aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/tools/perf/scripts/python/exported-sql-viewer.py (unfollow)
AgeCommit message (Collapse)AuthorFilesLines
2022-12-08Revert "HID: logitech-hidpp: Remove special-casing of Bluetooth devices"Benjamin Tissoires1-0/+9
This reverts commit 8544c812e43ab7bdf40458411b83987b8cba924d. We need to revert commit 532223c8ac57 ("HID: logitech-hidpp: Enable HID++ for all the Logitech Bluetooth devices") because that commit might make hid-logitech-hidpp bind on mice that are not well enough supported by hid-logitech-hidpp, and the end result is that the probe of those mice is now returning -ENODEV, leaving the end user with a dead mouse. Given that commit 8544c812e43a ("HID: logitech-hidpp: Remove special-casing of Bluetooth devices") is a direct dependency of 532223c8ac57, revert it too. Note that this also adapt according to commit 908d325e1665 ("HID: logitech-hidpp: Detect hi-res scrolling support") to re-add support of the devices that were removed from that commit too. I have locally an MX Master and I tested this device with that revert, ensuring we still have high-res scrolling. Reported-by: Rafael J . Wysocki <rjw@rjwysocki.net> Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-22HID: usbhid: Add ALWAYS_POLL quirk for some miceAnkit Patel2-0/+6
Some additional USB mouse devices are needing ALWAYS_POLL quirk without which they disconnect and reconnect every 60s. Add below devices to the known quirk list. CHERRY VID 0x046a, PID 0x000c MICROSOFT VID 0x045e, PID 0x0783 PRIMAX VID 0x0461, PID 0x4e2a Signed-off-by: Ankit Patel <anpatel@nvidia.com> Signed-off-by: Haotien Hsu <haotienh@nvidia.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-21HID: logitech HID++: Send SwID in GetProtocolVersionAndreas Bergmeier1-1/+1
According to docs a SwID should be sent for GetProtocolVersion. > 0x10.DeviceIndex.0x00.0x1n where n is SwID. Signed-off-by: Andreas Bergmeier <abergmeier@gmx.net> Reviewed-by: Bastien Nocera <hadess@hadess.net> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-21HID: hid-elan: use default remove for hid deviceMarcus Folkesson1-6/+0
hid_device_remove() will call hid_hw_stop() as default .remove function if no function is specified. Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-21HID: hid-alps: use default remove for hid deviceMarcus Folkesson1-6/+0
hid_device_remove() will call hid_hw_stop() as default .remove function if no function is specified. Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-21HID: core: fix shift-out-of-bounds in hid_report_raw_eventZhangPeng1-0/+3
Syzbot reported shift-out-of-bounds in hid_report_raw_event. microsoft 0003:045E:07DA.0001: hid_field_extract() called with n (128) > 32! (swapper/0) ====================================================================== UBSAN: shift-out-of-bounds in drivers/hid/hid-core.c:1323:20 shift exponent 127 is too large for 32-bit type 'int' CPU: 0 PID: 0 Comm: swapper/0 Not tainted 6.1.0-rc4-syzkaller-00159-g4bbf3422df78 #0 Hardware name: Google Compute Engine/Google Compute Engine, BIOS Google 10/26/2022 Call Trace: <IRQ> __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x1e3/0x2cb lib/dump_stack.c:106 ubsan_epilogue lib/ubsan.c:151 [inline] __ubsan_handle_shift_out_of_bounds+0x3a6/0x420 lib/ubsan.c:322 snto32 drivers/hid/hid-core.c:1323 [inline] hid_input_fetch_field drivers/hid/hid-core.c:1572 [inline] hid_process_report drivers/hid/hid-core.c:1665 [inline] hid_report_raw_event+0xd56/0x18b0 drivers/hid/hid-core.c:1998 hid_input_report+0x408/0x4f0 drivers/hid/hid-core.c:2066 hid_irq_in+0x459/0x690 drivers/hid/usbhid/hid-core.c:284 __usb_hcd_giveback_urb+0x369/0x530 drivers/usb/core/hcd.c:1671 dummy_timer+0x86b/0x3110 drivers/usb/gadget/udc/dummy_hcd.c:1988 call_timer_fn+0xf5/0x210 kernel/time/timer.c:1474 expire_timers kernel/time/timer.c:1519 [inline] __run_timers+0x76a/0x980 kernel/time/timer.c:1790 run_timer_softirq+0x63/0xf0 kernel/time/timer.c:1803 __do_softirq+0x277/0x75b kernel/softirq.c:571 __irq_exit_rcu+0xec/0x170 kernel/softirq.c:650 irq_exit_rcu+0x5/0x20 kernel/softirq.c:662 sysvec_apic_timer_interrupt+0x91/0xb0 arch/x86/kernel/apic/apic.c:1107 ====================================================================== If the size of the integer (unsigned n) is bigger than 32 in snto32(), shift exponent will be too large for 32-bit type 'int', resulting in a shift-out-of-bounds bug. Fix this by adding a check on the size of the integer (unsigned n) in snto32(). To add support for n greater than 32 bits, set n to 32, if n is greater than 32. Reported-by: syzbot+8b1641d2f14732407e23@syzkaller.appspotmail.com Fixes: dde5845a529f ("[PATCH] Generic HID layer - code split") Signed-off-by: ZhangPeng <zhangpeng362@huawei.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-21HID: uclogic: Add HID_QUIRK_HIDINPUT_FORCE quirkJosé Expósito1-0/+1
Commit f7d8e387d9ae ("HID: uclogic: Switch to Digitizer usage for styluses") changed the usage used in UCLogic from "Pen" to "Digitizer". However, the IS_INPUT_APPLICATION() macro evaluates to false for HID_DG_DIGITIZER causing issues with the XP-Pen Star G640 tablet. Add the HID_QUIRK_HIDINPUT_FORCE quirk to bypass the IS_INPUT_APPLICATION() check. Reported-by: Torge Matthies <openglfreak@googlemail.com> Reported-by: Alexander Zhang <alex@alexyzhang.dev> Tested-by: Alexander Zhang <alex@alexyzhang.dev> Signed-off-by: José Expósito <jose.exposito89@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-15HID: fix I2C_HID not selected when I2C_HID_OF_ELAN isBenjamin Tissoires1-2/+2
