Hi,
> Hi, Konstantin > > To illustrate this issue, I made the following modifications to the test > cases: > Modify the “test_data” content of the function "test_acl.c: > test_build_ports_range": > delete the second test data, and leave only one test message data. This way, > when > the function "acl_run_scalar.c: rte_acl_classify_scalar" processes the second > message > data, it will use the idle data. Then modify the idle data in the function > "acl_run.h: > acl_start_next_trie" to (to modify the idle, the const attribute was removed): > parms[n].data = (uint8_t *)idle; > parms[n].data[0] = 0x04; > parms[n].data[4] = 0x01; > When the function "acl_run_scalar.c: scalar_transition" processing this idle > message, > because the ranges are 0 and the message data is not 0(0x01), the message > data will > exceed all ranges. The calculated result x is 4, while the correct result > should be 0. The > following is a comparison of the settlement results of x before and after > modification > (I have printed each line of the calculation results of the function > "acl_run_stcalar. c: scalar_translation"): Sorry, it is a bit too cryptic for me. Why do you modify the internal ACL tables contents? If you can, please create a new test-case in app/test/test_acl.c to demonstrate the issue. Thanks Konstantin > ==before==: > Line 87: transition: 0x20000100 > Line 88: ranges: 0x0 > Line 89: index: 0x20000000 > Line 90: addr: 0x100 > Line 91: input: 0x0 > Line 96: c: 0x0 > Line 99: a: 0x80808080 > Line 102: a: 0x80808080 > Line 105: b: 0x0 > Line 108: a: 0x80808080 > Line 111: a: 0x80808080 > Line 114: x: 0x0 > Line 125: addr: 0x100 > > ==after==: > Line 87: transition: 0x20000100 > Line 88: ranges: 0x0 > Line 89: index: 0x20000000 > Line 90: addr: 0x100 > Line 91: input: 0x1 > Line 96: c: 0x1010101 > Line 99: a: 0x80808080 > Line 102: a: 0x7f7f7f7f > Line 105: b: 0x0 > Line 108: a: 0x0 > Line 111: a: 0x0 > Line 114: x: 0x4 > Line 125: addr: 0x104