On Mo, 2013-10-07 at 17:12 +, KY Srinivasan wrote:
>
> > -Original Message-
> > From: Gerd Hoffmann [mailto:kra...@redhat.com]
> > Sent: Sunday, October 06, 2013 11:51 PM
> > To: KY Srinivasan
> > Cc: Haiyang Zhang; Jean-Christophe Plagniol-Villard; Tomi Valkeinen; open
> > list:Hyper-
On Tue, Oct 08, 2013 at 10:43:07AM +1100, Dave Chinner wrote:
> > Maybe, writeboost should disable deferring barriers
> > if barrier_deadline_ms parameter is especially 0.
> > Linux kernel's layered architecture is obviously not always perfect
> > so there are similar cases in other boundaries
> >
This patch adds a MMU dependency to configure the ZSMALLOC in
drivers/staging/zsmalloc/Kconfig. Without this patch, build
system can lead to build failure. This was observed during
randconfig testing, in which ZSMALLOC was enabled w/o MMU being
enabled. Following was the error:
LD vmlinux
dri
Christoph,
> You can detect O_DIRECT writes by second guession a special combination
> of REQ_ flags only used there, as cfg tries to treat it special:
>
> #define WRITE_SYNC (WRITE | REQ_SYNC | REQ_NOIDLE)
> #define WRITE_ODIRECT (WRITE | REQ_SYNC)
>
> the lack of REQ_NOI
Dave,
> i.e. there's no point justifying a behaviour with "we could do this
> in future so lets ignore the impact on current users"...
Sure, I am happy if we find a solution that
is good for both of us or filesystem and block in other word.
> e.g. what happens if a user has a mixed workload - one
> -Original Message-
> From: Gerd Hoffmann [mailto:kra...@redhat.com]
> Sent: Tuesday, October 08, 2013 1:41 AM
> To: KY Srinivasan
> Cc: Haiyang Zhang; Jean-Christophe Plagniol-Villard; Tomi Valkeinen; open
> list:Hyper-V CORE AND...; open list:FRAMEBUFFER LAYER; open list
> Subject: Re:
Mikulas,
Let me ask you about this comment
of choosing the best API.
For the rest, I will reply later.
> BTW. You should use wait_event_interruptible_lock_irq instead of
> wait_event_interruptible and wait_event_interruptible_lock_irq_timeout
> instead of wait_event_interruptible_timeout. The f
On Tue, Oct 08 2013 at 6:37am -0400,
Akira Hayakawa wrote:
> Christoph,
>
> > You can detect O_DIRECT writes by second guession a special combination
> > of REQ_ flags only used there, as cfg tries to treat it special:
> >
> > #define WRITE_SYNC (WRITE | REQ_SYNC | REQ_NOIDLE)
> >
The `set_mode_a()` and `set_mode_b()` functions use a local variable
`setup` which is initialized to the value of parameter `Setup` which is
not used further. Get rid of the local variable and rename the
parameter to `setup`.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c |
The `DMAHandle` member of `struct bufferDMA` in "s626.h" is unused, so
remove it.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/staging/comedi/drivers/s626.h
b/drivers/staging/comedi/drivers/s626.h
index c4b8078..f
Rename the CamelCase members of `struct s626_private`.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 99 ++-
1 file changed, 50 insertions(+), 49 deletions(-)
diff --git a/drivers/staging/comedi/drivers/s626.c
b/drivers/staging/comedi/dri
The `I2C_B0(ATTR, VAL)`, `I2C_B1(ATTR, VAL)` and `I2C_B2(ATTR, VAL)`
macros are defined identically in "s626.h" and "s626.c". Remove the
duplicates from "s626.c".
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/drivers/
Rename `struct bufferDMA` and its members to avoid CamelCase.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 46 +--
drivers/staging/comedi/drivers/s626.h | 6 ++---
2 files changed, 26 insertions(+), 26 deletions(-)
diff --git a/drivers/s
Tidy up the "s626.h" header file a bit, cleaning up the whitespace and
fixing overlength lines. Add multiple inclusion protection (even though
it's only included once and only by "s626.c"). Remove unnecessary
driver comment near the start of the file as it is more or less the same
as the one in "
The Sensoray 626 is based on the Philips SAA7146 chip using the Philips
vendor and device ID for the chip, but with custom subdevice and
subvendor IDs. Use the Philips IDs in the PCI device table and replace
the macros for the subvendor and subdevice IDs with open-coded values.
Signed-off-by: Ian
Tidy up the s626 comedi driver a bit.
