.399314] RIP: 0033:0x7f5c11b38d7d [ 266.399360] Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 7b d0 0b 00 f7 d8 64 89 01 48Quoted 8 lines
[ 266.399364] RSP: 002b:00007ffe30e15b88 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 [ 266.399370] RAX: ffffffffffffffda RBX: 000055a8d48d6c10 RCX: 00007f5c11b38d7d [ 266.399372] RDX: 0000000000000000 RSI: 000055a8d3077d8b RDI: 0000000000000003 [ 266.399375] RBP: 000055a8d3077d8b R08: 00007f5c11bf6b00 R09: 00007ffe30e15bd0 [ 266.399376] R10: 0000000000000050 R11: 0000000000000246 R12: 0000000000040000 [ 266.399378] R13: 000055a8d48d6c90 R14: 000055a8d48d6390 R15: 000055a8d48d7090 [ 266.399382] </TASK> [ 266.399410] System76 ACPI Driver: probe of 17761776:00 failed with error -17Quoted 6 lines
See Bugzilla for the full thread and attached dmesg output.
Anyway, I'm adding this regression to regzbot:
#regzbot introduced: v6.5..v6.6-rc7 https://bugzilla.kernel.org/show_bug.cgi?id=218045
Quoted 6 lines
The reporter had narrowed down the culprit. He said on Bugzilla:
Quoted 1 line
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
Hi,
The offending commit stores the color in struct led_classdev and exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.
JJ
Quoted 7 lines
Anyway, telling regzbot:
#regzbot introduced:
c7d80059b086c4Thanks.
Quoted 8 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
Quoted 8 lines
Hi,
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.
OK, see you in the patch!
Signature · 1 line
An old man doll... just what I always wanted! - Clara
Quoted 12 lines · Bagas Sanjaya
Quoted 10 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
Hi,
Quoted 8 lines
Quoted 6 lines
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.OK, see you in the patch!
Is there a patch available?
This bug hits for a few Logitech keyboard models, too, and it makes 6.6 kernel unsable for them, as hid-lg-g15 driver probe fails due to this bug:
https://bugzilla.kernel.org/show_bug.cgi?id=218155
We need a quick fix for 6.6.x.
Thanks!
Takashi
Takashi, Jean-Jacques Hiblot, Lee,
Quoted 16 lines · Takashi Iwai
Quoted 14 lines · Bagas Sanjaya
Quoted 12 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.
Quoted 3 lines
Quoted 1 line
OK, see you in the patch!
Is there a patch available?
Not that I know of. Could not find anything on lore either.
Quoted 6 lines
This bug hits for a few Logitech keyboard models, too, and it makes 6.6 kernel unsable for them, as hid-lg-g15 driver probe fails due to this bug:
https://bugzilla.kernel.org/show_bug.cgi?id=218155We need a quick fix for 6.6.x.
Given that Jean-Jacques Hiblot (the author of the culprit) and Lee (who committed it and sent it to Linus) know about this for a while already without doing anything about it, I wonder if someone should just send a revert to Linus (unless of course that is likely to introduce a regression on its own).
Takashi, could you maybe do this, unless a fix shows up real soon?
Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
P.S.: /me should have followed up on this earlier... :-/ --
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
If I did something stupid, please tell me, as explained on that page.
Quoted 33 lines · Thorsten Leemhuis
Takashi, Jean-Jacques Hiblot, Lee,
Quoted 16 lines · Takashi Iwai
Quoted 14 lines · Bagas Sanjaya
Quoted 12 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.Quoted 3 lines
Quoted 1 line
OK, see you in the patch!
Is there a patch available?
Not that I know of. Could not find anything on lore either.
Quoted 6 lines
This bug hits for a few Logitech keyboard models, too, and it makes 6.6 kernel unsable for them, as hid-lg-g15 driver probe fails due to this bug:
https://bugzilla.kernel.org/show_bug.cgi?id=218155We need a quick fix for 6.6.x.
Quoted 7 lines
Given that Jean-Jacques Hiblot (the author of the culprit) and Lee (who committed it and sent it to Linus) know about this for a while already without doing anything about it, I wonder if someone should just send a revert to Linus (unless of course that is likely to introduce a regression on its own).
Takashi, could you maybe do this, unless a fix shows up real soon?
I can, but we need to decide which way to go.
There are several options:
1. Revert the commit c7d80059b086;
this drops led class color sysfs entries. Also the store of led_cdev->color from fwnode is dropped, too.
2. Drop only led class color sysfs entries;
a partial revert of c7d80059b086 above
3. Rename conflicting sysfs entries in drivers; e.g. color -> kb_color for hid-lg-g15 and system76_acpi
In either way, we'd break user-space (sysfs).
IMO, 2 would be the least harm, as the class sysfs entry was introduced since 6.6. I guess this is what Jean-Jacques suggested. But I'm not sure how important this new class sysfs entry is; it has to be clarified from leds people who actually use / wanted the feature.
3 was already confirmed to work on both bug reports. OTOH, it's not clear whether we can neglect (potentially) already existing users.
Takashi
Quoted 9 lines
Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
P.S.: /me should have followed up on this earlier... :-/ --
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
If I did something stupid, please tell me, as explained on that page.
Quoted 51 lines · Takashi Iwai
Quoted 33 lines · Thorsten Leemhuis
Takashi, Jean-Jacques Hiblot, Lee,
Quoted 16 lines · Takashi Iwai
Quoted 14 lines · Bagas Sanjaya
Quoted 12 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.Quoted 3 lines
Quoted 1 line
OK, see you in the patch!
Is there a patch available?
Not that I know of. Could not find anything on lore either.
Quoted 6 lines
This bug hits for a few Logitech keyboard models, too, and it makes 6.6 kernel unsable for them, as hid-lg-g15 driver probe fails due to this bug:
https://bugzilla.kernel.org/show_bug.cgi?id=218155We need a quick fix for 6.6.x.
Quoted 7 lines
Given that Jean-Jacques Hiblot (the author of the culprit) and Lee (who committed it and sent it to Linus) know about this for a while already without doing anything about it, I wonder if someone should just send a revert to Linus (unless of course that is likely to introduce a regression on its own).
Takashi, could you maybe do this, unless a fix shows up real soon?
I can, but we need to decide which way to go.
There are several options:1. Revert the commit
c7d80059b086;
this drops led class color sysfs entries. Also the store ofled_cdev->colorfrom fwnode is dropped, too.
Quoted 5 lines
2. Drop only led class color sysfs entries;
a partial revert ofc7d80059b086above3. Rename conflicting sysfs entries in drivers; e.g. color ->
kb_colorfor hid-lg-g15 andsystem76_acpi
... and
4. Rename conflicting sysfs class color entry
Quoted 1 line
In either way, we'd break user-space (sysfs).
It still holds :)
Takashi
Hi,
Quoted 46 lines · Takashi Iwai
Quoted 33 lines · Thorsten Leemhuis
Takashi, Jean-Jacques Hiblot, Lee,
Quoted 16 lines · Takashi Iwai
Quoted 14 lines · Bagas Sanjaya
Quoted 12 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.Quoted 3 lines
Quoted 1 line
OK, see you in the patch!
Is there a patch available?
Not that I know of. Could not find anything on lore either.
Quoted 6 lines
This bug hits for a few Logitech keyboard models, too, and it makes 6.6 kernel unsable for them, as hid-lg-g15 driver probe fails due to this bug:
https://bugzilla.kernel.org/show_bug.cgi?id=218155We need a quick fix for 6.6.x.
Quoted 7 lines
Given that Jean-Jacques Hiblot (the author of the culprit) and Lee (who committed it and sent it to Linus) know about this for a while already without doing anything about it, I wonder if someone should just send a revert to Linus (unless of course that is likely to introduce a regression on its own).
Takashi, could you maybe do this, unless a fix shows up real soon?
I can, but we need to decide which way to go.
