Skip to contentSkip to conversation

riscv: Cache Population Code Refactoring for ACPI and DT Support

RFC In review Sia Jee Heng 3 participants 8 messages Jan 29, 2024 – Jan 30, 2024 RISC-V

lore
Share & tools
lore
#0 lore raw

cover riscv: Cache Population Code Refactoring for ACPI and DT Support

This series of patches refactors the cache population function to seamlessly accommodate both DT and ACPI-based platforms. Additionally, to streamline the code, the unused parameter in the ci_leaf_init() function has been removed.

Sia Jee Heng (2):
riscv: cacheinfo: Remove unused parameter
riscv: cacheinfo: Refactor populate_cache_leaves()

 arch/riscv/kernel/cacheinfo.c | 50 ++++++++++++++---------------------
 1 file changed, 20 insertions(+), 30 deletions(-)

base-commit: 41bccc98fb7931d63d03f326a746ac4d429c1dd3

Signature · 1 line
2.34.1
#1 lore raw

RFC 1/2 riscv: cacheinfo: Remove unused parameter

Removing the unused parameter in ci_leaf_init() to simplify the code.

---
 arch/riscv/kernel/cacheinfo.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)
arch/riscv/kernel/cacheinfo.c +6 -7
index 09e9b88110d1..30a6878287ad 100644
@@ -64,7 +64,6 @@ uintptr_t get_cache_geometry(u32 level, enum cache_type type)
 }
 
 static void ci_leaf_init(struct cacheinfo *this_leaf,
-			 struct device_node *node,
 			 enum cache_type type, unsigned int level)
 {
 	this_leaf->level = level;
@@ -80,11 +79,11 @@ int populate_cache_leaves(unsigned int cpu)
 	int levels = 1, level = 1;
 
 	if (of_property_read_bool(np, "cache-size"))
-		ci_leaf_init(this_leaf++, np, CACHE_TYPE_UNIFIED, level);
+		ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
 	if (of_property_read_bool(np, "i-cache-size"))
-		ci_leaf_init(this_leaf++, np, CACHE_TYPE_INST, level);
+		ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
 	if (of_property_read_bool(np, "d-cache-size"))
-		ci_leaf_init(this_leaf++, np, CACHE_TYPE_DATA, level);
+		ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level);
 
 	prev = np;
 	while ((np = of_find_next_cache_node(np))) {
@@ -97,11 +96,11 @@ int populate_cache_leaves(unsigned int cpu)
 		if (level <= levels)
 			break;
 		if (of_property_read_bool(np, "cache-size"))
-			ci_leaf_init(this_leaf++, np, CACHE_TYPE_UNIFIED, level);
+			ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
 		if (of_property_read_bool(np, "i-cache-size"))
-			ci_leaf_init(this_leaf++, np, CACHE_TYPE_INST, level);
+			ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
 		if (of_property_read_bool(np, "d-cache-size"))
-			ci_leaf_init(this_leaf++, np, CACHE_TYPE_DATA, level);
+			ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level);
 		levels = level;
 	}
 	of_node_put(np);
Signature · 1 line
2.34.1
#2 lore raw

RFC 2/2 riscv: cacheinfo: Refactor populate_cache_leaves()

In reply to Sia Jee Heng

Refactoring the cache population function to support both DT and ACPI-based platforms.

