linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 00/14] alpha: cleanups for 6.10
@ 2024-05-03  8:11 Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 01/14] alpha: sort scr_mem{cpy,move}w() out Arnd Bergmann
                   ` (15 more replies)
  0 siblings, 16 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Arnd Bergmann <arnd@arndb.de>

I had investigated dropping support for alpha EV5 and earlier a while
ago after noticing that this is the only supported CPU family
in the kernel without native byte access and that Debian has already
dropped support for this generation last year [1] after it turned
out to be broken.

This topic came up again when Paul E. McKenney noticed that
parts of the RCU code already rely on byte access and do not
work on alpha EV5 reliably, so I refreshed my series now for
inclusion into the next merge window.

Al Viro did another series for alpha to address all the known build
issues. I rebased his patches without any further changes and included
it as a baseline for my work here to avoid conflicts.

      Arnd

[1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1036158
[2] https://lore.kernel.org/lkml/b67e79d4-06cb-4a45-a906-b9e0fbae22c5@paulmck-laptop/

Al Viro (9):
  alpha: sort scr_mem{cpy,move}w() out
  alpha: fix modversions for strcpy() et.al.
  alpha: add clone3() support
  alpha: don't make functions public without a reason
  alpha: sys_sio: fix misspelled ifdefs
  alpha: missing includes
  alpha: core_lca: take the unused functions out
  alpha: jensen, t2 - make __EXTERN_INLINE same as for the rest
  alpha: trim the unused stuff from asm-offsets.c

Arnd Bergmann (5):
  alpha: remove DECpc AXP150 (Jensen) support
  alpha: sable: remove early machine support
  alpha: remove LCA and APECS based machines
  alpha: cabriolet: remove EV5 CPU support
  alpha: drop pre-EV56 support

 Documentation/driver-api/eisa.rst      |   4 +-
 arch/alpha/Kconfig                     | 175 +-------
 arch/alpha/Makefile                    |   8 +-
 arch/alpha/include/asm/core_apecs.h    | 534 -------------------------
 arch/alpha/include/asm/core_lca.h      | 378 -----------------
 arch/alpha/include/asm/core_t2.h       |   8 -
 arch/alpha/include/asm/dma-mapping.h   |   4 -
 arch/alpha/include/asm/dma.h           |   9 +-
 arch/alpha/include/asm/elf.h           |   4 +-
 arch/alpha/include/asm/io.h            |  26 +-
 arch/alpha/include/asm/irq.h           |  10 +-
 arch/alpha/include/asm/jensen.h        | 363 -----------------
 arch/alpha/include/asm/machvec.h       |   9 -
 arch/alpha/include/asm/mmu_context.h   |  45 +--
 arch/alpha/include/asm/special_insns.h |   5 +-
 arch/alpha/include/asm/tlbflush.h      |  41 +-
 arch/alpha/include/asm/uaccess.h       |  80 ----
 arch/alpha/include/asm/vga.h           |   2 +
 arch/alpha/include/uapi/asm/compiler.h |  18 -
 arch/alpha/kernel/Makefile             |  25 +-
 arch/alpha/kernel/asm-offsets.c        |  21 +-
 arch/alpha/kernel/bugs.c               |   1 +
 arch/alpha/kernel/console.c            |   1 +
 arch/alpha/kernel/core_apecs.c         | 420 -------------------
 arch/alpha/kernel/core_cia.c           |   6 +-
 arch/alpha/kernel/core_irongate.c      |   1 -
 arch/alpha/kernel/core_lca.c           | 517 ------------------------
 arch/alpha/kernel/core_marvel.c        |   2 +-
 arch/alpha/kernel/core_t2.c            |   2 +-
 arch/alpha/kernel/core_wildfire.c      |   8 +-
 arch/alpha/kernel/entry.S              |   1 +
 arch/alpha/kernel/io.c                 |  19 +
 arch/alpha/kernel/irq.c                |   1 +
 arch/alpha/kernel/irq_i8259.c          |   4 -
 arch/alpha/kernel/machvec_impl.h       |  25 +-
 arch/alpha/kernel/pci-noop.c           | 113 ------
 arch/alpha/kernel/pci_impl.h           |   4 +-
 arch/alpha/kernel/perf_event.c         |   2 +-
 arch/alpha/kernel/proto.h              |  44 +-
 arch/alpha/kernel/setup.c              | 109 +----
 arch/alpha/kernel/smc37c669.c          |   6 +-
 arch/alpha/kernel/smc37c93x.c          |   2 +
 arch/alpha/kernel/smp.c                |   1 +
 arch/alpha/kernel/srmcons.c            |   2 +
 arch/alpha/kernel/sys_cabriolet.c      |  87 +---
 arch/alpha/kernel/sys_eb64p.c          | 238 -----------
 arch/alpha/kernel/sys_jensen.c         | 237 -----------
 arch/alpha/kernel/sys_mikasa.c         |  57 ---
 arch/alpha/kernel/sys_nautilus.c       |   8 +-
 arch/alpha/kernel/sys_noritake.c       |  60 ---
 arch/alpha/kernel/sys_sable.c          | 294 +-------------
 arch/alpha/kernel/sys_sio.c            | 486 ----------------------
 arch/alpha/kernel/syscalls/syscall.tbl |   2 +-
 arch/alpha/kernel/traps.c              |  64 ---
 arch/alpha/lib/Makefile                |  14 -
 arch/alpha/lib/checksum.c              |   1 +
 arch/alpha/lib/fpreg.c                 |   1 +
 arch/alpha/lib/memcpy.c                |   3 +
 arch/alpha/lib/stycpy.S                |  11 +
 arch/alpha/lib/styncpy.S               |  11 +
 arch/alpha/math-emu/math.c             |   7 +-
 arch/alpha/mm/init.c                   |   2 +-
 drivers/char/agp/alpha-agp.c           |   2 +-
 drivers/eisa/Kconfig                   |   9 +-
 drivers/eisa/virtual_root.c            |   2 +-
 drivers/input/serio/i8042-io.h         |   5 +-
 drivers/tty/serial/8250/8250.h         |   3 -
 drivers/tty/serial/8250/8250_alpha.c   |  21 -
 drivers/tty/serial/8250/8250_core.c    |   4 -
 drivers/tty/serial/8250/Makefile       |   2 -
 include/linux/blk_types.h              |   6 -
 include/linux/tty.h                    |  14 +-
 72 files changed, 166 insertions(+), 4545 deletions(-)
 delete mode 100644 arch/alpha/include/asm/core_apecs.h
 delete mode 100644 arch/alpha/include/asm/core_lca.h
 delete mode 100644 arch/alpha/include/asm/jensen.h
 delete mode 100644 arch/alpha/kernel/core_apecs.c
 delete mode 100644 arch/alpha/kernel/core_lca.c
 delete mode 100644 arch/alpha/kernel/pci-noop.c
 delete mode 100644 arch/alpha/kernel/sys_eb64p.c
 delete mode 100644 arch/alpha/kernel/sys_jensen.c
 delete mode 100644 arch/alpha/kernel/sys_sio.c
 create mode 100644 arch/alpha/lib/stycpy.S
 create mode 100644 arch/alpha/lib/styncpy.S
 delete mode 100644 drivers/tty/serial/8250/8250_alpha.c

-- 
2.39.2

Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: "Paul E. McKenney" <paulmck@kernel.org>
Cc: linux-alpha@vger.kernel.org
Cc: linux-kernel@vger.kernel.org

^ permalink raw reply	[flat|nested] 55+ messages in thread

* [PATCH 01/14] alpha: sort scr_mem{cpy,move}w() out
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 02/14] alpha: fix modversions for strcpy() et.al Arnd Bergmann
                   ` (14 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Al Viro <viro@zeniv.linux.org.uk>

Take scr_memmove() out of line, make both it and scr_memcpyw()
conditional upon VGA_CONSOLE or MDA_CONSOLE (if neither is
selected, we are certain to be working with the kernel-allocated
buffer rather than VRAM and defaults will work just fine).

That allows to clean vt_buffer.h, but that's a separate story

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/include/asm/vga.h |  2 ++
 arch/alpha/kernel/io.c       | 19 +++++++++++++++++++
 2 files changed, 21 insertions(+)

diff --git a/arch/alpha/include/asm/vga.h b/arch/alpha/include/asm/vga.h
index 4c347a8454c7..919931cb5b63 100644
--- a/arch/alpha/include/asm/vga.h
+++ b/arch/alpha/include/asm/vga.h
@@ -13,6 +13,7 @@
 #define VT_BUF_HAVE_RW
 #define VT_BUF_HAVE_MEMSETW
 #define VT_BUF_HAVE_MEMCPYW
+#define VT_BUF_HAVE_MEMMOVEW
 
 static inline void scr_writew(u16 val, volatile u16 *addr)
 {
@@ -40,6 +41,7 @@ static inline void scr_memsetw(u16 *s, u16 c, unsigned int count)
 
 /* Do not trust that the usage will be correct; analyze the arguments.  */
 extern void scr_memcpyw(u16 *d, const u16 *s, unsigned int count);
+extern void scr_memmovew(u16 *d, const u16 *s, unsigned int count);
 
 /* ??? These are currently only used for downloading character sets.  As
    such, they don't need memory barriers.  Is this all they are intended
diff --git a/arch/alpha/kernel/io.c b/arch/alpha/kernel/io.c
index eda09778268f..c28035d6d1e6 100644
--- a/arch/alpha/kernel/io.c
+++ b/arch/alpha/kernel/io.c
@@ -647,6 +647,10 @@ void _memset_c_io(volatile void __iomem *to, unsigned long c, long count)
 
 EXPORT_SYMBOL(_memset_c_io);
 
+#if IS_ENABLED(CONFIG_VGA_CONSOLE) || IS_ENABLED(CONFIG_MDA_CONSOLE)
+
+#include <asm/vga.h>
+
 /* A version of memcpy used by the vga console routines to move data around
    arbitrarily between screen and main memory.  */
 
@@ -681,6 +685,21 @@ scr_memcpyw(u16 *d, const u16 *s, unsigned int count)
 
 EXPORT_SYMBOL(scr_memcpyw);
 
+void scr_memmovew(u16 *d, const u16 *s, unsigned int count)
+{
+	if (d < s)
+		scr_memcpyw(d, s, count);
+	else {
+		count /= 2;
+		d += count;
+		s += count;
+		while (count--)
+			scr_writew(scr_readw(--s), --d);
+	}
+}
+EXPORT_SYMBOL(scr_memmovew);
+#endif
+
 void __iomem *ioport_map(unsigned long port, unsigned int size)
 {
 	return IO_CONCAT(__IO_PREFIX,ioportmap) (port);
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 02/14] alpha: fix modversions for strcpy() et.al.
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 01/14] alpha: sort scr_mem{cpy,move}w() out Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 03/14] alpha: add clone3() support Arnd Bergmann
                   ` (13 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Al Viro <viro@zeniv.linux.org.uk>

        On alpha str{n,}{cpy,cat}() implementations are playing
fun games with shared chunks of code.  The problem is, they are
using direct branches and need to be next to each other.
        Currently it's done by building them in separate object
files, then using ld -r to link those together.  Unfortunately,
genksyms machinery has no idea what to do with that - we have
generated in arch/alpha/lib/.strcat.S.cmd, but there's nothing
to propagate that into .stycpy.S.cmd, so modpost doesn't find
anything for those symbols, resulting in
WARNING: modpost: EXPORT symbol "strcpy" [vmlinux] version generation failed, symbol will not be versioned.
Is "strcpy" prototyped in <asm/asm-prototypes.h>?
WARNING: modpost: EXPORT symbol "strcat" [vmlinux] version generation failed, symbol will not be versioned.
Is "strcat" prototyped in <asm/asm-prototypes.h>?
WARNING: modpost: EXPORT symbol "strncpy" [vmlinux] version generation failed, symbol will not be versioned.
Is "strncpy" prototyped in <asm/asm-prototypes.h>?
WARNING: modpost: EXPORT symbol "strncat" [vmlinux] version generation failed, symbol will not be versioned.
Is "strncat" prototyped in <asm/asm-prototypes.h>?
spew on modversion-enabled builds (all 4 functions in question
are in fact prototyped in asm-prototypes.h)

        Fixing doesn't require messing with kbuild, thankfully -
just build one object (i.e. have sty{n,}cpy.S with includes of relevant
*.S instead of playing with ld -r) and that's it.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/lib/Makefile  | 14 --------------
 arch/alpha/lib/stycpy.S  | 11 +++++++++++
 arch/alpha/lib/styncpy.S | 11 +++++++++++
 3 files changed, 22 insertions(+), 14 deletions(-)
 create mode 100644 arch/alpha/lib/stycpy.S
 create mode 100644 arch/alpha/lib/styncpy.S

diff --git a/arch/alpha/lib/Makefile b/arch/alpha/lib/Makefile
index 6a779b9018fd..84046e730e6d 100644
--- a/arch/alpha/lib/Makefile
+++ b/arch/alpha/lib/Makefile
@@ -44,17 +44,3 @@ AFLAGS___remlu.o =       -DREM -DINTSIZE
 $(addprefix $(obj)/,__divqu.o __remqu.o __divlu.o __remlu.o): \
 						$(src)/$(ev6-y)divide.S FORCE
 	$(call if_changed_rule,as_o_S)
-
-# There are direct branches between {str*cpy,str*cat} and stx*cpy.
-# Ensure the branches are within range by merging these objects.
-
-LDFLAGS_stycpy.o := -r
-LDFLAGS_styncpy.o := -r
-
-$(obj)/stycpy.o: $(obj)/strcpy.o $(obj)/$(ev67-y)strcat.o \
-		 $(obj)/$(ev6-y)stxcpy.o FORCE
-	$(call if_changed,ld)
-
-$(obj)/styncpy.o: $(obj)/strncpy.o $(obj)/$(ev67-y)strncat.o \
-		 $(obj)/$(ev6-y)stxncpy.o FORCE
-	$(call if_changed,ld)
diff --git a/arch/alpha/lib/stycpy.S b/arch/alpha/lib/stycpy.S
new file mode 100644
index 000000000000..32ecd9c5f90d
--- /dev/null
+++ b/arch/alpha/lib/stycpy.S
@@ -0,0 +1,11 @@
+#include "strcpy.S"
+#ifdef CONFIG_ALPHA_EV67
+#include "ev67-strcat.S"
+#else
+#include "strcat.S"
+#endif
+#ifdef CONFIG_ALPHA_EV6
+#include "ev6-stxcpy.S"
+#else
+#include "stxcpy.S"
+#endif
diff --git a/arch/alpha/lib/styncpy.S b/arch/alpha/lib/styncpy.S
new file mode 100644
index 000000000000..72fc2754eb57
--- /dev/null
+++ b/arch/alpha/lib/styncpy.S
@@ -0,0 +1,11 @@
+#include "strncpy.S"
+#ifdef CONFIG_ALPHA_EV67
+#include "ev67-strncat.S"
+#else
+#include "strncat.S"
+#endif
+#ifdef CONFIG_ALPHA_EV6
+#include "ev6-stxncpy.S"
+#else
+#include "stxncpy.S"
+#endif
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 03/14] alpha: add clone3() support
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 01/14] alpha: sort scr_mem{cpy,move}w() out Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 02/14] alpha: fix modversions for strcpy() et.al Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 04/14] alpha: don't make functions public without a reason Arnd Bergmann
                   ` (12 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Al Viro <viro@zeniv.linux.org.uk>

Since clone3() needs the full register state saved for copying into
the child, it needs the same kind of wrapper as fork(), vfork() and
clone().  Exact same wrapper works, actually...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/kernel/entry.S              | 1 +
 arch/alpha/kernel/syscalls/syscall.tbl | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/alpha/kernel/entry.S b/arch/alpha/kernel/entry.S
index eb51f93a70c8..dd26062d75b3 100644
--- a/arch/alpha/kernel/entry.S
+++ b/arch/alpha/kernel/entry.S
@@ -811,6 +811,7 @@ alpha_\name:
 fork_like fork
 fork_like vfork
 fork_like clone
+fork_like clone3
 
 .macro	sigreturn_like name
 	.align	4
diff --git a/arch/alpha/kernel/syscalls/syscall.tbl b/arch/alpha/kernel/syscalls/syscall.tbl
index 8ff110826ce2..26cce7e7f70b 100644
--- a/arch/alpha/kernel/syscalls/syscall.tbl
+++ b/arch/alpha/kernel/syscalls/syscall.tbl
@@ -474,7 +474,7 @@
 542	common	fsmount				sys_fsmount
 543	common	fspick				sys_fspick
 544	common	pidfd_open			sys_pidfd_open
-# 545 reserved for clone3
+545	common	clone3				alpha_clone3
 546	common	close_range			sys_close_range
 547	common	openat2				sys_openat2
 548	common	pidfd_getfd			sys_pidfd_getfd
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 04/14] alpha: don't make functions public without a reason
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (2 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 03/14] alpha: add clone3() support Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 05/14] alpha: sys_sio: fix misspelled ifdefs Arnd Bergmann
                   ` (11 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Al Viro <viro@zeniv.linux.org.uk>

if it's really used only inside the same source file, make it
static...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/kernel/core_cia.c      | 6 +++---
 arch/alpha/kernel/core_marvel.c   | 2 +-
 arch/alpha/kernel/core_wildfire.c | 8 ++++----
 arch/alpha/kernel/perf_event.c    | 2 +-
 arch/alpha/kernel/smc37c669.c     | 4 +++-
 arch/alpha/kernel/sys_nautilus.c  | 6 +++---
 arch/alpha/kernel/sys_sio.c       | 2 +-
 drivers/char/agp/alpha-agp.c      | 2 +-
 8 files changed, 17 insertions(+), 15 deletions(-)

diff --git a/arch/alpha/kernel/core_cia.c b/arch/alpha/kernel/core_cia.c
index 12926e9538b8..ca3d9c732b61 100644
--- a/arch/alpha/kernel/core_cia.c
+++ b/arch/alpha/kernel/core_cia.c
@@ -280,7 +280,7 @@ cia_pci_tbi(struct pci_controller *hose, dma_addr_t start, dma_addr_t end)
 #define CIA_BROKEN_TBIA_SIZE	1024
 
 /* Always called with interrupts disabled */
-void
+static void
 cia_pci_tbi_try2(struct pci_controller *hose,
 		 dma_addr_t start, dma_addr_t end)
 {
@@ -576,7 +576,7 @@ struct
     } window[4];
 } saved_config __attribute((common));
 
-void
+static void
 cia_save_srm_settings(int is_pyxis)
 {
 	int i;
@@ -602,7 +602,7 @@ cia_save_srm_settings(int is_pyxis)
 	mb();
 }
 
-void
+static void
 cia_restore_srm_settings(void)
 {
 	int i;
diff --git a/arch/alpha/kernel/core_marvel.c b/arch/alpha/kernel/core_marvel.c
index e9348aec4649..b22248044bf0 100644
--- a/arch/alpha/kernel/core_marvel.c
+++ b/arch/alpha/kernel/core_marvel.c
@@ -355,7 +355,7 @@ marvel_init_io7(struct io7 *io7)
 	}
 }
 
-void __init
+static void __init
 marvel_io7_present(gct6_node *node)
 {
 	int pe;
diff --git a/arch/alpha/kernel/core_wildfire.c b/arch/alpha/kernel/core_wildfire.c
index 3a804b67f9da..8dd08c5e4270 100644
--- a/arch/alpha/kernel/core_wildfire.c
+++ b/arch/alpha/kernel/core_wildfire.c
@@ -59,7 +59,7 @@ unsigned long wildfire_pca_mask;
 unsigned long wildfire_cpu_mask;
 unsigned long wildfire_mem_mask;
 
-void __init
+static void __init
 wildfire_init_hose(int qbbno, int hoseno)
 {
 	struct pci_controller *hose;
@@ -137,7 +137,7 @@ wildfire_init_hose(int qbbno, int hoseno)
 	wildfire_pci_tbi(hose, 0, 0); /* Flush TLB at the end. */
 }
 
-void __init
+static void __init
 wildfire_init_pca(int qbbno, int pcano)
 {
 
@@ -154,7 +154,7 @@ wildfire_init_pca(int qbbno, int pcano)
 	wildfire_init_hose(qbbno, (pcano << 1) + 1);
 }
 
-void __init
+static void __init
 wildfire_init_qbb(int qbbno)
 {
 	int pcano;
@@ -176,7 +176,7 @@ wildfire_init_qbb(int qbbno)
 	}
 }
 
-void __init
+static void __init
 wildfire_hardware_probe(void)
 {
 	unsigned long temp;
diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c
index ccdb508c1516..1f0eb4f25c0f 100644
--- a/arch/alpha/kernel/perf_event.c
+++ b/arch/alpha/kernel/perf_event.c
@@ -870,7 +870,7 @@ static void alpha_perf_event_irq_handler(unsigned long la_ptr,
 /*
  * Init call to initialise performance events at kernel startup.
  */
-int __init init_hw_perf_events(void)
+static int __init init_hw_perf_events(void)
 {
 	pr_info("Performance events: ");
 
diff --git a/arch/alpha/kernel/smc37c669.c b/arch/alpha/kernel/smc37c669.c
index bbbd34586de0..cb8726b667da 100644
--- a/arch/alpha/kernel/smc37c669.c
+++ b/arch/alpha/kernel/smc37c669.c
@@ -2430,13 +2430,15 @@ int __init smcc669_write( struct FILE *fp, int size, int number, unsigned char *
 }
 #endif
 
-void __init
+#if SMC_DEBUG
+static void __init
 SMC37c669_dump_registers(void)
 {
   int i;
   for (i = 0; i <= 0x29; i++)
     printk("-- CR%02x : %02x\n", i, SMC37c669_read_config(i));
 }
+#endif
 /*+
  * ============================================================================
  * = SMC_init - SMC37c669 Super I/O controller initialization                 =
diff --git a/arch/alpha/kernel/sys_nautilus.c b/arch/alpha/kernel/sys_nautilus.c
index 96fd6ff3fe81..4dcee81606a5 100644
--- a/arch/alpha/kernel/sys_nautilus.c
+++ b/arch/alpha/kernel/sys_nautilus.c
@@ -78,7 +78,7 @@ nautilus_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
 	return irq;
 }
 
-void
+static void
 nautilus_kill_arch(int mode)
 {
 	struct pci_bus *bus = pci_isa_hose->bus;
@@ -127,7 +127,7 @@ naut_sys_machine_check(unsigned long vector, unsigned long la_ptr,
 /* Machine checks can come from two sources - those on the CPU and those
    in the system.  They are analysed separately but all starts here.  */
 
-void
+static void
 nautilus_machine_check(unsigned long vector, unsigned long la_ptr)
 {
 	char *mchk_class;
@@ -197,7 +197,7 @@ static struct resource busn_resource = {
 	.flags	= IORESOURCE_BUS,
 };
 
-void __init
+static void __init
 nautilus_init_pci(void)
 {
 	struct pci_controller *hose = hose_head;
diff --git a/arch/alpha/kernel/sys_sio.c b/arch/alpha/kernel/sys_sio.c
index 086488ed83a7..0bf38f868665 100644
--- a/arch/alpha/kernel/sys_sio.c
+++ b/arch/alpha/kernel/sys_sio.c
@@ -310,7 +310,7 @@ alphabook1_init_pci(void)
 	outb(0x0f, 0x3ce); outb(orig, 0x3cf); /* (re)lock PR0-4 */
 }
 
-void
+static void
 sio_kill_arch(int mode)
 {
 #if defined(ALPHA_RESTORE_SRM_SETUP)
diff --git a/drivers/char/agp/alpha-agp.c b/drivers/char/agp/alpha-agp.c
index c9bf2c219841..f0d0c044731c 100644
--- a/drivers/char/agp/alpha-agp.c
+++ b/drivers/char/agp/alpha-agp.c
@@ -149,7 +149,7 @@ struct agp_bridge_driver alpha_core_agp_driver = {
 
 struct agp_bridge_data *alpha_bridge;
 
-int __init
+static int __init
 alpha_core_agp_setup(void)
 {
 	alpha_agp_info *agp = alpha_mv.agp_info();
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 05/14] alpha: sys_sio: fix misspelled ifdefs
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (3 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 04/14] alpha: don't make functions public without a reason Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 06/14] alpha: missing includes Arnd Bergmann
                   ` (10 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Al Viro <viro@zeniv.linux.org.uk>

definitions of avanti_mv and noname_mv (and associated ALIAS_MV)
are conditional upon the wrong thing - it should be
CONFIG_ALPHA_{AVANTI,NONAME}_CH, not CONFIG_ALPHA_{AVANTI,NONAME}.

The former is a system type; the latter is for the bits shared
by AVANTI with XL and NONAME with ALPHA_BOOK1 resp.

We want all those machine vectors defined (but not aliased - see
ALIAS_MV() definition for details) for GENERIC build; for
system-specfic builds we want only one mv, so avanti_mv should *not*
be there for XL; it certainly should not be have alpha_mv aliased to
it on such config - xl_mv will be there and alpha_mv can't be aliased
to both of those.

The same goes for Noname vs. Alphabook1.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/kernel/sys_sio.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/alpha/kernel/sys_sio.c b/arch/alpha/kernel/sys_sio.c
index 0bf38f868665..c28a6ae1d989 100644
--- a/arch/alpha/kernel/sys_sio.c
+++ b/arch/alpha/kernel/sys_sio.c
@@ -360,7 +360,7 @@ struct alpha_machine_vector alphabook1_mv __initmv = {
 ALIAS_MV(alphabook1)
 #endif
 
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_AVANTI)
+#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_AVANTI_CH)
 struct alpha_machine_vector avanti_mv __initmv = {
 	.vector_name		= "Avanti",
 	DO_EV4_MMU,
@@ -389,7 +389,7 @@ struct alpha_machine_vector avanti_mv __initmv = {
 ALIAS_MV(avanti)
 #endif
 
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_NONAME)
+#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_NONAME_CH)
 struct alpha_machine_vector noname_mv __initmv = {
 	.vector_name		= "Noname",
 	DO_EV4_MMU,
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 06/14] alpha: missing includes
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (4 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 05/14] alpha: sys_sio: fix misspelled ifdefs Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 07/14] alpha: core_lca: take the unused functions out Arnd Bergmann
                   ` (9 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Al Viro <viro@zeniv.linux.org.uk>

... and missing externs in proto.h

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/kernel/bugs.c          |  1 +
 arch/alpha/kernel/console.c       |  1 +
 arch/alpha/kernel/core_irongate.c |  1 -
 arch/alpha/kernel/irq.c           |  1 +
 arch/alpha/kernel/pci-noop.c      |  1 +
 arch/alpha/kernel/proto.h         | 31 +++++++++++++++++++++++++++----
 arch/alpha/kernel/smc37c669.c     |  2 ++
 arch/alpha/kernel/smc37c93x.c     |  2 ++
 arch/alpha/kernel/smp.c           |  1 +
 arch/alpha/kernel/srmcons.c       |  2 ++
 arch/alpha/kernel/sys_nautilus.c  |  2 --
 arch/alpha/lib/checksum.c         |  1 +
 arch/alpha/lib/fpreg.c            |  1 +
 arch/alpha/lib/memcpy.c           |  3 +++
 arch/alpha/math-emu/math.c        |  7 +------
 arch/alpha/mm/init.c              |  2 +-
 16 files changed, 45 insertions(+), 14 deletions(-)

diff --git a/arch/alpha/kernel/bugs.c b/arch/alpha/kernel/bugs.c
index 08cc10d7fa17..e8c51089325f 100644
--- a/arch/alpha/kernel/bugs.c
+++ b/arch/alpha/kernel/bugs.c
@@ -1,6 +1,7 @@
 
 #include <asm/hwrpb.h>
 #include <linux/device.h>
+#include <linux/cpu.h>
 
 
 #ifdef CONFIG_SYSFS
diff --git a/arch/alpha/kernel/console.c b/arch/alpha/kernel/console.c
index 5476279329a6..4193f76e9633 100644
--- a/arch/alpha/kernel/console.c
+++ b/arch/alpha/kernel/console.c
@@ -15,6 +15,7 @@
 #include <asm/machvec.h>
 
 #include "pci_impl.h"
+#include "proto.h"
 
 #ifdef CONFIG_VGA_HOSE
 
diff --git a/arch/alpha/kernel/core_irongate.c b/arch/alpha/kernel/core_irongate.c
index 6b8ed12936b6..05dc4c1b9074 100644
--- a/arch/alpha/kernel/core_irongate.c
+++ b/arch/alpha/kernel/core_irongate.c
@@ -226,7 +226,6 @@ albacore_init_arch(void)
 	if (memtop > pci_mem) {
 #ifdef CONFIG_BLK_DEV_INITRD
 		extern unsigned long initrd_start, initrd_end;
-		extern void *move_initrd(unsigned long);
 
 		/* Move the initrd out of the way. */
 		if (initrd_end && __pa(initrd_end) > pci_mem) {
diff --git a/arch/alpha/kernel/irq.c b/arch/alpha/kernel/irq.c
index 15f2effd6baf..c67047c5d830 100644
--- a/arch/alpha/kernel/irq.c
+++ b/arch/alpha/kernel/irq.c
@@ -28,6 +28,7 @@
 
 #include <asm/io.h>
 #include <linux/uaccess.h>
+#include "irq_impl.h"
 
 volatile unsigned long irq_err_count;
 DEFINE_PER_CPU(unsigned long, irq_pmi_count);
diff --git a/arch/alpha/kernel/pci-noop.c b/arch/alpha/kernel/pci-noop.c
index ae82061edae9..32ca59a5a704 100644
--- a/arch/alpha/kernel/pci-noop.c
+++ b/arch/alpha/kernel/pci-noop.c
@@ -17,6 +17,7 @@
 #include <linux/scatterlist.h>
 #include <linux/syscalls.h>
 
+#include "pci_impl.h"
 #include "proto.h"
 
 
diff --git a/arch/alpha/kernel/proto.h b/arch/alpha/kernel/proto.h
index 2c89c1c55712..5b1d0d71d479 100644
--- a/arch/alpha/kernel/proto.h
+++ b/arch/alpha/kernel/proto.h
@@ -15,6 +15,7 @@ struct pt_regs;
 struct task_struct;
 struct pci_dev;
 struct pci_controller;
+struct pci_bus;
 
 /* core_apecs.c */
 extern struct pci_ops apecs_pci_ops;
@@ -114,6 +115,9 @@ extern int boot_cpuid;
 #ifdef CONFIG_VERBOSE_MCHECK
 extern unsigned long alpha_verbose_mcheck;
 #endif
+#ifdef CONFIG_BLK_DEV_INITRD
+extern void * __init move_initrd(unsigned long);
+#endif
 extern struct screen_info vgacon_screen_info;
 
 /* srmcons.c */
@@ -128,6 +132,7 @@ extern void unregister_srm_console(void);
 /* smp.c */
 extern void setup_smp(void);
 extern void handle_ipi(struct pt_regs *);
+extern void __init smp_callin(void);
 
 /* bios32.c */
 /* extern void reset_for_srm(void); */
@@ -139,13 +144,13 @@ extern void common_init_rtc(void);
 extern unsigned long est_cycle_freq;
 
 /* smc37c93x.c */
-extern void SMC93x_Init(void);
+extern int __init SMC93x_Init(void);
 
 /* smc37c669.c */
-extern void SMC669_Init(int);
+extern void __init SMC669_Init(int);
 
 /* es1888.c */
-extern void es1888_init(void);
+extern void __init es1888_init(void);
 
 /* ../lib/fpreg.c */
 extern void alpha_write_fp_reg (unsigned long reg, unsigned long val);
@@ -166,19 +171,37 @@ extern void entSys(void);
 extern void entUna(void);
 extern void entDbg(void);
 
+/* pci.c */
+extern void pcibios_claim_one_bus(struct pci_bus *);
+
 /* ptrace.c */
 extern int ptrace_set_bpt (struct task_struct *child);
 extern int ptrace_cancel_bpt (struct task_struct *child);
+extern void syscall_trace_leave(void);
+extern unsigned long syscall_trace_enter(void);
+
+/* signal.c */
+struct sigcontext;
+extern void do_sigreturn(struct sigcontext __user *);
+struct rt_sigframe;
+extern void do_rt_sigreturn(struct rt_sigframe __user *);
+extern void do_work_pending(struct pt_regs *, unsigned long, unsigned long, unsigned long);
 
 /* traps.c */
 extern void dik_show_regs(struct pt_regs *regs, unsigned long *r9_15);
 extern void die_if_kernel(char *, struct pt_regs *, long, unsigned long *);
+extern void do_entInt(unsigned long, unsigned long, unsigned long, struct pt_regs *);
+extern void do_entArith(unsigned long, unsigned long, struct pt_regs *);
+extern void do_entIF(unsigned long, struct pt_regs *);
+extern void do_entDbg(struct pt_regs *);
+struct allregs;
+extern void do_entUna(void *, unsigned long, unsigned long, struct allregs *);
+extern void do_entUnaUser(void __user *, unsigned long, unsigned long, struct pt_regs *);
 
 /* sys_titan.c */
 extern void titan_dispatch_irqs(u64);
 
 /* ../mm/init.c */
-extern void switch_to_system_map(void);
 extern void srm_paging_stop(void);
 
 static inline int
diff --git a/arch/alpha/kernel/smc37c669.c b/arch/alpha/kernel/smc37c669.c
index cb8726b667da..a5a6ed97a6ce 100644
--- a/arch/alpha/kernel/smc37c669.c
+++ b/arch/alpha/kernel/smc37c669.c
@@ -11,6 +11,8 @@
 #include <asm/hwrpb.h>
 #include <asm/io.h>
 
+#include "proto.h"
+
 #if 0
 # define DBG_DEVS(args)         printk args
 #else
diff --git a/arch/alpha/kernel/smc37c93x.c b/arch/alpha/kernel/smc37c93x.c
index 71cd7aca38ce..8028273f0d16 100644
--- a/arch/alpha/kernel/smc37c93x.c
+++ b/arch/alpha/kernel/smc37c93x.c
@@ -12,6 +12,8 @@
 #include <asm/hwrpb.h>
 #include <asm/io.h>
 
+#include "proto.h"
+
 #define SMC_DEBUG 0
 
 #if SMC_DEBUG
diff --git a/arch/alpha/kernel/smp.c b/arch/alpha/kernel/smp.c
index 8e9dd63b220c..ed06367ece57 100644
--- a/arch/alpha/kernel/smp.c
+++ b/arch/alpha/kernel/smp.c
@@ -38,6 +38,7 @@
 #include <asm/irq.h>
 #include <asm/mmu_context.h>
 #include <asm/tlbflush.h>
+#include <asm/cacheflush.h>
 
 #include "proto.h"
 #include "irq_impl.h"
diff --git a/arch/alpha/kernel/srmcons.c b/arch/alpha/kernel/srmcons.c
index feaf89f6936b..3e61073f4b30 100644
--- a/arch/alpha/kernel/srmcons.c
+++ b/arch/alpha/kernel/srmcons.c
@@ -21,6 +21,8 @@
 #include <asm/console.h>
 #include <linux/uaccess.h>
 
+#include "proto.h"
+
 
 static DEFINE_SPINLOCK(srmcons_callback_lock);
 static int srm_is_registered_console = 0;
diff --git a/arch/alpha/kernel/sys_nautilus.c b/arch/alpha/kernel/sys_nautilus.c
index 4dcee81606a5..13b79960b4b9 100644
--- a/arch/alpha/kernel/sys_nautilus.c
+++ b/arch/alpha/kernel/sys_nautilus.c
@@ -184,8 +184,6 @@ nautilus_machine_check(unsigned long vector, unsigned long la_ptr)
 	mb();
 }
 
-extern void pcibios_claim_one_bus(struct pci_bus *);
-
 static struct resource irongate_mem = {
 	.name	= "Irongate PCI MEM",
 	.flags	= IORESOURCE_MEM,
diff --git a/arch/alpha/lib/checksum.c b/arch/alpha/lib/checksum.c
index 3f35c3ed6948..17fa230baeef 100644
--- a/arch/alpha/lib/checksum.c
+++ b/arch/alpha/lib/checksum.c
@@ -12,6 +12,7 @@
  
 #include <linux/module.h>
 #include <linux/string.h>
+#include <net/checksum.h>
 
 #include <asm/byteorder.h>
 
diff --git a/arch/alpha/lib/fpreg.c b/arch/alpha/lib/fpreg.c
index 7c08b225261c..eee11fb4c7f1 100644
--- a/arch/alpha/lib/fpreg.c
+++ b/arch/alpha/lib/fpreg.c
@@ -9,6 +9,7 @@
 #include <linux/export.h>
 #include <linux/preempt.h>
 #include <asm/thread_info.h>
+#include <asm/fpu.h>
 
 #if defined(CONFIG_ALPHA_EV6) || defined(CONFIG_ALPHA_EV67)
 #define STT(reg,val)  asm volatile ("ftoit $f"#reg",%0" : "=r"(val));
diff --git a/arch/alpha/lib/memcpy.c b/arch/alpha/lib/memcpy.c
index cbac3dc6d963..78b6850a9d53 100644
--- a/arch/alpha/lib/memcpy.c
+++ b/arch/alpha/lib/memcpy.c
@@ -18,6 +18,7 @@
 
 #include <linux/types.h>
 #include <linux/export.h>
+#include <linux/string.h>
 
 /*
  * This should be done in one go with ldq_u*2/mask/stq_u. Do it
@@ -150,6 +151,8 @@ static inline void __memcpy_aligned_dn (unsigned long d, unsigned long s,
 	DO_REST_ALIGNED_DN(d,s,n);
 }
 
+#undef memcpy
+
 void * memcpy(void * dest, const void *src, size_t n)
 {
 	if (!(((unsigned long) dest ^ (unsigned long) src) & 7)) {
diff --git a/arch/alpha/math-emu/math.c b/arch/alpha/math-emu/math.c
index 4212258f3cfd..68d420bfd3c0 100644
--- a/arch/alpha/math-emu/math.c
+++ b/arch/alpha/math-emu/math.c
@@ -4,6 +4,7 @@
 #include <linux/kernel.h>
 #include <linux/sched.h>
 #include <asm/ptrace.h>
+#include <asm/fpu.h>
 
 #include <linux/uaccess.h>
 
@@ -45,12 +46,6 @@
 #define MISC_TRAPB	0x0000
 #define MISC_EXCB	0x0400
 
-extern unsigned long alpha_read_fp_reg (unsigned long reg);
-extern void alpha_write_fp_reg (unsigned long reg, unsigned long val);
-extern unsigned long alpha_read_fp_reg_s (unsigned long reg);
-extern void alpha_write_fp_reg_s (unsigned long reg, unsigned long val);
-
-
 #ifdef MODULE
 
 MODULE_DESCRIPTION("FP Software completion module");
diff --git a/arch/alpha/mm/init.c b/arch/alpha/mm/init.c
index a155180d7a83..4fe618446e4c 100644
--- a/arch/alpha/mm/init.c
+++ b/arch/alpha/mm/init.c
@@ -33,7 +33,7 @@
 #include <asm/setup.h>
 #include <asm/sections.h>
 
-extern void die_if_kernel(char *,struct pt_regs *,long);
+#include "../kernel/proto.h"
 
 static struct pcb_struct original_pcb;
 
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 07/14] alpha: core_lca: take the unused functions out
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (5 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 06/14] alpha: missing includes Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 08/14] alpha: jensen, t2 - make __EXTERN_INLINE same as for the rest Arnd Bergmann
                   ` (8 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Al Viro <viro@zeniv.linux.org.uk>

the only user had been drivers/char/h8.c, and that got taken out
and shot back in 2004...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/kernel/core_lca.c | 44 ------------------------------------
 1 file changed, 44 deletions(-)

diff --git a/arch/alpha/kernel/core_lca.c b/arch/alpha/kernel/core_lca.c
index 57e0750419f2..4616b11643d5 100644
--- a/arch/alpha/kernel/core_lca.c
+++ b/arch/alpha/kernel/core_lca.c
@@ -471,47 +471,3 @@ lca_machine_check(unsigned long vector, unsigned long la_ptr)
 	}
 #endif /* CONFIG_VERBOSE_MCHECK */
 }
-
-/*
- * The following routines are needed to support the SPEED changing
- * necessary to successfully manage the thermal problem on the AlphaBook1.
- */
-
-void
-lca_clock_print(void)
-{
-        long    pmr_reg;
-
-        pmr_reg = LCA_READ_PMR;
-
-        printk("Status of clock control:\n");
-        printk("\tPrimary clock divisor\t0x%lx\n", LCA_GET_PRIMARY(pmr_reg));
-        printk("\tOverride clock divisor\t0x%lx\n", LCA_GET_OVERRIDE(pmr_reg));
-        printk("\tInterrupt override is %s\n",
-	       (pmr_reg & LCA_PMR_INTO) ? "on" : "off"); 
-        printk("\tDMA override is %s\n",
-	       (pmr_reg & LCA_PMR_DMAO) ? "on" : "off"); 
-
-}
-
-int
-lca_get_clock(void)
-{
-        long    pmr_reg;
-
-        pmr_reg = LCA_READ_PMR;
-        return(LCA_GET_PRIMARY(pmr_reg));
-
-}
-
-void
-lca_clock_fiddle(int divisor)
-{
-        long    pmr_reg;
-
-        pmr_reg = LCA_READ_PMR;
-        LCA_SET_PRIMARY_CLOCK(pmr_reg, divisor);
-	/* lca_norm_clock = divisor; */
-        LCA_WRITE_PMR(pmr_reg);
-        mb();
-}
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 08/14] alpha: jensen, t2 - make __EXTERN_INLINE same as for the rest
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (6 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 07/14] alpha: core_lca: take the unused functions out Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 09/14] alpha: trim the unused stuff from asm-offsets.c Arnd Bergmann
                   ` (7 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Al Viro <viro@zeniv.linux.org.uk>

We want io.h primitives (readb(), etc.) to be extern inline.
However, that requires the backing out-of-line implementation
somewhere, preferably kept in sync with the inline ones.

The way it's done is __EXTERN_INLINE macro that defaults to
extern inline, but can be overridden in compilation unit where
the out-of-line instance will be.

That works, but it's brittle - we *must* make sure that asm/io.h
is the very first include in such compilation units.  There'd
been a bunch of bugs of that sort in the past.

Another issue is the choice of overriding definition for
__EXTERN_INLINE; it must be either 'inline' or empty.  Either
will do for compilation purposes - inline void foo(...) {...}
(without extern or static) is going to generate out-of-line
instance.  The difference is that 'definition without a
prototype' heuristics trigger on
void foo(void)
{
	...
}
but not on
inline void foo(void)
{
	...
}

Most of the overrides go for 'inline'; in two cases (sys_jensen
and core_t2) __EXTERN_INLINE is defined as empty.  Without
-Wmissing-prototypes it didn't matter, but now that we have
that thing always on...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/kernel/core_t2.c    | 2 +-
 arch/alpha/kernel/sys_jensen.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/alpha/kernel/core_t2.c b/arch/alpha/kernel/core_t2.c
index 98d5b6ff8a76..3d72d90624f1 100644
--- a/arch/alpha/kernel/core_t2.c
+++ b/arch/alpha/kernel/core_t2.c
@@ -10,7 +10,7 @@
  * Code common to all T2 core logic chips.
  */
 
-#define __EXTERN_INLINE
+#define __EXTERN_INLINE inline
 #include <asm/io.h>
 #include <asm/core_t2.h>
 #undef __EXTERN_INLINE
diff --git a/arch/alpha/kernel/sys_jensen.c b/arch/alpha/kernel/sys_jensen.c
index 5c9c88428124..a1bb1c4a7e93 100644
--- a/arch/alpha/kernel/sys_jensen.c
+++ b/arch/alpha/kernel/sys_jensen.c
@@ -7,7 +7,7 @@
  *
  * Code supporting the Jensen.
  */
-#define __EXTERN_INLINE
+#define __EXTERN_INLINE inline
 #include <asm/io.h>
 #include <asm/jensen.h>
 #undef  __EXTERN_INLINE
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 09/14] alpha: trim the unused stuff from asm-offsets.c
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (7 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 08/14] alpha: jensen, t2 - make __EXTERN_INLINE same as for the rest Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 10/14] alpha: remove DECpc AXP150 (Jensen) support Arnd Bergmann
                   ` (6 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Al Viro <viro@zeniv.linux.org.uk>

Out of 21 constants, only 6 are used...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/kernel/asm-offsets.c | 21 +--------------------
 1 file changed, 1 insertion(+), 20 deletions(-)

diff --git a/arch/alpha/kernel/asm-offsets.c b/arch/alpha/kernel/asm-offsets.c
index bf1eedd27cf7..4cfeae42c79a 100644
--- a/arch/alpha/kernel/asm-offsets.c
+++ b/arch/alpha/kernel/asm-offsets.c
@@ -10,35 +10,16 @@
 #include <linux/sched.h>
 #include <linux/ptrace.h>
 #include <linux/kbuild.h>
-#include <asm/io.h>
+#include <asm/machvec.h>
 
 static void __used foo(void)
 {
-	DEFINE(TI_TASK, offsetof(struct thread_info, task));
 	DEFINE(TI_FLAGS, offsetof(struct thread_info, flags));
-	DEFINE(TI_CPU, offsetof(struct thread_info, cpu));
 	DEFINE(TI_FP, offsetof(struct thread_info, fp));
 	DEFINE(TI_STATUS, offsetof(struct thread_info, status));
 	BLANK();
 
-        DEFINE(TASK_BLOCKED, offsetof(struct task_struct, blocked));
-        DEFINE(TASK_CRED, offsetof(struct task_struct, cred));
-        DEFINE(TASK_REAL_PARENT, offsetof(struct task_struct, real_parent));
-        DEFINE(TASK_GROUP_LEADER, offsetof(struct task_struct, group_leader));
-        DEFINE(TASK_TGID, offsetof(struct task_struct, tgid));
-        BLANK();
-
-        DEFINE(CRED_UID,  offsetof(struct cred, uid));
-        DEFINE(CRED_EUID, offsetof(struct cred, euid));
-        DEFINE(CRED_GID,  offsetof(struct cred, gid));
-        DEFINE(CRED_EGID, offsetof(struct cred, egid));
-        BLANK();
-
 	DEFINE(SIZEOF_PT_REGS, sizeof(struct pt_regs));
-	DEFINE(PT_PTRACED, PT_PTRACED);
-	DEFINE(CLONE_VM, CLONE_VM);
-	DEFINE(CLONE_UNTRACED, CLONE_UNTRACED);
-	DEFINE(SIGCHLD, SIGCHLD);
 	BLANK();
 
 	DEFINE(HAE_CACHE, offsetof(struct alpha_machine_vector, hae_cache));
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 10/14] alpha: remove DECpc AXP150 (Jensen) support
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (8 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 09/14] alpha: trim the unused stuff from asm-offsets.c Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03 16:07   ` Linus Torvalds
  2024-05-03 17:00   ` Al Viro
  2024-05-03  8:11 ` [PATCH 11/14] alpha: sable: remove early machine support Arnd Bergmann
                   ` (5 subsequent siblings)
  15 siblings, 2 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Arnd Bergmann <arnd@arndb.de>

This is one of the hackiest Alpha machines, and the only one without
PCI support. Removing this allows cleaning up code in eise and tty
drivers in addition to the architecture code.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 Documentation/driver-api/eisa.rst    |   4 +-
 arch/alpha/Kconfig                   |  18 +-
 arch/alpha/include/asm/dma-mapping.h |   4 -
 arch/alpha/include/asm/io.h          |  20 +-
 arch/alpha/include/asm/jensen.h      | 363 ---------------------------
 arch/alpha/kernel/Makefile           |   3 +-
 arch/alpha/kernel/pci-noop.c         | 114 ---------
 arch/alpha/kernel/setup.c            |  19 +-
 arch/alpha/kernel/sys_jensen.c       | 237 -----------------
 arch/alpha/kernel/traps.c            |  33 ---
 drivers/eisa/Kconfig                 |   9 +-
 drivers/eisa/virtual_root.c          |   2 +-
 drivers/input/serio/i8042-io.h       |   5 +-
 drivers/tty/serial/8250/8250.h       |   3 -
 drivers/tty/serial/8250/8250_alpha.c |  21 --
 drivers/tty/serial/8250/8250_core.c  |   4 -
 drivers/tty/serial/8250/Makefile     |   2 -
 17 files changed, 16 insertions(+), 845 deletions(-)
 delete mode 100644 arch/alpha/include/asm/jensen.h
 delete mode 100644 arch/alpha/kernel/pci-noop.c
 delete mode 100644 arch/alpha/kernel/sys_jensen.c
 delete mode 100644 drivers/tty/serial/8250/8250_alpha.c

diff --git a/Documentation/driver-api/eisa.rst b/Documentation/driver-api/eisa.rst
index 3eac11b7eb01..b33ebe1ec9ed 100644
--- a/Documentation/driver-api/eisa.rst
+++ b/Documentation/driver-api/eisa.rst
@@ -196,8 +196,8 @@ eisa_bus.disable_dev
 virtual_root.force_probe
 	Force the probing code to probe EISA slots even when it cannot find an
 	EISA compliant mainboard (nothing appears on slot 0). Defaults to 0
-	(don't force), and set to 1 (force probing) when either
-	CONFIG_ALPHA_JENSEN or CONFIG_EISA_VLB_PRIMING are set.
+	(don't force), and set to 1 (force probing) when
+	CONFIG_EISA_VLB_PRIMING is set.
 
 Random notes
 ============
diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 3afd042150f8..9b8d3ce40c28 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -10,7 +10,7 @@ config ALPHA
 	select ARCH_NO_SG_CHAIN
 	select ARCH_USE_CMPXCHG_LOCKREF
 	select DMA_OPS if PCI
-	select FORCE_PCI if !ALPHA_JENSEN
+	select FORCE_PCI
 	select PCI_DOMAINS if PCI
 	select PCI_SYSCALL if PCI
 	select HAVE_ASM_MODVERSIONS
@@ -99,7 +99,6 @@ choice
 	  EB64+               EB64+ 21064 evaluation board
 	  EB66                EB66 21066 evaluation board
 	  EB66+               EB66+ 21066 evaluation board
-	  Jensen              DECpc 150, DEC 2000 models 300, 500
 	  LX164               AlphaPC164-LX
 	  Lynx                AS 2100A
 	  Miata               Personal Workstation 433/500/600 a/au
@@ -198,17 +197,6 @@ config ALPHA_EIGER
 	  Apparently an obscure OEM single-board computer based on the
 	  Typhoon/Tsunami chipset family. Information on it is scanty.
 
