Fix aai percentage issue

[radare]

to show the % of code analyzed (considering data and basic blocks ranges) inside the current section (or just .text?)

[radare]

related to #2958

[Maijin]

Isn't this aai ?

[radare]

I dont see any % in this output

[Maijin]

[radare]

:D

…

On 4 Mar 2017, at 14:03, Maijin ***@***.***> wrote: — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

[xarkes]

Can you detail what you want ?

[radare]

The output of the command aai shows a bigger percentage than 100%, so there's a wrong computation here. this issue should track to fix that thing and p-% lgtm already

[Maijin]

p-% lgtm already
p-% doesn't exist, should aai be replaced by p-% ?

[radare]

No. Aai should be fixed. Just the percentage

…

On 7 Mar 2017, at 20:57, Maijin ***@***.***> wrote: p-% lgtm already p-% doesn't exist, should aai be replaced by p-% ? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

[Maijin]

So this may be a problem with realsize see the following:

https://github.com/radare/radare2/blob/master/libr/core/cmd_anal.c#L4962 => https://github.com/radare/radare2/blob/master/libr/core/cmd_anal.c#L4938

#4868

[radare]

Indeed

…

On 8 Mar 2017, at 01:26, Maijin ***@***.***> wrote: So this may be a problem with realsize see the following: https://github.com/radare/radare2/blob/master/libr/core/cmd_anal.c#L4962 => https://github.com/radare/radare2/blob/master/libr/core/cmd_anal.c#L4938 #4868 — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

[alvarofe]

Just testing since yesterday I couldn't (forgot)

air:pe alvaro$ r2 /bin/ls
Module version mismatch /Users/alvaro/.config/radare2/plugins/io_frida.dylib (1.3.0-git) vs (1.4.0-git)
 -- Use hasher to calculate hashes of portion blocks of a file
[0x100001174]> aaa
[x] Analyze all flags starting with sym. and entry0 (aa)
[x] Analyze len bytes of instructions for references (aar)
[x] Analyze function calls (aac)
[ ] [*] Use -AA or aaaa to perform additional experimental analysis.
[x] Constructing a function name for fcn.* and sym.func.* functions (aan))
[0x100001174]> aai
fcns    107
xrefs   1178
calls   0
strings 108
symbols 90
imports 76
covrage 0
codesz  16742
percent 0%

[xarkes]

Here's mine

 -- May the segfault be with you.
[0x00404030]> aaa
[x] Analyze all flags starting with sym. and entry0 (aa)
[x] Analyze len bytes of instructions for references (aar)
[x] Analyze function calls (aac)
[ ] [*] Use -AA or aaaa to perform additional experimental analysis.
[x] Constructing a function name for fcn.* and sym.func.* functions (aan))
[0x00404030]> aai
fcns    246
xrefs   6470
calls   76
strings 414
symbols 117
imports 109
covrage 109682
codesz  195549
percent 56%

[xarkes]

Actually the compute_coverage function does not work with Mach-O ls

[alvarofe]

maybe it doesn't work with mach0 since s->rwx iirc in is expressed in another way so s->rwx & 1 it fails.

[alvarofe]

so we should ensure the same flags throughout all sections