---
 arch/riscv/kernel/cacheinfo.c | 47 ++++++++++++++---------------------
 1 file changed, 19 insertions(+), 28 deletions(-)
arch/riscv/kernel/cacheinfo.c +19 -28
index 30a6878287ad..f10e26fb75b6 100644
@@ -74,36 +74,27 @@ int populate_cache_leaves(unsigned int cpu)
 {
 	struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
 	struct cacheinfo *this_leaf = this_cpu_ci->info_list;
-	struct device_node *np = of_cpu_device_node_get(cpu);
-	struct device_node *prev = NULL;
-	int levels = 1, level = 1;
-
-	if (of_property_read_bool(np, "cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
-	if (of_property_read_bool(np, "i-cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
-	if (of_property_read_bool(np, "d-cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level);
-
-	prev = np;
-	while ((np = of_find_next_cache_node(np))) {
-		of_node_put(prev);
-		prev = np;
-		if (!of_device_is_compatible(np, "cache"))
-			break;
-		if (of_property_read_u32(np, "cache-level", &level))
-			break;
-		if (level <= levels)
-			break;
-		if (of_property_read_bool(np, "cache-size"))
-			ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
-		if (of_property_read_bool(np, "i-cache-size"))
-			ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
-		if (of_property_read_bool(np, "d-cache-size"))
+	unsigned int level, idx;
+
+	for (idx = 0, level = 1; level <= this_cpu_ci->num_levels &&
+	     idx < this_cpu_ci->num_leaves; idx++, level++) {
+		/*
+		 * Since the RISC-V architecture doesn't provide any register for detecting the
+		 * Cache Level and Cache type, this assumes that:
+		 * - There cannot be any split caches (data/instruction) above a unified cache.
+		 * - Data/instruction caches come in pairs.
+		 * - Significant work is required elsewhere to fully support data/instruction-only
+		 *   type caches.
+		 * - The above assumptions are based on conventional system design and known
+		 *   examples.
+		 */
+		if (level == 1) {
 			ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level);
-		levels = level;
+			ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
+		} else {
+			ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
+		}
 	}
-	of_node_put(np);
 
 	return 0;
 }
Signature · 1 line
2.34.1
#3 lore raw

Re: [RFC v1 2/2] riscv: cacheinfo: Refactor populate_cache_leaves()

Hey,

Firstly, the $subject should really mention that the motivation for the refactoring is ACPI support.

Quoted 45 lines · Sia Jee Heng

Refactoring the cache population function to support both DT and ACPI-based platforms.

---
 arch/riscv/kernel/cacheinfo.c | 47 ++++++++++++++---------------------
 1 file changed, 19 insertions(+), 28 deletions(-)
arch/riscv/kernel/cacheinfo.c +3 -26
index 30a6878287ad..f10e26fb75b6 100644
@@ -74,36 +74,27 @@ int populate_cache_leaves(unsigned int cpu)
 {
 	struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
 	struct cacheinfo *this_leaf = this_cpu_ci->info_list;
-	struct device_node *np = of_cpu_device_node_get(cpu);
-	struct device_node *prev = NULL;
-	int levels = 1, level = 1;
-
-	if (of_property_read_bool(np, "cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
-	if (of_property_read_bool(np, "i-cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
-	if (of_property_read_bool(np, "d-cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level);
-
-	prev = np;
-	while ((np = of_find_next_cache_node(np))) {
-		of_node_put(prev);
-		prev = np;
-		if (!of_device_is_compatible(np, "cache"))
-			break;
-		if (of_property_read_u32(np, "cache-level", &level))
-			break;
-		if (level <= levels)
-			break;
-		if (of_property_read_bool(np, "cache-size"))
-			ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
-		if (of_property_read_bool(np, "i-cache-size"))
-			ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
-		if (of_property_read_bool(np, "d-cache-size"))
+	unsigned int level, idx;
+
+	for (idx = 0, level = 1; level <= this_cpu_ci->num_levels &&

arch/riscv/kernel/cacheinfo.c int populate_cache_leaves(unsigned int cpu) · line 89

Quoted 10 lines
arch/riscv/kernel/cacheinfo.c +10 -0
+	     idx < this_cpu_ci->num_leaves; idx++, level++) {
+		/*
+		 * Since the RISC-V architecture doesn't provide any register for detecting the
+		 * Cache Level and Cache type, this assumes that:
+		 * - There cannot be any split caches (data/instruction) above a unified cache.
+		 * - Data/instruction caches come in pairs.
+		 * - Significant work is required elsewhere to fully support data/instruction-only
+		 *   type caches.
+		 * - The above assumptions are based on conventional system design and known
+		 *   examples.

I don't think this comment matches what you are doing.

For example, the comment only requires that split caches cannot be above unified ones, but the code will always make a level 1 cache be split and higher level caches unified.

The place you took the comment about the split caches from does not enforce the type of cache layout that you do where the 1st level is always split and anything else is unified.

populate_cache_leaves() only gets called in a fallback path when the information has not already been configured by other means (and as you probably noticed on things like arm64 it uses some other means to fill in the data).

Is there a reason why we would not just return -ENOENT for ACPI systems if this has not been populated earlier in boot and leave the DT code here alone?

Thanks,
Conor.

Quoted 21 lines
arch/riscv/kernel/cacheinfo.c +6 -2
+		 */
+		if (level == 1) {
 			ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level);
-		levels = level;
+			ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
+		} else {
+			ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
+		}
 	}
-	of_node_put(np);
 
 	return 0;
 }
Signature · 7 lines
2.34.1


_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv
#4 lore raw

RE: [RFC v1 2/2] riscv: cacheinfo: Refactor populate_cache_leaves()

Quoted 10 lines

-----Original Message-----
From: Conor Dooley <[email protected]>
Sent: Monday, January 29, 2024 8:31 PM
To: JeeHeng Sia <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [RFC v1 2/2] riscv: cacheinfo: Refactor populate_cache_leaves()

Quoted 4 lines

Hey,

Firstly, the $subject should really mention that the motivation for the refactoring is ACPI support.

Noted. In fact, the main motivation is to support both DT and ACPI.

Quoted 60 lines · Sia Jee Heng
Quoted 45 lines

Refactoring the cache population function to support both DT and ACPI-based platforms.

---
 arch/riscv/kernel/cacheinfo.c | 47 ++++++++++++++---------------------
 1 file changed, 19 insertions(+), 28 deletions(-)
arch/riscv/kernel/cacheinfo.c +3 -26
index 30a6878287ad..f10e26fb75b6 100644
@@ -74,36 +74,27 @@ int populate_cache_leaves(unsigned int cpu)
 {
 	struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
 	struct cacheinfo *this_leaf = this_cpu_ci->info_list;
-	struct device_node *np = of_cpu_device_node_get(cpu);
-	struct device_node *prev = NULL;
-	int levels = 1, level = 1;
-
-	if (of_property_read_bool(np, "cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
-	if (of_property_read_bool(np, "i-cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
-	if (of_property_read_bool(np, "d-cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level);
-
-	prev = np;
-	while ((np = of_find_next_cache_node(np))) {
-		of_node_put(prev);
-		prev = np;
-		if (!of_device_is_compatible(np, "cache"))
-			break;
-		if (of_property_read_u32(np, "cache-level", &level))
-			break;
-		if (level <= levels)
-			break;
-		if (of_property_read_bool(np, "cache-size"))
-			ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
-		if (of_property_read_bool(np, "i-cache-size"))
-			ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
-		if (of_property_read_bool(np, "d-cache-size"))
+	unsigned int level, idx;
+
+	for (idx = 0, level = 1; level <= this_cpu_ci->num_levels &&

arch/riscv/kernel/cacheinfo.c int populate_cache_leaves(unsigned int cpu) · line 89

Quoted 10 lines
arch/riscv/kernel/cacheinfo.c +10 -0
+	     idx < this_cpu_ci->num_leaves; idx++, level++) {
+		/*
+		 * Since the RISC-V architecture doesn't provide any register for detecting the
+		 * Cache Level and Cache type, this assumes that:
+		 * - There cannot be any split caches (data/instruction) above a unified cache.
+		 * - Data/instruction caches come in pairs.
+		 * - Significant work is required elsewhere to fully support data/instruction-only
+		 *   type caches.
+		 * - The above assumptions are based on conventional system design and known
+		 *   examples.