01) staging: comedi: s626: tidy the header file a bit
02) staging: comedi: s626: tidy up comments at top of file
03) staging: comedi: s626: tidy up main source code file
04) staging: comedi: s626: convert a printk()
05) staging: comedi: s626: remove unused DMA
Move some functions to help avoid the forward declaration of
`s626_enc_chan_info[]`.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 284 +-
1 file changed, 142 insertions(+), 142 deletions(-)
diff --git a/drivers/staging/comedi/drivers/s626
Rename the CamelCase members of `struct enc_private`. Since most of
those are function pointers pointing to functions with similar names as
the members, rename the functions they point to as well.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 425 +---
The `encpriv` macro relies on a local variable `dev` (of type `struct
comedi_device *`) being set correctly. By a convoluted path involving
the `private` data pointer of subdevice 5 (the encoder (counter)
subdevice), the macro always yields a pointer to the first element of
the static array `enc_p
Move some more functions to help avoid the forward declaration of
`s626_enc_chan_info[]`.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 478 +-
1 file changed, 239 insertions(+), 239 deletions(-)
diff --git a/drivers/staging/comedi/drivers
Move the declaration of `struct buffer_dma` from "s626.h" to "s626.c" as
it seems more at home there. After this move, "s626.h" just contains
macros related to hardware registers.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 5 +
drivers/staging/comedi/drivers/s626.
`s626_mc_enable()` is often called to enable some function in a control
register after setting up some other registers. Precede the write to
the control register with a call to `mmiowb()` to preserve mmio write
ordering.
`s626_mc_disable()` is called to disable some function in a control
register
Move `s626_enc_chan_info[]` and remove its forward declaration.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 190 +-
1 file changed, 94 insertions(+), 96 deletions(-)
diff --git a/drivers/staging/comedi/drivers/s626.c
b/drivers/staging/c
Prefix the macros defined in "s626.c" with `S626_`. Macro `VECT0` is
unused, so remove it.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 75 +--
1 file changed, 37 insertions(+), 38 deletions(-)
diff --git a/drivers/staging/comedi/drivers
Add the prefix `s626_` to all the static variables and functions that
don't already have it.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 551 +-
1 file changed, 281 insertions(+), 270 deletions(-)
diff --git a/drivers/staging/comedi/driv
"s626.h" defines the macros `TRUE` and `FALSE` if they are not already
defined, yielding the expected numeric values. Remove the macros and
replace their usage with the values `true` and `false`, respectively.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 18 +---
Re-tag `struct buffer_dma` to `struct s626_buffer_dma` to avoid
potential namespace clashes.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/staging/comedi/drivers/s626.c
b/drivers/staging/co
Declare the static variables `trimchan[]` and `trimadrs[]` as constant
data.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/staging/comedi/drivers/s626.c
b/drivers/staging/comedi/drivers/s626.c
Rename the remaining CamelCase parameters and variables.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 258 +-
1 file changed, 129 insertions(+), 129 deletions(-)
diff --git a/drivers/staging/comedi/drivers/s626.c
b/drivers/staging/comedi
Move the final lot of encoder functions to help avoid the forward
declaration of `s626_enc_chan_info[]`.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 334 +-
1 file changed, 167 insertions(+), 167 deletions(-)
diff --git a/drivers/staging
Rename the remaining non-lower-case functions.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 293 +-
1 file changed, 147 insertions(+), 146 deletions(-)
diff --git a/drivers/staging/comedi/drivers/s626.c
b/drivers/staging/comedi/drivers/s
Convert a printk() to a dev_err().
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/staging/comedi/drivers/s626.c
b/drivers/staging/comedi/drivers/s626.c
index 99d1bed..e59769d 100644
--- a/driver
Rename the `ai_continous` member of `struct s626_private` to
`ai_continuous`.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/staging/comedi/drivers/s626.c
b/drivers/staging/comedi/drivers/s6
Make the comments at the top of the file conform to the CodingStyle.
Signed-off-by: Ian Abbott
---
drivers/staging/comedi/drivers/s626.c | 114 +-
1 file changed, 57 insertions(+), 57 deletions(-)
diff --git a/drivers/staging/comedi/drivers/s626.c
b/drivers/stag
The `I2CR` and `I2CW` macros expand to the I2C read and write addresses,
respectively. They are only used in one place each and include the name
of a local variable `devpriv` in their expansion.
Get rid of the macros and expand them in place.
Signed-off-by: Ian Abbott
---
drivers/staging/comed
As far as alloc_mux_rx() is called from probe() only
there is no need in GFP_ATOMIC here.
Found by Linux Driver Verification project (linuxtesting.org).
Signed-off-by: Alexey Khoroshilov
---
drivers/staging/gdm724x/gdm_mux.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --gi
init_usb() may fail after some of mux_rxes already allocated.
So we need to release them on the failure path.
Found by Linux Driver Verification project (linuxtesting.org).
Signed-off-by: Alexey Khoroshilov
---
drivers/staging/gdm724x/gdm_mux.c | 4 ++--
1 file changed, 2 insertions(+), 2 delet
Mike,
I am happy to see that
guys from filesystem to the block subsystem
have been discussing how to handle barriers in each layer
almost independently.
>> Merging the barriers and replacing it with a single FLUSH
>> by accepting a lot of writes
>> is the reason for deferring barriers in writeboo
Mikulas,
> Waking up every 100ms in flush_proc is not good because it wastes CPU time
> and energy if the driver is idle.
Yes, 100ms is too short. I will change it to 1sec then.
We can wait for 1 sec in termination.
> The problem is that if you fill up the whole cache device in less time
> than
39 matches
Mail list logo