When I2C_HID_OF_ELAN is set, we need to turn on I2C_HID_CORE to ensure we get all the HID requirements. Fixes: bd3cba00dcc6 ("HID: i2c-hid: elan: Add support for Elan eKTH6915 i2c-hid touchscreens") Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-14HID: hid-lg4ff: Add check for empty lbufAnastasia Belova1-0/+6
If an empty buf is received, lbuf is also empty. So lbuf is accessed by index -1. Found by Linux Verification Center (linuxtesting.org) with SVACE. Fixes: f31a2de3fe36 ("HID: hid-lg4ff: Allow switching of Logitech gaming wheels between compatibility modes") Signed-off-by: Anastasia Belova <abelova@astralinux.ru> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-14HID: ite: Enable QUIRK_TOUCHPAD_ON_OFF_REPORT on Acer Aspire Switch V 10Hans de Goede2-0/+6
The Acer Aspire Switch V 10 (SW5-017)'s keyboard-dock uses the same ITE controller setup as other Acer Switch 2-in-1's. This needs special handling for the wifi on/off toggle hotkey as well as to properly report touchpad on/off keypresses. Add the USB-ids for the SW5-017's keyboard-dock with a quirk setting of QUIRK_TOUCHPAD_ON_OFF_REPORT to fix both issues. Cc: Rudolf Polzer <rpolzer@google.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-14HID: apple: Swap Control and Command keys on Apple keyboardsAditya Garg1-1/+22
This patch allows users to swap the control and command keys. This can be useful for the Mac users who are used to using Command instead of Control in macOS for various commonly used shortcuts. Signed-off-by: Aditya Garg <gargaditya08@live.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-14HID: intel-ish-hid: ishtp: remove variable rb_countColin Ian King1-3/+0
The variable rb_count is being incremented but it is never referenced, it is redundant and can be removed. Signed-off-by: Colin Ian King <colin.i.king@gmail.com> Acked-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-14HID: uclogic: Fix frame templates for big endian architecturesJosé Expósito1-1/+1
When parsing a frame template with a placeholder indicating the number of buttons present on the frame its value was incorrectly set on big endian architectures due to double little endian conversion. In order to reproduce the issue and verify the fix, run the HID KUnit tests on the PowerPC architecture: $ ./tools/testing/kunit/kunit.py run --kunitconfig=drivers/hid \ --arch=powerpc --cross_compile=powerpc64-linux-gnu- Fixes: 867c89254425 ("HID: uclogic: Allow to generate frame templates") Signed-off-by: José Expósito <jose.exposito89@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: fix 'cast to restricted' kernel CI bot warningsMichael Zaidman1-7/+7
Fix 'cast to restricted' sparse warnings reported by kernel test robot in https://lore.kernel.org/all/202211021607.ssjymlKi-lkp@intel.com/ Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: missed NACK from busy deviceMichael Zaidman1-5/+22
When writing into a slow device like an EEPROM chip, the controller may exit the busy state before the device releases the bus. In this case, the ft260_xfer_status returns success before the data transfer completion. The patch fixes it by returning from the ft260_xfer_status() with the "-EAGAIN" on both controller and bus busy status when appropriate. It does not apply to the i2c combined transactions when after the write IO, the controller keeps the bus busy until the read IO and then between reading IOs to ensure an atomic operation. Co-developed-by: Germain Hebert <germain.hebert@ca.abb.com> Signed-off-by: Germain Hebert <germain.hebert@ca.abb.com> Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: fix a NULL pointer dereference in ft260_i2c_writeMichael Zaidman1-0/+3
The zero-length passed into the ft260_i2c_write() triggered the NULL pointer dereference in the debug message on data[0] access. Since the controller does not support a write of zero length, let's not allow it. Before: $ sudo i2ctransfer -y 13 w0@0x51 Killed After: $ sudo i2ctransfer -y 13 w0@0x51 Error: Sending messages failed: Invalid argument Reported-by: Enrik Berkhan <Enrik.Berkhan@inka.de> Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: wake up device from power saving modeMichael Zaidman1-0/+17
The FT260 can enter a power saving mode after being idle for longer than 5 seconds. When being woken up from power saving mode by an I2C write request, a possible NACK is not correctly reported by the controller. As a workaround, the driver will issue an I2C status report two times in ft260_xfer_status() after the chip has been idle for more than 5s. Co-developed-by: Enrik Berkhan <Enrik.Berkhan@inka.de> Signed-off-by: Enrik Berkhan <Enrik.Berkhan@inka.de> Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: missed NACK from big i2c readMichael Zaidman1-2/+4
