tracing/eprobe: Replace kzalloc with kmalloc
PATCH 1/1 tracing/eprobe: Replace kzalloc with kmalloc
Since this memory will be filled soon below, I feel that there is no need for a memory of all zeros here. 'snprintf' does not return negative num according to ISO C99, so I feel that no judgment is needed here.
No functional change intended.
--- kernel/trace/trace_eprobe.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
kernel/trace/trace_eprobe.c +1 -3
@@ -917,15 +917,13 @@ static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const chfor (i =0 ; i < argc; i++) len +=strlen (argv[i]) +1 ; - ep->filter_str =kzalloc (len,GFP_KERNEL ); + ep->filter_str =kmalloc (len,GFP_KERNEL );if (!ep->filter_str)return -ENOMEM ; p = ep->filter_str;for (i =0 ; i < argc; i++) { ret =snprintf (p, len,"%s " , argv[i]); -if (ret <0 ) - goto error;if (ret > len) { ret = -E2BIG ; goto error;
Signature · 1 line
2.31.1
Le 07/01/2023 à 04:45, Quanfa Fu a écrit :
Quoted 21 lines
Since this memory will be filled soon below, I feel that there is no need for a memory of all zeros here. 'snprintf' does not return negative num according to ISO C99, so I feel that no judgment is needed here.
No functional change intended.
Signed-off-by Quanfa Fu <[email protected]>--- kernel/trace/trace_eprobe.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)kernel/trace/trace_eprobe.c +0
-1@@ -917,15 +917,13 @@ static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const chfor (i =0 ; i < argc; i++) len +=strlen (argv[i]) +1 ; - ep->filter_str =kzalloc (len,GFP_KERNEL );
kernel/trace/trace_eprobe.c static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const ch · line 927
Quoted 10 lines
kernel/trace/trace_eprobe.c +1
-2+ ep->filter_str =kmalloc (len,GFP_KERNEL );if (!ep->filter_str)return -ENOMEM ; p = ep->filter_str;for (i =0 ; i < argc; i++) { ret =snprintf (p, len,"%s " , argv[i]); -if (ret <0 ) - goto error;if (ret > len) {
Hi,
as per [1]:
* The return value is the number of characters which would be
* generated for the given input, excluding the trailing null,
* as per ISO C99. If the return is greater than *or equal* to
* @size, the resulting string is truncated.
So, should this test be:
if (ret >= len)
~~~~Also, isn't the p[-1] = '\0' after the loop eating the last character?
argc = 1;
argv[0] = "a";
Before the loop:
===============
len = 1 + 1 = 2;
ep->filter_str = 0x00 0x00
^
|___ p
After the loop:
===============
ep->filter_str = 0x61 0x00
^
|___ p
len = 1; After p[-1]:
============
ep->filter_str = 0x00 0x00
~~ ^
|___ pDid I miss something obvious?
I don't know the intent here, or if it is an issue at all, but it looks odd.
CJ
[1]: https://elixir.bootlin.com/linux/v6.2-rc1/source/lib/vsprintf.c#L2925
Quoted 2 lines
kernel/trace/trace_eprobe.c +0
-0ret = -E2BIG ; goto error;
Quoted 71 lines · Christophe JAILLET
Quoted 21 lines · Quanfa Fu
Since this memory will be filled soon below, I feel that there is no need for a memory of all zeros here. 'snprintf' does not return negative num according to ISO C99, so I feel that no judgment is needed here.
No functional change intended.
Signed-off-by Quanfa Fu <[email protected]>--- kernel/trace/trace_eprobe.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)kernel/trace/trace_eprobe.c +0
-1@@ -917,15 +917,13 @@ static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const chfor (i =0 ; i < argc; i++) len +=strlen (argv[i]) +1 ; - ep->filter_str =kzalloc (len,GFP_KERNEL );kernel/trace/trace_eprobe.c static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const ch · line 927
Quoted 10 lines
kernel/trace/trace_eprobe.c +1
-2+ ep->filter_str =kmalloc (len,GFP_KERNEL );if (!ep->filter_str)return -ENOMEM ; p = ep->filter_str;for (i =0 ; i < argc; i++) { ret =snprintf (p, len,"%s " , argv[i]); -if (ret <0 ) - goto error;if (ret > len) {Hi,
as per [1]:
* The return value is the number of characters which would be
* generated for the given input, excluding the trailing null,
* as per ISO C99. If the return is greater than *or equal* to
* @size, the resulting string is truncated.So, should this test be:
if (ret >= len) ~~~~Also, isn't the p[-1] = '\0' after the loop eating the last character?
argc = 1; argv[0] = "a"; Before the loop: =============== len = 1 + 1 = 2; ep->filter_str = 0x00 0x00 ^ |___ p After the loop: =============== ep->filter_str = 0x61 0x00 ^ |___ p len = 1; After p[-1]: ============ ep->filter_str = 0x00 0x00 ~~ ^ |___ p
Quoted 10 lines
Did I miss something obvious?
I don't know the intent here, or if it is an issue at all, but it looks odd.CJ
[1]: https://elixir.bootlin.com/linux/v6.2-rc1/source/lib/vsprintf.c#L2925
Quoted 2 lines
kernel/trace/trace_eprobe.c +0
-0ret = -E2BIG ; goto error;
I think that:
for example, argc = 2, argv = {"a", "b"};
Before the loop
===============
len = (1 + 1) + (1 + 1) = 4;ep->filter_str = 0x00 0x00 0x00 0x00
^
|__ pAfter the loop:
===============
i = 1, snprintf write: 'a' and ' ', so ret = 2
i = 2, snprintf write: 'b' and ' ', so ret = 2
===============ep->filter_str = 0x61 0x20 0x62 0x00 // Since the length of the
last argv is not enough to write, the space is replaced by null
^
|__ pp = ep->fiter_str + 2 (ret1) +2 (ret2) = ep->filter_str + 4
===============
so After p[-1] = *(ep->filter_str + 3) = '\0'; ep->filter_str = 0x61 0x20 0x62 0x00
According to ISO C99: " If the output was truncated due to this limit hen the return value
is the number of characters (excluding the terminating null byte) which would have been
written to the final string if enough space had been available"
The last snprintf will end with 'NULL', so I think p[-1] = '\0' can also be deleted
Quoted 71 lines · Christophe JAILLET
Quoted 21 lines · Quanfa Fu
Since this memory will be filled soon below, I feel that there is no need for a memory of all zeros here. 'snprintf' does not return negative num according to ISO C99, so I feel that no judgment is needed here.
No functional change intended.
Signed-off-by Quanfa Fu <[email protected]>--- kernel/trace/trace_eprobe.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)kernel/trace/trace_eprobe.c +0
-1@@ -917,15 +917,13 @@ static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const chfor (i =0 ; i < argc; i++) len +=strlen (argv[i]) +1 ; - ep->filter_str =kzalloc (len,GFP_KERNEL );kernel/trace/trace_eprobe.c static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const ch · line 927
Quoted 10 lines
kernel/trace/trace_eprobe.c +1
-2+ ep->filter_str =kmalloc (len,GFP_KERNEL );if (!ep->filter_str)return -ENOMEM ; p = ep->filter_str;for (i =0 ; i < argc; i++) { ret =snprintf (p, len,"%s " , argv[i]); -if (ret <0 ) - goto error;if (ret > len) {Hi,
as per [1]:
* The return value is the number of characters which would be
* generated for the given input, excluding the trailing null,
* as per ISO C99. If the return is greater than *or equal* to
* @size, the resulting string is truncated.So, should this test be:
if (ret >= len) ~~~~Also, isn't the p[-1] = '\0' after the loop eating the last character?
argc = 1; argv[0] = "a"; Before the loop: =============== len = 1 + 1 = 2; ep->filter_str = 0x00 0x00 ^ |___ p After the loop: =============== ep->filter_str = 0x61 0x00 ^ |___ p len = 1; After p[-1]: ============ ep->filter_str = 0x00 0x00 ~~ ^ |___ p
Quoted 10 lines
Did I miss something obvious?
I don't know the intent here, or if it is an issue at all, but it looks odd.CJ
[1]: https://elixir.bootlin.com/linux/v6.2-rc1/source/lib/vsprintf.c#L2925
Quoted 2 lines
kernel/trace/trace_eprobe.c +0
-0ret = -E2BIG ; goto error;
I think that:
for example, argc = 2, argv = {"a", "b"};
Before the loop
===============
len = (1 + 1) + (1 + 1) = 4;ep->filter_str = 0x00 0x00 0x00 0x00
^
|__ pAfter the loop:
===============
i = 1, snprintf write: 'a' and ' ', so ret = 2
i = 2, snprintf write: 'b' and ' ', so ret = 2
===============
Since the length of the last argv is not enough
to write, the space is replaced by null
ep->filter_str = 0x61 0x20 0x62 0x00
^
|__ pp = ep->fiter_str + 2 (ret1) +2 (ret2) = ep->filter_str + 4
===============
so After p[-1] = *(ep->filter_str + 3) = '\0'; ep->filter_str = 0x61 0x20 0x62 0x00
According to ISO C99: " If the output was truncated due to this limit then the return value is the number of characters (excluding the terminating null byte) which would have been written to the final string if enough space had been available"
The last snprintf will end with 'NULL', so I think p[-1] = '\0' can also be deleted
Le 07/01/2023 à 13:23, Quanfa Fu a écrit :
Quoted 45 lines
Quoted 27 lines · Christophe JAILLET
Quoted 25 lines · Quanfa Fu
Since this memory will be filled soon below, I feel that there is no need for a memory of all zeros here. 'snprintf' does not return negative num according to ISO C99, so I feel that no judgment is needed here.