-config ALPHA_JENSEN
-	bool "Jensen"
-	select HAVE_EISA
-	help
-	  DEC PC 150 AXP (aka Jensen): This is a very old Digital system - one
-	  of the first-generation Alpha systems. A number of these systems
-	  seem to be available on the second- hand market. The Jensen is a
-	  floor-standing tower system which originally used a 150MHz 21064 It
-	  used programmable logic to interface a 486 EISA I/O bridge to the
-	  CPU.
-
 config ALPHA_LX164
 	bool "LX164"
 	help
@@ -337,7 +325,7 @@ config ALPHA_NONAME
 
 config ALPHA_EV4
 	bool
-	depends on ALPHA_JENSEN || (ALPHA_SABLE && !ALPHA_GAMMA) || ALPHA_LYNX || ALPHA_NORITAKE && !ALPHA_PRIMO || ALPHA_MIKASA && !ALPHA_PRIMO || ALPHA_CABRIOLET || ALPHA_AVANTI_CH || ALPHA_EB64P_CH || ALPHA_XL || ALPHA_NONAME || ALPHA_EB66 || ALPHA_EB66P || ALPHA_P2K
+	depends on (ALPHA_SABLE && !ALPHA_GAMMA) || ALPHA_LYNX || ALPHA_NORITAKE && !ALPHA_PRIMO || ALPHA_MIKASA && !ALPHA_PRIMO || ALPHA_CABRIOLET || ALPHA_AVANTI_CH || ALPHA_EB64P_CH || ALPHA_XL || ALPHA_NONAME || ALPHA_EB66 || ALPHA_EB66P || ALPHA_P2K
 	default y if !ALPHA_LYNX
 	default y if !ALPHA_EV5
 
@@ -483,7 +471,7 @@ config ALPHA_QEMU
 config ALPHA_SRM
 	bool "Use SRM as bootloader" if ALPHA_CABRIOLET || ALPHA_AVANTI_CH || ALPHA_EB64P || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_NAUTILUS || ALPHA_NONAME
 	depends on TTY
-	default y if ALPHA_JENSEN || ALPHA_MIKASA || ALPHA_SABLE || ALPHA_LYNX || ALPHA_NORITAKE || ALPHA_DP264 || ALPHA_RAWHIDE || ALPHA_EIGER || ALPHA_WILDFIRE || ALPHA_TITAN || ALPHA_SHARK || ALPHA_MARVEL
+	default y if ALPHA_MIKASA || ALPHA_SABLE || ALPHA_LYNX || ALPHA_NORITAKE || ALPHA_DP264 || ALPHA_RAWHIDE || ALPHA_EIGER || ALPHA_WILDFIRE || ALPHA_TITAN || ALPHA_SHARK || ALPHA_MARVEL
 	help
 	  There are two different types of booting firmware on Alphas: SRM,
 	  which is command line driven, and ARC, which uses menus and arrow
diff --git a/arch/alpha/include/asm/dma-mapping.h b/arch/alpha/include/asm/dma-mapping.h
index 6ce7e2041685..ad5a59b035cb 100644
--- a/arch/alpha/include/asm/dma-mapping.h
+++ b/arch/alpha/include/asm/dma-mapping.h
@@ -6,11 +6,7 @@ extern const struct dma_map_ops alpha_pci_ops;
 
 static inline const struct dma_map_ops *get_arch_dma_ops(void)
 {
-#ifdef CONFIG_ALPHA_JENSEN
-	return NULL;
-#else
 	return &alpha_pci_ops;
-#endif
 }
 
 #endif	/* _ALPHA_DMA_MAPPING_H */
diff --git a/arch/alpha/include/asm/io.h b/arch/alpha/include/asm/io.h
index 4f47a5003fe8..0c54ddb5b33e 100644
--- a/arch/alpha/include/asm/io.h
+++ b/arch/alpha/include/asm/io.h
@@ -209,8 +209,6 @@ static inline int generic_is_mmio(const volatile void __iomem *a)
 # include <asm/core_cia.h>
 #elif defined(CONFIG_ALPHA_IRONGATE)
 # include <asm/core_irongate.h>
-#elif defined(CONFIG_ALPHA_JENSEN)
-# include <asm/jensen.h>
 #elif defined(CONFIG_ALPHA_LCA)
 # include <asm/core_lca.h>
 #elif defined(CONFIG_ALPHA_MARVEL)
@@ -631,23 +629,7 @@ extern void outsl (unsigned long port, const void *src, unsigned long count);
 #define outsw outsw
 #define outsl outsl
 
-/*
- * The Alpha Jensen hardware for some rather strange reason puts
- * the RTC clock at 0x170 instead of 0x70. Probably due to some
- * misguided idea about using 0x70 for NMI stuff.
- *
- * These defines will override the defaults when doing RTC queries
- */
-
-#ifdef CONFIG_ALPHA_GENERIC
-# define RTC_PORT(x)	((x) + alpha_mv.rtc_port)
-#else
-# ifdef CONFIG_ALPHA_JENSEN
-#  define RTC_PORT(x)	(0x170+(x))
-# else
-#  define RTC_PORT(x)	(0x70 + (x))
-# endif
-#endif
+#define RTC_PORT(x)	(0x70 + (x))
 #define RTC_ALWAYS_BCD	0
 
 /*
diff --git a/arch/alpha/include/asm/jensen.h b/arch/alpha/include/asm/jensen.h
deleted file mode 100644
index 66eb049eb421..000000000000
--- a/arch/alpha/include/asm/jensen.h
+++ /dev/null
@@ -1,363 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef __ALPHA_JENSEN_H
-#define __ALPHA_JENSEN_H
-
-#include <asm/compiler.h>
-
-/*
- * Defines for the AlphaPC EISA IO and memory address space.
- */
-
-/*
- * NOTE! The memory operations do not set any memory barriers, as it's
- * not needed for cases like a frame buffer that is essentially memory-like.
- * You need to do them by hand if the operations depend on ordering.
- *
- * Similarly, the port IO operations do a "mb" only after a write operation:
- * if an mb is needed before (as in the case of doing memory mapped IO
- * first, and then a port IO operation to the same device), it needs to be
- * done by hand.
- *
- * After the above has bitten me 100 times, I'll give up and just do the
- * mb all the time, but right now I'm hoping this will work out.  Avoiding
- * mb's may potentially be a noticeable speed improvement, but I can't
- * honestly say I've tested it.
- *
- * Handling interrupts that need to do mb's to synchronize to non-interrupts
- * is another fun race area.  Don't do it (because if you do, I'll have to
- * do *everything* with interrupts disabled, ugh).
- */
-
-/*
- * EISA Interrupt Acknowledge address
- */
-#define EISA_INTA		(IDENT_ADDR + 0x100000000UL)
-
-/*
- * FEPROM addresses
- */
-#define EISA_FEPROM0		(IDENT_ADDR + 0x180000000UL)
-#define EISA_FEPROM1		(IDENT_ADDR + 0x1A0000000UL)
-
-/*
- * VL82C106 base address
- */
-#define EISA_VL82C106		(IDENT_ADDR + 0x1C0000000UL)
-
-/*
- * EISA "Host Address Extension" address (bits 25-31 of the EISA address)
- */
-#define EISA_HAE		(IDENT_ADDR + 0x1D0000000UL)
-
-/*
- * "SYSCTL" register address
- */
-#define EISA_SYSCTL		(IDENT_ADDR + 0x1E0000000UL)
-
-/*
- * "spare" register address
- */
-#define EISA_SPARE		(IDENT_ADDR + 0x1F0000000UL)
-
-/*
- * EISA memory address offset
- */
-#define EISA_MEM		(IDENT_ADDR + 0x200000000UL)
-
-/*
- * EISA IO address offset
- */
-#define EISA_IO			(IDENT_ADDR + 0x300000000UL)
-
-
-#ifdef __KERNEL__
-
-#ifndef __EXTERN_INLINE
-#define __EXTERN_INLINE extern inline
-#define __IO_EXTERN_INLINE
-#endif
-
-/*
- * Handle the "host address register". This needs to be set
- * to the high 7 bits of the EISA address.  This is also needed
- * for EISA IO addresses, which are only 16 bits wide (the
- * hae needs to be set to 0).
- *
- * HAE isn't needed for the local IO operations, though.
- */
-
-#define JENSEN_HAE_ADDRESS	EISA_HAE
-#define JENSEN_HAE_MASK		0x1ffffff
-
-__EXTERN_INLINE void jensen_set_hae(unsigned long addr)
-{
-	/* hae on the Jensen is bits 31:25 shifted right */
-	addr >>= 25;
-	if (addr != alpha_mv.hae_cache)
-		set_hae(addr);
-}
-
-#define vuip	volatile unsigned int *
-#define vulp	volatile unsigned long *
-
-/*
- * IO functions
- *
- * The "local" functions are those that don't go out to the EISA bus,
- * but instead act on the VL82C106 chip directly.. This is mainly the
- * keyboard, RTC,  printer and first two serial lines..
- *
- * The local stuff makes for some complications, but it seems to be
- * gone in the PCI version. I hope I can get DEC suckered^H^H^H^H^H^H^H^H
- * convinced that I need one of the newer machines.
- */
-
-__EXTERN_INLINE unsigned int jensen_local_inb(unsigned long addr)
-{
-	return 0xff & *(vuip)((addr << 9) + EISA_VL82C106);
-}
-
-__EXTERN_INLINE void jensen_local_outb(u8 b, unsigned long addr)
-{
-	*(vuip)((addr << 9) + EISA_VL82C106) = b;
-	mb();
-}
-
-__EXTERN_INLINE unsigned int jensen_bus_inb(unsigned long addr)
-{
-	long result;
-
-	jensen_set_hae(0);
-	result = *(volatile int *)((addr << 7) + EISA_IO + 0x00);
-	return __kernel_extbl(result, addr & 3);
-}
-
-__EXTERN_INLINE void jensen_bus_outb(u8 b, unsigned long addr)
-{
-	jensen_set_hae(0);
-	*(vuip)((addr << 7) + EISA_IO + 0x00) = b * 0x01010101;
-	mb();
-}
-
-/*
- * It seems gcc is not very good at optimizing away logical
- * operations that result in operations across inline functions.
- * Which is why this is a macro.
- */
-
-#define jensen_is_local(addr) ( \
-/* keyboard */	(addr == 0x60 || addr == 0x64) || \
-/* RTC */	(addr == 0x170 || addr == 0x171) || \
-/* mb COM2 */	(addr >= 0x2f8 && addr <= 0x2ff) || \
-/* mb LPT1 */	(addr >= 0x3bc && addr <= 0x3be) || \
-/* mb COM2 */	(addr >= 0x3f8 && addr <= 0x3ff))
-
-__EXTERN_INLINE u8 jensen_inb(unsigned long addr)
-{
-	if (jensen_is_local(addr))
-		return jensen_local_inb(addr);
-	else
-		return jensen_bus_inb(addr);
-}
-
-__EXTERN_INLINE void jensen_outb(u8 b, unsigned long addr)
-{
-	if (jensen_is_local(addr))
-		jensen_local_outb(b, addr);
-	else
-		jensen_bus_outb(b, addr);
-}
-
-__EXTERN_INLINE u16 jensen_inw(unsigned long addr)
-{
-	long result;
-
-	jensen_set_hae(0);
-	result = *(volatile int *) ((addr << 7) + EISA_IO + 0x20);
-	result >>= (addr & 3) * 8;
-	return 0xffffUL & result;
-}
-
-__EXTERN_INLINE u32 jensen_inl(unsigned long addr)
-{
-	jensen_set_hae(0);
-	return *(vuip) ((addr << 7) + EISA_IO + 0x60);
-}
-
-__EXTERN_INLINE u64 jensen_inq(unsigned long addr)
-{
-	jensen_set_hae(0);
-	return *(vulp) ((addr << 7) + EISA_IO + 0x60);
-}
-
-__EXTERN_INLINE void jensen_outw(u16 b, unsigned long addr)
-{
-	jensen_set_hae(0);
-	*(vuip) ((addr << 7) + EISA_IO + 0x20) = b * 0x00010001;
-	mb();
-}
-
-__EXTERN_INLINE void jensen_outl(u32 b, unsigned long addr)
-{
-	jensen_set_hae(0);
-	*(vuip) ((addr << 7) + EISA_IO + 0x60) = b;
-	mb();
-}
-
-__EXTERN_INLINE void jensen_outq(u64 b, unsigned long addr)
-{
-	jensen_set_hae(0);
-	*(vulp) ((addr << 7) + EISA_IO + 0x60) = b;
-	mb();
-}
-
-/*
- * Memory functions.
- */
-
-__EXTERN_INLINE u8 jensen_readb(const volatile void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	long result;
-
-	jensen_set_hae(addr);
-	addr &= JENSEN_HAE_MASK;
-	result = *(volatile int *) ((addr << 7) + EISA_MEM + 0x00);
-	result >>= (addr & 3) * 8;
-	return 0xffUL & result;
-}
-
-__EXTERN_INLINE u16 jensen_readw(const volatile void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	long result;
-
-	jensen_set_hae(addr);
-	addr &= JENSEN_HAE_MASK;
-	result = *(volatile int *) ((addr << 7) + EISA_MEM + 0x20);
-	result >>= (addr & 3) * 8;
-	return 0xffffUL & result;
-}
-
-__EXTERN_INLINE u32 jensen_readl(const volatile void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	jensen_set_hae(addr);
-	addr &= JENSEN_HAE_MASK;
-	return *(vuip) ((addr << 7) + EISA_MEM + 0x60);
-}
-
-__EXTERN_INLINE u64 jensen_readq(const volatile void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	unsigned long r0, r1;
-
-	jensen_set_hae(addr);
-	addr &= JENSEN_HAE_MASK;
-	addr = (addr << 7) + EISA_MEM + 0x60;
-	r0 = *(vuip) (addr);
-	r1 = *(vuip) (addr + (4 << 7));
-	return r1 << 32 | r0;
-}
-
-__EXTERN_INLINE void jensen_writeb(u8 b, volatile void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	jensen_set_hae(addr);
-	addr &= JENSEN_HAE_MASK;
-	*(vuip) ((addr << 7) + EISA_MEM + 0x00) = b * 0x01010101;
-}
-
-__EXTERN_INLINE void jensen_writew(u16 b, volatile void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	jensen_set_hae(addr);
-	addr &= JENSEN_HAE_MASK;
-	*(vuip) ((addr << 7) + EISA_MEM + 0x20) = b * 0x00010001;
-}
-
-__EXTERN_INLINE void jensen_writel(u32 b, volatile void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	jensen_set_hae(addr);
-	addr &= JENSEN_HAE_MASK;
-	*(vuip) ((addr << 7) + EISA_MEM + 0x60) = b;
-}
-
-__EXTERN_INLINE void jensen_writeq(u64 b, volatile void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	jensen_set_hae(addr);
-	addr &= JENSEN_HAE_MASK;
-	addr = (addr << 7) + EISA_MEM + 0x60;
-	*(vuip) (addr) = b;
-	*(vuip) (addr + (4 << 7)) = b >> 32;
-}
-
-__EXTERN_INLINE void __iomem *jensen_ioportmap(unsigned long addr)
-{
-	return (void __iomem *)addr;
-}
-
-__EXTERN_INLINE void __iomem *jensen_ioremap(unsigned long addr,
-					     unsigned long size)
-{
-	return (void __iomem *)(addr + 0x100000000ul);
-}
-
-__EXTERN_INLINE int jensen_is_ioaddr(unsigned long addr)
-{
-	return (long)addr >= 0;
-}
-
-__EXTERN_INLINE int jensen_is_mmio(const volatile void __iomem *addr)
-{
-	return (unsigned long)addr >= 0x100000000ul;
-}
-
-/* New-style ioread interface.  All the routines are so ugly for Jensen
-   that it doesn't make sense to merge them.  */
-
-#define IOPORT(OS, NS)							\
-__EXTERN_INLINE u##NS jensen_ioread##NS(const void __iomem *xaddr)	\
-{									\
-	if (jensen_is_mmio(xaddr))					\
-		return jensen_read##OS(xaddr - 0x100000000ul);		\
-	else								\
-		return jensen_in##OS((unsigned long)xaddr);		\
-}									\
-__EXTERN_INLINE void jensen_iowrite##NS(u##NS b, void __iomem *xaddr)	\
-{									\
-	if (jensen_is_mmio(xaddr))					\
-		jensen_write##OS(b, xaddr - 0x100000000ul);		\
-	else								\
-		jensen_out##OS(b, (unsigned long)xaddr);		\
-}
-
-IOPORT(b, 8)
-IOPORT(w, 16)
-IOPORT(l, 32)
-IOPORT(q, 64)
-
-#undef IOPORT
-
-#undef vuip
-#undef vulp
-
-#undef __IO_PREFIX
-#define __IO_PREFIX		jensen
-#define jensen_trivial_rw_bw	0
-#define jensen_trivial_rw_lq	0
-#define jensen_trivial_io_bw	0
-#define jensen_trivial_io_lq	0
-#define jensen_trivial_iounmap	1
-#include <asm/io_trivial.h>
-
-#ifdef __IO_EXTERN_INLINE
-#undef __EXTERN_INLINE
-#undef __IO_EXTERN_INLINE
-#endif
-
-#endif /* __KERNEL__ */
-
-#endif /* __ALPHA_JENSEN_H */
diff --git a/arch/alpha/kernel/Makefile b/arch/alpha/kernel/Makefile
index fb4efec7cbc7..4b31060e1711 100644
--- a/arch/alpha/kernel/Makefile
+++ b/arch/alpha/kernel/Makefile
@@ -27,7 +27,7 @@ obj-y 	 += core_apecs.o core_cia.o core_irongate.o core_lca.o \
 	    core_tsunami.o
 
 obj-y	 += sys_alcor.o sys_cabriolet.o sys_dp264.o sys_eb64p.o sys_eiger.o \
-	    sys_jensen.o sys_miata.o sys_mikasa.o sys_nautilus.o \
+	    sys_miata.o sys_mikasa.o sys_nautilus.o \
 	    sys_noritake.o sys_rawhide.o sys_ruffian.o sys_rx164.o \
 	    sys_sable.o sys_sio.o sys_sx164.o sys_takara.o
 
@@ -78,7 +78,6 @@ obj-$(CONFIG_ALPHA_TITAN)	+= sys_titan.o irq_i8259.o smc37c669.o
 obj-$(CONFIG_ALPHA_EB64P)	+= sys_eb64p.o irq_i8259.o
 obj-$(CONFIG_ALPHA_EB66)	+= sys_eb64p.o irq_i8259.o
 obj-$(CONFIG_ALPHA_EIGER)	+= sys_eiger.o irq_i8259.o
-obj-$(CONFIG_ALPHA_JENSEN)	+= sys_jensen.o pci-noop.o irq_i8259.o
 obj-$(CONFIG_ALPHA_MARVEL)	+= sys_marvel.o 
 obj-$(CONFIG_ALPHA_MIATA)	+= sys_miata.o irq_pyxis.o irq_i8259.o \
 				   es1888.o smc37c669.o
diff --git a/arch/alpha/kernel/pci-noop.c b/arch/alpha/kernel/pci-noop.c
deleted file mode 100644
index 32ca59a5a704..000000000000
--- a/arch/alpha/kernel/pci-noop.c
+++ /dev/null
@@ -1,114 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- *	linux/arch/alpha/kernel/pci-noop.c
- *
- * Stub PCI interfaces for Jensen-specific kernels.
- */
-
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/memblock.h>
-#include <linux/gfp.h>
-#include <linux/capability.h>
-#include <linux/mm.h>
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/dma-mapping.h>
-#include <linux/scatterlist.h>
-#include <linux/syscalls.h>
-
-#include "pci_impl.h"
-#include "proto.h"
-
-
-/*
- * The PCI controller list.
- */
-
-struct pci_controller *hose_head, **hose_tail = &hose_head;
-struct pci_controller *pci_isa_hose;
-
-
-struct pci_controller * __init
-alloc_pci_controller(void)
-{
-	struct pci_controller *hose;
-
-	hose = memblock_alloc(sizeof(*hose), SMP_CACHE_BYTES);
-	if (!hose)
-		panic("%s: Failed to allocate %zu bytes\n", __func__,
-		      sizeof(*hose));
-
-	*hose_tail = hose;
-	hose_tail = &hose->next;
-
-	return hose;
-}
-
-struct resource * __init
-alloc_resource(void)
-{
-	void *ptr = memblock_alloc(sizeof(struct resource), SMP_CACHE_BYTES);
-
-	if (!ptr)
-		panic("%s: Failed to allocate %zu bytes\n", __func__,
-		      sizeof(struct resource));
-
-	return ptr;
-}
-
-SYSCALL_DEFINE3(pciconfig_iobase, long, which, unsigned long, bus,
-		unsigned long, dfn)
-{
-	struct pci_controller *hose;
-
-	/* from hose or from bus.devfn */
-	if (which & IOBASE_FROM_HOSE) {
-		for (hose = hose_head; hose; hose = hose->next)
-			if (hose->index == bus)
-				break;
-		if (!hose)
-			return -ENODEV;
-	} else {
-		/* Special hook for ISA access.  */
-		if (bus == 0 && dfn == 0)
-			hose = pci_isa_hose;
-		else
-			return -ENODEV;
-	}
-
-	switch (which & ~IOBASE_FROM_HOSE) {
-	case IOBASE_HOSE:
-		return hose->index;
-	case IOBASE_SPARSE_MEM:
-		return hose->sparse_mem_base;
-	case IOBASE_DENSE_MEM:
-		return hose->dense_mem_base;
-	case IOBASE_SPARSE_IO:
-		return hose->sparse_io_base;
-	case IOBASE_DENSE_IO:
-		return hose->dense_io_base;
-	case IOBASE_ROOT_BUS:
-		return hose->bus->number;
-	}
-
-	return -EOPNOTSUPP;
-}
-
-SYSCALL_DEFINE5(pciconfig_read, unsigned long, bus, unsigned long, dfn,
-		unsigned long, off, unsigned long, len, void __user *, buf)
-{
-	if (!capable(CAP_SYS_ADMIN))
-		return -EPERM;
-	else
-		return -ENODEV;
-}
-
-SYSCALL_DEFINE5(pciconfig_write, unsigned long, bus, unsigned long, dfn,
-		unsigned long, off, unsigned long, len, void __user *, buf)
-{
-	if (!capable(CAP_SYS_ADMIN))
-		return -EPERM;
-	else
-		return -ENODEV;
-}
diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index 0738f9396f95..6e6c842ef07c 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -181,7 +181,6 @@ WEAK(eb64p_mv);
 WEAK(eb66_mv);
 WEAK(eb66p_mv);
 WEAK(eiger_mv);
-WEAK(jensen_mv);
 WEAK(lx164_mv);
 WEAK(lynx_mv);
 WEAK(marvel_ev7_mv);
@@ -224,7 +223,7 @@ static void __init
 reserve_std_resources(void)
 {
 	static struct resource standard_io_resources[] = {
-		{ .name = "rtc", .start = -1, .end = -1 },
+		{ .name = "rtc", .start = 0x70, .end =  0x7f},
         	{ .name = "dma1", .start = 0x00, .end = 0x1f },
         	{ .name = "pic1", .start = 0x20, .end = 0x3f },
         	{ .name = "timer", .start = 0x40, .end = 0x5f },
@@ -246,10 +245,6 @@ reserve_std_resources(void)
 			}
 	}
 
-	/* Fix up for the Jensen's queer RTC placement.  */
-	standard_io_resources[0].start = RTC_PORT(0);
-	standard_io_resources[0].end = RTC_PORT(0) + 0x0f;
-
 	for (i = 0; i < ARRAY_SIZE(standard_io_resources); ++i)
 		request_resource(io, standard_io_resources+i);
 }
@@ -486,14 +481,7 @@ setup_arch(char **cmdline_p)
 	/* 
 	 * Locate the command line.
 	 */
-	/* Hack for Jensen... since we're restricted to 8 or 16 chars for
-	   boot flags depending on the boot mode, we need some shorthand.
-	   This should do for installation.  */
-	if (strcmp(COMMAND_LINE, "INSTALL") == 0) {
-		strscpy(command_line, "root=/dev/fd0 load_ramdisk=1", sizeof(command_line));
-	} else {
-		strscpy(command_line, COMMAND_LINE, sizeof(command_line));
-	}
+	strscpy(command_line, COMMAND_LINE, sizeof(command_line));
 	strcpy(boot_command_line, command_line);
 	*cmdline_p = command_line;
 
@@ -745,7 +733,7 @@ get_sysvec(unsigned long type, unsigned long variation, unsigned long cpu)
 		NULL,		/* Ruby */
 		NULL,		/* Flamingo */
 		NULL,		/* Mannequin */
-		&jensen_mv,
+		NULL,		/* Jensens */
 		NULL, 		/* Pelican */
 		NULL,		/* Morgan */
 		NULL,		/* Sable -- see below.  */
@@ -943,7 +931,6 @@ get_sysvec_byname(const char *name)
 		&eb66_mv,
 		&eb66p_mv,
 		&eiger_mv,
-		&jensen_mv,
 		&lx164_mv,
 		&lynx_mv,
 		&miata_mv,
diff --git a/arch/alpha/kernel/sys_jensen.c b/arch/alpha/kernel/sys_jensen.c
deleted file mode 100644
index a1bb1c4a7e93..000000000000
--- a/arch/alpha/kernel/sys_jensen.c
+++ /dev/null
@@ -1,237 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- *	linux/arch/alpha/kernel/sys_jensen.c
- *
- *	Copyright (C) 1995 Linus Torvalds
- *	Copyright (C) 1998, 1999 Richard Henderson
- *
- * Code supporting the Jensen.
- */
-#define __EXTERN_INLINE inline
-#include <asm/io.h>
-#include <asm/jensen.h>
-#undef  __EXTERN_INLINE
-
-#include <linux/interrupt.h>
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/sched.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-
-#include <asm/ptrace.h>
-
-#include <asm/dma.h>
-#include <asm/irq.h>
-#include <asm/mmu_context.h>
-#include <asm/tlbflush.h>
-
-#include "proto.h"
-#include "irq_impl.h"
-#include "pci_impl.h"
-#include "machvec_impl.h"
-
-
-/*
- * Jensen is special: the vector is 0x8X0 for EISA interrupt X, and
- * 0x9X0 for the local motherboard interrupts.
- *
- * Note especially that those local interrupts CANNOT be masked,
- * which causes much of the pain below...
- *
- *	0x660 - NMI
- *
- *	0x800 - IRQ0  interval timer (not used, as we use the RTC timer)
- *	0x810 - IRQ1  line printer (duh..)
- *	0x860 - IRQ6  floppy disk
- *
- *	0x900 - COM1
- *	0x920 - COM2
- *	0x980 - keyboard
- *	0x990 - mouse
- *
- * PCI-based systems are more sane: they don't have the local
- * interrupts at all, and have only normal PCI interrupts from
- * devices.  Happily it's easy enough to do a sane mapping from the
- * Jensen.
- * 
- * Note that this means that we may have to do a hardware
- * "local_op" to a different interrupt than we report to the rest of the
- * world.
- */
-
-static void
-jensen_local_enable(struct irq_data *d)
-{
-	/* the parport is really hw IRQ 1, silly Jensen.  */
-	if (d->irq == 7)
-		i8259a_enable_irq(d);
-}
-
-static void
-jensen_local_disable(struct irq_data *d)
-{
-	/* the parport is really hw IRQ 1, silly Jensen.  */
-	if (d->irq == 7)
-		i8259a_disable_irq(d);
-}
-
-static void
-jensen_local_mask_ack(struct irq_data *d)
-{
-	/* the parport is really hw IRQ 1, silly Jensen.  */
-	if (d->irq == 7)
-		i8259a_mask_and_ack_irq(d);
-}
-
-static struct irq_chip jensen_local_irq_type = {
-	.name		= "LOCAL",
-	.irq_unmask	= jensen_local_enable,
-	.irq_mask	= jensen_local_disable,
-	.irq_mask_ack	= jensen_local_mask_ack,
-};
-
-static void 
-jensen_device_interrupt(unsigned long vector)
-{
-	int irq;
-
-	switch (vector) {
-	case 0x660:
-		printk("Whee.. NMI received. Probable hardware error\n");
-		printk("61=%02x, 461=%02x\n", inb(0x61), inb(0x461));
-		return;
-
-	/* local device interrupts: */
-	case 0x900: irq = 4; break;		/* com1 -> irq 4 */
-	case 0x920: irq = 3; break;		/* com2 -> irq 3 */
-	case 0x980: irq = 1; break;		/* kbd -> irq 1 */
-	case 0x990: irq = 9; break;		/* mouse -> irq 9 */
-
-	default:
-		if (vector > 0x900) {
-			printk("Unknown local interrupt %lx\n", vector);
-			return;
-		}
-
-		irq = (vector - 0x800) >> 4;
-		if (irq == 1)
-			irq = 7;
-		break;
-	}
-
-	/* If there is no handler yet... */
-	if (!irq_has_action(irq)) {
-	    /* If it is a local interrupt that cannot be masked... */
-	    if (vector >= 0x900)
-	    {
-	        /* Clear keyboard/mouse state */
-	    	inb(0x64);
-		inb(0x60);
-		/* Reset serial ports */
-		inb(0x3fa);
-		inb(0x2fa);
-		outb(0x0c, 0x3fc);
-		outb(0x0c, 0x2fc);
-		/* Clear NMI */
-		outb(0,0x61);
-		outb(0,0x461);
-	    }
-	}
-
-#if 0
-        /* A useful bit of code to find out if an interrupt is going wild.  */
-        {
-          static unsigned int last_msg = 0, last_cc = 0;
-          static int last_irq = -1, count = 0;
-          unsigned int cc;
-
-          __asm __volatile("rpcc %0" : "=r"(cc));
-          ++count;
-#define JENSEN_CYCLES_PER_SEC	(150000000)
-          if (cc - last_msg > ((JENSEN_CYCLES_PER_SEC) * 3) ||
-	      irq != last_irq) {
-                printk(KERN_CRIT " irq %d count %d cc %u @ %lx\n",
-                       irq, count, cc-last_cc, get_irq_regs()->pc);
-                count = 0;
-                last_msg = cc;
-                last_irq = irq;
-          }
-          last_cc = cc;
-        }
-#endif
-
-	handle_irq(irq);
-}
-
-static void __init
-jensen_init_irq(void)
-{
-	init_i8259a_irqs();
-
-	irq_set_chip_and_handler(1, &jensen_local_irq_type, handle_level_irq);
-	irq_set_chip_and_handler(4, &jensen_local_irq_type, handle_level_irq);
-	irq_set_chip_and_handler(3, &jensen_local_irq_type, handle_level_irq);
-	irq_set_chip_and_handler(7, &jensen_local_irq_type, handle_level_irq);
-	irq_set_chip_and_handler(9, &jensen_local_irq_type, handle_level_irq);
-
-	common_init_isa_dma();
-}
-
-static void __init
-jensen_init_arch(void)
-{
-	struct pci_controller *hose;
-#ifdef CONFIG_PCI
-	static struct pci_dev fake_isa_bridge = { .dma_mask = 0xffffffffUL, };
-
-	isa_bridge = &fake_isa_bridge;
-#endif
-
-	/* Create a hose so that we can report i/o base addresses to
-	   userland.  */
-
-	pci_isa_hose = hose = alloc_pci_controller();
-	hose->io_space = &ioport_resource;
-	hose->mem_space = &iomem_resource;
-	hose->index = 0;
-
-	hose->sparse_mem_base = EISA_MEM - IDENT_ADDR;
-	hose->dense_mem_base = 0;
-	hose->sparse_io_base = EISA_IO - IDENT_ADDR;
-	hose->dense_io_base = 0;
-
-	hose->sg_isa = hose->sg_pci = NULL;
-	__direct_map_base = 0;
-	__direct_map_size = 0xffffffff;
-}
-
-static void
-jensen_machine_check(unsigned long vector, unsigned long la)
-{
-	printk(KERN_CRIT "Machine check\n");
-}
-
-/*
- * The System Vector
- */
-
-struct alpha_machine_vector jensen_mv __initmv = {
-	.vector_name		= "Jensen",
-	DO_EV4_MMU,
-	IO_LITE(JENSEN,jensen),
-	.machine_check		= jensen_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.rtc_port		= 0x170,
-
-	.nr_irqs		= 16,
-	.device_interrupt	= jensen_device_interrupt,
-
-	.init_arch		= jensen_init_arch,
-	.init_irq		= jensen_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= NULL,
-	.kill_arch		= NULL,
-};
-ALIAS_MV(jensen)
diff --git a/arch/alpha/kernel/traps.c b/arch/alpha/kernel/traps.c
index 7fc72aeb7398..a9a8e9ab0f52 100644
--- a/arch/alpha/kernel/traps.c
+++ b/arch/alpha/kernel/traps.c
@@ -30,39 +30,6 @@
 
 #include "proto.h"
 
-/* Work-around for some SRMs which mishandle opDEC faults.  */
-
-static int opDEC_fix;
-
-static void
-opDEC_check(void)
-{
-	__asm__ __volatile__ (
-	/* Load the address of... */
-	"	br	$16, 1f\n"
-	/* A stub instruction fault handler.  Just add 4 to the
-	   pc and continue.  */
-	"	ldq	$16, 8($sp)\n"
-	"	addq	$16, 4, $16\n"
-	"	stq	$16, 8($sp)\n"
-	"	call_pal %[rti]\n"
-	/* Install the instruction fault handler.  */
-	"1:	lda	$17, 3\n"
-	"	call_pal %[wrent]\n"
-	/* With that in place, the fault from the round-to-minf fp
-	   insn will arrive either at the "lda 4" insn (bad) or one
-	   past that (good).  This places the correct fixup in %0.  */
-	"	lda %[fix], 0\n"
-	"	cvttq/svm $f31,$f31\n"
-	"	lda %[fix], 4"
-	: [fix] "=r" (opDEC_fix)
-	: [rti] "n" (PAL_rti), [wrent] "n" (PAL_wrent)
-	: "$0", "$1", "$16", "$17", "$22", "$23", "$24", "$25");
-
-	if (opDEC_fix)
-		printk("opDEC fixup enabled.\n");
-}
-
 void
 dik_show_regs(struct pt_regs *regs, unsigned long *r9_15)
 {
diff --git a/drivers/eisa/Kconfig b/drivers/eisa/Kconfig
index c8bbf90209f5..a66b3be502a9 100644
--- a/drivers/eisa/Kconfig
+++ b/drivers/eisa/Kconfig
@@ -44,17 +44,16 @@ config EISA_PCI_EISA
 
 	  When in doubt, say Y.
 
-# Using EISA_VIRTUAL_ROOT on something other than an Alpha or
-# an X86 may lead to crashes...
+# Using EISA_VIRTUAL_ROOT on something other than an X86 may lead
+# to crashes...
 
 config EISA_VIRTUAL_ROOT
 	bool "EISA virtual root device"
-	depends on EISA && (ALPHA || X86)
+	depends on EISA && X86
 	default y
 	help
 	  Activate this option if your system only have EISA bus
-	  (no PCI slots). The Alpha Jensen is an example of such
-	  a system.
+	  (no PCI slots).
 
 	  When in doubt, say Y.
 
diff --git a/drivers/eisa/virtual_root.c b/drivers/eisa/virtual_root.c
index 37e6dd219c37..cd9515d9d8f0 100644
--- a/drivers/eisa/virtual_root.c
+++ b/drivers/eisa/virtual_root.c
@@ -13,7 +13,7 @@
 #include <linux/moduleparam.h>
 #include <linux/init.h>
 
-#if defined(CONFIG_ALPHA_JENSEN) || defined(CONFIG_EISA_VLB_PRIMING)
+#if defined(CONFIG_EISA_VLB_PRIMING)
 #define EISA_FORCE_PROBE_DEFAULT 1
 #else
 #define EISA_FORCE_PROBE_DEFAULT 0
diff --git a/drivers/input/serio/i8042-io.h b/drivers/input/serio/i8042-io.h
index 64590b86eb37..a8f4b2d70e59 100644
--- a/drivers/input/serio/i8042-io.h
+++ b/drivers/input/serio/i8042-io.h
@@ -15,10 +15,7 @@
  * IRQs.
  */
 
-#ifdef __alpha__
-# define I8042_KBD_IRQ	1
-# define I8042_AUX_IRQ	(RTC_PORT(0) == 0x170 ? 9 : 12)	/* Jensen is special */
-#elif defined(__arm__)
+#if defined(__arm__)
 /* defined in include/asm-arm/arch-xxx/irqs.h */
 #include <asm/irq.h>
 #elif defined(CONFIG_PPC)
diff --git a/drivers/tty/serial/8250/8250.h b/drivers/tty/serial/8250/8250.h
index 1aa3e55c8b47..6a7b286f6f5a 100644
--- a/drivers/tty/serial/8250/8250.h
+++ b/drivers/tty/serial/8250/8250.h
@@ -293,9 +293,6 @@ static inline int serial8250_in_MCR(struct uart_8250_port *up)
 	return mctrl;
 }
 
-bool alpha_jensen(void);
-void alpha_jensen_set_mctrl(struct uart_port *port, unsigned int mctrl);
-
 #ifdef CONFIG_SERIAL_8250_PNP
 int serial8250_pnp_init(void);
 void serial8250_pnp_exit(void);
diff --git a/drivers/tty/serial/8250/8250_alpha.c b/drivers/tty/serial/8250/8250_alpha.c
deleted file mode 100644
index 58e70328aa4d..000000000000
--- a/drivers/tty/serial/8250/8250_alpha.c
+++ /dev/null
@@ -1,21 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0+
-
-#include <asm/machvec.h>
-#include "8250.h"
-
-bool alpha_jensen(void)
-{
-	return !strcmp(alpha_mv.vector_name, "Jensen");
-}
-
-void alpha_jensen_set_mctrl(struct uart_port *port, unsigned int mctrl)
-{
-	/*
-	 * Digital did something really horribly wrong with the OUT1 and OUT2
-	 * lines on Alpha Jensen.  The failure mode is that if either is
-	 * cleared, the machine locks up with endless interrupts.
-	 */
-	mctrl |= TIOCM_OUT1 | TIOCM_OUT2;
-
-	serial8250_do_set_mctrl(port, mctrl);
-}
diff --git a/drivers/tty/serial/8250/8250_core.c b/drivers/tty/serial/8250/8250_core.c
index b62ad9006780..2504e0455875 100644
--- a/drivers/tty/serial/8250/8250_core.c
+++ b/drivers/tty/serial/8250/8250_core.c
@@ -508,10 +508,6 @@ static struct uart_8250_port *serial8250_setup_port(int index)
 
 	up->ops = &univ8250_driver_ops;
 
-	if (IS_ENABLED(CONFIG_ALPHA_JENSEN) ||
-	    (IS_ENABLED(CONFIG_ALPHA_GENERIC) && alpha_jensen()))
-		up->port.set_mctrl = alpha_jensen_set_mctrl;
-
 	serial8250_set_defaults(up);
 
 	return up;
diff --git a/drivers/tty/serial/8250/Makefile b/drivers/tty/serial/8250/Makefile
index ea2e81f58eac..69ac00270547 100644
--- a/drivers/tty/serial/8250/Makefile
+++ b/drivers/tty/serial/8250/Makefile
@@ -5,8 +5,6 @@
 
 obj-$(CONFIG_SERIAL_8250)		+= 8250.o 8250_base.o
 8250-y					:= 8250_core.o
-8250-$(CONFIG_ALPHA_GENERIC)		+= 8250_alpha.o
-8250-$(CONFIG_ALPHA_JENSEN)		+= 8250_alpha.o
 8250-$(CONFIG_SERIAL_8250_PNP)		+= 8250_pnp.o
 8250_base-y				:= 8250_port.o
 8250_base-$(CONFIG_SERIAL_8250_DMA)	+= 8250_dma.o
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 11/14] alpha: sable: remove early machine support
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (9 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 10/14] alpha: remove DECpc AXP150 (Jensen) support Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 12/14] alpha: remove LCA and APECS based machines Arnd Bergmann
                   ` (4 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Arnd Bergmann <arnd@arndb.de>

The sable family (Alphaserver 2000 and 2100) comes in variants for
EV4, EV45, EV5 and EV56. Drop support for the earlier ones that
lack support for the BWX extension but keep the later 'gamma'
variant around since that works with EV56 CPUs.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/Kconfig               |  29 +--
 arch/alpha/include/asm/core_t2.h |   8 -
 arch/alpha/include/asm/irq.h     |   1 -
 arch/alpha/kernel/Makefile       |   1 -
 arch/alpha/kernel/setup.c        |  11 +-
 arch/alpha/kernel/sys_sable.c    | 294 +------------------------------
 6 files changed, 11 insertions(+), 333 deletions(-)

diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 9b8d3ce40c28..951a9c712361 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -100,7 +100,6 @@ choice
 	  EB66                EB66 21066 evaluation board
 	  EB66+               EB66+ 21066 evaluation board
 	  LX164               AlphaPC164-LX
-	  Lynx                AS 2100A
 	  Miata               Personal Workstation 433/500/600 a/au
 	  Marvel              AlphaServer ES47 / ES80 / GS1280
 	  Mikasa              AS 1000
@@ -203,12 +202,6 @@ config ALPHA_LX164
 	  A technical overview of this board is available at
 	  <http://www.unix-ag.org/Linux-Alpha/Architectures/LX164.html>.
 
-config ALPHA_LYNX
-	bool "Lynx"
-	select HAVE_EISA
-	help
-	  AlphaServer 2100A-based systems.
-
 config ALPHA_MARVEL
 	bool "Marvel"
 	help
@@ -325,9 +318,8 @@ config ALPHA_NONAME
 
 config ALPHA_EV4
 	bool
-	depends on (ALPHA_SABLE && !ALPHA_GAMMA) || ALPHA_LYNX || ALPHA_NORITAKE && !ALPHA_PRIMO || ALPHA_MIKASA && !ALPHA_PRIMO || ALPHA_CABRIOLET || ALPHA_AVANTI_CH || ALPHA_EB64P_CH || ALPHA_XL || ALPHA_NONAME || ALPHA_EB66 || ALPHA_EB66P || ALPHA_P2K
-	default y if !ALPHA_LYNX
-	default y if !ALPHA_EV5
+	depends on ALPHA_NORITAKE && !ALPHA_PRIMO || ALPHA_MIKASA && !ALPHA_PRIMO || ALPHA_CABRIOLET || ALPHA_AVANTI_CH || ALPHA_EB64P_CH || ALPHA_XL || ALPHA_NONAME || ALPHA_EB66 || ALPHA_EB66P || ALPHA_P2K
+	default y
 
 config ALPHA_LCA
 	bool
@@ -352,8 +344,8 @@ config ALPHA_EB64P
 	  Runs from standard PC power supply.
 
 config ALPHA_EV5
-	bool "EV5 CPU(s) (model 5/xxx)?" if ALPHA_LYNX
-	default y if ALPHA_RX164 || ALPHA_RAWHIDE || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_SABLE && ALPHA_GAMMA || ALPHA_NORITAKE && ALPHA_PRIMO || ALPHA_MIKASA && ALPHA_PRIMO || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR
+	bool
+	default y if ALPHA_RX164 || ALPHA_RAWHIDE || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_SABLE || ALPHA_NORITAKE && ALPHA_PRIMO || ALPHA_MIKASA && ALPHA_PRIMO || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR
 
 config ALPHA_CIA
 	bool
@@ -378,16 +370,9 @@ config ALPHA_PRIMO
 	help
 	  Say Y if you have an AS 1000 5/xxx or an AS 1000A 5/xxx.
 
-config ALPHA_GAMMA
-	bool "EV5 CPU(s) (model 5/xxx)?" if ALPHA_SABLE
-	depends on ALPHA_SABLE || ALPHA_LYNX
-	default ALPHA_LYNX
-	help
-	  Say Y if you have an AS 2000 5/xxx or an AS 2100 5/xxx.
-
 config ALPHA_T2
 	bool
-	depends on ALPHA_SABLE || ALPHA_LYNX
+	depends on ALPHA_SABLE
 	default y
 
 config ALPHA_PYXIS
@@ -471,7 +456,7 @@ config ALPHA_QEMU
 config ALPHA_SRM
 	bool "Use SRM as bootloader" if ALPHA_CABRIOLET || ALPHA_AVANTI_CH || ALPHA_EB64P || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_NAUTILUS || ALPHA_NONAME
 	depends on TTY
-	default y if ALPHA_MIKASA || ALPHA_SABLE || ALPHA_LYNX || ALPHA_NORITAKE || ALPHA_DP264 || ALPHA_RAWHIDE || ALPHA_EIGER || ALPHA_WILDFIRE || ALPHA_TITAN || ALPHA_SHARK || ALPHA_MARVEL
+	default y if ALPHA_MIKASA || ALPHA_SABLE || ALPHA_NORITAKE || ALPHA_DP264 || ALPHA_RAWHIDE || ALPHA_EIGER || ALPHA_WILDFIRE || ALPHA_TITAN || ALPHA_SHARK || ALPHA_MARVEL
 	help
 	  There are two different types of booting firmware on Alphas: SRM,
 	  which is command line driven, and ARC, which uses menus and arrow
@@ -497,7 +482,7 @@ config ARCH_MAY_HAVE_PC_FDC
 
 config SMP
 	bool "Symmetric multi-processing support"
-	depends on ALPHA_SABLE || ALPHA_LYNX || ALPHA_RAWHIDE || ALPHA_DP264 || ALPHA_WILDFIRE || ALPHA_TITAN || ALPHA_GENERIC || ALPHA_SHARK || ALPHA_MARVEL
+	depends on ALPHA_SABLE || ALPHA_RAWHIDE || ALPHA_DP264 || ALPHA_WILDFIRE || ALPHA_TITAN || ALPHA_GENERIC || ALPHA_SHARK || ALPHA_MARVEL
 	help
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, say N. If you have a system with more
diff --git a/arch/alpha/include/asm/core_t2.h b/arch/alpha/include/asm/core_t2.h
index ab956b1625b5..ca9b091d9c5f 100644
--- a/arch/alpha/include/asm/core_t2.h
+++ b/arch/alpha/include/asm/core_t2.h
@@ -25,16 +25,8 @@
 #define T2_MEM_R1_MASK 0x07ffffff  /* Mem sparse region 1 mask is 27 bits */
 
 /* GAMMA-SABLE is a SABLE with EV5-based CPUs */
-/* All LYNX machines, EV4 or EV5, use the GAMMA bias also */
 #define _GAMMA_BIAS		0x8000000000UL
-
-#if defined(CONFIG_ALPHA_GENERIC)
-#define GAMMA_BIAS		alpha_mv.sys.t2.gamma_bias
-#elif defined(CONFIG_ALPHA_GAMMA)
 #define GAMMA_BIAS		_GAMMA_BIAS
-#else
-#define GAMMA_BIAS		0
-#endif
 
 /*
  * Memory spaces:
diff --git a/arch/alpha/include/asm/irq.h b/arch/alpha/include/asm/irq.h
index 432402c8e47f..442097c56934 100644
--- a/arch/alpha/include/asm/irq.h
+++ b/arch/alpha/include/asm/irq.h
@@ -55,7 +55,6 @@
 # define NR_IRQS	40
 
 #elif defined(CONFIG_ALPHA_DP264) || \
-      defined(CONFIG_ALPHA_LYNX)  || \
       defined(CONFIG_ALPHA_SHARK)
 # define NR_IRQS	64
 
diff --git a/arch/alpha/kernel/Makefile b/arch/alpha/kernel/Makefile
index 4b31060e1711..706ce8625e35 100644
--- a/arch/alpha/kernel/Makefile
+++ b/arch/alpha/kernel/Makefile
@@ -88,7 +88,6 @@ obj-$(CONFIG_ALPHA_RAWHIDE)	+= sys_rawhide.o irq_i8259.o
 obj-$(CONFIG_ALPHA_RUFFIAN)	+= sys_ruffian.o irq_pyxis.o irq_i8259.o
 obj-$(CONFIG_ALPHA_RX164)	+= sys_rx164.o irq_i8259.o
 obj-$(CONFIG_ALPHA_SABLE)	+= sys_sable.o
-obj-$(CONFIG_ALPHA_LYNX)	+= sys_sable.o
 obj-$(CONFIG_ALPHA_BOOK1)	+= sys_sio.o irq_i8259.o irq_srm.o pc873xx.o
 obj-$(CONFIG_ALPHA_AVANTI)	+= sys_sio.o irq_i8259.o irq_srm.o pc873xx.o
 obj-$(CONFIG_ALPHA_NONAME)	+= sys_sio.o irq_i8259.o irq_srm.o pc873xx.o
diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index 6e6c842ef07c..3b833bd97817 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -182,7 +182,6 @@ WEAK(eb66_mv);
 WEAK(eb66p_mv);
 WEAK(eiger_mv);
 WEAK(lx164_mv);
-WEAK(lynx_mv);
 WEAK(marvel_ev7_mv);
 WEAK(miata_mv);
 WEAK(mikasa_mv);
@@ -198,7 +197,6 @@ WEAK(privateer_mv);
 WEAK(rawhide_mv);
 WEAK(ruffian_mv);
 WEAK(rx164_mv);
-WEAK(sable_mv);
 WEAK(sable_gamma_mv);
 WEAK(shark_mv);
 WEAK(sx164_mv);
@@ -751,7 +749,7 @@ get_sysvec(unsigned long type, unsigned long variation, unsigned long cpu)
 		&alphabook1_mv,
 		&rawhide_mv,
 		NULL,		/* K2 */
-		&lynx_mv,	/* Lynx */
+		NULL,		/* Lynx */
 		&xl_mv,
 		NULL,		/* EB164 -- see variation.  */
 		NULL,		/* Noritake -- see below.  */
@@ -905,10 +903,7 @@ get_sysvec(unsigned long type, unsigned long variation, unsigned long cpu)
 				vec = &noritake_mv;
 			break;
 		case ST_DEC_2100_A500:
-			if (cpu == EV5_CPU || cpu == EV56_CPU)
-				vec = &sable_gamma_mv;
-			else
-				vec = &sable_mv;
+			vec = &sable_gamma_mv;
 			break;
 		}
 	}
@@ -932,7 +927,6 @@ get_sysvec_byname(const char *name)
 		&eb66p_mv,
 		&eiger_mv,
 		&lx164_mv,
-		&lynx_mv,
 		&miata_mv,
 		&mikasa_mv,
 		&mikasa_primo_mv,
@@ -947,7 +941,6 @@ get_sysvec_byname(const char *name)
 		&rawhide_mv,
 		&ruffian_mv,
 		&rx164_mv,
-		&sable_mv,
 		&sable_gamma_mv,
 		&shark_mv,
 		&sx164_mv,
diff --git a/arch/alpha/kernel/sys_sable.c b/arch/alpha/kernel/sys_sable.c
index 930005b2f630..49f5c75134ec 100644
--- a/arch/alpha/kernel/sys_sable.c
+++ b/arch/alpha/kernel/sys_sable.c
@@ -212,232 +212,6 @@ sable_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
 }
 #endif /* defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_SABLE) */
 
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_LYNX)
-
-/***********************************************************************/
-/* LYNX hardware specifics
- */
-/*
- *   For LYNX, which is also baroque, we manage 64 IRQs, via a custom IC.
- *
- * Bit      Meaning               Kernel IRQ
- *------------------------------------------
- * 0        
- * 1        
- * 2        
- * 3        mouse			12
- * 4        
- * 5        
- * 6        keyboard			1
- * 7        floppy			6
- * 8        COM2			3
- * 9        parallel port		7
- *10        EISA irq 3			-
- *11        EISA irq 4			-
- *12        EISA irq 5			5
- *13        EISA irq 6			-
- *14        EISA irq 7			-
- *15        COM1			4
- *16        EISA irq 9			9
- *17        EISA irq 10			10
- *18        EISA irq 11			11
- *19        EISA irq 12			-
- *20        
- *21        EISA irq 14			14
- *22        EISA irq 15			15
- *23        IIC				-
- *24        VGA (builtin)               -
- *25
- *26
- *27
- *28        NCR810 (builtin)		28
- *29
- *30
- *31
- *32        PCI 0 slot 4 A primary bus  32
- *33        PCI 0 slot 4 B primary bus  33
- *34        PCI 0 slot 4 C primary bus  34
- *35        PCI 0 slot 4 D primary bus
- *36        PCI 0 slot 5 A primary bus
- *37        PCI 0 slot 5 B primary bus
- *38        PCI 0 slot 5 C primary bus
- *39        PCI 0 slot 5 D primary bus
- *40        PCI 0 slot 6 A primary bus
- *41        PCI 0 slot 6 B primary bus
- *42        PCI 0 slot 6 C primary bus
- *43        PCI 0 slot 6 D primary bus
- *44        PCI 0 slot 7 A primary bus
- *45        PCI 0 slot 7 B primary bus
- *46        PCI 0 slot 7 C primary bus
- *47        PCI 0 slot 7 D primary bus
- *48        PCI 0 slot 0 A secondary bus
- *49        PCI 0 slot 0 B secondary bus
- *50        PCI 0 slot 0 C secondary bus
- *51        PCI 0 slot 0 D secondary bus
- *52        PCI 0 slot 1 A secondary bus
- *53        PCI 0 slot 1 B secondary bus
- *54        PCI 0 slot 1 C secondary bus
- *55        PCI 0 slot 1 D secondary bus
- *56        PCI 0 slot 2 A secondary bus
- *57        PCI 0 slot 2 B secondary bus
- *58        PCI 0 slot 2 C secondary bus
- *59        PCI 0 slot 2 D secondary bus
- *60        PCI 0 slot 3 A secondary bus
- *61        PCI 0 slot 3 B secondary bus
- *62        PCI 0 slot 3 C secondary bus
- *63        PCI 0 slot 3 D secondary bus
- */
-
-static void
-lynx_update_irq_hw(unsigned long bit, unsigned long mask)
-{
-	/*
-	 * Write the AIR register on the T3/T4 with the
-	 * address of the IC mask register (offset 0x40)
-	 */
-	*(vulp)T2_AIR = 0x40;
-	mb();
-	*(vulp)T2_AIR; /* re-read to force write */
-	mb();
-	*(vulp)T2_DIR = mask;    
-	mb();
-	mb();
-}
-
-static void
-lynx_ack_irq_hw(unsigned long bit)
-{
-	*(vulp)T2_VAR = (u_long) bit;
-	mb();
-	mb();
-}
-
-static irq_swizzle_t lynx_irq_swizzle = {
-	{ /* irq_to_mask */
-		-1,  6, -1,  8, 15, 12,  7,  9,	/* pseudo PIC  0-7  */
-		-1, 16, 17, 18,  3, -1, 21, 22,	/* pseudo PIC  8-15 */
-		-1, -1, -1, -1, -1, -1, -1, -1,	/* pseudo */
-		-1, -1, -1, -1, 28, -1, -1, -1,	/* pseudo */
-		32, 33, 34, 35, 36, 37, 38, 39,	/* mask 32-39 */
-		40, 41, 42, 43, 44, 45, 46, 47,	/* mask 40-47 */
-		48, 49, 50, 51, 52, 53, 54, 55,	/* mask 48-55 */
-		56, 57, 58, 59, 60, 61, 62, 63	/* mask 56-63 */
-	},
-	{ /* mask_to_irq */
-		-1, -1, -1, 12, -1, -1,  1,  6,	/* mask 0-7   */
-		 3,  7, -1, -1,  5, -1, -1,  4,	/* mask 8-15  */
-		 9, 10, 11, -1, -1, 14, 15, -1,	/* mask 16-23 */
-		-1, -1, -1, -1, 28, -1, -1, -1,	/* mask 24-31 */
-		32, 33, 34, 35, 36, 37, 38, 39,	/* mask 32-39 */
-		40, 41, 42, 43, 44, 45, 46, 47,	/* mask 40-47 */
-		48, 49, 50, 51, 52, 53, 54, 55,	/* mask 48-55 */
-		56, 57, 58, 59, 60, 61, 62, 63	/* mask 56-63 */
-	},
-	-1,
-	lynx_update_irq_hw,
-	lynx_ack_irq_hw
-};
-
-static void __init
-lynx_init_irq(void)
-{
-	sable_lynx_irq_swizzle = &lynx_irq_swizzle;
-	sable_lynx_init_irq(64);
-}
-
-/*
- * PCI Fixup configuration for ALPHA LYNX (2100A)
- *
- * The device to slot mapping looks like:
- *
- * Slot     Device
- *  0       none
- *  1       none
- *  2       PCI-EISA bridge
- *  3       PCI-PCI bridge
- *  4       NCR 810 (Demi-Lynx only)
- *  5       none
- *  6       PCI on board slot 4
- *  7       PCI on board slot 5
- *  8       PCI on board slot 6
- *  9       PCI on board slot 7
- *
- * And behind the PPB we have:
- *
- * 11       PCI on board slot 0
- * 12       PCI on board slot 1
- * 13       PCI on board slot 2
- * 14       PCI on board slot 3
- */
-/*
- * NOTE: the IRQ assignments below are arbitrary, but need to be consistent
- * with the values in the irq swizzling tables above.
- */
-
-static int
-lynx_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
-{
-	static char irq_tab[19][5] = {
-		/*INT    INTA   INTB   INTC   INTD */
-		{   -1,    -1,    -1,    -1,    -1},  /* IdSel 13,  PCEB   */
-		{   -1,    -1,    -1,    -1,    -1},  /* IdSel 14,  PPB    */
-		{   28,    28,    28,    28,    28},  /* IdSel 15,  NCR demi */
-		{   -1,    -1,    -1,    -1,    -1},  /* IdSel 16,  none   */
-		{   32,    32,    33,    34,    35},  /* IdSel 17,  slot 4 */
-		{   36,    36,    37,    38,    39},  /* IdSel 18,  slot 5 */
-		{   40,    40,    41,    42,    43},  /* IdSel 19,  slot 6 */
-		{   44,    44,    45,    46,    47},  /* IdSel 20,  slot 7 */
-		{   -1,    -1,    -1,    -1,    -1},  /* IdSel 22,  none   */
-		/* The following are actually behind the PPB. */
-		{   -1,    -1,    -1,    -1,    -1},  /* IdSel 16   none */
-		{   28,    28,    28,    28,    28},  /* IdSel 17   NCR lynx */
-		{   -1,    -1,    -1,    -1,    -1},  /* IdSel 18   none */
-		{   -1,    -1,    -1,    -1,    -1},  /* IdSel 19   none */
-		{   -1,    -1,    -1,    -1,    -1},  /* IdSel 20   none */
-		{   -1,    -1,    -1,    -1,    -1},  /* IdSel 21   none */
-		{   48,    48,    49,    50,    51},  /* IdSel 22   slot 0 */
-		{   52,    52,    53,    54,    55},  /* IdSel 23   slot 1 */
-		{   56,    56,    57,    58,    59},  /* IdSel 24   slot 2 */
-		{   60,    60,    61,    62,    63}   /* IdSel 25   slot 3 */
-	};
-	const long min_idsel = 2, max_idsel = 20, irqs_per_slot = 5;
-	return COMMON_TABLE_LOOKUP;
-}
-
-static u8
-lynx_swizzle(struct pci_dev *dev, u8 *pinp)
-{
-	int slot, pin = *pinp;
-
-	if (dev->bus->number == 0) {
-		slot = PCI_SLOT(dev->devfn);
-	}
-	/* Check for the built-in bridge */
-	else if (PCI_SLOT(dev->bus->self->devfn) == 3) {
-		slot = PCI_SLOT(dev->devfn) + 11;
-	}
-	else
-	{
-		/* Must be a card-based bridge.  */
-		do {
-			if (PCI_SLOT(dev->bus->self->devfn) == 3) {
-				slot = PCI_SLOT(dev->devfn) + 11;
-				break;
-			}
-			pin = pci_swizzle_interrupt_pin(dev, pin);
-
-			/* Move up the chain of bridges.  */
-			dev = dev->bus->self;
-			/* Slot of the next bridge.  */
-			slot = PCI_SLOT(dev->devfn);
-		} while (dev->bus->self);
-	}
-	*pinp = pin;
-	return slot;
-}
-
-#endif /* defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_LYNX) */
-
 /***********************************************************************/
 /* GENERIC irq routines */
 
@@ -539,40 +313,7 @@ sable_lynx_init_pci(void)
  * these games with GAMMA_BIAS.
  */
 
-#if defined(CONFIG_ALPHA_GENERIC) || \
-    (defined(CONFIG_ALPHA_SABLE) && !defined(CONFIG_ALPHA_GAMMA))
-#undef GAMMA_BIAS
-#define GAMMA_BIAS 0
-struct alpha_machine_vector sable_mv __initmv = {
-	.vector_name		= "Sable",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_T2_IO,
-	.machine_check		= t2_machine_check,
-	.max_isa_dma_address	= ALPHA_SABLE_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= EISA_DEFAULT_IO_BASE,
-	.min_mem_address	= T2_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 40,
-	.device_interrupt	= sable_lynx_srm_device_interrupt,
-
-	.init_arch		= t2_init_arch,
-	.init_irq		= sable_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= sable_lynx_init_pci,
-	.kill_arch		= t2_kill_arch,
-	.pci_map_irq		= sable_map_irq,
-	.pci_swizzle		= common_swizzle,
-
-	.sys = { .t2 = {
-	    .gamma_bias		= 0
-	} }
-};
-ALIAS_MV(sable)
-#endif /* GENERIC || (SABLE && !GAMMA) */
-
-#if defined(CONFIG_ALPHA_GENERIC) || \
-    (defined(CONFIG_ALPHA_SABLE) && defined(CONFIG_ALPHA_GAMMA))
+#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_SABLE)
 #undef GAMMA_BIAS
 #define GAMMA_BIAS _GAMMA_BIAS
 struct alpha_machine_vector sable_gamma_mv __initmv = {
@@ -601,35 +342,4 @@ struct alpha_machine_vector sable_gamma_mv __initmv = {
 	} }
 };
 ALIAS_MV(sable_gamma)
-#endif /* GENERIC || (SABLE && GAMMA) */
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_LYNX)
-#undef GAMMA_BIAS
-#define GAMMA_BIAS _GAMMA_BIAS
-struct alpha_machine_vector lynx_mv __initmv = {
-	.vector_name		= "Lynx",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_T2_IO,
-	.machine_check		= t2_machine_check,
-	.max_isa_dma_address	= ALPHA_SABLE_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= EISA_DEFAULT_IO_BASE,
-	.min_mem_address	= T2_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 64,
-	.device_interrupt	= sable_lynx_srm_device_interrupt,
-
-	.init_arch		= t2_init_arch,
-	.init_irq		= lynx_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= sable_lynx_init_pci,
-	.kill_arch		= t2_kill_arch,
-	.pci_map_irq		= lynx_map_irq,
-	.pci_swizzle		= lynx_swizzle,
-
-	.sys = { .t2 = {
-	    .gamma_bias		= _GAMMA_BIAS
-	} }
-};
-ALIAS_MV(lynx)
-#endif /* GENERIC || LYNX */
+#endif /* GENERIC || SABLE */
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 12/14] alpha: remove LCA and APECS based machines
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (10 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 11/14] alpha: sable: remove early machine support Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 13/14] alpha: cabriolet: remove EV5 CPU support Arnd Bergmann
                   ` (3 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Arnd Bergmann <arnd@arndb.de>

APECS is the DECchip 21071x chipset for the EV4 and EV45 generation, while
LCA is the integrated I/O support on the corresponding low-cost alpha
machines of that generation.

All of these CPUs lack the BWX extension for byte and word access, so
drop the chipset support and all associated machines.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/Kconfig                  | 114 +-----
 arch/alpha/include/asm/core_apecs.h | 534 ----------------------------
 arch/alpha/include/asm/core_lca.h   | 378 --------------------
 arch/alpha/include/asm/dma.h        |   9 +-
 arch/alpha/include/asm/io.h         |   6 +-
 arch/alpha/include/asm/irq.h        |   8 +-
 arch/alpha/kernel/Makefile          |  19 +-
 arch/alpha/kernel/core_apecs.c      | 420 ----------------------
 arch/alpha/kernel/core_lca.c        | 473 ------------------------
 arch/alpha/kernel/irq_i8259.c       |   4 -
 arch/alpha/kernel/pci_impl.h        |   4 +-
 arch/alpha/kernel/proto.h           |  13 -
 arch/alpha/kernel/setup.c           |  79 +---
 arch/alpha/kernel/sys_cabriolet.c   |  59 ---
 arch/alpha/kernel/sys_eb64p.c       | 238 -------------
 arch/alpha/kernel/sys_mikasa.c      |  57 ---
 arch/alpha/kernel/sys_noritake.c    |  60 ----
 arch/alpha/kernel/sys_sio.c         | 486 -------------------------
 18 files changed, 21 insertions(+), 2940 deletions(-)
 delete mode 100644 arch/alpha/include/asm/core_apecs.h
 delete mode 100644 arch/alpha/include/asm/core_lca.h
 delete mode 100644 arch/alpha/kernel/core_apecs.c
 delete mode 100644 arch/alpha/kernel/core_lca.c
 delete mode 100644 arch/alpha/kernel/sys_eb64p.c
 delete mode 100644 arch/alpha/kernel/sys_sio.c

diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 951a9c712361..ee209e895505 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -90,20 +90,12 @@ choice
 	  <http://www.alphalinux.org/>. In summary:
 
 	  Alcor/Alpha-XLT     AS 600, AS 500, XL-300, XL-366
-	  Alpha-XL            XL-233, XL-266
-	  AlphaBook1          Alpha laptop
-	  Avanti              AS 200, AS 205, AS 250, AS 255, AS 300, AS 400
-	  Cabriolet           AlphaPC64, AlphaPCI64
 	  DP264               DP264 / DS20 / ES40 / DS10 / DS10L
 	  EB164               EB164 21164 evaluation board
-	  EB64+               EB64+ 21064 evaluation board
-	  EB66                EB66 21066 evaluation board
-	  EB66+               EB66+ 21066 evaluation board
 	  LX164               AlphaPC164-LX
 	  Miata               Personal Workstation 433/500/600 a/au
 	  Marvel              AlphaServer ES47 / ES80 / GS1280
 	  Mikasa              AS 1000
-	  Noname              AXPpci33, UDB (Multia)
 	  Noritake            AS 1000A, AS 600A, AS 800
 	  PC164               AlphaPC164
 	  Rawhide             AS 1200, AS 4000, AS 4100
@@ -135,27 +127,6 @@ config ALPHA_ALCOR
 	  all the work required to support an external Bcache and to maintain
 	  memory coherence when a PCI device DMAs into (or out of) memory.
 
-config ALPHA_XL
-	bool "Alpha-XL"
-	help
-	  XL-233 and XL-266-based Alpha systems.
-
-config ALPHA_BOOK1
-	bool "AlphaBook1"
-	help
-	  Dec AlphaBook1/Burns Alpha-based laptops.
-
-config ALPHA_AVANTI_CH
-	bool "Avanti"
-
-config ALPHA_CABRIOLET
-	bool "Cabriolet"
-	help
-	  Cabriolet AlphaPC64, AlphaPCI64 systems.  Derived from EB64+ but now
-	  baby-AT with Flash boot ROM, no on-board SCSI or Ethernet. 3 ISA
-	  slots, 4 PCI slots (one pair are on a shared slot), uses plug-in
-	  Bcache SIMMs.  Requires power supply with 3.3V output.
-
 config ALPHA_DP264
 	bool "DP264"
 	help
@@ -173,23 +144,6 @@ config ALPHA_EB164
 	  MOUSE (PS2 style), RTC/NVRAM.  Boot ROM is Flash.  PC-AT-sized
 	  motherboard.  Requires power supply with 3.3V output.
 
-config ALPHA_EB64P_CH
-	bool "EB64+"
-
-config ALPHA_EB66
-	bool "EB66"
-	help
-	  A Digital DS group board.  Uses 21066 or 21066A.  I/O sub-system is
-	  identical to EB64+.  Baby PC-AT size.  Runs from standard PC power
-	  supply.  The EB66 schematic was published as a marketing poster
-	  advertising the 21066 as "the first microprocessor in the world with
-	  embedded PCI".
-
-config ALPHA_EB66P
-	bool "EB66+"
-	help
-	  Later variant of the EB66 board.
-
 config ALPHA_EIGER
 	bool "Eiger"
 	help
@@ -224,9 +178,6 @@ config ALPHA_NAUTILUS
 	help
 	  Alpha systems based on the AMD 751 & ALI 1543C chipsets.
 
-config ALPHA_NONAME_CH
-	bool "Noname"
-
 config ALPHA_NORITAKE
 	bool "Noritake"
 	select HAVE_EISA
@@ -237,9 +188,6 @@ config ALPHA_NORITAKE
 config ALPHA_PC164
 	bool "PC164"
 
-config ALPHA_P2K
-	bool "Platform2000"
-
 config ALPHA_RAWHIDE
 	bool "Rawhide"
 	select HAVE_EISA
@@ -303,53 +251,16 @@ config ISA_DMA_API
 	bool
 	default y
 
-config ALPHA_NONAME
-	bool
-	depends on ALPHA_BOOK1 || ALPHA_NONAME_CH
-	default y
-	help
-	  The AXPpci33 (aka NoName), is based on the EB66 (includes the Multia
-	  UDB).  This design was produced by Digital's Technical OEM (TOEM)
-	  group. It uses the 21066 processor running at 166MHz or 233MHz. It
-	  is a baby-AT size, and runs from a standard PC power supply. It has
-	  5 ISA slots and 3 PCI slots (one pair are a shared slot). There are
-	  2 versions, with either PS/2 or large DIN connectors for the
-	  keyboard.
-
 config ALPHA_EV4
 	bool
-	depends on ALPHA_NORITAKE && !ALPHA_PRIMO || ALPHA_MIKASA && !ALPHA_PRIMO || ALPHA_CABRIOLET || ALPHA_AVANTI_CH || ALPHA_EB64P_CH || ALPHA_XL || ALPHA_NONAME || ALPHA_EB66 || ALPHA_EB66P || ALPHA_P2K
-	default y
-
-config ALPHA_LCA
-	bool
-	depends on ALPHA_NONAME || ALPHA_EB66 || ALPHA_EB66P || ALPHA_P2K
-	default y
-
-config ALPHA_APECS
-	bool
-	depends on !ALPHA_PRIMO && (ALPHA_NORITAKE || ALPHA_MIKASA) || ALPHA_CABRIOLET || ALPHA_AVANTI_CH || ALPHA_EB64P_CH || ALPHA_XL
-	default y
-
-config ALPHA_EB64P
-	bool
-	depends on ALPHA_CABRIOLET || ALPHA_EB64P_CH
-	default y
-	help
-	  Uses 21064 or 21064A and APECs.  Has ISA and PCI expansion (3 ISA,
-	  2 PCI, one pair are on a shared slot). Supports 36-bit DRAM SIMs.
-	  ISA bus generated by Intel SaturnI/O PCI-ISA bridge. On-board SCSI
-	  (NCR 810 on PCI) Ethernet (Digital 21040), KBD, MOUSE (PS2 style),
-	  SuperI/O (2S, 1P, FD), RTC/NVRAM. Boot ROM is EPROM.  PC-AT size.
-	  Runs from standard PC power supply.
 
 config ALPHA_EV5
 	bool
-	default y if ALPHA_RX164 || ALPHA_RAWHIDE || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_SABLE || ALPHA_NORITAKE && ALPHA_PRIMO || ALPHA_MIKASA && ALPHA_PRIMO || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR
+	default y if ALPHA_RX164 || ALPHA_RAWHIDE || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_SABLE || ALPHA_NORITAKE || ALPHA_MIKASA || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR
 
 config ALPHA_CIA
 	bool
-	depends on ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_NORITAKE && ALPHA_PRIMO || ALPHA_MIKASA && ALPHA_PRIMO || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR
+	depends on ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_NORITAKE || ALPHA_MIKASA || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR
 	default y
 
 config ALPHA_EV56
@@ -358,18 +269,12 @@ config ALPHA_EV56
 
 config ALPHA_EV56
 	prompt "EV56 CPU (speed >= 333MHz)?"
-	depends on ALPHA_NORITAKE || ALPHA_PRIMO
+	depends on ALPHA_NORITAKE || ALPHA_MIKASA
 
 config ALPHA_EV56
 	prompt "EV56 CPU (speed >= 400MHz)?"
 	depends on ALPHA_RAWHIDE
 
-config ALPHA_PRIMO
-	bool "EV5 CPU daughtercard (model 5/xxx)?"
-	depends on ALPHA_NORITAKE || ALPHA_MIKASA
-	help
-	  Say Y if you have an AS 1000 5/xxx or an AS 1000A 5/xxx.
-
 config ALPHA_T2
 	bool
 	depends on ALPHA_SABLE
@@ -416,15 +321,6 @@ config GENERIC_HWEIGHT
 	bool
 	default y if !ALPHA_EV67
 
-config ALPHA_AVANTI
-	bool
-	depends on ALPHA_XL || ALPHA_AVANTI_CH
-	default y
-	help
-	  Avanti AS 200, AS 205, AS 250, AS 255, AS 300, and AS 400-based
-	  Alphas. Info at
-	  <http://www.unix-ag.org/Linux-Alpha/Architectures/Avanti.html>.
-
 config ALPHA_BROKEN_IRQ_MASK
 	bool
 	depends on ALPHA_GENERIC || ALPHA_PC164
@@ -454,7 +350,7 @@ config ALPHA_QEMU
 
 
 config ALPHA_SRM
-	bool "Use SRM as bootloader" if ALPHA_CABRIOLET || ALPHA_AVANTI_CH || ALPHA_EB64P || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_NAUTILUS || ALPHA_NONAME
+	bool "Use SRM as bootloader" if ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_NAUTILUS
 	depends on TTY
 	default y if ALPHA_MIKASA || ALPHA_SABLE || ALPHA_NORITAKE || ALPHA_DP264 || ALPHA_RAWHIDE || ALPHA_EIGER || ALPHA_WILDFIRE || ALPHA_TITAN || ALPHA_SHARK || ALPHA_MARVEL
 	help
@@ -518,7 +414,7 @@ config ARCH_SPARSEMEM_ENABLE
 config ALPHA_WTINT
 	bool "Use WTINT" if ALPHA_SRM || ALPHA_GENERIC
 	default y if ALPHA_QEMU
-	default n if ALPHA_EV5 || ALPHA_EV56 || (ALPHA_EV4 && !ALPHA_LCA)
+	default n if ALPHA_EV5 || ALPHA_EV56
 	default n if !ALPHA_SRM && !ALPHA_GENERIC
 	default y if SMP
 	help
diff --git a/arch/alpha/include/asm/core_apecs.h b/arch/alpha/include/asm/core_apecs.h
deleted file mode 100644
index 69a2fc62c9c3..000000000000
--- a/arch/alpha/include/asm/core_apecs.h
+++ /dev/null
@@ -1,534 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef __ALPHA_APECS__H__
-#define __ALPHA_APECS__H__
-
-#include <linux/types.h>
-#include <asm/compiler.h>
-
-/*
- * APECS is the internal name for the 2107x chipset which provides
- * memory controller and PCI access for the 21064 chip based systems.
- *
- * This file is based on:
- *
- * DECchip 21071-AA and DECchip 21072-AA Core Logic Chipsets
- * Data Sheet
- *
- * EC-N0648-72
- *
- *
- * david.rusling@reo.mts.dec.com Initial Version.
- *
- */
-
-/*
-   An AVANTI *might* be an XL, and an XL has only 27 bits of ISA address
-   that get passed through the PCI<->ISA bridge chip. So we've gotta use
-   both windows to max out the physical memory we can DMA to. Sigh...
-
-   If we try a window at 0 for 1GB as a work-around, we run into conflicts
-   with ISA/PCI bus memory which can't be relocated, like VGA aperture and
-   BIOS ROMs. So we must put the windows high enough to avoid these areas.
-
-   We put window 1 at BUS 64Mb for 64Mb, mapping physical 0 to 64Mb-1,
-   and window 2 at BUS 1Gb for 1Gb, mapping physical 0 to 1Gb-1.
-   Yes, this does map 0 to 64Mb-1 twice, but only window 1 will actually
-   be used for that range (via virt_to_bus()).
-
-   Note that we actually fudge the window 1 maximum as 48Mb instead of 64Mb,
-   to keep virt_to_bus() from returning an address in the first window, for
-   a data area that goes beyond the 64Mb first DMA window.  Sigh...
-   The fudge factor MUST match with <asm/dma.h> MAX_DMA_ADDRESS, but
-   we can't just use that here, because of header file looping... :-(
-
-   Window 1 will be used for all DMA from the ISA bus; yes, that does
-   limit what memory an ISA floppy or sound card or Ethernet can touch, but
-   it's also a known limitation on other platforms as well. We use the
-   same technique that is used on INTEL platforms with similar limitation:
-   set MAX_DMA_ADDRESS and clear some pages' DMAable flags during mem_init().
-   We trust that any ISA bus device drivers will *always* ask for DMAable
-   memory explicitly via kmalloc()/get_free_pages() flags arguments.
-
-   Note that most PCI bus devices' drivers do *not* explicitly ask for
-   DMAable memory; they count on being able to DMA to any memory they
-   get from kmalloc()/get_free_pages(). They will also use window 1 for
-   any physical memory accesses below 64Mb; the rest will be handled by
-   window 2, maxing out at 1Gb of memory. I trust this is enough... :-)
-
-   We hope that the area before the first window is large enough so that
-   there will be no overlap at the top end (64Mb). We *must* locate the
-   PCI cards' memory just below window 1, so that there's still the
-   possibility of being able to access it via SPARSE space. This is
-   important for cards such as the Matrox Millennium, whose Xserver
-   wants to access memory-mapped registers in byte and short lengths.
-
-   Note that the XL is treated differently from the AVANTI, even though
-   for most other things they are identical. It didn't seem reasonable to
-   make the AVANTI support pay for the limitations of the XL. It is true,
-   however, that an XL kernel will run on an AVANTI without problems.
-
-   %%% All of this should be obviated by the ability to route
-   everything through the iommu.
-*/
-
-/*
- * 21071-DA Control and Status registers.
- * These are used for PCI memory access.
- */
-#define APECS_IOC_DCSR                  (IDENT_ADDR + 0x1A0000000UL)
-#define APECS_IOC_PEAR                  (IDENT_ADDR + 0x1A0000020UL)
-#define APECS_IOC_SEAR                  (IDENT_ADDR + 0x1A0000040UL)
-#define APECS_IOC_DR1                   (IDENT_ADDR + 0x1A0000060UL)
-#define APECS_IOC_DR2                   (IDENT_ADDR + 0x1A0000080UL)
-#define APECS_IOC_DR3                   (IDENT_ADDR + 0x1A00000A0UL)
-
-#define APECS_IOC_TB1R                  (IDENT_ADDR + 0x1A00000C0UL)
-#define APECS_IOC_TB2R                  (IDENT_ADDR + 0x1A00000E0UL)
-
-#define APECS_IOC_PB1R                  (IDENT_ADDR + 0x1A0000100UL)
-#define APECS_IOC_PB2R                  (IDENT_ADDR + 0x1A0000120UL)
-
-#define APECS_IOC_PM1R                  (IDENT_ADDR + 0x1A0000140UL)
-#define APECS_IOC_PM2R                  (IDENT_ADDR + 0x1A0000160UL)
-
-#define APECS_IOC_HAXR0                 (IDENT_ADDR + 0x1A0000180UL)
-#define APECS_IOC_HAXR1                 (IDENT_ADDR + 0x1A00001A0UL)
-#define APECS_IOC_HAXR2                 (IDENT_ADDR + 0x1A00001C0UL)
-
-#define APECS_IOC_PMLT                  (IDENT_ADDR + 0x1A00001E0UL)
-
-#define APECS_IOC_TLBTAG0               (IDENT_ADDR + 0x1A0000200UL)
-#define APECS_IOC_TLBTAG1               (IDENT_ADDR + 0x1A0000220UL)
-#define APECS_IOC_TLBTAG2               (IDENT_ADDR + 0x1A0000240UL)
-#define APECS_IOC_TLBTAG3               (IDENT_ADDR + 0x1A0000260UL)
-#define APECS_IOC_TLBTAG4               (IDENT_ADDR + 0x1A0000280UL)
-#define APECS_IOC_TLBTAG5               (IDENT_ADDR + 0x1A00002A0UL)
-#define APECS_IOC_TLBTAG6               (IDENT_ADDR + 0x1A00002C0UL)
-#define APECS_IOC_TLBTAG7               (IDENT_ADDR + 0x1A00002E0UL)
-
-#define APECS_IOC_TLBDATA0              (IDENT_ADDR + 0x1A0000300UL)
-#define APECS_IOC_TLBDATA1              (IDENT_ADDR + 0x1A0000320UL)
-#define APECS_IOC_TLBDATA2              (IDENT_ADDR + 0x1A0000340UL)
-#define APECS_IOC_TLBDATA3              (IDENT_ADDR + 0x1A0000360UL)
-#define APECS_IOC_TLBDATA4              (IDENT_ADDR + 0x1A0000380UL)
-#define APECS_IOC_TLBDATA5              (IDENT_ADDR + 0x1A00003A0UL)
-#define APECS_IOC_TLBDATA6              (IDENT_ADDR + 0x1A00003C0UL)
-#define APECS_IOC_TLBDATA7              (IDENT_ADDR + 0x1A00003E0UL)
-
-#define APECS_IOC_TBIA                  (IDENT_ADDR + 0x1A0000400UL)
-
-
-/*
- * 21071-CA Control and Status registers.
- * These are used to program memory timing,
- *  configure memory and initialise the B-Cache.
- */
-#define APECS_MEM_GCR		        (IDENT_ADDR + 0x180000000UL)
-#define APECS_MEM_EDSR		        (IDENT_ADDR + 0x180000040UL)
-#define APECS_MEM_TAR  		        (IDENT_ADDR + 0x180000060UL)
-#define APECS_MEM_ELAR		        (IDENT_ADDR + 0x180000080UL)
-#define APECS_MEM_EHAR  		(IDENT_ADDR + 0x1800000a0UL)
-#define APECS_MEM_SFT_RST		(IDENT_ADDR + 0x1800000c0UL)
-#define APECS_MEM_LDxLAR 		(IDENT_ADDR + 0x1800000e0UL)
-#define APECS_MEM_LDxHAR 		(IDENT_ADDR + 0x180000100UL)
-#define APECS_MEM_GTR    		(IDENT_ADDR + 0x180000200UL)
-#define APECS_MEM_RTR    		(IDENT_ADDR + 0x180000220UL)
-#define APECS_MEM_VFPR   		(IDENT_ADDR + 0x180000240UL)
-#define APECS_MEM_PDLDR  		(IDENT_ADDR + 0x180000260UL)
-#define APECS_MEM_PDhDR  		(IDENT_ADDR + 0x180000280UL)
-
-/* Bank x Base Address Register */
-#define APECS_MEM_B0BAR  		(IDENT_ADDR + 0x180000800UL)
-#define APECS_MEM_B1BAR  		(IDENT_ADDR + 0x180000820UL)
-#define APECS_MEM_B2BAR  		(IDENT_ADDR + 0x180000840UL)
-#define APECS_MEM_B3BAR  		(IDENT_ADDR + 0x180000860UL)
-#define APECS_MEM_B4BAR  		(IDENT_ADDR + 0x180000880UL)
-#define APECS_MEM_B5BAR  		(IDENT_ADDR + 0x1800008A0UL)
-#define APECS_MEM_B6BAR  		(IDENT_ADDR + 0x1800008C0UL)
-#define APECS_MEM_B7BAR  		(IDENT_ADDR + 0x1800008E0UL)
-#define APECS_MEM_B8BAR  		(IDENT_ADDR + 0x180000900UL)
-
-/* Bank x Configuration Register */
-#define APECS_MEM_B0BCR  		(IDENT_ADDR + 0x180000A00UL)
-#define APECS_MEM_B1BCR  		(IDENT_ADDR + 0x180000A20UL)
-#define APECS_MEM_B2BCR  		(IDENT_ADDR + 0x180000A40UL)
-#define APECS_MEM_B3BCR  		(IDENT_ADDR + 0x180000A60UL)
-#define APECS_MEM_B4BCR  		(IDENT_ADDR + 0x180000A80UL)
-#define APECS_MEM_B5BCR  		(IDENT_ADDR + 0x180000AA0UL)
-#define APECS_MEM_B6BCR  		(IDENT_ADDR + 0x180000AC0UL)
-#define APECS_MEM_B7BCR  		(IDENT_ADDR + 0x180000AE0UL)
-#define APECS_MEM_B8BCR  		(IDENT_ADDR + 0x180000B00UL)
-
-/* Bank x Timing Register A */
-#define APECS_MEM_B0TRA  		(IDENT_ADDR + 0x180000C00UL)
-#define APECS_MEM_B1TRA  		(IDENT_ADDR + 0x180000C20UL)
-#define APECS_MEM_B2TRA  		(IDENT_ADDR + 0x180000C40UL)
-#define APECS_MEM_B3TRA  		(IDENT_ADDR + 0x180000C60UL)
-#define APECS_MEM_B4TRA  		(IDENT_ADDR + 0x180000C80UL)
-#define APECS_MEM_B5TRA  		(IDENT_ADDR + 0x180000CA0UL)
-#define APECS_MEM_B6TRA  		(IDENT_ADDR + 0x180000CC0UL)
-#define APECS_MEM_B7TRA  		(IDENT_ADDR + 0x180000CE0UL)
-#define APECS_MEM_B8TRA  		(IDENT_ADDR + 0x180000D00UL)
-
-/* Bank x Timing Register B */
-#define APECS_MEM_B0TRB                 (IDENT_ADDR + 0x180000E00UL)
-#define APECS_MEM_B1TRB  		(IDENT_ADDR + 0x180000E20UL)
-#define APECS_MEM_B2TRB  		(IDENT_ADDR + 0x180000E40UL)
-#define APECS_MEM_B3TRB  		(IDENT_ADDR + 0x180000E60UL)
-#define APECS_MEM_B4TRB  		(IDENT_ADDR + 0x180000E80UL)
-#define APECS_MEM_B5TRB  		(IDENT_ADDR + 0x180000EA0UL)
-#define APECS_MEM_B6TRB  		(IDENT_ADDR + 0x180000EC0UL)
-#define APECS_MEM_B7TRB  		(IDENT_ADDR + 0x180000EE0UL)
-#define APECS_MEM_B8TRB  		(IDENT_ADDR + 0x180000F00UL)
-
-
-/*
- * Memory spaces:
- */
-#define APECS_IACK_SC		        (IDENT_ADDR + 0x1b0000000UL)
-#define APECS_CONF		        (IDENT_ADDR + 0x1e0000000UL)
-#define APECS_IO			(IDENT_ADDR + 0x1c0000000UL)
-#define APECS_SPARSE_MEM		(IDENT_ADDR + 0x200000000UL)
-#define APECS_DENSE_MEM		        (IDENT_ADDR + 0x300000000UL)
-
-
-/*
- * Bit definitions for I/O Controller status register 0:
- */
-#define APECS_IOC_STAT0_CMD		0xf
-#define APECS_IOC_STAT0_ERR		(1<<4)
-#define APECS_IOC_STAT0_LOST		(1<<5)
-#define APECS_IOC_STAT0_THIT		(1<<6)
-#define APECS_IOC_STAT0_TREF		(1<<7)
-#define APECS_IOC_STAT0_CODE_SHIFT	8
-#define APECS_IOC_STAT0_CODE_MASK	0x7
-#define APECS_IOC_STAT0_P_NBR_SHIFT	13
-#define APECS_IOC_STAT0_P_NBR_MASK	0x7ffff
-
-#define APECS_HAE_ADDRESS		APECS_IOC_HAXR1
-
-
-/*
- * Data structure for handling APECS machine checks:
- */
-
-struct el_apecs_mikasa_sysdata_mcheck
-{
-	unsigned long coma_gcr;
-	unsigned long coma_edsr;
-	unsigned long coma_ter;
-	unsigned long coma_elar;
-	unsigned long coma_ehar;
-	unsigned long coma_ldlr;
-	unsigned long coma_ldhr;
-	unsigned long coma_base0;
-	unsigned long coma_base1;
-	unsigned long coma_base2;
-	unsigned long coma_base3;
-	unsigned long coma_cnfg0;
-	unsigned long coma_cnfg1;
-	unsigned long coma_cnfg2;
-	unsigned long coma_cnfg3;
-	unsigned long epic_dcsr;
-	unsigned long epic_pear;
-	unsigned long epic_sear;
-	unsigned long epic_tbr1;
-	unsigned long epic_tbr2;
-	unsigned long epic_pbr1;
-	unsigned long epic_pbr2;
-	unsigned long epic_pmr1;
-	unsigned long epic_pmr2;
-	unsigned long epic_harx1;
-	unsigned long epic_harx2;
-	unsigned long epic_pmlt;
-	unsigned long epic_tag0;
-	unsigned long epic_tag1;
-	unsigned long epic_tag2;
-	unsigned long epic_tag3;
-	unsigned long epic_tag4;
-	unsigned long epic_tag5;
-	unsigned long epic_tag6;
-	unsigned long epic_tag7;
-	unsigned long epic_data0;
-	unsigned long epic_data1;
-	unsigned long epic_data2;
-	unsigned long epic_data3;
-	unsigned long epic_data4;
-	unsigned long epic_data5;
-	unsigned long epic_data6;
-	unsigned long epic_data7;
-
-	unsigned long pceb_vid;
-	unsigned long pceb_did;
-	unsigned long pceb_revision;
-	unsigned long pceb_command;
-	unsigned long pceb_status;
-	unsigned long pceb_latency;
-	unsigned long pceb_control;
-	unsigned long pceb_arbcon;
-	unsigned long pceb_arbpri;
-
-	unsigned long esc_id;
-	unsigned long esc_revision;
-	unsigned long esc_int0;
-	unsigned long esc_int1;
-	unsigned long esc_elcr0;
-	unsigned long esc_elcr1;
-	unsigned long esc_last_eisa;
-	unsigned long esc_nmi_stat;
-
-	unsigned long pci_ir;
-	unsigned long pci_imr;
-	unsigned long svr_mgr;
-};
-
-/* This for the normal APECS machines.  */
-struct el_apecs_sysdata_mcheck
-{
-	unsigned long coma_gcr;
-	unsigned long coma_edsr;
-	unsigned long coma_ter;
-	unsigned long coma_elar;
-	unsigned long coma_ehar;
-	unsigned long coma_ldlr;
-	unsigned long coma_ldhr;
-	unsigned long coma_base0;
-	unsigned long coma_base1;
-	unsigned long coma_base2;
-	unsigned long coma_cnfg0;
-	unsigned long coma_cnfg1;
-	unsigned long coma_cnfg2;
-	unsigned long epic_dcsr;
-	unsigned long epic_pear;
-	unsigned long epic_sear;
-	unsigned long epic_tbr1;
-	unsigned long epic_tbr2;
-	unsigned long epic_pbr1;
-	unsigned long epic_pbr2;
-	unsigned long epic_pmr1;
-	unsigned long epic_pmr2;
-	unsigned long epic_harx1;
-	unsigned long epic_harx2;
-	unsigned long epic_pmlt;
-	unsigned long epic_tag0;
-	unsigned long epic_tag1;
-	unsigned long epic_tag2;
-	unsigned long epic_tag3;
-	unsigned long epic_tag4;
-	unsigned long epic_tag5;
-	unsigned long epic_tag6;
-	unsigned long epic_tag7;
-	unsigned long epic_data0;
-	unsigned long epic_data1;
-	unsigned long epic_data2;
-	unsigned long epic_data3;
-	unsigned long epic_data4;
-	unsigned long epic_data5;
-	unsigned long epic_data6;
-	unsigned long epic_data7;
-};
-
-struct el_apecs_procdata
-{
-	unsigned long paltemp[32];  /* PAL TEMP REGS. */
-	/* EV4-specific fields */
-	unsigned long exc_addr;     /* Address of excepting instruction. */
-	unsigned long exc_sum;      /* Summary of arithmetic traps. */
-	unsigned long exc_mask;     /* Exception mask (from exc_sum). */
-	unsigned long iccsr;        /* IBox hardware enables. */
-	unsigned long pal_base;     /* Base address for PALcode. */
-	unsigned long hier;         /* Hardware Interrupt Enable. */
-	unsigned long hirr;         /* Hardware Interrupt Request. */
-	unsigned long csr;          /* D-stream fault info. */
-	unsigned long dc_stat;      /* D-cache status (ECC/Parity Err). */
-	unsigned long dc_addr;      /* EV3 Phys Addr for ECC/DPERR. */
-	unsigned long abox_ctl;     /* ABox Control Register. */
-	unsigned long biu_stat;     /* BIU Status. */
-	unsigned long biu_addr;     /* BUI Address. */
-	unsigned long biu_ctl;      /* BIU Control. */
-	unsigned long fill_syndrome;/* For correcting ECC errors. */
-	unsigned long fill_addr;    /* Cache block which was being read */
-	unsigned long va;           /* Effective VA of fault or miss. */
-	unsigned long bc_tag;       /* Backup Cache Tag Probe Results.*/
-};
-
-
-#ifdef __KERNEL__
-
-#ifndef __EXTERN_INLINE
-#define __EXTERN_INLINE extern inline
-#define __IO_EXTERN_INLINE
-#endif
-
-/*
- * I/O functions:
- *
- * Unlike Jensen, the APECS machines have no concept of local
- * I/O---everything goes over the PCI bus.
- *
- * There is plenty room for optimization here.  In particular,
- * the Alpha's insb/insw/extb/extw should be useful in moving
- * data to/from the right byte-lanes.
- */
-
-#define vip	volatile int __force *
-#define vuip	volatile unsigned int __force *
-#define vulp	volatile unsigned long __force *
-
-#define APECS_SET_HAE						\
-	do {							\
-		if (addr >= (1UL << 24)) {			\
-			unsigned long msb = addr & 0xf8000000;	\
-			addr -= msb;				\
-			set_hae(msb);				\
-		}						\
-	} while (0)
-
-__EXTERN_INLINE u8 apecs_ioread8(const void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	unsigned long result, base_and_type;
-
-	if (addr >= APECS_DENSE_MEM) {
-		addr -= APECS_DENSE_MEM;
-		APECS_SET_HAE;
-		base_and_type = APECS_SPARSE_MEM + 0x00;
-	} else {
-		addr -= APECS_IO;
-		base_and_type = APECS_IO + 0x00;
-	}
-
-	result = *(vip) ((addr << 5) + base_and_type);
-	return __kernel_extbl(result, addr & 3);
-}
-
-__EXTERN_INLINE void apecs_iowrite8(u8 b, void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	unsigned long w, base_and_type;
-
-	if (addr >= APECS_DENSE_MEM) {
-		addr -= APECS_DENSE_MEM;
-		APECS_SET_HAE;
-		base_and_type = APECS_SPARSE_MEM + 0x00;
-	} else {
-		addr -= APECS_IO;
-		base_and_type = APECS_IO + 0x00;
-	}
-
-	w = __kernel_insbl(b, addr & 3);
-	*(vuip) ((addr << 5) + base_and_type) = w;
-}
-
-__EXTERN_INLINE u16 apecs_ioread16(const void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	unsigned long result, base_and_type;
-
-	if (addr >= APECS_DENSE_MEM) {
-		addr -= APECS_DENSE_MEM;
-		APECS_SET_HAE;
-		base_and_type = APECS_SPARSE_MEM + 0x08;
-	} else {
-		addr -= APECS_IO;
-		base_and_type = APECS_IO + 0x08;
-	}
-
-	result = *(vip) ((addr << 5) + base_and_type);
-	return __kernel_extwl(result, addr & 3);
-}
-
-__EXTERN_INLINE void apecs_iowrite16(u16 b, void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	unsigned long w, base_and_type;
-
-	if (addr >= APECS_DENSE_MEM) {
-		addr -= APECS_DENSE_MEM;
-		APECS_SET_HAE;
-		base_and_type = APECS_SPARSE_MEM + 0x08;
-	} else {
-		addr -= APECS_IO;
-		base_and_type = APECS_IO + 0x08;
-	}
-
-	w = __kernel_inswl(b, addr & 3);
-	*(vuip) ((addr << 5) + base_and_type) = w;
-}
-
-__EXTERN_INLINE u32 apecs_ioread32(const void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	if (addr < APECS_DENSE_MEM)
-		addr = ((addr - APECS_IO) << 5) + APECS_IO + 0x18;
-	return *(vuip)addr;
-}
-
-__EXTERN_INLINE void apecs_iowrite32(u32 b, void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	if (addr < APECS_DENSE_MEM)
-		addr = ((addr - APECS_IO) << 5) + APECS_IO + 0x18;
-	*(vuip)addr = b;
-}
-
-__EXTERN_INLINE u64 apecs_ioread64(const void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	if (addr < APECS_DENSE_MEM)
-		addr = ((addr - APECS_IO) << 5) + APECS_IO + 0x18;
-	return *(vulp)addr;
-}
-
-__EXTERN_INLINE void apecs_iowrite64(u64 b, void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	if (addr < APECS_DENSE_MEM)
-		addr = ((addr - APECS_IO) << 5) + APECS_IO + 0x18;
-	*(vulp)addr = b;
-}
-
-__EXTERN_INLINE void __iomem *apecs_ioportmap(unsigned long addr)
-{
-	return (void __iomem *)(addr + APECS_IO);
-}
-
-__EXTERN_INLINE void __iomem *apecs_ioremap(unsigned long addr,
-					    unsigned long size)
-{
-	return (void __iomem *)(addr + APECS_DENSE_MEM);
-}
-
-__EXTERN_INLINE int apecs_is_ioaddr(unsigned long addr)
-{
-	return addr >= IDENT_ADDR + 0x180000000UL;
-}
-
-__EXTERN_INLINE int apecs_is_mmio(const volatile void __iomem *addr)
-{
-	return (unsigned long)addr >= APECS_DENSE_MEM;
-}
-
-#undef APECS_SET_HAE
-
-#undef vip
-#undef vuip
-#undef vulp
-
-#undef __IO_PREFIX
-#define __IO_PREFIX		apecs
-#define apecs_trivial_io_bw	0
-#define apecs_trivial_io_lq	0
-#define apecs_trivial_rw_bw	2
-#define apecs_trivial_rw_lq	1
-#define apecs_trivial_iounmap	1
-#include <asm/io_trivial.h>
-
-#ifdef __IO_EXTERN_INLINE
-#undef __EXTERN_INLINE
-#undef __IO_EXTERN_INLINE
-#endif
-
-#endif /* __KERNEL__ */
-
-#endif /* __ALPHA_APECS__H__ */
diff --git a/arch/alpha/include/asm/core_lca.h b/arch/alpha/include/asm/core_lca.h
deleted file mode 100644
index d8c3e72ef8f6..000000000000
--- a/arch/alpha/include/asm/core_lca.h
+++ /dev/null
@@ -1,378 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef __ALPHA_LCA__H__
-#define __ALPHA_LCA__H__
-
-#include <asm/compiler.h>
-#include <asm/mce.h>
-
-/*
- * Low Cost Alpha (LCA) definitions (these apply to 21066 and 21068,
- * for example).
- *
- * This file is based on:
- *
- *	DECchip 21066 and DECchip 21068 Alpha AXP Microprocessors
- *	Hardware Reference Manual; Digital Equipment Corp.; May 1994;
- *	Maynard, MA; Order Number: EC-N2681-71.
- */
-
-/*
- * NOTE: The LCA uses a Host Address Extension (HAE) register to access
- *	 PCI addresses that are beyond the first 27 bits of address
- *	 space.  Updating the HAE requires an external cycle (and
- *	 a memory barrier), which tends to be slow.  Instead of updating
- *	 it on each sparse memory access, we keep the current HAE value
- *	 cached in variable cache_hae.  Only if the cached HAE differs
- *	 from the desired HAE value do we actually updated HAE register.
- *	 The HAE register is preserved by the interrupt handler entry/exit
- *	 code, so this scheme works even in the presence of interrupts.
- *
- * Dense memory space doesn't require the HAE, but is restricted to
- * aligned 32 and 64 bit accesses.  Special Cycle and Interrupt
- * Acknowledge cycles may also require the use of the HAE.  The LCA
- * limits I/O address space to the bottom 24 bits of address space,
- * but this easily covers the 16 bit ISA I/O address space.
- */
-
-/*
- * NOTE 2! The memory operations do not set any memory barriers, as
- * it's not needed for cases like a frame buffer that is essentially
- * memory-like.  You need to do them by hand if the operations depend
- * on ordering.
- *
- * Similarly, the port I/O operations do a "mb" only after a write
- * operation: if an mb is needed before (as in the case of doing
- * memory mapped I/O first, and then a port I/O operation to the same
- * device), it needs to be done by hand.
- *
- * After the above has bitten me 100 times, I'll give up and just do
- * the mb all the time, but right now I'm hoping this will work out.
- * Avoiding mb's may potentially be a noticeable speed improvement,
- * but I can't honestly say I've tested it.
- *
- * Handling interrupts that need to do mb's to synchronize to
- * non-interrupts is another fun race area.  Don't do it (because if
- * you do, I'll have to do *everything* with interrupts disabled,
- * ugh).
- */
-
-/*
- * Memory Controller registers:
- */
-#define LCA_MEM_BCR0		(IDENT_ADDR + 0x120000000UL)
-#define LCA_MEM_BCR1		(IDENT_ADDR + 0x120000008UL)
-#define LCA_MEM_BCR2		(IDENT_ADDR + 0x120000010UL)
-#define LCA_MEM_BCR3		(IDENT_ADDR + 0x120000018UL)
-#define LCA_MEM_BMR0		(IDENT_ADDR + 0x120000020UL)
-#define LCA_MEM_BMR1		(IDENT_ADDR + 0x120000028UL)
-#define LCA_MEM_BMR2		(IDENT_ADDR + 0x120000030UL)
-#define LCA_MEM_BMR3		(IDENT_ADDR + 0x120000038UL)
-#define LCA_MEM_BTR0		(IDENT_ADDR + 0x120000040UL)
-#define LCA_MEM_BTR1		(IDENT_ADDR + 0x120000048UL)
-#define LCA_MEM_BTR2		(IDENT_ADDR + 0x120000050UL)
-#define LCA_MEM_BTR3		(IDENT_ADDR + 0x120000058UL)
-#define LCA_MEM_GTR		(IDENT_ADDR + 0x120000060UL)
-#define LCA_MEM_ESR		(IDENT_ADDR + 0x120000068UL)
-#define LCA_MEM_EAR		(IDENT_ADDR + 0x120000070UL)
-#define LCA_MEM_CAR		(IDENT_ADDR + 0x120000078UL)
-#define LCA_MEM_VGR		(IDENT_ADDR + 0x120000080UL)
-#define LCA_MEM_PLM		(IDENT_ADDR + 0x120000088UL)
-#define LCA_MEM_FOR		(IDENT_ADDR + 0x120000090UL)
-
-/*
- * I/O Controller registers:
- */
-#define LCA_IOC_HAE		(IDENT_ADDR + 0x180000000UL)
-#define LCA_IOC_CONF		(IDENT_ADDR + 0x180000020UL)
-#define LCA_IOC_STAT0		(IDENT_ADDR + 0x180000040UL)
-#define LCA_IOC_STAT1		(IDENT_ADDR + 0x180000060UL)
-#define LCA_IOC_TBIA		(IDENT_ADDR + 0x180000080UL)
-#define LCA_IOC_TB_ENA		(IDENT_ADDR + 0x1800000a0UL)
-#define LCA_IOC_SFT_RST		(IDENT_ADDR + 0x1800000c0UL)
-#define LCA_IOC_PAR_DIS		(IDENT_ADDR + 0x1800000e0UL)
-#define LCA_IOC_W_BASE0		(IDENT_ADDR + 0x180000100UL)
-#define LCA_IOC_W_BASE1		(IDENT_ADDR + 0x180000120UL)
-#define LCA_IOC_W_MASK0		(IDENT_ADDR + 0x180000140UL)
-#define LCA_IOC_W_MASK1		(IDENT_ADDR + 0x180000160UL)
-#define LCA_IOC_T_BASE0		(IDENT_ADDR + 0x180000180UL)
-#define LCA_IOC_T_BASE1		(IDENT_ADDR + 0x1800001a0UL)
-#define LCA_IOC_TB_TAG0		(IDENT_ADDR + 0x188000000UL)
-#define LCA_IOC_TB_TAG1		(IDENT_ADDR + 0x188000020UL)
-#define LCA_IOC_TB_TAG2		(IDENT_ADDR + 0x188000040UL)
-#define LCA_IOC_TB_TAG3		(IDENT_ADDR + 0x188000060UL)
-#define LCA_IOC_TB_TAG4		(IDENT_ADDR + 0x188000070UL)
-#define LCA_IOC_TB_TAG5		(IDENT_ADDR + 0x1880000a0UL)
-#define LCA_IOC_TB_TAG6		(IDENT_ADDR + 0x1880000c0UL)
-#define LCA_IOC_TB_TAG7		(IDENT_ADDR + 0x1880000e0UL)
-
-/*
- * Memory spaces:
- */
-#define LCA_IACK_SC		(IDENT_ADDR + 0x1a0000000UL)
-#define LCA_CONF		(IDENT_ADDR + 0x1e0000000UL)
-#define LCA_IO			(IDENT_ADDR + 0x1c0000000UL)
-#define LCA_SPARSE_MEM		(IDENT_ADDR + 0x200000000UL)
-#define LCA_DENSE_MEM		(IDENT_ADDR + 0x300000000UL)
-
-/*
- * Bit definitions for I/O Controller status register 0:
- */
-#define LCA_IOC_STAT0_CMD		0xf
-#define LCA_IOC_STAT0_ERR		(1<<4)
-#define LCA_IOC_STAT0_LOST		(1<<5)
-#define LCA_IOC_STAT0_THIT		(1<<6)
-#define LCA_IOC_STAT0_TREF		(1<<7)
-#define LCA_IOC_STAT0_CODE_SHIFT	8
-#define LCA_IOC_STAT0_CODE_MASK		0x7
-#define LCA_IOC_STAT0_P_NBR_SHIFT	13
-#define LCA_IOC_STAT0_P_NBR_MASK	0x7ffff
-
-#define LCA_HAE_ADDRESS		LCA_IOC_HAE
-
-/* LCA PMR Power Management register defines */
-#define LCA_PMR_ADDR	(IDENT_ADDR + 0x120000098UL)
-#define LCA_PMR_PDIV    0x7                     /* Primary clock divisor */
-#define LCA_PMR_ODIV    0x38                    /* Override clock divisor */
-#define LCA_PMR_INTO    0x40                    /* Interrupt override */
-#define LCA_PMR_DMAO    0x80                    /* DMA override */
-#define LCA_PMR_OCCEB   0xffff0000L             /* Override cycle counter - even bits */
-#define LCA_PMR_OCCOB   0xffff000000000000L     /* Override cycle counter - even bits */
-#define LCA_PMR_PRIMARY_MASK    0xfffffffffffffff8L
-
-/* LCA PMR Macros */
-
-#define LCA_READ_PMR        (*(volatile unsigned long *)LCA_PMR_ADDR)
-#define LCA_WRITE_PMR(d)    (*((volatile unsigned long *)LCA_PMR_ADDR) = (d))
-
-#define LCA_GET_PRIMARY(r)  ((r) & LCA_PMR_PDIV)
-#define LCA_GET_OVERRIDE(r) (((r) >> 3) & LCA_PMR_PDIV)
-#define LCA_SET_PRIMARY_CLOCK(r, c) ((r) = (((r) & LCA_PMR_PRIMARY_MASK)|(c)))
-
-/* LCA PMR Divisor values */
-#define LCA_PMR_DIV_1   0x0
-#define LCA_PMR_DIV_1_5 0x1
-#define LCA_PMR_DIV_2   0x2
-#define LCA_PMR_DIV_4   0x3
-#define LCA_PMR_DIV_8   0x4
-#define LCA_PMR_DIV_16  0x5
-#define LCA_PMR_DIV_MIN DIV_1
-#define LCA_PMR_DIV_MAX DIV_16
-
-
-/*
- * Data structure for handling LCA machine checks.  Correctable errors
- * result in a short logout frame, uncorrectable ones in a long one.
- */
-struct el_lca_mcheck_short {
-	struct el_common	h;		/* common logout header */
-	unsigned long		esr;		/* error-status register */
-	unsigned long		ear;		/* error-address register */
-	unsigned long		dc_stat;	/* dcache status register */
-	unsigned long		ioc_stat0;	/* I/O controller status register 0 */
-	unsigned long		ioc_stat1;	/* I/O controller status register 1 */
-};
-
-struct el_lca_mcheck_long {
-	struct el_common	h;		/* common logout header */
-	unsigned long		pt[31];		/* PAL temps */
-	unsigned long		exc_addr;	/* exception address */
-	unsigned long		pad1[3];
-	unsigned long		pal_base;	/* PALcode base address */
-	unsigned long		hier;		/* hw interrupt enable */
-	unsigned long		hirr;		/* hw interrupt request */
-	unsigned long		mm_csr;		/* MMU control & status */
-	unsigned long		dc_stat;	/* data cache status */
-	unsigned long		dc_addr;	/* data cache addr register */
-	unsigned long		abox_ctl;	/* address box control register */
-	unsigned long		esr;		/* error status register */
-	unsigned long		ear;		/* error address register */
-	unsigned long		car;		/* cache control register */
-	unsigned long		ioc_stat0;	/* I/O controller status register 0 */
-	unsigned long		ioc_stat1;	/* I/O controller status register 1 */
-	unsigned long		va;		/* virtual address register */
-};
-
-union el_lca {
-	struct el_common *		c;
-	struct el_lca_mcheck_long *	l;
-	struct el_lca_mcheck_short *	s;
-};
-
-#ifdef __KERNEL__
-
-#ifndef __EXTERN_INLINE
-#define __EXTERN_INLINE extern inline
-#define __IO_EXTERN_INLINE
-#endif
-
-/*
- * I/O functions:
- *
- * Unlike Jensen, the Noname machines have no concept of local
- * I/O---everything goes over the PCI bus.
- *
- * There is plenty room for optimization here.  In particular,
- * the Alpha's insb/insw/extb/extw should be useful in moving
- * data to/from the right byte-lanes.
- */
-
-#define vip	volatile int __force *
-#define vuip	volatile unsigned int __force *
-#define vulp	volatile unsigned long __force *
-
-#define LCA_SET_HAE						\
-	do {							\
-		if (addr >= (1UL << 24)) {			\
-			unsigned long msb = addr & 0xf8000000;	\
-			addr -= msb;				\
-			set_hae(msb);				\
-		}						\
-	} while (0)
-
-
-__EXTERN_INLINE u8 lca_ioread8(const void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	unsigned long result, base_and_type;
-
-	if (addr >= LCA_DENSE_MEM) {
-		addr -= LCA_DENSE_MEM;
-		LCA_SET_HAE;
-		base_and_type = LCA_SPARSE_MEM + 0x00;
-	} else {
-		addr -= LCA_IO;
-		base_and_type = LCA_IO + 0x00;
-	}
-
-	result = *(vip) ((addr << 5) + base_and_type);
-	return __kernel_extbl(result, addr & 3);
-}
-
-__EXTERN_INLINE void lca_iowrite8(u8 b, void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	unsigned long w, base_and_type;
-
-	if (addr >= LCA_DENSE_MEM) {
-		addr -= LCA_DENSE_MEM;
-		LCA_SET_HAE;
-		base_and_type = LCA_SPARSE_MEM + 0x00;
-	} else {
-		addr -= LCA_IO;
-		base_and_type = LCA_IO + 0x00;
-	}
-
-	w = __kernel_insbl(b, addr & 3);
-	*(vuip) ((addr << 5) + base_and_type) = w;
-}
-
-__EXTERN_INLINE u16 lca_ioread16(const void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	unsigned long result, base_and_type;
-
-	if (addr >= LCA_DENSE_MEM) {
-		addr -= LCA_DENSE_MEM;
-		LCA_SET_HAE;
-		base_and_type = LCA_SPARSE_MEM + 0x08;
-	} else {
-		addr -= LCA_IO;
-		base_and_type = LCA_IO + 0x08;
-	}
-
-	result = *(vip) ((addr << 5) + base_and_type);
-	return __kernel_extwl(result, addr & 3);
-}
-
-__EXTERN_INLINE void lca_iowrite16(u16 b, void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	unsigned long w, base_and_type;
-
-	if (addr >= LCA_DENSE_MEM) {
-		addr -= LCA_DENSE_MEM;
-		LCA_SET_HAE;
-		base_and_type = LCA_SPARSE_MEM + 0x08;
-	} else {
-		addr -= LCA_IO;
-		base_and_type = LCA_IO + 0x08;
-	}
-
-	w = __kernel_inswl(b, addr & 3);
-	*(vuip) ((addr << 5) + base_and_type) = w;
-}
-
-__EXTERN_INLINE u32 lca_ioread32(const void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	if (addr < LCA_DENSE_MEM)
-		addr = ((addr - LCA_IO) << 5) + LCA_IO + 0x18;
-	return *(vuip)addr;
-}
-
-__EXTERN_INLINE void lca_iowrite32(u32 b, void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	if (addr < LCA_DENSE_MEM)
-		addr = ((addr - LCA_IO) << 5) + LCA_IO + 0x18;
-	*(vuip)addr = b;
-}
-
-__EXTERN_INLINE u64 lca_ioread64(const void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	if (addr < LCA_DENSE_MEM)
-		addr = ((addr - LCA_IO) << 5) + LCA_IO + 0x18;
-	return *(vulp)addr;
-}
-
-__EXTERN_INLINE void lca_iowrite64(u64 b, void __iomem *xaddr)
-{
-	unsigned long addr = (unsigned long) xaddr;
-	if (addr < LCA_DENSE_MEM)
-		addr = ((addr - LCA_IO) << 5) + LCA_IO + 0x18;
-	*(vulp)addr = b;
-}
-
-__EXTERN_INLINE void __iomem *lca_ioportmap(unsigned long addr)
-{
-	return (void __iomem *)(addr + LCA_IO);
-}
-
-__EXTERN_INLINE void __iomem *lca_ioremap(unsigned long addr,
-					  unsigned long size)
-{
-	return (void __iomem *)(addr + LCA_DENSE_MEM);
-}
-
-__EXTERN_INLINE int lca_is_ioaddr(unsigned long addr)
-{
-	return addr >= IDENT_ADDR + 0x120000000UL;
-}
-
-__EXTERN_INLINE int lca_is_mmio(const volatile void __iomem *addr)
-{
-	return (unsigned long)addr >= LCA_DENSE_MEM;
-}
-
-#undef vip
-#undef vuip
-#undef vulp
-
-#undef __IO_PREFIX
-#define __IO_PREFIX		lca
-#define lca_trivial_rw_bw	2
-#define lca_trivial_rw_lq	1
-#define lca_trivial_io_bw	0
-#define lca_trivial_io_lq	0
-#define lca_trivial_iounmap	1
-#include <asm/io_trivial.h>
-
-#ifdef __IO_EXTERN_INLINE
-#undef __EXTERN_INLINE
-#undef __IO_EXTERN_INLINE
-#endif
-
-#endif /* __KERNEL__ */
-
-#endif /* __ALPHA_LCA__H__ */
diff --git a/arch/alpha/include/asm/dma.h b/arch/alpha/include/asm/dma.h
index a04d76b96089..3a88812b7165 100644
--- a/arch/alpha/include/asm/dma.h
+++ b/arch/alpha/include/asm/dma.h
@@ -82,11 +82,6 @@
   just a wiring limit.
 */
 
-/* The maximum address for ISA DMA transfer on Alpha XL, due to an
-   hardware SIO limitation, is 64MB.
-*/
-#define ALPHA_XL_MAX_ISA_DMA_ADDRESS		0x04000000UL
-
 /* The maximum address for ISA DMA transfer on RUFFIAN,
    due to an hardware SIO limitation, is 16MB.
 */
@@ -107,9 +102,7 @@
 #ifdef CONFIG_ALPHA_GENERIC
 # define MAX_ISA_DMA_ADDRESS		(alpha_mv.max_isa_dma_address)
 #else
-# if defined(CONFIG_ALPHA_XL)
-#  define MAX_ISA_DMA_ADDRESS		ALPHA_XL_MAX_ISA_DMA_ADDRESS
-# elif defined(CONFIG_ALPHA_RUFFIAN)
+# if defined(CONFIG_ALPHA_RUFFIAN)
 #  define MAX_ISA_DMA_ADDRESS		ALPHA_RUFFIAN_MAX_ISA_DMA_ADDRESS
 # elif defined(CONFIG_ALPHA_SABLE)
 #  define MAX_ISA_DMA_ADDRESS		ALPHA_SABLE_MAX_ISA_DMA_ADDRESS
diff --git a/arch/alpha/include/asm/io.h b/arch/alpha/include/asm/io.h
index 0c54ddb5b33e..2bb8cbeedf91 100644
--- a/arch/alpha/include/asm/io.h
+++ b/arch/alpha/include/asm/io.h
@@ -203,14 +203,10 @@ static inline int generic_is_mmio(const volatile void __iomem *a)
 
 #else
 
-#if defined(CONFIG_ALPHA_APECS)
-# include <asm/core_apecs.h>
-#elif defined(CONFIG_ALPHA_CIA)
+#if defined(CONFIG_ALPHA_CIA)
 # include <asm/core_cia.h>
 #elif defined(CONFIG_ALPHA_IRONGATE)
 # include <asm/core_irongate.h>
-#elif defined(CONFIG_ALPHA_LCA)
-# include <asm/core_lca.h>
 #elif defined(CONFIG_ALPHA_MARVEL)
 # include <asm/core_marvel.h>
 #elif defined(CONFIG_ALPHA_MCPCIA)
diff --git a/arch/alpha/include/asm/irq.h b/arch/alpha/include/asm/irq.h
index 442097c56934..e7d22357a5a6 100644
--- a/arch/alpha/include/asm/irq.h
+++ b/arch/alpha/include/asm/irq.h
@@ -31,16 +31,12 @@
 #  define NR_IRQS      (32768 + 16)    /* marvel - 32 pids */
 # endif
 
-#elif defined(CONFIG_ALPHA_CABRIOLET) || \
-      defined(CONFIG_ALPHA_EB66P)     || \
-      defined(CONFIG_ALPHA_EB164)     || \
+#elif defined(CONFIG_ALPHA_EB164)     || \
       defined(CONFIG_ALPHA_PC164)     || \
       defined(CONFIG_ALPHA_LX164)
 # define NR_IRQS	35
 
-#elif defined(CONFIG_ALPHA_EB66)      || \
-      defined(CONFIG_ALPHA_EB64P)     || \
-      defined(CONFIG_ALPHA_MIKASA)
+#elif defined(CONFIG_ALPHA_MIKASA)
 # define NR_IRQS	32
 
 #elif defined(CONFIG_ALPHA_ALCOR)     || \
diff --git a/arch/alpha/kernel/Makefile b/arch/alpha/kernel/Makefile
index 706ce8625e35..bd5d62fa4a9a 100644
--- a/arch/alpha/kernel/Makefile
+++ b/arch/alpha/kernel/Makefile
@@ -22,14 +22,14 @@ obj-$(CONFIG_AUDIT)	+= audit.o
 
 ifdef CONFIG_ALPHA_GENERIC
 
-obj-y 	 += core_apecs.o core_cia.o core_irongate.o core_lca.o \
+obj-y 	 += core_cia.o core_irongate.o \
 	    core_mcpcia.o core_polaris.o core_t2.o \
 	    core_tsunami.o
 
-obj-y	 += sys_alcor.o sys_cabriolet.o sys_dp264.o sys_eb64p.o sys_eiger.o \
+obj-y	 += sys_alcor.o sys_cabriolet.o sys_dp264.o sys_eiger.o \
 	    sys_miata.o sys_mikasa.o sys_nautilus.o \
 	    sys_noritake.o sys_rawhide.o sys_ruffian.o sys_rx164.o \
-	    sys_sable.o sys_sio.o sys_sx164.o sys_takara.o
+	    sys_sable.o sys_sx164.o sys_takara.o
 
 ifndef CONFIG_ALPHA_LEGACY_START_ADDRESS
 obj-y	 += core_marvel.o core_titan.o core_wildfire.o
@@ -48,10 +48,8 @@ else
 obj-$(CONFIG_ALPHA_SRM)		+= srmcons.o
 
 # Core logic support
-obj-$(CONFIG_ALPHA_APECS)	+= core_apecs.o
 obj-$(CONFIG_ALPHA_CIA)		+= core_cia.o
 obj-$(CONFIG_ALPHA_IRONGATE)	+= core_irongate.o
-obj-$(CONFIG_ALPHA_LCA)		+= core_lca.o
 obj-$(CONFIG_ALPHA_MARVEL)	+= core_marvel.o gct.o
 obj-$(CONFIG_ALPHA_MCPCIA)	+= core_mcpcia.o
 obj-$(CONFIG_ALPHA_POLARIS)	+= core_polaris.o
@@ -62,12 +60,8 @@ obj-$(CONFIG_ALPHA_WILDFIRE)	+= core_wildfire.o
 
 # Board support
 obj-$(CONFIG_ALPHA_ALCOR)	+= sys_alcor.o irq_i8259.o irq_srm.o
-obj-$(CONFIG_ALPHA_CABRIOLET)	+= sys_cabriolet.o irq_i8259.o irq_srm.o \
-				   pc873xx.o
 obj-$(CONFIG_ALPHA_EB164)	+= sys_cabriolet.o irq_i8259.o irq_srm.o \
 				   pc873xx.o
-obj-$(CONFIG_ALPHA_EB66P)	+= sys_cabriolet.o irq_i8259.o irq_srm.o \
-				   pc873xx.o
 obj-$(CONFIG_ALPHA_LX164)	+= sys_cabriolet.o irq_i8259.o irq_srm.o \
 				   smc37c93x.o
 obj-$(CONFIG_ALPHA_PC164)	+= sys_cabriolet.o irq_i8259.o irq_srm.o \
@@ -75,8 +69,6 @@ obj-$(CONFIG_ALPHA_PC164)	+= sys_cabriolet.o irq_i8259.o irq_srm.o \
 obj-$(CONFIG_ALPHA_DP264)	+= sys_dp264.o irq_i8259.o es1888.o smc37c669.o
 obj-$(CONFIG_ALPHA_SHARK)	+= sys_dp264.o irq_i8259.o es1888.o smc37c669.o
 obj-$(CONFIG_ALPHA_TITAN)	+= sys_titan.o irq_i8259.o smc37c669.o
-obj-$(CONFIG_ALPHA_EB64P)	+= sys_eb64p.o irq_i8259.o
-obj-$(CONFIG_ALPHA_EB66)	+= sys_eb64p.o irq_i8259.o
 obj-$(CONFIG_ALPHA_EIGER)	+= sys_eiger.o irq_i8259.o
 obj-$(CONFIG_ALPHA_MARVEL)	+= sys_marvel.o 
 obj-$(CONFIG_ALPHA_MIATA)	+= sys_miata.o irq_pyxis.o irq_i8259.o \
@@ -88,11 +80,6 @@ obj-$(CONFIG_ALPHA_RAWHIDE)	+= sys_rawhide.o irq_i8259.o
 obj-$(CONFIG_ALPHA_RUFFIAN)	+= sys_ruffian.o irq_pyxis.o irq_i8259.o
 obj-$(CONFIG_ALPHA_RX164)	+= sys_rx164.o irq_i8259.o
 obj-$(CONFIG_ALPHA_SABLE)	+= sys_sable.o
-obj-$(CONFIG_ALPHA_BOOK1)	+= sys_sio.o irq_i8259.o irq_srm.o pc873xx.o
-obj-$(CONFIG_ALPHA_AVANTI)	+= sys_sio.o irq_i8259.o irq_srm.o pc873xx.o
-obj-$(CONFIG_ALPHA_NONAME)	+= sys_sio.o irq_i8259.o irq_srm.o pc873xx.o
-obj-$(CONFIG_ALPHA_P2K)		+= sys_sio.o irq_i8259.o irq_srm.o pc873xx.o
-obj-$(CONFIG_ALPHA_XL)		+= sys_sio.o irq_i8259.o irq_srm.o pc873xx.o
 obj-$(CONFIG_ALPHA_SX164)	+= sys_sx164.o irq_pyxis.o irq_i8259.o \
 				   irq_srm.o smc37c669.o
 obj-$(CONFIG_ALPHA_TAKARA)	+= sys_takara.o irq_i8259.o pc873xx.o
diff --git a/arch/alpha/kernel/core_apecs.c b/arch/alpha/kernel/core_apecs.c
deleted file mode 100644
index 6df765ff2b10..000000000000
--- a/arch/alpha/kernel/core_apecs.c
+++ /dev/null
@@ -1,420 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- *	linux/arch/alpha/kernel/core_apecs.c
- *
- * Rewritten for Apecs from the lca.c from:
- *
- * Written by David Mosberger (davidm@cs.arizona.edu) with some code
- * taken from Dave Rusling's (david.rusling@reo.mts.dec.com) 32-bit
- * bios code.
- *
- * Code common to all APECS core logic chips.
- */
-
-#define __EXTERN_INLINE inline
-#include <asm/io.h>
-#include <asm/core_apecs.h>
-#undef __EXTERN_INLINE
-
-#include <linux/types.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-
-#include <asm/ptrace.h>
-#include <asm/smp.h>
-#include <asm/mce.h>
-
-#include "proto.h"
-#include "pci_impl.h"
-
-/*
- * NOTE: Herein lie back-to-back mb instructions.  They are magic. 
- * One plausible explanation is that the i/o controller does not properly
- * handle the system transaction.  Another involves timing.  Ho hum.
- */
-
-/*
- * BIOS32-style PCI interface:
- */
-
-#define DEBUG_CONFIG 0
-
-#if DEBUG_CONFIG
-# define DBGC(args)	printk args
-#else
-# define DBGC(args)
-#endif
-
-#define vuip	volatile unsigned int  *
-
-/*
- * Given a bus, device, and function number, compute resulting
- * configuration space address and setup the APECS_HAXR2 register
- * accordingly.  It is therefore not safe to have concurrent
- * invocations to configuration space access routines, but there
- * really shouldn't be any need for this.
- *
- * Type 0:
- *
- *  3 3|3 3 2 2|2 2 2 2|2 2 2 2|1 1 1 1|1 1 1 1|1 1 
- *  3 2|1 0 9 8|7 6 5 4|3 2 1 0|9 8 7 6|5 4 3 2|1 0 9 8|7 6 5 4|3 2 1 0
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- * | | | | | | | | | | | | | | | | | | | | | | | |F|F|F|R|R|R|R|R|R|0|0|
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- *
- *	31:11	Device select bit.
- * 	10:8	Function number
- * 	 7:2	Register number
- *
- * Type 1:
- *
- *  3 3|3 3 2 2|2 2 2 2|2 2 2 2|1 1 1 1|1 1 1 1|1 1 
- *  3 2|1 0 9 8|7 6 5 4|3 2 1 0|9 8 7 6|5 4 3 2|1 0 9 8|7 6 5 4|3 2 1 0
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- * | | | | | | | | | | |B|B|B|B|B|B|B|B|D|D|D|D|D|F|F|F|R|R|R|R|R|R|0|1|
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- *
- *	31:24	reserved
- *	23:16	bus number (8 bits = 128 possible buses)
- *	15:11	Device number (5 bits)
- *	10:8	function number
- *	 7:2	register number
- *  
- * Notes:
- *	The function number selects which function of a multi-function device 
- *	(e.g., SCSI and Ethernet).
- * 
- *	The register selects a DWORD (32 bit) register offset.  Hence it
- *	doesn't get shifted by 2 bits as we want to "drop" the bottom two
- *	bits.
- */
-
-static int
-mk_conf_addr(struct pci_bus *pbus, unsigned int device_fn, int where,
-	     unsigned long *pci_addr, unsigned char *type1)
-{
-	unsigned long addr;
-	u8 bus = pbus->number;
-
-	DBGC(("mk_conf_addr(bus=%d ,device_fn=0x%x, where=0x%x,"
-	      " pci_addr=0x%p, type1=0x%p)\n",
-	      bus, device_fn, where, pci_addr, type1));
-
-	if (bus == 0) {
-		int device = device_fn >> 3;
-
-		/* type 0 configuration cycle: */
-
-		if (device > 20) {
-			DBGC(("mk_conf_addr: device (%d) > 20, returning -1\n",
-			      device));
-			return -1;
-		}
-
-		*type1 = 0;
-		addr = (device_fn << 8) | (where);
-	} else {
-		/* type 1 configuration cycle: */
-		*type1 = 1;
-		addr = (bus << 16) | (device_fn << 8) | (where);
-	}
-	*pci_addr = addr;
-	DBGC(("mk_conf_addr: returning pci_addr 0x%lx\n", addr));
-	return 0;
-}
-
-static unsigned int
-conf_read(unsigned long addr, unsigned char type1)
-{
-	unsigned long flags;
-	unsigned int stat0, value;
-	unsigned int haxr2 = 0;
-
-	local_irq_save(flags);	/* avoid getting hit by machine check */
-
-	DBGC(("conf_read(addr=0x%lx, type1=%d)\n", addr, type1));
-
-	/* Reset status register to avoid losing errors.  */
-	stat0 = *(vuip)APECS_IOC_DCSR;
-	*(vuip)APECS_IOC_DCSR = stat0;
-	mb();
-	DBGC(("conf_read: APECS DCSR was 0x%x\n", stat0));
-
-	/* If Type1 access, must set HAE #2. */
-	if (type1) {
-		haxr2 = *(vuip)APECS_IOC_HAXR2;
-		mb();
-		*(vuip)APECS_IOC_HAXR2 = haxr2 | 1;
-		DBGC(("conf_read: TYPE1 access\n"));
-	}
-
-	draina();
-	mcheck_expected(0) = 1;
-	mcheck_taken(0) = 0;
-	mb();
-
-	/* Access configuration space.  */
-
-	/* Some SRMs step on these registers during a machine check.  */
-	asm volatile("ldl %0,%1; mb; mb" : "=r"(value) : "m"(*(vuip)addr)
-		     : "$9", "$10", "$11", "$12", "$13", "$14", "memory");
-
-	if (mcheck_taken(0)) {
-		mcheck_taken(0) = 0;
-		value = 0xffffffffU;
-		mb();
-	}
-	mcheck_expected(0) = 0;
-	mb();
-
-#if 1
-	/*
-	 * david.rusling@reo.mts.dec.com.  This code is needed for the
-	 * EB64+ as it does not generate a machine check (why I don't
-	 * know).  When we build kernels for one particular platform
-	 * then we can make this conditional on the type.
-	 */
-	draina();
-
-	/* Now look for any errors.  */
-	stat0 = *(vuip)APECS_IOC_DCSR;
-	DBGC(("conf_read: APECS DCSR after read 0x%x\n", stat0));
-
-	/* Is any error bit set? */
-	if (stat0 & 0xffe0U) {
-		/* If not NDEV, print status.  */
-		if (!(stat0 & 0x0800)) {
-			printk("apecs.c:conf_read: got stat0=%x\n", stat0);
-		}
-
-		/* Reset error status.  */
-		*(vuip)APECS_IOC_DCSR = stat0;
-		mb();
-		wrmces(0x7);			/* reset machine check */
-		value = 0xffffffff;
-	}
-#endif
-
-	/* If Type1 access, must reset HAE #2 so normal IO space ops work.  */
-	if (type1) {
-		*(vuip)APECS_IOC_HAXR2 = haxr2 & ~1;
-		mb();
-	}
-	local_irq_restore(flags);
-
-	return value;
-}
-
-static void
-conf_write(unsigned long addr, unsigned int value, unsigned char type1)
-{
-	unsigned long flags;
-	unsigned int stat0;
-	unsigned int haxr2 = 0;
-
-	local_irq_save(flags);	/* avoid getting hit by machine check */
-
-	/* Reset status register to avoid losing errors.  */
-	stat0 = *(vuip)APECS_IOC_DCSR;
-	*(vuip)APECS_IOC_DCSR = stat0;
-	mb();
-
-	/* If Type1 access, must set HAE #2. */
-	if (type1) {
-		haxr2 = *(vuip)APECS_IOC_HAXR2;
-		mb();
-		*(vuip)APECS_IOC_HAXR2 = haxr2 | 1;
-	}
-
-	draina();
-	mcheck_expected(0) = 1;
-	mb();
-
-	/* Access configuration space.  */
-	*(vuip)addr = value;
-	mb();
-	mb();  /* magic */
-	mcheck_expected(0) = 0;
-	mb();
-
-#if 1
-	/*
-	 * david.rusling@reo.mts.dec.com.  This code is needed for the
-	 * EB64+ as it does not generate a machine check (why I don't
-	 * know).  When we build kernels for one particular platform
-	 * then we can make this conditional on the type.
-	 */
-	draina();
-
-	/* Now look for any errors.  */
-	stat0 = *(vuip)APECS_IOC_DCSR;
-
-	/* Is any error bit set? */
-	if (stat0 & 0xffe0U) {
-		/* If not NDEV, print status.  */
-		if (!(stat0 & 0x0800)) {
-			printk("apecs.c:conf_write: got stat0=%x\n", stat0);
-		}
-
-		/* Reset error status.  */
-		*(vuip)APECS_IOC_DCSR = stat0;
-		mb();
-		wrmces(0x7);			/* reset machine check */
-	}
-#endif
-
-	/* If Type1 access, must reset HAE #2 so normal IO space ops work.  */
-	if (type1) {
-		*(vuip)APECS_IOC_HAXR2 = haxr2 & ~1;
-		mb();
-	}
-	local_irq_restore(flags);
-}
-
-static int
-apecs_read_config(struct pci_bus *bus, unsigned int devfn, int where,
-		  int size, u32 *value)
-{
-	unsigned long addr, pci_addr;
-	unsigned char type1;
-	long mask;
-	int shift;
-
-	if (mk_conf_addr(bus, devfn, where, &pci_addr, &type1))
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	mask = (size - 1) * 8;
-	shift = (where & 3) * 8;
-	addr = (pci_addr << 5) + mask + APECS_CONF;
-	*value = conf_read(addr, type1) >> (shift);
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int
-apecs_write_config(struct pci_bus *bus, unsigned int devfn, int where,
-		   int size, u32 value)
-{
-	unsigned long addr, pci_addr;
-	unsigned char type1;
-	long mask;
-
-	if (mk_conf_addr(bus, devfn, where, &pci_addr, &type1))
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	mask = (size - 1) * 8;
-	addr = (pci_addr << 5) + mask + APECS_CONF;
-	conf_write(addr, value << ((where & 3) * 8), type1);
-	return PCIBIOS_SUCCESSFUL;
-}
-
-struct pci_ops apecs_pci_ops = 
-{
-	.read =		apecs_read_config,
-	.write =	apecs_write_config,
-};
-\f
-void
-apecs_pci_tbi(struct pci_controller *hose, dma_addr_t start, dma_addr_t end)
-{
-	wmb();
-	*(vip)APECS_IOC_TBIA = 0;
-	mb();
-}
-\f
-void __init
-apecs_init_arch(void)
-{
-	struct pci_controller *hose;
-
-	/*
-	 * Create our single hose.
-	 */
-
-	pci_isa_hose = hose = alloc_pci_controller();
-	hose->io_space = &ioport_resource;
-	hose->mem_space = &iomem_resource;
-	hose->index = 0;
-
-	hose->sparse_mem_base = APECS_SPARSE_MEM - IDENT_ADDR;
-	hose->dense_mem_base = APECS_DENSE_MEM - IDENT_ADDR;
-	hose->sparse_io_base = APECS_IO - IDENT_ADDR;
-	hose->dense_io_base = 0;
-
-	/*
-	 * Set up the PCI to main memory translation windows.
-	 *
-	 * Window 1 is direct access 1GB at 1GB
-	 * Window 2 is scatter-gather 8MB at 8MB (for isa)
-	 */
-	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000,
-				       SMP_CACHE_BYTES);
-	hose->sg_pci = NULL;
-	__direct_map_base = 0x40000000;
-	__direct_map_size = 0x40000000;
-
-	*(vuip)APECS_IOC_PB1R = __direct_map_base | 0x00080000;
-	*(vuip)APECS_IOC_PM1R = (__direct_map_size - 1) & 0xfff00000U;
-	*(vuip)APECS_IOC_TB1R = 0;
-
-	*(vuip)APECS_IOC_PB2R = hose->sg_isa->dma_base | 0x000c0000;
-	*(vuip)APECS_IOC_PM2R = (hose->sg_isa->size - 1) & 0xfff00000;
-	*(vuip)APECS_IOC_TB2R = virt_to_phys(hose->sg_isa->ptes) >> 1;
-
-	apecs_pci_tbi(hose, 0, -1);
-
-	/*
-	 * Finally, clear the HAXR2 register, which gets used
-	 * for PCI Config Space accesses. That is the way
-	 * we want to use it, and we do not want to depend on
-	 * what ARC or SRM might have left behind...
-	 */
-	*(vuip)APECS_IOC_HAXR2 = 0;
-	mb();
-}
-
-void
-apecs_pci_clr_err(void)
-{
-	unsigned int jd;
-
-	jd = *(vuip)APECS_IOC_DCSR;
-	if (jd & 0xffe0L) {
-		*(vuip)APECS_IOC_SEAR;
-		*(vuip)APECS_IOC_DCSR = jd | 0xffe1L;
-		mb();
-		*(vuip)APECS_IOC_DCSR;
-	}
-	*(vuip)APECS_IOC_TBIA = (unsigned int)APECS_IOC_TBIA;
-	mb();
-	*(vuip)APECS_IOC_TBIA;
-}
-
-void
-apecs_machine_check(unsigned long vector, unsigned long la_ptr)
-{
-	struct el_common *mchk_header;
-	struct el_apecs_procdata *mchk_procdata;
-	struct el_apecs_sysdata_mcheck *mchk_sysdata;
-
-	mchk_header = (struct el_common *)la_ptr;
-
-	mchk_procdata = (struct el_apecs_procdata *)
-		(la_ptr + mchk_header->proc_offset
-		 - sizeof(mchk_procdata->paltemp));
-
-	mchk_sysdata = (struct el_apecs_sysdata_mcheck *)
-		(la_ptr + mchk_header->sys_offset);
-
-
-	/* Clear the error before any reporting.  */
-	mb();
-	mb(); /* magic */
-	draina();
-	apecs_pci_clr_err();
-	wrmces(0x7);		/* reset machine check pending flag */
-	mb();
-
-	process_mcheck_info(vector, la_ptr, "APECS",
-			    (mcheck_expected(0)
-			     && (mchk_sysdata->epic_dcsr & 0x0c00UL)));
-}
diff --git a/arch/alpha/kernel/core_lca.c b/arch/alpha/kernel/core_lca.c
deleted file mode 100644
index 4616b11643d5..000000000000
--- a/arch/alpha/kernel/core_lca.c
+++ /dev/null
@@ -1,473 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- *	linux/arch/alpha/kernel/core_lca.c
- *
- * Written by David Mosberger (davidm@cs.arizona.edu) with some code
- * taken from Dave Rusling's (david.rusling@reo.mts.dec.com) 32-bit
- * bios code.
- *
- * Code common to all LCA core logic chips.
- */
-
-#define __EXTERN_INLINE inline
-#include <asm/io.h>
-#include <asm/core_lca.h>
-#undef __EXTERN_INLINE
-
-#include <linux/types.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/tty.h>
-
-#include <asm/ptrace.h>
-#include <asm/irq_regs.h>
-#include <asm/smp.h>
-
-#include "proto.h"
-#include "pci_impl.h"
-
-
-/*
- * BIOS32-style PCI interface:
- */
-
-/*
- * Machine check reasons.  Defined according to PALcode sources
- * (osf.h and platform.h).
- */
-#define MCHK_K_TPERR		0x0080
-#define MCHK_K_TCPERR		0x0082
-#define MCHK_K_HERR		0x0084
-#define MCHK_K_ECC_C		0x0086
-#define MCHK_K_ECC_NC		0x0088
-#define MCHK_K_UNKNOWN		0x008A
-#define MCHK_K_CACKSOFT		0x008C
-#define MCHK_K_BUGCHECK		0x008E
-#define MCHK_K_OS_BUGCHECK	0x0090
-#define MCHK_K_DCPERR		0x0092
-#define MCHK_K_ICPERR		0x0094
-
-
-/*
- * Platform-specific machine-check reasons:
- */
-#define MCHK_K_SIO_SERR		0x204	/* all platforms so far */
-#define MCHK_K_SIO_IOCHK	0x206	/* all platforms so far */
-#define MCHK_K_DCSR		0x208	/* all but Noname */
-
-
-/*
- * Given a bus, device, and function number, compute resulting
- * configuration space address and setup the LCA_IOC_CONF register
- * accordingly.  It is therefore not safe to have concurrent
- * invocations to configuration space access routines, but there
- * really shouldn't be any need for this.
- *
- * Type 0:
- *
- *  3 3|3 3 2 2|2 2 2 2|2 2 2 2|1 1 1 1|1 1 1 1|1 1 
- *  3 2|1 0 9 8|7 6 5 4|3 2 1 0|9 8 7 6|5 4 3 2|1 0 9 8|7 6 5 4|3 2 1 0
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- * | | | | | | | | | | | | | | | | | | | | | | | |F|F|F|R|R|R|R|R|R|0|0|
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- *
- *	31:11	Device select bit.
- * 	10:8	Function number
- * 	 7:2	Register number
- *
- * Type 1:
- *
- *  3 3|3 3 2 2|2 2 2 2|2 2 2 2|1 1 1 1|1 1 1 1|1 1 
- *  3 2|1 0 9 8|7 6 5 4|3 2 1 0|9 8 7 6|5 4 3 2|1 0 9 8|7 6 5 4|3 2 1 0
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- * | | | | | | | | | | |B|B|B|B|B|B|B|B|D|D|D|D|D|F|F|F|R|R|R|R|R|R|0|1|
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- *
- *	31:24	reserved
- *	23:16	bus number (8 bits = 128 possible buses)
- *	15:11	Device number (5 bits)
- *	10:8	function number
- *	 7:2	register number
- *  
- * Notes:
- *	The function number selects which function of a multi-function device 
- *	(e.g., SCSI and Ethernet).
- * 
- *	The register selects a DWORD (32 bit) register offset.  Hence it
- *	doesn't get shifted by 2 bits as we want to "drop" the bottom two
- *	bits.
- */
-
-static int
-mk_conf_addr(struct pci_bus *pbus, unsigned int device_fn, int where,
-	     unsigned long *pci_addr)
-{
-	unsigned long addr;
-	u8 bus = pbus->number;
-
-	if (bus == 0) {
-		int device = device_fn >> 3;
-		int func = device_fn & 0x7;
-
-		/* Type 0 configuration cycle.  */
-
-		if (device > 12) {
-			return -1;
-		}
-
-		*(vulp)LCA_IOC_CONF = 0;
-		addr = (1 << (11 + device)) | (func << 8) | where;
-	} else {
-		/* Type 1 configuration cycle.  */
-		*(vulp)LCA_IOC_CONF = 1;
-		addr = (bus << 16) | (device_fn << 8) | where;
-	}
-	*pci_addr = addr;
-	return 0;
-}
-
-static unsigned int
-conf_read(unsigned long addr)
-{
-	unsigned long flags, code, stat0;
-	unsigned int value;
-
-	local_irq_save(flags);
-
-	/* Reset status register to avoid losing errors.  */
-	stat0 = *(vulp)LCA_IOC_STAT0;
-	*(vulp)LCA_IOC_STAT0 = stat0;
-	mb();
-
-	/* Access configuration space.  */
-	value = *(vuip)addr;
-	draina();
-
-	stat0 = *(vulp)LCA_IOC_STAT0;
-	if (stat0 & LCA_IOC_STAT0_ERR) {
-		code = ((stat0 >> LCA_IOC_STAT0_CODE_SHIFT)
-			& LCA_IOC_STAT0_CODE_MASK);
-		if (code != 1) {
-			printk("lca.c:conf_read: got stat0=%lx\n", stat0);
-		}
-
-		/* Reset error status.  */
-		*(vulp)LCA_IOC_STAT0 = stat0;
-		mb();
-
-		/* Reset machine check.  */
-		wrmces(0x7);
-
-		value = 0xffffffff;
-	}
-	local_irq_restore(flags);
-	return value;
-}
-
-static void
-conf_write(unsigned long addr, unsigned int value)
-{
-	unsigned long flags, code, stat0;
-
-	local_irq_save(flags);	/* avoid getting hit by machine check */
-
-	/* Reset status register to avoid losing errors.  */
-	stat0 = *(vulp)LCA_IOC_STAT0;
-	*(vulp)LCA_IOC_STAT0 = stat0;
-	mb();
-
-	/* Access configuration space.  */
-	*(vuip)addr = value;
-	draina();
-
-	stat0 = *(vulp)LCA_IOC_STAT0;
-	if (stat0 & LCA_IOC_STAT0_ERR) {
-		code = ((stat0 >> LCA_IOC_STAT0_CODE_SHIFT)
-			& LCA_IOC_STAT0_CODE_MASK);
-		if (code != 1) {
-			printk("lca.c:conf_write: got stat0=%lx\n", stat0);
-		}
-
-		/* Reset error status.  */
-		*(vulp)LCA_IOC_STAT0 = stat0;
-		mb();
-
-		/* Reset machine check. */
-		wrmces(0x7);
-	}
-	local_irq_restore(flags);
-}
-
-static int
-lca_read_config(struct pci_bus *bus, unsigned int devfn, int where,
-		int size, u32 *value)
-{
-	unsigned long addr, pci_addr;
-	long mask;
-	int shift;
-
-	if (mk_conf_addr(bus, devfn, where, &pci_addr))
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	shift = (where & 3) * 8;
-	mask = (size - 1) * 8;
-	addr = (pci_addr << 5) + mask + LCA_CONF;
-	*value = conf_read(addr) >> (shift);
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int 
-lca_write_config(struct pci_bus *bus, unsigned int devfn, int where, int size,
-		 u32 value)
-{
-	unsigned long addr, pci_addr;
-	long mask;
-
-	if (mk_conf_addr(bus, devfn, where, &pci_addr))
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	mask = (size - 1) * 8;
-	addr = (pci_addr << 5) + mask + LCA_CONF;
-	conf_write(addr, value << ((where & 3) * 8));
-	return PCIBIOS_SUCCESSFUL;
-}
-
-struct pci_ops lca_pci_ops = 
-{
-	.read =		lca_read_config,
-	.write =	lca_write_config,
-};
-\f
-void
-lca_pci_tbi(struct pci_controller *hose, dma_addr_t start, dma_addr_t end)
-{
-	wmb();
-	*(vulp)LCA_IOC_TBIA = 0;
-	mb();
-}
-\f
-void __init
-lca_init_arch(void)
-{
-	struct pci_controller *hose;
-
-	/*
-	 * Create our single hose.
-	 */
-
-	pci_isa_hose = hose = alloc_pci_controller();
-	hose->io_space = &ioport_resource;
-	hose->mem_space = &iomem_resource;
-	hose->index = 0;
-
-	hose->sparse_mem_base = LCA_SPARSE_MEM - IDENT_ADDR;
-	hose->dense_mem_base = LCA_DENSE_MEM - IDENT_ADDR;
-	hose->sparse_io_base = LCA_IO - IDENT_ADDR;
-	hose->dense_io_base = 0;
-
-	/*
-	 * Set up the PCI to main memory translation windows.
-	 *
-	 * Mimic the SRM settings for the direct-map window.
-	 *   Window 0 is scatter-gather 8MB at 8MB (for isa).
-	 *   Window 1 is direct access 1GB at 1GB.
-	 *
-	 * Note that we do not try to save any of the DMA window CSRs
-	 * before setting them, since we cannot read those CSRs on LCA.
-	 */
-	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000,
-				       SMP_CACHE_BYTES);
-	hose->sg_pci = NULL;
-	__direct_map_base = 0x40000000;
-	__direct_map_size = 0x40000000;
-
-	*(vulp)LCA_IOC_W_BASE0 = hose->sg_isa->dma_base | (3UL << 32);
-	*(vulp)LCA_IOC_W_MASK0 = (hose->sg_isa->size - 1) & 0xfff00000;
-	*(vulp)LCA_IOC_T_BASE0 = virt_to_phys(hose->sg_isa->ptes);
-
-	*(vulp)LCA_IOC_W_BASE1 = __direct_map_base | (2UL << 32);
-	*(vulp)LCA_IOC_W_MASK1 = (__direct_map_size - 1) & 0xfff00000;
-	*(vulp)LCA_IOC_T_BASE1 = 0;
-
-	*(vulp)LCA_IOC_TB_ENA = 0x80;
-
-	lca_pci_tbi(hose, 0, -1);
-
-	/*
-	 * Disable PCI parity for now.  The NCR53c810 chip has
-	 * troubles meeting the PCI spec which results in
-	 * data parity errors.
-	 */
-	*(vulp)LCA_IOC_PAR_DIS = 1UL<<5;
-
-	/*
-	 * Finally, set up for restoring the correct HAE if using SRM.
-	 * Again, since we cannot read many of the CSRs on the LCA,
-	 * one of which happens to be the HAE, we save the value that
-	 * the SRM will expect...
-	 */
-	if (alpha_using_srm)
-		srm_hae = 0x80000000UL;
-}
-
-/*
- * Constants used during machine-check handling.  I suppose these
- * could be moved into lca.h but I don't see much reason why anybody
- * else would want to use them.
- */
-
-#define ESR_EAV		(1UL<< 0)	/* error address valid */
-#define ESR_CEE		(1UL<< 1)	/* correctable error */
-#define ESR_UEE		(1UL<< 2)	/* uncorrectable error */
-#define ESR_WRE		(1UL<< 3)	/* write-error */
-#define ESR_SOR		(1UL<< 4)	/* error source */
-#define ESR_CTE		(1UL<< 7)	/* cache-tag error */
-#define ESR_MSE		(1UL<< 9)	/* multiple soft errors */
-#define ESR_MHE		(1UL<<10)	/* multiple hard errors */
-#define ESR_NXM		(1UL<<12)	/* non-existent memory */
-
-#define IOC_ERR		(  1<<4)	/* ioc logs an error */
-#define IOC_CMD_SHIFT	0
-#define IOC_CMD		(0xf<<IOC_CMD_SHIFT)
-#define IOC_CODE_SHIFT	8
-#define IOC_CODE	(0xf<<IOC_CODE_SHIFT)
-#define IOC_LOST	(  1<<5)
-#define IOC_P_NBR	((__u32) ~((1<<13) - 1))
-
-static void
-mem_error(unsigned long esr, unsigned long ear)
-{
-	printk("    %s %s error to %s occurred at address %x\n",
-	       ((esr & ESR_CEE) ? "Correctable" :
-		(esr & ESR_UEE) ? "Uncorrectable" : "A"),
-	       (esr & ESR_WRE) ? "write" : "read",
-	       (esr & ESR_SOR) ? "memory" : "b-cache",
-	       (unsigned) (ear & 0x1ffffff8));
-	if (esr & ESR_CTE) {
-		printk("    A b-cache tag parity error was detected.\n");
-	}
-	if (esr & ESR_MSE) {
-		printk("    Several other correctable errors occurred.\n");
-	}
-	if (esr & ESR_MHE) {
-		printk("    Several other uncorrectable errors occurred.\n");
-	}
-	if (esr & ESR_NXM) {
-		printk("    Attempted to access non-existent memory.\n");
-	}
-}
-
-static void
-ioc_error(__u32 stat0, __u32 stat1)
-{
-	static const char * const pci_cmd[] = {
-		"Interrupt Acknowledge", "Special", "I/O Read", "I/O Write",
-		"Rsvd 1", "Rsvd 2", "Memory Read", "Memory Write", "Rsvd3",
-		"Rsvd4", "Configuration Read", "Configuration Write",
-		"Memory Read Multiple", "Dual Address", "Memory Read Line",
-		"Memory Write and Invalidate"
-	};
-	static const char * const err_name[] = {
-		"exceeded retry limit", "no device", "bad data parity",
-		"target abort", "bad address parity", "page table read error",
-		"invalid page", "data error"
-	};
-	unsigned code = (stat0 & IOC_CODE) >> IOC_CODE_SHIFT;
-	unsigned cmd  = (stat0 & IOC_CMD)  >> IOC_CMD_SHIFT;
-
-	printk("    %s initiated PCI %s cycle to address %x"
-	       " failed due to %s.\n",
-	       code > 3 ? "PCI" : "CPU", pci_cmd[cmd], stat1, err_name[code]);
-
-	if (code == 5 || code == 6) {
-		printk("    (Error occurred at PCI memory address %x.)\n",
-		       (stat0 & ~IOC_P_NBR));
-	}
-	if (stat0 & IOC_LOST) {
-		printk("    Other PCI errors occurred simultaneously.\n");
-	}
-}
-
-void
-lca_machine_check(unsigned long vector, unsigned long la_ptr)
-{
-	const char * reason;
-	union el_lca el;
-
-	el.c = (struct el_common *) la_ptr;
-
-	wrmces(rdmces());	/* reset machine check pending flag */
-
-	printk(KERN_CRIT "LCA machine check: vector=%#lx pc=%#lx code=%#x\n",
-	       vector, get_irq_regs()->pc, (unsigned int) el.c->code);
-
-	/*
-	 * The first quadword after the common header always seems to
-	 * be the machine check reason---don't know why this isn't
-	 * part of the common header instead.  In the case of a long
-	 * logout frame, the upper 32 bits is the machine check
-	 * revision level, which we ignore for now.
-	 */
-	switch ((unsigned int) el.c->code) {
-	case MCHK_K_TPERR:	reason = "tag parity error"; break;
-	case MCHK_K_TCPERR:	reason = "tag control parity error"; break;
-	case MCHK_K_HERR:	reason = "access to non-existent memory"; break;
-	case MCHK_K_ECC_C:	reason = "correctable ECC error"; break;
-	case MCHK_K_ECC_NC:	reason = "non-correctable ECC error"; break;
-	case MCHK_K_CACKSOFT:	reason = "MCHK_K_CACKSOFT"; break;
-	case MCHK_K_BUGCHECK:	reason = "illegal exception in PAL mode"; break;
-	case MCHK_K_OS_BUGCHECK: reason = "callsys in kernel mode"; break;
-	case MCHK_K_DCPERR:	reason = "d-cache parity error"; break;
-	case MCHK_K_ICPERR:	reason = "i-cache parity error"; break;
-	case MCHK_K_SIO_SERR:	reason = "SIO SERR occurred on PCI bus"; break;
-	case MCHK_K_SIO_IOCHK:	reason = "SIO IOCHK occurred on ISA bus"; break;
-	case MCHK_K_DCSR:	reason = "MCHK_K_DCSR"; break;
-	case MCHK_K_UNKNOWN:
-	default:		reason = "unknown"; break;
-	}
-
-	switch (el.c->size) {
-	case sizeof(struct el_lca_mcheck_short):
-		printk(KERN_CRIT
-		       "  Reason: %s (short frame%s, dc_stat=%#lx):\n",
-		       reason, el.c->retry ? ", retryable" : "",
-		       el.s->dc_stat);
-		if (el.s->esr & ESR_EAV) {
-			mem_error(el.s->esr, el.s->ear);
-		}
-		if (el.s->ioc_stat0 & IOC_ERR) {
-			ioc_error(el.s->ioc_stat0, el.s->ioc_stat1);
-		}
-		break;
-
-	case sizeof(struct el_lca_mcheck_long):
-		printk(KERN_CRIT "  Reason: %s (long frame%s):\n",
-		       reason, el.c->retry ? ", retryable" : "");
-		printk(KERN_CRIT
-		       "    reason: %#lx  exc_addr: %#lx  dc_stat: %#lx\n", 
-		       el.l->pt[0], el.l->exc_addr, el.l->dc_stat);
-		printk(KERN_CRIT "    car: %#lx\n", el.l->car);
-		if (el.l->esr & ESR_EAV) {
-			mem_error(el.l->esr, el.l->ear);
-		}
-		if (el.l->ioc_stat0 & IOC_ERR) {
-			ioc_error(el.l->ioc_stat0, el.l->ioc_stat1);
-		}
-		break;
-
-	default:
-		printk(KERN_CRIT "  Unknown errorlog size %d\n", el.c->size);
-	}
-
-	/* Dump the logout area to give all info.  */
-#ifdef CONFIG_VERBOSE_MCHECK
-	if (alpha_verbose_mcheck > 1) {
-		unsigned long * ptr = (unsigned long *) la_ptr;
-		long i;
-		for (i = 0; i < el.c->size / sizeof(long); i += 2) {
-			printk(KERN_CRIT " +%8lx %016lx %016lx\n",
-			       i*sizeof(long), ptr[i], ptr[i+1]);
-		}
-	}
-#endif /* CONFIG_VERBOSE_MCHECK */
-}
diff --git a/arch/alpha/kernel/irq_i8259.c b/arch/alpha/kernel/irq_i8259.c
index 1dcf0d9038fd..29c6c477ac35 100644
--- a/arch/alpha/kernel/irq_i8259.c
+++ b/arch/alpha/kernel/irq_i8259.c
@@ -98,10 +98,6 @@ init_i8259a_irqs(void)
 
 #if defined(CONFIG_ALPHA_GENERIC)
 # define IACK_SC	alpha_mv.iack_sc
-#elif defined(CONFIG_ALPHA_APECS)
-# define IACK_SC	APECS_IACK_SC
-#elif defined(CONFIG_ALPHA_LCA)
-# define IACK_SC	LCA_IACK_SC
 #elif defined(CONFIG_ALPHA_CIA)
 # define IACK_SC	CIA_IACK_SC
 #elif defined(CONFIG_ALPHA_PYXIS)
diff --git a/arch/alpha/kernel/pci_impl.h b/arch/alpha/kernel/pci_impl.h
index 18043af45e2b..a16325ce21c4 100644
--- a/arch/alpha/kernel/pci_impl.h
+++ b/arch/alpha/kernel/pci_impl.h
@@ -143,9 +143,7 @@ struct pci_iommu_arena
 	unsigned int align_entry;
 };
 
-#if defined(CONFIG_ALPHA_SRM) && \
-    (defined(CONFIG_ALPHA_CIA) || defined(CONFIG_ALPHA_LCA) || \
-     defined(CONFIG_ALPHA_AVANTI))
+#if defined(CONFIG_ALPHA_SRM) && defined(CONFIG_ALPHA_CIA)
 # define NEED_SRM_SAVE_RESTORE
 #else
 # undef NEED_SRM_SAVE_RESTORE
diff --git a/arch/alpha/kernel/proto.h b/arch/alpha/kernel/proto.h
index 5b1d0d71d479..a8bc3ead776b 100644
--- a/arch/alpha/kernel/proto.h
+++ b/arch/alpha/kernel/proto.h
@@ -17,13 +17,6 @@ struct pci_dev;
 struct pci_controller;
 struct pci_bus;
 
-/* core_apecs.c */
-extern struct pci_ops apecs_pci_ops;
-extern void apecs_init_arch(void);
-extern void apecs_pci_clr_err(void);
-extern void apecs_machine_check(unsigned long vector, unsigned long la_ptr);
-extern void apecs_pci_tbi(struct pci_controller *, dma_addr_t, dma_addr_t);
-
 /* core_cia.c */
 extern struct pci_ops cia_pci_ops;
 extern void cia_init_pci(void);
@@ -39,12 +32,6 @@ extern int irongate_pci_clr_err(void);
 extern void irongate_init_arch(void);
 #define irongate_pci_tbi ((void *)0)
 
-/* core_lca.c */
-extern struct pci_ops lca_pci_ops;
-extern void lca_init_arch(void);
-extern void lca_machine_check(unsigned long vector, unsigned long la_ptr);
-extern void lca_pci_tbi(struct pci_controller *, dma_addr_t, dma_addr_t);
-
 /* core_marvel.c */
 extern struct pci_ops marvel_pci_ops;
 extern void marvel_init_arch(void);
diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index 3b833bd97817..bebdffafaee8 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -171,27 +171,17 @@ EXPORT_SYMBOL(__direct_map_size);
 	asm(".weak "#X)
 
 WEAK(alcor_mv);
-WEAK(alphabook1_mv);
-WEAK(avanti_mv);
-WEAK(cabriolet_mv);
 WEAK(clipper_mv);
 WEAK(dp264_mv);
 WEAK(eb164_mv);
-WEAK(eb64p_mv);
-WEAK(eb66_mv);
-WEAK(eb66p_mv);
 WEAK(eiger_mv);
 WEAK(lx164_mv);
 WEAK(marvel_ev7_mv);
 WEAK(miata_mv);
-WEAK(mikasa_mv);
 WEAK(mikasa_primo_mv);
 WEAK(monet_mv);
 WEAK(nautilus_mv);
-WEAK(noname_mv);
-WEAK(noritake_mv);
 WEAK(noritake_primo_mv);
-WEAK(p2k_mv);
 WEAK(pc164_mv);
 WEAK(privateer_mv);
 WEAK(rawhide_mv);
@@ -204,7 +194,6 @@ WEAK(takara_mv);
 WEAK(titan_mv);
 WEAK(webbrick_mv);
 WEAK(wildfire_mv);
-WEAK(xl_mv);
 WEAK(xlt_mv);
 
 #undef WEAK
@@ -692,12 +681,6 @@ static int eb164_indices[] = {0,0,0,1,1,1,1,1,2,2,2,2,3,3,3,3,4};
 static char alcor_names[][16] = {"Alcor", "Maverick", "Bret"};
 static int alcor_indices[] = {0,0,0,1,1,1,0,0,0,0,0,0,2,2,2,2,2,2};
 
-static char eb64p_names[][16] = {"EB64+", "Cabriolet", "AlphaPCI64"};
-static int eb64p_indices[] = {0,0,1,2};
-
-static char eb66_names[][8] = {"EB66", "EB66+"};
-static int eb66_indices[] = {0,0,1};
-
 static char marvel_names[][16] = {
 	"Marvel/EV7"
 };
@@ -736,21 +719,21 @@ get_sysvec(unsigned long type, unsigned long variation, unsigned long cpu)
 		NULL,		/* Morgan */
 		NULL,		/* Sable -- see below.  */
 		NULL,		/* Medulla */
-		&noname_mv,
+		NULL,		/* Noname */
 		NULL,		/* Turbolaser */
-		&avanti_mv,
+		NULL,		/* Avanti */
 		NULL,		/* Mustang */
 		NULL,		/* Alcor, Bret, Maverick. HWRPB inaccurate? */
 		NULL,		/* Tradewind */
 		NULL,		/* Mikasa -- see below.  */
 		NULL,		/* EB64 */
-		NULL,		/* EB66 -- see variation.  */
-		NULL,		/* EB64+ -- see variation.  */
-		&alphabook1_mv,
+		NULL,		/* EB66 */
+		NULL,		/* EB64+ */
+		NULL,		/* Alphabook1 */
 		&rawhide_mv,
 		NULL,		/* K2 */
 		NULL,		/* Lynx */
-		&xl_mv,
+		NULL,		/* XL */
 		NULL,		/* EB164 -- see variation.  */
 		NULL,		/* Noritake -- see below.  */
 		NULL,		/* Cortex */
@@ -789,19 +772,6 @@ get_sysvec(unsigned long type, unsigned long variation, unsigned long cpu)
 		&eb164_mv, &pc164_mv, &lx164_mv, &sx164_mv, &rx164_mv
 	};
 
-	static struct alpha_machine_vector *eb64p_vecs[] __initdata =
-	{
-		&eb64p_mv,
-		&cabriolet_mv,
-		&cabriolet_mv		/* AlphaPCI64 */
-	};
-
-	static struct alpha_machine_vector *eb66_vecs[] __initdata =
-	{
-		&eb66_mv,
-		&eb66p_mv
-	};
-
 	static struct alpha_machine_vector *marvel_vecs[] __initdata =
 	{
 		&marvel_ev7_mv,
@@ -869,14 +839,6 @@ get_sysvec(unsigned long type, unsigned long variation, unsigned long cpu)
 			if (vec == &eb164_mv && cpu == EV56_CPU)
 				vec = &pc164_mv;
 			break;
-		case ST_DEC_EB64P:
-			if (member < ARRAY_SIZE(eb64p_indices))
-				vec = eb64p_vecs[eb64p_indices[member]];
-			break;
-		case ST_DEC_EB66:
-			if (member < ARRAY_SIZE(eb66_indices))
-				vec = eb66_vecs[eb66_indices[member]];
-			break;
 		case ST_DEC_MARVEL:
 			if (member < ARRAY_SIZE(marvel_indices))
 				vec = marvel_vecs[marvel_indices[member]];
@@ -891,16 +853,10 @@ get_sysvec(unsigned long type, unsigned long variation, unsigned long cpu)
 				vec = tsunami_vecs[tsunami_indices[member]];
 			break;
 		case ST_DEC_1000:
-			if (cpu == EV5_CPU || cpu == EV56_CPU)
-				vec = &mikasa_primo_mv;
-			else
-				vec = &mikasa_mv;
+			vec = &mikasa_primo_mv;
 			break;
 		case ST_DEC_NORITAKE:
-			if (cpu == EV5_CPU || cpu == EV56_CPU)
-				vec = &noritake_primo_mv;
-			else
-				vec = &noritake_mv;
+			vec = &noritake_primo_mv;
 			break;
 		case ST_DEC_2100_A500:
 			vec = &sable_gamma_mv;
@@ -916,26 +872,16 @@ get_sysvec_byname(const char *name)
 	static struct alpha_machine_vector *all_vecs[] __initdata =
 	{
 		&alcor_mv,
-		&alphabook1_mv,
-		&avanti_mv,
-		&cabriolet_mv,
 		&clipper_mv,
 		&dp264_mv,
 		&eb164_mv,
-		&eb64p_mv,
-		&eb66_mv,
-		&eb66p_mv,
 		&eiger_mv,
 		&lx164_mv,
 		&miata_mv,
-		&mikasa_mv,
 		&mikasa_primo_mv,
 		&monet_mv,
 		&nautilus_mv,
-		&noname_mv,
-		&noritake_mv,
 		&noritake_primo_mv,
-		&p2k_mv,
 		&pc164_mv,
 		&privateer_mv,
 		&rawhide_mv,
@@ -947,7 +893,6 @@ get_sysvec_byname(const char *name)
 		&takara_mv,
 		&webbrick_mv,
 		&wildfire_mv,
-		&xl_mv,
 		&xlt_mv
 	};
 
@@ -1009,14 +954,6 @@ get_sysnames(unsigned long type, unsigned long variation, unsigned long cpu,
 		if (member < ARRAY_SIZE(alcor_indices))
 			*variation_name = alcor_names[alcor_indices[member]];
 		break;
-	case ST_DEC_EB64P:
-		if (member < ARRAY_SIZE(eb64p_indices))
-			*variation_name = eb64p_names[eb64p_indices[member]];
-		break;
-	case ST_DEC_EB66:
-		if (member < ARRAY_SIZE(eb66_indices))
-			*variation_name = eb66_names[eb66_indices[member]];
-		break;
 	case ST_DEC_MARVEL:
 		if (member < ARRAY_SIZE(marvel_indices))
 			*variation_name = marvel_names[marvel_indices[member]];
diff --git a/arch/alpha/kernel/sys_cabriolet.c b/arch/alpha/kernel/sys_cabriolet.c
index 47459b73cdb7..06a950072f82 100644
--- a/arch/alpha/kernel/sys_cabriolet.c
+++ b/arch/alpha/kernel/sys_cabriolet.c
@@ -23,9 +23,7 @@
 #include <asm/irq.h>
 #include <asm/mmu_context.h>
 #include <asm/io.h>
-#include <asm/core_apecs.h>
 #include <asm/core_cia.h>
-#include <asm/core_lca.h>
 #include <asm/tlbflush.h>
 
 #include "proto.h"
@@ -232,13 +230,6 @@ cabriolet_enable_ide(void)
 	}
 }
 
-static inline void __init
-cabriolet_init_pci(void)
-{
-	common_init_pci();
-	cabriolet_enable_ide();
-}
-
 static inline void __init
 cia_cab_init_pci(void)
 {
@@ -317,32 +308,6 @@ alphapc164_init_pci(void)
  * The System Vector
  */
 
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_CABRIOLET)
-struct alpha_machine_vector cabriolet_mv __initmv = {
-	.vector_name		= "Cabriolet",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_APECS_IO,
-	.machine_check		= apecs_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= APECS_AND_LCA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 35,
-	.device_interrupt	= cabriolet_device_interrupt,
-
-	.init_arch		= apecs_init_arch,
-	.init_irq		= cabriolet_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= cabriolet_init_pci,
-	.pci_map_irq		= cabriolet_map_irq,
-	.pci_swizzle		= common_swizzle,
-};
-#ifndef CONFIG_ALPHA_EB64P
-ALIAS_MV(cabriolet)
-#endif
-#endif
-
 #if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_EB164)
 struct alpha_machine_vector eb164_mv __initmv = {
 	.vector_name		= "EB164",
@@ -368,30 +333,6 @@ struct alpha_machine_vector eb164_mv __initmv = {
 ALIAS_MV(eb164)
 #endif
 
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_EB66P)
-struct alpha_machine_vector eb66p_mv __initmv = {
-	.vector_name		= "EB66+",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_LCA_IO,
-	.machine_check		= lca_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= APECS_AND_LCA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 35,
-	.device_interrupt	= cabriolet_device_interrupt,
-
-	.init_arch		= lca_init_arch,
-	.init_irq		= cabriolet_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= cabriolet_init_pci,
-	.pci_map_irq		= eb66p_map_irq,
-	.pci_swizzle		= common_swizzle,
-};
-ALIAS_MV(eb66p)
-#endif
-
 #if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_LX164)
 struct alpha_machine_vector lx164_mv __initmv = {
 	.vector_name		= "LX164",
diff --git a/arch/alpha/kernel/sys_eb64p.c b/arch/alpha/kernel/sys_eb64p.c
deleted file mode 100644
index 3c43fd347526..000000000000
--- a/arch/alpha/kernel/sys_eb64p.c
+++ /dev/null
@@ -1,238 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- *	linux/arch/alpha/kernel/sys_eb64p.c
- *
- *	Copyright (C) 1995 David A Rusling
- *	Copyright (C) 1996 Jay A Estabrook
- *	Copyright (C) 1998, 1999 Richard Henderson
- *
- * Code supporting the EB64+ and EB66.
- */
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/sched.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/bitops.h>
-
-#include <asm/ptrace.h>
-#include <asm/dma.h>
-#include <asm/irq.h>
-#include <asm/mmu_context.h>
-#include <asm/io.h>
-#include <asm/core_apecs.h>
-#include <asm/core_lca.h>
-#include <asm/hwrpb.h>
-#include <asm/tlbflush.h>
-
-#include "proto.h"
-#include "irq_impl.h"
-#include "pci_impl.h"
-#include "machvec_impl.h"
-
-
-/* Note mask bit is true for DISABLED irqs.  */
-static unsigned int cached_irq_mask = -1;
-
-static inline void
-eb64p_update_irq_hw(unsigned int irq, unsigned long mask)
-{
-	outb(mask >> (irq >= 24 ? 24 : 16), (irq >= 24 ? 0x27 : 0x26));
-}
-
-static inline void
-eb64p_enable_irq(struct irq_data *d)
-{
-	eb64p_update_irq_hw(d->irq, cached_irq_mask &= ~(1 << d->irq));
-}
-
-static void
-eb64p_disable_irq(struct irq_data *d)
-{
-	eb64p_update_irq_hw(d->irq, cached_irq_mask |= 1 << d->irq);
-}
-
-static struct irq_chip eb64p_irq_type = {
-	.name		= "EB64P",
-	.irq_unmask	= eb64p_enable_irq,
-	.irq_mask	= eb64p_disable_irq,
-	.irq_mask_ack	= eb64p_disable_irq,
-};
-
-static void 
-eb64p_device_interrupt(unsigned long vector)
-{
-	unsigned long pld;
-	unsigned int i;
-
-	/* Read the interrupt summary registers */
-	pld = inb(0x26) | (inb(0x27) << 8);
-
-	/*
-	 * Now, for every possible bit set, work through
-	 * them and call the appropriate interrupt handler.
-	 */
-	while (pld) {
-		i = ffz(~pld);
-		pld &= pld - 1;	/* clear least bit set */
-
-		if (i == 5) {
-			isa_device_interrupt(vector);
-		} else {
-			handle_irq(16 + i);
-		}
-	}
-}
-
-static void __init
-eb64p_init_irq(void)
-{
-	long i;
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_CABRIOLET)
-	/*
-	 * CABRIO SRM may not set variation correctly, so here we test
-	 * the high word of the interrupt summary register for the RAZ
-	 * bits, and hope that a true EB64+ would read all ones...
-	 */
-	if (inw(0x806) != 0xffff) {
-		extern struct alpha_machine_vector cabriolet_mv;
-
-		printk("Detected Cabriolet: correcting HWRPB.\n");
-
-		hwrpb->sys_variation |= 2L << 10;
-		hwrpb_update_checksum(hwrpb);
-
-		alpha_mv = cabriolet_mv;
-		alpha_mv.init_irq();
-		return;
-	}
-#endif /* GENERIC */
-
-	outb(0xff, 0x26);
-	outb(0xff, 0x27);
-
-	init_i8259a_irqs();
-
-	for (i = 16; i < 32; ++i) {
-		irq_set_chip_and_handler(i, &eb64p_irq_type, handle_level_irq);
-		irq_set_status_flags(i, IRQ_LEVEL);
-	}
-
-	common_init_isa_dma();
-	if (request_irq(16 + 5, no_action, 0, "isa-cascade", NULL))
-		pr_err("Failed to register isa-cascade interrupt\n");
-}
-
-/*
- * PCI Fixup configuration.
- *
- * There are two 8 bit external summary registers as follows:
- *
- * Summary @ 0x26:
- * Bit      Meaning
- * 0        Interrupt Line A from slot 0
- * 1        Interrupt Line A from slot 1
- * 2        Interrupt Line B from slot 0
- * 3        Interrupt Line B from slot 1
- * 4        Interrupt Line C from slot 0
- * 5        Interrupt line from the two ISA PICs
- * 6        Tulip
- * 7        NCR SCSI
- *
- * Summary @ 0x27
- * Bit      Meaning
- * 0        Interrupt Line C from slot 1
- * 1        Interrupt Line D from slot 0
- * 2        Interrupt Line D from slot 1
- * 3        RAZ
- * 4        RAZ
- * 5        RAZ
- * 6        RAZ
- * 7        RAZ
- *
- * The device to slot mapping looks like:
- *
- * Slot     Device
- *  5       NCR SCSI controller
- *  6       PCI on board slot 0
- *  7       PCI on board slot 1
- *  8       Intel SIO PCI-ISA bridge chip
- *  9       Tulip - DECchip 21040 Ethernet controller
- *   
- *
- * This two layered interrupt approach means that we allocate IRQ 16 and 
- * above for PCI interrupts.  The IRQ relates to which bit the interrupt
- * comes in on.  This makes interrupt processing much easier.
- */
-
-static int
-eb64p_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
-{
-	static char irq_tab[5][5] = {
-		/*INT  INTA  INTB  INTC   INTD */
-		{16+7, 16+7, 16+7, 16+7,  16+7},  /* IdSel 5,  slot ?, ?? */
-		{16+0, 16+0, 16+2, 16+4,  16+9},  /* IdSel 6,  slot ?, ?? */
-		{16+1, 16+1, 16+3, 16+8, 16+10},  /* IdSel 7,  slot ?, ?? */
-		{  -1,   -1,   -1,   -1,    -1},  /* IdSel 8,  SIO */
-		{16+6, 16+6, 16+6, 16+6,  16+6},  /* IdSel 9,  TULIP */
-	};
-	const long min_idsel = 5, max_idsel = 9, irqs_per_slot = 5;
-	return COMMON_TABLE_LOOKUP;
-}
-
-
-/*
- * The System Vector
- */
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_EB64P)
-struct alpha_machine_vector eb64p_mv __initmv = {
-	.vector_name		= "EB64+",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_APECS_IO,
-	.machine_check		= apecs_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= APECS_AND_LCA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 32,
-	.device_interrupt	= eb64p_device_interrupt,
-
-	.init_arch		= apecs_init_arch,
-	.init_irq		= eb64p_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= common_init_pci,
-	.kill_arch		= NULL,
-	.pci_map_irq		= eb64p_map_irq,
-	.pci_swizzle		= common_swizzle,
-};
-ALIAS_MV(eb64p)
-#endif
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_EB66)
-struct alpha_machine_vector eb66_mv __initmv = {
-	.vector_name		= "EB66",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_LCA_IO,
-	.machine_check		= lca_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= APECS_AND_LCA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 32,
-	.device_interrupt	= eb64p_device_interrupt,
-
-	.init_arch		= lca_init_arch,
-	.init_irq		= eb64p_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= common_init_pci,
-	.pci_map_irq		= eb64p_map_irq,
-	.pci_swizzle		= common_swizzle,
-};
-ALIAS_MV(eb66)
-#endif
diff --git a/arch/alpha/kernel/sys_mikasa.c b/arch/alpha/kernel/sys_mikasa.c
index 7690dfd57cb6..557802398231 100644
--- a/arch/alpha/kernel/sys_mikasa.c
+++ b/arch/alpha/kernel/sys_mikasa.c
@@ -23,7 +23,6 @@
 #include <asm/irq.h>
 #include <asm/mmu_context.h>
 #include <asm/io.h>
-#include <asm/core_apecs.h>
 #include <asm/core_cia.h>
 #include <asm/tlbflush.h>
 
@@ -164,64 +163,9 @@ mikasa_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
 }
 
 
-#if defined(CONFIG_ALPHA_GENERIC) || !defined(CONFIG_ALPHA_PRIMO)
-static void
-mikasa_apecs_machine_check(unsigned long vector, unsigned long la_ptr)
-{
-#define MCHK_NO_DEVSEL 0x205U
-#define MCHK_NO_TABT 0x204U
-
-	struct el_common *mchk_header;
-	unsigned int code;
-
-	mchk_header = (struct el_common *)la_ptr;
-
-	/* Clear the error before any reporting.  */
-	mb();
-	mb(); /* magic */
-	draina();
-	apecs_pci_clr_err();
-	wrmces(0x7);
-	mb();
-
-	code = mchk_header->code;
-	process_mcheck_info(vector, la_ptr, "MIKASA APECS",
-			    (mcheck_expected(0)
-			     && (code == MCHK_NO_DEVSEL
-			         || code == MCHK_NO_TABT)));
-}
-#endif
-
-
 /*
  * The System Vector
  */
-
-#if defined(CONFIG_ALPHA_GENERIC) || !defined(CONFIG_ALPHA_PRIMO)
-struct alpha_machine_vector mikasa_mv __initmv = {
-	.vector_name		= "Mikasa",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_APECS_IO,
-	.machine_check		= mikasa_apecs_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= APECS_AND_LCA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 32,
-	.device_interrupt	= mikasa_device_interrupt,
-
-	.init_arch		= apecs_init_arch,
-	.init_irq		= mikasa_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= common_init_pci,
-	.pci_map_irq		= mikasa_map_irq,
-	.pci_swizzle		= common_swizzle,
-};
-ALIAS_MV(mikasa)
-#endif
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_PRIMO)
 struct alpha_machine_vector mikasa_primo_mv __initmv = {
 	.vector_name		= "Mikasa-Primo",
 	DO_EV5_MMU,
@@ -244,4 +188,3 @@ struct alpha_machine_vector mikasa_primo_mv __initmv = {
 	.pci_swizzle		= common_swizzle,
 };
 ALIAS_MV(mikasa_primo)
-#endif
diff --git a/arch/alpha/kernel/sys_noritake.c b/arch/alpha/kernel/sys_noritake.c
index 47f3ce4f719a..eed3f16561c0 100644
--- a/arch/alpha/kernel/sys_noritake.c
+++ b/arch/alpha/kernel/sys_noritake.c
@@ -24,7 +24,6 @@
 #include <asm/irq.h>
 #include <asm/mmu_context.h>
 #include <asm/io.h>
-#include <asm/core_apecs.h>
 #include <asm/core_cia.h>
 #include <asm/tlbflush.h>
 
@@ -253,64 +252,6 @@ noritake_swizzle(struct pci_dev *dev, u8 *pinp)
 	return slot;
 }
 
-#if defined(CONFIG_ALPHA_GENERIC) || !defined(CONFIG_ALPHA_PRIMO)
-static void
-noritake_apecs_machine_check(unsigned long vector, unsigned long la_ptr)
-{
-#define MCHK_NO_DEVSEL 0x205U
-#define MCHK_NO_TABT 0x204U
-
-        struct el_common *mchk_header;
-        unsigned int code;
-
-        mchk_header = (struct el_common *)la_ptr;
-
-        /* Clear the error before any reporting.  */
-        mb();
-        mb(); /* magic */
-        draina();
-        apecs_pci_clr_err();
-        wrmces(0x7);
-        mb();
-
-        code = mchk_header->code;
-        process_mcheck_info(vector, la_ptr, "NORITAKE APECS",
-                            (mcheck_expected(0)
-                             && (code == MCHK_NO_DEVSEL
-                                 || code == MCHK_NO_TABT)));
-}
-#endif
-
-
-/*
- * The System Vectors
- */
-
-#if defined(CONFIG_ALPHA_GENERIC) || !defined(CONFIG_ALPHA_PRIMO)
-struct alpha_machine_vector noritake_mv __initmv = {
-	.vector_name		= "Noritake",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_APECS_IO,
-	.machine_check		= noritake_apecs_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= EISA_DEFAULT_IO_BASE,
-	.min_mem_address	= APECS_AND_LCA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 48,
-	.device_interrupt	= noritake_device_interrupt,
-
-	.init_arch		= apecs_init_arch,
-	.init_irq		= noritake_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= common_init_pci,
-	.pci_map_irq		= noritake_map_irq,
-	.pci_swizzle		= noritake_swizzle,
-};
-ALIAS_MV(noritake)
-#endif
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_PRIMO)
 struct alpha_machine_vector noritake_primo_mv __initmv = {
 	.vector_name		= "Noritake-Primo",
 	DO_EV5_MMU,
@@ -333,4 +274,3 @@ struct alpha_machine_vector noritake_primo_mv __initmv = {
 	.pci_swizzle		= noritake_swizzle,
 };
 ALIAS_MV(noritake_primo)
-#endif
diff --git a/arch/alpha/kernel/sys_sio.c b/arch/alpha/kernel/sys_sio.c
deleted file mode 100644
index c28a6ae1d989..000000000000
--- a/arch/alpha/kernel/sys_sio.c
+++ /dev/null
@@ -1,486 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- *	linux/arch/alpha/kernel/sys_sio.c
- *
- *	Copyright (C) 1995 David A Rusling
- *	Copyright (C) 1996 Jay A Estabrook
- *	Copyright (C) 1998, 1999 Richard Henderson
- *
- * Code for all boards that route the PCI interrupts through the SIO
- * PCI/ISA bridge.  This includes Noname (AXPpci33), Multia (UDB),
- * Kenetics's Platform 2000, Avanti (AlphaStation), XL, and AlphaBook1.
- */
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/sched.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/screen_info.h>
-
-#include <asm/compiler.h>
-#include <asm/ptrace.h>
-#include <asm/dma.h>
-#include <asm/irq.h>
-#include <asm/mmu_context.h>
-#include <asm/io.h>
-#include <asm/core_apecs.h>
-#include <asm/core_lca.h>
-#include <asm/tlbflush.h>
-
-#include "proto.h"
-#include "irq_impl.h"
-#include "pci_impl.h"
-#include "machvec_impl.h"
-#include "pc873xx.h"
-
-#if defined(ALPHA_RESTORE_SRM_SETUP)
-/* Save LCA configuration data as the console had it set up.  */
-struct 
-{
-	unsigned int orig_route_tab; /* for SAVE/RESTORE */
-} saved_config __attribute((common));
-#endif
-
-
-static void __init
-sio_init_irq(void)
-{
-	if (alpha_using_srm)
-		alpha_mv.device_interrupt = srm_device_interrupt;
-
-	init_i8259a_irqs();
-	common_init_isa_dma();
-}
-
-static inline void __init
-alphabook1_init_arch(void)
-{
-#ifdef CONFIG_VGA_CONSOLE
-	/* The AlphaBook1 has LCD video fixed at 800x600,
-	   37 rows and 100 cols. */
-	vgacon_screen_info.orig_y = 37;
-	vgacon_screen_info.orig_video_cols = 100;
-	vgacon_screen_info.orig_video_lines = 37;
-#endif
-
-	lca_init_arch();
-}
-
-
-/*
- * sio_route_tab selects irq routing in PCI/ISA bridge so that:
- *		PIRQ0 -> irq 15
- *		PIRQ1 -> irq  9
- *		PIRQ2 -> irq 10
- *		PIRQ3 -> irq 11
- *
- * This probably ought to be configurable via MILO.  For
- * example, sound boards seem to like using IRQ 9.
- *
- * This is NOT how we should do it. PIRQ0-X should have
- * their own IRQs, the way intel uses the IO-APIC IRQs.
- */
-
-static void __init
-sio_pci_route(void)
-{
-	unsigned int orig_route_tab;
-
-	/* First, ALWAYS read and print the original setting. */
-	pci_bus_read_config_dword(pci_isa_hose->bus, PCI_DEVFN(7, 0), 0x60,
-				  &orig_route_tab);
-	printk("%s: PIRQ original 0x%x new 0x%x\n", __func__,
-	       orig_route_tab, alpha_mv.sys.sio.route_tab);
-
-#if defined(ALPHA_RESTORE_SRM_SETUP)
-	saved_config.orig_route_tab = orig_route_tab;
-#endif
-
-	/* Now override with desired setting. */
-	pci_bus_write_config_dword(pci_isa_hose->bus, PCI_DEVFN(7, 0), 0x60,
-				   alpha_mv.sys.sio.route_tab);
-}
-
-static bool sio_pci_dev_irq_needs_level(const struct pci_dev *dev)
-{
-	if ((dev->class >> 16 == PCI_BASE_CLASS_BRIDGE) &&
-	    (dev->class >> 8 != PCI_CLASS_BRIDGE_PCMCIA))
-		return false;
-
-	return true;
-}
-
-static unsigned int __init
-sio_collect_irq_levels(void)
-{
-	unsigned int level_bits = 0;
-	struct pci_dev *dev = NULL;
-
-	/* Iterate through the devices, collecting IRQ levels.  */
-	for_each_pci_dev(dev) {
-		if (!sio_pci_dev_irq_needs_level(dev))
-			continue;
-
-		if (dev->irq)
-			level_bits |= (1 << dev->irq);
-	}
-	return level_bits;
-}
-
-static void __sio_fixup_irq_levels(unsigned int level_bits, bool reset)
-{
-	unsigned int old_level_bits;
-
-	/*
-	 * Now, make all PCI interrupts level sensitive.  Notice:
-	 * these registers must be accessed byte-wise.  inw()/outw()
-	 * don't work.
-	 *
-	 * Make sure to turn off any level bits set for IRQs 9,10,11,15,
-	 *  so that the only bits getting set are for devices actually found.
-	 * Note that we do preserve the remainder of the bits, which we hope
-	 *  will be set correctly by ARC/SRM.
-	 *
-	 * Note: we at least preserve any level-set bits on AlphaBook1
-	 */
-	old_level_bits = inb(0x4d0) | (inb(0x4d1) << 8);
-
-	if (reset)
-		old_level_bits &= 0x71ff;
-
-	level_bits |= old_level_bits;
-
-	outb((level_bits >> 0) & 0xff, 0x4d0);
-	outb((level_bits >> 8) & 0xff, 0x4d1);
-}
-
-static inline void
-sio_fixup_irq_levels(unsigned int level_bits)
-{
-	__sio_fixup_irq_levels(level_bits, true);
-}
-
-static inline int
-noname_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
-{
-	/*
-	 * The Noname board has 5 PCI slots with each of the 4
-	 * interrupt pins routed to different pins on the PCI/ISA
-	 * bridge (PIRQ0-PIRQ3).  The table below is based on
-	 * information available at:
-	 *
-	 *   http://ftp.digital.com/pub/DEC/axppci/ref_interrupts.txt
-	 *
-	 * I have no information on the Avanti interrupt routing, but
-	 * the routing seems to be identical to the Noname except
-	 * that the Avanti has an additional slot whose routing I'm
-	 * unsure of.
-	 *
-	 * pirq_tab[0] is a fake entry to deal with old PCI boards
-	 * that have the interrupt pin number hardwired to 0 (meaning
-	 * that they use the default INTA line, if they are interrupt
-	 * driven at all).
-	 */
-	static char irq_tab[][5] = {
-		/*INT A   B   C   D */
-		{ 3,  3,  3,  3,  3}, /* idsel  6 (53c810) */ 
-		{-1, -1, -1, -1, -1}, /* idsel  7 (SIO: PCI/ISA bridge) */
-		{ 2,  2, -1, -1, -1}, /* idsel  8 (Hack: slot closest ISA) */
-		{-1, -1, -1, -1, -1}, /* idsel  9 (unused) */
-		{-1, -1, -1, -1, -1}, /* idsel 10 (unused) */
-		{ 0,  0,  2,  1,  0}, /* idsel 11 KN25_PCI_SLOT0 */
-		{ 1,  1,  0,  2,  1}, /* idsel 12 KN25_PCI_SLOT1 */
-		{ 2,  2,  1,  0,  2}, /* idsel 13 KN25_PCI_SLOT2 */
-		{ 0,  0,  0,  0,  0}, /* idsel 14 AS255 TULIP */
-	};
-	const long min_idsel = 6, max_idsel = 14, irqs_per_slot = 5;
-	int irq = COMMON_TABLE_LOOKUP, tmp;
-	tmp = __kernel_extbl(alpha_mv.sys.sio.route_tab, irq);
-
-	irq = irq >= 0 ? tmp : -1;
-
-	/* Fixup IRQ level if an actual IRQ mapping is detected */
-	if (sio_pci_dev_irq_needs_level(dev) && irq >= 0)
-		__sio_fixup_irq_levels(1 << irq, false);
-
-	return irq;
-}
-
-static inline int
-p2k_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
-{
-	static char irq_tab[][5] = {
-		/*INT A   B   C   D */
-		{ 0,  0, -1, -1, -1}, /* idsel  6 (53c810) */
-		{-1, -1, -1, -1, -1}, /* idsel  7 (SIO: PCI/ISA bridge) */
-		{ 1,  1,  2,  3,  0}, /* idsel  8 (slot A) */
-		{ 2,  2,  3,  0,  1}, /* idsel  9 (slot B) */
-		{-1, -1, -1, -1, -1}, /* idsel 10 (unused) */
-		{-1, -1, -1, -1, -1}, /* idsel 11 (unused) */
-		{ 3,  3, -1, -1, -1}, /* idsel 12 (CMD0646) */
-	};
-	const long min_idsel = 6, max_idsel = 12, irqs_per_slot = 5;
-	int irq = COMMON_TABLE_LOOKUP, tmp;
-	tmp = __kernel_extbl(alpha_mv.sys.sio.route_tab, irq);
-	return irq >= 0 ? tmp : -1;
-}
-
-static inline void __init
-noname_init_pci(void)
-{
-	common_init_pci();
-	sio_pci_route();
-	sio_fixup_irq_levels(sio_collect_irq_levels());
-
-	if (pc873xx_probe() == -1) {
-		printk(KERN_ERR "Probing for PC873xx Super IO chip failed.\n");
-	} else {
-		printk(KERN_INFO "Found %s Super IO chip at 0x%x\n",
-			pc873xx_get_model(), pc873xx_get_base());
-
-		/* Enabling things in the Super IO chip doesn't actually
-		 * configure and enable things, the legacy drivers still
-		 * need to do the actual configuration and enabling.
-		 * This only unblocks them.
-		 */
-
-#if !defined(CONFIG_ALPHA_AVANTI)
-		/* Don't bother on the Avanti family.
-		 * None of them had on-board IDE.
-		 */
-		pc873xx_enable_ide();
-#endif
-		pc873xx_enable_epp19();
-	}
-}
-
-static inline void __init
-alphabook1_init_pci(void)
-{
-	struct pci_dev *dev;
-	unsigned char orig, config;
-
-	common_init_pci();
-	sio_pci_route();
-
-	/*
-	 * On the AlphaBook1, the PCMCIA chip (Cirrus 6729)
-	 * is sensitive to PCI bus bursts, so we must DISABLE
-	 * burst mode for the NCR 8xx SCSI... :-(
-	 *
-	 * Note that the NCR810 SCSI driver must preserve the
-	 * setting of the bit in order for this to work.  At the
-	 * moment (2.0.29), ncr53c8xx.c does NOT do this, but
-	 * 53c7,8xx.c DOES.
-	 */
-
-	dev = NULL;
-	while ((dev = pci_get_device(PCI_VENDOR_ID_NCR, PCI_ANY_ID, dev))) {
-		if (dev->device == PCI_DEVICE_ID_NCR_53C810
-		    || dev->device == PCI_DEVICE_ID_NCR_53C815
-		    || dev->device == PCI_DEVICE_ID_NCR_53C820
-		    || dev->device == PCI_DEVICE_ID_NCR_53C825) {
-			unsigned long io_port;
-			unsigned char ctest4;
-
-			io_port = dev->resource[0].start;
-			ctest4 = inb(io_port+0x21);
-			if (!(ctest4 & 0x80)) {
-				printk("AlphaBook1 NCR init: setting"
-				       " burst disable\n");
-				outb(ctest4 | 0x80, io_port+0x21);
-			}
-                }
-	}
-
-	/* Do not set *ANY* level triggers for AlphaBook1. */
-	sio_fixup_irq_levels(0);
-
-	/* Make sure that register PR1 indicates 1Mb mem */
-	outb(0x0f, 0x3ce); orig = inb(0x3cf);   /* read PR5  */
-	outb(0x0f, 0x3ce); outb(0x05, 0x3cf);   /* unlock PR0-4 */
-	outb(0x0b, 0x3ce); config = inb(0x3cf); /* read PR1 */
-	if ((config & 0xc0) != 0xc0) {
-		printk("AlphaBook1 VGA init: setting 1Mb memory\n");
-		config |= 0xc0;
-		outb(0x0b, 0x3ce); outb(config, 0x3cf); /* write PR1 */
-	}
-	outb(0x0f, 0x3ce); outb(orig, 0x3cf); /* (re)lock PR0-4 */
-}
-
-static void
-sio_kill_arch(int mode)
-{
-#if defined(ALPHA_RESTORE_SRM_SETUP)
-	/* Since we cannot read the PCI DMA Window CSRs, we
-	 * cannot restore them here.
-	 *
-	 * However, we CAN read the PIRQ route register, so restore it
-	 * now...
-	 */
- 	pci_bus_write_config_dword(pci_isa_hose->bus, PCI_DEVFN(7, 0), 0x60,
-				   saved_config.orig_route_tab);
-#endif
-}
-
-
-/*
- * The System Vectors
- */
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_BOOK1)
-struct alpha_machine_vector alphabook1_mv __initmv = {
-	.vector_name		= "AlphaBook1",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_LCA_IO,
-	.machine_check		= lca_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= APECS_AND_LCA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 16,
-	.device_interrupt	= isa_device_interrupt,
-
-	.init_arch		= alphabook1_init_arch,
-	.init_irq		= sio_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= alphabook1_init_pci,
-	.kill_arch		= sio_kill_arch,
-	.pci_map_irq		= noname_map_irq,
-	.pci_swizzle		= common_swizzle,
-
-	.sys = { .sio = {
-		/* NCR810 SCSI is 14, PCMCIA controller is 15.  */
-		.route_tab	= 0x0e0f0a0a,
-	}}
-};
-ALIAS_MV(alphabook1)
-#endif
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_AVANTI_CH)
-struct alpha_machine_vector avanti_mv __initmv = {
-	.vector_name		= "Avanti",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_APECS_IO,
-	.machine_check		= apecs_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= APECS_AND_LCA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 16,
-	.device_interrupt	= isa_device_interrupt,
-
-	.init_arch		= apecs_init_arch,
-	.init_irq		= sio_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= noname_init_pci,
-	.kill_arch		= sio_kill_arch,
-	.pci_map_irq		= noname_map_irq,
-	.pci_swizzle		= common_swizzle,
-
-	.sys = { .sio = {
-		.route_tab	= 0x0b0a050f, /* leave 14 for IDE, 9 for SND */
-	}}
-};
-ALIAS_MV(avanti)
-#endif
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_NONAME_CH)
-struct alpha_machine_vector noname_mv __initmv = {
-	.vector_name		= "Noname",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_LCA_IO,
-	.machine_check		= lca_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= APECS_AND_LCA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 16,
-	.device_interrupt	= srm_device_interrupt,
-
-	.init_arch		= lca_init_arch,
-	.init_irq		= sio_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= noname_init_pci,
-	.kill_arch		= sio_kill_arch,
-	.pci_map_irq		= noname_map_irq,
-	.pci_swizzle		= common_swizzle,
-
-	.sys = { .sio = {
-		/* For UDB, the only available PCI slot must not map to IRQ 9,
-		   since that's the builtin MSS sound chip. That PCI slot
-		   will map to PIRQ1 (for INTA at least), so we give it IRQ 15
-		   instead.
-
-		   Unfortunately we have to do this for NONAME as well, since
-		   they are co-indicated when the platform type "Noname" is
-		   selected... :-(  */
-
-		.route_tab	= 0x0b0a0f0d,
-	}}
-};
-ALIAS_MV(noname)
-#endif
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_P2K)
-struct alpha_machine_vector p2k_mv __initmv = {
-	.vector_name		= "Platform2000",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_LCA_IO,
-	.machine_check		= lca_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= APECS_AND_LCA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 16,
-	.device_interrupt	= srm_device_interrupt,
-
-	.init_arch		= lca_init_arch,
-	.init_irq		= sio_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= noname_init_pci,
-	.kill_arch		= sio_kill_arch,
-	.pci_map_irq		= p2k_map_irq,
-	.pci_swizzle		= common_swizzle,
-
-	.sys = { .sio = {
-		.route_tab	= 0x0b0a090f,
-	}}
-};
-ALIAS_MV(p2k)
-#endif
-
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_XL)
-struct alpha_machine_vector xl_mv __initmv = {
-	.vector_name		= "XL",
-	DO_EV4_MMU,
-	DO_DEFAULT_RTC,
-	DO_APECS_IO,
-	.machine_check		= apecs_machine_check,
-	.max_isa_dma_address	= ALPHA_XL_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= XL_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 16,
-	.device_interrupt	= isa_device_interrupt,
-
-	.init_arch		= apecs_init_arch,
-	.init_irq		= sio_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= noname_init_pci,
-	.kill_arch		= sio_kill_arch,
-	.pci_map_irq		= noname_map_irq,
-	.pci_swizzle		= common_swizzle,
-
-	.sys = { .sio = {
-		.route_tab	= 0x0b0a090f,
-	}}
-};
-ALIAS_MV(xl)
-#endif
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 13/14] alpha: cabriolet: remove EV5 CPU support
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (11 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 12/14] alpha: remove LCA and APECS based machines Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-03  8:11 ` [PATCH 14/14] alpha: drop pre-EV56 support Arnd Bergmann
                   ` (2 subsequent siblings)
  15 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Arnd Bergmann <arnd@arndb.de>

The sys_cabriolet.c file includes support for multiple evaluation
boards. pc164 and lx164 are for ev56 CPUs, while the eb164 is
now the last supported machine that only supports ev5 but not
ev56.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/Kconfig                | 15 ++-------------
 arch/alpha/include/asm/irq.h      |  3 +--
 arch/alpha/kernel/Makefile        |  2 --
 arch/alpha/kernel/sys_cabriolet.c | 28 +---------------------------
 4 files changed, 4 insertions(+), 44 deletions(-)

diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index ee209e895505..7987bb548652 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -91,7 +91,6 @@ choice
 
 	  Alcor/Alpha-XLT     AS 600, AS 500, XL-300, XL-366
 	  DP264               DP264 / DS20 / ES40 / DS10 / DS10L
-	  EB164               EB164 21164 evaluation board
 	  LX164               AlphaPC164-LX
 	  Miata               Personal Workstation 433/500/600 a/au
 	  Marvel              AlphaServer ES47 / ES80 / GS1280
@@ -134,16 +133,6 @@ config ALPHA_DP264
 	  API Networks: 264DP, UP2000(+), CS20;
 	  Compaq: DS10(E,L), XP900, XP1000, DS20(E), ES40.
 
-config ALPHA_EB164
-	bool "EB164"
-	help
-	  EB164 21164 evaluation board from DEC.  Uses 21164 and ALCOR.  Has
-	  ISA and PCI expansion (3 ISA slots, 2 64-bit PCI slots (one is
-	  shared with an ISA slot) and 2 32-bit PCI slots.  Uses plus-in
-	  Bcache SIMMs. I/O sub-system provides SuperI/O (2S, 1P, FD), KBD,
-	  MOUSE (PS2 style), RTC/NVRAM.  Boot ROM is Flash.  PC-AT-sized
-	  motherboard.  Requires power supply with 3.3V output.
-
 config ALPHA_EIGER
 	bool "Eiger"
 	help
@@ -260,7 +249,7 @@ config ALPHA_EV5
 
 config ALPHA_CIA
 	bool
-	depends on ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_NORITAKE || ALPHA_MIKASA || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR
+	depends on ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_NORITAKE || ALPHA_MIKASA || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_ALCOR
 	default y
 
 config ALPHA_EV56
@@ -350,7 +339,7 @@ config ALPHA_QEMU
 
 
 config ALPHA_SRM
-	bool "Use SRM as bootloader" if ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_NAUTILUS
+	bool "Use SRM as bootloader" if ALPHA_PC164 || ALPHA_TAKARA || ALPHA_ALCOR || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_NAUTILUS
 	depends on TTY
 	default y if ALPHA_MIKASA || ALPHA_SABLE || ALPHA_NORITAKE || ALPHA_DP264 || ALPHA_RAWHIDE || ALPHA_EIGER || ALPHA_WILDFIRE || ALPHA_TITAN || ALPHA_SHARK || ALPHA_MARVEL
 	help
diff --git a/arch/alpha/include/asm/irq.h b/arch/alpha/include/asm/irq.h
index e7d22357a5a6..d83b26b6660f 100644
--- a/arch/alpha/include/asm/irq.h
+++ b/arch/alpha/include/asm/irq.h
@@ -31,8 +31,7 @@
 #  define NR_IRQS      (32768 + 16)    /* marvel - 32 pids */
 # endif
 
-#elif defined(CONFIG_ALPHA_EB164)     || \
-      defined(CONFIG_ALPHA_PC164)     || \
+#elif defined(CONFIG_ALPHA_PC164)     || \
       defined(CONFIG_ALPHA_LX164)
 # define NR_IRQS	35
 
diff --git a/arch/alpha/kernel/Makefile b/arch/alpha/kernel/Makefile
index bd5d62fa4a9a..b6c862dff1f6 100644
--- a/arch/alpha/kernel/Makefile
+++ b/arch/alpha/kernel/Makefile
@@ -60,8 +60,6 @@ obj-$(CONFIG_ALPHA_WILDFIRE)	+= core_wildfire.o
 
 # Board support
 obj-$(CONFIG_ALPHA_ALCOR)	+= sys_alcor.o irq_i8259.o irq_srm.o
-obj-$(CONFIG_ALPHA_EB164)	+= sys_cabriolet.o irq_i8259.o irq_srm.o \
-				   pc873xx.o
 obj-$(CONFIG_ALPHA_LX164)	+= sys_cabriolet.o irq_i8259.o irq_srm.o \
 				   smc37c93x.o
 obj-$(CONFIG_ALPHA_PC164)	+= sys_cabriolet.o irq_i8259.o irq_srm.o \
diff --git a/arch/alpha/kernel/sys_cabriolet.c b/arch/alpha/kernel/sys_cabriolet.c
index 06a950072f82..54e75d4fdbe3 100644
--- a/arch/alpha/kernel/sys_cabriolet.c
+++ b/arch/alpha/kernel/sys_cabriolet.c
@@ -6,8 +6,7 @@
  *	Copyright (C) 1996 Jay A Estabrook
  *	Copyright (C) 1998, 1999, 2000 Richard Henderson
  *
- * Code supporting the Cabriolet (AlphaPC64), EB66+, and EB164,
- * PC164 and LX164.
+ * Code supporting the PC164 and LX164.
  */
 
 #include <linux/kernel.h>
@@ -308,31 +307,6 @@ alphapc164_init_pci(void)
  * The System Vector
  */
 
-#if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_EB164)
-struct alpha_machine_vector eb164_mv __initmv = {
-	.vector_name		= "EB164",
-	DO_EV5_MMU,
-	DO_DEFAULT_RTC,
-	DO_CIA_IO,
-	.machine_check		= cia_machine_check,
-	.max_isa_dma_address	= ALPHA_MAX_ISA_DMA_ADDRESS,
-	.min_io_address		= DEFAULT_IO_BASE,
-	.min_mem_address	= CIA_DEFAULT_MEM_BASE,
-
-	.nr_irqs		= 35,
-	.device_interrupt	= cabriolet_device_interrupt,
-
-	.init_arch		= cia_init_arch,
-	.init_irq		= cabriolet_init_irq,
-	.init_rtc		= common_init_rtc,
-	.init_pci		= cia_cab_init_pci,
-	.kill_arch		= cia_kill_arch,
-	.pci_map_irq		= cabriolet_map_irq,
-	.pci_swizzle		= common_swizzle,
-};
-ALIAS_MV(eb164)
-#endif
-
 #if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_LX164)
 struct alpha_machine_vector lx164_mv __initmv = {
 	.vector_name		= "LX164",
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* [PATCH 14/14] alpha: drop pre-EV56 support
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (12 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 13/14] alpha: cabriolet: remove EV5 CPU support Arnd Bergmann
@ 2024-05-03  8:11 ` Arnd Bergmann
  2024-05-04 15:00   ` Richard Henderson
  2024-06-03  6:02   ` Jiri Slaby
  2024-05-03 16:06 ` [PATCH 00/14] alpha: cleanups for 6.10 Matt Turner
  2024-05-03 16:53 ` John Paul Adrian Glaubitz
  15 siblings, 2 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03  8:11 UTC (permalink / raw)
  To: linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

From: Arnd Bergmann <arnd@arndb.de>

All EV4 machines are already gone, and the remaining EV5 based machines
all support the slightly more modern EV56 generation as well.
Debian only supports EV56 and later.

Drop both of these and build kernels optimized for EV56 and higher
when the "generic" options is selected, tuning for an out-of-order
EV6 pipeline, same as Debian userspace.

Since this was the only supported architecture without 8-bit and
16-bit stores, common kernel code no longer has to worry about
aligning struct members, and existing workarounds from the block
and tty layers can be removed.

The alpha memory management code no longer needs an abstraction
for the differences between EV4 and EV5+.

Link: https://lists.debian.org/debian-alpha/2023/05/msg00009.html
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/alpha/Kconfig                     | 21 +------
 arch/alpha/Makefile                    |  8 +--
 arch/alpha/include/asm/elf.h           |  4 +-
 arch/alpha/include/asm/machvec.h       |  9 ---
 arch/alpha/include/asm/mmu_context.h   | 45 +--------------
 arch/alpha/include/asm/special_insns.h |  5 +-
 arch/alpha/include/asm/tlbflush.h      | 41 +------------
 arch/alpha/include/asm/uaccess.h       | 80 --------------------------
 arch/alpha/include/uapi/asm/compiler.h | 18 ------
 arch/alpha/kernel/machvec_impl.h       | 25 +-------
 arch/alpha/kernel/traps.c              | 31 ----------
 include/linux/blk_types.h              |  6 --
 include/linux/tty.h                    | 14 ++---
 13 files changed, 19 insertions(+), 288 deletions(-)

diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 7987bb548652..50ff06d5b799 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -240,29 +240,14 @@ config ISA_DMA_API
 	bool
 	default y
 
-config ALPHA_EV4
-	bool
-
-config ALPHA_EV5
-	bool
-	default y if ALPHA_RX164 || ALPHA_RAWHIDE || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_SABLE || ALPHA_NORITAKE || ALPHA_MIKASA || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR
-
 config ALPHA_CIA
 	bool
 	depends on ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_NORITAKE || ALPHA_MIKASA || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_ALCOR
 	default y
 
 config ALPHA_EV56
-	bool "EV56 CPU (speed >= 366MHz)?" if ALPHA_ALCOR
-	default y if ALPHA_RX164 || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_PC164 || ALPHA_TAKARA
-
-config ALPHA_EV56
-	prompt "EV56 CPU (speed >= 333MHz)?"
-	depends on ALPHA_NORITAKE || ALPHA_MIKASA
-
-config ALPHA_EV56
-	prompt "EV56 CPU (speed >= 400MHz)?"
-	depends on ALPHA_RAWHIDE
+	bool
+	default y if ALPHA_ALCOR || ALPHA_RX164 || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_RUFFIAN || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_NORITAKE || ALPHA_MIKASA || ALPHA_RAWHIDE || ALPHA_SABLE
 
 config ALPHA_T2
 	bool
@@ -403,7 +388,7 @@ config ARCH_SPARSEMEM_ENABLE
 config ALPHA_WTINT
 	bool "Use WTINT" if ALPHA_SRM || ALPHA_GENERIC
 	default y if ALPHA_QEMU
-	default n if ALPHA_EV5 || ALPHA_EV56
+	default n if ALPHA_EV56
 	default n if !ALPHA_SRM && !ALPHA_GENERIC
 	default y if SMP
 	help
diff --git a/arch/alpha/Makefile b/arch/alpha/Makefile
index 45158024085e..35445ff2e489 100644
--- a/arch/alpha/Makefile
+++ b/arch/alpha/Makefile
@@ -15,18 +15,14 @@ CHECKFLAGS	+= -D__alpha__
 cflags-y	:= -pipe -mno-fp-regs -ffixed-8
 cflags-y	+= $(call cc-option, -fno-jump-tables)
 
-cpuflags-$(CONFIG_ALPHA_EV4)		:= -mcpu=ev4
-cpuflags-$(CONFIG_ALPHA_EV5)		:= -mcpu=ev5
 cpuflags-$(CONFIG_ALPHA_EV56)		:= -mcpu=ev56
 cpuflags-$(CONFIG_ALPHA_POLARIS)	:= -mcpu=pca56
 cpuflags-$(CONFIG_ALPHA_SX164)		:= -mcpu=pca56
 cpuflags-$(CONFIG_ALPHA_EV6)		:= -mcpu=ev6
 cpuflags-$(CONFIG_ALPHA_EV67)		:= -mcpu=ev67
 # If GENERIC, make sure to turn off any instruction set extensions that
-# the host compiler might have on by default.  Given that EV4 and EV5
-# have the same instruction set, prefer EV5 because an EV5 schedule is
-# more likely to keep an EV4 processor busy than vice-versa.
-cpuflags-$(CONFIG_ALPHA_GENERIC)	:= -mcpu=ev5
+# the host compiler might have on by default.
+cpuflags-$(CONFIG_ALPHA_GENERIC)	:= -mcpu=ev56 -mtune=ev6
 
 cflags-y				+= $(cpuflags-y)
 
diff --git a/arch/alpha/include/asm/elf.h b/arch/alpha/include/asm/elf.h
index e6da23f1da83..4d7c46f50382 100644
--- a/arch/alpha/include/asm/elf.h
+++ b/arch/alpha/include/asm/elf.h
@@ -133,9 +133,7 @@ extern int dump_elf_task(elf_greg_t *dest, struct task_struct *task);
 #define ELF_PLATFORM				\
 ({						\
 	enum implver_enum i_ = implver();	\
-	( i_ == IMPLVER_EV4 ? "ev4"		\
-	: i_ == IMPLVER_EV5			\
-	  ? (amask(AMASK_BWX) ? "ev5" : "ev56")	\
+	( i_ == IMPLVER_EV5 ? "ev56"			\
 	: amask (AMASK_CIX) ? "ev6" : "ev67");	\
 })
 
diff --git a/arch/alpha/include/asm/machvec.h b/arch/alpha/include/asm/machvec.h
index 8623f995d34c..490fc880bb3f 100644
--- a/arch/alpha/include/asm/machvec.h
+++ b/arch/alpha/include/asm/machvec.h
@@ -72,15 +72,6 @@ struct alpha_machine_vector
 	int (*mv_is_ioaddr)(unsigned long);
 	int (*mv_is_mmio)(const volatile void __iomem *);
 
-	void (*mv_switch_mm)(struct mm_struct *, struct mm_struct *,
-			     struct task_struct *);
-	void (*mv_activate_mm)(struct mm_struct *, struct mm_struct *);
-
-	void (*mv_flush_tlb_current)(struct mm_struct *);
-	void (*mv_flush_tlb_current_page)(struct mm_struct * mm,
-					  struct vm_area_struct *vma,
-					  unsigned long addr);
-
 	void (*update_irq_hw)(unsigned long, unsigned long, int);
 	void (*ack_irq)(unsigned long);
 	void (*device_interrupt)(unsigned long vector);
diff --git a/arch/alpha/include/asm/mmu_context.h b/arch/alpha/include/asm/mmu_context.h
index 29a3e3a1f02b..eee8fe836a59 100644
--- a/arch/alpha/include/asm/mmu_context.h
+++ b/arch/alpha/include/asm/mmu_context.h
@@ -71,9 +71,7 @@ __reload_thread(struct pcb_struct *pcb)
 #ifdef CONFIG_ALPHA_GENERIC
 # define MAX_ASN	(alpha_mv.max_asn)
 #else
-# ifdef CONFIG_ALPHA_EV4
-#  define MAX_ASN	EV4_MAX_ASN
-# elif defined(CONFIG_ALPHA_EV5)
+# if defined(CONFIG_ALPHA_EV56)
 #  define MAX_ASN	EV5_MAX_ASN
 # else
 #  define MAX_ASN	EV6_MAX_ASN
@@ -162,26 +160,6 @@ ev5_switch_mm(struct mm_struct *prev_mm, struct mm_struct *next_mm,
 	task_thread_info(next)->pcb.asn = mmc & HARDWARE_ASN_MASK;
 }
 
-__EXTERN_INLINE void
-ev4_switch_mm(struct mm_struct *prev_mm, struct mm_struct *next_mm,
-	      struct task_struct *next)
-{
-	/* As described, ASN's are broken for TLB usage.  But we can
-	   optimize for switching between threads -- if the mm is
-	   unchanged from current we needn't flush.  */
-	/* ??? May not be needed because EV4 PALcode recognizes that
-	   ASN's are broken and does a tbiap itself on swpctx, under
-	   the "Must set ASN or flush" rule.  At least this is true
-	   for a 1992 SRM, reports Joseph Martin (jmartin@hlo.dec.com).
-	   I'm going to leave this here anyway, just to Be Sure.  -- r~  */
-	if (prev_mm != next_mm)
-		tbiap();
-
-	/* Do continue to allocate ASNs, because we can still use them
-	   to avoid flushing the icache.  */
-	ev5_switch_mm(prev_mm, next_mm, next);
-}
-
 extern void __load_new_mm_context(struct mm_struct *);
 asmlinkage void do_page_fault(unsigned long address, unsigned long mmcsr,
 			      long cause, struct pt_regs *regs);
@@ -209,25 +187,8 @@ ev5_activate_mm(struct mm_struct *prev_mm, struct mm_struct *next_mm)
 	__load_new_mm_context(next_mm);
 }
 
-__EXTERN_INLINE void
-ev4_activate_mm(struct mm_struct *prev_mm, struct mm_struct *next_mm)
-{
-	__load_new_mm_context(next_mm);
-	tbiap();
-}
-
-#ifdef CONFIG_ALPHA_GENERIC
-# define switch_mm(a,b,c)	alpha_mv.mv_switch_mm((a),(b),(c))
-# define activate_mm(x,y)	alpha_mv.mv_activate_mm((x),(y))
-#else
-# ifdef CONFIG_ALPHA_EV4
-#  define switch_mm(a,b,c)	ev4_switch_mm((a),(b),(c))
-#  define activate_mm(x,y)	ev4_activate_mm((x),(y))
-# else
-#  define switch_mm(a,b,c)	ev5_switch_mm((a),(b),(c))
-#  define activate_mm(x,y)	ev5_activate_mm((x),(y))
-# endif
-#endif
+#define switch_mm(a,b,c)	ev5_switch_mm((a),(b),(c))
+#define activate_mm(x,y)	ev5_activate_mm((x),(y))
 
 #define init_new_context init_new_context
 static inline int
diff --git a/arch/alpha/include/asm/special_insns.h b/arch/alpha/include/asm/special_insns.h
index ca2c5c30b22e..798d0bdb11f9 100644
--- a/arch/alpha/include/asm/special_insns.h
+++ b/arch/alpha/include/asm/special_insns.h
@@ -15,10 +15,7 @@ enum implver_enum {
    (enum implver_enum) __implver; })
 #else
 /* Try to eliminate some dead code.  */
-#ifdef CONFIG_ALPHA_EV4
-#define implver() IMPLVER_EV4
-#endif
-#ifdef CONFIG_ALPHA_EV5
+#ifdef CONFIG_ALPHA_EV56
 #define implver() IMPLVER_EV5
 #endif
 #if defined(CONFIG_ALPHA_EV6)
diff --git a/arch/alpha/include/asm/tlbflush.h b/arch/alpha/include/asm/tlbflush.h
index 94dc37cf873a..02ccac5c5916 100644
--- a/arch/alpha/include/asm/tlbflush.h
+++ b/arch/alpha/include/asm/tlbflush.h
@@ -14,39 +14,12 @@
 extern void __load_new_mm_context(struct mm_struct *);
 
 
-/* Use a few helper functions to hide the ugly broken ASN
-   numbers on early Alphas (ev4 and ev45).  */
-
-__EXTERN_INLINE void
-ev4_flush_tlb_current(struct mm_struct *mm)
-{
-	__load_new_mm_context(mm);
-	tbiap();
-}
-
 __EXTERN_INLINE void
 ev5_flush_tlb_current(struct mm_struct *mm)
 {
 	__load_new_mm_context(mm);
 }
 
-/* Flush just one page in the current TLB set.  We need to be very
-   careful about the icache here, there is no way to invalidate a
-   specific icache page.  */
-
-__EXTERN_INLINE void
-ev4_flush_tlb_current_page(struct mm_struct * mm,
-			   struct vm_area_struct *vma,
-			   unsigned long addr)
-{
-	int tbi_flag = 2;
-	if (vma->vm_flags & VM_EXEC) {
-		__load_new_mm_context(mm);
-		tbi_flag = 3;
-	}
-	tbi(tbi_flag, addr);
-}
-
 __EXTERN_INLINE void
 ev5_flush_tlb_current_page(struct mm_struct * mm,
 			   struct vm_area_struct *vma,
@@ -59,18 +32,8 @@ ev5_flush_tlb_current_page(struct mm_struct * mm,
 }
 
 
-#ifdef CONFIG_ALPHA_GENERIC
-# define flush_tlb_current		alpha_mv.mv_flush_tlb_current
-# define flush_tlb_current_page		alpha_mv.mv_flush_tlb_current_page
-#else
-# ifdef CONFIG_ALPHA_EV4
-#  define flush_tlb_current		ev4_flush_tlb_current
-#  define flush_tlb_current_page	ev4_flush_tlb_current_page
-# else
-#  define flush_tlb_current		ev5_flush_tlb_current
-#  define flush_tlb_current_page	ev5_flush_tlb_current_page
-# endif
-#endif
+#define flush_tlb_current	ev5_flush_tlb_current
+#define flush_tlb_current_page	ev5_flush_tlb_current_page
 
 #ifdef __MMU_EXTERN_INLINE
 #undef __EXTERN_INLINE
diff --git a/arch/alpha/include/asm/uaccess.h b/arch/alpha/include/asm/uaccess.h
index c32c2584c0b7..ef295cbb797c 100644
--- a/arch/alpha/include/asm/uaccess.h
+++ b/arch/alpha/include/asm/uaccess.h
@@ -96,9 +96,6 @@ struct __large_struct { unsigned long buf[100]; };
 		: "=r"(__gu_val), "=r"(__gu_err)	\
 		: "m"(__m(addr)), "1"(__gu_err))
 
-#ifdef __alpha_bwx__
-/* Those lucky bastards with ev56 and later CPUs can do byte/word moves.  */
-
 #define __get_user_16(addr)				\
 	__asm__("1: ldwu %0,%2\n"			\
 	"2:\n"						\
@@ -112,33 +109,6 @@ struct __large_struct { unsigned long buf[100]; };
 	EXC(1b,2b,%0,%1)				\
 		: "=r"(__gu_val), "=r"(__gu_err)	\
 		: "m"(__m(addr)), "1"(__gu_err))
-#else
-/* Unfortunately, we can't get an unaligned access trap for the sub-word
-   load, so we have to do a general unaligned operation.  */
-
-#define __get_user_16(addr)						\
-{									\
-	long __gu_tmp;							\
-	__asm__("1: ldq_u %0,0(%3)\n"					\
-	"2:	ldq_u %1,1(%3)\n"					\
-	"	extwl %0,%3,%0\n"					\
-	"	extwh %1,%3,%1\n"					\
-	"	or %0,%1,%0\n"						\
-	"3:\n"								\
-	EXC(1b,3b,%0,%2)						\
-	EXC(2b,3b,%0,%2)						\
-		: "=&r"(__gu_val), "=&r"(__gu_tmp), "=r"(__gu_err)	\
-		: "r"(addr), "2"(__gu_err));				\
-}
-
-#define __get_user_8(addr)						\
-	__asm__("1: ldq_u %0,0(%2)\n"					\
-	"	extbl %0,%2,%0\n"					\
-	"2:\n"								\
-	EXC(1b,2b,%0,%1)						\
-		: "=&r"(__gu_val), "=r"(__gu_err)			\
-		: "r"(addr), "1"(__gu_err))
-#endif
 
 extern void __put_user_unknown(void);
 
@@ -192,9 +162,6 @@ __asm__ __volatile__("1: stl %r2,%1\n"				\
 		: "=r"(__pu_err)				\
 		: "m"(__m(addr)), "rJ"(x), "0"(__pu_err))
 
-#ifdef __alpha_bwx__
-/* Those lucky bastards with ev56 and later CPUs can do byte/word moves.  */
-
 #define __put_user_16(x, addr)					\
 __asm__ __volatile__("1: stw %r2,%1\n"				\
 	"2:\n"							\
@@ -208,53 +175,6 @@ __asm__ __volatile__("1: stb %r2,%1\n"				\
 	EXC(1b,2b,$31,%0)					\
 		: "=r"(__pu_err)				\
 		: "m"(__m(addr)), "rJ"(x), "0"(__pu_err))
-#else
-/* Unfortunately, we can't get an unaligned access trap for the sub-word
-   write, so we have to do a general unaligned operation.  */
-
-#define __put_user_16(x, addr)					\
-{								\
-	long __pu_tmp1, __pu_tmp2, __pu_tmp3, __pu_tmp4;	\
-	__asm__ __volatile__(					\
-	"1:	ldq_u %2,1(%5)\n"				\
-	"2:	ldq_u %1,0(%5)\n"				\
-	"	inswh %6,%5,%4\n"				\
-	"	inswl %6,%5,%3\n"				\
-	"	mskwh %2,%5,%2\n"				\
-	"	mskwl %1,%5,%1\n"				\
-	"	or %2,%4,%2\n"					\
-	"	or %1,%3,%1\n"					\
-	"3:	stq_u %2,1(%5)\n"				\
-	"4:	stq_u %1,0(%5)\n"				\
-	"5:\n"							\
-	EXC(1b,5b,$31,%0)					\
-	EXC(2b,5b,$31,%0)					\
-	EXC(3b,5b,$31,%0)					\
-	EXC(4b,5b,$31,%0)					\
-		: "=r"(__pu_err), "=&r"(__pu_tmp1), 		\
-		  "=&r"(__pu_tmp2), "=&r"(__pu_tmp3), 		\
-		  "=&r"(__pu_tmp4)				\
-		: "r"(addr), "r"((unsigned long)(x)), "0"(__pu_err)); \
-}
-
-#define __put_user_8(x, addr)					\
-{								\
-	long __pu_tmp1, __pu_tmp2;				\
-	__asm__ __volatile__(					\
-	"1:	ldq_u %1,0(%4)\n"				\
-	"	insbl %3,%4,%2\n"				\
-	"	mskbl %1,%4,%1\n"				\
-	"	or %1,%2,%1\n"					\
-	"2:	stq_u %1,0(%4)\n"				\
-	"3:\n"							\
-	EXC(1b,3b,$31,%0)					\
-	EXC(2b,3b,$31,%0)					\
-		: "=r"(__pu_err), 				\
-	  	  "=&r"(__pu_tmp1), "=&r"(__pu_tmp2)		\
-		: "r"((unsigned long)(x)), "r"(addr), "0"(__pu_err)); \
-}
-#endif
-
 
 /*
  * Complex access routines
diff --git a/arch/alpha/include/uapi/asm/compiler.h b/arch/alpha/include/uapi/asm/compiler.h
index 0e00c0e13374..8c03740966b4 100644
--- a/arch/alpha/include/uapi/asm/compiler.h
+++ b/arch/alpha/include/uapi/asm/compiler.h
@@ -95,24 +95,6 @@
 #define __kernel_ldwu(mem)	(mem)
 #define __kernel_stb(val,mem)	((mem) = (val))
 #define __kernel_stw(val,mem)	((mem) = (val))
-#else
-#define __kernel_ldbu(mem)				\
-  ({ unsigned char __kir;				\
-     __asm__(".arch ev56;				\
-	      ldbu %0,%1" : "=r"(__kir) : "m"(mem));	\
-     __kir; })
-#define __kernel_ldwu(mem)				\
-  ({ unsigned short __kir;				\
-     __asm__(".arch ev56;				\
-	      ldwu %0,%1" : "=r"(__kir) : "m"(mem));	\
-     __kir; })
-#define __kernel_stb(val,mem)				\
-  __asm__(".arch ev56;					\
-	   stb %1,%0" : "=m"(mem) : "r"(val))
-#define __kernel_stw(val,mem)				\
-  __asm__(".arch ev56;					\
-	   stw %1,%0" : "=m"(mem) : "r"(val))
 #endif
 
-
 #endif /* _UAPI__ALPHA_COMPILER_H */
diff --git a/arch/alpha/kernel/machvec_impl.h b/arch/alpha/kernel/machvec_impl.h
index c2ebcb39e589..129ae36b8e6d 100644
--- a/arch/alpha/kernel/machvec_impl.h
+++ b/arch/alpha/kernel/machvec_impl.h
@@ -44,33 +44,14 @@
 
 #define DO_DEFAULT_RTC			.rtc_port = 0x70
 
-#define DO_EV4_MMU							\
-	.max_asn =			EV4_MAX_ASN,			\
-	.mv_switch_mm =			ev4_switch_mm,			\
-	.mv_activate_mm =		ev4_activate_mm,		\
-	.mv_flush_tlb_current =		ev4_flush_tlb_current,		\
-	.mv_flush_tlb_current_page =	ev4_flush_tlb_current_page
-
 #define DO_EV5_MMU							\
-	.max_asn =			EV5_MAX_ASN,			\
-	.mv_switch_mm =			ev5_switch_mm,			\
-	.mv_activate_mm =		ev5_activate_mm,		\
-	.mv_flush_tlb_current =		ev5_flush_tlb_current,		\
-	.mv_flush_tlb_current_page =	ev5_flush_tlb_current_page
+	.max_asn =			EV5_MAX_ASN			\
 
 #define DO_EV6_MMU							\
-	.max_asn =			EV6_MAX_ASN,			\
-	.mv_switch_mm =			ev5_switch_mm,			\
-	.mv_activate_mm =		ev5_activate_mm,		\
-	.mv_flush_tlb_current =		ev5_flush_tlb_current,		\
-	.mv_flush_tlb_current_page =	ev5_flush_tlb_current_page
+	.max_asn =			EV6_MAX_ASN			\
 
 #define DO_EV7_MMU							\
-	.max_asn =			EV6_MAX_ASN,			\
-	.mv_switch_mm =			ev5_switch_mm,			\
-	.mv_activate_mm =		ev5_activate_mm,		\
-	.mv_flush_tlb_current =		ev5_flush_tlb_current,		\
-	.mv_flush_tlb_current_page =	ev5_flush_tlb_current_page
+	.max_asn =			EV6_MAX_ASN			\
 
 #define IO_LITE(UP,low)							\
 	.hae_register =		(unsigned long *) CAT(UP,_HAE_ADDRESS),	\
diff --git a/arch/alpha/kernel/traps.c b/arch/alpha/kernel/traps.c
index a9a8e9ab0f52..6afae65e9a8b 100644
--- a/arch/alpha/kernel/traps.c
+++ b/arch/alpha/kernel/traps.c
@@ -320,32 +320,6 @@ do_entIF(unsigned long type, struct pt_regs *regs)
 		return;
 
 	      case 4: /* opDEC */
-		if (implver() == IMPLVER_EV4) {
-			long si_code;
-
-			/* The some versions of SRM do not handle
-			   the opDEC properly - they return the PC of the
-			   opDEC fault, not the instruction after as the
-			   Alpha architecture requires.  Here we fix it up.
-			   We do this by intentionally causing an opDEC
-			   fault during the boot sequence and testing if
-			   we get the correct PC.  If not, we set a flag
-			   to correct it every time through.  */
-			regs->pc += opDEC_fix; 
-			
-			/* EV4 does not implement anything except normal
-			   rounding.  Everything else will come here as
-			   an illegal instruction.  Emulate them.  */
-			si_code = alpha_fp_emul(regs->pc - 4);
-			if (si_code == 0)
-				return;
-			if (si_code > 0) {
-				send_sig_fault_trapno(SIGFPE, si_code,
-						      (void __user *) regs->pc,
-						      0, current);
-				return;
-			}
-		}
 		break;
 
 	      case 5: /* illoc */
@@ -946,11 +920,6 @@ trap_init(void)
 	register unsigned long gptr __asm__("$29");
 	wrkgp(gptr);
 
-	/* Hack for Multia (UDB) and JENSEN: some of their SRMs have
-	   a bug in the handling of the opDEC fault.  Fix it up if so.  */
-	if (implver() == IMPLVER_EV4)
-		opDEC_check();
-
 	wrent(entArith, 1);
 	wrent(entMM, 2);
 	wrent(entIF, 3);
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index cb1526ec44b5..c3e098b21c16 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -88,15 +88,9 @@ struct block_device {
 
 /*
  * Block error status values.  See block/blk-core:blk_errors for the details.
- * Alpha cannot write a byte atomically, so we need to use 32-bit value.
  */
-#if defined(CONFIG_ALPHA) && !defined(__alpha_bwx__)
-typedef u32 __bitwise blk_status_t;
-typedef u32 blk_short_t;
-#else
 typedef u8 __bitwise blk_status_t;
 typedef u16 blk_short_t;
-#endif
 #define	BLK_STS_OK 0
 #define BLK_STS_NOTSUPP		((__force blk_status_t)1)
 #define BLK_STS_TIMEOUT		((__force blk_status_t)2)
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 2b2e6f0a54d6..2372f9357240 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -145,15 +145,12 @@ struct tty_operations;
  * @count: count of open processes, reaching zero cancels all the work for
  *	   this tty and drops a @kref too (but does not free this tty)
  * @winsize: size of the terminal "window" (cf. @winsize_mutex)
- * @flow: flow settings grouped together, see also @flow.unused
+ * @flow: flow settings grouped together
  * @flow.lock: lock for @flow members
  * @flow.stopped: tty stopped/started by stop_tty()/start_tty()
  * @flow.tco_stopped: tty stopped/started by %TCOOFF/%TCOON ioctls (it has
  *		      precedence over @flow.stopped)
- * @flow.unused: alignment for Alpha, so that no members other than @flow.* are
- *		 modified by the same 64b word store. The @flow's __aligned is
- *		 there for the very same reason.
- * @ctrl: control settings grouped together, see also @ctrl.unused
+ * @ctrl: control settings grouped together
  * @ctrl.lock: lock for @ctrl members
  * @ctrl.pgrp: process group of this tty (setpgrp(2))
  * @ctrl.session: session of this tty (setsid(2)). Writes are protected by both
@@ -161,7 +158,6 @@ struct tty_operations;
  *		  them.
  * @ctrl.pktstatus: packet mode status (bitwise OR of %TIOCPKT_ constants)
  * @ctrl.packet: packet mode enabled
- * @ctrl.unused: alignment for Alpha, see @flow.unused for explanation
  * @hw_stopped: not controlled by the tty layer, under @driver's control for CTS
  *		handling
  * @receive_room: bytes permitted to feed to @ldisc without any being lost
@@ -216,8 +212,7 @@ struct tty_struct {
 		spinlock_t lock;
 		bool stopped;
 		bool tco_stopped;
-		unsigned long unused[0];
-	} __aligned(sizeof(unsigned long)) flow;
+	} flow;
 
 	struct {
 		struct pid *pgrp;
@@ -225,8 +220,7 @@ struct tty_struct {
 		spinlock_t lock;
 		unsigned char pktstatus;
 		bool packet;
-		unsigned long unused[0];
-	} __aligned(sizeof(unsigned long)) ctrl;
+	} ctrl;
 
 	bool hw_stopped;
 	bool closing;
-- 
2.39.2


^ permalink raw reply related	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (13 preceding siblings ...)
  2024-05-03  8:11 ` [PATCH 14/14] alpha: drop pre-EV56 support Arnd Bergmann
@ 2024-05-03 16:06 ` Matt Turner
  2024-05-03 20:15   ` Arnd Bergmann
  2024-05-03 16:53 ` John Paul Adrian Glaubitz
  15 siblings, 1 reply; 55+ messages in thread
From: Matt Turner @ 2024-05-03 16:06 UTC (permalink / raw)
  To: Arnd Bergmann
  Cc: linux-alpha, Arnd Bergmann, Richard Henderson, Ivan Kokshaysky,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

On Fri, May 3, 2024 at 4:12 AM Arnd Bergmann <arnd@kernel.org> wrote:
>
> From: Arnd Bergmann <arnd@arndb.de>
>
> I had investigated dropping support for alpha EV5 and earlier a while
> ago after noticing that this is the only supported CPU family
> in the kernel without native byte access and that Debian has already
> dropped support for this generation last year [1] after it turned
> out to be broken.
>
> This topic came up again when Paul E. McKenney noticed that
> parts of the RCU code already rely on byte access and do not
> work on alpha EV5 reliably, so I refreshed my series now for
> inclusion into the next merge window.
>
> Al Viro did another series for alpha to address all the known build
> issues. I rebased his patches without any further changes and included
> it as a baseline for my work here to avoid conflicts.

Thanks for all this. Removing support for non-BWX alphas makes a lot
of sense to me.

The whole series is

Acked-by: Matt Turner <mattst88@gmail.com>

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 10/14] alpha: remove DECpc AXP150 (Jensen) support
  2024-05-03  8:11 ` [PATCH 10/14] alpha: remove DECpc AXP150 (Jensen) support Arnd Bergmann
@ 2024-05-03 16:07   ` Linus Torvalds
  2024-05-03 17:00   ` Al Viro
  1 sibling, 0 replies; 55+ messages in thread
From: Linus Torvalds @ 2024-05-03 16:07 UTC (permalink / raw)
  To: Arnd Bergmann
  Cc: linux-alpha, Arnd Bergmann, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, Paul E. McKenney,
	linux-kernel

On Fri, 3 May 2024 at 01:12, Arnd Bergmann <arnd@kernel.org> wrote:
>
> From: Arnd Bergmann <arnd@arndb.de>
>
> This is one of the hackiest Alpha machines, and the only one without
> PCI support. Removing this allows cleaning up code in eise and tty
> drivers in addition to the architecture code.

Oh well, The axp150 was the machine I used originally, so it's a bit
sad to see it go.

But yeah, good riddance. The lack of byte and word operations were a
fundamental mistake and made those early alphas very painful.

The design team obviously made other technical mistakes (sw fill tlb
etc, with memory ordering being the one that never got fixed), but the
byte were the killer for any sanity both on the IO side and the code
generation side.

                  Linus

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
                   ` (14 preceding siblings ...)
  2024-05-03 16:06 ` [PATCH 00/14] alpha: cleanups for 6.10 Matt Turner
@ 2024-05-03 16:53 ` John Paul Adrian Glaubitz
  2024-05-03 17:19   ` Paul E. McKenney
  2024-05-27 23:49   ` Maciej W. Rozycki
  15 siblings, 2 replies; 55+ messages in thread
From: John Paul Adrian Glaubitz @ 2024-05-03 16:53 UTC (permalink / raw)
  To: Arnd Bergmann, linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel, Michael Cree, Frank Scheiner

Hello Arnd,

On Fri, 2024-05-03 at 10:11 +0200, Arnd Bergmann wrote:
> I had investigated dropping support for alpha EV5 and earlier a while
> ago after noticing that this is the only supported CPU family
> in the kernel without native byte access and that Debian has already
> dropped support for this generation last year [1] after it turned
> out to be broken.

That's not quite correct. Support for older Alphas is not broken and
always worked when I tested it. It's just that some people wanted to
raise the baseline in order to improve code performance on newer machines
with the hope to fix some minor issues we saw on Alpha here and there.

> This topic came up again when Paul E. McKenney noticed that
> parts of the RCU code already rely on byte access and do not
> work on alpha EV5 reliably, so I refreshed my series now for
> inclusion into the next merge window.

Hrrrm? That sounds like like Paul ran tests on EV5, did he?

> Al Viro did another series for alpha to address all the known build
> issues. I rebased his patches without any further changes and included
> it as a baseline for my work here to avoid conflicts.

It's somewhat strange that Al improves code on the older machines only
to be axed by your series. I would prefer such removals to aimed at an
LTS release, if possible.

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 10/14] alpha: remove DECpc AXP150 (Jensen) support
  2024-05-03  8:11 ` [PATCH 10/14] alpha: remove DECpc AXP150 (Jensen) support Arnd Bergmann
  2024-05-03 16:07   ` Linus Torvalds
@ 2024-05-03 17:00   ` Al Viro
  2024-05-03 20:07     ` Arnd Bergmann
  1 sibling, 1 reply; 55+ messages in thread
From: Al Viro @ 2024-05-03 17:00 UTC (permalink / raw)
  To: Arnd Bergmann
  Cc: linux-alpha, Arnd Bergmann, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel

On Fri, May 03, 2024 at 10:11:21AM +0200, Arnd Bergmann wrote:
> diff --git a/arch/alpha/kernel/traps.c b/arch/alpha/kernel/traps.c
> index 7fc72aeb7398..a9a8e9ab0f52 100644
> --- a/arch/alpha/kernel/traps.c
> +++ b/arch/alpha/kernel/traps.c
> @@ -30,39 +30,6 @@
>  
>  #include "proto.h"
>  
> -/* Work-around for some SRMs which mishandle opDEC faults.  */
> -
> -static int opDEC_fix;
> -
> -static void
> -opDEC_check(void)
> -{
> -	__asm__ __volatile__ (
> -	/* Load the address of... */
> -	"	br	$16, 1f\n"
> -	/* A stub instruction fault handler.  Just add 4 to the
> -	   pc and continue.  */
> -	"	ldq	$16, 8($sp)\n"
> -	"	addq	$16, 4, $16\n"
> -	"	stq	$16, 8($sp)\n"
> -	"	call_pal %[rti]\n"
> -	/* Install the instruction fault handler.  */
> -	"1:	lda	$17, 3\n"
> -	"	call_pal %[wrent]\n"
> -	/* With that in place, the fault from the round-to-minf fp
> -	   insn will arrive either at the "lda 4" insn (bad) or one
> -	   past that (good).  This places the correct fixup in %0.  */
> -	"	lda %[fix], 0\n"
> -	"	cvttq/svm $f31,$f31\n"
> -	"	lda %[fix], 4"
> -	: [fix] "=r" (opDEC_fix)
> -	: [rti] "n" (PAL_rti), [wrent] "n" (PAL_wrent)
> -	: "$0", "$1", "$16", "$17", "$22", "$23", "$24", "$25");
> -
> -	if (opDEC_fix)
> -		printk("opDEC fixup enabled.\n");
> -}
> -

That should be moved into commit that removes the caller...

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-03 16:53 ` John Paul Adrian Glaubitz
@ 2024-05-03 17:19   ` Paul E. McKenney
  2024-05-27 23:49   ` Maciej W. Rozycki
  1 sibling, 0 replies; 55+ messages in thread
From: Paul E. McKenney @ 2024-05-03 17:19 UTC (permalink / raw)
  To: John Paul Adrian Glaubitz
  Cc: Arnd Bergmann, linux-alpha, Arnd Bergmann, Richard Henderson,
	Ivan Kokshaysky, Matt Turner, Alexander Viro, Marc Zyngier,
	Linus Torvalds, linux-kernel, Michael Cree, Frank Scheiner

On Fri, May 03, 2024 at 06:53:45PM +0200, John Paul Adrian Glaubitz wrote:
> Hello Arnd,
> 
> On Fri, 2024-05-03 at 10:11 +0200, Arnd Bergmann wrote:
> > I had investigated dropping support for alpha EV5 and earlier a while
> > ago after noticing that this is the only supported CPU family
> > in the kernel without native byte access and that Debian has already
> > dropped support for this generation last year [1] after it turned
> > out to be broken.
> 
> That's not quite correct. Support for older Alphas is not broken and
> always worked when I tested it. It's just that some people wanted to
> raise the baseline in order to improve code performance on newer machines
> with the hope to fix some minor issues we saw on Alpha here and there.
> 
> > This topic came up again when Paul E. McKenney noticed that
> > parts of the RCU code already rely on byte access and do not
> > work on alpha EV5 reliably, so I refreshed my series now for
> > inclusion into the next merge window.
> 
> Hrrrm? That sounds like like Paul ran tests on EV5, did he?

Arnd does say "noticed", not "tested".  No Alpha CPUs here, and I don't
run Alpha emulators.  There is only so much time in each day and only
so much budget for electricity.  ;-)

For the series: Acked-by: Paul E. McKenney <paulmck@kernel.org>

> > Al Viro did another series for alpha to address all the known build
> > issues. I rebased his patches without any further changes and included
> > it as a baseline for my work here to avoid conflicts.
> 
> It's somewhat strange that Al improves code on the older machines only
> to be axed by your series. I would prefer such removals to aimed at an
> LTS release, if possible.

Once they are in mainline, you are within your rights to send Al's
code-improvement patches to -stable, which should get them to the LTS
releases.  It might well be that Arnd was planning to do just that.

							Thanx, Paul

> Adrian
> 
> -- 
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer
> `. `'   Physicist
>   `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 10/14] alpha: remove DECpc AXP150 (Jensen) support
  2024-05-03 17:00   ` Al Viro
@ 2024-05-03 20:07     ` Arnd Bergmann
  0 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03 20:07 UTC (permalink / raw)
  To: Alexander Viro, Arnd Bergmann
  Cc: linux-alpha, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Marc Zyngier, Linus Torvalds, Paul E. McKenney, linux-kernel

On Fri, May 3, 2024, at 19:00, Al Viro wrote:
>
> That should be moved into commit that removes the caller...

Fixed now, thanks!

    Arnd

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-03 16:06 ` [PATCH 00/14] alpha: cleanups for 6.10 Matt Turner
@ 2024-05-03 20:15   ` Arnd Bergmann
  2024-05-06  9:16     ` Michael Cree
  0 siblings, 1 reply; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-03 20:15 UTC (permalink / raw)
  To: Matt Turner, Arnd Bergmann
  Cc: linux-alpha, Richard Henderson, Ivan Kokshaysky, Alexander Viro,
	Marc Zyngier, Linus Torvalds, Paul E. McKenney, linux-kernel

On Fri, May 3, 2024, at 18:06, Matt Turner wrote:
> On Fri, May 3, 2024 at 4:12 AM Arnd Bergmann <arnd@kernel.org> wrote:
>>
>> Al Viro did another series for alpha to address all the known build
>> issues. I rebased his patches without any further changes and included
>> it as a baseline for my work here to avoid conflicts.
>
> Thanks for all this. Removing support for non-BWX alphas makes a lot
> of sense to me.
>
> The whole series is
>
> Acked-by: Matt Turner <mattst88@gmail.com>

Thanks!

I've pushed out the series with the additional Acks to
https://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic.git/log/?h=alpha-cleanup-6.9
and merged it into the main asm-generic branch for 6.10.

Can you give this a quick test on one of your machines to make
sure I didn't introduce a stupid regression somewhere?

     Arnd

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 14/14] alpha: drop pre-EV56 support
  2024-05-03  8:11 ` [PATCH 14/14] alpha: drop pre-EV56 support Arnd Bergmann
@ 2024-05-04 15:00   ` Richard Henderson
  2024-05-06 10:06     ` Arnd Bergmann
  2024-06-03  6:02   ` Jiri Slaby
  1 sibling, 1 reply; 55+ messages in thread
From: Richard Henderson @ 2024-05-04 15:00 UTC (permalink / raw)
  To: Arnd Bergmann, linux-alpha
  Cc: Arnd Bergmann, Ivan Kokshaysky, Matt Turner, Alexander Viro,
	Marc Zyngier, Linus Torvalds, Paul E. McKenney, linux-kernel

On 5/3/24 01:11, Arnd Bergmann wrote:
>   
> -/* Flush just one page in the current TLB set.  We need to be very
> -   careful about the icache here, there is no way to invalidate a
> -   specific icache page.  */
> -
> -__EXTERN_INLINE void
> -ev4_flush_tlb_current_page(struct mm_struct * mm,
> -			   struct vm_area_struct *vma,
> -			   unsigned long addr)
> -{
> -	int tbi_flag = 2;
> -	if (vma->vm_flags & VM_EXEC) {
> -		__load_new_mm_context(mm);
> -		tbi_flag = 3;
> -	}
> -	tbi(tbi_flag, addr);
> -}
> -
>   __EXTERN_INLINE void
>   ev5_flush_tlb_current_page(struct mm_struct * mm,
>   			   struct vm_area_struct *vma,

The comment being removed applied to both functions and still applies to 
ev5_flush_tlb_current_page.  (Thus __load_new_mm_context instead of a tbi.)


r~

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-03 20:15   ` Arnd Bergmann
@ 2024-05-06  9:16     ` Michael Cree
  2024-05-06 10:11       ` Arnd Bergmann
  0 siblings, 1 reply; 55+ messages in thread
From: Michael Cree @ 2024-05-06  9:16 UTC (permalink / raw)
  To: Arnd Bergmann
  Cc: Matt Turner, Arnd Bergmann, linux-alpha, Richard Henderson,
	Ivan Kokshaysky, Alexander Viro, Marc Zyngier, Linus Torvalds,
	Paul E. McKenney, linux-kernel

On Fri, May 03, 2024 at 10:15:10PM +0200, Arnd Bergmann wrote:
> On Fri, May 3, 2024, at 18:06, Matt Turner wrote:
> > On Fri, May 3, 2024 at 4:12 AM Arnd Bergmann <arnd@kernel.org> wrote:
> >>
> >> Al Viro did another series for alpha to address all the known build
> >> issues. I rebased his patches without any further changes and included
> >> it as a baseline for my work here to avoid conflicts.
> 
> I've pushed out the series with the additional Acks to
> https://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic.git/log/?h=alpha-cleanup-6.9
> and merged it into the main asm-generic branch for 6.10.
> 
> Can you give this a quick test on one of your machines to make
> sure I didn't introduce a stupid regression somewhere?

I built a dp264 specific kernel and its working fine on an XP1000
(EV67 arch).  Just built a generic kernel (as that's probably a more
important test) and that is also working fine on the XP1000.

I also built a titan specific kernel to test on an ES45 (SMP EV68
arch) but that OOPSes early in the boot process with a kernel null
pointer access.  I suspect that has nothing to do with your patches
as I have a recollection that I have seen that OOPS before. So I
tried the same generic kernel that I have running on the XP1000 but
that fails to unpack at the initial boot stage (!) with:

aboot> 9
aboot: loading uncompressed vmlinuz.test...
aboot: loading compressed vmlinuz.test...
aboot: Can't load kernel.
  Memory at fffffc0000310000 - fffffc0000f45df7 (PHDR 0) is Busy (Reserved)

unzip: invalid exec header
aboot> 

Cheers,
Michael.

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 14/14] alpha: drop pre-EV56 support
  2024-05-04 15:00   ` Richard Henderson
@ 2024-05-06 10:06     ` Arnd Bergmann
  0 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-06 10:06 UTC (permalink / raw)
  To: Richard Henderson, Arnd Bergmann, linux-alpha
  Cc: Ivan Kokshaysky, Matt Turner, Alexander Viro, Marc Zyngier,
	Linus Torvalds, Paul E. McKenney, linux-kernel

On Sat, May 4, 2024, at 17:00, Richard Henderson wrote:
> On 5/3/24 01:11, Arnd Bergmann wrote:
>>   
>> -/* Flush just one page in the current TLB set.  We need to be very
>> -   careful about the icache here, there is no way to invalidate a
>> -   specific icache page.  */
>> -
>> -__EXTERN_INLINE void
>> -ev4_flush_tlb_current_page(struct mm_struct * mm,
>> -			   struct vm_area_struct *vma,
>> -			   unsigned long addr)
>> -{
>> -	int tbi_flag = 2;
>> -	if (vma->vm_flags & VM_EXEC) {
>> -		__load_new_mm_context(mm);
>> -		tbi_flag = 3;
>> -	}
>> -	tbi(tbi_flag, addr);
>> -}
>> -
>>   __EXTERN_INLINE void
>>   ev5_flush_tlb_current_page(struct mm_struct * mm,
>>   			   struct vm_area_struct *vma,
>
> The comment being removed applied to both functions and still applies to 
> ev5_flush_tlb_current_page.  (Thus __load_new_mm_context instead of a tbi.)

Fixed now, thanks for taking a look!

     Arnd

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-06  9:16     ` Michael Cree
@ 2024-05-06 10:11       ` Arnd Bergmann
  0 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-06 10:11 UTC (permalink / raw)
  To: Michael Cree
  Cc: Matt Turner, Arnd Bergmann, linux-alpha, Richard Henderson,
	Ivan Kokshaysky, Alexander Viro, Marc Zyngier, Linus Torvalds,
	Paul E. McKenney, linux-kernel

On Mon, May 6, 2024, at 11:16, Michael Cree wrote:
> On Fri, May 03, 2024 at 10:15:10PM +0200, Arnd Bergmann wrote:
>> On Fri, May 3, 2024, at 18:06, Matt Turner wrote:
>> > On Fri, May 3, 2024 at 4:12 AM Arnd Bergmann <arnd@kernel.org> wrote:
>> >>
>> >> Al Viro did another series for alpha to address all the known build
>> >> issues. I rebased his patches without any further changes and included
>> >> it as a baseline for my work here to avoid conflicts.
>> 
>> I've pushed out the series with the additional Acks to
>> https://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic.git/log/?h=alpha-cleanup-6.9
>> and merged it into the main asm-generic branch for 6.10.
>> 
>> Can you give this a quick test on one of your machines to make
>> sure I didn't introduce a stupid regression somewhere?
>
> I built a dp264 specific kernel and its working fine on an XP1000
> (EV67 arch).  Just built a generic kernel (as that's probably a more
> important test) and that is also working fine on the XP1000.

Thanks a lot for testing!

> I also built a titan specific kernel to test on an ES45 (SMP EV68
> arch) but that OOPSes early in the boot process with a kernel null
> pointer access.  I suspect that has nothing to do with your patches
> as I have a recollection that I have seen that OOPS before.

Ok

> So I tried the same generic kernel that I have running on the XP1000
> but that fails to unpack at the initial boot stage (!) with:

Just to be sure: this is not a regression from my patches either,
right? I would expect that any EV6 (or later) target specific
kernel didn't change much at all from my patches, while the
generic kernel changing from -march=ev4 to -march=ev56 is
certainly meant to be different (in a good way, with smaller
size and faster execution).

     Arnd

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-03 16:53 ` John Paul Adrian Glaubitz
  2024-05-03 17:19   ` Paul E. McKenney
@ 2024-05-27 23:49   ` Maciej W. Rozycki
  2024-05-28 14:43     ` Paul E. McKenney
  1 sibling, 1 reply; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-05-27 23:49 UTC (permalink / raw)
  To: John Paul Adrian Glaubitz
  Cc: Arnd Bergmann, linux-alpha, Arnd Bergmann, Richard Henderson,
	Ivan Kokshaysky, Matt Turner, Alexander Viro, Marc Zyngier,
	Linus Torvalds, Paul E. McKenney, linux-kernel, Michael Cree,
	Frank Scheiner

On Fri, 3 May 2024, John Paul Adrian Glaubitz wrote:

> > I had investigated dropping support for alpha EV5 and earlier a while
> > ago after noticing that this is the only supported CPU family
> > in the kernel without native byte access and that Debian has already
> > dropped support for this generation last year [1] after it turned
> > out to be broken.
> 
> That's not quite correct. Support for older Alphas is not broken and
> always worked when I tested it. It's just that some people wanted to
> raise the baseline in order to improve code performance on newer machines
> with the hope to fix some minor issues we saw on Alpha here and there.

 I'm not quite happy to see pre-EV5 support go as EV45 is all the Alpha 
hardware I have and it's only owing to issues with the firmware of my 
console manager hardware that I haven't deployed it at my lab yet for 
Linux and GNU toolchain verification.  I'd rather I wasn't stuck with an 
obsolete version of Linux.

> > This topic came up again when Paul E. McKenney noticed that
> > parts of the RCU code already rely on byte access and do not
> > work on alpha EV5 reliably, so I refreshed my series now for
> > inclusion into the next merge window.
> 
> Hrrrm? That sounds like like Paul ran tests on EV5, did he?

 What exactly is required to make it work?

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-27 23:49   ` Maciej W. Rozycki
@ 2024-05-28 14:43     ` Paul E. McKenney
  2024-05-29 18:50       ` Maciej W. Rozycki
  0 siblings, 1 reply; 55+ messages in thread
From: Paul E. McKenney @ 2024-05-28 14:43 UTC (permalink / raw)
  To: Maciej W. Rozycki
  Cc: John Paul Adrian Glaubitz, Arnd Bergmann, linux-alpha,
	Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, linux-kernel,
	Michael Cree, Frank Scheiner

On Tue, May 28, 2024 at 12:49:16AM +0100, Maciej W. Rozycki wrote:
> On Fri, 3 May 2024, John Paul Adrian Glaubitz wrote:
> 
> > > I had investigated dropping support for alpha EV5 and earlier a while
> > > ago after noticing that this is the only supported CPU family
> > > in the kernel without native byte access and that Debian has already
> > > dropped support for this generation last year [1] after it turned
> > > out to be broken.
> > 
> > That's not quite correct. Support for older Alphas is not broken and
> > always worked when I tested it. It's just that some people wanted to
> > raise the baseline in order to improve code performance on newer machines
> > with the hope to fix some minor issues we saw on Alpha here and there.
> 
>  I'm not quite happy to see pre-EV5 support go as EV45 is all the Alpha 
> hardware I have and it's only owing to issues with the firmware of my 
> console manager hardware that I haven't deployed it at my lab yet for 
> Linux and GNU toolchain verification.  I'd rather I wasn't stuck with an 
> obsolete version of Linux.
> 
> > > This topic came up again when Paul E. McKenney noticed that
> > > parts of the RCU code already rely on byte access and do not
> > > work on alpha EV5 reliably, so I refreshed my series now for
> > > inclusion into the next merge window.
> > 
> > Hrrrm? That sounds like like Paul ran tests on EV5, did he?
> 
>  What exactly is required to make it work?

Whatever changes are needed to prevent the data corruption that can
currently result in code generated by single-byte stores.  For but one
example, consider a pair of tasks (or one task and an interrupt handler
in the CONFIG_SMP=n case) do a single-byte store to a pair of bytes
in the same machine word.  As I understand it, in code generated for
older Alphas, both "stores" will load the word containing that byte,
update their own byte, and store the updated word.

If two such single-byte stores run concurrently, one or the other of those
two stores will be lost, as in overwritten by the other.  This is a bug,
even in kernels built for single-CPU systems.  And a rare bug at that, one
that tends to disappear as you add debug code in an attempt to find it.

So if you want to run current kernels on old Alphas, you will need to
do something to fix this.

There might well be other things in need of fixing, for but one example,
it might be that the same issue will soon need to be addressed for
two-byte stores.  You will therefore need to carefully investigate this
issue to determine the full extent of work required to solve it.

							Thanx, Paul

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-28 14:43     ` Paul E. McKenney
@ 2024-05-29 18:50       ` Maciej W. Rozycki
  2024-05-29 22:09         ` Paul E. McKenney
                           ` (2 more replies)
  0 siblings, 3 replies; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-05-29 18:50 UTC (permalink / raw)
  To: Paul E. McKenney
  Cc: John Paul Adrian Glaubitz, Arnd Bergmann, linux-alpha,
	Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, linux-kernel,
	Michael Cree, Frank Scheiner

On Tue, 28 May 2024, Paul E. McKenney wrote:

> > > > This topic came up again when Paul E. McKenney noticed that
> > > > parts of the RCU code already rely on byte access and do not
> > > > work on alpha EV5 reliably, so I refreshed my series now for
> > > > inclusion into the next merge window.
> > > 
> > > Hrrrm? That sounds like like Paul ran tests on EV5, did he?
> > 
> >  What exactly is required to make it work?
> 
> Whatever changes are needed to prevent the data corruption that can
> currently result in code generated by single-byte stores.  For but one
> example, consider a pair of tasks (or one task and an interrupt handler
> in the CONFIG_SMP=n case) do a single-byte store to a pair of bytes
> in the same machine word.  As I understand it, in code generated for
> older Alphas, both "stores" will load the word containing that byte,
> update their own byte, and store the updated word.
> 
> If two such single-byte stores run concurrently, one or the other of those
> two stores will be lost, as in overwritten by the other.  This is a bug,
> even in kernels built for single-CPU systems.  And a rare bug at that, one
> that tends to disappear as you add debug code in an attempt to find it.

 Thank you for the detailed description of the problematic scenario.

 I hope someone will find it useful, however for the record I have been 
familiar with the intricacies of the Alpha architecture as well as their 
implications for software for decades now.  The Alpha port of Linux was 
the first non-x86 Linux platform I have used and actually (and I've chased 
that as a matter of interest) my first ever contribution to Linux was for 
Alpha platform code:

On Mon, 30 Mar 1998, Jay.Estabrook@digital.com wrote:

> Hi, sorry about the delay in answering, but you'll be happy to know, I took
> your patches and merged them into my latest SMP patches, and submitted them
> to Linus just last night. He promises them to (mostly) be in 2.1.92, so we
> can look forward to that... :-)

so I find the scenario you have described more than obvious.

 Mind that the read-modify-write sequence that software does for sub-word 
write accesses with original Alpha hardware is precisely what hardware 
would have to do anyway and support for that was deliberately omitted by 
the architecture designers from the ISA to give it performance advantages 
quoted in the architecture manual.  The only difference here is that with 
hardware read-modify-write operations atomicity for sub-word accesses is 
guaranteed by the ISA, however for software read-modify-write it has to be 
explictly coded using the usual load-locked/store-conditional sequence in 
a loop.  I don't think it's a big deal really, it should be trivial to do 
in the relevant accessors, along with the memory barriers that are needed 
anyway for EV56+ and possibly other ports such as the MIPS one.

 What I have been after actually is: can you point me at a piece of code 
in our tree that will cause an issue with a non-BWX Alpha as described in 
your scenario, so that I have a starting point?  Mind that I'm completely 
new to RCU as I didn't have a need to research it before (though from a 
skim over Documentation/RCU/rcu.rst I understand what its objective is).

 FWIW even if it was only me I think that depriving the already thin Alpha 
port developer base of any quantity of the remaining manpower, by dropping 
support for a subset of the hardware available, and then a subset that is 
not just as exotic as the original i386 became to the x86 platform at the 
time support for it was dropped, is only going to lead to further demise 
and eventual drop of the entire port.

 And I think it would be good if we kept the port, just as we keep other 
ports of historical significance only, for educational reasons if nothing 
else, such as to let people understand based on an actual example, once 
mainstream, the implications of weakly ordered memory systems.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-29 18:50       ` Maciej W. Rozycki
@ 2024-05-29 22:09         ` Paul E. McKenney
  2024-05-30 22:59           ` Maciej W. Rozycki
  2024-05-31  3:56           ` Maciej W. Rozycki
  2024-05-30  1:08         ` Linus Torvalds
  2024-05-31 15:48         ` Arnd Bergmann
  2 siblings, 2 replies; 55+ messages in thread
From: Paul E. McKenney @ 2024-05-29 22:09 UTC (permalink / raw)
  To: Maciej W. Rozycki
  Cc: John Paul Adrian Glaubitz, Arnd Bergmann, linux-alpha,
	Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, linux-kernel,
	Michael Cree, Frank Scheiner

On Wed, May 29, 2024 at 07:50:28PM +0100, Maciej W. Rozycki wrote:
> On Tue, 28 May 2024, Paul E. McKenney wrote:
> 
> > > > > This topic came up again when Paul E. McKenney noticed that
> > > > > parts of the RCU code already rely on byte access and do not
> > > > > work on alpha EV5 reliably, so I refreshed my series now for
> > > > > inclusion into the next merge window.
> > > > 
> > > > Hrrrm? That sounds like like Paul ran tests on EV5, did he?
> > > 
> > >  What exactly is required to make it work?
> > 
> > Whatever changes are needed to prevent the data corruption that can
> > currently result in code generated by single-byte stores.  For but one
> > example, consider a pair of tasks (or one task and an interrupt handler
> > in the CONFIG_SMP=n case) do a single-byte store to a pair of bytes
> > in the same machine word.  As I understand it, in code generated for
> > older Alphas, both "stores" will load the word containing that byte,
> > update their own byte, and store the updated word.
> > 
> > If two such single-byte stores run concurrently, one or the other of those
> > two stores will be lost, as in overwritten by the other.  This is a bug,
> > even in kernels built for single-CPU systems.  And a rare bug at that, one
> > that tends to disappear as you add debug code in an attempt to find it.
> 
>  Thank you for the detailed description of the problematic scenario.
> 
>  I hope someone will find it useful, however for the record I have been 
> familiar with the intricacies of the Alpha architecture as well as their 
> implications for software for decades now.  The Alpha port of Linux was 
> the first non-x86 Linux platform I have used and actually (and I've chased 
> that as a matter of interest) my first ever contribution to Linux was for 
> Alpha platform code:
> 
> On Mon, 30 Mar 1998, Jay.Estabrook@digital.com wrote:
> 
> > Hi, sorry about the delay in answering, but you'll be happy to know, I took
> > your patches and merged them into my latest SMP patches, and submitted them
> > to Linus just last night. He promises them to (mostly) be in 2.1.92, so we
> > can look forward to that... :-)
> 
> so I find the scenario you have described more than obvious.

Glad that it helped.

>  Mind that the read-modify-write sequence that software does for sub-word 
> write accesses with original Alpha hardware is precisely what hardware 
> would have to do anyway and support for that was deliberately omitted by 
> the architecture designers from the ISA to give it performance advantages 
> quoted in the architecture manual.  The only difference here is that with 
> hardware read-modify-write operations atomicity for sub-word accesses is 
> guaranteed by the ISA, however for software read-modify-write it has to be 
> explictly coded using the usual load-locked/store-conditional sequence in 
> a loop.  I don't think it's a big deal really, it should be trivial to do 
> in the relevant accessors, along with the memory barriers that are needed 
> anyway for EV56+ and possibly other ports such as the MIPS one.

There shouldn't be any memory barriers required, and don't EV56+ have
single-byte loads and stores?

>  What I have been after actually is: can you point me at a piece of code 
> in our tree that will cause an issue with a non-BWX Alpha as described in 
> your scenario, so that I have a starting point?  Mind that I'm completely 
> new to RCU as I didn't have a need to research it before (though from a 
> skim over Documentation/RCU/rcu.rst I understand what its objective is).

See the uses of the fields of the current->rcu_read_unlock_special.b
anonymous structure for the example that led us here.  And who knows how
many other pieces of the Linux kernel that assume that it is possible
to atomically store a single byte.

Many of which use a normal C-language store, in which case there are
no accessors.  This can be a problem even in the case where there are no
data races to either byte, because the need for the read-modify-write
sequence on older Alpha systems results in implicit data races at the
machine-word level.

>  FWIW even if it was only me I think that depriving the already thin Alpha 
> port developer base of any quantity of the remaining manpower, by dropping 
> support for a subset of the hardware available, and then a subset that is 
> not just as exotic as the original i386 became to the x86 platform at the 
> time support for it was dropped, is only going to lead to further demise 
> and eventual drop of the entire port.

Yes, support has been dropped for some of the older x86 CPUs as well,
for example, Linux-kernel support for multiprocessor 80386 systems was
dropped a great many years ago, in part because those CPUs do not have
a cmpxchg instruction.  So it is not like we are picking on Alpha.

>  And I think it would be good if we kept the port, just as we keep other 
> ports of historical significance only, for educational reasons if nothing 
> else, such as to let people understand based on an actual example, once 
> mainstream, the implications of weakly ordered memory systems.

I don't know of any remaining issues with the newer Alpha systems that do
support single-byte and double-byte load and store instructions, and so
I am not aware of any reason for dropping Linux-kernel support for them.

							Thanx, Paul

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-29 18:50       ` Maciej W. Rozycki
  2024-05-29 22:09         ` Paul E. McKenney
@ 2024-05-30  1:08         ` Linus Torvalds
  2024-05-30 22:57           ` Maciej W. Rozycki
  2024-05-31 15:48         ` Arnd Bergmann
  2 siblings, 1 reply; 55+ messages in thread
From: Linus Torvalds @ 2024-05-30  1:08 UTC (permalink / raw)
  To: Maciej W. Rozycki
  Cc: Paul E. McKenney, John Paul Adrian Glaubitz, Arnd Bergmann,
	linux-alpha, Arnd Bergmann, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Wed, 29 May 2024 at 11:50, Maciej W. Rozycki <macro@orcam.me.uk> wrote:
>
>              The only difference here is that with
> hardware read-modify-write operations atomicity for sub-word accesses is
> guaranteed by the ISA, however for software read-modify-write it has to be
> explictly coded using the usual load-locked/store-conditional sequence in
> a loop.

I have some bad news for you: the old alpha CPU's not only screwed up
the byte/word design, they _also_ screwed up the
load-locked/store-conditional.

You'd think that LL/SC would be done at a cacheline level, like any
sane person would do.

But no.

The 21064 actually did atomicity with an external pin on the bus, the
same way people used to do before caches even existed.

Yes, it has an internal L1 D$, but it is a write-through cache, and
clearly things like cache coherency weren't designed for. In fact,
LL/SC is even documented to not work in the external L2 cache
("Bcache" - don't ask me why the odd naming).

So LL/SC on the 21064 literally works on external memory.

Quoting the reference manual:

  "A.6 Load Locked and Store Conditional
  The 21064 provides the ability to perform locked memory accesses through
  the LDxL (Load_Locked) and STxC (Store_Conditional) cycle command pair.
  The LDxL command forces the 21064 to bypass the Bcache and request data
  directly from the external memory interface. The memory interface logic must
  set a special interlock flag as it returns the data, and may
optionally keep the
  locked address"

End result: a LL/SC pair is very very slow. It was incredibly slow
even for the time. I had benchmarks, I can't recall them, but I'd like
to say "hundreds of cycles". Maybe thousands.

So actual reliable byte operations are not realistically possible on
the early alpha CPU's. You can do them with LL/SC, sure, but
performance would be so horrendously bad that it would be just sad.

The 21064A had some "fast lock" mode which allows the data from the
LDQ_L to come from the Bcache. So it still isn't exactly fast, and it
still didn't work at CPU core speeds, but at least it worked with the
external cache.

Compilers will generate the sequence that DEC specified, which isn't
thread-safe.

In fact, it's worse than "not thread safe". It's not even safe on UP
with interrupts, or even signals in user space.

It's one of those "technically valid POSIX", since there's
"sig_atomic_t" and if you do any concurrent signal stuff you're
supposed to only use that type. But it's another of those "Yeah, you'd
better make sure your structure members are either 'int' or bigger, or
never accessed from signals or interrupts, or they might clobber
nearby values".

           Linus

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-30  1:08         ` Linus Torvalds
@ 2024-05-30 22:57           ` Maciej W. Rozycki
  2024-05-31  0:10             ` Linus Torvalds
  0 siblings, 1 reply; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-05-30 22:57 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Paul E. McKenney, John Paul Adrian Glaubitz, Arnd Bergmann,
	linux-alpha, Arnd Bergmann, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Wed, 29 May 2024, Linus Torvalds wrote:

> >              The only difference here is that with
> > hardware read-modify-write operations atomicity for sub-word accesses is
> > guaranteed by the ISA, however for software read-modify-write it has to be
> > explictly coded using the usual load-locked/store-conditional sequence in
> > a loop.
> 
> I have some bad news for you: the old alpha CPU's not only screwed up
> the byte/word design, they _also_ screwed up the
> load-locked/store-conditional.
> 
> You'd think that LL/SC would be done at a cacheline level, like any
> sane person would do.
> 
> But no.
> 
> The 21064 actually did atomicity with an external pin on the bus, the
> same way people used to do before caches even existed.

 Umm, 8086's LOCK#, anyone?

> Yes, it has an internal L1 D$, but it is a write-through cache, and
> clearly things like cache coherency weren't designed for. In fact,
> LL/SC is even documented to not work in the external L2 cache
> ("Bcache" - don't ask me why the odd naming).

 Board cache, I suppose.

> So LL/SC on the 21064 literally works on external memory.
> 
> Quoting the reference manual:
> 
>   "A.6 Load Locked and Store Conditional
>   The 21064 provides the ability to perform locked memory accesses through
>   the LDxL (Load_Locked) and STxC (Store_Conditional) cycle command pair.
>   The LDxL command forces the 21064 to bypass the Bcache and request data
>   directly from the external memory interface. The memory interface logic must
>   set a special interlock flag as it returns the data, and may
> optionally keep the
>   locked address"
> 
> End result: a LL/SC pair is very very slow. It was incredibly slow
> even for the time. I had benchmarks, I can't recall them, but I'd like
> to say "hundreds of cycles". Maybe thousands.

 Interesting and disappointing, given how many years the Alpha designers 
had to learn from the MIPS R4000.  Which they borrowed from already after 
all and which they had first-hand experience with present onboard, from 
the R4000 DECstation systems built at their WSE facility.  Hmm, I wonder 
if there was patent avoidance involved.

> So actual reliable byte operations are not realistically possible on
> the early alpha CPU's. You can do them with LL/SC, sure, but
> performance would be so horrendously bad that it would be just sad.

 Hmm, performance with a 30 years old system?  Who cares!  It mattered 30 
years ago, maybe 25.  And the performance of a system that runs slowly is 
still infinitely better than one of a system that doesn't boot anymore, 
isn't it?

> The 21064A had some "fast lock" mode which allows the data from the
> LDQ_L to come from the Bcache. So it still isn't exactly fast, and it
> still didn't work at CPU core speeds, but at least it worked with the
> external cache.
> 
> Compilers will generate the sequence that DEC specified, which isn't
> thread-safe.
> 
> In fact, it's worse than "not thread safe". It's not even safe on UP
> with interrupts, or even signals in user space.

 Ouch, I find it a surprising oversight.  Come to think of it indeed the 
plain unlocked read-modify-write sequences are unsafe.  I don't suppose 
any old DECies are still around, but any idea how this was sorted in DEC's 
own commercial operating systems (DU and OVMS)?

 So this seems like something that needs to be sorted in the compiler, by 
always using a locked sequence for 8-bit and 16-bit writes with non-BWX 
targets.  I can surely do it myself, not a big deal, and I reckon such a 
change to GCC should be pretty compact and self-contained, as all the bits 
are already within `alpha_expand_mov_nobwx' anyway.

 I'm not sure if Richard will be happy to accept it, but it seems to me 
the right thing to do at this point and with that in place there should be 
no safety concern for RCU or anything with the old Alphas, with no effort 
at all on the Linux side as all the burden will be on the compiler.  We 
may want to probe for the associated compiler option though and bail out 
if unsupported.

 Will it be enough to keep Linux support at least until the next obstacle?

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-29 22:09         ` Paul E. McKenney
@ 2024-05-30 22:59           ` Maciej W. Rozycki
  2024-05-31  3:56           ` Maciej W. Rozycki
  1 sibling, 0 replies; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-05-30 22:59 UTC (permalink / raw)
  To: Paul E. McKenney
  Cc: John Paul Adrian Glaubitz, Arnd Bergmann, linux-alpha,
	Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, linux-kernel,
	Michael Cree, Frank Scheiner

On Wed, 29 May 2024, Paul E. McKenney wrote:

> >  What I have been after actually is: can you point me at a piece of code 
> > in our tree that will cause an issue with a non-BWX Alpha as described in 
> > your scenario, so that I have a starting point?  Mind that I'm completely 
> > new to RCU as I didn't have a need to research it before (though from a 
> > skim over Documentation/RCU/rcu.rst I understand what its objective is).
> 
> See the uses of the fields of the current->rcu_read_unlock_special.b
> anonymous structure for the example that led us here.  And who knows how
> many other pieces of the Linux kernel that assume that it is possible
> to atomically store a single byte.

 Thanks, that helps.

> Many of which use a normal C-language store, in which case there are
> no accessors.  This can be a problem even in the case where there are no
> data races to either byte, because the need for the read-modify-write
> sequence on older Alpha systems results in implicit data races at the
> machine-word level.

 Ack.

> >  FWIW even if it was only me I think that depriving the already thin Alpha 
> > port developer base of any quantity of the remaining manpower, by dropping 
> > support for a subset of the hardware available, and then a subset that is 
> > not just as exotic as the original i386 became to the x86 platform at the 
> > time support for it was dropped, is only going to lead to further demise 
> > and eventual drop of the entire port.
> 
> Yes, support has been dropped for some of the older x86 CPUs as well,
> for example, Linux-kernel support for multiprocessor 80386 systems was
> dropped a great many years ago, in part because those CPUs do not have
> a cmpxchg instruction.  So it is not like we are picking on Alpha.

 That's what I mentioned (and for the record i386 wasn't dropped for the 
lack of CMPXCHG, as we never supported i386 SMP, exceedingly rare, anyway, 
but for the lack of page-level write-protection in the kernel mode, which 
implied painful manual checks).  At the time our support for the i386 was 
dropped its population outside embedded use was minuscule and certainly 
compared to non-i386 x86 Linux user base.  And the supply of modern x86 
systems was not an issue either.

 Conversely no new Alpha systems are made and I suspect the ratio between 
BWX and non-BWX Alpha Linux users is not as high as between post-i386 x86 
and original i386 Linux users at the time of the drop.

> >  And I think it would be good if we kept the port, just as we keep other 
> > ports of historical significance only, for educational reasons if nothing 
> > else, such as to let people understand based on an actual example, once 
> > mainstream, the implications of weakly ordered memory systems.
> 
> I don't know of any remaining issues with the newer Alpha systems that do
> support single-byte and double-byte load and store instructions, and so
> I am not aware of any reason for dropping Linux-kernel support for them.

 Well, the lack of developers to maintain the port would be the reason I 
refer to.  If you let developers drop by preventing them from using their 
hardware to work on the port, then eventually we'll have none.

 Anyway it seems like an issue to be sorted in the compiler, transparently 
to RCU, so it shouldn't be a reason to drop support for non-BWX Alpha CPUs 
anymore.  See my reply to Linus in this thread.

 Thank you for your input, always appreciated.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-30 22:57           ` Maciej W. Rozycki
@ 2024-05-31  0:10             ` Linus Torvalds
  2024-06-03 11:09               ` Maciej W. Rozycki
  0 siblings, 1 reply; 55+ messages in thread
From: Linus Torvalds @ 2024-05-31  0:10 UTC (permalink / raw)
  To: Maciej W. Rozycki
  Cc: Paul E. McKenney, John Paul Adrian Glaubitz, Arnd Bergmann,
	linux-alpha, Arnd Bergmann, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Thu, 30 May 2024 at 15:57, Maciej W. Rozycki <macro@orcam.me.uk> wrote:
>
> On Wed, 29 May 2024, Linus Torvalds wrote:
> >
> > The 21064 actually did atomicity with an external pin on the bus, the
> > same way people used to do before caches even existed.
>
>  Umm, 8086's LOCK#, anyone?

Well, yes and no.

So yes, exactly like 8086 did before having caches.

But no, not like the alpha contemporary PPro that did have caches. The
PPro already did locked cycles in the caches.

Yes, the PPro still did have an external lock pin (and in fact current
much more modern x86 CPUs do too), but it's only used for locked IO
accesses or possibly cacheline crossing accesses.

So x86 has supported atomic accesses on IO - and it is very very slow,
to this day. So slow, and problematic, in fact, that Intel is only now
trying to remove it (look  up "split lock"

But the 21064 explicitly did not support locking on IO - and unaligned
LL/SC accesses obviously also did not work.

So I really feel the 21064 was broken.

It's probably related to the whole cache coherency being designed to
be external to the built-in caches - or even the Bcache. The caches
basically are write-through, and the weak memory ordering was designed
for allowing this horrible model.

> > In fact, it's worse than "not thread safe". It's not even safe on UP
> > with interrupts, or even signals in user space.
>
>  Ouch, I find it a surprising oversight.

The sad part is that it doesn't seem to have been an oversight. It
really was broken-as-designed.

Basically, the CPU was designed for single-threaded Spec benchmarks
and absolutely nothing else. Classic RISC where you recompile to fix
problems like the atomicity thing - "just use a 32-bit sig_atomic_t
and you're fine")

The original alpha architecture handbook makes a big deal of how
clever the lack of byte and word operations is. I also remember
reading an article by Dick Sites - one of the main designers - talking
a lot about how the lack of byte operations is great, and encourages
vectorizing byte accesses and doing string operations in whole words.

          Linus

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-29 22:09         ` Paul E. McKenney
  2024-05-30 22:59           ` Maciej W. Rozycki
@ 2024-05-31  3:56           ` Maciej W. Rozycki
  2024-05-31 19:33             ` Paul E. McKenney
  1 sibling, 1 reply; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-05-31  3:56 UTC (permalink / raw)
  To: Paul E. McKenney
  Cc: John Paul Adrian Glaubitz, Arnd Bergmann, linux-alpha,
	Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, linux-kernel,
	Michael Cree, Frank Scheiner

On Wed, 29 May 2024, Paul E. McKenney wrote:

> >  Mind that the read-modify-write sequence that software does for sub-word 
> > write accesses with original Alpha hardware is precisely what hardware 
> > would have to do anyway and support for that was deliberately omitted by 
> > the architecture designers from the ISA to give it performance advantages 
> > quoted in the architecture manual.  The only difference here is that with 
> > hardware read-modify-write operations atomicity for sub-word accesses is 
> > guaranteed by the ISA, however for software read-modify-write it has to be 
> > explictly coded using the usual load-locked/store-conditional sequence in 
> > a loop.  I don't think it's a big deal really, it should be trivial to do 
> > in the relevant accessors, along with the memory barriers that are needed 
> > anyway for EV56+ and possibly other ports such as the MIPS one.
> 
> There shouldn't be any memory barriers required, and don't EV56+ have
> single-byte loads and stores?

 I should have commented on this in my original reply.

 You're the RCU expert so you know the answer.  I don't.  If it's OK for
successive writes to get reordered, or readers to see a stale value, then 
you don't need memory barriers.  Otherwise you do.  Whether byte accesses 
are available or not does not matter, the CPU *will* do reordering if it's 
allowed to (or more specifically, it won't do anything to prevent it from 
happening, especially in SMP configurations; I can't remember offhand if 
there are cases with UP).  Also adjacent byte writes may be merged, but I 
suppose it does not matter, or does it?

 NB MIPS has similar architectural arrangements (and a bunch of barriers 
defined in the ISA), it's just most implementations are actually strongly 
ordered, so most people can't see the effects of this.  With MIPS I know 
for sure there are cases of UP reordering, but they only really matter for 
MMIO use cases and not regular memory.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-29 18:50       ` Maciej W. Rozycki
  2024-05-29 22:09         ` Paul E. McKenney
  2024-05-30  1:08         ` Linus Torvalds
@ 2024-05-31 15:48         ` Arnd Bergmann
  2024-05-31 16:32           ` Linus Torvalds
  2024-06-03 11:33           ` Maciej W. Rozycki
  2 siblings, 2 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-31 15:48 UTC (permalink / raw)
  To: Maciej W. Rozycki, Paul E. McKenney
  Cc: John Paul Adrian Glaubitz, Arnd Bergmann, linux-alpha,
	Richard Henderson, Ivan Kokshaysky, Matt Turner, Alexander Viro,
	Marc Zyngier, Linus Torvalds, linux-kernel, Michael Cree,
	Frank Scheiner

On Wed, May 29, 2024, at 20:50, Maciej W. Rozycki wrote:
> On Tue, 28 May 2024, Paul E. McKenney wrote:

>  What I have been after actually is: can you point me at a piece of code 
> in our tree that will cause an issue with a non-BWX Alpha as described in 
> your scenario, so that I have a starting point?  Mind that I'm completely 
> new to RCU as I didn't have a need to research it before (though from a 
> skim over Documentation/RCU/rcu.rst I understand what its objective is).

I tried to look for examples and started with users of WRITE_ONCE()
on small variables by only allowing it to be used on types that
can be written natively:

--- a/include/linux/compiler_types.h
+++ b/include/linux/compiler_types.h
@@ -451,8 +451,7 @@ struct ftrace_likely_data {
 
 /* Is this type a native word size -- useful for atomic operations */
 #define __native_word(t) \
-       (sizeof(t) == sizeof(char) || sizeof(t) == sizeof(short) || \
-        sizeof(t) == sizeof(int) || sizeof(t) == sizeof(long))
+       (sizeof(t) == sizeof(int) || sizeof(t) == sizeof(long))
 
 #ifdef __OPTIMIZE__
 # define __compiletime_assert(condition, msg, prefix, suffix)          \

The WRITE_ONCE() calls tend to be there in order to avoid
expensive atomic or locking when something can be expressed
with a store that known to be visible atomically (on all other
architectures).

I then tried changing the underlying variables to 32-bit ones
to see how many changes are needed, but I gave up after around
150 of them, as I was only scratching the surface. To do this
right, you'd need to go through each one of them and come up
with a solution that is the best trade-off in terms of memory
usage and performance for that one. There are of course
others that should be using WRITE_ONCE() and are missing
this, so the list is not complete either. See below for
the ones I could find quickly.

>  FWIW even if it was only me I think that depriving the already thin Alpha 
> port developer base of any quantity of the remaining manpower, by dropping 
> support for a subset of the hardware available, and then a subset that is 
> not just as exotic as the original i386 became to the x86 platform at the 
> time support for it was dropped, is only going to lead to further demise 
> and eventual drop of the entire port.

I know you like you museum pieces to be older than everyone
else's, and I'm sorry that my patch series is causing you
problems, but I don't think the more general criticism is
valid here. My hope was mainly to help our with both keeping
Alpha viable for a few more years while also allowing Paul
to continue with his RCU changes.

As far as I can tell, nobody else is actually using EV4
machines or has been for years now, but the presence of that
code did affect both the performance and correctness of the
kernel code for all EV56+ users since distros have no way
of picking the ISA level on alpha for a generic kernel.

As I wrote in my patch notes, Debian already changed their
userspace to be built for EV56 or higher, as they had
determined that this was a significant improvement for
their users, so there is no binary distro left to ship
with a modern kernel.

Matt still maintains the Gentoo port (in addition to alpha
kernel), which seems to still support EV4, but all eight
of his machines on https://mattst88.com/computers/ are
EV56 or higher.

The strongest argument I see for assuming non-BWX alphas
are long dead is that gcc-4.4 added support for C11 style
_Atomic variables for alpha, but got the stores wrong
without anyone ever noticing the problem. Even one makes
the argument that normal byte stores and volatiles ones
should not need atomic ll/st sequenes, the atomics
clearly do. Building BWX-enabled kernels and userland
completely avoids this problem, which make debugging
easier for the remaining users when stuff breaks.

   Arnd

----
below: partial patch to change types for WRITE_ONCE() variables,

diff --git a/include/linux/compiler_types.h b/include/linux/compiler_types.h
index 5d1779759c21..11f1368808fe 100644
--- a/include/linux/compiler_types.h
+++ b/include/linux/compiler_types.h
@@ -451,8 +451,7 @@ struct ftrace_likely_data {
 
 /* Is this type a native word size -- useful for atomic operations */
 #define __native_word(t) \
-	(sizeof(t) == sizeof(char) || sizeof(t) == sizeof(short) || \
-	 sizeof(t) == sizeof(int) || sizeof(t) == sizeof(long))
+	(sizeof(t) == sizeof(int) || sizeof(t) == sizeof(long))
 
 #ifdef __OPTIMIZE__
 # define __compiletime_assert(condition, msg, prefix, suffix)		\
diff --git a/block/blk-crypto-fallback.c b/block/blk-crypto-fallback.c
index b1e7415f8439..86402997df21 100644
--- a/block/blk-crypto-fallback.c
+++ b/block/blk-crypto-fallback.c
@@ -71,7 +71,7 @@ static mempool_t *bio_fallback_crypt_ctx_pool;
  * be used at a time - the rest of the unused tfms have their keys cleared.
  */
 static DEFINE_MUTEX(tfms_init_lock);
-static bool tfms_inited[BLK_ENCRYPTION_MODE_MAX];
+static int tfms_inited[BLK_ENCRYPTION_MODE_MAX];
 
 static struct blk_crypto_fallback_keyslot {
 	enum blk_crypto_mode_num crypto_mode;diff --git a/include/linux/hrtimer_types.h b/include/linux/hrtimer_types.h
index ad66a3081735..4e675dc1ea29 100644
--- a/include/linux/hrtimer_types.h
+++ b/include/linux/hrtimer_types.h
@@ -41,10 +41,10 @@ struct hrtimer {
 	ktime_t				_softexpires;
 	enum hrtimer_restart		(*function)(struct hrtimer *);
 	struct hrtimer_clock_base	*base;
-	u8				state;
-	u8				is_rel;
-	u8				is_soft;
-	u8				is_hard;
+	u32				state;
+	u32				is_rel;
+	u32				is_soft;
+	u32				is_hard;
 };
 
 #endif /* _LINUX_HRTIMER_TYPES_H */
diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h
index 91224bbcfa73..11d2b5da3982 100644
--- a/include/linux/io_uring_types.h
+++ b/include/linux/io_uring_types.h
@@ -411,8 +411,8 @@ struct io_ring_ctx {
 
 	/* napi busy poll default timeout */
 	unsigned int		napi_busy_poll_to;
-	bool			napi_prefer_busy_poll;
-	bool			napi_enabled;
+	int			napi_prefer_busy_poll;
+	int			napi_enabled;
 
 	DECLARE_HASHTABLE(napi_ht, 4);
 #endif
@@ -605,7 +605,7 @@ struct io_kiocb {
 
 	u8				opcode;
 	/* polled IO has completed */
-	u8				iopoll_completed;
+	u32				iopoll_completed;
 	/*
 	 * Can be either a fixed buffer index, or used with provided buffers.
 	 * For the latter, before issue it points to the buffer group ID,
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 383a0ea2ab91..0f72f411b520 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -254,7 +254,7 @@ struct ipv6_pinfo {
 						 * 010: prefer public address
 						 * 100: prefer care-of address
 						 */
-	__u8			pmtudisc;
+	__u32			pmtudisc;
 	__u8			min_hopcount;
 	__u8			tclass;
 	__be32			rcv_flowinfo;
diff --git a/include/linux/key.h b/include/linux/key.h
index 943a432da3ae..8809e797417e 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -218,7 +218,7 @@ struct key {
 						 * - may not match RCU dereferenced payload
 						 * - payload should contain own length
 						 */
-	short			state;		/* Key state (+) or rejection error (-) */
+	int			state;		/* Key state (+) or rejection error (-) */
 
 #ifdef KEY_DEBUGGING
 	unsigned		magic;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 546de9cf46df..a7ef1e3aa9d0 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2047,7 +2047,7 @@ struct net_device {
 	 * and to use WRITE_ONCE() to annotate the writes.
 	 */
 	unsigned int		mtu;
-	unsigned short		needed_headroom;
+	unsigned int		needed_headroom;
 	struct netdev_tc_txq	tc_to_txq[TC_MAX_QUEUE];
 #ifdef CONFIG_XPS
 	struct xps_dev_maps __rcu *xps_maps[XPS_MAPS_MAX];
@@ -2298,7 +2298,7 @@ struct net_device {
 
 	struct list_head	link_watch_list;
 
-	u8 reg_state;
+	u32 reg_state;
 
 	bool dismantle;
 
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 61591ac6eab6..5eadcdfcf089 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -719,10 +719,10 @@ struct uclamp_se {
 
 union rcu_special {
 	struct {
-		u8			blocked;
-		u8			need_qs;
-		u8			exp_hint; /* Hint for performance. */
-		u8			need_mb; /* Readers need smp_mb(). */
+		u32			blocked;
+		u32			need_qs;
+		u32			exp_hint; /* Hint for performance. */
+		u32			need_mb; /* Readers need smp_mb(). */
 	} b; /* Bits. */
 	u32 s; /* Set of bits. */
 };
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 6a5e08b937b3..8dafb3e49546 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -125,7 +125,7 @@ struct tcp_options_received {
 	u8	saw_unknown:1,	/* Received unknown option		*/
 		unused:7;
 	u8	num_sacks;	/* Number of SACK blocks		*/
-	u16	user_mss;	/* mss requested by user in ioctl	*/
+	u32	user_mss;	/* mss requested by user in ioctl	*/
 	u16	mss_clamp;	/* Maximal mss, negotiated at connection setup */
 };
 
@@ -237,8 +237,8 @@ struct tcp_sock {
 	u32	tlp_high_seq;	/* snd_nxt at the time of TLP */
 	u32	rttvar_us;	/* smoothed mdev_max			*/
 	u32	retrans_out;	/* Retransmitted packets out		*/
-	u16	advmss;		/* Advertised MSS			*/
-	u16	urg_data;	/* Saved octet of OOB data and control flags */
+	u32	advmss;		/* Advertised MSS			*/
+	u32	urg_data;	/* Saved octet of OOB data and control flags */
 	u32	lost;		/* Total data packets lost incl. rexmits */
 	struct  minmax rtt_min;
 	/* OOO segments go in this rbtree. Socket lock must be held. */
@@ -383,7 +383,7 @@ struct tcp_sock {
 		syn_fastopen_ch:1, /* Active TFO re-enabling probe */
 		syn_data_acked:1;/* data in SYN is acked by SYN-ACK */
 
-	u8	keepalive_probes; /* num of allowed keep alive probes	*/
+	u32	keepalive_probes; /* num of allowed keep alive probes	*/
 	u32	tcp_tx_delay;	/* delay (in usec) added to TX packets */
 
 /* RTT measurement */
diff --git a/include/linux/udp.h b/include/linux/udp.h
index 3eb3f2b9a2a0..2bba12ee545e 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -54,19 +54,19 @@ struct udp_sock {
 	unsigned long	 udp_flags;
 
 	int		 pending;	/* Any pending frames ? */
-	__u8		 encap_type;	/* Is this an Encapsulation socket? */
+	__u32		 encap_type;	/* Is this an Encapsulation socket? */
 
 	/*
 	 * Following member retains the information to create a UDP header
 	 * when the socket is uncorked.
 	 */
 	__u16		 len;		/* total length of pending frames */
-	__u16		 gso_size;
+	__u32		 gso_size;
 	/*
 	 * Fields specific to UDP-Lite.
 	 */
-	__u16		 pcslen;
-	__u16		 pcrlen;
+	__u32		 pcslen;
+	__u32		 pcrlen;
 	/*
 	 * For encapsulation sockets.
 	 */
@@ -94,7 +94,7 @@ struct udp_sock {
 	int		forward_threshold;
 
 	/* Cache friendly copy of sk->sk_peek_off >= 0 */
-	bool		peeking_with_offset;
+	int		peeking_with_offset;
 };
 
 #define udp_test_bit(nr, sk)			\
diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index 7d6b1254c92d..b549a39360ec 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -105,11 +105,11 @@ struct inet_connection_sock {
 	__u8			  icsk_retransmits;
 	__u8			  icsk_pending;
 	__u8			  icsk_backoff;
-	__u8			  icsk_syn_retries;
+	__u32			  icsk_syn_retries;
 	__u8			  icsk_probes_out;
 	__u16			  icsk_ext_hdr_len;
 	struct {
-		__u8		  pending;	 /* ACK is pending			   */
+		__u32		  pending;	 /* ACK is pending			   */
 		__u8		  quick;	 /* Scheduled number of quick acks	   */
 		__u8		  pingpong;	 /* The session is interactive		   */
 		__u8		  retry;	 /* Number of attempts			   */
@@ -120,7 +120,7 @@ struct inet_connection_sock {
 		unsigned long	  timeout;	 /* Currently scheduled timeout		   */
 		__u32		  lrcvtime;	 /* timestamp of last received data packet */
 		__u16		  last_seg_size; /* Size of last incoming segment	   */
-		__u16		  rcv_mss;	 /* MSS used for delayed ACK decisions	   */
+		__u32		  rcv_mss;	 /* MSS used for delayed ACK decisions	   */
 	} icsk_ack;
 	struct {
 		/* Range of MTUs to search */
diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index 5af6eb14c5db..5d368156e437 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -18,7 +18,7 @@ struct fqdir {
 	int			max_dist;
 	struct inet_frags	*f;
 	struct net		*net;
-	bool			dead;
+	int			dead;
 
 	struct rhashtable       rhashtable ____cacheline_aligned_in_smp;
 
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index f9ddd47dc4f8..98f2c745a34e 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -220,17 +220,17 @@ struct inet_sock {
 
 	unsigned long		inet_flags;
 	__be32			inet_saddr;
-	__s16			uc_ttl;
+	__s32			uc_ttl;
 	__be16			inet_sport;
 	struct ip_options_rcu __rcu	*inet_opt;
 	atomic_t		inet_id;
 
-	__u8			tos;
-	__u8			min_ttl;
-	__u8			mc_ttl;
-	__u8			pmtudisc;
-	__u8			rcv_tos;
-	__u8			convert_csum;
+	__u32			tos;
+	__u32			min_ttl;
+	__u32			mc_ttl;
+	__u32			pmtudisc;
+	__u32			rcv_tos;
+	__u32			convert_csum;
 	int			uc_index;
 	int			mc_index;
 	__be32			mc_addr;
diff --git a/include/net/ip.h b/include/net/ip.h
index 6d735e00d3f3..e0cb7e0bfec9 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -81,7 +81,7 @@ struct ipcm_cookie {
 	__u8			ttl;
 	__s16			tos;
 	char			priority;
-	__u16			gso_size;
+	__u32			gso_size;
 };
 
 static inline void ipcm_init(struct ipcm_cookie *ipcm)
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 9b2f69ba5e49..182b7eade5c0 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -139,7 +139,7 @@ struct fib_info {
 	refcount_t		fib_treeref;
 	refcount_t		fib_clntref;
 	unsigned int		fib_flags;
-	unsigned char		fib_dead;
+	u32			fib_dead;
 	unsigned char		fib_protocol;
 	unsigned char		fib_scope;
 	unsigned char		fib_type;
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 88a8e554f7a1..4732c3084e10 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -358,7 +358,7 @@ struct ipcm6_cookie {
 	struct sockcm_cookie sockc;
 	__s16 hlimit;
 	__s16 tclass;
-	__u16 gso_size;
+	__u32 gso_size;
 	__s8  dontfrag;
 	struct ipv6_txoptions *opt;
 };
diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index 0d28172193fa..e2f580880977 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -147,10 +147,10 @@ struct neighbour {
 	struct timer_list	timer;
 	unsigned long		used;
 	atomic_t		probes;
-	u8			nud_state;
-	u8			type;
-	u8			dead;
-	u8			protocol;
+	u32			nud_state;
+	u32			type;
+	u32			dead;
+	u32			protocol;
 	u32			flags;
 	seqlock_t		ha_lock;
 	unsigned char		ha[ALIGN(MAX_ADDR_LEN, sizeof(unsigned long))] __aligned(8);
diff --git a/include/net/netns/core.h b/include/net/netns/core.h
index 78214f1b43a2..46d9b3966c5b 100644
--- a/include/net/netns/core.h
+++ b/include/net/netns/core.h
@@ -14,7 +14,7 @@ struct netns_core {
 
 	int	sysctl_somaxconn;
 	int	sysctl_optmem_max;
-	u8	sysctl_txrehash;
+	u32	sysctl_txrehash;
 
 #ifdef CONFIG_PROC_FS
 	struct prot_inuse __percpu *prot_inuse;
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index c356c458b340..3df67aa03c3c 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -48,27 +48,27 @@ struct netns_ipv4 {
 
 	/* TX readonly hotpath cache lines */
 	__cacheline_group_begin(netns_ipv4_read_tx);
-	u8 sysctl_tcp_early_retrans;
-	u8 sysctl_tcp_tso_win_divisor;
-	u8 sysctl_tcp_tso_rtt_log;
-	u8 sysctl_tcp_autocorking;
+	u32 sysctl_tcp_early_retrans;
+	u32 sysctl_tcp_tso_win_divisor;
+	u32 sysctl_tcp_tso_rtt_log;
+	u32 sysctl_tcp_autocorking;
 	int sysctl_tcp_min_snd_mss;
 	unsigned int sysctl_tcp_notsent_lowat;
 	int sysctl_tcp_limit_output_bytes;
 	int sysctl_tcp_min_rtt_wlen;
 	int sysctl_tcp_wmem[3];
-	u8 sysctl_ip_fwd_use_pmtu;
+	u32 sysctl_ip_fwd_use_pmtu;
 	__cacheline_group_end(netns_ipv4_read_tx);
 
 	/* TXRX readonly hotpath cache lines */
 	__cacheline_group_begin(netns_ipv4_read_txrx);
-	u8 sysctl_tcp_moderate_rcvbuf;
+	u32 sysctl_tcp_moderate_rcvbuf;
 	__cacheline_group_end(netns_ipv4_read_txrx);
 
 	/* RX readonly hotpath cache line */
 	__cacheline_group_begin(netns_ipv4_read_rx);
-	u8 sysctl_ip_early_demux;
-	u8 sysctl_tcp_early_demux;
+	u32 sysctl_ip_early_demux;
+	u32 sysctl_tcp_early_demux;
 	int sysctl_tcp_reordering;
 	int sysctl_tcp_rmem[3];
 	__cacheline_group_end(netns_ipv4_read_rx);
@@ -96,7 +96,7 @@ struct netns_ipv4 {
 #endif
 	bool			fib_has_custom_local_routes;
 	bool			fib_offload_disabled;
-	u8			sysctl_tcp_shrink_window;
+	u32			sysctl_tcp_shrink_window;
 #ifdef CONFIG_IP_ROUTE_CLASSID
 	atomic_t		fib_num_tclassid_users;
 #endif
@@ -108,11 +108,11 @@ struct netns_ipv4 {
 	struct inet_peer_base	*peers;
 	struct fqdir		*fqdir;
 
-	u8 sysctl_icmp_echo_ignore_all;
-	u8 sysctl_icmp_echo_enable_probe;
-	u8 sysctl_icmp_echo_ignore_broadcasts;
-	u8 sysctl_icmp_ignore_bogus_error_responses;
-	u8 sysctl_icmp_errors_use_inbound_ifaddr;
+	u32 sysctl_icmp_echo_ignore_all;
+	u32 sysctl_icmp_echo_enable_probe;
+	u32 sysctl_icmp_echo_ignore_broadcasts;
+	u32 sysctl_icmp_ignore_bogus_error_responses;
+	u32 sysctl_icmp_errors_use_inbound_ifaddr;
 	int sysctl_icmp_ratelimit;
 	int sysctl_icmp_ratemask;
 
@@ -122,29 +122,29 @@ struct netns_ipv4 {
 
 	struct local_ports ip_local_ports;
 
-	u8 sysctl_tcp_ecn;
-	u8 sysctl_tcp_ecn_fallback;
+	u32 sysctl_tcp_ecn;
+	u32 sysctl_tcp_ecn_fallback;
 
-	u8 sysctl_ip_default_ttl;
-	u8 sysctl_ip_no_pmtu_disc;
-	u8 sysctl_ip_fwd_update_priority;
-	u8 sysctl_ip_nonlocal_bind;
-	u8 sysctl_ip_autobind_reuse;
+	u32 sysctl_ip_default_ttl;
+	u32 sysctl_ip_no_pmtu_disc;
+	u32 sysctl_ip_fwd_update_priority;
+	u32 sysctl_ip_nonlocal_bind;
+	u32 sysctl_ip_autobind_reuse;
 	/* Shall we try to damage output packets if routing dev changes? */
-	u8 sysctl_ip_dynaddr;
+	u32 sysctl_ip_dynaddr;
 #ifdef CONFIG_NET_L3_MASTER_DEV
-	u8 sysctl_raw_l3mdev_accept;
+	u32 sysctl_raw_l3mdev_accept;
 #endif
-	u8 sysctl_udp_early_demux;
+	u32 sysctl_udp_early_demux;
 
-	u8 sysctl_nexthop_compat_mode;
+	u32 sysctl_nexthop_compat_mode;
 
-	u8 sysctl_fwmark_reflect;
-	u8 sysctl_tcp_fwmark_accept;
+	u32 sysctl_fwmark_reflect;
+	u32 sysctl_tcp_fwmark_accept;
 #ifdef CONFIG_NET_L3_MASTER_DEV
-	u8 sysctl_tcp_l3mdev_accept;
+	u32 sysctl_tcp_l3mdev_accept;
 #endif
-	u8 sysctl_tcp_mtu_probing;
+	u32 sysctl_tcp_mtu_probing;
 	int sysctl_tcp_mtu_probe_floor;
 	int sysctl_tcp_base_mss;
 	int sysctl_tcp_probe_threshold;
@@ -152,43 +152,43 @@ struct netns_ipv4 {
 
 	int sysctl_tcp_keepalive_time;
 	int sysctl_tcp_keepalive_intvl;
-	u8 sysctl_tcp_keepalive_probes;
-
-	u8 sysctl_tcp_syn_retries;
-	u8 sysctl_tcp_synack_retries;
-	u8 sysctl_tcp_syncookies;
-	u8 sysctl_tcp_migrate_req;
-	u8 sysctl_tcp_comp_sack_nr;
-	u8 sysctl_tcp_backlog_ack_defer;
-	u8 sysctl_tcp_pingpong_thresh;
-
-	u8 sysctl_tcp_retries1;
-	u8 sysctl_tcp_retries2;
-	u8 sysctl_tcp_orphan_retries;
-	u8 sysctl_tcp_tw_reuse;
+	u32 sysctl_tcp_keepalive_probes;
+
+	u32 sysctl_tcp_syn_retries;
+	u32 sysctl_tcp_synack_retries;
+	u32 sysctl_tcp_syncookies;
+	u32 sysctl_tcp_migrate_req;
+	u32 sysctl_tcp_comp_sack_nr;
+	u32 sysctl_tcp_backlog_ack_defer;
+	u32 sysctl_tcp_pingpong_thresh;
+
+	u32 sysctl_tcp_retries1;
+	u32 sysctl_tcp_retries2;
+	u32 sysctl_tcp_orphan_retries;
+	u32 sysctl_tcp_tw_reuse;
 	int sysctl_tcp_fin_timeout;
-	u8 sysctl_tcp_sack;
-	u8 sysctl_tcp_window_scaling;
-	u8 sysctl_tcp_timestamps;
-	u8 sysctl_tcp_recovery;
-	u8 sysctl_tcp_thin_linear_timeouts;
-	u8 sysctl_tcp_slow_start_after_idle;
-	u8 sysctl_tcp_retrans_collapse;
-	u8 sysctl_tcp_stdurg;
-	u8 sysctl_tcp_rfc1337;
-	u8 sysctl_tcp_abort_on_overflow;
-	u8 sysctl_tcp_fack; /* obsolete */
+	u32 sysctl_tcp_sack;
+	u32 sysctl_tcp_window_scaling;
+	u32 sysctl_tcp_timestamps;
+	u32 sysctl_tcp_recovery;
+	u32 sysctl_tcp_thin_linear_timeouts;
+	u32 sysctl_tcp_slow_start_after_idle;
+	u32 sysctl_tcp_retrans_collapse;
+	u32 sysctl_tcp_stdurg;
+	u32 sysctl_tcp_rfc1337;
+	u32 sysctl_tcp_abort_on_overflow;
+	u32 sysctl_tcp_fack; /* obsolete */
 	int sysctl_tcp_max_reordering;
 	int sysctl_tcp_adv_win_scale; /* obsolete */
-	u8 sysctl_tcp_dsack;
-	u8 sysctl_tcp_app_win;
-	u8 sysctl_tcp_frto;
-	u8 sysctl_tcp_nometrics_save;
-	u8 sysctl_tcp_no_ssthresh_metrics_save;
-	u8 sysctl_tcp_workaround_signed_windows;
+	u32 sysctl_tcp_dsack;
+	u32 sysctl_tcp_app_win;
+	u32 sysctl_tcp_frto;
+	u32 sysctl_tcp_nometrics_save;
+	u32 sysctl_tcp_no_ssthresh_metrics_save;
+	u32 sysctl_tcp_workaround_signed_windows;
 	int sysctl_tcp_challenge_ack_limit;
-	u8 sysctl_tcp_min_tso_segs;
-	u8 sysctl_tcp_reflect_tos;
+	u32 sysctl_tcp_min_tso_segs;
+	u32 sysctl_tcp_reflect_tos;
 	int sysctl_tcp_invalid_ratelimit;
 	int sysctl_tcp_pacing_ss_ratio;
 	int sysctl_tcp_pacing_ca_ratio;
@@ -204,23 +204,23 @@ struct netns_ipv4 {
 	unsigned long tfo_active_disable_stamp;
 	u32 tcp_challenge_timestamp;
 	u32 tcp_challenge_count;
-	u8 sysctl_tcp_plb_enabled;
-	u8 sysctl_tcp_plb_idle_rehash_rounds;
-	u8 sysctl_tcp_plb_rehash_rounds;
-	u8 sysctl_tcp_plb_suspend_rto_sec;
+	u32 sysctl_tcp_plb_enabled;
+	u32 sysctl_tcp_plb_idle_rehash_rounds;
+	u32 sysctl_tcp_plb_rehash_rounds;
+	u32 sysctl_tcp_plb_suspend_rto_sec;
 	int sysctl_tcp_plb_cong_thresh;
 
 	int sysctl_udp_wmem_min;
 	int sysctl_udp_rmem_min;
 
-	u8 sysctl_fib_notify_on_flag_change;
-	u8 sysctl_tcp_syn_linear_timeouts;
+	u32 sysctl_fib_notify_on_flag_change;
+	u32 sysctl_tcp_syn_linear_timeouts;
 
 #ifdef CONFIG_NET_L3_MASTER_DEV
-	u8 sysctl_udp_l3mdev_accept;
+	u32 sysctl_udp_l3mdev_accept;
 #endif
 
-	u8 sysctl_igmp_llm_reports;
+	u32 sysctl_igmp_llm_reports;
 	int sysctl_igmp_max_memberships;
 	int sysctl_igmp_max_msf;
 	int sysctl_igmp_qrv;
@@ -246,8 +246,8 @@ struct netns_ipv4 {
 #endif
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 	u32 sysctl_fib_multipath_hash_fields;
-	u8 sysctl_fib_multipath_use_neigh;
-	u8 sysctl_fib_multipath_hash_policy;
+	u32 sysctl_fib_multipath_use_neigh;
+	u32 sysctl_fib_multipath_hash_policy;
 #endif
 
 	struct fib_notifier_ops	*notifier_ops;
diff --git a/include/net/request_sock.h b/include/net/request_sock.h
index ebcb8896bffc..662804623036 100644
--- a/include/net/request_sock.h
+++ b/include/net/request_sock.h
@@ -221,7 +221,7 @@ struct fastopen_queue {
  */
 struct request_sock_queue {
 	spinlock_t		rskq_lock;
-	u8			rskq_defer_accept;
+	u32			rskq_defer_accept;
 
 	u32			synflood_warned;
 	atomic_t		qlen;
diff --git a/include/net/sock.h b/include/net/sock.h
index 5f4d0629348f..729fa16dd29f 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -168,8 +168,8 @@ struct sock_common {
 		};
 	};
 
-	unsigned short		skc_family;
-	volatile unsigned char	skc_state;
+	u32		skc_family;
+	volatile unsigned int	skc_state;
 	unsigned char		skc_reuse:4;
 	unsigned char		skc_reuseport:1;
 	unsigned char		skc_ipv6only:1;
@@ -210,9 +210,9 @@ struct sock_common {
 		struct hlist_node	skc_node;
 		struct hlist_nulls_node skc_nulls_node;
 	};
-	unsigned short		skc_tx_queue_mapping;
+	unsigned int		skc_tx_queue_mapping;
 #ifdef CONFIG_SOCK_RX_QUEUE_MAPPING
-	unsigned short		skc_rx_queue_mapping;
+	unsigned int		skc_rx_queue_mapping;
 #endif
 	union {
 		int		skc_incoming_cpu;
@@ -411,8 +411,8 @@ struct sock {
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	unsigned int		sk_ll_usec;
 	unsigned int		sk_napi_id;
-	u16			sk_busy_poll_budget;
-	u8			sk_prefer_busy_poll;
+	u32			sk_busy_poll_budget;
+	u32			sk_prefer_busy_poll;
 #endif
 	u8			sk_userlocks;
 	int			sk_rcvbuf;
@@ -486,7 +486,7 @@ struct sock {
 	unsigned int		sk_gso_max_size;
 	gfp_t			sk_allocation;
 	u32			sk_txhash;
-	u8			sk_pacing_shift;
+	u32			sk_pacing_shift;
 	bool			sk_use_task_frag;
 	__cacheline_group_end(sock_read_tx);
 
@@ -498,7 +498,7 @@ struct sock {
 				sk_kern_sock : 1,
 				sk_no_check_tx : 1,
 				sk_no_check_rx : 1;
-	u8			sk_shutdown;
+	u32			sk_shutdown;
 	u16			sk_type;
 	u16			sk_protocol;
 	unsigned long	        sk_lingertime;
@@ -519,7 +519,7 @@ struct sock {
 #endif
 	int			sk_disconnects;
 
-	u8			sk_txrehash;
+	u32			sk_txrehash;
 	u8			sk_clockid;
 	u8			sk_txtime_deadline_mode : 1,
 				sk_txtime_report_errors : 1,
diff --git a/include/net/sock_reuseport.h b/include/net/sock_reuseport.h
index 6ec140b0a61b..1c79170538f3 100644
--- a/include/net/sock_reuseport.h
+++ b/include/net/sock_reuseport.h
@@ -15,8 +15,8 @@ struct sock_reuseport {
 
 	u16			max_socks;		/* length of socks */
 	u16			num_socks;		/* elements in socks */
-	u16			num_closed_socks;	/* closed elements in socks */
-	u16			incoming_cpu;
+	u32			num_closed_socks;	/* closed elements in socks */
+	u32			incoming_cpu;
 	/* The last synq overflow event timestamp of this
 	 * reuse->socks[] group.
 	 */
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 060e95b331a2..4c068da5d085 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1744,7 +1744,7 @@ static inline void tcp_clear_all_retrans_hints(struct tcp_sock *tp)
 /* - key database */
 struct tcp_md5sig_key {
 	struct hlist_node	node;
-	u8			keylen;
+	u32			keylen;
 	u8			family; /* AF_INET or AF_INET6 */
 	u8			prefixlen;
 	u8			flags;
diff --git a/include/net/udp.h b/include/net/udp.h
index c4e05b14b648..2794e6b75f86 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -279,7 +279,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len);
 void udp_splice_eof(struct socket *sock);
 int udp_push_pending_frames(struct sock *sk);
 void udp_flush_pending_frames(struct sock *sk);
-int udp_cmsg_send(struct sock *sk, struct msghdr *msg, u16 *gso_size);
+int udp_cmsg_send(struct sock *sk, struct msghdr *msg, u32 *gso_size);
 void udp4_hwcsum(struct sk_buff *skb, __be32 src, __be32 dst);
 int udp_rcv(struct sk_buff *skb);
 int udp_ioctl(struct sock *sk, int cmd, int *karg);
diff --git a/include/sound/ump.h b/include/sound/ump.h
index 91238dabe307..c6ce23d2db00 100644
--- a/include/sound/ump.h
+++ b/include/sound/ump.h
@@ -26,7 +26,7 @@ struct snd_ump_endpoint {
 
 	/* UMP Stream message processing */
 	u32 stream_wait_for;	/* expected stream message status */
-	bool stream_finished;	/* set when message has been processed */
+	u32 stream_finished;	/* set when message has been processed */
 	bool parsed;		/* UMP / FB parse finished? */
 	bool no_process_stream;	/* suppress UMP stream messages handling */
 	wait_queue_head_t stream_wait;
diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
index 994bf7af0efe..647503be0999 100644
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -30,7 +30,7 @@ extern "C" {
 struct io_uring_sqe {
 	__u8	opcode;		/* type of operation for this sqe */
 	__u8	flags;		/* IOSQE_ flags */
-	__u16	ioprio;		/* ioprio for the request */
+	__u32	ioprio;		/* ioprio for the request */
 	__s32	fd;		/* file descriptor to do IO on */
 	union {
 		__u64	off;	/* offset into file */
@@ -78,9 +78,9 @@ struct io_uring_sqe {
 	/* pack this to avoid bogus arm OABI complaints */
 	union {
 		/* index into fixed buffers, if used */
-		__u16	buf_index;
+		__u32	buf_index;
 		/* for grouped buffer selection */
-		__u16	buf_group;
+		__u32	buf_group;
 	} __attribute__((packed));
 	/* personality to use, if used */
 	__u16	personality;
@@ -89,8 +89,8 @@ struct io_uring_sqe {
 		__u32	file_index;
 		__u32	optlen;
 		struct {
-			__u16	addr_len;
-			__u16	__pad3[1];
+			__u32	addr_len;
+			__u32	__pad3[1];
 		};
 	};
 	union {
diff --git a/io_uring/io_uring.h b/io_uring/io_uring.h
index 624ca9076a50..b14acb4de822 100644
--- a/io_uring/io_uring.h
+++ b/io_uring/io_uring.h
@@ -44,7 +44,7 @@ struct io_wait_queue {
 
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	unsigned int napi_busy_poll_to;
-	bool napi_prefer_busy_poll;
+	int napi_prefer_busy_poll;
 #endif
 };
 
diff --git a/io_uring/notif.h b/io_uring/notif.h
index f3589cfef4a9..e3f4f2462c4a 100644
--- a/io_uring/notif.h
+++ b/io_uring/notif.h
@@ -18,9 +18,9 @@ struct io_notif_data {
 	struct io_notif_data	*head;
 
 	unsigned		account_pages;
-	bool			zc_report;
-	bool			zc_used;
-	bool			zc_copied;
+	int			zc_report;
+	int			zc_used;
+	int			zc_copied;
 };
 
 struct io_kiocb *io_alloc_notif(struct io_ring_ctx *ctx);
diff --git a/kernel/locking/locktorture.c b/kernel/locking/locktorture.c
index de95ec07e477..1b485f4041a2 100644
--- a/kernel/locking/locktorture.c
+++ b/kernel/locking/locktorture.c
@@ -123,7 +123,7 @@ struct lock_stress_stats {
 
 struct call_rcu_chain {
 	struct rcu_head crc_rh;
-	bool crc_stop;
+	int crc_stop;
 };
 struct call_rcu_chain *call_rcu_chain_list;
 
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index cefa27f92bb6..1f690da4a6e4 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -872,10 +872,10 @@ struct root_domain {
 	 * - More than one runnable task
 	 * - Running task is misfit
 	 */
-	bool			overloaded;
+	int			overloaded;
 
 	/* Indicate one or more CPUs over-utilized (tipping point) */
-	bool			overutilized;
+	int			overutilized;
 
 	/*
 	 * The bit corresponding to a CPU gets set here if such CPU has more
diff --git a/lib/vsprintf.c b/lib/vsprintf.c
index cdd4e2314bfc..a79464cba03d 100644
--- a/lib/vsprintf.c
+++ b/lib/vsprintf.c
@@ -753,7 +753,7 @@ static int __init debug_boot_weak_hash_enable(char *str)
 }
 early_param("debug_boot_weak_hash", debug_boot_weak_hash_enable);
 
-static bool filled_random_ptr_key __read_mostly;
+static int filled_random_ptr_key __read_mostly;
 static siphash_key_t ptr_key __read_mostly;
 
 static int fill_ptr_key(struct notifier_block *nb, unsigned long action, void *data)
diff --git a/mm/swap_state.c b/mm/swap_state.c
index a5dae40523ab..cbf325c8afc4 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -40,7 +40,7 @@ static const struct address_space_operations swap_aops = {
 
 struct address_space *swapper_spaces[MAX_SWAPFILES] __read_mostly;
 static unsigned int nr_swapper_spaces[MAX_SWAPFILES] __read_mostly;
-static bool enable_vma_readahead __read_mostly = true;
+static int enable_vma_readahead __read_mostly = true;
 
 #define SWAP_RA_WIN_SHIFT	(PAGE_SHIFT / 2)
 #define SWAP_RA_HITS_MASK	((1UL << SWAP_RA_WIN_SHIFT) - 1)
diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h
index f9b9e26c32c1..5c5b69e5fb8b 100644
--- a/net/ipv4/fib_lookup.h
+++ b/net/ipv4/fib_lookup.h
@@ -17,9 +17,9 @@ struct fib_alias {
 	u8			fa_slen;
 	u32			tb_id;
 	s16			fa_default;
-	u8			offload;
-	u8			trap;
-	u8			offload_failed;
+	u32			offload;
+	u32			trap;
+	u32			offload_failed;
 	struct rcu_head		rcu;
 };
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 681b54e1f3a6..24ebb76dc7b5 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -4729,8 +4633,6 @@ void __init tcp_init(void)
 	BUILD_BUG_ON(sizeof(struct tcp_skb_cb) >
 		     sizeof_field(struct sk_buff, cb));
 
-	tcp_struct_check();
-
 	percpu_counter_init(&tcp_sockets_allocated, 0, GFP_KERNEL);
 
 	timer_setup(&tcp_orphan_timer, tcp_orphan_update, TIMER_DEFERRABLE);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 9c04a9c8be9d..7531e52c5e5d 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -568,7 +568,7 @@ static void tcp_init_buffer_space(struct sock *sk)
 
 		if (tcp_app_win && maxwin > 4 * tp->advmss)
 			WRITE_ONCE(tp->window_clamp,
-				   max(maxwin - (maxwin >> tcp_app_win),
+				   max_t(u32, maxwin - (maxwin >> tcp_app_win),
 				       4 * tp->advmss));
 	}
 
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 189c9113fe9a..47bf7dc201ab 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1015,7 +1015,7 @@ int udp_push_pending_frames(struct sock *sk)
 }
 EXPORT_SYMBOL(udp_push_pending_frames);
 
-static int __udp_cmsg_send(struct cmsghdr *cmsg, u16 *gso_size)
+static int __udp_cmsg_send(struct cmsghdr *cmsg, u32 *gso_size)
 {
 	switch (cmsg->cmsg_type) {
 	case UDP_SEGMENT:
@@ -1028,7 +1028,7 @@ static int __udp_cmsg_send(struct cmsghdr *cmsg, u16 *gso_size)
 	}
 }
 
-int udp_cmsg_send(struct sock *sk, struct msghdr *msg, u16 *gso_size)
+int udp_cmsg_send(struct sock *sk, struct msghdr *msg, u32 *gso_size)
 {
 	struct cmsghdr *cmsg;
 	bool need_ip = false;
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 7aa47e2dd52b..803ca22eadef 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -219,17 +219,17 @@ struct mptcp_pm_data {
 
 	spinlock_t	lock;		/*protects the whole PM data */
 
-	u8		addr_signal;
-	bool		server_side;
-	bool		work_pending;
-	bool		accept_addr;
-	bool		accept_subflow;
-	bool		remote_deny_join_id0;
+	u32		addr_signal;
+	u32		server_side;
+	u32		work_pending;
+	u32		accept_addr;
+	u32		accept_subflow;
+	u32		remote_deny_join_id0;
 	u8		add_addr_signaled;
 	u8		add_addr_accepted;
 	u8		local_addr_used;
-	u8		pm_type;
-	u8		subflows;
+	u32		pm_type;
+	u32		subflows;
 	u8		status;
 	DECLARE_BITMAP(id_avail_bitmap, MPTCP_PM_MAX_ADDR_ID + 1);
 	struct mptcp_rm_list rm_list_tx;
@@ -290,14 +290,14 @@ struct mptcp_sock {
 	unsigned long	flags;
 	unsigned long	cb_flags;
 	bool		recovery;		/* closing subflow write queue reinjected */
-	bool		can_ack;
-	bool		fully_established;
-	bool		rcv_data_fin;
-	bool		snd_data_fin_enable;
-	bool		rcv_fastclose;
-	bool		use_64bit_ack; /* Set when we received a 64-bit DSN */
-	bool		csum_enabled;
-	bool		allow_infinite_fallback;
+	int		can_ack;
+	int		fully_established;
+	u32		rcv_data_fin;
+	u32		snd_data_fin_enable;
+	u32		rcv_fastclose;
+	u32		use_64bit_ack; /* Set when we received a 64-bit DSN */
+	u32		csum_enabled;
+	u32		allow_infinite_fallback;
 	u8		pending_state; /* A subflow asked to set this sk_state,
 					* protected by the msk data lock
 					*/
@@ -445,8 +445,8 @@ struct mptcp_subflow_request_sock {
 		backup : 1,
 		csum_reqd : 1,
 		allow_join_id0 : 1;
-	u8	local_id;
-	u8	remote_id;
+	u16	local_id;
+	u32	remote_id;
 	u64	local_key;
 	u64	idsn;
 	u32	token;
@@ -519,8 +519,8 @@ struct mptcp_subflow_context {
 		valid_csum_seen : 1,        /* at least one csum validated */
 		is_mptfo : 1,	    /* subflow is doing TFO */
 		__unused : 10;
-	bool	data_avail;
-	bool	scheduled;
+	u32	data_avail;
+	u32	scheduled;
 	u32	remote_nonce;
 	u64	thmac;
 	u32	local_nonce;
@@ -529,8 +529,8 @@ struct mptcp_subflow_context {
 		u8	hmintac[MPTCPOPT_HMAC_LEN]; /* MPJ subflow only */
 		u64	iasn;	    /* initial ack sequence number, MPC subflows only */
 	};
-	s16	local_id;	    /* if negative not initialized yet */
-	u8	remote_id;
+	u32	local_id;	    /* if negative not initialized yet */
+	u32	remote_id;
 	u8	reset_seen:1;
 	u8	reset_transient:1;
 	u8	reset_reason:4;
diff --git a/scripts/Makefile.extrawarn b/scripts/Makefile.extrawarn
index 55f6e6917033..02c54a8b2f2f 100644
--- a/scripts/Makefile.extrawarn
+++ b/scripts/Makefile.extrawarn
@@ -25,7 +25,7 @@ endif
 
 KBUILD_CPPFLAGS-$(CONFIG_WERROR) += -Werror
 KBUILD_CPPFLAGS += $(KBUILD_CPPFLAGS-y)
-KBUILD_CFLAGS-$(CONFIG_CC_NO_ARRAY_BOUNDS) += -Wno-array-bounds
+#KBUILD_CFLAGS-$(CONFIG_CC_NO_ARRAY_BOUNDS) += -Wno-array-bounds
 
 ifdef CONFIG_CC_IS_CLANG
 # The kernel builds with '-std=gnu11' so use of GNU extensions is acceptable.
diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h
index 289bf9233f71..92dff1fd4ac3 100644
--- a/security/selinux/include/security.h
+++ b/security/selinux/include/security.h
@@ -91,7 +91,7 @@ struct selinux_policy;
 
 struct selinux_state {
 #ifdef CONFIG_SECURITY_SELINUX_DEVELOP
-	bool enforcing;
+	int enforcing;
 #endif
 	bool initialized;
 	bool policycap[__POLICYDB_CAP_MAX];
diff --git a/security/tomoyo/common.h b/security/tomoyo/common.h
index 0e8e2e959aef..ebca489c729f 100644
--- a/security/tomoyo/common.h
+++ b/security/tomoyo/common.h
@@ -772,7 +772,7 @@ struct tomoyo_inet_acl {
 struct tomoyo_unix_acl {
 	struct tomoyo_acl_info head; /* type = TOMOYO_TYPE_UNIX_ACL */
 	u8 protocol;
-	u8 perm; /* Bitmask of values in "enum tomoyo_network_acl_index" */
+	u32 perm; /* Bitmask of values in "enum tomoyo_network_acl_index" */
 	struct tomoyo_name_union name;
 };
 
diff --git a/security/tomoyo/network.c b/security/tomoyo/network.c
index 8dc61335f65e..dccd873b5bf8 100644
--- a/security/tomoyo/network.c
+++ b/security/tomoyo/network.c
@@ -257,7 +257,7 @@ static bool tomoyo_merge_unix_acl(struct tomoyo_acl_info *a,
 				  struct tomoyo_acl_info *b,
 				  const bool is_delete)
 {
-	u8 * const a_perm =
+	u32 * const a_perm =
 		&container_of(a, struct tomoyo_unix_acl, head)->perm;
 	u8 perm = READ_ONCE(*a_perm);
 	const u8 b_perm = container_of(b, struct tomoyo_unix_acl, head)->perm;

^ permalink raw reply related	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-31 15:48         ` Arnd Bergmann
@ 2024-05-31 16:32           ` Linus Torvalds
  2024-05-31 16:54             ` Arnd Bergmann
                               ` (2 more replies)
  2024-06-03 11:33           ` Maciej W. Rozycki
  1 sibling, 3 replies; 55+ messages in thread
From: Linus Torvalds @ 2024-05-31 16:32 UTC (permalink / raw)
  To: Arnd Bergmann
  Cc: Maciej W. Rozycki, Paul E. McKenney, John Paul Adrian Glaubitz,
	Arnd Bergmann, linux-alpha, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Fri, 31 May 2024 at 08:48, Arnd Bergmann <arnd@arndb.de> wrote:
>
>  /* Is this type a native word size -- useful for atomic operations */
>  #define __native_word(t) \
> -       (sizeof(t) == sizeof(char) || sizeof(t) == sizeof(short) || \
> -        sizeof(t) == sizeof(int) || sizeof(t) == sizeof(long))
> +       (sizeof(t) == sizeof(int) || sizeof(t) == sizeof(long))
>
>  #ifdef __OPTIMIZE__
>  # define __compiletime_assert(condition, msg, prefix, suffix)          \
>
> The WRITE_ONCE() calls tend to be there in order to avoid
> expensive atomic or locking when something can be expressed
> with a store that known to be visible atomically (on all other
> architectures).

No, if you go down this road, then you would want to do the same thing
we do for READ_ONCE() - but for a different reason - hook into it for
alpha, and add a memory barrier to get rid of the crazy alpha memory
ordering:

  /*
   * Alpha is apparently daft enough to reorder address-dependent loads
   * on some CPU implementations. Knock some common sense into it with
   * a memory barrier in READ_ONCE().
   *
   * For the curious, more information about this unusual reordering is
   * available in chapter 15 of the "perfbook":
   *
   *  https://kernel.org/pub/linux/kernel/people/paulmck/perfbook/perfbook.html
   *
   */
  #define __READ_ONCE(x)                                                  \
  ({                                                                      \
        __unqual_scalar_typeof(x) __x =                                 \
                (*(volatile typeof(__x) *)(&(x)));                      \
        mb();                                                           \
        (typeof(x))__x;                                                 \
  })

and the solution would be to make a __WRITE_ONCE() that then uses
"sizeof()" to decide at compile-time whether it can just do it as a
regular write, or whether it needs to do it as a LL/SC loop.

Because we're definitely not changing hundreds - probably thousands -
of random generic data structures.

That said, the above fixes WRITE_ONCE() without changing the
definition of what a native word size is, but doesn't actually *fix*
the problem.

Let's take a really simple example:

    struct net_device {
        ...
        u8 reg_state;

        bool dismantle;

        enum {
                RTNL_LINK_INITIALIZED,
                RTNL_LINK_INITIALIZING,
        } rtnl_link_state:16;
        ...

are all in the same 32-bit word, and we intentionally have code
without locking like this:

        WRITE_ONCE(dev->reg_state, NETREG_RELEASED);
...
        return READ_ONCE(dev->reg_state) <= NETREG_REGISTERED;

because the code knows the state machine ordering requirements (ie
once it has moved past NETREG_REGISTERED, it won't move back).

So now - assuming we fix WRITE_ONCE() to use LL/SC, these READ_ONCE()
and WRITE_ONCE() games work fine on alpha

BUT.

Code that then does something like this:

        dev->dismantle = true;

which is all nice and good (accesses are done under the RTNL lock) now
will possibly race with the unlocked reg_state accesses.

So it's still fundamentally buggy.

And before you say "that's why I wanted to fix the __native_word()
definition", please realize that the above happens EVEN WITH the
READ_ONCE/WRITE_ONCE being done on an "int".

Yes, really. The READ_ONCE and WRITE_ONCE will be individual
instructions. But lookie here, if we have

        u32 reg_state;
        bool dismantle;

and they happen to share the same 8-byte word, and somebody passes
'&dismantle' off to something that does byte writes to it, guess what
the canonical byte write sequence is?

That's right, it looks something like this (excuse any bugs, this is
from memory and looking up the ops in the architecture manual):

        LDQ_U tmp,(addr)
        INSBL byte,addr,tmp2
        MSKBL tmp,addr,tmp
        OR tmp,tmp2,tmp
        STQ_U tmp,(addr)

and notice how in the process it read and then wrote that supposedly
atomic 'req_state" that was otherwise accessed purely with 32-bit
atomic instructions?

There are no LDL_U/STL_U instructions. The unaligned memory ops are
always 8 bytes wide (you can obviously always do address masking
manually and "emulate" a LDL_U/STL_U model, but then you make already
bad code generation even *worse*).

So no. Even 32-bit values aren't "atomic" in alpha, because of the
complete sh*t-show that is lack of byte ops.

NOTE NOTE NOTE! Note how I said "pass off the address of
'dev->dismantle' to something that does byte ops"? If you *know* the
alignment of the byte in a structure, so you don't just get a random
pointer to a byte, you can - and should - generate better code on
alpha, which may in fact involve just doing a 32-bit load, masking off
the low bits, and doing the 32-bit store.

So that LDQ_U/STQ_U sequence is for the generic case, with various
simpler sub-cases that don't necessarily require it.

The fact is, the original alpha is the worst architecture ever made.
The lack of byte instructions and the absolutely horrendous memory
ordering are fatal flaws. And while the memory ordering arguably had
excuses for it ("they didn't know better"), the lack of byte ops was
wilful misdesign that the designers were proud of, and made a central
tenet of their mess.

And I say that as somebody who *loved* it originally. Yes, the lack of
byte operations always was a pain, because it really caused the IO
subsystem to be a nightmare, but I was young, I was stupid, it was
interesting, and I had bought into the kool aid.

But alpha without BWX really is shit. People think x86 is bad. Those
people have NO CLUE.

               Linus

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-31 16:32           ` Linus Torvalds
@ 2024-05-31 16:54             ` Arnd Bergmann
  2024-06-01 13:51             ` David Laight
  2024-07-01 23:48             ` Maciej W. Rozycki
  2 siblings, 0 replies; 55+ messages in thread
From: Arnd Bergmann @ 2024-05-31 16:54 UTC (permalink / raw)
  To: Linus Torvalds, Arnd Bergmann
  Cc: Maciej W. Rozycki, Paul E. McKenney, John Paul Adrian Glaubitz,
	linux-alpha, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, linux-kernel, Michael Cree,
	Frank Scheiner

On Fri, May 31, 2024, at 18:32, Linus Torvalds wrote:
> On Fri, 31 May 2024 at 08:48, Arnd Bergmann <arnd@arndb.de> wrote:
>
> and the solution would be to make a __WRITE_ONCE() that then uses
> "sizeof()" to decide at compile-time whether it can just do it as a
> regular write, or whether it needs to do it as a LL/SC loop.

Just to clarify: I did not mean to suggest we actually
apply the patch from my email or anything close to it, this was
meant as a demonstration for instances that are likely broken but
each one would need to be individually documented if we were to
try to fix them, as Maciej asked about what it would take.

My "alpha: drop pre-EV56 support" actually removed two workarounds
for non-BWX alphas, and neither of them even uses WRITE_ONCE,
so these would also have gone unnoticed by any changes we do
to WRITE_ONCE.

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/include/linux/?id=a4184174be36369c3af8d937e165f28a43ef1e02

    Arnd

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-31  3:56           ` Maciej W. Rozycki
@ 2024-05-31 19:33             ` Paul E. McKenney
  2024-06-03 16:22               ` Maciej W. Rozycki
  0 siblings, 1 reply; 55+ messages in thread
From: Paul E. McKenney @ 2024-05-31 19:33 UTC (permalink / raw)
  To: Maciej W. Rozycki
  Cc: John Paul Adrian Glaubitz, Arnd Bergmann, linux-alpha,
	Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, linux-kernel,
	Michael Cree, Frank Scheiner

On Fri, May 31, 2024 at 04:56:28AM +0100, Maciej W. Rozycki wrote:
> On Wed, 29 May 2024, Paul E. McKenney wrote:
> 
> > >  Mind that the read-modify-write sequence that software does for sub-word 
> > > write accesses with original Alpha hardware is precisely what hardware 
> > > would have to do anyway and support for that was deliberately omitted by 
> > > the architecture designers from the ISA to give it performance advantages 
> > > quoted in the architecture manual.  The only difference here is that with 
> > > hardware read-modify-write operations atomicity for sub-word accesses is 
> > > guaranteed by the ISA, however for software read-modify-write it has to be 
> > > explictly coded using the usual load-locked/store-conditional sequence in 
> > > a loop.  I don't think it's a big deal really, it should be trivial to do 
> > > in the relevant accessors, along with the memory barriers that are needed 
> > > anyway for EV56+ and possibly other ports such as the MIPS one.
> > 
> > There shouldn't be any memory barriers required, and don't EV56+ have
> > single-byte loads and stores?
> 
>  I should have commented on this in my original reply.
> 
>  You're the RCU expert so you know the answer.  I don't.  If it's OK for
> successive writes to get reordered, or readers to see a stale value, then 
> you don't need memory barriers.  Otherwise you do.  Whether byte accesses 
> are available or not does not matter, the CPU *will* do reordering if it's 
> allowed to (or more specifically, it won't do anything to prevent it from 
> happening, especially in SMP configurations; I can't remember offhand if 
> there are cases with UP).  Also adjacent byte writes may be merged, but I 
> suppose it does not matter, or does it?

RCU uses whichever wrapper is required.  For example, if ordering is
required, it might use smp_store_release() and smp_load_acquire().
If ordering does not matter, it might use WRITE_ONCE() and READ_ONCE().
If tearing/fusing/merging does not matter, as in there are not concurrent
accesses, it uses plain C-language loads and stores.

>  NB MIPS has similar architectural arrangements (and a bunch of barriers 
> defined in the ISA), it's just most implementations are actually strongly 
> ordered, so most people can't see the effects of this.  With MIPS I know 
> for sure there are cases of UP reordering, but they only really matter for 
> MMIO use cases and not regular memory.

Any given architecture is required to provide architecture-specific
implementations of the various functions that meet the requirements of
Linux-kernel memory model.  See tools/memory-model for more information.

							Thanx, Paul

^ permalink raw reply	[flat|nested] 55+ messages in thread

* RE: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-31 16:32           ` Linus Torvalds
  2024-05-31 16:54             ` Arnd Bergmann
@ 2024-06-01 13:51             ` David Laight
  2024-07-01 23:48             ` Maciej W. Rozycki
  2 siblings, 0 replies; 55+ messages in thread
From: David Laight @ 2024-06-01 13:51 UTC (permalink / raw)
  To: 'Linus Torvalds', Arnd Bergmann
  Cc: Maciej W. Rozycki, Paul E. McKenney, John Paul Adrian Glaubitz,
	Arnd Bergmann, linux-alpha, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

...
> The fact is, the original alpha is the worst architecture ever made.
> The lack of byte instructions and the absolutely horrendous memory
> ordering are fatal flaws. And while the memory ordering arguably had
> excuses for it ("they didn't know better"), the lack of byte ops was
> wilful misdesign that the designers were proud of, and made a central
> tenet of their mess.

If it wasn't from DEC (where the pdp-11 and vax were fine) I'd think
it was someone harking back to the old mainframe days where is was
perfectly normal to only have 'word addressing' and, for example,
to put three 6-bit characters into an 18-bit word (hi Univac!).
(Don't even think how 18-bit words got written to mag tape!)

It is almost as is someone assumed that the only use for byte accesses
was within character arrays - and they can jolly well align the arrays.

Mind you, all the byte shifting needed to get the data onto the
right data bus lines is a PITA and will affect the max cpu frequency [1].
So perhaps they decided it was a 'software problem' so some benchmarks
could run faster.

	David

[1] I've been busy re-implementing the Nios-II cpu.

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 14/14] alpha: drop pre-EV56 support
  2024-05-03  8:11 ` [PATCH 14/14] alpha: drop pre-EV56 support Arnd Bergmann
  2024-05-04 15:00   ` Richard Henderson
@ 2024-06-03  6:02   ` Jiri Slaby
  2024-06-04 13:58     ` Greg KH
  1 sibling, 1 reply; 55+ messages in thread
From: Jiri Slaby @ 2024-06-03  6:02 UTC (permalink / raw)
  To: Arnd Bergmann, linux-alpha
  Cc: Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, Paul E. McKenney,
	linux-kernel, Greg KH

Cc Greg for awareness too.

On 03. 05. 24, 10:11, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> All EV4 machines are already gone, and the remaining EV5 based machines
> all support the slightly more modern EV56 generation as well.
> Debian only supports EV56 and later.
> 
> Drop both of these and build kernels optimized for EV56 and higher
> when the "generic" options is selected, tuning for an out-of-order
> EV6 pipeline, same as Debian userspace.
> 
> Since this was the only supported architecture without 8-bit and
> 16-bit stores, common kernel code no longer has to worry about
> aligning struct members, and existing workarounds from the block
> and tty layers can be removed.

Yes!

For TTY pieces below:
Reviewed-by: Jiri Slaby <jirislaby@kernel.org>

> diff --git a/include/linux/tty.h b/include/linux/tty.h
> index 2b2e6f0a54d6..2372f9357240 100644
> --- a/include/linux/tty.h
> +++ b/include/linux/tty.h
> @@ -145,15 +145,12 @@ struct tty_operations;
>    * @count: count of open processes, reaching zero cancels all the work for
>    *	   this tty and drops a @kref too (but does not free this tty)
>    * @winsize: size of the terminal "window" (cf. @winsize_mutex)
> - * @flow: flow settings grouped together, see also @flow.unused
> + * @flow: flow settings grouped together
>    * @flow.lock: lock for @flow members
>    * @flow.stopped: tty stopped/started by stop_tty()/start_tty()
>    * @flow.tco_stopped: tty stopped/started by %TCOOFF/%TCOON ioctls (it has
>    *		      precedence over @flow.stopped)
> - * @flow.unused: alignment for Alpha, so that no members other than @flow.* are
> - *		 modified by the same 64b word store. The @flow's __aligned is
> - *		 there for the very same reason.
> - * @ctrl: control settings grouped together, see also @ctrl.unused
> + * @ctrl: control settings grouped together
>    * @ctrl.lock: lock for @ctrl members
>    * @ctrl.pgrp: process group of this tty (setpgrp(2))
>    * @ctrl.session: session of this tty (setsid(2)). Writes are protected by both
> @@ -161,7 +158,6 @@ struct tty_operations;
>    *		  them.
>    * @ctrl.pktstatus: packet mode status (bitwise OR of %TIOCPKT_ constants)
>    * @ctrl.packet: packet mode enabled
> - * @ctrl.unused: alignment for Alpha, see @flow.unused for explanation
>    * @hw_stopped: not controlled by the tty layer, under @driver's control for CTS
>    *		handling
>    * @receive_room: bytes permitted to feed to @ldisc without any being lost
> @@ -216,8 +212,7 @@ struct tty_struct {
>   		spinlock_t lock;
>   		bool stopped;
>   		bool tco_stopped;
> -		unsigned long unused[0];
> -	} __aligned(sizeof(unsigned long)) flow;
> +	} flow;
>   
>   	struct {
>   		struct pid *pgrp;
> @@ -225,8 +220,7 @@ struct tty_struct {
>   		spinlock_t lock;
>   		unsigned char pktstatus;
>   		bool packet;
> -		unsigned long unused[0];
> -	} __aligned(sizeof(unsigned long)) ctrl;
> +	} ctrl;
>   
>   	bool hw_stopped;
>   	bool closing;

-- 
js
suse labs


^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-31  0:10             ` Linus Torvalds
@ 2024-06-03 11:09               ` Maciej W. Rozycki
  2024-06-03 11:36                 ` John Paul Adrian Glaubitz
  2024-06-03 16:57                 ` Linus Torvalds
  0 siblings, 2 replies; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-06-03 11:09 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Paul E. McKenney, John Paul Adrian Glaubitz, Arnd Bergmann,
	linux-alpha, Arnd Bergmann, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Thu, 30 May 2024, Linus Torvalds wrote:

> > > The 21064 actually did atomicity with an external pin on the bus, the
> > > same way people used to do before caches even existed.
> >
> >  Umm, 8086's LOCK#, anyone?
> 
> Well, yes and no.
> 
> So yes, exactly like 8086 did before having caches.

 Well I wrote 8086 specifically, not x86.

> But no, not like the alpha contemporary PPro that did have caches. The
> PPro already did locked cycles in the caches.

 But the 21064 does predate the PPro by a couple of years: Feb 1992 vs Nov 
1995, so surely Intel folks had extra time to resolve this stuff properly.  

 Conversely the R4000 came about in Oct 1991, so before the 21064.  But 
only so slightly and not as much as I remembered (I thought the 21064 was 
more like 1993), so it seems like DEC couldn't have had enough time after 
all to figure out what SGI did (patents notwithstanding).  Surely the 
R4000MC cache coherency protocol was complex for the silicon technology of 
the time, but it's just MOESI in modern terms AFAICT, and LL/SC is handled 
there (and is in fact undefined for uncached accesses).

 I'm not sure what else was out there at the time, but going back to x86 
the i486 was contemporary, the original write-through cache version, which 
if memory serves, was not any better in this respect (and the "write-back 
enhanced" DX2/DX4 models with proper MESI cache protocol came out much 
later, after Pentium only, which they borrowed from).

> So I really feel the 21064 was broken.
> 
> It's probably related to the whole cache coherency being designed to
> be external to the built-in caches - or even the Bcache. The caches
> basically are write-through, and the weak memory ordering was designed
> for allowing this horrible model.

 In retrospect perhaps it wasn't the best design, but they have learnt 
from their mistakes.

> > > In fact, it's worse than "not thread safe". It's not even safe on UP
> > > with interrupts, or even signals in user space.
> >
> >  Ouch, I find it a surprising oversight.
> 
> The sad part is that it doesn't seem to have been an oversight. It
> really was broken-as-designed.
> 
> Basically, the CPU was designed for single-threaded Spec benchmarks
> and absolutely nothing else. Classic RISC where you recompile to fix
> problems like the atomicity thing - "just use a 32-bit sig_atomic_t
> and you're fine")

 Not OK however, as you correctly point out, for plain ordinary non-atomic 
stuff.  Point me at any document that claims that a pair of threads poking 
at even and odd byte vector elements each is not allowed.  Caches may not 
enjoy it, but there's nothing AFAIK saying this is UB or whatever.

> The original alpha architecture handbook makes a big deal of how
> clever the lack of byte and word operations is. I also remember

 I've seen that; dropped in v3 with the addition of the BWX extension.

> reading an article by Dick Sites - one of the main designers - talking
> a lot about how the lack of byte operations is great, and encourages
> vectorizing byte accesses and doing string operations in whole words.

 Yeah, the software folks at DEC must have been delighted porting all the 
VAX VMS software.  But pehaps this was the last attempt to try something 
different from the CPU architecture standards established back in 1970s 
(by the VAX among others) that make current designs so similar to one 
another.

 Anyway, back to my point.  A feasible solution non-intrusive for Linux 
and low-overhead for GCC has been found.  I can expedite implementation 
and I'll see if I can regression-test it too, but I may have to rely on 
other people to complete it after all, as I haven't been prepared for this 
effort in the light of certain issues I have recently suffered from in my 
lab.

 Is that going to be enough to bring the platform bits back?

 FAOD, with all the hacks so eagerly being removed now happily left in the 
dust bin where they belong, and which I wholeheartedly agree with: we 
shouldn't be suffering from design mistakes of systems that are no longer 
relevant, but I fail to see the reason why we should disallow their use 
where the burden is confined or plain elsewhere.

 For example we continue supporting old UP MIPS platforms that predate 
LL/SC, by just trapping and emulating these instructions.  Surely it sucks 
performance-wise and it's possibly hundreds of cycles too, but it works 
and the burden is confined to the exception handler, so not a big deal.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-31 15:48         ` Arnd Bergmann
  2024-05-31 16:32           ` Linus Torvalds
@ 2024-06-03 11:33           ` Maciej W. Rozycki
  1 sibling, 0 replies; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-06-03 11:33 UTC (permalink / raw)
  To: Arnd Bergmann
  Cc: Paul E. McKenney, John Paul Adrian Glaubitz, Arnd Bergmann,
	linux-alpha, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, linux-kernel,
	Michael Cree, Frank Scheiner

On Fri, 31 May 2024, Arnd Bergmann wrote:

> I then tried changing the underlying variables to 32-bit ones
> to see how many changes are needed, but I gave up after around
> 150 of them, as I was only scratching the surface. To do this
> right, you'd need to go through each one of them and come up
> with a solution that is the best trade-off in terms of memory
> usage and performance for that one. There are of course
> others that should be using WRITE_ONCE() and are missing
> this, so the list is not complete either. See below for
> the ones I could find quickly.

 Thank you for your attempt, and I agree this is excessive and beyond what 
we can reasonably handle.

> >  FWIW even if it was only me I think that depriving the already thin Alpha 
> > port developer base of any quantity of the remaining manpower, by dropping 
> > support for a subset of the hardware available, and then a subset that is 
> > not just as exotic as the original i386 became to the x86 platform at the 
> > time support for it was dropped, is only going to lead to further demise 
> > and eventual drop of the entire port.
> 
> I know you like you museum pieces to be older than everyone
> else's, and I'm sorry that my patch series is causing you
> problems, but I don't think the more general criticism is
> valid here. My hope was mainly to help our with both keeping
> Alpha viable for a few more years while also allowing Paul
> to continue with his RCU changes.

 Appreciated and thank you for your appreciation as well.

> As far as I can tell, nobody else is actually using EV4
> machines or has been for years now, but the presence of that
> code did affect both the performance and correctness of the
> kernel code for all EV56+ users since distros have no way
> of picking the ISA level on alpha for a generic kernel.

 Well, at least John Paul Adrian complained as well, and who knows who 
else is there downstream.  I'd expect most people (i.e. all except for 
core Linux developers) not to track upstream development in a continuous 
manner.

> The strongest argument I see for assuming non-BWX alphas
> are long dead is that gcc-4.4 added support for C11 style
> _Atomic variables for alpha, but got the stores wrong
> without anyone ever noticing the problem. Even one makes
> the argument that normal byte stores and volatiles ones
> should not need atomic ll/st sequenes, the atomics
> clearly do. Building BWX-enabled kernels and userland
> completely avoids this problem, which make debugging
> easier for the remaining users when stuff breaks.

 This only shows the lack of proper verification here rather than just 
use.  I'm not even sure if the nature of this problem is going to make it 
trigger in GCC regression testing.  Which BTW I have wired my EV45 system 
for in my lab last year and which would be going by now if not for issues 
with support network automation equipment (FAOD, state of the art and 
supported by the manufacturer).  We shall see once I'm done.

 As John Paul Adrian has pointed out the removal was expedited with no 
attempt made to find a proper solution that would not affect other users.  
As you can see it took me one e-mail exchange with Linus to understand 
what the underlying issue has been and then just a little bit of thinking, 
maybe half an hour, likely even less, to identify a feasible solution.

 Yes, I could have come up with it maybe a month ago if I wasn't so much 
behind on mailing list traffic.  But it's not my day job and since we had 
this issue for years now, it wasn't something that had to be handled as a 
matter of urgency.  We all are people and have our limitations.  We could 
have waited with the RFC out for another development cycle.  This has been 
the point of my complaint.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-06-03 11:09               ` Maciej W. Rozycki
@ 2024-06-03 11:36                 ` John Paul Adrian Glaubitz
  2024-06-03 16:57                 ` Linus Torvalds
  1 sibling, 0 replies; 55+ messages in thread
From: John Paul Adrian Glaubitz @ 2024-06-03 11:36 UTC (permalink / raw)
  To: Maciej W. Rozycki, Linus Torvalds
  Cc: Paul E. McKenney, Arnd Bergmann, linux-alpha, Arnd Bergmann,
	Richard Henderson, Ivan Kokshaysky, Matt Turner, Alexander Viro,
	Marc Zyngier, linux-kernel, Michael Cree, Frank Scheiner

Hi Maciej,

On Mon, 2024-06-03 at 12:09 +0100, Maciej W. Rozycki wrote:
>  Anyway, back to my point.  A feasible solution non-intrusive for Linux 
> and low-overhead for GCC has been found.  I can expedite implementation 
> and I'll see if I can regression-test it too, but I may have to rely on 
> other people to complete it after all, as I haven't been prepared for this 
> effort in the light of certain issues I have recently suffered from in my 
> lab.

That's really great to hear! Please let me know if you have something to test,
I would love to help with this effort.

>  Is that going to be enough to bring the platform bits back?

That would be awesome. Would love to be able to keep running a current kernel
on my AlphaStation 233 which is pre-EV56.

>  FAOD, with all the hacks so eagerly being removed now happily left in the 
> dust bin where they belong, and which I wholeheartedly agree with: we 
> shouldn't be suffering from design mistakes of systems that are no longer 
> relevant, but I fail to see the reason why we should disallow their use 
> where the burden is confined or plain elsewhere.

Agreed.

>  For example we continue supporting old UP MIPS platforms that predate 
> LL/SC, by just trapping and emulating these instructions.  Surely it sucks 
> performance-wise and it's possibly hundreds of cycles too, but it works 
> and the burden is confined to the exception handler, so not a big deal.

Fully agreed.

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-31 19:33             ` Paul E. McKenney
@ 2024-06-03 16:22               ` Maciej W. Rozycki
  2024-06-03 17:08                 ` Paul E. McKenney
  0 siblings, 1 reply; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-06-03 16:22 UTC (permalink / raw)
  To: Paul E. McKenney
  Cc: John Paul Adrian Glaubitz, Arnd Bergmann, linux-alpha,
	Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, linux-kernel,
	Michael Cree, Frank Scheiner

On Fri, 31 May 2024, Paul E. McKenney wrote:

> >  You're the RCU expert so you know the answer.  I don't.  If it's OK for
> > successive writes to get reordered, or readers to see a stale value, then 
> > you don't need memory barriers.  Otherwise you do.  Whether byte accesses 
> > are available or not does not matter, the CPU *will* do reordering if it's 
> > allowed to (or more specifically, it won't do anything to prevent it from 
> > happening, especially in SMP configurations; I can't remember offhand if 
> > there are cases with UP).  Also adjacent byte writes may be merged, but I 
> > suppose it does not matter, or does it?
> 
> RCU uses whichever wrapper is required.  For example, if ordering is
> required, it might use smp_store_release() and smp_load_acquire().
> If ordering does not matter, it might use WRITE_ONCE() and READ_ONCE().
> If tearing/fusing/merging does not matter, as in there are not concurrent
> accesses, it uses plain C-language loads and stores.

 Fair enough.

> >  NB MIPS has similar architectural arrangements (and a bunch of barriers 
> > defined in the ISA), it's just most implementations are actually strongly 
> > ordered, so most people can't see the effects of this.  With MIPS I know 
> > for sure there are cases of UP reordering, but they only really matter for 
> > MMIO use cases and not regular memory.
> 
> Any given architecture is required to provide architecture-specific
> implementations of the various functions that meet the requirements of
> Linux-kernel memory model.  See tools/memory-model for more information.

 This is a fairly recent addition, thank you for putting it all together.  
I used to rely solely on Documentation/memory-barriers.txt.  Thanks for 
the reference.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-06-03 11:09               ` Maciej W. Rozycki
  2024-06-03 11:36                 ` John Paul Adrian Glaubitz
@ 2024-06-03 16:57                 ` Linus Torvalds
  2024-07-01 23:48                   ` Maciej W. Rozycki
  1 sibling, 1 reply; 55+ messages in thread
From: Linus Torvalds @ 2024-06-03 16:57 UTC (permalink / raw)
  To: Maciej W. Rozycki
  Cc: Paul E. McKenney, John Paul Adrian Glaubitz, Arnd Bergmann,
	linux-alpha, Arnd Bergmann, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Mon, 3 Jun 2024 at 04:09, Maciej W. Rozycki <macro@orcam.me.uk> wrote:
>
>  Anyway, back to my point.  A feasible solution non-intrusive for Linux
> and low-overhead for GCC has been found.  I can expedite implementation
> and I'll see if I can regression-test it too, but I may have to rely on
> other people to complete it after all, as I haven't been prepared for this
> effort in the light of certain issues I have recently suffered from in my
> lab.

Yeah, if compiler support makes us not have to care, then I don't
think the difference between pre-BWX and BWX is going to matter much
for the kernel.

The real pain with alpha has been that it's special enough that it
affects non-alpha code, and BWX was one big piece of that.

That said, some of the EV4 misfeatures end up being a huge pain inside
the alpha code either because of the horrible hoops that the IO
accessors have to jump through, or because of the broken ASID's.

So even with enw compiler support, maybe it's worth trying to
re-introduce any support for older cpu's incrementally.

For example, the ASID hw issue is _claimed_ to have been fixed in
PALcode, and maybe the games we played for ev4-era cpus aren't
actually needed any more?

And the various odd IO platforms should only be re-introduced when
there are people who actually have access to the relevant hardware and
will test.

           Linus

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-06-03 16:22               ` Maciej W. Rozycki
@ 2024-06-03 17:08                 ` Paul E. McKenney
  2024-07-01 23:50                   ` Maciej W. Rozycki
  0 siblings, 1 reply; 55+ messages in thread
From: Paul E. McKenney @ 2024-06-03 17:08 UTC (permalink / raw)
  To: Maciej W. Rozycki
  Cc: John Paul Adrian Glaubitz, Arnd Bergmann, linux-alpha,
	Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, linux-kernel,
	Michael Cree, Frank Scheiner

On Mon, Jun 03, 2024 at 05:22:22PM +0100, Maciej W. Rozycki wrote:
> On Fri, 31 May 2024, Paul E. McKenney wrote:
> 
> > >  You're the RCU expert so you know the answer.  I don't.  If it's OK for
> > > successive writes to get reordered, or readers to see a stale value, then 
> > > you don't need memory barriers.  Otherwise you do.  Whether byte accesses 
> > > are available or not does not matter, the CPU *will* do reordering if it's 
> > > allowed to (or more specifically, it won't do anything to prevent it from 
> > > happening, especially in SMP configurations; I can't remember offhand if 
> > > there are cases with UP).  Also adjacent byte writes may be merged, but I 
> > > suppose it does not matter, or does it?
> > 
> > RCU uses whichever wrapper is required.  For example, if ordering is
> > required, it might use smp_store_release() and smp_load_acquire().
> > If ordering does not matter, it might use WRITE_ONCE() and READ_ONCE().
> > If tearing/fusing/merging does not matter, as in there are not concurrent
> > accesses, it uses plain C-language loads and stores.
> 
>  Fair enough.
> 
> > >  NB MIPS has similar architectural arrangements (and a bunch of barriers 
> > > defined in the ISA), it's just most implementations are actually strongly 
> > > ordered, so most people can't see the effects of this.  With MIPS I know 
> > > for sure there are cases of UP reordering, but they only really matter for 
> > > MMIO use cases and not regular memory.
> > 
> > Any given architecture is required to provide architecture-specific
> > implementations of the various functions that meet the requirements of
> > Linux-kernel memory model.  See tools/memory-model for more information.
> 
>  This is a fairly recent addition, thank you for putting it all together.  
> I used to rely solely on Documentation/memory-barriers.txt.  Thanks for 
> the reference.

It has been in the kernel since April 2018, but OK.  And a big "thank you"
to all the people who made this possible and who continue contributing
to it.  And Documentation/memory-barriers.txt still matters, though the
long-term goal is for it to be subsumed into tools/memory-model.  Things
like compiler optimizations make this difficult, but not impossible.

Another precaution is to ensure that any contraints of a non-common-case
architecture be tested for.  For example, if I add a 64-bit divide, I
get yelled at promptly.  In contrast, that long list of byte accesses
that Arnd posted were suffered in silence.  So they accumulated well
past the point where they can reasonably be backed out.

							Thanx, Paul

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 14/14] alpha: drop pre-EV56 support
  2024-06-03  6:02   ` Jiri Slaby
@ 2024-06-04 13:58     ` Greg KH
  0 siblings, 0 replies; 55+ messages in thread
From: Greg KH @ 2024-06-04 13:58 UTC (permalink / raw)
  To: Jiri Slaby
  Cc: Arnd Bergmann, linux-alpha, Arnd Bergmann, Richard Henderson,
	Ivan Kokshaysky, Matt Turner, Alexander Viro, Marc Zyngier,
	Linus Torvalds, Paul E. McKenney, linux-kernel

On Mon, Jun 03, 2024 at 08:02:03AM +0200, Jiri Slaby wrote:
> Cc Greg for awareness too.
> 
> On 03. 05. 24, 10:11, Arnd Bergmann wrote:
> > From: Arnd Bergmann <arnd@arndb.de>
> > 
> > All EV4 machines are already gone, and the remaining EV5 based machines
> > all support the slightly more modern EV56 generation as well.
> > Debian only supports EV56 and later.
> > 
> > Drop both of these and build kernels optimized for EV56 and higher
> > when the "generic" options is selected, tuning for an out-of-order
> > EV6 pipeline, same as Debian userspace.
> > 
> > Since this was the only supported architecture without 8-bit and
> > 16-bit stores, common kernel code no longer has to worry about
> > aligning struct members, and existing workarounds from the block
> > and tty layers can be removed.
> 
> Yes!
> 
> For TTY pieces below:
> Reviewed-by: Jiri Slaby <jirislaby@kernel.org>

Nice!

Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-05-31 16:32           ` Linus Torvalds
  2024-05-31 16:54             ` Arnd Bergmann
  2024-06-01 13:51             ` David Laight
@ 2024-07-01 23:48             ` Maciej W. Rozycki
  2024-07-02  1:13               ` Linus Torvalds
  2 siblings, 1 reply; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-07-01 23:48 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Arnd Bergmann, Paul E. McKenney, John Paul Adrian Glaubitz,
	Arnd Bergmann, linux-alpha, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Fri, 31 May 2024, Linus Torvalds wrote:

> The fact is, the original alpha is the worst architecture ever made.
> The lack of byte instructions and the absolutely horrendous memory
> ordering are fatal flaws. And while the memory ordering arguably had
> excuses for it ("they didn't know better"), the lack of byte ops was
> wilful misdesign that the designers were proud of, and made a central
> tenet of their mess.
> 
> And I say that as somebody who *loved* it originally. Yes, the lack of
> byte operations always was a pain, because it really caused the IO
> subsystem to be a nightmare, but I was young, I was stupid, it was
> interesting, and I had bought into the kool aid.

 Looking from today's perspective it was clearly a bad choice.  However it 
was 30+ years ago, it wasn't so certain as it is now that x86 was there to 
stay -- indeed as I recall it DEC had the ambition to phase x86 out with 
their Alpha (whether they approached it the right way business-wise is 
another matter) -- so the notion of having a fully byte-addressed machine 
perhaps wasn't yet so obvious to DEC engineers as it is now, when most if 
not all the current CPU architectures have these fundamentals the same.  

 As I say it may have been the final attempt to do something differently 
before x86 domination forced everyone to be at least remotely compatible.

 And there used to be weirder architectures before people moved away from 
them and settled on the current paradigm, just as nobody wants to build a 
general-purpose atmospheric railway anymore and yet a while ago not only 
it didn't appear ridiculous, but such stuff was actually built and run, 
such as the South Devon Railway.

 And then it wasn't the only failed attempt: remember the i860 or the 
iAPX432?  At least the Alphas weren't a total disaster, they made their 
impact, the worst mistakes have been fixed as the architecture evolved, 
and the engineering legacy remains, often in unexpected places.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-06-03 16:57                 ` Linus Torvalds
@ 2024-07-01 23:48                   ` Maciej W. Rozycki
  0 siblings, 0 replies; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-07-01 23:48 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Paul E. McKenney, John Paul Adrian Glaubitz, Arnd Bergmann,
	linux-alpha, Arnd Bergmann, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Mon, 3 Jun 2024, Linus Torvalds wrote:

> >  Anyway, back to my point.  A feasible solution non-intrusive for Linux
> > and low-overhead for GCC has been found.  I can expedite implementation
> > and I'll see if I can regression-test it too, but I may have to rely on
> > other people to complete it after all, as I haven't been prepared for this
> > effort in the light of certain issues I have recently suffered from in my
> > lab.
> 
> Yeah, if compiler support makes us not have to care, then I don't
> think the difference between pre-BWX and BWX is going to matter much
> for the kernel.
> 
> The real pain with alpha has been that it's special enough that it
> affects non-alpha code, and BWX was one big piece of that.

 Understood, that's burden beyond justification for an obsolete legacy 
platform.

> That said, some of the EV4 misfeatures end up being a huge pain inside
> the alpha code either because of the horrible hoops that the IO
> accessors have to jump through, or because of the broken ASID's.
> 
> So even with enw compiler support, maybe it's worth trying to
> re-introduce any support for older cpu's incrementally.

 Ack.

> For example, the ASID hw issue is _claimed_ to have been fixed in
> PALcode, and maybe the games we played for ev4-era cpus aren't
> actually needed any more?

 Actually my system seems to be an odd relic that has very old PALcode:

[...]
X3.7-10895, built on Sep 15 1994 at 10:19:05
>>>sh conf

SRM Console X3.7-10895  VMS PALcode X5.48-60, OSF PALcode X1.35-42
[...]

-- which is dated well before the system's release date.  It has been 
heavily patched with extra components retrofitted on the PCB as if an 
early hardware revision and the part number labelled on the PCB it's an 
AlphaStation 250 and yet it came packaged as AlphaServer 300 (the only 
documented difference between the PCBs of the two systems is the maximum 
amount of DRAM supported), with a vast mismatch between the dates given on 
the PCB and the case.  I don't know what's the story behind it, maybe it 
once was a DEC engineering machine.

 And for instance its SRM cannot netboot over BOOTP/TFTP, it can only 
use MOP.  Not an issue for me, and I feel a bit uneasy about upgrading the 
firmware, I'd rather I didn't brick the machine.  I guess we shall see 
whether it matters and if so, then what can be done about it.

 I used an AlphaServer 300 before that was purchased brand new and I can't 
recall any such patching on the PCB, and I reckon SRM was more modern too.
Indeed having checked old logs I found:

[...]
version                 X6.2-165 Nov  4 1996 10:06:10
>>>sh conf

Firmware
SRM Console:    X6.2-165
ARC Console:    4.49
PALcode:        VMS PALcode V5.56-2, OSF PALcode X1.46-2
[...]

> And the various odd IO platforms should only be re-introduced when
> there are people who actually have access to the relevant hardware and
> will test.

 Absolutely, what's the point of keeping something we have no way to
verify?  I'll begin with what I'm interested in myself and will gather 
input from people willing to verify stuff with other hardware they may 
have.

 Anyway, it's been a hectic month for me and I have my Alpha machine in 
the remote lab fully ready for this effort now, with a number of issues 
fixed, most importantly rather tricky GCC PR rtl-optimization/115565, 
<https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115565> which prevented the 
userland from being used as installed.

 With that in place the system was able to complete GCC 15 verification, 
so now it should be able to do pretty much anything.  I ran some glibc 
upstream master testing too.

 With that ticked off I do hope to work on the GCC part throughout July, 
and then the kernel bits will follow.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-06-03 17:08                 ` Paul E. McKenney
@ 2024-07-01 23:50                   ` Maciej W. Rozycki
  0 siblings, 0 replies; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-07-01 23:50 UTC (permalink / raw)
  To: Paul E. McKenney
  Cc: John Paul Adrian Glaubitz, Arnd Bergmann, linux-alpha,
	Arnd Bergmann, Richard Henderson, Ivan Kokshaysky, Matt Turner,
	Alexander Viro, Marc Zyngier, Linus Torvalds, linux-kernel,
	Michael Cree, Frank Scheiner

On Mon, 3 Jun 2024, Paul E. McKenney wrote:

> >  This is a fairly recent addition, thank you for putting it all together.  
> > I used to rely solely on Documentation/memory-barriers.txt.  Thanks for 
> > the reference.
> 
> It has been in the kernel since April 2018, but OK.  And a big "thank you"

 When you've been around for 25+ years, 5 years back seems like yesterday.

> to all the people who made this possible and who continue contributing
> to it.  And Documentation/memory-barriers.txt still matters, though the
> long-term goal is for it to be subsumed into tools/memory-model.  Things
> like compiler optimizations make this difficult, but not impossible.

 I realise these are tough matters and I second your gratitude.

> Another precaution is to ensure that any contraints of a non-common-case
> architecture be tested for.  For example, if I add a 64-bit divide, I
> get yelled at promptly.  In contrast, that long list of byte accesses
> that Arnd posted were suffered in silence.  So they accumulated well
> past the point where they can reasonably be backed out.

 Well, it's easy to notice and yell when you get an unresolved link-time 
reference to __divdi3 or suchlike.  While such heisenbugs as those caused 
by the race condition from concurrent unprotected rmw accesses may all be 
too easily blamed on cosmic rays or any other random instability.

 Take for example the GCC bug I mentioned in my reply to Linus in this 
thread, GCC PR rtl-optimization/115565.  It took 20 years to spot, even 
though it's in heavily used code and it does not depend on timing: with 
the right conditions it will trigger every time.

 If I were aware of these issues, I would definitely have got at them 
sooner.  Anyway, as mentioned in the other reply, I've overcome system 
setup issues now and will be working on the problem discussed here.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-07-01 23:48             ` Maciej W. Rozycki
@ 2024-07-02  1:13               ` Linus Torvalds
  2024-07-03  0:12                 ` Maciej W. Rozycki
  0 siblings, 1 reply; 55+ messages in thread
From: Linus Torvalds @ 2024-07-02  1:13 UTC (permalink / raw)
  To: Maciej W. Rozycki
  Cc: Arnd Bergmann, Paul E. McKenney, John Paul Adrian Glaubitz,
	Arnd Bergmann, linux-alpha, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Mon, 1 Jul 2024 at 16:48, Maciej W. Rozycki <macro@orcam.me.uk> wrote:
>
>  Looking from today's perspective it was clearly a bad choice.  However it
> was 30+ years ago, it wasn't so certain as it is now that x86 was there to
> stay

No.

The thing is, it was objectively the wrong thing to do even 30 years
ago, and has nothing to do with x86.

The lack of byte operations literally means that even _word_
operations aren't reliable.

Because when you emulate byte operations with quad-word operations -
which is the way the alpha instruction set was literally designed -
you mess with the adjacent word too.

So even word accesses aren't safe. And I'm pretty sure that
'sig_atomic_t' was just 32-bit on alpha (that's what glibc had, and
I'm pretty sure OSF/1 did too). So...

And that's an issue even just UP, and just completely bog-standard
POSIX 1003.1 and C.

You really can't get much more basic than that.

So supposedly portable programs would have subtle bugs because the
architecture was bad, and the workarounds for that badness were
incomplete.

SMP and IO - which are a thing, and which were things that the
architecture was allegedly designed for - are then only much worse.

The architecture was wrong 30 years ago. It's not that it "became"
wrong in hindsight. It was wrong originally, and it's just that people
hadn't thought things through enough to realize how wrong it was.

The only way it's not wrong is if you say "byte accesses do not
matter". That's a very Cray way of looking at things - Cray 1 had a
64-bit "char" in C, because there were no byte accesses.

That's fine if your only goal in life is to do HPC.

So if you simply don't care about bytes, and you *only* work with
words and quad-words, then alpha looks ok.

But honestly, that's basically saying "in a different universe, alpha
is not a mis-design".

That's not the universe we live in, and it's entirely unrelated to
x86. Bytes were very much a thing 30 years ago, and they will be a
thing 30 years from now even if x86 is dead and buried.

Basically, the fundamental mistake of thinking that you can do byte
operations by just masking quad-words screwed up POSIX compatibility,
screwed up SMP, and majorly screwed up the alpha IO layer too.

And by the time it was fixed, it was too late.

Don't make excuses for it. It's not ok today, but it really wasn't ok
30 years ago either.

It's ok to have rose-colored glasses and have a weak spot in your
heart for an architecture. But let's not make that weak spot in your
heart be a weak spot in your mind.

            Linus

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-07-02  1:13               ` Linus Torvalds
@ 2024-07-03  0:12                 ` Maciej W. Rozycki
  2024-07-03  0:50                   ` Linus Torvalds
  0 siblings, 1 reply; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-07-03  0:12 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Arnd Bergmann, Paul E. McKenney, John Paul Adrian Glaubitz,
	Arnd Bergmann, linux-alpha, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Mon, 1 Jul 2024, Linus Torvalds wrote:

> The architecture was wrong 30 years ago. It's not that it "became"
> wrong in hindsight. It was wrong originally, and it's just that people
> hadn't thought things through enough to realize how wrong it was.
> 
> The only way it's not wrong is if you say "byte accesses do not
> matter". That's a very Cray way of looking at things - Cray 1 had a
> 64-bit "char" in C, because there were no byte accesses.
> 
> That's fine if your only goal in life is to do HPC.
> 
> So if you simply don't care about bytes, and you *only* work with
> words and quad-words, then alpha looks ok.
> 
> But honestly, that's basically saying "in a different universe, alpha
> is not a mis-design".

 Precisely my point!  We got so used to think in multiples of 8 bits that 
other approaches seem ridiculous.

 The PDP-10 operated on 36-bit quantities and strings were essentially 
clusters of 6-bit characters packed into 6-packs (which is also allegedly 
where the C language's original limitation of using at most six characters 
for identifiers came from -- so that the PDP-10 could compare a pair with 
a single machine instruction).

 So there was already legacy of doing things this way at DEC back in ~1990 
and I can envisage engineers there actually thought that to have a machine 
that in C terms has 32-bit shorts and ints, 64-bit longs and pointers, and 
strings as clusters of 8-bit characters packed into 4-packs or 8-packs was 
not at all unreasonable.  Or maybe just plain 32-bit characters.  After 
all you don't absolutely *have* to use data types of 8 or 16 bits exactly 
in width for anything, do you?  NB for strings nowadays we have Unicode 
and we could just use UTF-32 if not to waste memory.

 And even now ISO C is very flexible on data type widths and only requires 
the character data type to be at least 8 bits wide, and 16-bit and 24-bit 
examples are actually given in the standard itself.  Yes, POSIX requires 
the character data type to be 8 bits wide exactly now, but POSIX.1-1988 
deferred to ANSI C AFAICT.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-07-03  0:12                 ` Maciej W. Rozycki
@ 2024-07-03  0:50                   ` Linus Torvalds
  2024-07-04 22:21                     ` Maciej W. Rozycki
  0 siblings, 1 reply; 55+ messages in thread
From: Linus Torvalds @ 2024-07-03  0:50 UTC (permalink / raw)
  To: Maciej W. Rozycki
  Cc: Arnd Bergmann, Paul E. McKenney, John Paul Adrian Glaubitz,
	Arnd Bergmann, linux-alpha, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Tue, 2 Jul 2024 at 17:12, Maciej W. Rozycki <macro@orcam.me.uk> wrote:
>
> On Mon, 1 Jul 2024, Linus Torvalds wrote:
> >
> > But honestly, that's basically saying "in a different universe, alpha
> > is not a mis-design".
>
>  Precisely my point!  We got so used to think in multiples of 8 bits that
> other approaches seem ridiculous.

But Maciej - alpha *was* designed for bytes. It wasn't a Cray 1. It
wasn't a PDP-10. It was designed by the time people knew that bytes
were the dominant thing, and that bytes were important and the main
use case.

But it was designed BADLY. The architecture sucked.

Give it up. If alpha had been designed in the 60s or 70s when the
whole issue of bytes were was debatable, it would have been
incredible.

But no. It was designed for byte accesses, and it FAILED AT THEM.

              Linus

^ permalink raw reply	[flat|nested] 55+ messages in thread

* Re: [PATCH 00/14] alpha: cleanups for 6.10
  2024-07-03  0:50                   ` Linus Torvalds
@ 2024-07-04 22:21                     ` Maciej W. Rozycki
  0 siblings, 0 replies; 55+ messages in thread
From: Maciej W. Rozycki @ 2024-07-04 22:21 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Arnd Bergmann, Paul E. McKenney, John Paul Adrian Glaubitz,
	Arnd Bergmann, linux-alpha, Richard Henderson, Ivan Kokshaysky,
	Matt Turner, Alexander Viro, Marc Zyngier, linux-kernel,
	Michael Cree, Frank Scheiner

On Tue, 2 Jul 2024, Linus Torvalds wrote:

> >  Precisely my point!  We got so used to think in multiples of 8 bits that
> > other approaches seem ridiculous.
> 
> But Maciej - alpha *was* designed for bytes. It wasn't a Cray 1. It
> wasn't a PDP-10. It was designed by the time people knew that bytes
> were the dominant thing, and that bytes were important and the main
> use case.
> 
> But it was designed BADLY. The architecture sucked.

 OK, perhaps it was those people who decided to make it that way that 
lived in a parallel universe.

> Give it up. If alpha had been designed in the 60s or 70s when the
> whole issue of bytes were was debatable, it would have been
> incredible.
> 
> But no. It was designed for byte accesses, and it FAILED AT THEM.

 I guess they decided that trading byte and word accesses for simpler bus 
logic that does not have all the bits required to issue an RMW operation 
to recalculate the ECC syndrome on such accesses was a good deal, and I 
guess they did not realise data race implications or thought they could be 
sorted in a reasonable way.  The avoidance of RMWs is explicitly mentioned 
in the preface to the Alpha ARM.

 And I guess you are aware that getting an asynchronous multi-bit ECC 
error interrupt for a partial write the origin of which has long gone and 
all you have is the physical address is also a horror to handle.

 Bad choice I guess anyway.  Too many guesses I guess too.

  Maciej

^ permalink raw reply	[flat|nested] 55+ messages in thread

end of thread, other threads:[~2024-07-04 22:21 UTC | newest]

Thread overview: 55+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-05-03  8:11 [PATCH 00/14] alpha: cleanups for 6.10 Arnd Bergmann
2024-05-03  8:11 ` [PATCH 01/14] alpha: sort scr_mem{cpy,move}w() out Arnd Bergmann
2024-05-03  8:11 ` [PATCH 02/14] alpha: fix modversions for strcpy() et.al Arnd Bergmann
2024-05-03  8:11 ` [PATCH 03/14] alpha: add clone3() support Arnd Bergmann
2024-05-03  8:11 ` [PATCH 04/14] alpha: don't make functions public without a reason Arnd Bergmann
2024-05-03  8:11 ` [PATCH 05/14] alpha: sys_sio: fix misspelled ifdefs Arnd Bergmann
2024-05-03  8:11 ` [PATCH 06/14] alpha: missing includes Arnd Bergmann
2024-05-03  8:11 ` [PATCH 07/14] alpha: core_lca: take the unused functions out Arnd Bergmann
2024-05-03  8:11 ` [PATCH 08/14] alpha: jensen, t2 - make __EXTERN_INLINE same as for the rest Arnd Bergmann
2024-05-03  8:11 ` [PATCH 09/14] alpha: trim the unused stuff from asm-offsets.c Arnd Bergmann
2024-05-03  8:11 ` [PATCH 10/14] alpha: remove DECpc AXP150 (Jensen) support Arnd Bergmann
2024-05-03 16:07   ` Linus Torvalds
2024-05-03 17:00   ` Al Viro
2024-05-03 20:07     ` Arnd Bergmann
2024-05-03  8:11 ` [PATCH 11/14] alpha: sable: remove early machine support Arnd Bergmann
2024-05-03  8:11 ` [PATCH 12/14] alpha: remove LCA and APECS based machines Arnd Bergmann
2024-05-03  8:11 ` [PATCH 13/14] alpha: cabriolet: remove EV5 CPU support Arnd Bergmann
2024-05-03  8:11 ` [PATCH 14/14] alpha: drop pre-EV56 support Arnd Bergmann
2024-05-04 15:00   ` Richard Henderson
2024-05-06 10:06     ` Arnd Bergmann
2024-06-03  6:02   ` Jiri Slaby
2024-06-04 13:58     ` Greg KH
2024-05-03 16:06 ` [PATCH 00/14] alpha: cleanups for 6.10 Matt Turner
2024-05-03 20:15   ` Arnd Bergmann
2024-05-06  9:16     ` Michael Cree
2024-05-06 10:11       ` Arnd Bergmann
2024-05-03 16:53 ` John Paul Adrian Glaubitz
2024-05-03 17:19   ` Paul E. McKenney
2024-05-27 23:49   ` Maciej W. Rozycki
2024-05-28 14:43     ` Paul E. McKenney
2024-05-29 18:50       ` Maciej W. Rozycki
2024-05-29 22:09         ` Paul E. McKenney
2024-05-30 22:59           ` Maciej W. Rozycki
2024-05-31  3:56           ` Maciej W. Rozycki
2024-05-31 19:33             ` Paul E. McKenney
2024-06-03 16:22               ` Maciej W. Rozycki
2024-06-03 17:08                 ` Paul E. McKenney
2024-07-01 23:50                   ` Maciej W. Rozycki
2024-05-30  1:08         ` Linus Torvalds
2024-05-30 22:57           ` Maciej W. Rozycki
2024-05-31  0:10             ` Linus Torvalds
2024-06-03 11:09               ` Maciej W. Rozycki
2024-06-03 11:36                 ` John Paul Adrian Glaubitz
2024-06-03 16:57                 ` Linus Torvalds
2024-07-01 23:48                   ` Maciej W. Rozycki
2024-05-31 15:48         ` Arnd Bergmann
2024-05-31 16:32           ` Linus Torvalds
2024-05-31 16:54             ` Arnd Bergmann
2024-06-01 13:51             ` David Laight
2024-07-01 23:48             ` Maciej W. Rozycki
2024-07-02  1:13               ` Linus Torvalds
2024-07-03  0:12                 ` Maciej W. Rozycki
2024-07-03  0:50                   ` Linus Torvalds
2024-07-04 22:21                     ` Maciej W. Rozycki
2024-06-03 11:33           ` Maciej W. Rozycki

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).