The FT260 controller does not return NACK when performing a big read (of multiple hid reports size) from a non-existing device or from the device responding with NACK when it is not ready to serve the request. However, it responds correctly with NACK to a read of up to a single hid report size. To overcome this issue, we split the muli-report read request into a read of a single HID report of 60 bytes size and a multi-report read. Big read of 256 bytes with first read of 60 bytes: $ sudo ./i2cperf -d 2 -o 2 -s 256 -r 0-0xff 1 0x50 -S [ +5.633280] ft260_i2c_write_read: off 0x0 rlen 255 wlen 2 [ +0.000006] ft260_i2c_write: rep 0xd0 addr 0x50 off 0 len 2 wlen 2 flag 0x2 d[0] 0x0 [ +0.013205] ft260_xfer_status: bus_status 0x20, clock 100 [ +0.000007] ft260_i2c_read: rep 0xc2 addr 0x50 len 255 rlen 60 flag 0x3 [ +0.010932] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.004733] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000006] ft260_i2c_read: rep 0xc2 addr 0x50 len 195 rlen 128 flag 0x0 [ +0.012572] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.005789] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.003189] ft260_raw_event: i2c resp: rep 0xd1 len 8 [ +0.004092] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000010] ft260_i2c_read: rep 0xc2 addr 0x50 len 67 rlen 67 flag 0x4 [ +0.011688] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.004700] ft260_raw_event: i2c resp: rep 0xd1 len 7 [ +0.004858] ft260_xfer_status: bus_status 0x20, clock 100 Read from non-existing device at address 8. The first 60 read responded with NACK. $ sudo ./i2cperf -d 2 -o 2 -s 256 -r 0-0xff 1 0x8 -S [Oct19 15:37] ft260_i2c_write_read: off 0x0 rlen 255 wlen 2 [ +0.000007] ft260_i2c_write: rep 0xd0 addr 0x8 off 0 len 2 wlen 2 flag 0x2 d[0] 0x0 [ +0.022820] ft260_xfer_status: bus_status 0x20, clock 100 [ +0.000007] ft260_i2c_read: rep 0xc2 addr 0x8 len 255 rlen 60 flag 0x3 [ +0.010658] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.005965] ft260_xfer_status: bus_status 0x46, clock 100 <-- NACK [ +0.000009] ft260 0003:0403:6030.0004: i2c bus error: 0x46 [ +0.007784] ft260_i2c_reset: done Co-developed-by: Enrik Berkhan <Enrik.Berkhan@inka.de> Signed-off-by: Enrik Berkhan <Enrik.Berkhan@inka.de> Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: remove SMBus Quick command supportMichael Zaidman1-9/+1
The i2cdetect uses the SMBus Quick command by default to scan devices on the I2C bus. The FT260 implements an I2C bus controller. The SMBus is derived from I2C, but there are several differences between the specifications of the two buses in the areas of timing, protocols, operation modes, and electrical characteristics. One of the differences is that the I2C devices allow the slave not to ACK its slave address, but SMBus requires it to always ACK it as a mechanism to detect a detachable device’s presence on the bus. Since FT260 is the I2C bus controller, it does not acknowledge the SMBus Quick write command, which sends a single bit to the device at the place of the RD/WR bit. The ft260 driver attempted to mimic the SMBus Quick Write functionality by writing a single byte as the SMBus Byte Write command does. Usually, one byte in the SMBus Quick Write will be fine. However, it may cause problems with devices with a control register at offset 0, like i2c muxes, for example, when scanned with the i2cdetect utility. The i2cdetect with the "-r" option uses the SMBus Read Byte command, which is a reasonable workaround. To prevent the I2C bus from locking at write-only devices (most notably clock chips at address 0x69), use the "-r" option in conjunction with scanning range parameters. This patch removes the SMBus Quick command support. $ sudo i2cdetect -y 13 Warning: Can't use SMBus Quick Write command, will skip some addresses 0 1 2 3 4 5 6 7 8 9 a b c d e f 00: 10: 20: 30: -- -- -- -- -- -- -- -- 40: 50: 50 51 -- -- -- -- -- -- -- -- -- -- -- -- -- -- 60: 70: $ sudo i2cdetect -y -r 13 0 1 2 3 4 5 6 7 8 9 a b c d e f 00: -- -- -- -- -- -- -- -- 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 50: 50 51 -- -- -- -- -- -- -- -- -- -- -- -- -- -- 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 70: -- -- -- -- -- -- -- -- Reported-by: Vince Asbridge <VAsbridge@sanblaze.com> Reported-by: Stephen Shirron <SShirron@sanblaze.com> Reported-by: Enrik Berkhan <Enrik.Berkhan@inka.de> Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: skip unexpected HID input reportsMichael Zaidman1-12/+24
The FT260 is not supposed to generate unexpected HID reports. However, in theory, the unsolicited HID Input reports can be issued by a specially crafted malicious USB device masquerading as FT260 when the attacker has physical access to the USB port. In this case, the read_buf pointer points to the final data portion of the previous I2C Read transfer, and the memcpy invoked in the ft260_raw_event() will try copying the content of the unexpected report into the wrong location. This commit sets the Read buffer pointer to NULL on the I2C Read transaction completion and checks it in the ft260_raw_event() to detect and skip the unsolicited Input report. Reported-by: Enrik Berkhan <Enrik.Berkhan@inka.de> Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: do not populate /dev/hidraw deviceMichael Zaidman1-3/+6
Do not populate the /dev/hidraw on ft260 interfaces when the hid-ft260 driver is loaded. $ sudo insmod hid-ft260.ko $ ls /dev/hidraw* /dev/hidraw0 $ sudo rmmod hid-ft260.ko $ ls /dev/hidraw* /dev/hidraw0 /dev/hidraw1 /dev/hidraw2 Reported-by: Enrik Berkhan <Enrik.Berkhan@inka.de> Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: improve i2c large reads performanceMichael Zaidman1-5/+12
