OpenSBI Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Anup Patel <apatel@ventanamicro.com>
To: opensbi@lists.infradead.org
Subject: [PATCH 2/7] lib: sbi_pmu: Replace sbi_pmu_ctr_read() with sbi_pmu_ctr_fw_read()
Date: Thu, 25 Aug 2022 10:21:39 +0530	[thread overview]
Message-ID: <20220825045144.752619-3-apatel@ventanamicro.com> (raw)
In-Reply-To: <20220825045144.752619-1-apatel@ventanamicro.com>

The "read a firmware counter" SBI call should only work for firmware
counters so let us replace sbi_pmu_ctr_read() with sbi_pmu_ctr_fw_read()
which works only on firmware counters.

Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
---
 include/sbi/sbi_pmu.h   |  2 +-
 lib/sbi/sbi_ecall_pmu.c |  3 ++-
 lib/sbi/sbi_pmu.c       | 45 +++++++----------------------------------
 3 files changed, 10 insertions(+), 40 deletions(-)

diff --git a/include/sbi/sbi_pmu.h b/include/sbi/sbi_pmu.h
index f5e3dbd..34336f7 100644
--- a/include/sbi/sbi_pmu.h
+++ b/include/sbi/sbi_pmu.h
@@ -53,7 +53,7 @@ int sbi_pmu_add_hw_event_counter_map(u32 eidx_start, u32 eidx_end, u32 cmap);
 
 int sbi_pmu_add_raw_event_counter_map(uint64_t select, uint64_t select_mask, u32 cmap);
 
-int sbi_pmu_ctr_read(uint32_t cidx, unsigned long *cval);
+int sbi_pmu_ctr_fw_read(uint32_t cidx, uint64_t *cval);
 
 int sbi_pmu_ctr_stop(unsigned long cidx_base, unsigned long cidx_mask,
 		     unsigned long flag);
diff --git a/lib/sbi/sbi_ecall_pmu.c b/lib/sbi/sbi_ecall_pmu.c
index 9ee9e81..826c8a8 100644
--- a/lib/sbi/sbi_ecall_pmu.c
+++ b/lib/sbi/sbi_ecall_pmu.c
@@ -51,7 +51,8 @@ static int sbi_ecall_pmu_handler(unsigned long extid, unsigned long funcid,
 
 		break;
 	case SBI_EXT_PMU_COUNTER_FW_READ:
-		ret = sbi_pmu_ctr_read(regs->a0, out_val);
+		ret = sbi_pmu_ctr_fw_read(regs->a0, &temp);
+		*out_val = temp;
 		break;
 	case SBI_EXT_PMU_COUNTER_START:
 
diff --git a/lib/sbi/sbi_pmu.c b/lib/sbi/sbi_pmu.c
index 535e5cc..250808e 100644
--- a/lib/sbi/sbi_pmu.c
+++ b/lib/sbi/sbi_pmu.c
@@ -167,50 +167,19 @@ static int pmu_ctr_validate(uint32_t cidx, uint32_t *event_idx_code)
 	return event_idx_type;
 }
 
-static int pmu_ctr_read_fw(uint32_t cidx, unsigned long *cval,
-			       uint32_t fw_evt_code)
-{
-	u32 hartid = current_hartid();
-	struct sbi_pmu_fw_event fevent;
-
-	fevent = fw_event_map[hartid][fw_evt_code];
-	*cval = fevent.curr_count;
-
-	return 0;
-}
-
-/* Add a hardware counter read for completeness for future purpose */
-static int pmu_ctr_read_hw(uint32_t cidx, uint64_t *cval)
-{
-	/* Check for invalid hw counter read requests */
-	if (unlikely(cidx == 1))
-		return SBI_EINVAL;
-#if __riscv_xlen == 32
-	uint32_t temp, temph = 0;
-
-	temp = csr_read_num(CSR_MCYCLE + cidx);
-	temph = csr_read_num(CSR_MCYCLEH + cidx);
-	*cval = ((uint64_t)temph << 32) | temp;
-#else
-	*cval = csr_read_num(CSR_MCYCLE + cidx);
-#endif
-
-	return 0;
-}
-
-int sbi_pmu_ctr_read(uint32_t cidx, unsigned long *cval)
+int sbi_pmu_ctr_fw_read(uint32_t cidx, uint64_t *cval)
 {
 	int event_idx_type;
 	uint32_t event_code;
-	uint64_t cval64;
+	u32 hartid = current_hartid();
+	struct sbi_pmu_fw_event *fevent;
 
 	event_idx_type = pmu_ctr_validate(cidx, &event_code);
-	if (event_idx_type < 0)
+	if (event_idx_type != SBI_PMU_EVENT_TYPE_FW)
 		return SBI_EINVAL;
-	else if (event_idx_type == SBI_PMU_EVENT_TYPE_FW)
-		pmu_ctr_read_fw(cidx, cval, event_code);
-	else
-		pmu_ctr_read_hw(cidx, &cval64);
+
+	fevent = &fw_event_map[hartid][event_code];
+	*cval = fevent->curr_count;
 
 	return 0;
 }
-- 
2.34.1



  parent reply	other threads:[~2022-08-25  4:51 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-25  4:51 [PATCH 0/7] OpenSBI PMU improvements Anup Patel
2022-08-25  4:51 ` [PATCH 1/7] lib: sbi_pmu: Remove "event_idx" member from struct sbi_pmu_fw_event Anup Patel
2022-08-25  4:51 ` Anup Patel [this message]
2022-08-25  4:51 ` [PATCH 3/7] lib: sbi_pmu: Firmware counters are always 64 bits wide Anup Patel
2022-08-25  4:51 ` [PATCH 4/7] lib: sbi_pmu: Simplify FW counters to reduce memory usage Anup Patel
2022-08-25  4:51 ` [PATCH 5/7] lib: sbi_pmu: Add custom PMU device operations Anup Patel
2022-08-25  8:05   ` Andrew Jones
2022-08-25  4:51 ` [PATCH 6/7] lib: sbi: Print platform PMU device at boot-time Anup Patel
     [not found]   ` <CAOnJCUL2kD_8dbwyBOX1XbaGizUqRAymPUhRWddavtG=aLojMA@mail.gmail.com>
2022-09-01 11:29     ` Anup Patel
2022-08-25  4:51 ` [PATCH 7/7] include: sbi: Reduce includes in sbi_pmu.h Anup Patel
2022-09-01 11:30 ` [PATCH 0/7] OpenSBI PMU improvements Anup Patel

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=20220825045144.752619-3-apatel@ventanamicro.com \
    --to=apatel@ventanamicro.com \
    --cc=opensbi@lists.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