* [PATCH v7 0/3] Enable Zicbom in usermode
@ 2025-02-26 6:32 Yunhui Cui
2025-02-26 6:32 ` [PATCH v7 1/3] RISC-V: Enable cbo.clean/flush " Yunhui Cui
` (3 more replies)
0 siblings, 4 replies; 5+ messages in thread
From: Yunhui Cui @ 2025-02-26 6:32 UTC (permalink / raw)
To: ajones, alexghiti, andybnac, aou, charlie, cleger, conor.dooley,
conor, corbet, cuiyunhui, evan, jesse, linux-doc, linux-kernel,
linux-kselftest, linux-riscv, palmer, paul.walmsley,
samuel.holland, shuah
v1/v2:
There is only the first patch: RISC-V: Enable cbo.clean/flush in usermode,
which mainly removes the enabling of cbo.inval in user mode.
v3:
Add the functionality of Expose Zicbom and selftests for Zicbom.
v4:
Modify the order of macros, The test_no_cbo_inval function is added
separately.
v5:
1. Modify the order of RISCV_HWPROBE_KEY_ZICBOM_BLOCK_SIZE in hwprobe.rst
2. "TEST_NO_ZICBOINVAL" -> "TEST_NO_CBO_INVAL"
v6:
Change hwprobe_ext0_has's second param to u64.
v7:
Rebase to the latest code of linux-next.
Yunhui Cui (3):
RISC-V: Enable cbo.clean/flush in usermode
RISC-V: hwprobe: Expose Zicbom extension and its block size
RISC-V: selftests: Add TEST_ZICBOM into CBO tests
Documentation/arch/riscv/hwprobe.rst | 6 ++
arch/riscv/include/asm/hwprobe.h | 2 +-
arch/riscv/include/uapi/asm/hwprobe.h | 2 +
arch/riscv/kernel/cpufeature.c | 8 +++
arch/riscv/kernel/sys_hwprobe.c | 8 ++-
tools/testing/selftests/riscv/hwprobe/cbo.c | 66 +++++++++++++++++----
6 files changed, 79 insertions(+), 13 deletions(-)
--
2.39.2
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v7 1/3] RISC-V: Enable cbo.clean/flush in usermode
2025-02-26 6:32 [PATCH v7 0/3] Enable Zicbom in usermode Yunhui Cui
@ 2025-02-26 6:32 ` Yunhui Cui
2025-02-26 6:32 ` [PATCH v7 2/3] RISC-V: hwprobe: Expose Zicbom extension and its block size Yunhui Cui
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Yunhui Cui @ 2025-02-26 6:32 UTC (permalink / raw)
To: ajones, alexghiti, andybnac, aou, charlie, cleger, conor.dooley,
conor, corbet, cuiyunhui, evan, jesse, linux-doc, linux-kernel,
linux-kselftest, linux-riscv, palmer, paul.walmsley,
samuel.holland, shuah
Enabling cbo.clean and cbo.flush in user mode makes it more
convenient to manage the cache state and achieve better performance.
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Yunhui Cui <cuiyunhui@bytedance.com>
---
arch/riscv/kernel/cpufeature.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
index 40ac72e407b6..a8b726d0dab8 100644
--- a/arch/riscv/kernel/cpufeature.c
+++ b/arch/riscv/kernel/cpufeature.c
@@ -32,6 +32,7 @@
#define NUM_ALPHA_EXTS ('z' - 'a' + 1)
static bool any_cpu_has_zicboz;
+static bool any_cpu_has_zicbom;
unsigned long elf_hwcap __read_mostly;
@@ -91,6 +92,8 @@ static int riscv_ext_zicbom_validate(const struct riscv_isa_ext_data *data,
pr_err("Zicbom disabled as cbom-block-size present, but is not a power-of-2\n");
return -EINVAL;
}
+
+ any_cpu_has_zicbom = true;
return 0;
}
@@ -1001,6 +1004,11 @@ void __init riscv_user_isa_enable(void)
current->thread.envcfg |= ENVCFG_CBZE;
else if (any_cpu_has_zicboz)
pr_warn("Zicboz disabled as it is unavailable on some harts\n");
+
+ if (riscv_has_extension_unlikely(RISCV_ISA_EXT_ZICBOM))
+ current->thread.envcfg |= ENVCFG_CBCFE;
+ else if (any_cpu_has_zicbom)
+ pr_warn("Zicbom disabled as it is unavailable on some harts\n");
}
#ifdef CONFIG_RISCV_ALTERNATIVE
--
2.39.2
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v7 2/3] RISC-V: hwprobe: Expose Zicbom extension and its block size
2025-02-26 6:32 [PATCH v7 0/3] Enable Zicbom in usermode Yunhui Cui
2025-02-26 6:32 ` [PATCH v7 1/3] RISC-V: Enable cbo.clean/flush " Yunhui Cui
@ 2025-02-26 6:32 ` Yunhui Cui
2025-02-26 6:32 ` [PATCH v7 3/3] RISC-V: selftests: Add TEST_ZICBOM into CBO tests Yunhui Cui
2025-03-27 3:24 ` [PATCH v7 0/3] Enable Zicbom in usermode patchwork-bot+linux-riscv
3 siblings, 0 replies; 5+ messages in thread
From: Yunhui Cui @ 2025-02-26 6:32 UTC (permalink / raw)
To: ajones, alexghiti, andybnac, aou, charlie, cleger, conor.dooley,
conor, corbet, cuiyunhui, evan, jesse, linux-doc, linux-kernel,
linux-kselftest, linux-riscv, palmer, paul.walmsley,
samuel.holland, shuah
Expose Zicbom through hwprobe and also provide a key to extract its
respective block size.
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Samuel Holland <samuel.holland@sifive.com>
Signed-off-by: Yunhui Cui <cuiyunhui@bytedance.com>
---
Documentation/arch/riscv/hwprobe.rst | 6 ++++++
arch/riscv/include/asm/hwprobe.h | 2 +-
arch/riscv/include/uapi/asm/hwprobe.h | 2 ++
arch/riscv/kernel/sys_hwprobe.c | 8 +++++++-
4 files changed, 16 insertions(+), 2 deletions(-)
diff --git a/Documentation/arch/riscv/hwprobe.rst b/Documentation/arch/riscv/hwprobe.rst
index f273ea15a8e8..d91e39e19ea0 100644
--- a/Documentation/arch/riscv/hwprobe.rst
+++ b/Documentation/arch/riscv/hwprobe.rst
@@ -242,6 +242,9 @@ The following keys are defined:
* :c:macro:`RISCV_HWPROBE_EXT_SUPM`: The Supm extension is supported as
defined in version 1.0 of the RISC-V Pointer Masking extensions.
+ * :c:macro:`RISCV_HWPROBE_EXT_ZICBOM`: The Zicbom extension is supported, as
+ ratified in commit 3dd606f ("Create cmobase-v1.0.pdf") of riscv-CMOs.
+
* :c:macro:`RISCV_HWPROBE_KEY_CPUPERF_0`: Deprecated. Returns similar values to
:c:macro:`RISCV_HWPROBE_KEY_MISALIGNED_SCALAR_PERF`, but the key was
mistakenly classified as a bitmask rather than a value.
@@ -303,3 +306,6 @@ The following keys are defined:
* :c:macro:`RISCV_HWPROBE_VENDOR_EXT_XTHEADVECTOR`: The xtheadvector vendor
extension is supported in the T-Head ISA extensions spec starting from
commit a18c801634 ("Add T-Head VECTOR vendor extension. ").
+
+* :c:macro:`RISCV_HWPROBE_KEY_ZICBOM_BLOCK_SIZE`: An unsigned int which
+ represents the size of the Zicbom block in bytes.
