You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Many columns, e.g. LIMB columns in represent 16B data limbs. For these the leading zeros are part of the data and ought to be shown upon inspection. There should be an option (similar to the :display option for opcodes) to display these columns in full, leading zeros included.
This option may be extremely useful for debugging certain modules. Examples of modules containing LIMB columns are: SHAKIRA, ECDATA, RLP_TXN, BLAKE_MODEXP, RLP_ADDR, etc ... and most importantly of all: MMIO.
One could even imagine to have a link between columns of the form
When running corset inspect I thought initially the data was badly provided to the SHAKIRA module as the outputs looked identical for everything, even in terms of colours of the LIMB column, e.g.
KECCAK256(0xDEADBEEF)
KECCAK256(0x00DEADBEEF)
Both pictures look identical although we aren't hashing the same stuff.
And in full:
The text was updated successfully, but these errors were encountered:
Main issue
Many columns, e.g.
LIMB
columns in represent 16B data limbs. For these the leading zeros are part of the data and ought to be shown upon inspection. There should be an option (similar to the:display
option for opcodes) to display these columns in full, leading zeros included.This option may be extremely useful for debugging certain modules. Examples of modules containing
LIMB
columns are:SHAKIRA
,ECDATA
,RLP_TXN
,BLAKE_MODEXP
,RLP_ADDR
, etc ... and most importantly of all:MMIO
.One could even imagine to have a link between columns of the form
Where in
corset inspect
theLIMB
would be displayed like sonBYTES
are normal color16 - nBYTES
bytes are grayed outProbably that's asking a little much :)
Illustration of the issue
The following example illustrates the issue. I wrote a test where I was computing KECCAK's of variations on the hex string
deadbeef
:When running
corset inspect
I thought initially the data was badly provided to the SHAKIRA module as the outputs looked identical for everything, even in terms of colours of theLIMB
column, e.g.KECCAK256(
0xDEADBEEF
)KECCAK256(
0x00DEADBEEF
)Both pictures look identical although we aren't hashing the same stuff.
And in full:
The text was updated successfully, but these errors were encountered: