From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
To: "Harry Wentland" <harry.wentland@amd.com>,
"Alex Deucher" <alexander.deucher@amd.com>,
"Christian König" <christian.koenig@amd.com>,
"Daniel Vetter" <daniel@ffwll.ch>,
"Peter Zijlstra" <peterz@infradead.org>,
aric.cyr@amd.com
Cc: dri-devel@lists.freedesktop.org, amd-gfx@lists.freedesktop.org
Subject: [PATCH 3/4] drm/amd/display: Add ref count control for FPU utilization
Date: Wed, 31 Mar 2021 08:25:01 -0400 [thread overview]
Message-ID: <20210331122502.1031073-4-Rodrigo.Siqueira@amd.com> (raw)
In-Reply-To: <20210331122502.1031073-1-Rodrigo.Siqueira@amd.com>
DC invokes DC_FPU_START/END in multiple parts of the code; this can
create a situation where we invoke this FPU operation in a nested way.
For avoiding this situation, this commit adds a mechanism where
dc_fpu_begin/end manages the access to kernel_fpu_begin/end.
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
.../amd/display/amdgpu_dm/amdgpu_dm_trace.h | 13 ++++---
.../gpu/drm/amd/display/amdgpu_dm/dc_fpu.c | 34 ++++++++++++++++---
drivers/gpu/drm/amd/display/dc/dc_trace.h | 4 +--
3 files changed, 40 insertions(+), 11 deletions(-)
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h
index 230bb12c405e..cd4f0d3f37fb 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h
@@ -638,23 +638,26 @@ TRACE_EVENT(amdgpu_refresh_rate_track,
);
TRACE_EVENT(dcn_fpu,
- TP_PROTO(bool begin, const char *function, const int line),
- TP_ARGS(begin, function, line),
+ TP_PROTO(bool begin, const char *function, const int line, const int ref_count),
+ TP_ARGS(begin, function, line, ref_count),
TP_STRUCT__entry(
__field(bool, begin)
__field(const char *, function)
__field(int, line)
+ __field(int, ref_count)
),
TP_fast_assign(
__entry->begin = begin;
__entry->function = function;
__entry->line = line;
+ __entry->ref_count = ref_count;
),
- TP_printk("%s()+%d: %s",
+ TP_printk("%s: ref_count: %d: %s()+%d:",
+ __entry->begin ? "begin" : "end",
+ __entry->ref_count,
__entry->function,
- __entry->line,
- __entry->begin ? "begin" : "end"
+ __entry->line
)
);
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c
index ff34007509de..5dcefe193523 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c
@@ -28,6 +28,19 @@
#include <asm/fpu/api.h>
+/**
+ * DOC: Overview
+ *
+ * DC core uses FPU operations in multiple parts of the code, which requires a
+ * more specialized way to manage these areas' entrance. To fulfill this
+ * requirement, we created some wrapper functions that encapsulate
+ * kernel_fpu_begin/end to better fit our need in the display component. In
+ * summary, in this file, you can find functions related to FPU operation
+ * management.
+ */
+
+static DEFINE_PER_CPU(atomic_t, fpu_ref);
+
/**
* dc_fpu_begin - Enables FPU protection
* @function_name: A string containing the function name for debug purposes
@@ -40,8 +53,14 @@
*/
void dc_fpu_begin(const char *function_name, const int line)
{
- TRACE_DCN_FPU(true, function_name, line);
- kernel_fpu_begin();
+ int ret;
+ atomic_t *local_fpu_ref = this_cpu_ptr(&fpu_ref);
+
+ ret = atomic_inc_return(local_fpu_ref);
+ TRACE_DCN_FPU(true, function_name, line, ret);
+
+ if (ret == 1)
+ kernel_fpu_begin();
}
/**
@@ -56,6 +75,13 @@ void dc_fpu_begin(const char *function_name, const int line)
*/
void dc_fpu_end(const char *function_name, const int line)
{
- TRACE_DCN_FPU(false, function_name, line);
- kernel_fpu_end();
+
+ int ret;
+ atomic_t *local_fpu_ref = this_cpu_ptr(&fpu_ref);
+
+ ret = atomic_dec_return(local_fpu_ref);
+ TRACE_DCN_FPU(false, function_name, line, ret);
+
+ if (!ret)
+ kernel_fpu_end();
}
diff --git a/drivers/gpu/drm/amd/display/dc/dc_trace.h b/drivers/gpu/drm/amd/display/dc/dc_trace.h
index d598ba697e45..c711797e5c9e 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_trace.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_trace.h
@@ -38,5 +38,5 @@
#define TRACE_DCN_CLOCK_STATE(dcn_clocks) \
trace_amdgpu_dm_dc_clocks_state(dcn_clocks)
-#define TRACE_DCN_FPU(begin, function, line) \
- trace_dcn_fpu(begin, function, line)
+#define TRACE_DCN_FPU(begin, function, line, ref_count) \
+ trace_dcn_fpu(begin, function, line, ref_count)
--
2.25.1
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2021-03-31 12:25 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-31 12:24 [PATCH 0/4] drm/amd/display: Base changes for isolating FPU operation in a single place Rodrigo Siqueira
2021-03-31 12:24 ` [PATCH 1/4] drm/amd/display: Introduce FPU directory inside DC Rodrigo Siqueira
2021-03-31 17:25 ` kernel test robot
2021-03-31 12:25 ` [PATCH 2/4] drm/amd/display: Add FPU event trace Rodrigo Siqueira
2021-03-31 19:28 ` kernel test robot
2021-03-31 20:26 ` kernel test robot
2021-03-31 12:25 ` Rodrigo Siqueira [this message]
2021-03-31 12:44 ` [PATCH 3/4] drm/amd/display: Add ref count control for FPU utilization Christian König
2021-03-31 12:25 ` [PATCH 4/4] drm/amd/display: Add DC_FP helper to check FPU state Rodrigo Siqueira
2021-03-31 12:47 ` Christian König
2021-03-31 12:49 ` [PATCH 0/4] drm/amd/display: Base changes for isolating FPU operation in a single place Christian König
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210331122502.1031073-4-Rodrigo.Siqueira@amd.com \
--to=rodrigo.siqueira@amd.com \
--cc=alexander.deucher@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=aric.cyr@amd.com \
--cc=christian.koenig@amd.com \
--cc=daniel@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=harry.wentland@amd.com \
--cc=peterz@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox