Skip to content

L
linux
Switch branch/tag
  1. 17 Dec, 2010 2 commits
  3. 16 Dec, 2010 4 commits
  5. 10 Dec, 2010 8 commits
    • Ondrej Puzman's avatar
      8250: fix uninitialized FIFOs · e4f05af1
      Ondrej Puzman authored 
      I have found a bug in 8250.c driver which causes that 16550A uart FIFOs
are not turned on during initialization if they are manually configured
by setserial. UART is then working only as plain 16450 without FIFOs. On
systems with higher interrupt latency this causes buffer overruns and
loss of received data when using higher communication speeds.

I'm working for a company which produces industrial computers. These
devices typically contain high number (8 or more) of traditional 16550A
uarts - we use TL16C554A chips, but that is not much relevant. UARTs are
connected to the CPU by ISA bus (Celeron based devices) or LPC bus (Atom
based devices).

In the Linux the UARTs are using standard 8250.c driver and are
initialized using setserial command:
setserial /dev/ttyS4 uart 16550A port 0x3E0 irq 10 baud_base 115200

This executes the UART initialization through serial8250_startup()
function. At the beginning of the function up->capabilities is
initialized from uart_config:
 up->capabilities = uart_config[up->port.type].flags;
Please note that neither up->port.fifosize nor up->tx_loadsz is
initialized here!!

Later in the same function serial8250_clear_fifos() is called and
disables FIFOs. The above comment says that they will be reenabled in
set_termios (they won't ...)

After serial8250_startup() the serial8250_set_termios() is called. In
this function the following check fails because up->port.fifosize is
zero because it is not initialized correctly.

        if (up->capabilities & UART_CAP_FIFO && up->port.fifosize > 1) {
                if (baud < 2400)
                        fcr = UART_FCR_ENABLE_FIFO | UART_FCR_TRIGGER_1;
                else
                        fcr = uart_config[up->port.type].fcr;
        }

fcr variable remains zero and in the end the FCR register is set to zero
which results in disabled FIFOs even if the UART type is 16550A. This is
also true for other types of UARTs with FIFOs.

If the UART is autoconfigured via 'setserial /dev/ttySx autoconfig' then
port.fifosize and tx_loadsz are initialized correctly in the
autoconfig() function and the UART is working correctly then.

I checked the source codes and I can say that this bug is present in
2.6.x series of kernels for a couple of years. Namely I can confirm its
presence in 2.6.16.57, 2.6.32.24 and 2.6.36.1 (tested all of them on our
hardware).

I think it was not noticed before because not many people use manually
configured non PNP UARTs on ISA/LPC bus these days. Also the data loss
caused by buffer overruns occures only if  IRQ latency is higher then
time needed to receive one character on given communication speed.
For example our hardware looses received characters only if the UARTs
are connected throught LPC bus with SERIRQ (serial IRQ transport) and
not if they are connected to ISA bus because LPC SERIRQ has higher
interrupt latency then parallel ISA interupt lines.

Here is the patch to correct the bug created against 2.6.36.1:
Signed-off-by: default avatarOndrej Puzman <puzman@gmail.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@suse.de>
      e4f05af1
    • Namhyung Kim's avatar
      ip2: fix compiler warning on ip2main_pci_tbl · 53139e36
      Namhyung Kim authored 
      Annotate ip2main_pci_tbl as '__used' to fix following warning:

  CC      drivers/char/ip2/ip2main.o
drivers/char/ip2/ip2main.c:3227: warning: ‘ip2main_pci_tbl’ defined but not used
Signed-off-by: default avatarNamhyung Kim <namhyung@gmail.com>
Cc: "Michael H. Warfield" <mhw@wittsend.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@suse.de>
      53139e36
    • Namhyung Kim's avatar
      specialix: fix compiler warning on specialix_pci_tbl · fd0f5c54
      Namhyung Kim authored 
      Annotate specialx_pci_tbl as '__used' to fix following warning:

  CC      drivers/char/specialix.o
drivers/char/specialix.c:2358: warning: ‘specialx_pci_tbl’ defined but not used
Signed-off-by: default avatarNamhyung Kim <namhyung@gmail.com>
Cc: Roger Wolff <R.E.Wolff@BitWizard.nl>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@suse.de>
      fd0f5c54
    • Namhyung Kim's avatar
      rocket: fix compiler warning on rocket_pci_ids · 6835a209
      Namhyung Kim authored 
      Annotate rocket_pci_ids as '__used' to fix following warning:

  CC      drivers/char/rocket.o
drivers/char/rocket.c:1767: warning: ‘rocket_pci_ids’ defined but not used
Signed-off-by: default avatarNamhyung Kim <namhyung@gmail.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@suse.de>
      6835a209
    • Jamie Iles's avatar
      8250: add a UPIO_DWAPB32 for 32 bit accesses · a3ae0fc3
      Jamie Iles authored 
      Some platforms contain a Synopsys DesignWare APB UART that is attached
to a 32-bit APB bus where sub-word accesses are not allowed. Add a new
IO type (UPIO_DWAPB32) that performs 32 bit acccesses to the UART.

v2:
	- don't test for 32 bit in the output fast path, provide a
	  separate dwabp32_serial_out() function. Refactor
	  dwabp_serial_out() so that we can reuse the LCR saving
	  code.
v3:
	- rebased on top of "8250: use container_of() instead of
	  casting"
Signed-off-by: default avatarJamie Iles <jamie@jamieiles.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@suse.de>
      a3ae0fc3
    • Jamie Iles's avatar
      8250: use container_of() instead of casting · 49d5741b
      Jamie Iles authored 
      The 8250 driver structure uart_8250_port took advantage of the fact
that the struct uart_port was the first member of its structure and
used an explicit cast to convert to the derived class. Replace the
explicit casts with container_of() for safety and clarity.
Signed-off-by: default avatarJamie Iles <jamie@jamieiles.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@suse.de>
      49d5741b
    • Cosmin Cojocar's avatar
      serial: omap-serial: Add support for kernel debugger · 1b41dbc1
      Cosmin Cojocar authored 
      The kgdb invokes the poll_put_char and poll_get_char when communicating
with the host. This patch also changes the initialization order because the
kgdb will check at the very beginning, if there is a valid serial
driver.
Signed-off-by: default avatarCosmin Cojocar <cosmin.cojocar@gmail.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@suse.de>
      1b41dbc1
    • Randy Dunlap's avatar
      serial: fix pch_uart kconfig & build · 5ac387d9
      Randy Dunlap authored 
      The dma_request_channel/dma_release_channel interfaces are not
built when DMADEVICES is not enabled, so make the driver depend on
DMADEVICES.  Also, the help text says that the driver enables & uses
PCH_DMA, which is not enabled, so select that.

ERROR: "__dma_request_channel" [drivers/serial/pch_uart.ko] undefined!
ERROR: "dma_release_channel" [drivers/serial/pch_uart.ko] undefined!
Signed-off-by: default avatarRandy Dunlap <randy.dunlap@oracle.com>
Cc: Tomoya MORINAGA <tomoya-linux@dsn.okisemi.com>
Reported-by: default avatarZimny Lech <napohybelskurwysynom2010@gmail.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@suse.de>
      5ac387d9
  7. 01 Dec, 2010 6 commits
  9. 16 Nov, 2010 8 commits
  11. 11 Nov, 2010 7 commits
  13. 08 Nov, 2010 5 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7