Hello modders, since I have modded my router recently, I'd like to share some info about it. I am currently running fresh tomato-slodki-ND-1.28.02 firmware with SDHC mod. This is my set up: router WRT54GL v1.1 CPU speed 250 MHz SDHC: 16GB Class 10 (manufacturer A-DATA) SDHC has 2 partitions: 1st partition is 64MB type swapfs, 2nd partition with all remaining space is now ext2 type. Optware is installed on this partition too. I decided to use swapfs to expand router's available memory. I was getting 'Memory allocation error' when I tried to install some programs from optware. (e.g. mediatomb) Installing swap space did the trick. Command for adding a swap space is: Code: swapon /dev/mmc/disc0/part1 Based on well-known testing method: Code: sync; date; dd if=/dev/zero of=/mmc/test.bin bs=1024 count=102400; sync; date sync; date; dd if=/mmc/test.bin of=/dev/null bs=1024 count=102400; sync; date I measured these interesting results concerning speed of SDHC card: partition type: VFAT, writing speed 931 kB/s, reading speed 528 kB/s partition type: EXT2, writing speed 826 kB/s, reading speed 531 kB/s partition type: EXT3, writing speed 731 kB/s, reading speed 485 kB/s Tests were repeated 2 times for each partition type with similar results. I took the best results for each partition. I decided to keep ext2 type because of speed. It was not possible to use VFAT for installation of optware, so etx2 was a compromise for me. So far, so good. But there is a conflict between SD mod and tomato firmware as well. The conflict is not a big issue. However it might be annoying and makes scripts based on holding of SES button unusable. For example I used to have SSH started after holding SES for couple of seconds. Problem itself is caused by linking GPIO4 signal from CPU to DO (data out) pin of SD/SDHC card. This is a standard modification when we want SD card running. As far as I know, WRT54GL has no other option for this but GPIO4. So far nobody was able to find where e.g. GPIO5 is connected. Never mind. Standard purpose of GPIO4 is to react on SES button located on front panel. But if SD card is connected to the same pin in parallel, every communication with SD card acts as if SES button was pressed. System log is then overfull by messages 'SES pushed. Count was...'. Enabled scripts or events linked to SES button are run randomly too. It is not a big deal when you have active 'Reboot' or 'Toggle Wireless' in 'SES/AOSS Button' menu in such case. As a temporary solution I kill buttons process in startup. Maybe a permanent solution would be to modify buttons.c in firmware source to check if SD card communication is active and thus allow evaluate pressing of SES button only in case when there is no communication with SD card for some time. On the other hand pressing of SES button during communication with SD card must definitely cause an error in data sent from the card. Probably corrupted files on SD card too. So the best solution would be to find mentioned GPIO5 - if ever possible. Happy modding !