Invoke cn_netlink_send() in a work context as opposed being called
in the context of channel callback. On entry into the channel callback
code the channel inbound spin lock is held and deferring to a work
context avoids having to invoke cn_netlink_send() while holding
the inbound lock. As part of
From: wwang
Fix a bug that SDIO and SD normal card would appear simultaneously if a SDIO
card inserted.
Signed-off-by: wwang
---
drivers/staging/rts_pstor/sd.c | 18 +-
1 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/drivers/staging/rts_pstor/sd.c b/drivers/st
Dear Carpenter:
I will resend the patch
Best Regards,
wwang
于 2011年10月09日 21:57, Dan Carpenter 写道:
> What I'm saying is, it's confusing to add another variable which is
> only subtly different from retval. Here is what it looks like after
> we apply the patch.
>
> int reset_pass = 0;
>
>
Hi,
On Sun, 9 Oct 2011, Dan Carpenter wrote:
On Sun, Oct 09, 2011 at 11:22:07PM +0300, Aaro Koskinen wrote:
On Sun, 9 Oct 2011, Dan Carpenter wrote:
On Sun, Oct 09, 2011 at 07:42:51PM +0300, Aaro Koskinen wrote:
- switch (xgi_video_info.disp_state & DISPTYPE_DISP2) {
+ switch (xgi
Reviewed-by: Dan Carpenter
regards,
dan carpenter
___
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
On Sun, Oct 09, 2011 at 11:22:07PM +0300, Aaro Koskinen wrote:
> Hi,
>
> On Sun, 9 Oct 2011, Dan Carpenter wrote:
> >On Sun, Oct 09, 2011 at 07:42:51PM +0300, Aaro Koskinen wrote:
> >>- switch (xgi_video_info.disp_state & DISPTYPE_DISP2) {
> >>+ switch (xgifb_info->disp_state & DISPTYPE_DISP2)
Hi,
On Sun, 9 Oct 2011, Dan Carpenter wrote:
On Sun, Oct 09, 2011 at 07:42:51PM +0300, Aaro Koskinen wrote:
- switch (xgi_video_info.disp_state & DISPTYPE_DISP2) {
+ switch (xgifb_info->disp_state & DISPTYPE_DISP2) {
This isn't related to this patch, but this data type is weird.
Hi Arnaud,
Your mandriva email address in MAINTAINERS has been dead for a long
time. Do you want us to update it to a different address?
regards,
dan carpenter
___
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.or
On Sun, Oct 09, 2011 at 07:43:03PM +0300, Aaro Koskinen wrote:
> + xgifb_reg_set(P3c4,
> + 0x18,
> + pVBInfo->SR15[2][pVBInfo->ram_type]); /* SR18 */
> xgifb_reg_set(P3c4, 0x19, 0x01);
> xgifb_reg_set(P3c4, 0x16, pVBInfo->SR16[0]);
> xgifb_r
On Sun, Oct 09, 2011 at 07:43:02PM +0300, Aaro Koskinen wrote:
> @@ -2292,7 +2294,9 @@ static int __devinit xgifb_probe(struct pci_dev *pdev,
> XGIbios_mode[xgifb_info->mode_idx].xres) &&
> (XGI21_LCDCapList[m].LVDSVDE ==
>
On Sun, Oct 09, 2011 at 07:42:51PM +0300, Aaro Koskinen wrote:
> @@ -1163,17 +1166,17 @@ static int XGIfb_do_set_var(struct fb_var_screeninfo
> *var, int isactive,
> if (var->pixclock && htotal && vtotal) {
> drate = 10 / var->pixclock;
> hrate = (drate *
On Sun, Oct 09, 2011 at 07:42:51PM +0300, Aaro Koskinen wrote:
> - switch (xgi_video_info.disp_state & DISPTYPE_DISP2) {
> + switch (xgifb_info->disp_state & DISPTYPE_DISP2) {
This isn't related to this patch, but this data type is weird.
It would be probably be more natural to break up -
Move vb_device_info into xgifb_video_info.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h| 21 +
drivers/staging/xgifb/XGI_main_26.c | 15 ++-
drivers/staging/xgifb/XGIfb.h |2 ++
3 files changed, 21 insertions(+), 17 deletions
DAC tables are read-only and can be made const.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/vb_setmode.c | 11 ++-
1 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/staging/xgifb/vb_setmode.c
b/drivers/staging/xgifb/vb_setmode.c
index 8f3a812..81c0cc4 1
RAM type tables are read-only and can be made const.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/vb_init.c | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/staging/xgifb/vb_init.c b/drivers/staging/xgifb/vb_init.c
index ce472dc..9e890a1 100644
Eliminate default_var and initialize needed fields of fb_info->var
in probe().
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h| 29 --
drivers/staging/xgifb/XGI_main_26.c | 56 ---
2 files changed, 32 insertions(+), 53 del
Comments are wrong, the table is read-only and can be made const.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h |5 ++---
1 files changed, 2 insertions(+), 3 deletions(-)
diff --git a/drivers/staging/xgifb/XGI_main.h b/drivers/staging/xgifb/XGI_main.h
index 78f4c93..6c1f
filter_tb is only used inside a single function, and it does not need
to be static.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h|1 -
drivers/staging/xgifb/XGI_main_26.c |3 +++
2 files changed, 3 insertions(+), 1 deletions(-)
diff --git a/drivers/staging/xgifb/
Move the memory channel info into vb_device_info.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/vb_init.c | 52 ++--
drivers/staging/xgifb/vb_struct.h |1 +
2 files changed, 27 insertions(+), 26 deletions(-)
diff --git a/drivers/staging/xgifb/vb_
The current refresh rate index should be stored in the device-specific
data allocated in probe().