The patch increases the read buffer size to 180 bytes. It reduces the number of ft260_i2c_read() calls by three, improving the big reads performance. $ sudo i2ctransfer -y -f 13 w2@0x51 0x0 0x0 r180 Before: [ +4.071878] ft260_i2c_write_read: off 0x0 rlen 180 wlen 2 [ +0.000005] ft260_i2c_write: rep 0xd0 addr 0x51 off 0 len 2 wlen 2 flag 0x2 d[0] 0x0 [ +0.001097] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000175] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000004] ft260_i2c_read: rep 0xc2 addr 0x51 len 180 rlen 60 flag 0x3 [ +0.008579] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.000208] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000001] ft260_i2c_read: rep 0xc2 addr 0x51 len 120 rlen 60 flag 0x0 [ +0.008794] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.000181] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000002] ft260_i2c_read: rep 0xc2 addr 0x51 len 60 rlen 60 flag 0x4 [ +0.008817] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.000223] ft260_xfer_status: bus_status 0x20, clock 100 After: [ +11.611642] ft260_i2c_write_read: off 0x0 rlen 180 wlen 2 [ +0.000005] ft260_i2c_write: rep 0xd0 addr 0x51 off 0 len 2 wlen 2 flag 0x2 d[0] 0x0 [ +0.008001] ft260_xfer_status: bus_status 0x20, clock 100 [ +0.000001] ft260_i2c_read: rep 0xc2 addr 0x51 len 180 rlen 180 flag 0x7 [ +0.008994] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.007987] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.007992] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.000206] ft260_xfer_status: bus_status 0x20, clock 100 Suggested-by: Enrik Berkhan <Enrik.Berkhan@inka.de> Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: support i2c reads greater than HID report sizeMichael Zaidman1-61/+66
A random i2c read operation in EEPROM devices is implemented as a dummy write operation, followed by a current address read operation. The dummy write operation is used to load the target byte or word address (a.k.a offset) into the offset counter, from which the subsequent read operation then reads. To support longer than one HID report size random read, the ft260 driver issues multiple pairs of i2c write offset + read data transactions of HID report size so that the EEPROM device sees many i2c random read requests from different offsets. Two issues with the current implementation: - This approach suffers from extra overhead caused by writing offset requests. - Necessity to handle offset per HID report in big-endian representation as EEPROM devices expect. The current implementation does not do it and correctly handles the reads up to 60 bytes only. This patch addresses both issues by implementing more efficient approach. It issues a single i2c read request of up to the EEPROM page size and then waits for the data to arrive in multiple HID reports. For example, to read the 256 bytes from a 24LC512 chip, which has 128 bytes page size, the old method performs six ft260_i2c_write_read transactions while the new - two only. Before: $ sudo ./i2cperf -d 2 -o 2 -s 128 -r 0-0xff 13 0x51 -S Read block via i2ctransfer by chunks ------------------------------------------------------------------- data rate(bps) efficiency(%) data size(B) total IOs IO size(B) ------------------------------------------------------------------- 40803 85 256 2 128 Kernel log of a single 128 bytes read request: [ +2.376308] ft260_i2c_write_read: read_off 0x0 left_len 128 len 60 [ +0.000002] ft260_i2c_write: rep 0xd0 addr 0x51 off 0 len 2 wlen 2 flag 0x2 d[0] 0x0 [ +0.000707] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000173] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000001] ft260_i2c_read: rep 0xc2 addr 0x51 len 60 [ +0.008660] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.000156] ft260_xfer_status: bus_status 0x20, clock 100 [ +0.000001] ft260_i2c_write_read: read_off 0x3c left_len 68 len 60 [ +0.000001] ft260_i2c_write: rep 0xd0 addr 0x51 off 0 len 2 wlen 2 flag 0x2 d[0] 0x3c [ +0.001034] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000191] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000001] ft260_i2c_read: rep 0xc2 addr 0x51 len 60 [ +0.008614] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.000203] ft260_xfer_status: bus_status 0x20, clock 100 [ +0.000001] ft260_i2c_write_read: read_off 0x78 left_len 8 len 8 [ +0.000001] ft260_i2c_write: rep 0xd0 addr 0x51 off 0 len 2 wlen 2 flag 0x2 d[0] 0x78 [ +0.000987] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000192] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000001] ft260_i2c_read: rep 0xc2 addr 0x51 len 8 [ +0.002614] ft260_raw_event: i2c resp: rep 0xd1 len 8 [ +0.000200] ft260_xfer_status: bus_status 0x20, clock 100 After: $ sudo ./i2cperf -d 2 -o 2 -s 128 -r 0-0xff 13 0x51 -S Read block via i2ctransfer by chunks ------------------------------------------------------------------- data rate(bps) efficiency(%) data size(B) total IOs IO size(B) ------------------------------------------------------------------- 43990 85 256 2 128 Kernel log of a single 128 bytes read request: [ +1.464346] ft260_i2c_write_read: off 0x0 rlen 128 wlen 2 [ +0.000002] ft260_i2c_write: rep 0xd0 addr 0x51 off 0 len 2 wlen 2 flag 0x2 d[0] 0x0 [ +0.001653] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000188] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000002] ft260_i2c_read: rep 0xc2 addr 0x51 len 128 rlen 60 flag 0x3 [ +0.008609] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.000157] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000002] ft260_i2c_read: rep 0xc2 addr 0x51 len 68 rlen 60 flag 0x0 [ +0.008840] ft260_raw_event: i2c resp: rep 0xde len 60 [ +0.000203] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000002] ft260_i2c_read: rep 0xc2 addr 0x51 len 8 rlen 8 flag 0x4 [ +0.002794] ft260_raw_event: i2c resp: rep 0xd1 len 8 [ +0.000201] ft260_xfer_status: bus_status 0x20, clock 100 Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Tested-by: Guillaume Champagne <champagne.guillaume.c@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: support i2c writes larger than HID report sizeMichael Zaidman1-18/+23
