Skip to contentSkip to conversation

eal: increase TRACE CTF SIZE to recommended size

PATCH In review Timothy McDaniel 4 participants 9 messages Oct 5, 2020 – Oct 8, 2020

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 eal: increase TRACE CTF SIZE to recommended size

Increase TRACE_CTF_FIELD_SIZE to 448, the recommended size. Fixes "CTF field is too long" error when running with trace enabled.

---
 lib/librte_eal/common/eal_trace.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
lib/librte_eal/common/eal_trace.h +1 -1
index 92c5951..438c2b7 100644
@@ -24,7 +24,7 @@
 
 #define TRACE_PREFIX_LEN 12
 #define TRACE_DIR_STR_LEN (sizeof("YYYY-mm-dd-AM-HH-MM-SS") + TRACE_PREFIX_LEN)
-#define TRACE_CTF_FIELD_SIZE 384
+#define TRACE_CTF_FIELD_SIZE 448
 #define TRACE_POINT_NAME_SIZE 64
 #define TRACE_CTF_MAGIC 0xC1FC1FC1
 #define TRACE_MAX_ARGS	32
Signature · 1 line
2.6.4
#1 lore raw
Quoted 2 lines · Timothy McDaniel

Increase TRACE_CTF_FIELD_SIZE to 448, the recommended size.

Repeating the same sentence in the title and the commitlog does not give much info.

Plus, what is this "recommendation"?

Quoted 1 line

Fixes "CTF field is too long" error when running with trace enabled.

Ok, you hit this limit, but it would help to get some context here. Looking at this patch in the future, we won't know why it was necessary.

Quoted 22 lines
---
 lib/librte_eal/common/eal_trace.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
lib/librte_eal/common/eal_trace.h +1 -2
index 92c5951..438c2b7 100644
@@ -24,7 +24,7 @@
 
 #define TRACE_PREFIX_LEN 12
 #define TRACE_DIR_STR_LEN (sizeof("YYYY-mm-dd-AM-HH-MM-SS") + TRACE_PREFIX_LEN)
-#define TRACE_CTF_FIELD_SIZE 384
+#define TRACE_CTF_FIELD_SIZE 448
 #define TRACE_POINT_NAME_SIZE 64
 #define TRACE_CTF_MAGIC 0xC1FC1FC1
 #define TRACE_MAX_ARGS 32
--

2.6.4

Signature · 1 line
David Marchand
#2 lore raw

Regards
Sunil Kumar Kori

Quoted 17 lines

-----Original Message-----
From: David Marchand <[email protected]>
Sent: Tuesday, October 6, 2020 2:15 PM
To: Timothy McDaniel <[email protected]>
Cc: Jerin Jacob Kollanukkaran <[email protected]>; Sunil Kumar Kori <[email protected]>; dev <[email protected]>; Erik Gabriel Carrillo <[email protected]>; Gage Eads <[email protected]>; Van Haaren Harry <[email protected]>
Subject: [EXT] Re: [dpdk-dev] [PATCH 1/1] eal: increase TRACE CTF SIZE to recommended size

External Email

----------------------------------------------------------------------

Quoted 1 line · Timothy McDaniel
Quoted 6 lines
Quoted 1 line

Increase TRACE_CTF_FIELD_SIZE to 448, the recommended size.

Repeating the same sentence in the title and the commitlog does not give much info.

Plus, what is this "recommendation"?

When analyzed this issue, only one more byte was needed to fix this issue but in future similar issue can occur again. So increasing this value by 64 bytes which actually equals to a cache line. That’s why we have suggested this size.

Quoted 35 lines
Quoted 3 lines

Fixes "CTF field is too long" error when running with trace enabled.

Ok, you hit this limit, but it would help to get some context here. Looking at this patch in the future, we won't know why it was necessary.

Quoted 15 lines
---
 lib/librte_eal/common/eal_trace.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
lib/librte_eal/common/eal_trace.h +0 -0
index 92c5951..438c2b7 100644
@@ -24,7 +24,7 @@
 
 #define TRACE_PREFIX_LEN 12
 #define TRACE_DIR_STR_LEN (sizeof("YYYY-mm-dd-AM-HH-MM-SS") +
Quoted 8 lines

TRACE_PREFIX_LEN) -#define TRACE_CTF_FIELD_SIZE 384 +#define TRACE_CTF_FIELD_SIZE 448
#define TRACE_POINT_NAME_SIZE 64
#define TRACE_CTF_MAGIC 0xC1FC1FC1
#define TRACE_MAX_ARGS 32
--
2.6.4

--
David Marchand

#3 lore raw
Quoted 11 lines · Sunil Kumar Kori
Quoted 9 lines
Quoted 2 lines · Timothy McDaniel

Increase TRACE_CTF_FIELD_SIZE to 448, the recommended size.

Repeating the same sentence in the title and the commitlog does not give much info.

Plus, what is this "recommendation"?

When analyzed this issue, only one more byte was needed to fix this issue but in future similar issue can occur again. So increasing this value by 64 bytes which actually equals to a cache line. That’s why we have suggested this size.

384 is aligned to both 64 and 128 bytes cache lines. 448 is only aligned to 64 bytes.

Should we care about 128 bytes cache lines systems?

Quoted 7 lines
Quoted 7 lines
Quoted 4 lines

Fixes "CTF field is too long" error when running with trace enabled.

Ok, you hit this limit, but it would help to get some context here. Looking at this patch in the future, we won't know why it was necessary.

How about following commitlog:

"""
trace: increase trace point buffer size

The current buffer size is not big enough to accomodate traces for new additions in the eventdev subsystem.
Increase this buffer size by XXX for reason YYY.
"""

Signature · 1 line
David Marchand
#4 lore raw
Quoted 10 lines · David Marchand
Quoted 8 lines · Sunil Kumar Kori
Quoted 8 lines
Quoted 2 lines · Timothy McDaniel

Increase TRACE_CTF_FIELD_SIZE to 448, the recommended size.

Repeating the same sentence in the title and the commitlog does not give much info.

Quoted 8 lines
Quoted 3 lines
Quoted 1 line

Plus, what is this "recommendation"?

When analyzed this issue, only one more byte was needed to fix this issue but in future similar issue can occur again. So increasing this value by 64 bytes which actually equals to a cache line. That’s why we have suggested this size.

384 is aligned to both 64 and 128 bytes cache lines. 448 is only aligned to 64 bytes.

Should we care about 128 bytes cache lines systems?

it is on a slow path. 448 is OK.

Quoted 11 lines
Quoted 8 lines
Quoted 8 lines
Quoted 5 lines

Fixes "CTF field is too long" error when running with trace enabled.

Ok, you hit this limit, but it would help to get some context here. Looking at this patch in the future, we won't know why it was necessary.

How about following commitlog:

Quoted 7 lines

"""
trace: increase trace point buffer size

The current buffer size is not big enough to accomodate traces for new additions in the eventdev subsystem.
Increase this buffer size by XXX for reason YYY.
"""

Looks good to me.

Quoted 5 lines

--
David Marchand

#5 lore raw
Quoted 16 lines · Jerin Jacob
Quoted 14 lines · David Marchand
Quoted 11 lines · Sunil Kumar Kori
Quoted 9 lines
Quoted 2 lines · Timothy McDaniel

Increase TRACE_CTF_FIELD_SIZE to 448, the recommended size.

Repeating the same sentence in the title and the commitlog does not give much info.

Plus, what is this "recommendation"?

When analyzed this issue, only one more byte was needed to fix this issue but in future similar issue can occur again. So increasing this value by 64 bytes which actually equals to a cache line. That’s why we have suggested this size.

Quoted 6 lines
Quoted 4 lines

384 is aligned to both 64 and 128 bytes cache lines. 448 is only aligned to 64 bytes.

Should we care about 128 bytes cache lines systems?

it is on a slow path. 448 is OK.

Ah yes, this is for the ctf description.
Could it be changed to rely on dynamic allocations and we simply remove this limit?

Quoted 15 lines
Quoted 15 lines
Quoted 9 lines
Quoted 9 lines
Quoted 6 lines

Fixes "CTF field is too long" error when running with trace enabled.

Ok, you hit this limit, but it would help to get some context here. Looking at this patch in the future, we won't know why it was necessary.

How about following commitlog:

"""
trace: increase trace point buffer size

Quoted 6 lines
Quoted 4 lines

The current buffer size is not big enough to accomodate traces for new additions in the eventdev subsystem.
Increase this buffer size by XXX for reason YYY.
"""

Looks good to me.

Well in this case, there is no actual reason.
The increased value is deemed "enough for now", unless we change this to dynamic allocations.

Signature · 1 line
David Marchand
#6 lore raw

Regards
Sunil Kumar Kori

Quoted 36 lines

-----Original Message-----
From: David Marchand <[email protected]>
Sent: Wednesday, October 7, 2020 2:34 PM
To: Jerin Jacob <[email protected]>
Cc: Sunil Kumar Kori <[email protected]>; Timothy McDaniel <[email protected]>; Jerin Jacob Kollanukkaran <[email protected]>; dev <[email protected]>; Erik Gabriel Carrillo <[email protected]>; Gage Eads <[email protected]>; Van Haaren Harry <[email protected]>
Subject: Re: [dpdk-dev] [EXT] Re: [PATCH 1/1] eal: increase TRACE CTF SIZE to recommended size

Quoted 2 lines · Jerin Jacob

On Tue, Oct 6, 2020 at 3:10 PM David Marchand

Quoted 2 lines · <[email protected]> wrote:
Quoted 2 lines

On Tue, Oct 6, 2020 at 11:22 AM Sunil Kumar Kori <[email protected]>

Quoted 10 lines · wrote:
Quoted 10 lines
Quoted 10 lines
Quoted 9 lines
Quoted 2 lines · Timothy McDaniel

Increase TRACE_CTF_FIELD_SIZE to 448, the recommended size.

Repeating the same sentence in the title and the commitlog does not give much info.

Plus, what is this "recommendation"?

When analyzed this issue, only one more byte was needed to fix this

issue but in future similar issue can occur again.

Quoted 1 line
Quoted 1 line
Quoted 1 line

So increasing this value by 64 bytes which actually equals to a cache line.

That’s why we have suggested this size.

Quoted 4 lines
Quoted 4 lines

384 is aligned to both 64 and 128 bytes cache lines. 448 is only aligned to 64 bytes.

Quoted 7 lines
Quoted 3 lines
Quoted 1 line

Should we care about 128 bytes cache lines systems?

it is on a slow path. 448 is OK.

Ah yes, this is for the ctf description.
Could it be changed to rely on dynamic allocations and we simply remove this limit?

Changing it to dynamic allocation is difficult because if we do this then every time memory is to reallocated. So IMO, It is okay to keep it static with enough size.

Quoted 33 lines
Quoted 11 lines
Quoted 11 lines
Quoted 11 lines
Quoted 11 lines
Quoted 8 lines

Fixes "CTF field is too long" error when running with trace enabled.

Ok, you hit this limit, but it would help to get some context here. Looking at this patch in the future, we won't know why it was

necessary.

Quoted 6 lines
Quoted 6 lines

How about following commitlog:

"""
trace: increase trace point buffer size

Quoted 6 lines
Quoted 4 lines

The current buffer size is not big enough to accomodate traces for new additions in the eventdev subsystem.
Increase this buffer size by XXX for reason YYY.
"""

Looks good to me.

Well in this case, there is no actual reason.
The increased value is deemed "enough for now", unless we change this to dynamic allocations.

--
David Marchand

#7 lore raw
Quoted 7 lines · Sunil Kumar Kori
Quoted 5 lines

it is on a slow path. 448 is OK.

Ah yes, this is for the ctf description.
Could it be changed to rely on dynamic allocations and we simply remove this limit?

Changing it to dynamic allocation is difficult because if we do this then every time memory is to reallocated. So IMO, It is okay to keep it static with enough size.

Ok.
I will take this as a ack from both Jerin and you and push with the suggested commitlog.

Signature · 1 line
David Marchand