Re: Please report status of tested systems for sane-pre1-1.0.5

From: Paul Floyd (paulf@free.fr)
Date: Sat Jun 30 2001 - 15:29:18 PDT

  • Next message: G.LeeJ: "sane/canoscanFB620U"

    Oliver Rauch wrote:
    >
    > Hello.
    >
    > To make the "supported platforms" list for sane-1.0.5 up to date:
    > Please could you report the status of all systems you tried
    > to compile sane-pre1-1.0.5 on?

    Not working:

    OS/2 4, snapscan 1236s, emx (gcc 2.8.1)

    This code:

    static SANE_Status read_calibration_data (SnapScan_Scanner *pss, void
    *buf, u_char num_lines)
    {
        static const char *me = "read_calibration_data";
        SANE_Status status;
        size_t expected_read_bytes = num_lines *
    calibration_line_length(pss);
        size_t read_bytes;

        DBG (DL_CALL_TRACE, "%s\n", me);
        zero_buf (pss->cmd, MAX_SCSI_CMD_LEN);
        pss->cmd[0] = READ;
        pss->cmd[2] = READ_CALIBRATION;
        pss->cmd[5] = num_lines;
        u_int_to_u_char3p (expected_read_bytes, pss->cmd + 6);
        read_bytes = expected_read_bytes;

        status = snapscan_cmd (pss->pdev->bus, pss->fd, pss->cmd,
                                 READ_LEN, buf, &read_bytes);

    tries to call snapscan_cmd to obtain 244800 bytes, but the maximum that
    sanei_scsi_cmd2 (USE_OS2) will cope with is 32K. This either causes
    a sigsegv or an assertion error (depending on whether debug mode or
    not).

    Increasing the value of MAX_DATA causes the aspi driver to trap the OS.

    A bientot
    Paul

    -- 
    Paul Floyd                 http://paulf.free.fr (for what it's worth)
                               MailTo:paulf@free.fr
    If more is better, are double standards better than single ones?
    

    -- Source code, list archive, and docs: http://www.mostang.com/sane/ To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com



    This archive was generated by hypermail 2b29 : Sat Jun 30 2001 - 15:11:12 PDT