To support longer than one HID report size write, the driver splits a single i2c message data payload into multiple i2c messages of HID report size. However, it does not replicate the offset bytes within the EEPROM chip in every consequent HID report because it is not and should not be aware of the EEPROM type. It breaks the i2c write message integrity and causes the EEPROM device not to acknowledge the second HID report keeping the i2c bus busy until the ft260 controller reports failure. This patch preserves the i2c write message integrity by manipulating the i2c flag bits across multiple HID reports to be seen by the EEPROM device as a single i2c write transfer. Before: $ sudo ./i2cperf -f 2 -o 2 -s 64 -r 0-0xff 13 0x51 -S Error: Sending messages failed: Input/output error [ +3.667741] ft260_i2c_write: rep 0xde addr 0x51 off 0 len 60 d[0] 0x0 [ +0.007330] ft260_hid_output_report_check_status: wait 6400 usec, len 64 [ +0.000203] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.000001] ft260_i2c_write: rep 0xd1 addr 0x51 off 60 len 6 d[0] 0x0 [ +0.002337] ft260_hid_output_report_check_status: wait 1000 usec, len 10 [ +0.000157] ft260_xfer_status: bus_status 0x2e, clock 100 [ +0.000241] ft260_i2c_reset: done [ +0.000003] ft260_i2c_write: failed to start transfer, ret -5 After: $ sudo ./i2cperf -f 2 -o 2 -s 128 -r 0-0xff 13 0x51 -S Fill block with increment via i2ctransfer by chunks ------------------------------------------------------------------- data rate(bps) efficiency(%) data size(B) total IOs IO size(B) ------------------------------------------------------------------- 71260 86 256 2 128 Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Tested-by: Guillaume Champagne <champagne.guillaume.c@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: improve i2c write performanceMichael Zaidman1-5/+9
The patch improves the I2C write performance by 20 - 30 percent by revising the sleep time in the ft260_hid_output_report_check_status() in the following ways: 1. Reduce the wait time and start to poll earlier. Sending a large amount of data at a low I2C clock rate saturates the internal FT260 buffer and causes hiccups in status readiness, as shown below in the log fragment. Aligning the status check wait time to the worst case significantly reduces the write performance. [Oct22 10:28] ft260_i2c_write: rep 0xd8 addr 0x51 off 0 len 34 d[0] 0x0 [ +0.005296] ft260_xfer_status: bus_status 0x20, clock 100 [ +0.013460] ft260_i2c_write: rep 0xd8 addr 0x51 off 0 len 34 d[0] 0x0 [ +0.003244] ft260_hid_output_report_check_status: wait 1920 usec, len 38 [ +0.000190] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.015324] ft260_i2c_write: rep 0xd8 addr 0x51 off 0 len 34 d[0] 0x0 [ +0.003491] ft260_hid_output_report_check_status: wait 1920 usec, len 38 [ +0.000202] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.016047] ft260_i2c_write: rep 0xd8 addr 0x51 off 0 len 34 d[0] 0x0 [ +0.002768] ft260_hid_output_report_check_status: wait 1920 usec, len 38 [ +0.000150] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.011389] ft260_i2c_write: rep 0xd8 addr 0x51 off 0 len 34 d[0] 0x0 [ +0.003467] ft260_hid_output_report_check_status: wait 1920 usec, len 38 [ +0.000191] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000172] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000131] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000241] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000233] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000190] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000196] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.011314] ft260_i2c_write: rep 0xd8 addr 0x51 off 0 len 34 d[0] 0x0 [ +0.003334] ft260_hid_output_report_check_status: wait 1920 usec, len 38 [ +0.000227] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000204] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000198] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000147] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.011060] ft260_i2c_write: rep 0xd8 addr 0x51 off 0 len 34 d[0] 0x0 Before: $ sudo ./i2cperf -f 2 -o 2 -s 32 -r 0-0xff 13 0x51 -S Fill block with increment via i2ctransfer by chunks ------------------------------------------------------------------- data rate(bps) efficiency(%) data size(B) total IOs IO size(B) ------------------------------------------------------------------- 40510 80 256 8 32 After: $ sudo ./i2cperf -f 2 -o 2 -s 32 -r 0-0xff 13 0x51 -S Fill block with increment via i2ctransfer by chunks ------------------------------------------------------------------- data rate(bps) efficiency(%) data size(B) total IOs IO size(B) ------------------------------------------------------------------- 52584 80 256 8 32 2. Do not sleep if the estimated I2C transfer time is below 2 ms since the first xfer status query frequently takes around 1.5 ms, and the following status queries take about 200us on average. So we usually return from the routine after the first 1 - 3 status checks. [Oct22 