There are several options:
Quoted 6 lines
1. Revert the commit
c7d80059b086;
this drops led class color sysfs entries. Also the store ofled_cdev->colorfrom fwnode is dropped, too.2. Drop only led class color sysfs entries;
a partial revert ofc7d80059b086above
AFAIK further up in the thread (or a related thread) there already was consensus to do this. Someone just needs to write the patch.
Quoted 4 lines
3. Rename conflicting sysfs entries in drivers; e.g. color ->
kb_colorfor hid-lg-g15 andsystem76_acpiIn either way, we'd break user-space (sysfs).
The new color attribute causing the conflict has only been in 6.6 so there likely aren't any users of it yet and since dropping it would be backported to 6.6.y there shouldn't be any future users of it either, since any 6.6 users presumably will use 6.6.y and not 6.6.0
Quoted 2 lines
IMO, 2 would be the least harm, as the class sysfs entry was introduced since 6.6.
Ack.
Quoted 1 line
I guess this is what Jean-Jacques suggested.
Right.
Quoted 3 lines
But I'm not sure how important this new class sysfs entry is; it has to be clarified from leds people who actually use / wanted the feature.
If I have read the thread correctly it is not important the value it represents is used internally in the LED subsystem and userspace does not really need it.
Regards,
Hans
Quoted 54 lines · Hans de Goede
Hi,
Quoted 50 lines · Takashi Iwai
Quoted 33 lines · Thorsten Leemhuis
Takashi, Jean-Jacques Hiblot, Lee,
Quoted 16 lines · Takashi Iwai
Quoted 14 lines · Bagas Sanjaya
Quoted 12 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.Quoted 3 lines
Quoted 1 line
OK, see you in the patch!
Is there a patch available?
Not that I know of. Could not find anything on lore either.
Quoted 6 lines
This bug hits for a few Logitech keyboard models, too, and it makes 6.6 kernel unsable for them, as hid-lg-g15 driver probe fails due to this bug:
https://bugzilla.kernel.org/show_bug.cgi?id=218155We need a quick fix for 6.6.x.
Quoted 7 lines
Given that Jean-Jacques Hiblot (the author of the culprit) and Lee (who committed it and sent it to Linus) know about this for a while already without doing anything about it, I wonder if someone should just send a revert to Linus (unless of course that is likely to introduce a regression on its own).
Takashi, could you maybe do this, unless a fix shows up real soon?
I can, but we need to decide which way to go.
There are several options:1. Revert the commit
c7d80059b086;
this drops led class color sysfs entries. Also the store ofled_cdev->colorfrom fwnode is dropped, too.
Quoted 6 lines
Quoted 2 lines
2. Drop only led class color sysfs entries;
a partial revert ofc7d80059b086aboveAFAIK further up in the thread (or a related thread) there already was consensus to do this. Someone just needs to write the patch.
Well, is there any user of this new led_classdev.color field? The value read from fwnode is stored there, but as far as I see, there seems no real user, so far. If it's still unused, we can do the whole revert -- which is cleaner.
Takashi
Quoted 34 lines
Quoted 4 lines
3. Rename conflicting sysfs entries in drivers; e.g. color ->
kb_colorfor hid-lg-g15 andsystem76_acpiIn either way, we'd break user-space (sysfs).
The new color attribute causing the conflict has only been in 6.6 so there likely aren't any users of it yet and since dropping it would be backported to 6.6.y there shouldn't be any future users of it either, since any 6.6 users presumably will use 6.6.y and not 6.6.0
Quoted 2 lines
IMO, 2 would be the least harm, as the class sysfs entry was introduced since 6.6.
Ack.
Quoted 1 line
I guess this is what Jean-Jacques suggested.
Right.
Quoted 3 lines
But I'm not sure how important this new class sysfs entry is; it has to be clarified from leds people who actually use / wanted the feature.
If I have read the thread correctly it is not important the value it represents is used internally in the LED subsystem and userspace does not really need it.
Regards,
Hans
Quoted 59 lines · Takashi Iwai
Quoted 57 lines · Hans de Goede
Hi,
Quoted 52 lines · Takashi Iwai
Quoted 33 lines · Thorsten Leemhuis
Takashi, Jean-Jacques Hiblot, Lee,
Quoted 16 lines · Takashi Iwai
Quoted 14 lines · Bagas Sanjaya
Quoted 12 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.Quoted 3 lines
Quoted 1 line
OK, see you in the patch!