diff --git a/arch/riscv/include/asm/hwprobe.h b/arch/riscv/include/asm/hwprobe.h
index dd624523981c..1f690fea0e03 100644
--- a/arch/riscv/include/asm/hwprobe.h
+++ b/arch/riscv/include/asm/hwprobe.h
@@ -8,7 +8,7 @@
#include <uapi/asm/hwprobe.h>
-#define RISCV_HWPROBE_MAX_KEY 11
+#define RISCV_HWPROBE_MAX_KEY 12
static inline bool riscv_hwprobe_key_is_valid(__s64 key)
{
diff --git a/arch/riscv/include/uapi/asm/hwprobe.h b/arch/riscv/include/uapi/asm/hwprobe.h
index c3c1cc951cb9..8c706cde3777 100644
--- a/arch/riscv/include/uapi/asm/hwprobe.h
+++ b/arch/riscv/include/uapi/asm/hwprobe.h
@@ -73,6 +73,7 @@ struct riscv_hwprobe {
#define RISCV_HWPROBE_EXT_ZCMOP (1ULL << 47)
#define RISCV_HWPROBE_EXT_ZAWRS (1ULL << 48)
#define RISCV_HWPROBE_EXT_SUPM (1ULL << 49)
+#define RISCV_HWPROBE_EXT_ZICBOM (1ULL << 50)
#define RISCV_HWPROBE_KEY_CPUPERF_0 5
#define RISCV_HWPROBE_MISALIGNED_UNKNOWN (0 << 0)
#define RISCV_HWPROBE_MISALIGNED_EMULATED (1 << 0)
@@ -95,6 +96,7 @@ struct riscv_hwprobe {
#define RISCV_HWPROBE_MISALIGNED_VECTOR_FAST 3
#define RISCV_HWPROBE_MISALIGNED_VECTOR_UNSUPPORTED 4
#define RISCV_HWPROBE_KEY_VENDOR_EXT_THEAD_0 11
+#define RISCV_HWPROBE_KEY_ZICBOM_BLOCK_SIZE 12
/* Increase RISCV_HWPROBE_MAX_KEY when adding items. */
/* Flags */
diff --git a/arch/riscv/kernel/sys_hwprobe.c b/arch/riscv/kernel/sys_hwprobe.c
index 04a4e5495512..69fe9da320dc 100644
--- a/arch/riscv/kernel/sys_hwprobe.c
+++ b/arch/riscv/kernel/sys_hwprobe.c
@@ -107,6 +107,7 @@ static void hwprobe_isa_ext0(struct riscv_hwprobe *pair,
EXT_KEY(ZCA);
EXT_KEY(ZCB);
EXT_KEY(ZCMOP);
+ EXT_KEY(ZICBOM);
EXT_KEY(ZICBOZ);
EXT_KEY(ZICOND);
EXT_KEY(ZIHINTNTL);
@@ -161,7 +162,7 @@ static void hwprobe_isa_ext0(struct riscv_hwprobe *pair,
pair->value &= ~missing;
}
-static bool hwprobe_ext0_has(const struct cpumask *cpus, unsigned long ext)
+static bool hwprobe_ext0_has(const struct cpumask *cpus, u64 ext)
{
struct riscv_hwprobe pair;
@@ -279,6 +280,11 @@ static void hwprobe_one_pair(struct riscv_hwprobe *pair,
if (hwprobe_ext0_has(cpus, RISCV_HWPROBE_EXT_ZICBOZ))
pair->value = riscv_cboz_block_size;
break;
+ case RISCV_HWPROBE_KEY_ZICBOM_BLOCK_SIZE:
+ pair->value = 0;
+ if (hwprobe_ext0_has(cpus, RISCV_HWPROBE_EXT_ZICBOM))
+ pair->value = riscv_cbom_block_size;
+ break;
case RISCV_HWPROBE_KEY_HIGHEST_VIRT_ADDRESS:
pair->value = user_max_virt_addr();
break;
--
2.39.2
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v7 3/3] RISC-V: selftests: Add TEST_ZICBOM into CBO tests
2025-02-26 6:32 [PATCH v7 0/3] Enable Zicbom in usermode Yunhui Cui
2025-02-26 6:32 ` [PATCH v7 1/3] RISC-V: Enable cbo.clean/flush " Yunhui Cui
2025-02-26 6:32 ` [PATCH v7 2/3] RISC-V: hwprobe: Expose Zicbom extension and its block size Yunhui Cui
@ 2025-02-26 6:32 ` Yunhui Cui
2025-03-27 3:24 ` [PATCH v7 0/3] Enable Zicbom in usermode patchwork-bot+linux-riscv
3 siblings, 0 replies; 5+ messages in thread
From: Yunhui Cui @ 2025-02-26 6:32 UTC (permalink / raw)
To: ajones, alexghiti, andybnac, aou, charlie, cleger, conor.dooley,
conor, corbet, cuiyunhui, evan, jesse, linux-doc, linux-kernel,
linux-kselftest, linux-riscv, palmer, paul.walmsley,
samuel.holland, shuah
Add test for Zicbom and its block size into CBO tests, when
Zicbom is present, test that cbo.clean/flush may be issued and works.
As the software can't verify the clean/flush functions, we just judged
that cbo.clean/flush isn't executed illegally.
