aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/mxc/mlb
Commit message (Collapse)AuthorAge
* ENGR00314144 mxc: mlb: Rename the mxc_mlb150 to mxc_mlb.luweizhou2014-06-24
| | | | | | | | | Since i.MX6SX doesn't supports MLB150 , it is not strictly explicit to name driver module as mxc_mlb150.ko. Rename it to mxc_mlb.ko.It would be more common. Signed-off-by: Luwei Zhou <b45643@freescale.com> (cherry picked from commit dbe1a350e2f61e827501fd322951b9ea255461a5)
* ENGR00303300-1 mxc: mlb: Add MLB50 driver support on i.MX6SX chipLuwei Zhou2014-04-16
| | | | | | | Add MLB50 driver supported on i.MX6SX chip. MLB50 will not support frame rate over 1024 per second. Signed-off-by: Luwei Zhou <b45643@freescale.com>
* ENGR00290679 mxc: mlb Fix the MLB150 not stable issue in ISOC mode.Luwei Zhou2014-04-16
| | | | | | | | | | | The MLB test bench has bug when testing the ISOC mode.When we press stop test button when completing test, MITB will cause MLB150 on ARD error.The mlbintr ISR handler is used to handle the error interrupt. In the ISR handler, the MLB150_MS0,MLB150MS1 should be cleared. The spec doesn't give detailed description. The spec only says that the registers need to be cleared before enabling interrupt. Signed-off-by: Luwei Zhou <b45643@freescale.com>
* ENGR00290176 mxc: mlb: Fix MLB suspend/resume issueLuwei Zhou2014-04-16
| | | | | | | There is risk that mlb register will be access when clock is closed. This patch fix it. Signed-off-by: Luwei Zhou <b45643@freescale.com>
* ENGR00270037 mxc: mlb: Fix MLB crash when testing abnormallyLuwei Zhou2014-04-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If quit the test program via CTRL+c during the test and leaving the MITB still running, kernel crash sometimes happen when launching the test program for a second time. This patch fix this issue. The main modification is: * Initialize the wait queue head dynamically not statically * Enable/Disalbe IRQ when necessary Unable to handle kernel NULL pointer dereference at virtual address 00000000 pgd = 80004000 [00000000] *pgd=00000000 Internal error: Oops: 80000007 [#1] SMP ARM Modules linked in: mxc_mlb150 CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.10.17-16879-g5d48ab5 #227 task: 80c2d908 ti: 80c22000 task.ti: 80c22000 PC is at 0x0 LR is at __wake_up_common+0x54/0x94 pc : [<00000000>] lr : [<8004b9fc>] psr: 90000193 sp : 80c23e18 ip : dc86ff1c fp : 80c23e44 r10: 00000000 r9 : 00000001 r8 : 00000000 r7 : 00000000 r6 : 7f002fe0 r5 : 7f0017fc r4 : dcaff0f4 r3 : 00000000 r2 : 00000000 r1 : 00000001 r0 : dc86ff1c Flags: NzcV IRQs off FIQs on Mode SVC_32 ISA ARM Segment kernel Control: 10c53c7d Table: 6c90004a DAC: 00000015 Process swapper/0 (pid: 0, stack limit = 0x80c22238) Stack: (0x80c23e18 to 0x80c24000) 3e00: 00000000 00000001 3e20: dc1d14c0 7f002fdc 20000193 00000001 00000001 00000000 00000000 80c7018e 3e40: 80c23e6c 8004bbf4 00000000 8004bbf4 00000004 0091a840 7f002f80 7f002e1c 3e60: 00000004 fffffff9 00000001 7f001054 ffffae63 00000009 0000005a 00000000 3e80: ffffffff 00000010 00000095 00000000 00000000 00000095 dc011180 7f001168 3ea0: dc482e40 80073c08 00000015 80c2a770 80c1e7e0 dc011180 00000095 00000000 3ec0: f4000100 00000000 00000000 80c22000 80c2a4d8 80073d70 00000000 dc011180 3ee0: 00000095 80076ae8 00000095 800733d0 80c1ee3c 8000e848 f400010c 80c2a8b8 3f00: 80c23f20 80008570 8005a15c 804299d0 60000013 ffffffff 80c23f54 8000dbc0 3f20: 80c23f68 0000005a 3437dc5e 00000015 34373d83 00000015 81aef080 80c30050 3f40: 00000000 00000000 80c22000 80c2a4d8 00000017 80c23f68 8005a15c 804299d0 3f60: 60000013 ffffffff 3437dc5e 00000015 80cc41a4 806152ac 81aef080 80cc41a4 3f80: 00000000 80c30050 00000000 80429b10 00000001 80c7017a 80c2a524 806152ac 3fa0: 80c22000 80c7017a 80c22000 8000eb7c 00067162 800599f0 000000d9 80c12ef0 3fc0: 00000000 80bd6a9c ffffffff ffffffff 80bd6548 00000000 00000000 80c12ef0 3fe0: 10c53c7d 80c2a4a0 80c12eec 80c2e6ec 1000406a 10008074 00000000 00000000 [<8004b9fc>] (__wake_up_common+0x54/0x94) from [<8004bbf4>] (__wake_up+0x3c/0x50) [<8004bbf4>] (__wake_up+0x3c/0x50) from [<7f001054>] (mlb_tx_isr+0xa0/0xf4 [mxc_mlb150]) [<7f001054>] (mlb_tx_isr+0xa0/0xf4 [mxc_mlb150]) from [<7f001168>] (mlb_ahb_isr+0xc0/0x134 [mxc_mlb150]) [<7f001168>] (mlb_ahb_isr+0xc0/0x134 [mxc_mlb150]) from [<80073c08>] (handle_irq_event_percpu+0x54/0x17c) [<80073c08>] (handle_irq_event_percpu+0x54/0x17c) from [<80073d70>] (handle_irq_event+0x40/0x60) [<80073d70>] (handle_irq_event+0x40/0x60) from [<80076ae8>] (handle_fasteoi_irq+0x80/0x158) [<80076ae8>] (handle_fasteoi_irq+0x80/0x158) from [<800733d0>] (generic_handle_irq+0x2c/0x3c) [<800733d0>] (generic_handle_irq+0x2c/0x3c) from [<8000e848>] (handle_IRQ+0x40/0x90) [<8000e848>] (handle_IRQ+0x40/0x90) from [<80008570>] (gic_handle_irq+0x2c/0x5c) [<80008570>] (gic_handle_irq+0x2c/0x5c) from [<8000dbc0>] (__irq_svc+0x40/0x50 Signed-off-by: Luwei Zhou <b45643@freescale.com>
* ENGR00287826 mxc: mlb: Fix MLB rcu stall issue when MITB is not set correctlyLuwei Zhou2014-04-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | MLB lock and clock check has logic issue in code. The timeout cann't work and it will cause rcu-stall. Exception stack(0xdca21e68 to 0xdca21eb0) 1e60: 00000001 00000000 ffffff7f 0000270f 00002710 0000ffc0 1e80: 00000000 00000004 00000002 00000001 00000001 7f003b50 f0378000 dca21eb0 1ea0: 7f003b50 7f0009cc 20000013 ffffffff [<8000dc04>] (__irq_svc+0x44/0x58) from [<7f0009cc>] (mlb150_dev_unmute_syn_ch [<7f0009cc>] (mlb150_dev_unmute_syn_ch+0x34/0xd4 [mxc_mlb150]) from [<7f002584 [<7f002584>] (mxc_mlb150_ioctl+0x85c/0x898 [mxc_mlb150]) from [<800db86c>] (do_ [<800db86c>] (do_vfs_ioctl+0x40c/0x5e4) from [<800dba7c>] (SyS_ioctl+0x38/0x64) [<800dba7c>] (SyS_ioctl+0x38/0x64) from [<8000e000>] (ret_fast_syscall+0x0/0x48 INFO: rcu_sched self-detected stall on CPU 0: (14684 ticks this GP) idle=fdf/140000000000001/0 softirq=1555/1555 (t=14720 jiffies g=2 c=1 q=22) CPU: 0 PID: 628 Comm: mxc_mlb150_test Not tainted 3.10.17-16837-g187ed79-dirty [<80013d7c>] (unwind_backtrace+0x0/0xf4) from [<80011798>] (show_stack+0x10/0x [<80011798>] (show_stack+0x10/0x14) from [<8008575c>] (rcu_check_callbacks+0x3 [<8008575c>] (rcu_check_callbacks+0x3d0/0x7f8) from [<80033884>] (update_proce [<80033884>] (update_process_times+0x40/0x6c) from [<80064f48>] (tick_sched_ti [<80064f48>] (tick_sched_timer+0x4c/0x78) from [<80048508>] (__run_hrtimer.isr [<80048508>] (__run_hrtimer.isra.32+0x44/0xd4) from [<80048de4>] (hrtimer_inte [<80048de4>] (hrtimer_interrupt+0x108/0x294) from [<800136f0>] (twd_handler+0x [<800136f0>] (twd_handler+0x34/0x44) from [<80080208>] (handle_percpu_devid_ir [<80080208>] (handle_percpu_devid_irq+0x6c/0x84) from [<8007c958>] (generic_ha [<8007c958>] (generic_handle_irq+0x2c/0x3c) from [<8000e908>] (handle_IRQ+0x40 [<8000e908>] (handle_IRQ+0x40/0x90) from [<8000856c>] (gic_handle_irq+0x2c/0x5 [<8000856c>] (gic_handle_irq+0x2c/0x5c) from [<8000dc04>] (__irq_svc+0x44/0x5. Signed-off-by: Luwei Zhou <b45643@freescale.com>
* ENGR00287278 mxc: mlb: Enable the clock when open MLB device.Luwei Zhou2014-04-16
| | | | | | | The driver enalbe/disable the clock in probe()/remove(). It should be done in open()/close() hook function in fact. Signed-off-by: Luwei Zhou <b45643@freescale.com>
* ENGR00287256 mxc: mlb: Fix the SYNC mode hang issue when connected with MITBLuwei Zhou2014-04-16
| | | | | | | | | | | This patch fixes the hang and crash issue of MLB SYNC mode in the driver. The MITB will casue Sabreauto to hang and crash when testing the SYNC mode. It is because MITB will cause something error on MLB bus when stopping the SYNC test. The Sabreauto will keep entering error ISR and hang. Since we don't know the details about MITB, we make drivers provide IO_CTRL interface to disable the interrupt in SYNC mode. Signed-off-by: Luwei Zhou <b45643@freescale.com>
* ENGR00285441 mxc: mlb: Fix the receive issue of ISOC mode on SabreautoLuwei Zhou2014-04-16
| | | | | | | | | | | This patch fixes the receive issue of MLB ISO mode on Sabreauto side. During debug, it is found that the Sabreauto can only receive 1/5 packets of MITB sends.The MITB sends out 188 bytes per packet. The Sabreauto receives 188*5 bytes. It seems that the packet length on Sabreauto side is 5 times larger than MITB side. After changing the defautl block number to 1, Sabreauto passes the test. Signed-off-by: Luwei Zhou <b45643@freescale.com>
* ENGR00284648 mxc: mlb: clean up the MLB driver code.Luwei Zhou2014-04-16
| | | | | | | | | | | This patch clean up the MLB driver code of i.MX6Q/DL AUTO. * Reorder the head file alphabetically * Fix code comments format * Fix the lines over 80 characters * Fix the return type to void for some private functions Signed-off-by: Luwei Zhou <b45643@freescale.com>
* ENGR00283475 mxc: mlb : add MLB150 driver support.Luwei Zhou2014-04-16
This patch add MLB150 driver support.Copy files from 3.0.35 kernel. * Using of_xxx API to get resource from dts * Using devm_xxx API to simply the code * Change clk_enable()/clk_disable() to clk_prepare_enable()/ clk_disable_unprepare() * Fix some warning when compiling * Fix issue when handling errors in mxc_mlb150_probe() * Remove declaration of gpio_mlb_active(), gpio_mlb_inactive() out of mxc_mlb.h, because the functions are not invoked. * Add UAPI support. Signed-off-by: Luwei Zhou <b45643@freescale.com>
generated by cgit v1.2.2 (git 2.25.0) at 2025-04-04 15:26:57 -0400