Is there a patch available?
Not that I know of. Could not find anything on lore either.
Quoted 6 lines
This bug hits for a few Logitech keyboard models, too, and it makes 6.6 kernel unsable for them, as hid-lg-g15 driver probe fails due to this bug:
https://bugzilla.kernel.org/show_bug.cgi?id=218155We need a quick fix for 6.6.x.
Quoted 7 lines
Given that Jean-Jacques Hiblot (the author of the culprit) and Lee (who committed it and sent it to Linus) know about this for a while already without doing anything about it, I wonder if someone should just send a revert to Linus (unless of course that is likely to introduce a regression on its own).
Takashi, could you maybe do this, unless a fix shows up real soon?
I can, but we need to decide which way to go.
There are several options:1. Revert the commit
c7d80059b086;
this drops led class color sysfs entries. Also the store ofled_cdev->colorfrom fwnode is dropped, too.2. Drop only led class color sysfs entries;
a partial revert ofc7d80059b086above
Quoted 8 lines
Quoted 3 lines
AFAIK further up in the thread (or a related thread) there already was consensus to do this. Someone just needs to write the patch.
Well, is there any user of this new led_classdev.color field? The value read from fwnode is stored there, but as far as I see, there seems no real user, so far. If it's still unused, we can do the whole revert -- which is cleaner.
ACK for the revert. The original commit should have been designed to not have sysfs conflict like this, though.
Thanks.
Signature · 1 line
An old man doll... just what I always wanted! - Clara
Hi,
Quoted 59 lines · Takashi Iwai
Quoted 57 lines · Hans de Goede
Hi,
Quoted 52 lines · Takashi Iwai
Quoted 33 lines · Thorsten Leemhuis
Takashi, Jean-Jacques Hiblot, Lee,
Quoted 16 lines · Takashi Iwai
Quoted 14 lines · Bagas Sanjaya
Quoted 12 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.Quoted 3 lines
Quoted 1 line
OK, see you in the patch!
Is there a patch available?
Not that I know of. Could not find anything on lore either.
Quoted 6 lines
This bug hits for a few Logitech keyboard models, too, and it makes 6.6 kernel unsable for them, as hid-lg-g15 driver probe fails due to this bug:
https://bugzilla.kernel.org/show_bug.cgi?id=218155We need a quick fix for 6.6.x.
Quoted 7 lines
Given that Jean-Jacques Hiblot (the author of the culprit) and Lee (who committed it and sent it to Linus) know about this for a while already without doing anything about it, I wonder if someone should just send a revert to Linus (unless of course that is likely to introduce a regression on its own).
Takashi, could you maybe do this, unless a fix shows up real soon?
I can, but we need to decide which way to go.
There are several options:1. Revert the commit
c7d80059b086;
this drops led class color sysfs entries. Also the store ofled_cdev->colorfrom fwnode is dropped, too.2. Drop only led class color sysfs entries;
a partial revert ofc7d80059b086above
Quoted 8 lines
Quoted 3 lines
AFAIK further up in the thread (or a related thread) there already was consensus to do this. Someone just needs to write the patch.
Well, is there any user of this new led_classdev.color field? The value read from fwnode is stored there, but as far as I see, there seems no real user, so far. If it's still unused, we can do the whole revert -- which is cleaner.
I honestly don't know. I've mostly just been reading along. I think there may be some future in kernel use planned (not sure at all though).
If there are no current in kernel users then I agree we should just go with a full revert now to fix the regression. If some later in kernel users do come along then they can always re-introduce the change minus the sysfs attr addition.
Regards,
Hans
Quoted 72 lines · Hans de Goede
Hi,
Quoted 59 lines · Takashi Iwai
Quoted 57 lines · Hans de Goede
Hi,
Quoted 52 lines · Takashi Iwai
Quoted 33 lines · Thorsten Leemhuis
Takashi, Jean-Jacques Hiblot, Lee,
Quoted 16 lines · Takashi Iwai
Quoted 14 lines · Bagas Sanjaya
Quoted 12 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.Quoted 3 lines
Quoted 1 line
OK, see you in the patch!
