On Fri, 24 Jun 2011 09:36:56 -0400, Alex Deucher wrote:
On Fri, Jun 24, 2011 at 12:02 AM, Thomas Reim thomas.reim@nepomuc.de wrote:
diff --git a/drivers/gpu/drm/radeon/radeon_i2c.c b/drivers/gpu/drm/radeon/radeon_i2c.c index 781196d..7e93cf9 100644 --- a/drivers/gpu/drm/radeon/radeon_i2c.c +++ b/drivers/gpu/drm/radeon/radeon_i2c.c @@ -34,7 +34,7 @@ */ bool radeon_ddc_probe(struct radeon_connector *radeon_connector) {
- u8 out_buf[] = { 0x0, 0x0};
- u8 out = 0x0;
u8 buf[2]; int ret; struct i2c_msg msgs[] = { @@ -42,7 +42,7 @@ bool radeon_ddc_probe(struct radeon_connector *radeon_connector) .addr = 0x50, .flags = 0, .len = 1,
- .buf = out_buf,
- .buf = &out,
}, { .addr = 0x50,
The change above doesn't seem to be related.
This was a comment from Jean who complained about the ineffective usage of the i2c bus. But I can also restore the old code. What's your preference?
Ah, I missed that. Let's make that a separate patch, or fix it when you add support for the extended edid check.
Thanks for fixing this up.
I'll send a patch for that one, as I found it. It is indeed unrelated to the problem, I mentioned it only to avoid the same mistake in a newly added function.
It's a very minor cleanup / optimization, BTW.