I don't think this comment matches what you are doing.

Quoted 7 lines

For example, the comment only requires that split caches cannot be above unified ones, but the code will always make a level 1 cache be split and higher level caches unified.

The place you took the comment about the split caches from does not enforce the type of cache layout that you do where the 1st level is always split and anything else is unified.

Correct, I meant to say 1st level is always split and anything else is unified. But, do we agree with the statement?

Quoted 7 lines

populate_cache_leaves() only gets called in a fallback path when the information has not already been configured by other means (and as you probably noticed on things like arm64 it uses some other means to fill in the data).

Is there a reason why we would not just return -ENOENT for ACPI systems

I don't think that we should return -ENOENT otherwise the cacheinfo framework would failed.

Quoted 2 lines

if this has not been populated earlier in boot and leave the DT code here alone?

This function is shared by both ACPI and DT.

Quoted 21 lines

Thanks,
Conor.

Quoted 17 lines
arch/riscv/kernel/cacheinfo.c +6 -3
+		 */
+		if (level == 1) {
 			ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level);
-		levels = level;
+			ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
+		} else {
+			ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
+		}
 	}
-	of_node_put(np);
 
 	return 0;
 }
--

2.34.1

Quoted 4 lines
Quoted 4 lines

_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv

_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv

#5 lore raw

Re: [RFC v1 2/2] riscv: cacheinfo: Refactor populate_cache_leaves()

Quoted 65 lines · JeeHeng Sia
Quoted 65 lines

From: Conor Dooley <[email protected]>
Sent: Monday, January 29, 2024 8:31 PM

Quoted 45 lines · Sia Jee Heng

Refactoring the cache population function to support both DT and ACPI-based platforms.

---
 arch/riscv/kernel/cacheinfo.c | 47 ++++++++++++++---------------------
 1 file changed, 19 insertions(+), 28 deletions(-)
arch/riscv/kernel/cacheinfo.c +3 -26
index 30a6878287ad..f10e26fb75b6 100644
@@ -74,36 +74,27 @@ int populate_cache_leaves(unsigned int cpu)
 {
 	struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
 	struct cacheinfo *this_leaf = this_cpu_ci->info_list;
-	struct device_node *np = of_cpu_device_node_get(cpu);
-	struct device_node *prev = NULL;
-	int levels = 1, level = 1;
-
-	if (of_property_read_bool(np, "cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
-	if (of_property_read_bool(np, "i-cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
-	if (of_property_read_bool(np, "d-cache-size"))
-		ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level);
-
-	prev = np;
-	while ((np = of_find_next_cache_node(np))) {
-		of_node_put(prev);
-		prev = np;
-		if (!of_device_is_compatible(np, "cache"))
-			break;
-		if (of_property_read_u32(np, "cache-level", &level))
-			break;
-		if (level <= levels)
-			break;
-		if (of_property_read_bool(np, "cache-size"))
-			ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
-		if (of_property_read_bool(np, "i-cache-size"))
-			ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
-		if (of_property_read_bool(np, "d-cache-size"))
+	unsigned int level, idx;
+
+	for (idx = 0, level = 1; level <= this_cpu_ci->num_levels &&

arch/riscv/kernel/cacheinfo.c int populate_cache_leaves(unsigned int cpu) · line 89

Quoted 10 lines
arch/riscv/kernel/cacheinfo.c +10 -0
+	     idx < this_cpu_ci->num_leaves; idx++, level++) {
+		/*
+		 * Since the RISC-V architecture doesn't provide any register for detecting the
+		 * Cache Level and Cache type, this assumes that:
+		 * - There cannot be any split caches (data/instruction) above a unified cache.
+		 * - Data/instruction caches come in pairs.
+		 * - Significant work is required elsewhere to fully support data/instruction-only
+		 *   type caches.
+		 * - The above assumptions are based on conventional system design and known
+		 *   examples.