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Samuel Holland <samuel.holland@sifive.com>
Signed-off-by: Yunhui Cui <cuiyunhui@bytedance.com>
---
tools/testing/selftests/riscv/hwprobe/cbo.c | 66 +++++++++++++++++----
1 file changed, 55 insertions(+), 11 deletions(-)
diff --git a/tools/testing/selftests/riscv/hwprobe/cbo.c b/tools/testing/selftests/riscv/hwprobe/cbo.c
index a40541bb7c7d..5e96ef785d0d 100644
--- a/tools/testing/selftests/riscv/hwprobe/cbo.c
+++ b/tools/testing/selftests/riscv/hwprobe/cbo.c
@@ -50,6 +50,14 @@ static void cbo_clean(char *base) { cbo_insn(base, 1); }
static void cbo_flush(char *base) { cbo_insn(base, 2); }
static void cbo_zero(char *base) { cbo_insn(base, 4); }
+static void test_no_cbo_inval(void *arg)
+{
+ ksft_print_msg("Testing cbo.inval instruction remain privileged\n");
+ illegal_insn = false;
+ cbo_inval(&mem[0]);
+ ksft_test_result(illegal_insn, "No cbo.inval\n");
+}
+
static void test_no_zicbom(void *arg)
{
ksft_print_msg("Testing Zicbom instructions remain privileged\n");
@@ -61,10 +69,6 @@ static void test_no_zicbom(void *arg)
illegal_insn = false;
cbo_flush(&mem[0]);
ksft_test_result(illegal_insn, "No cbo.flush\n");
-
- illegal_insn = false;
- cbo_inval(&mem[0]);
- ksft_test_result(illegal_insn, "No cbo.inval\n");
}
static void test_no_zicboz(void *arg)
@@ -81,6 +85,30 @@ static bool is_power_of_2(__u64 n)
return n != 0 && (n & (n - 1)) == 0;
}
+static void test_zicbom(void *arg)
+{
+ struct riscv_hwprobe pair = {
+ .key = RISCV_HWPROBE_KEY_ZICBOM_BLOCK_SIZE,
+ };
+ cpu_set_t *cpus = (cpu_set_t *)arg;
+ __u64 block_size;
+ long rc;
+
+ rc = riscv_hwprobe(&pair, 1, sizeof(cpu_set_t), (unsigned long *)cpus, 0);
+ block_size = pair.value;
+ ksft_test_result(rc == 0 && pair.key == RISCV_HWPROBE_KEY_ZICBOM_BLOCK_SIZE &&
+ is_power_of_2(block_size), "Zicbom block size\n");
+ ksft_print_msg("Zicbom block size: %llu\n", block_size);
+
+ illegal_insn = false;
+ cbo_clean(&mem[block_size]);
+ ksft_test_result(!illegal_insn, "cbo.clean\n");
+
+ illegal_insn = false;
+ cbo_flush(&mem[block_size]);
+ ksft_test_result(!illegal_insn, "cbo.flush\n");
+}
+
static void test_zicboz(void *arg)
{
struct riscv_hwprobe pair = {
@@ -129,7 +157,7 @@ static void test_zicboz(void *arg)
ksft_test_result_pass("cbo.zero check\n");
}
-static void check_no_zicboz_cpus(cpu_set_t *cpus)
+static void check_no_zicbo_cpus(cpu_set_t *cpus, __u64 cbo)
{
struct riscv_hwprobe pair = {
.key = RISCV_HWPROBE_KEY_IMA_EXT_0,
@@ -137,6 +165,7 @@ static void check_no_zicboz_cpus(cpu_set_t *cpus)
cpu_set_t one_cpu;
int i = 0, c = 0;
long rc;
+ char *cbostr;
while (i++ < CPU_COUNT(cpus)) {
while (!CPU_ISSET(c, cpus))
@@ -148,10 +177,13 @@ static void check_no_zicboz_cpus(cpu_set_t *cpus)
rc = riscv_hwprobe(&pair, 1, sizeof(cpu_set_t), (unsigned long *)&one_cpu, 0);
assert(rc == 0 && pair.key == RISCV_HWPROBE_KEY_IMA_EXT_0);
- if (pair.value & RISCV_HWPROBE_EXT_ZICBOZ)
- ksft_exit_fail_msg("Zicboz is only present on a subset of harts.\n"
- "Use taskset to select a set of harts where Zicboz\n"
- "presence (present or not) is consistent for each hart\n");
+ cbostr = cbo == RISCV_HWPROBE_EXT_ZICBOZ ? "Zicboz" : "Zicbom";
+
+ if (pair.value & cbo)
+ ksft_exit_fail_msg("%s is only present on a subset of harts.\n"
+ "Use taskset to select a set of harts where %s\n"
+ "presence (present or not) is consistent for each hart\n",
+ cbostr, cbostr);
++c;
}
}
@@ -159,7 +191,9 @@ static void check_no_zicboz_cpus(cpu_set_t *cpus)
enum {
TEST_ZICBOZ,
TEST_NO_ZICBOZ,
+ TEST_ZICBOM,
TEST_NO_ZICBOM,
+ TEST_NO_CBO_INVAL,
};
static struct test_info {
@@ -169,7 +203,9 @@ static struct test_info {
} tests[] = {
[TEST_ZICBOZ] = { .nr_tests = 3, test_zicboz },
[TEST_NO_ZICBOZ] = { .nr_tests = 1, test_no_zicboz },
- [TEST_NO_ZICBOM] = { .nr_tests = 3, test_no_zicbom },
+ [TEST_ZICBOM] = { .nr_tests = 3, test_zicbom },
+ [TEST_NO_ZICBOM] = { .nr_tests = 2, test_no_zicbom },
+ [TEST_NO_CBO_INVAL] = { .nr_tests = 1, test_no_cbo_inval },
};
int main(int argc, char **argv)
@@ -189,6 +225,7 @@ int main(int argc, char **argv)
assert(rc == 0);
tests[TEST_NO_ZICBOZ].enabled = true;
tests[TEST_NO_ZICBOM].enabled = true;
+ tests[TEST_NO_CBO_INVAL].enabled = true;
}
rc = sched_getaffinity(0, sizeof(cpu_set_t), &cpus);
@@ -206,7 +243,14 @@ int main(int argc, char **argv)
tests[TEST_ZICBOZ].enabled = true;
tests[TEST_NO_ZICBOZ].enabled = false;
} else {
- check_no_zicboz_cpus(&cpus);
+ check_no_zicbo_cpus(&cpus, RISCV_HWPROBE_EXT_ZICBOZ);
+ }
+
+ if (pair.value & RISCV_HWPROBE_EXT_ZICBOM) {
+ tests[TEST_ZICBOM].enabled = true;
+ tests[TEST_NO_ZICBOM].enabled = false;
+ } else {
+ check_no_zicbo_cpus(&cpus, RISCV_HWPROBE_EXT_ZICBOM);
}
for (i = 0; i < ARRAY_SIZE(tests); ++i)
--
2.39.2
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v7 0/3] Enable Zicbom in usermode
2025-02-26 6:32 [PATCH v7 0/3] Enable Zicbom in usermode Yunhui Cui
` (2 preceding siblings ...)
2025-02-26 6:32 ` [PATCH v7 3/3] RISC-V: selftests: Add TEST_ZICBOM into CBO tests Yunhui Cui
@ 2025-03-27 3:24 ` patchwork-bot+linux-riscv
3 siblings, 0 replies; 5+ messages in thread
From: patchwork-bot+linux-riscv @ 2025-03-27 3:24 UTC (permalink / raw)
To: yunhui cui
Cc: linux-riscv, ajones, alexghiti, andybnac, aou, charlie, cleger,
conor.dooley, conor, corbet, evan, jesse, linux-doc,
linux-kernel, linux-kselftest, palmer, paul.walmsley,
samuel.holland, shuah
Hello:
This series was applied to riscv/linux.git (for-next)
by Alexandre Ghiti <alexghiti@rivosinc.com>:
On Wed, 26 Feb 2025 14:32:03 +0800 you wrote:
> v1/v2:
> There is only the first patch: RISC-V: Enable cbo.clean/flush in usermode,
> which mainly removes the enabling of cbo.inval in user mode.
>
> v3:
> Add the functionality of Expose Zicbom and selftests for Zicbom.
>
> [...]
Here is the summary with links:
- [v7,1/3] RISC-V: Enable cbo.clean/flush in usermode
https://git.kernel.org/riscv/c/de70b532f91b
- [v7,2/3] RISC-V: hwprobe: Expose Zicbom extension and its block size
https://git.kernel.org/riscv/c/eb1003970940
- [v7,3/3] RISC-V: selftests: Add TEST_ZICBOM into CBO tests
https://git.kernel.org/riscv/c/36dec9e44805
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2025-03-27 3:24 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2025-02-26 6:32 [PATCH v7 0/3] Enable Zicbom in usermode Yunhui Cui
2025-02-26 6:32 ` [PATCH v7 1/3] RISC-V: Enable cbo.clean/flush " Yunhui Cui
2025-02-26 6:32 ` [PATCH v7 2/3] RISC-V: hwprobe: Expose Zicbom extension and its block size Yunhui Cui
2025-02-26 6:32 ` [PATCH v7 3/3] RISC-V: selftests: Add TEST_ZICBOM into CBO tests Yunhui Cui
2025-03-27 3:24 ` [PATCH v7 0/3] Enable Zicbom in usermode patchwork-bot+linux-riscv
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®