openkylin
/
linux
mirror of https://gitee.com/openkylin/linux.git
9
0
Fork 0
Code Issues Projects Releases Wiki Activity
linux/arch/blackfin/kernel/cplb-nompu/cplbmgr.c

209 lines
4.6 KiB
C
Raw Normal View History

Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
/*
* Based on: arch/blackfin/kernel/cplb-mpu/cplbmgr.c
* Author: Michael McTernan <mmcternan@airvana.com>
*
* Description: CPLB miss handler.
*
* Modified:
* Copyright 2008 Airvana Inc.
Blackfin: mass clean up of copyright/licensing info Bill Gatliff & David Brownell pointed out we were missing some copyrights, and licensing terms in some of the files in ./arch/blackfin, so this fixes things, and cleans them up. It also removes: - verbose GPL text(refer to the top level ./COPYING file) - file names (you are looking at the file) - bug url (it's in the ./MAINTAINERS file) - "or later" on GPL-2, when we did not have that right It also allows some Blackfin-specific assembly files to be under a BSD like license (for people to use them outside of Linux). Signed-off-by: Robin Getz <robin.getz@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2009-09-24 22:11:24 +08:00
* Copyright 2008-2009 Analog Devices Inc.
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
*
Blackfin: mass clean up of copyright/licensing info Bill Gatliff & David Brownell pointed out we were missing some copyrights, and licensing terms in some of the files in ./arch/blackfin, so this fixes things, and cleans them up. It also removes: - verbose GPL text(refer to the top level ./COPYING file) - file names (you are looking at the file) - bug url (it's in the ./MAINTAINERS file) - "or later" on GPL-2, when we did not have that right It also allows some Blackfin-specific assembly files to be under a BSD like license (for people to use them outside of Linux). Signed-off-by: Robin Getz <robin.getz@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2009-09-24 22:11:24 +08:00
* Licensed under the GPL-2 or later
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
*/
#include <linux/kernel.h>
#include <asm/blackfin.h>
#include <asm/cplbinit.h>
#include <asm/cplb.h>
#include <asm/mmu_context.h>
Blackfin: only handle CPLB protection violations when MPU is enabled We don't need to handle CPLB protection violations unless we are running with the MPU on. Fix the entry code to call common trap_c, and remove the code which is never run. This allows the traps test suite to run on older boards with the MPU disabled. URL: http://blackfin.uclinux.org/gf/tracker/5129 Signed-off-by: Robin Getz <robin.getz@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2009-06-09 09:18:41 +08:00
#include <asm/traps.h>
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
/*
* WARNING
*
* This file is compiled with certain -ffixed-reg options. We have to
* make sure not to call any functions here that could clobber these
* registers.
*/
int nr_dcplb_miss[NR_CPUS], nr_icplb_miss[NR_CPUS];
int nr_dcplb_supv_miss[NR_CPUS], nr_icplb_supv_miss[NR_CPUS];
int nr_cplb_flush[NR_CPUS], nr_dcplb_prot[NR_CPUS];
#ifdef CONFIG_EXCPT_IRQ_SYSC_L1
#define MGR_ATTR __attribute__((l1_text))
#else
#define MGR_ATTR
#endif
static inline void write_dcplb_data(int cpu, int idx, unsigned long data,
unsigned long addr)
{
Blackfin: cleanup sync handling when enabling/disabling cplbs The handling of updating the [DI]MEM_CONTROL MMRs does not follow proper sync procedures as laid out in the Blackfin programming manual. So rather than audit/fix every call location, create helper functions that do the right things in order to safely update these MMRs. Then convert all call sites to use these new helper functions. While we're fixing the code, drop the workaround for anomaly 05000125 as that anomaly applies to old versions of silicon that we do not support. Signed-off-by: Yi Li <yi.li@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2009-08-07 09:20:58 +08:00
_disable_dcplb();
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
bfin_write32(DCPLB_DATA0 + idx * 4, data);
bfin_write32(DCPLB_ADDR0 + idx * 4, addr);
Blackfin: cleanup sync handling when enabling/disabling cplbs The handling of updating the [DI]MEM_CONTROL MMRs does not follow proper sync procedures as laid out in the Blackfin programming manual. So rather than audit/fix every call location, create helper functions that do the right things in order to safely update these MMRs. Then convert all call sites to use these new helper functions. While we're fixing the code, drop the workaround for anomaly 05000125 as that anomaly applies to old versions of silicon that we do not support. Signed-off-by: Yi Li <yi.li@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2009-08-07 09:20:58 +08:00
_enable_dcplb();
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
#ifdef CONFIG_CPLB_INFO
dcplb_tbl[cpu][idx].addr = addr;
dcplb_tbl[cpu][idx].data = data;
#endif
}
static inline void write_icplb_data(int cpu, int idx, unsigned long data,
unsigned long addr)
{
Blackfin: cleanup sync handling when enabling/disabling cplbs The handling of updating the [DI]MEM_CONTROL MMRs does not follow proper sync procedures as laid out in the Blackfin programming manual. So rather than audit/fix every call location, create helper functions that do the right things in order to safely update these MMRs. Then convert all call sites to use these new helper functions. While we're fixing the code, drop the workaround for anomaly 05000125 as that anomaly applies to old versions of silicon that we do not support. Signed-off-by: Yi Li <yi.li@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2009-08-07 09:20:58 +08:00
_disable_icplb();
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
bfin_write32(ICPLB_DATA0 + idx * 4, data);
bfin_write32(ICPLB_ADDR0 + idx * 4, addr);
Blackfin: cleanup sync handling when enabling/disabling cplbs The handling of updating the [DI]MEM_CONTROL MMRs does not follow proper sync procedures as laid out in the Blackfin programming manual. So rather than audit/fix every call location, create helper functions that do the right things in order to safely update these MMRs. Then convert all call sites to use these new helper functions. While we're fixing the code, drop the workaround for anomaly 05000125 as that anomaly applies to old versions of silicon that we do not support. Signed-off-by: Yi Li <yi.li@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2009-08-07 09:20:58 +08:00
_enable_icplb();
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
#ifdef CONFIG_CPLB_INFO
icplb_tbl[cpu][idx].addr = addr;
icplb_tbl[cpu][idx].data = data;
#endif
}
/* Counters to implement round-robin replacement. */
static int icplb_rr_index[NR_CPUS] PDT_ATTR;
static int dcplb_rr_index[NR_CPUS] PDT_ATTR;
/*
* Find an ICPLB entry to be evicted and return its index.
*/
static int evict_one_icplb(int cpu)
{
int i = first_switched_icplb + icplb_rr_index[cpu];
if (i >= MAX_CPLBS) {
i -= MAX_CPLBS - first_switched_icplb;
icplb_rr_index[cpu] -= MAX_CPLBS - first_switched_icplb;
}
icplb_rr_index[cpu]++;
return i;
}
static int evict_one_dcplb(int cpu)
{
int i = first_switched_dcplb + dcplb_rr_index[cpu];
if (i >= MAX_CPLBS) {
i -= MAX_CPLBS - first_switched_dcplb;
dcplb_rr_index[cpu] -= MAX_CPLBS - first_switched_dcplb;
}
dcplb_rr_index[cpu]++;
return i;
}
MGR_ATTR static int icplb_miss(int cpu)
{
unsigned long addr = bfin_read_ICPLB_FAULT_ADDR();
int status = bfin_read_ICPLB_STATUS();
int idx;
unsigned long i_data, base, addr1, eaddr;
nr_icplb_miss[cpu]++;
if (unlikely(status & FAULT_USERSUPV))
nr_icplb_supv_miss[cpu]++;
base = 0;
Blackfin arch: cplb mananger: use a do...while loop rather than a for loop use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-02-04 16:49:45 +08:00
idx = 0;
do {
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
eaddr = icplb_bounds[idx].eaddr;
if (addr < eaddr)
break;
base = eaddr;
Blackfin arch: cplb mananger: use a do...while loop rather than a for loop use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-02-04 16:49:45 +08:00
} while (++idx < icplb_nr_bounds);
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
if (unlikely(idx == icplb_nr_bounds))
return CPLB_NO_ADDR_MATCH;
i_data = icplb_bounds[idx].data;
if (unlikely(i_data == 0))
return CPLB_NO_ADDR_MATCH;
addr1 = addr & ~(SIZE_4M - 1);
addr &= ~(SIZE_1M - 1);
i_data |= PAGE_SIZE_1MB;
if (addr1 >= base && (addr1 + SIZE_4M) <= eaddr) {
/*
* This works because
* (PAGE_SIZE_4MB & PAGE_SIZE_1MB) == PAGE_SIZE_1MB.
*/
i_data |= PAGE_SIZE_4MB;
addr = addr1;
}
/* Pick entry to evict */
idx = evict_one_icplb(cpu);
write_icplb_data(cpu, idx, i_data, addr);
return CPLB_RELOADED;
}
MGR_ATTR static int dcplb_miss(int cpu)
{
unsigned long addr = bfin_read_DCPLB_FAULT_ADDR();
int status = bfin_read_DCPLB_STATUS();
int idx;
unsigned long d_data, base, addr1, eaddr;
nr_dcplb_miss[cpu]++;
if (unlikely(status & FAULT_USERSUPV))
nr_dcplb_supv_miss[cpu]++;
base = 0;
Blackfin arch: cplb mananger: use a do...while loop rather than a for loop use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-02-04 16:49:45 +08:00
idx = 0;
do {
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
eaddr = dcplb_bounds[idx].eaddr;
if (addr < eaddr)
break;
base = eaddr;
Blackfin arch: cplb mananger: use a do...while loop rather than a for loop use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-02-04 16:49:45 +08:00
} while (++idx < dcplb_nr_bounds);
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
if (unlikely(idx == dcplb_nr_bounds))
return CPLB_NO_ADDR_MATCH;
d_data = dcplb_bounds[idx].data;
if (unlikely(d_data == 0))
return CPLB_NO_ADDR_MATCH;
addr1 = addr & ~(SIZE_4M - 1);
addr &= ~(SIZE_1M - 1);
d_data |= PAGE_SIZE_1MB;
if (addr1 >= base && (addr1 + SIZE_4M) <= eaddr) {
/*
* This works because
* (PAGE_SIZE_4MB & PAGE_SIZE_1MB) == PAGE_SIZE_1MB.
*/
d_data |= PAGE_SIZE_4MB;
addr = addr1;
}
blackfin: cplb: add support for bf60x Bf60x support big CPLB pages, this commit enable it. Signed-off-by: Bob Liu <lliubbo@gmail.com>
2012-05-16 18:03:47 +08:00
#ifdef CONFIG_BF60x
if ((addr >= ASYNC_BANK0_BASE)
&& (addr < ASYNC_BANK3_BASE + ASYNC_BANK3_SIZE))
d_data |= PAGE_SIZE_64MB;
#endif
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
/* Pick entry to evict */
idx = evict_one_dcplb(cpu);
write_dcplb_data(cpu, idx, d_data, addr);
return CPLB_RELOADED;
}
MGR_ATTR int cplb_hdr(int seqstat, struct pt_regs *regs)
{
int cause = seqstat & 0x3f;
Blackfin: use raw_smp_processor_id() in exception code When preempt debugging is enabled, smp_processor_id() may utilize the "current" structure. This may not be safe to access under all exceptions due to it being in dynamically allocated memory. So in exception code, make sure we use raw_smp_processor_id() instead to get at the real value directly. Signed-off-by: Yi Li <yi.li@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2009-08-20 12:17:47 +08:00
unsigned int cpu = raw_smp_processor_id();
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
switch (cause) {
Blackfin: only handle CPLB protection violations when MPU is enabled We don't need to handle CPLB protection violations unless we are running with the MPU on. Fix the entry code to call common trap_c, and remove the code which is never run. This allows the traps test suite to run on older boards with the MPU disabled. URL: http://blackfin.uclinux.org/gf/tracker/5129 Signed-off-by: Robin Getz <robin.getz@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2009-06-09 09:18:41 +08:00
case VEC_CPLB_I_M:
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
return icplb_miss(cpu);
Blackfin: only handle CPLB protection violations when MPU is enabled We don't need to handle CPLB protection violations unless we are running with the MPU on. Fix the entry code to call common trap_c, and remove the code which is never run. This allows the traps test suite to run on older boards with the MPU disabled. URL: http://blackfin.uclinux.org/gf/tracker/5129 Signed-off-by: Robin Getz <robin.getz@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2009-06-09 09:18:41 +08:00
case VEC_CPLB_M:
Blackfin arch: Faster C implementation of no-MPU CPLB handler This is a mixture ofcMichael McTernan's patch and the existing cplb-mpu code. We ditch the old cplb-nompu implementation, which is a good example of why a good algorithm in a HLL is preferrable to a bad algorithm written in assembly. Rather than try to construct a table of all posible CPLBs and search it, we just create a (smaller) table of memory regions and their attributes. Some of the data structures are now unified for both the mpu and nompu cases. A lot of needless complexity in cplbinit.c is removed. Further optimizations: * compile cplbmgr.c with a lot of -ffixed-reg options, and omit saving these registers on the stack when entering a CPLB exception. * lose cli/nop/nop/sti sequences for some workarounds - these don't * make sense in an exception context Additional code unification should be possible after this. [Mike Frysinger <vapier.adi@gmail.com>: - convert CPP if statements to C if statements - remove redundant statements - use a do...while loop rather than a for loop to get slightly better optimization and to avoid gcc "may be used uninitialized" warnings ... we know that the [id]cplb_nr_bounds variables will never be 0, so this is OK - the no-mpu code was the last user of MAX_MEM_SIZE and with that rewritten, we can punt it - add some BUG_ON() checks to make sure we dont overflow the small cplb_bounds array - add i/d cplb entries for the bootrom because there is functions/data in there we want to access - we do not need a NULL trailing entry as any time we access the bounds arrays, we use the nr_bounds variable ] Signed-off-by: Michael McTernan <mmcternan@airvana.com> Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de> Signed-off-by: Bryan Wu <cooloney@kernel.org>
2009-01-07 23:14:38 +08:00
return dcplb_miss(cpu);
default:
return CPLB_UNKNOWN_ERR;
}
}