I don't think this comment matches what you are doing.

For example, the comment only requires that split caches cannot be above unified ones, but the code will always make a level 1 cache be split and higher level caches unified.

Quoted 5 lines
Quoted 3 lines

The place you took the comment about the split caches from does not enforce the type of cache layout that you do where the 1st level is always split and anything else is unified.

Correct, I meant to say 1st level is always split and anything else is unified. But, do we agree with the statement?

That the first level is always split and anything else is always unified? No, but I think the assumption /in the comment/ is reasonable however.

This is your patch, you need to justify the changes you are making here, not ask me if it is okay after I noticed that your comments and code do not match.

Quoted 8 lines
Quoted 6 lines

populate_cache_leaves() only gets called in a fallback path when the information has not already been configured by other means (and as you probably noticed on things like arm64 it uses some other means to fill in the data).

Is there a reason why we would not just return -ENOENT for ACPI systems

I don't think that we should return -ENOENT otherwise the cacheinfo framework would failed.

If you don't have a way to determine the cache layout, what makes -ENOENT worse than making something up?

Why does your system not get information from its ACPI tables?

Quoted 4 lines
Quoted 2 lines

if this has not been populated earlier in boot and leave the DT code here alone?

This function is shared by both ACPI and DT.

I don't see how that answers my question.

Why should the DT systems stop trying to parse for the information?

Why must ACPI and DT do the same thing here?

Thanks,
Conor.

Quoted 21 lines
Quoted 21 lines
Quoted 21 lines
arch/riscv/kernel/cacheinfo.c +6 -3
+		 */
+		if (level == 1) {
 			ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level);
-		levels = level;
+			ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level);
+		} else {
+			ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level);
+		}
 	}
-	of_node_put(np);
 
 	return 0;
 }
--

2.34.1

_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv

#6 lore raw
In reply to Sia Jee Heng
Quoted 4 lines · Sia Jee Heng

This series of patches refactors the cache population function to seamlessly accommodate both DT and ACPI-based platforms. Additionally, to streamline the code, the unused parameter in the ci_leaf_init() function has been removed.

Why's this an RFC? I don't see any mention of why.

Thanks,
Conor.

Quoted 18 lines

Sia Jee Heng (2):
riscv: cacheinfo: Remove unused parameter
riscv: cacheinfo: Refactor populate_cache_leaves()

 arch/riscv/kernel/cacheinfo.c | 50 ++++++++++++++---------------------
 1 file changed, 20 insertions(+), 30 deletions(-)

base-commit: 41bccc98fb7931d63d03f326a746ac4d429c1dd3

Signature · 7 lines
2.34.1


_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv
#7 lore raw
Quoted 10 lines

-----Original Message-----
From: Conor Dooley <[email protected]>
Sent: Monday, January 29, 2024 7:50 PM
To: JeeHeng Sia <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [RFC v1 0/2] riscv: Cache Population Code Refactoring for ACPI and DT Support

Quoted 7 lines
Quoted 4 lines · Sia Jee Heng

This series of patches refactors the cache population function to seamlessly accommodate both DT and ACPI-based platforms. Additionally, to streamline the code, the unused parameter in the ci_leaf_init() function has been removed.

Why's this an RFC? I don't see any mention of why.

My bad, I should have mentioned the reason in the cover letter. Btw, the reason treating this as RFC patch is because I am seeking for more opinions as it is the first try for riscv acpi based cacheinfo.

Quoted 22 lines

Thanks,
Conor.

Quoted 18 lines

Sia Jee Heng (2):
riscv: cacheinfo: Remove unused parameter
riscv: cacheinfo: Refactor populate_cache_leaves()

 arch/riscv/kernel/cacheinfo.c | 50 ++++++++++++++---------------------
 1 file changed, 20 insertions(+), 30 deletions(-)

base-commit: 41bccc98fb7931d63d03f326a746ac4d429c1dd3 --
2.34.1

_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv