[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [RFC v3 2/4] xen/linux_compat: Add a Linux compat header




On 12/5/2017 5:31 AM, Julien Grall wrote:
> Hi Sameer,
> 
> On 05/12/17 03:59, Sameer Goel wrote:
>> For porting files from Linux it is useful to have a Linux API to Xen API
>> mapping header at a common location.
>> This file adds common API functions and other defines that are needed for
>> porting arm SMMU drivers.
>>
>> ---
>>   xen/include/xen/linux_compat.h | 106 
>> +++++++++++++++++++++++++++++++++++++++++
>>   1 file changed, 106 insertions(+)
>>   create mode 100644 xen/include/xen/linux_compat.h
>>
>> diff --git a/xen/include/xen/linux_compat.h b/xen/include/xen/linux_compat.h
>> new file mode 100644
>> index 0000000..217e0cc
>> --- /dev/null
>> +++ b/xen/include/xen/linux_compat.h
>> @@ -0,0 +1,106 @@
>> +/******************************************************************************
>> + * include/xen/linux_compat.h
>> + *
>> + * Compatibility defines for porting code from Linux to Xen
>> + *
>> + * Copyright (c) 2017 Linaro Limited
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License as published by
>> + * the Free Software Foundation; either version 2 of the License, or
>> + * (at your option) any later version.
>> + *
>> + * This program is distributed in the hope that it will be useful,
>> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>> + * GNU General Public License for more details.
>> + *
>> + * You should have received a copy of the GNU General Public License
>> + * along with this program; If not, see <http://www.gnu.org/licenses/>.
>> + */
>> +
>> +#ifndef __XEN_LINUX_COMPAT_H__
>> +#define __XEN_LINUX_COMPAT_H__
>> +
>> +#include <asm/types.h>
>> +
>> +typedef paddr_t phys_addr_t;
>> +typedef paddr_t dma_addr_t;
>> +
>> +/* Alias to Xen device tree helpers */
>> +#define device_node dt_device_node
>> +#define of_phandle_args dt_phandle_args
>> +#define of_device_id dt_device_match
>> +#define of_match_node dt_match_node
>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, 
>> pname, out))
>> +#define of_property_read_bool dt_property_read_bool
>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
>> +/* The user should consider if it is safe to treat mutex as a spinlock */
> 
> I am against defining mutex as spinlock in a generic header. People will 
> overlook it and it is hardly going to be detected in a verbatim port.
> 
> This should be done on the case by case basis.
> 
>> +#define mutex spinlock_t
>> +#define mutex_init spin_lock_init
>> +#define mutex_lock spin_lock
>> +#define mutex_unlock spin_unlock
>> +
>> +#define ilog2 LOG_2
> 
> There is only one user of LOG_2 in Xen. So wouldn't it be better to rename 
> directly to ilog2?

Its used in a couple of places (x86_64/asm-offsets.c). I can change that file 
too, let me know what you think. I am keeping this for now.
>  
>> +
>> +#define readx_poll_timeout(op, addr, val, cond, sleep_us, timeout_us) \
>> +({ \
>> +    s_time_t deadline = NOW() + MICROSECS(timeout_us); \
>> +    for (;;) \
>> +    { \
>> +        (val) = op(addr); \
>> +        if ( cond ) \
>> +            break; \
>> +        if ( NOW() > deadline ) \
>> +        { \
>> +            (val) = op(addr); \
>> +            break; \
>> +        } \
>> +        udelay(sleep_us); \
>> +    } \
>> +    (cond) ? 0 : -ETIMEDOUT; \
>> +})
>> +
>> +#define readl_relaxed_poll_timeout(addr, val, cond, delay_us, timeout_us) \
>> +    readx_poll_timeout(readl_relaxed, addr, val, cond, delay_us, timeout_us)
> 
> I don't think putting read* macros in a common header is necessary. Their use 
> in Linux is very limited.
> 
>> +
>> +/* Xen: Helpers for IRQ functions */
>> +#define request_irq(irq, func, flags, name, dev) request_irq(irq, flags, 
>> func, name, dev)
>> +#define free_irq release_irq
>> +
>> +enum irqreturn {
>> +    IRQ_NONE    = (0 << 0),
>> +    IRQ_HANDLED = (1 << 0),
>> +    IRQ_WAKE_THREAD = (2 << 0),
>> +};
>> +
>> +typedef enum irqreturn irqreturn_t;
>> +
>> +/* Device logger functions */
>> +#define dev_print(dev, lvl, fmt, ...)                       \
>> +     printk(lvl fmt, ## __VA_ARGS__)
>> +
>> +#define dev_dbg(dev, fmt, ...) dev_print(dev, XENLOG_DEBUG, fmt, ## 
>> __VA_ARGS__)
>> +#define dev_notice(dev, fmt, ...) dev_print(dev, XENLOG_INFO, fmt, ## 
>> __VA_ARGS__)
>> +#define dev_warn(dev, fmt, ...) dev_print(dev, XENLOG_WARNING, fmt, ## 
>> __VA_ARGS__)
>> +#define dev_err(dev, fmt, ...) dev_print(dev, XENLOG_ERR, fmt, ## 
>> __VA_ARGS__)
>> +#define dev_info(dev, fmt, ...) dev_print(dev, XENLOG_INFO, fmt, ## 
>> __VA_ARGS__)
>> +
>> +#define dev_err_ratelimited(dev, fmt, ...)                  \
>> +     dev_print(dev, XENLOG_ERR, fmt, ## __VA_ARGS__)
>> +
>> +#define dev_name(dev) dt_node_full_name(dev_to_dt(dev))
>> +
>> +/* Alias to Xen allocation helpers */
>> +#define kfree xfree
>> +#define kmalloc(size, flags)        _xmalloc(size, sizeof(void *))
>> +#define kzalloc(size, flags)        _xzalloc(size, sizeof(void *))
>> +#define devm_kzalloc(dev, size, flags)  _xzalloc(size, sizeof(void *))
>> +#define kmalloc_array(size, n, flags)   _xmalloc_array(size, sizeof(void 
>> *), n)
>> +
>> +/* Alias to Xen time functions */
>> +#define ktime_t s_time_t
>> +#define ktime_add_us(t,i)       (NOW() + MICROSECS(i))
>> +#define ktime_compare(t,i)      (NOW() > (i))
>> +
>> +#endif /* __XEN_LINUX_COMPAT_H__ */
>>
> 
> Cheers,
> 

-- 
 Qualcomm Datacenter Technologies as an affiliate of Qualcomm Technologies, 
Inc. Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux 
Foundation Collaborative Project.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.