Hi everyone, Ik did some more testing en after a export to the LPCXpresso IDE logging is not locked after 1368 appends. So the possible problem is the Mbed compiler. I only have make a import and I changed nothing to the LPCXpresso IDE compiler settings. 26 june 2016 ; For sure its the online Mbed compiler, after export to LPCXpresso I did every try to lock up my logging. 2/3 time a second a write en remove random the SD card run the program for a few days and everything works rock stable.....

inline bool SDFileSystem::waitReady(int timeout) // timeout (ms)
{
    char resp;
    //Keep sending dummy clocks with DI held high until the card releases the DO line
    m_Timer.start();
    do {
        resp = m_Spi.write(0xFF);
    } while (resp == 0x00 && m_Timer.read_ms() < timeout);
    m_Timer.stop();
    m_Timer.reset();
    //Return success/failure
    return (resp > 0x00);
}

int Timer::read_us() {
    core_util_critical_section_enter();
    int time = _time + slicetime();
    core_util_critical_section_exit();
    return time;
}

void core_util_critical_section_enter(void)
{
    bool interrupts_disabled = !core_util_are_interrupts_enabled();
    __disable_irq();

    /* Save the interrupt disabled state as it was prior to any nested critical section lock use */
    if (!interrupt_enable_counter) {
        critical_interrupts_disabled = interrupts_disabled;
    }

    /* If the interrupt_enable_counter overflows or we are in a nested critical section and interrupts
       are enabled, then something has gone badly wrong thus assert an error.
    */
    MBED_ASSERT(interrupt_enable_counter < UINT32_MAX); 
// FIXME
#ifndef   FEATURE_UVISOR
    if (interrupt_enable_counter > 0) {
        MBED_ASSERT(interrupts_disabled);
    }
#else
#warning "core_util_critical_section_enter needs fixing to work from unprivileged code"
#endif /* FEATURE_UVISOR */
    interrupt_enable_counter++;
}

inline bool SDFileSystem::waitReady(int timeout) // timeout (ms)
{
    char resp;
    uint32_t start = us_ticker_read();
    //Keep sending dummy clocks with DI held high until the card releases the DO line
    //m_Timer.start();
    do {
        resp = m_Spi.write(0xFF);
    //} while (resp == 0x00 && m_Timer.read_ms() < timeout);
    //m_Timer.stop();
    //m_Timer.reset();
    } while (resp == 0x00 && ((us_ticker_read() - start) < (uint32_t)(timeout*1000)) ); // ms to us
    //Return success/failure
    return (resp > 0x00);
}

Hello, Neil. I use your SDFileSystem with LPCxpress1549 and it works great. I have one suggestion to your code. Your code uses mbed Timer API for wait or time out in some place like below. <<code title=SDFileSystem.cpp>> inline bool SDFileSystem::waitReady(int timeout) // timeout (ms) { char resp; //Keep sending dummy clocks with DI held high until the card releases the DO line m_Timer.start(); do { resp = m_Spi.write(0xFF); } while (resp == 0x00 && m_Timer.read_ms() < timeout); m_Timer.stop(); m_Timer.reset(); //Return success/failure return (resp > 0x00); } <</code>> read_us code in mbed Timer API is <<code title=Timer.cpp>> int Timer::read_us() { core_util_critical_section_enter(); int time = _time + slicetime(); core_util_critical_section_exit(); return time; } <</code>> I think "core_util_critical_section_enter()" is not good in some codes with crucial interrupt timing. Because it call _disable_irq(); <<code title=mbed_critical.c>> void core_util_critical_section_enter(void) { bool interrupts_disabled = !core_util_are_interrupts_enabled(); __disable_irq(); /* Save the interrupt disabled state as it was prior to any nested critical section lock use */ if (!interrupt_enable_counter) { critical_interrupts_disabled = interrupts_disabled; } /* If the interrupt_enable_counter overflows or we are in a nested critical section and interrupts are enabled, then something has gone badly wrong thus assert an error. */ MBED_ASSERT(interrupt_enable_counter < UINT32_MAX); // FIXME #ifndef FEATURE_UVISOR if (interrupt_enable_counter > 0) { MBED_ASSERT(interrupts_disabled); } #else #warning "core_util_critical_section_enter needs fixing to work from unprivileged code" #endif /* FEATURE_UVISOR */ interrupt_enable_counter++; } <</code>> So I changed your code without Timer API like this. <<code title=SDFileSystem.cpp>> inline bool SDFileSystem::waitReady(int timeout) // timeout (ms) { char resp; uint32_t start = us_ticker_read(); //Keep sending dummy clocks with DI held high until the card releases the DO line //m_Timer.start(); do { resp = m_Spi.write(0xFF); //} while (resp == 0x00 && m_Timer.read_ms() < timeout); //m_Timer.stop(); //m_Timer.reset(); } while (resp == 0x00 && ((us_ticker_read() - start) < (uint32_t)(timeout*1000)) ); // ms to us //Return success/failure return (resp > 0x00); } <</code>> I use LPCxpresso1549 and us_ticker_read(); doesn't use _disable_irq(); , only use 1MHz SCT3 timer. I really don't know us_ticker_read(); is good solution or not in any environment. Thank you.

Hi guys... does anyone know if sd file system can support two cards at once? I am trying to initialize two sd objects sd and sd2, and the program compiles fine. When I go to run it on my nucleo, it says it cannot create the second card. Anyone run into this problem or have suggestions?

Hello... I'm trying to use this with the mbed-rtos, and it is just not happening. It can see the card - sd.card_type() and sd.disk_sectors() return the correct values. sd.opendir always returns NULL. As does fopen. I tried moving all the sd operations to a separate thread with a large (8K - 4x default) stack. Still no joy. Where should I start looking? <edit> OK, I think I found something: The example provided works with mbed versions 137 and lower The example provided does NOT work with versions 138, 139 and 140. I see comments about 'interrupt in' being deprecated (on the cs line), this may or may not be part of the problem. Near as I can tell mbed-rtos is not part of the problem. Bottom line: It (SDFileSystem) does NOT work with the latest mbed release.

Press the button to perform tests: 
Mounting SD card...success!
.Card type: SDHC
.Sectors: 62333952
.Capacity: 30436.5MB
Testing 4096B write performance...done!
.Result: 374.50KB/s
Testing 4096B read performance...done!
.Result: 564.65KB/s

Neil. Thank you for your work on the SD card library, which is a great help. I tried your Hello World program ([[https://os.mbed.com/users/neilt6/code/SDFileSystem_HelloWorld/]]) and got the following output for a SanDisk 32GB MicroSD card: <<code>> Press the button to perform tests: Mounting SD card...success! .Card type: SDHC .Sectors: 62333952 .Capacity: 30436.5MB Testing 4096B write performance...done! .Result: 374.50KB/s Testing 4096B read performance...done! .Result: 564.65KB/s <</code>> It made no difference to the performance whether or not I connected a 47K resistor between 3.3V and MISO. Please may I ask what is the value of connecting this pull-up resistor? Thanks for your advice.