11:14] ft260_i2c_write: rep 0xd4 addr 0x51 off 0 len 18 d[0] 0x0 [ +0.004270] ft260_xfer_status: bus_status 0x20, clock 100 [ +0.013889] ft260_i2c_write: rep 0xd4 addr 0x51 off 0 len 18 d[0] 0x0 [ +0.000856] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000138] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.013352] ft260_i2c_write: rep 0xd4 addr 0x51 off 0 len 18 d[0] 0x0 [ +0.001501] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000177] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.014477] ft260_i2c_write: rep 0xd4 addr 0x51 off 0 len 18 d[0] 0x0 [ +0.001377] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000233] ft260_xfer_status: bus_status 0x41, clock 100 [ +0.000191] ft260_xfer_status: bus_status 0x40, clock 100 [ +0.013197] ft260_i2c_write: rep 0xd4 addr 0x51 off 0 len 18 d[0] 0x0 Before: $ sudo ./i2cperf -f 2 -o 2 -s 16 -r 0-0xff 13 0x51 -S Fill block with increment via i2ctransfer by chunks ------------------------------------------------------------------- data rate(bps) efficiency(%) data size(B) total IOs IO size(B) ------------------------------------------------------------------- 28826 73 256 16 16 After: $ sudo ./i2cperf -f 2 -o 2 -s 16 -r 0-0xff 13 0x51 -S Fill block with increment via i2ctransfer by chunks ------------------------------------------------------------------- data rate(bps) efficiency(%) data size(B) total IOs IO size(B) ------------------------------------------------------------------- 45138 73 256 16 16 Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Tested-by: Guillaume Champagne <champagne.guillaume.c@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-11HID: ft260: ft260_xfer_status routine cleanupMichael Zaidman1-20/+10
After clarifying with FTDI's support, it turned out that the error condition (bit 1) in byte 1 of the i2c status HID report is a status bit reflecting all error conditions. When bits 2, 3, or 4 are raised to 1, bit 1 is set to 1 also. Since the ft260_xfer_status routine tests the error condition bit and exits in the case of an error, the program flow never reaches the conditional expressions for 2, 3, and 4 bits when any of them indicates an error state. Though these expressions are never evaluated to true, they are checked several times per IO, increasing the ft260_xfer_status polling cycle duration. The patch removes the conditional expressions for 2, 3, and 4 bits in byte 1 of the i2c status HID report. Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com> Tested-by: Guillaume Champagne <champagne.guillaume.c@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2022-11-10eth: sp7021: drop free_netdev() from spl2sw_init_netdev()Wei Yongjun1-1/+0
It's not necessary to free netdev allocated with devm_alloc_etherdev() and using free_netdev() leads to double free. Fixes: fd3040b9394c ("net: ethernet: Add driver for Sunplus SP7021") Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Link: https://lore.kernel.org/r/20221109150116.2988194-1-weiyongjun@huaweicloud.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-11-10MAINTAINERS: Move Vivien to CREDITSFlorian Fainelli2-2/+5
Last patch from Vivien was nearly 3 years ago and he has not reviewed or responded to DSA patches since then, move to CREDITS. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Link: https://lore.kernel.org/r/20221109231907.621678-1-f.fainelli@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-11-10net: macvlan: fix memory leaks of macvlan_common_newlinkChuang Wang1-1/+3
kmemleak reports memory leaks in macvlan_common_newlink, as follows: ip link add link eth0 name .. type macvlan mode source macaddr add <MAC-ADDR> kmemleak reports: unreferenced object 0xffff8880109bb140 (size 64): comm "ip", pid 284, jiffies 4294986150 (age 430.108s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 b8 aa 5a 12 80 88 ff ff ..........Z..... 80 1b fa 0d 80 88 ff ff 1e ff ac af c7 c1 6b 6b ..............kk backtrace: [<ffffffff813e06a7>] kmem_cache_alloc_trace+0x1c7/0x300 [<ffffffff81b66025>] macvlan_hash_add_source+0x45/0xc0 [<ffffffff81b66a67>] macvlan_changelink_sources+0xd7/0x170 [<ffffffff81b6775c>] macvlan_common_newlink+0x38c/0x5a0 [<ffffffff81b6797e>] macvlan_newlink+0xe/0x20 [<ffffffff81d97f8f>] __rtnl_newlink+0x7af/0xa50 [<ffffffff81d98278>] rtnl_newlink+0x48/0x70 ... In the scenario where the macvlan mode is configured as 'source', macvlan_changelink_sources() will be execured to reconfigure list of remote source mac addresses, at the same time, if register_netdevice() return an error, the resource generated by macvlan_changelink_sources() is not cleaned up. Using this patch, in the case of an error, it will execute macvlan_flush_sources() to ensure that the resource is cleaned up. Fixes: aa5fd0fb7748 ("driver: macvlan: Destroy new macvlan port if macvlan_common_newlink failed.") Signed-off-by: Chuang Wang <nashuiliang@gmail.com> Link: https://lore.kernel.org/r/20221109090735.690500-1-nashuiliang@gmail.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10ethernet: tundra: free irq when alloc ring failed in tsi108_open()Zhengchao Shao1-1/+4
When alloc tx/rx ring failed in tsi108_open(), it doesn't free irq. Fix it. Fixes: 5e123b844a1c ("[PATCH] Add tsi108/9 On Chip Ethernet device driver support") Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> Link: https://lore.kernel.org/r/20221109044016.126866-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10ALSA: hda: fix potential memleak in 'add_widget_node'Ye Bin1-1/+3
