summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorcbbrowne <cbbrowne@cbbrowne.tor.int-afilias.info>2016-12-01 16:24:03 -0500
committercbbrowne <cbbrowne@cbbrowne.tor.int-afilias.info>2016-12-01 16:24:03 -0500
commitb787ed9dc6c81f47215c48d47240717afa3cd440 (patch)
treee422aef3b7a7d7ada490782f016c834df80fdc7b
parent1f290f6213c7c3bf001a85242cfae4f9b5a14942 (diff)
More notes on .hex size protection
-rw-r--r--readme.md19
1 files changed, 16 insertions, 3 deletions
diff --git a/readme.md b/readme.md
index a92ae4c1dc..5c82efcddd 100644
--- a/readme.md
+++ b/readme.md
@@ -1318,14 +1318,27 @@ You probably don't want to "brick" your keyboard, making it impossible
to rewrite firmware onto it. Here are some of the parameters to show
what things are (and likely aren't) too risky.
-- If a keyboard map does not include RESET, then, to get into DFU
+- If your keyboard map does not include RESET, then, to get into DFU
mode, you will need to press the reset button on the PCB, which
- requires unscrewing some bits.
+ requires unscrewing the bottom.
- Messing with tmk_core / common files might make the keyboard
inoperable
- Too large a .hex file is trouble; `make dfu` will erase the block,
test the size (oops, wrong order!), which errors out, failing to
- flash the keyboard
+ flash the keyboard, leaving it in DFU mode.
+ - To this end, note that the maximum .hex file size on Planck is
+ 7000h (28672 decimal)
+```Linking: .build/planck_rev4_cbbrowne.elf [OK]
+Creating load file for Flash: .build/planck_rev4_cbbrowne.hex [OK]
+
+Size after:
+ text data bss dec hex filename
+ 0 22396 0 22396 577c planck_rev4_cbbrowne.hex
+```
+ - The above file is of size 22396/577ch, which is less than
+ 28672/7000h
+ - As long as you have a suitable alternative .hex file around, you
+ can retry, loading that one
- DFU tools do /not/ allow you to write into the bootloader (unless
you throw in extra fruitsalad of options), so there is little risk
there.