No functional change intended.
Signed-off-by Quanfa Fu <[email protected]>--- kernel/trace/trace_eprobe.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)kernel/trace/trace_eprobe.c +1
-1@@ -917,15 +917,13 @@ static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const chfor (i =0 ; i < argc; i++) len +=strlen (argv[i]) +1 ; - ep->filter_str =kzalloc (len,GFP_KERNEL ); + ep->filter_str =kmalloc (len,GFP_KERNEL );if (!ep->filter_str)return -ENOMEM ;Quoted 7 lines
kernel/trace/trace_eprobe.c static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const ch · line 927
Quoted 6 lines
kernel/trace/trace_eprobe.c +0
-2p = ep->filter_str;for (i =0 ; i < argc; i++) { ret =snprintf (p, len,"%s " , argv[i]); -if (ret <0 ) - goto error;if (ret > len) {I think that:
for example, argc = 2, argv = {"a", "b"};
Before the loop
===============
len = (1 + 1) + (1 + 1) = 4;ep->filter_str=0x000x000x000x00
Quoted 8 lines
^ |__ pAfter the loop:
===============
i = 1, snprintf write: 'a' and ' ', so ret = 2
i = 2, snprintf write: 'b' and ' ', so ret = 2
===============
Ok, I missed the space after the %s.
Sorry for the noise.
Quoted 11 lines
Since the length of the last argv is not enough
to write, the space is replaced by null
ep->filter_str=0x610x200x620x00^ |__ pp =
ep->fiter_str+ 2 (ret1) +2 (ret2) =ep->filter_str+ 4
===============
so After p[-1] = *(ep->filter_str + 3) = '\0';ep->filter_str=0x610x200x620x00
Quoted 7 lines
According to ISO C99: " If the output was truncated due to this limit then the return value is the number of characters (excluding the terminating null byte) which would have been written to the final string if enough space had been available"
The last snprintf will end with 'NULL', so I think p[-1] = '\0' can also be deleted
Hmm, now that I see it, I think that it is there to remove the last space (even if there shouldn't be any because the last snprintf will be truncated).
Code LGTM as-is, even if puzzling.
CJ
Quoted 1 line
Quoted 7 lines · On Sat, 7 Jan 2023 11:45:57 +0800 Quanfa Fu <[email protected]> wrote:
Since this memory will be filled soon below, I feel that there is no need for a memory of all zeros here. 'snprintf' does not return negative num according to ISO C99, so I feel that no judgment is needed here.
No functional change intended.
Ah, good catch. I didn't notice that snprintf() doesn't return error code. (I confirmed that the linux internal snprintf() also doesn't return the error code)
Thank you!
Quoted 29 lines
Signed-off-by Quanfa Fu <[email protected]>--- kernel/trace/trace_eprobe.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)kernel/trace/trace_eprobe.c +1
-3@@ -917,15 +917,13 @@ static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const chfor (i =0 ; i < argc; i++) len +=strlen (argv[i]) +1 ; - ep->filter_str =kzalloc (len,GFP_KERNEL ); + ep->filter_str =kmalloc (len,GFP_KERNEL );if (!ep->filter_str)return -ENOMEM ; p = ep->filter_str;for (i =0 ; i < argc; i++) { ret =snprintf (p, len,"%s " , argv[i]); -if (ret <0 ) - goto error;if (ret > len) { ret = -E2BIG ; goto error;Signature · 1 line
2.31.1
Signature · 1 line
Masami Hiramatsu (Google) <[email protected]>
Steven Rostedt
maintainerQuoted 5 lines · On Sun, 8 Jan 2023 01:01:24 +0900 Masami Hiramatsu (Google)…
Ah, good catch. I didn't notice that
snprintf()doesn't return error code. (I confirmed that the linux internalsnprintf()also doesn't return the error code)
I replied to to the first instance of this patch ;-)
https://lore.kernel.org/all/[email protected]/
-- Steve