As 'kobject_add' may allocated memory for 'kobject->name' when return error. And in this function, if call 'kobject_add' failed didn't free kobject. So call 'kobject_put' to recycling resources. Signed-off-by: Ye Bin <yebin10@huawei.com> Cc: <stable@vger.kernel.org> Link: https://lore.kernel.org/r/20221110144539.2989354-1-yebin@huaweicloud.com Signed-off-by: Takashi Iwai <tiwai@suse.de>
2022-11-10ALSA: memalloc: Don't fall back for SG-buffer with IOMMUTakashi Iwai1-9/+11
When the non-contiguous page allocation for SG buffer allocation fails, the memalloc helper tries to fall back to the old page allocation methods. This would, however, result in the bogus page addresses when IOMMU is enabled. Usually in such a case, the fallback allocation should fail as well, but occasionally it succeeds and hitting a bad access. The fallback was thought for non-IOMMU case, and as the error from dma_alloc_noncontiguous() with IOMMU essentially implies a fatal memory allocation error, we should return the error straightforwardly without fallback. This avoids the corner case like the above. The patch also renames the local variable "dma_ops" with snd_ prefix for avoiding the name conflict. Fixes: a8d302a0b770 ("ALSA: memalloc: Revive x86-specific WC page allocations again") Reported-by: Kai Vehmanen <kai.vehmanen@linux.intel.com> Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com> Link: https://lore.kernel.org/r/alpine.DEB.2.22.394.2211041541090.3532114@eliteleevi.tm.intel.com Link: https://lore.kernel.org/r/20221110132216.30605-1-tiwai@suse.de Signed-off-by: Takashi Iwai <tiwai@suse.de>
2022-11-10net: mv643xx_eth: disable napi when init rxq or txq failed in mv643xx_eth_open()Zhengchao Shao1-0/+1
When failed to init rxq or txq in mv643xx_eth_open() for opening device, napi isn't disabled. When open mv643xx_eth device next time, it will trigger a BUG_ON() in napi_enable(). Compile tested only. Fixes: 2257e05c1705 ("mv643xx_eth: get rid of receive-side locking") Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> Link: https://lore.kernel.org/r/20221109025432.80900-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10ethernet: s2io: disable napi when start nic failed in s2io_card_up()Zhengchao Shao1-9/+20
When failed to start nic or add interrupt service routine in s2io_card_up() for opening device, napi isn't disabled. When open s2io device next time, it will trigger a BUG_ON()in napi_enable(). Compile tested only. Fixes: 5f490c968056 ("S2io: Fixed synchronization between scheduling of napi with card reset and close") Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> Link: https://lore.kernel.org/r/20221109023741.131552-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10net: atlantic: macsec: clear encryption keys from the stackAntoine Tenart2-7/+13
Commit aaab73f8fba4 ("macsec: clear encryption keys from the stack after setting up offload") made sure to clean encryption keys from the stack after setting up offloading, but the atlantic driver made a copy and did not clear it. Fix this. [4 Fixes tags below, all part of the same series, no need to split this] Fixes: 9ff40a751a6f ("net: atlantic: MACSec ingress offload implementation") Fixes: b8f8a0b7b5cb ("net: atlantic: MACSec ingress offload HW bindings") Fixes: 27736563ce32 ("net: atlantic: MACSec egress offload implementation") Fixes: 9d106c6dd81b ("net: atlantic: MACSec egress offload HW bindings") Signed-off-by: Antoine Tenart <atenart@kernel.org> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10net: phy: mscc: macsec: clear encryption keys when freeing a flowAntoine Tenart1-0/+1
Commit aaab73f8fba4 ("macsec: clear encryption keys from the stack after setting up offload") made sure to clean encryption keys from the stack after setting up offloading, but the MSCC PHY driver made a copy, kept it in the flow data and did not clear it when freeing a flow. Fix this. Fixes: 28c5107aa904 ("net: phy: mscc: macsec support") Signed-off-by: Antoine Tenart <atenart@kernel.org> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10stmmac: dwmac-loongson: fix missing of_node_put() while module exitingYang Yingliang1-5/+14
The node returned by of_get_child_by_name() with refcount decremented, of_node_put() needs be called when finish using it. So add it in the error path in loongson_dwmac_probe() and in loongson_dwmac_remove(). Fixes: 2ae34111fe4e ("stmmac: dwmac-loongson: fix invalid mdio_node") Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10stmmac: dwmac-loongson: fix missing pci_disable_device() in loongson_dwmac_probe()Yang Yingliang1-2/+5
Add missing pci_disable_device() in the error path in loongson_dwmac_probe(). Fixes: 30bba69d7db4 ("stmmac: pci: Add dwmac support for Loongson") Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10stmmac: dwmac-loongson: fix missing pci_disable_msi() while module exitingYang Yingliang1-1/+12
pci_enable_msi() has been called in loongson_dwmac_probe(), so pci_disable_msi() needs be called in remove path and error path of probe(). Fixes: 30bba69d7db4 ("stmmac: pci: Add dwmac support for Loongson") Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10ALSA: usb-audio: add quirk to fix Hamedal C20 disconnect issueAi Chao3-1/+7
For Hamedal C20, the current rate is different from the runtime rate, snd_usb_endpoint stop and close endpoint to resetting rate. if snd_usb_endpoint close the endpoint, sometimes usb will disconnect the device. Signed-off-by: Ai Chao <aichao@kylinos.cn> Link: https://lore.kernel.org/r/20221110063452.295110-1-aichao@kylinos.cn Signed-off-by: Takashi Iwai <tiwai@suse.de>
2022-11-10ALSA: hda/realtek: Add Positivo C6300 model quirkEdson Juliano Drosdeck1-0/+1
Positivo Master C6300 (1849:a233) require quirk for anabling headset-mic Signed-off-by: Edson Juliano Drosdeck <edson.drosdeck@gmail.com> Cc: <stable@vger.kernel.org> Link: https://lore.kernel.org/r/20221109171732.5417-1-edson.drosdeck@gmail.com Signed-off-by: Takashi Iwai <tiwai@suse.de>
2022-11-09cxgb4vf: shut down the adapter when t4vf_update_port_info() failed in cxgb4vf_open()Zhengchao Shao1-1/+1
When t4vf_update_port_info() failed in cxgb4vf_open(), resources applied during adapter goes up are not cleared. Fix it. Only be compiled, not be tested. Fixes: 18d79f721e0a ("cxgb4vf: Update port information in cxgb4vf_open()") Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> Link: https://lore.kernel.org/r/20221109012100.99132-1-shaozhengchao@huawei.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-11-09mctp: Fix an error handling path in mctp_init()Wei Yongjun2-2/+4
If mctp_neigh_init() return error, the routes resources should be released in the error handling path. Otherwise some resources leak. Fixes: 4d8b9319282a ("mctp: Add neighbour implementation") Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Acked-by: Matt Johnston <matt@codeconstruct.com.au> Link: https://lore.kernel.org/r/20221108095517.620115-1-weiyongjun@huaweicloud.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-11-09stmmac: intel: Update PCH PTP clock rate from 200MHz to 204.8MHzTan, Tee Min1-2/+9
Current Intel platform has an output of ~976ms interval when probed on 1 Pulse-per-Second(PPS) hardware pin. The correct PTP clock frequency for PCH GbE should be 204.8MHz instead of 200MHz. PSE GbE PTP clock rate remains at 200MHz. Fixes: 58da0cfa6cf1 ("net: stmmac: create dwmac-intel.c to contain all Intel platform") Signed-off-by: Ling Pei Lee <pei.lee.ling@intel.com> Signed-off-by: Tan, Tee Min <tee.min.tan@intel.com> Signed-off-by: Voon Weifeng <weifeng.voon@intel.com> Signed-off-by: Gan Yi Fang <yi.fang.gan@intel.com> Link: https://lore.kernel.org/r/20221108020811.12919-1-yi.fang.gan@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-11-09net: cxgb3_main: disable napi when bind qsets failed in cxgb_up()Zhengchao Shao1-0/+1
When failed to bind qsets in cxgb_up() for opening device, napi isn't disabled. When open cxgb3 device next time, it will trigger a BUG_ON() in napi_enable(). Compile tested only. Fixes: 48c4b6dbb7e2 ("cxgb3 - fix port up/down error path") Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> Link: https://lore.kernel.org/r/20221109021451.121490-1-shaozhengchao@huawei.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-11-09net: cpsw: disable napi in cpsw_ndo_open()Zhengchao Shao1-0/+2
When failed to create xdp rxqs or fill rx channels in cpsw_ndo_open() for opening device, napi isn't disabled. When open cpsw device next time, it will report a invalid opcode issue. Compiled tested only. Fixes: d354eb85d618 ("drivers: net: cpsw: dual_emac: simplify napi usage") Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> Link: https://lore.kernel.org/r/20221109011537.96975-1-shaozhengchao@huawei.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-11-09drm/amd/display: only fill dirty rectangles when PSR is enabledHamza Mahfooz1-3/+4
Currently, we are calling fill_dc_dirty_rects() even if PSR isn't supported by the relevant link in amdgpu_dm_commit_planes(), this is undesirable especially because when drm.debug is enabled we are printing messages in fill_dc_dirty_rects() that are only useful for debugging PSR (and confusing otherwise). So, we can instead limit the filling of dirty rectangles to only when PSR is enabled. Reviewed-by: Leo Li <sunpeng.li@amd.com> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-11-09drm/amdgpu: disable BACO on special BEIGE_GOBY cardGuchun Chen1-1/+3
Still avoid intermittent failure. Signed-off-by: Guchun Chen <guchun.chen@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Acked-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2022-11-09drm/amdgpu: Drop eviction lock when allocating PT BOPhilip Yang3-26/+28
Re-take the eviction lock immediately again after the allocation is completed, to fix circular locking warning with drm_buddy allocator. Move amdgpu_vm_eviction_lock/unlock/trylock to amdgpu_vm.h as they are called from multiple files. Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-11-09drm/amdgpu: Unlock bo_list_mutex after error handlingPhilip Yang1-0/+1
Get below kernel WARNING backtrace when pressing ctrl-C to kill kfdtest application. If amdgpu_cs_parser_bos returns error after taking bo_list_mutex, as caller amdgpu_cs_ioctl will not unlock bo_list_mutex, this generates the kernel WARNING. Add unlock bo_list_mutex after amdgpu_cs_parser_bos error handling to cleanup bo_list userptr bo. WARNING: kfdtest/2930 still has locks held! 1 lock held by kfdtest/2930: (&list->bo_list_mutex){+.+.}-{3:3}, at: amdgpu_cs_ioctl+0xce5/0x1f10 [amdgpu] stack backtrace: dump_stack_lvl+0x44/0x57 get_signal+0x79f/0xd00 arch_do_signal_or_restart+0x36/0x7b0 exit_to_user_mode_prepare+0xfd/0x1b0 syscall_exit_to_user_mode+0x19/0x40 do_syscall_64+0x40/0x80 Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.