On 4 April 2018 at 12:42, Philippe Mathieu-Daudé <f4...@amsat.org> wrote: > Hi Peter, > > On 03/19/2018 01:15 PM, Peter Maydell wrote: >> Add some tracepoints to the bcm2835_sdhost driver, to assist >> debugging.
>> +# hw/sd/bcm2835_sdhost.c >> +bcm2835_sdhost_read(uint64_t offset, uint64_t data, unsigned size) "offset >> 0x%" PRIx64 " data 0x%" PRIx64 " size %u" >> +bcm2835_sdhost_write(uint64_t offset, uint64_t data, unsigned size) "offset >> 0x%" PRIx64 " data 0x%" PRIx64 " size %u" > > Can you use the more explicit "size_t" and "%zu" please? The argument to the read/write functions which we're tracing here isn't a size_t, though (and since it's only ever 1/2/4/8 it makes sense that it isn't a size_t). What would be the point in casting it to an size_t here? A quick grep through hw/*/trace-events suggests we don't use size_t for tracing of mmio read/write functions in other devices. thanks -- PMM