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h|3 ---
drivers/staging/xgifb/XGI_main_26.c | 30 ++
drivers/staging/xgifb/XGIfb.h |1
Replace XGIfb_CRT2_write_enable with IND_XGI_CRT2_WRITE_ENABLE_315.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h|3 ---
drivers/staging/xgifb/XGI_main_26.c | 11 ---
2 files changed, 4 insertions(+), 10 deletions(-)
diff --git a/drivers/staging/xgifb/XGI_m
pseudo_palette should be dynamically allocated for each fb.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h|2 --
drivers/staging/xgifb/XGI_main_26.c |2 +-
drivers/staging/xgifb/XGIfb.h |2 ++
3 files changed, 3 insertions(+), 3 deletions(-)
diff --git a
The RAM type is device specific, so move it into vb_device_info.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/vb_init.c | 128
drivers/staging/xgifb/vb_struct.h |2 +
2 files changed, 73 insertions(+), 57 deletions(-)
diff --git a/drivers/st
The current mode index should be stored in the device-specific data
allocated in probe().
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h|1 -
drivers/staging/xgifb/XGI_main_26.c | 101 +++
drivers/staging/xgifb/XGIfb.h |2 +
3
Avoid direct references to global xgi_video_info. This will help changing
the data from statically allocated to dynamically allocated.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main_26.c | 512 ++-
1 files changed, 263 insertions(+), 249 deletion
Move the memory bus width info to vb_device_info.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/vb_init.c | 32 +++-
drivers/staging/xgifb/vb_struct.h |1 +
2 files changed, 16 insertions(+), 17 deletions(-)
diff --git a/drivers/staging/xgifb/vb_init
Eliminate XGIfb_fix and initialize needed fields of fb_info->fix
in probe().
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h|7 ---
drivers/staging/xgifb/XGI_main_26.c |6 +-
2 files changed, 5 insertions(+), 8 deletions(-)
diff --git a/drivers/staging/xgif
We can get the information from the table with mode index.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h|1 -
drivers/staging/xgifb/XGI_main_26.c | 14 ++
2 files changed, 6 insertions(+), 9 deletions(-)
diff --git a/drivers/staging/xgifb/XGI_main.h b/d
Delete global xgi_video_info and dynamically allocate the data.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main_26.c | 15 ++-
drivers/staging/xgifb/XGIfb.h |3 ---
2 files changed, 6 insertions(+), 12 deletions(-)
diff --git a/drivers/staging/xgifb/XGI_m
Rename struct video_info to xgifb_video_info.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main_26.c |4 ++--
drivers/staging/xgifb/XGIfb.h |4 ++--
drivers/staging/xgifb/vb_init.c |2 +-
3 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/s
Add a global variable for the default refresh rate. This is done to
get rid of references to xgi_video_info before the probe routine, which
should allocate the xgi_video_info dynamically in the future.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main_26.c |6 --
1 files ch
Don't access the global xgi_video_info directly in xgifb_remove().
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main_26.c | 19 +++
1 files changed, 11 insertions(+), 8 deletions(-)
diff --git a/drivers/staging/xgifb/XGI_main_26.c
b/drivers/staging/xgifb/XGI_mai
Move xgi_hw_device_info into xgi_video_info.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h|3 -
drivers/staging/xgifb/XGI_main_26.c | 94 ++-
drivers/staging/xgifb/XGIfb.h |3 +
3 files changed, 51 insertions(+), 49 deletion
XGIfb_query_VGA_config_space() is used only once during the init and
can be replaced with a single PCI configuration space read.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main_26.c | 44 +-
drivers/staging/xgifb/vb_init.c | 15 ++-
Hi,
These are the start of cleanups for the global variable and static data
usage. The driver should allocate all device-specific data in probe and
release it when the module is removed. Static data is used only for the
module and fb parameters.
The patches are based and tested on top of the curr
Move fb_info into xgi_video_info.
Signed-off-by: Aaro Koskinen
---
drivers/staging/xgifb/XGI_main.h|3 ---
drivers/staging/xgifb/XGI_main_26.c |3 +++
drivers/staging/xgifb/XGIfb.h |2 ++
3 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/staging/xgifb/X
Hi Mikael,
thanks for review. The nvec driver received some overhaul in the 3.1 release
cycle
and he fruids can be viewed in linux-next tree. The nvec client drivers were
converted to mfd children in the mean time. This solves the build error, but is
not
very flexible. We are planing to move
What I'm saying is, it's confusing to add another variable which is
only subtly different from retval. Here is what it looks like after
we apply the patch.
int reset_pass = 0;
/* skip 45 lines */
retval = sd_change_bank_voltage(chip, SD_IO_3V3);
if (retval != STA
Dear Carpenter:
Yes, the new code is exactly what I intended.
As to the following code, the old system responded fully wrong. It
should return STATUS_FAIL here instead.
3144 if (CHECK_PID(chip, 0x5209)) {
3145 retval = sd_change_bank_voltage(chip, SD_IO_3V3);
3146 if (retval != STATUS_SUCCESS) {
Adding the check for if (sd_check_err_code(chip, SD_NO_CARD)) is
good, but introducing the new "reset_pass" variable is wrong.
I don't think you are updating the new variable consistently on
all paths. For example, if it's a 0x5209 chip and the code is like
this:
retval = sd_ch
40 matches
Mail list logo
