Please find updated patch with the review comments given above implemented.
The compressed data now includes all backup blocks and their headers except the header of first backup block in WAL record. The first backup block header in WAL record is used to store the compression information. This is done in order to avoid adding compression information in WAL record header.
Memory allocation on SIGHUP in autovacuum is remaining. Working on it.
>>>Do we release the buffers for compressed data when fpw is changed from "compress" to "on"? >> The current code does not do this. >Don't we need to do that? Yes this needs to be done in order to avoid memory leak when compression is turned off at runtime while the backend session is running. >You don't need to make the processes except the startup process allocate >the memory for uncompressedPages when fpw=on. Only the startup process >uses it for the WAL decompression I see. fpw != on check can be put at the time of memory allocation of uncompressedPages in the backend code . And at the time of recovery uncompressedPages can be allocated separately if not already allocated. >BTW, what happens if the memory allocation for uncompressedPages for >the recovery fails? The current code does not handle this. This will be rectified. >Which would prevent the recovery at all, so PANIC should >happen in that case? IIUC, instead of reporting PANIC , palloc can be used to allocate memory for uncompressedPages at the time of recovery which will throw ERROR and abort startup process in case of failure.
