[[PATCH v2]] drm/ast: fixed reading monitor EDID not stable issue
陳雅正
yacheng600221 at gmail.com
Thu Nov 22 01:42:09 UTC 2018
Hi Dave,
Thanks for your feedback. No issue found actually if I remove "volatile" on
my platform. In my experience, if the value is volatile, adding "volatile"
will be safer and no harm, that is why I add it by default. If you think it
is not necessary, it's ok for me to remove it.
Regards,
Y.C. Chen
Dave Airlie <airlied at gmail.com> 於 2018年11月22日 週四 上午9:15寫道:
> On Wed, 31 Oct 2018 at 18:12, Y.C. Chen <yacheng600221 at gmail.com> wrote:
> >
> > From: "Y.C. Chen" <yc_chen at aspeedtech.com>
> >
> > v1: over-sample data to increase the stability with some specific
> monitors
> > v2: refine to avoid infinite loop
>
> This contains at least 4 whitespace breakages (val =) in a few
> places, also why the volatiles, I don't think they make much sense
> here, have you verified they are required?
>
> Dave.
>
> >
> > Signed-off-by: Y.C. Chen <yc_chen at aspeedtech.com>
> > ---
> > drivers/gpu/drm/ast/ast_mode.c | 34 ++++++++++++++++++++++++++++------
> > 1 file changed, 28 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/ast/ast_mode.c
> b/drivers/gpu/drm/ast/ast_mode.c
> > index 5e77d45..092e9a7 100644
> > --- a/drivers/gpu/drm/ast/ast_mode.c
> > +++ b/drivers/gpu/drm/ast/ast_mode.c
> > @@ -972,9 +972,20 @@ static int get_clock(void *i2c_priv)
> > {
> > struct ast_i2c_chan *i2c = i2c_priv;
> > struct ast_private *ast = i2c->dev->dev_private;
> > - uint32_t val;
> > + volatile uint32_t val, val2, count, pass;
> > +
> > + count = 0;
> > + pass = 0;
> > + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7,
> 0x10) >> 4) & 0x01;
> > + do {
> > + val2 = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT,
> 0xb7, 0x10) >> 4) & 0x01;
> > + if (val == val2) pass++;
> > + else {
> > + pass = 0;
> > + val = (ast_get_index_reg_mask(ast,
> AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4) & 0x01;
>
>
>
>
> > + }
> > + } while ((pass < 5) && (count++ < 0x10000));
> >
> > - val = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10)
> >> 4;
> > return val & 1 ? 1 : 0;
> > }
> >
> > @@ -982,9 +993,20 @@ static int get_data(void *i2c_priv)
> > {
> > struct ast_i2c_chan *i2c = i2c_priv;
> > struct ast_private *ast = i2c->dev->dev_private;
> > - uint32_t val;
> > + volatile uint32_t val, val2, count, pass;
> > +
> > + count = 0;
> > + pass = 0;
> > + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7,
> 0x20) >> 5) & 0x01;
> > + do {
> > + val2 = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT,
> 0xb7, 0x20) >> 5) & 0x01;
> > + if (val == val2) pass++;
> > + else {
> > + pass = 0;
> > + val = (ast_get_index_reg_mask(ast,
> AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5) & 0x01;
> > + }
> > + } while ((pass < 5) && (count++ < 0x10000));
> >
> > - val = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20)
> >> 5;
> > return val & 1 ? 1 : 0;
> > }
> >
> > @@ -997,7 +1019,7 @@ static void set_clock(void *i2c_priv, int clock)
> >
> > for (i = 0; i < 0x10000; i++) {
> > ujcrb7 = ((clock & 0x01) ? 0 : 1);
> > - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7,
> 0xfe, ujcrb7);
> > + ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7,
> 0xf4, ujcrb7);
> > jtemp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT,
> 0xb7, 0x01);
> > if (ujcrb7 == jtemp)
> > break;
> > @@ -1013,7 +1035,7 @@ static void set_data(void *i2c_priv, int data)
> >
> > for (i = 0; i < 0x10000; i++) {
> > ujcrb7 = ((data & 0x01) ? 0 : 1) << 2;
> > - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7,
> 0xfb, ujcrb7);
> > + ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7,
> 0xf1, ujcrb7);
> > jtemp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT,
> 0xb7, 0x04);
> > if (ujcrb7 == jtemp)
> > break;
> > --
> > 1.8.3.1
> >
> > _______________________________________________
> > dri-devel mailing list
> > dri-devel at lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/dri-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20181122/bb6de055/attachment-0001.html>
More information about the dri-devel
mailing list