Is there a patch available?
Not that I know of. Could not find anything on lore either.
Quoted 6 lines
This bug hits for a few Logitech keyboard models, too, and it makes 6.6 kernel unsable for them, as hid-lg-g15 driver probe fails due to this bug:
https://bugzilla.kernel.org/show_bug.cgi?id=218155We need a quick fix for 6.6.x.
Quoted 7 lines
Given that Jean-Jacques Hiblot (the author of the culprit) and Lee (who committed it and sent it to Linus) know about this for a while already without doing anything about it, I wonder if someone should just send a revert to Linus (unless of course that is likely to introduce a regression on its own).
Takashi, could you maybe do this, unless a fix shows up real soon?
I can, but we need to decide which way to go.
There are several options:1. Revert the commit
c7d80059b086;
this drops led class color sysfs entries. Also the store ofled_cdev->colorfrom fwnode is dropped, too.2. Drop only led class color sysfs entries;
a partial revert ofc7d80059b086aboveQuoted 8 lines
Quoted 3 lines
AFAIK further up in the thread (or a related thread) there already was consensus to do this. Someone just needs to write the patch.
Well, is there any user of this new led_classdev.color field? The value read from fwnode is stored there, but as far as I see, there seems no real user, so far. If it's still unused, we can do the whole revert -- which is cleaner.
Quoted 7 lines
I honestly don't know. I've mostly just been reading along. I think there may be some future in kernel use planned (not sure at all though).
If there are no current in kernel users then I agree we should just go with a full revert now to fix the regression. If some later in kernel users do come along then they can always re-introduce the change minus the sysfs attr addition.
Sounds reasonable.
OK, let me pitch the revert patch.
thanks,
Takashi
Lee Jones
maintainerQuoted 30 lines · Thorsten Leemhuis
Takashi, Jean-Jacques Hiblot, Lee,
Quoted 16 lines · Takashi Iwai
Quoted 14 lines · Bagas Sanjaya
Quoted 12 lines · Jean-Jacques Hiblot
Quoted 5 lines · Bagas Sanjaya
Quoted 1 line · Bagas Sanjaya
The culprit seems to be commit
c7d80059b086c4986cd994a1973ec7a5d75f8eea, which introduces a new 'color' attribute for led sysfs class devices. The problem is that the system76-acpi platform driver tries to create the exact same sysfs attribute itself for thesystem76_acpi::kbd_backlightdevice, leading to the conflict. For testing purposes, I've just rebuilt the kernel with the system76-apci color attribute renamed tokb_color, and that fixes the issue.Jean-Jacques Hiblot, would you like to take a look on this regression, since you authored the culprit?
The offending commit stores the color in struct
led_classdevand exposes it via sysfs. It was part of a series that create a RGB leds from multiple single-color LEDs. for this series, we need the color information but we don't really need to expose it it via sysfs. In order to fix the issue, we can remove the 'color' attribute from the sysfs.Quoted 3 lines
Quoted 1 line
OK, see you in the patch!
Is there a patch available?
Not that I know of. Could not find anything on lore either.
Quoted 5 lines
This bug hits for a few Logitech keyboard models, too, and it makes 6.6 kernel unsable for them, as hid-lg-g15 driver probe fails due to this bug:
https://bugzilla.kernel.org/show_bug.cgi?id=218155
Quoted 7 lines
Quoted 1 line
We need a quick fix for 6.6.x.
Given that Jean-Jacques Hiblot (the author of the culprit) and Lee (who committed it and sent it to Linus) know about this for a while already without doing anything about it, I wonder if someone should just send a revert to Linus (unless of course that is likely to introduce a regression on its own).
You seem to have gone from DEFCON 4 to DEFCON 2 there. The middle step is to submit a patch to fix the issue and follow the usual patch flow.
Signature · 1 line
Lee Jones [李琼斯]