mirror of
https://github.com/espressif/binutils-gdb.git
synced 2025-06-13 20:10:02 +08:00

This commit introduces the idea of loading only part of an array in order to print it, what I call "limited length" arrays. The motivation behind this work is to make it possible to print slices of very large arrays, where very large means bigger than `max-value-size'. Consider this GDB session with the current GDB: (gdb) set max-value-size 100 (gdb) p large_1d_array value requires 400 bytes, which is more than max-value-size (gdb) p -elements 10 -- large_1d_array value requires 400 bytes, which is more than max-value-size notice that the request to print 10 elements still fails, even though 10 elements should be less than the max-value-size. With a patched version of GDB: (gdb) p -elements 10 -- large_1d_array $1 = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9...} So now the print has succeeded. It also has loaded `max-value-size' worth of data into value history, so the recorded value can be accessed consistently: (gdb) p -elements 10 -- $1 $2 = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9...} (gdb) p $1 $3 = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, <unavailable> <repeats 75 times>} (gdb) Accesses with other languages work similarly, although for Ada only C-style [] array element/dimension accesses use history. For both Ada and Fortran () array element/dimension accesses go straight to the inferior, bypassing the value history just as with C pointers. Co-Authored-By: Maciej W. Rozycki <macro@embecosm.com>