- 22 Jul, 2008 40 commits
-
-
Paul Fulghum authored
Improve the accuracy of TIOCOUTQ value as implemented in chars_in_buffer() method by walking and counting tx DMA buffers, reading controller tx FIFO level and accounting for controller tx shift register. The greatly improves application control of transmit latency at lower data rates. Signed-off-by:
Paul Fulghum <paulkf@microgate.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Paul Fulghum authored
Add control of hardware serial bit order between LSB first (default/standard) and MSB first. Signed-off-by:
-
Paul Fulghum authored
Improve write method by allowing multiple HDLC frames to be loaded into tx DMA buffer ring for continuous frame transmission. This simplifies the transmit code by using the common procedures for all serial protocols. Signed-off-by:
-
Paul Fulghum authored
Make n_hdlc line discipline honor the O_NONBLOCK file flag on write. Signed-off-by:
-
Anton Vorontsov authored
With SMP kernels _irqsave spinlock disables only local interrupts, while the shared serial interrupt could be assigned to the CPU that is not currently starting up the serial port. This might cause issues because serial8250_startup() routine issues IRQ-triggering operations before registering the port in the IRQ chain (though, this is fine to do and done explicitly because we don't want to process any interrupts on the port startup). With RT kernels and preemptable hardirqs, _irqsave spinlock does not disable local hardirqs, and the bug could be reproduced much easily: $ cat /dev/ttyS0 & $ cat /dev/ttyS1 irq 42: nobody cared (try booting with the "irqpoll" option) Call Trace: [C0475EB0] [C0008A98] show_stack+0x4c/0x1ac (unreliable) [C0475EF0] [C004BBD4] __report_bad_irq+0x34/0xb8 [C0475F10] [C004BD38] note_interrupt+0xe0/0x308 [C0475F50] [C004B09C] thread_simple_irq+0xdc/0x104 [C0475F70] [C004B3FC] do_irqd+0x338/0x3c8 [C0475FC0] [C00398E0] kthread+0xf8/0x100 [C0475FF0] [C0011FE0] original_kernel_thread+0x44/0x60 handlers: [<c02112c4>] (serial8250_interrupt+0x0/0x138) Disabling IRQ #42 After this, all serial ports on the given IRQ are non-functional. To fix the issue we should explicitly disable shared IRQ before issuing any IRQ-triggering operations. I also changed spin_lock_irqsave to the ordinary spin_lock, since it seems to be safe: chain does not contain new port (yet), thus nobody will interfere us from the ISRs. Signed-off-by:
Anton Vorontsov <avorontsov@ru.mvista.com> Signed-off-by:
Alan Cox <alan@redhat.com> Signed-off-by:
-
David Howells authored
Fix the epca driver to call epca_setup() if digiepca=xxx is included on the command line and the epca driver is built in. epca_setup() used to be called from init/main.c in 2.2 kernels, but somewhere along the way that call was removed but not replaced. Signed-off-by:
David Howells <dhowells@redhat.com> Signed-off-by:
-
Jiri Slaby authored
Add support for CP-102UF moxa card (update to 1.12 original driver) and increment this driver version. (Somewhat reworked by alan@redhat.com to merge in with other patches) Signed-off-by:
-
Alan Cox authored
Use the hardware break support on the specialix driver Signed-off-by:
-
David Howells authored
Fix the Stallion driver's putchar() and break_ctl() ops and iStallion's putchar() to return values. Is it actually possible for putchar() or break_ctl() to be called with tty == NULL or can the check be discarded? Should stl_write() be returning 0 if tty->driver_data is NULL or tx.buf is NULL? Is this even possible? I've made Stallion's functions return -EINVAL as stli_breakctl() if the checks fail. Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Convert the driver to use the added hardware break support Signed-off-by:
-
Alan Cox authored
Some hardware needs to do break handling itself and may have partial support only. Make break_ctl return an error code. Add a tty driver flag so you can indicate driver hardware side break support. Signed-off-by:
-
Alan Cox authored
Nobody seems to use these drivers anyway so if they want them they can fix them up. I don't have the needed info to add break_ctl support to them. Send patches if you don't like it. Signed-off-by:
-
Alan Cox authored
Old style ioctl break code can go Signed-off-by:
-
Alan Cox authored
The ioctls it talks about are midlayer provided. Signed-off-by:
-
Alan Cox authored
Go through the inlines and other oddments that are iffy. Remove various bits of dead code and bogus debug. Turn the crtsdts compile time option into a runtime switch. Signed-off-by:
-
Alan Cox authored
Preparation for doing some real work on the driver. Do this first so we can easily identify if the cleanups accidentally broke something Signed-off-by:
-
Alan Cox authored
Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by:
-
Alan Cox authored
The edgeport reports negative error codes to functions that do not expect them. This can cause ports to jam forever Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
