When patching gathers, we don't need to check against
gathers with lower indices than the current one, as
they are guaranteed to already have been handled.
Signed-off-by: Erik Faye-Lund <[email protected]>
---
Here's a trivial optimization I have been running with for a while.
drivers/gpu/host1x/job.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c
index de5ec33..e965805 100644
--- a/drivers/gpu/host1x/job.c
+++ b/drivers/gpu/host1x/job.c
@@ -534,7 +534,7 @@ int host1x_job_pin(struct host1x_job *job, struct device *dev)
g->base = job->gather_addr_phys[i];
- for (j = 0; j < job->num_gathers; j++)
+ for (j = i + 1; j < job->num_gathers; j++)
if (job->gathers[j].bo == g->bo)
job->gathers[j].handled = true;
--
1.8.1.2
Ping?
On Tue, Jan 7, 2014 at 9:03 PM, Erik Faye-Lund <[email protected]> wrote:
> When patching gathers, we don't need to check against
> gathers with lower indices than the current one, as
> they are guaranteed to already have been handled.
>
> Signed-off-by: Erik Faye-Lund <[email protected]>
> ---
>
> Here's a trivial optimization I have been running with for a while.
>
> drivers/gpu/host1x/job.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c
> index de5ec33..e965805 100644
> --- a/drivers/gpu/host1x/job.c
> +++ b/drivers/gpu/host1x/job.c
> @@ -534,7 +534,7 @@ int host1x_job_pin(struct host1x_job *job, struct device *dev)
>
> g->base = job->gather_addr_phys[i];
>
> - for (j = 0; j < job->num_gathers; j++)
> + for (j = i + 1; j < job->num_gathers; j++)
> if (job->gathers[j].bo == g->bo)
> job->gathers[j].handled = true;
>
> --
> 1.8.1.2
>
On 07.01.2014 22:03, Erik Faye-Lund wrote:
> When patching gathers, we don't need to check against
> gathers with lower indices than the current one, as
> they are guaranteed to already have been handled.
>
> Signed-off-by: Erik Faye-Lund <[email protected]>
> ---
>
> Here's a trivial optimization I have been running with for a while.
>
> drivers/gpu/host1x/job.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c
> index de5ec33..e965805 100644
> --- a/drivers/gpu/host1x/job.c
> +++ b/drivers/gpu/host1x/job.c
> @@ -534,7 +534,7 @@ int host1x_job_pin(struct host1x_job *job, struct device *dev)
>
> g->base = job->gather_addr_phys[i];
>
> - for (j = 0; j < job->num_gathers; j++)
> + for (j = i + 1; j < job->num_gathers; j++)
> if (job->gathers[j].bo == g->bo)
> job->gathers[j].handled = true;
>
Hi,
Thanks. This looks good logically, and I ran some tests that agree.
Acked-By: Terje Bergstrom <[email protected]>
Terje
On Tue, Jan 07, 2014 at 09:03:06PM +0100, Erik Faye-Lund wrote:
> When patching gathers, we don't need to check against
> gathers with lower indices than the current one, as
> they are guaranteed to already have been handled.
>
> Signed-off-by: Erik Faye-Lund <[email protected]>
> ---
>
> Here's a trivial optimization I have been running with for a while.
>
> drivers/gpu/host1x/job.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
Applied, thanks!
Thierry
On Tue, Jan 07, 2014 at 09:03:06PM +0100, Erik Faye-Lund wrote:
> When patching gathers, we don't need to check against
> gathers with lower indices than the current one, as
> they are guaranteed to already have been handled.
>
> Signed-off-by: Erik Faye-Lund <[email protected]>
> ---
>
> Here's a trivial optimization I have been running with for a while.
>
> drivers/gpu/host1x/job.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
Applied, thanks!
Thierry