Re: [RFC PATCH 8/17] input: RMI4 F09 Built-In Self Test
flat view
From: Linus Walleij <hidden>
Date: 2012-08-27 22:07:52
Also in:
lkml
On Fri, Aug 17, 2012 at 3:17 PM, Christopher Heiny [off-list ref] wrote: Put in a verbose description of what this is. (...)
quoted hunk ↗ jump to hunk
+++ b/drivers/input/rmi4/rmi_f09.c
quoted hunk ↗ jump to hunk
+/* data specific to fn $09 that needs to be kept around */ +struct f09_query { + u8 limit_register_count; + union { + struct { + u8 result_register_count:3; + u8 reserved:3; + u8 internal_limits:1; + u8 host_test_enable:1; + }; + u8 f09_bist_query1; + }; +};
__attribute__((packed)); ?
quoted hunk ↗ jump to hunk
+struct f09_control { + union { + struct { + u8 test1_limit_low:8; + u8 test1_limit_high:8; + u8 test1_limit_diff:8; + }; + u8 f09_control_test1[3]; + }; + union { + struct { + u8 test2_limit_low:8; + u8 test2_limit_high:8; + u8 test2_limit_diff:8; + }; + u8 f09_control_test2[3]; + }; +};
__attribute__((packed)); ?
quoted hunk ↗ jump to hunk
+struct f09_data { + u8 test_number_control; + u8 overall_bist_result; + u8 test_result1; + u8 test_result2; + u8 transmitter_number; + + union { + struct { + u8 receiver_number:6; + u8 limit_failure_code:2; + }; + u8 f09_bist_data2; + }; +};
__attribute__((packed)); ?
quoted hunk ↗ jump to hunk
+struct f09_cmd { + union { + struct { + u8 run_bist:1; + }; + u8 f09_bist_cmd0; + }; +};
__attribute__((packed)); ? (...)
quoted hunk ↗ jump to hunk
+static struct device_attribute attrs[] = { + __ATTR(status, RMI_RW_ATTR, + rmi_f09_status_show, rmi_f09_status_store), + __ATTR(limitRegisterCount, RMI_RO_ATTR, + rmi_f09_limit_register_count_show, rmi_store_error), + __ATTR(hostTestEnable, RMI_RW_ATTR, + rmi_f09_host_test_enable_show, rmi_f09_host_test_enable_store), + __ATTR(internalLimits, RMI_RO_ATTR, + rmi_f09_internal_limits_show, rmi_store_error), + __ATTR(resultRegisterCount, RMI_RO_ATTR, + rmi_f09_result_register_count_show, rmi_store_error), + __ATTR(overall_bist_result, RMI_RO_ATTR, + rmi_f09_overall_bist_result_show, rmi_store_error), + __ATTR(test_number_control, RMI_RW_ATTR, + rmi_f09_test_number_control_show, + rmi_f09_test_number_control_store), + __ATTR(test_result1, RMI_RO_ATTR, + rmi_f09_test_result1_show, rmi_store_error), + __ATTR(test_result2, RMI_RO_ATTR, + rmi_f09_test_result2_show, rmi_store_error), + __ATTR(run_bist, RMI_RW_ATTR, + rmi_f09_run_bist_show, rmi_f09_run_bist_store), + __ATTR(f09_control_test1, RMI_RW_ATTR, + rmi_f09_control_test1_show, rmi_f09_control_test1_store), + __ATTR(f09_control_test2, RMI_RW_ATTR, + rmi_f09_control_test2_show, rmi_f09_control_test2_store), +};
If this is *only* for tests, then for sure this should be in debugfs?
+static int rmi_f09_alloc_memory(struct rmi_function_container *fc)
(...)
+static void rmi_f09_free_memory(struct rmi_function_container *fc)
Why do you need separate functions for these two?
If they are only used from one place (which I suspect) then just
put the code at that site.
(...)quoted hunk ↗ jump to hunk
+static int rmi_f09_initialize(struct rmi_function_container *fc) +{ + struct rmi_device *rmi_dev = fc->rmi_dev; + struct rmi_device_platform_data *pdata; + struct rmi_fn_09_data *f09 = fc->data; + u16 query_base_addr; + int rc; + + + pdata = to_rmi_platform_data(rmi_dev); + query_base_addr = fc->fd.query_base_addr; + + /* initial all default values for f09 query here */ + rc = rmi_read_block(rmi_dev, query_base_addr, + (u8 *)&f09->query, sizeof(f09->query)); + if (rc < 0) { + dev_err(&fc->dev, "Failed to read query register." + " from 0x%04x\n", query_base_addr); + return rc; + } + + return 0; +}
Similar here. Cannot this be brought into the only call site?
quoted hunk ↗ jump to hunk
+static int rmi_f09_config(struct rmi_function_container *fc) +{ + /*we do nothing here. instead reset should notify the user.*/ + return 0; +}
Make it optional and just don't define it.
quoted hunk ↗ jump to hunk
+static int rmi_f09_reset(struct rmi_function_container *fc) +{ + struct rmi_fn_09_data *instance_data = fc->data; + + instance_data->status = -ECONNRESET; + + return 0; +}
Dito. (Already remarked this at the last patch.) Yours, Linus Walleij