Commit d5902844 authored by Linus Torvalds's avatar Linus Torvalds
Browse files

Merge tag 's390-5.4-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull s390 updates from Vasily Gorbik:

 - Add support for IBM z15 machines.

 - Add SHA3 and CCA AES cipher key support in zcrypt and pkey
   refactoring.

 - Move to arch_stack_walk infrastructure for the stack unwinder.

 - Various kasan fixes and improvements.

 - Various command line parsing fixes.

 - Improve decompressor phase debuggability.

 - Lift no bss usage restriction for the early code.

 - Use refcount_t for reference counters for couple of places in mm
   code.

 - Logging improvements and return code fix in vfio-ccw code.

 - Couple of zpci fixes and minor refactoring.

 - Remove some outdated documentation.

 - Fix secure boot detection.

 - Other various minor code clean ups.

* tag 's390-5.4-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (48 commits)
  s390: remove pointless drivers-y in drivers/s390/Makefile
  s390/cpum_sf: Fix line length and format string
  s390/pci: fix MSI message data
  s390: add support for IBM z15 machines
  s390/crypto: Support for SHA3 via CPACF (MSA6)
  s390/startup: add pgm check info printing
  s390/crypto: xts-aes-s390 fix extra run-time crypto self tests finding
  vfio-ccw: fix error return code in vfio_ccw_sch_init()
  s390: vfio-ap: fix warning reset not completed
  s390/base: remove unused s390_base_mcck_handler
  s390/sclp: Fix bit checked for has_sipl
  s390/zcrypt: fix wrong handling of cca cipher keygenflags
  s390/kasan: add kdump support
  s390/setup: avoid using strncmp with hardcoded length
  s390/sclp: avoid using strncmp with hardcoded length
  s390/module: avoid using strncmp with hardcoded length
  s390/pci: avoid using strncmp with hardcoded length
  s390/kaslr: reserve memory for kasan usage
  s390/mem_detect: provide single get_mem_detect_end
  s390/cmma: reuse kstrtobool for option value parsing
  ...
parents 1e24aaab 2735913c
==================
DASD device driver
==================
S/390's disk devices (DASDs) are managed by Linux via the DASD device
driver. It is valid for all types of DASDs and represents them to
Linux as block devices, namely "dd". Currently the DASD driver uses a
single major number (254) and 4 minor numbers per volume (1 for the
physical volume and 3 for partitions). With respect to partitions see
below. Thus you may have up to 64 DASD devices in your system.
The kernel parameter 'dasd=from-to,...' may be issued arbitrary times
in the kernel's parameter line or not at all. The 'from' and 'to'
parameters are to be given in hexadecimal notation without a leading
0x.
If you supply kernel parameters the different instances are processed
in order of appearance and a minor number is reserved for any device
covered by the supplied range up to 64 volumes. Additional DASDs are
ignored. If you do not supply the 'dasd=' kernel parameter at all, the
DASD driver registers all supported DASDs of your system to a minor
number in ascending order of the subchannel number.
The driver currently supports ECKD-devices and there are stubs for
support of the FBA and CKD architectures. For the FBA architecture
only some smart data structures are missing to make the support
complete.
We performed our testing on 3380 and 3390 type disks of different
sizes, under VM and on the bare hardware (LPAR), using internal disks
of the multiprise as well as a RAMAC virtual array. Disks exported by
an Enterprise Storage Server (Seascape) should work fine as well.
We currently implement one partition per volume, which is the whole
volume, skipping the first blocks up to the volume label. These are
reserved for IPL records and IBM's volume label to assure
accessibility of the DASD from other OSs. In a later stage we will
provide support of partitions, maybe VTOC oriented or using a kind of
partition table in the label record.
Usage
=====
-Low-level format (?CKD only)
For using an ECKD-DASD as a Linux harddisk you have to low-level
format the tracks by issuing the BLKDASDFORMAT-ioctl on that
device. This will erase any data on that volume including IBM volume
labels, VTOCs etc. The ioctl may take a `struct format_data *` or
'NULL' as an argument::
typedef struct {
int start_unit;
int stop_unit;
int blksize;
} format_data_t;
When a NULL argument is passed to the BLKDASDFORMAT ioctl the whole
disk is formatted to a blocksize of 1024 bytes. Otherwise start_unit
and stop_unit are the first and last track to be formatted. If
stop_unit is -1 it implies that the DASD is formatted from start_unit
up to the last track. blksize can be any power of two between 512 and
4096. We recommend no blksize lower than 1024 because the ext2fs uses
1kB blocks anyway and you gain approx. 50% of capacity increasing your
blksize from 512 byte to 1kB.
Make a filesystem
=================
Then you can mk??fs the filesystem of your choice on that volume or
partition. For reasons of sanity you should build your filesystem on
the partition /dev/dd?1 instead of the whole volume. You only lose 3kB
but may be sure that you can reuse your data after introduction of a
real partition table.
Bugs
====
- Performance sometimes is rather low because we don't fully exploit clustering
TODO-List
=========
- Add IBM'S Disk layout to genhd
- Enhance driver to use more than one major number
- Enable usage as a module
- Support Cache fast write and DASD fast write (ECKD)
This diff is collapsed.
......@@ -7,7 +7,6 @@ s390 Architecture
cds
3270
debugging390
driver-model
monreader
qeth
......@@ -15,7 +14,6 @@ s390 Architecture
vfio-ap
vfio-ccw
zfcpdump
dasd
common_io
text_files
......
......@@ -105,6 +105,7 @@ config S390
select ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE
select ARCH_KEEP_MEMBLOCK
select ARCH_SAVE_PAGE_KEYS if HIBERNATION
select ARCH_STACKWALK
select ARCH_SUPPORTS_ATOMIC_RMW
select ARCH_SUPPORTS_NUMA_BALANCING
select ARCH_USE_BUILTIN_BSWAP
......@@ -236,6 +237,10 @@ config HAVE_MARCH_Z14_FEATURES
def_bool n
select HAVE_MARCH_Z13_FEATURES
config HAVE_MARCH_Z15_FEATURES
def_bool n
select HAVE_MARCH_Z14_FEATURES
choice
prompt "Processor type"
default MARCH_Z196
......@@ -307,6 +312,14 @@ config MARCH_Z14
and 3906 series). The kernel will be slightly faster but will not
work on older machines.
config MARCH_Z15
bool "IBM z15"
select HAVE_MARCH_Z15_FEATURES
help
Select this to enable optimizations for IBM z15 (8562
and 8561 series). The kernel will be slightly faster but will not
work on older machines.
endchoice
config MARCH_Z900_TUNE
......@@ -333,6 +346,9 @@ config MARCH_Z13_TUNE
config MARCH_Z14_TUNE
def_bool TUNE_Z14 || MARCH_Z14 && TUNE_DEFAULT
config MARCH_Z15_TUNE
def_bool TUNE_Z15 || MARCH_Z15 && TUNE_DEFAULT
choice
prompt "Tune code generation"
default TUNE_DEFAULT
......@@ -377,6 +393,9 @@ config TUNE_Z13
config TUNE_Z14
bool "IBM z14"
config TUNE_Z15
bool "IBM z15"
endchoice
config 64BIT
......
......@@ -45,6 +45,7 @@ mflags-$(CONFIG_MARCH_Z196) := -march=z196
mflags-$(CONFIG_MARCH_ZEC12) := -march=zEC12
mflags-$(CONFIG_MARCH_Z13) := -march=z13
mflags-$(CONFIG_MARCH_Z14) := -march=z14
mflags-$(CONFIG_MARCH_Z15) := -march=z15
export CC_FLAGS_MARCH := $(mflags-y)
......@@ -59,6 +60,7 @@ cflags-$(CONFIG_MARCH_Z196_TUNE) += -mtune=z196
cflags-$(CONFIG_MARCH_ZEC12_TUNE) += -mtune=zEC12
cflags-$(CONFIG_MARCH_Z13_TUNE) += -mtune=z13
cflags-$(CONFIG_MARCH_Z14_TUNE) += -mtune=z14
cflags-$(CONFIG_MARCH_Z15_TUNE) += -mtune=z15
cflags-y += -Wa,-I$(srctree)/arch/$(ARCH)/include
......
......@@ -36,7 +36,7 @@ CFLAGS_sclp_early_core.o += -I$(srctree)/drivers/s390/char
obj-y := head.o als.o startup.o mem_detect.o ipl_parm.o ipl_report.o
obj-y += string.o ebcdic.o sclp_early_core.o mem.o ipl_vmparm.o cmdline.o
obj-y += version.o ctype.o text_dma.o
obj-y += version.o pgm_check_info.o ctype.o text_dma.o
obj-$(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) += uv.o
obj-$(CONFIG_RELOCATABLE) += machine_kexec_reloc.o
obj-$(CONFIG_RANDOMIZE_BASE) += kaslr.o
......
......@@ -10,6 +10,7 @@ void parse_boot_command_line(void);
void setup_memory_end(void);
void verify_facilities(void);
void print_missing_facilities(void);
void print_pgm_check_info(void);
unsigned long get_random_base(unsigned long safe_addr);
extern int kaslr_enabled;
......
sizes.h
vmlinux
vmlinux.lds
vmlinux.scr.lds
vmlinux.bin.full
......@@ -37,9 +37,9 @@ SECTIONS
* .dma section for code, data, ex_table that need to stay below 2 GB,
* even when the kernel is relocate: above 2 GB.
*/
. = ALIGN(PAGE_SIZE);
_sdma = .;
.dma.text : {
. = ALIGN(PAGE_SIZE);
_stext_dma = .;
*(.dma.text)
. = ALIGN(PAGE_SIZE);
......@@ -52,6 +52,7 @@ SECTIONS
_stop_dma_ex_table = .;
}
.dma.data : { *(.dma.data) }
. = ALIGN(PAGE_SIZE);
_edma = .;
BOOT_DATA
......
......@@ -60,8 +60,10 @@ __HEAD
.long 0x02000690,0x60000050
.long 0x020006e0,0x20000050
.org 0x1a0
.org __LC_RST_NEW_PSW # 0x1a0
.quad 0,iplstart
.org __LC_PGM_NEW_PSW # 0x1d0
.quad 0x0000000180000000,startup_pgm_check_handler
.org 0x200
......@@ -351,6 +353,34 @@ ENTRY(startup_kdump)
#include "head_kdump.S"
#
# This program check is active immediately after kernel start
# and until early_pgm_check_handler is set in kernel/early.c
# It simply saves general/control registers and psw in
# the save area and does disabled wait with a faulty address.
#
ENTRY(startup_pgm_check_handler)
stmg %r0,%r15,__LC_SAVE_AREA_SYNC
la %r1,4095
stctg %c0,%c15,__LC_CREGS_SAVE_AREA-4095(%r1)
mvc __LC_GPREGS_SAVE_AREA-4095(128,%r1),__LC_SAVE_AREA_SYNC
mvc __LC_PSW_SAVE_AREA-4095(16,%r1),__LC_PGM_OLD_PSW
mvc __LC_RETURN_PSW(16),__LC_PGM_OLD_PSW
ni __LC_RETURN_PSW,0xfc # remove IO and EX bits
ni __LC_RETURN_PSW+1,0xfb # remove MCHK bit
oi __LC_RETURN_PSW+1,0x2 # set wait state bit
larl %r2,.Lold_psw_disabled_wait
stg %r2,__LC_PGM_NEW_PSW+8
l %r15,.Ldump_info_stack-.Lold_psw_disabled_wait(%r2)
brasl %r14,print_pgm_check_info
.Lold_psw_disabled_wait:
la %r1,4095
lmg %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1)
lpswe __LC_RETURN_PSW # disabled wait
.Ldump_info_stack:
.long 0x5000 + PAGE_SIZE - STACK_FRAME_OVERHEAD
ENDPROC(startup_pgm_check_handler)
#
# params at 10400 (setup.h)
# Must be keept in sync with struct parmarea in setup.h
......
......@@ -7,6 +7,7 @@
#include <asm/sections.h>
#include <asm/boot_data.h>
#include <asm/facility.h>
#include <asm/pgtable.h>
#include <asm/uv.h>
#include "boot.h"
......@@ -14,6 +15,7 @@ char __bootdata(early_command_line)[COMMAND_LINE_SIZE];
struct ipl_parameter_block __bootdata_preserved(ipl_block);
int __bootdata_preserved(ipl_block_valid);
unsigned long __bootdata(vmalloc_size) = VMALLOC_DEFAULT_SIZE;
unsigned long __bootdata(memory_end);
int __bootdata(memory_end_set);
int __bootdata(noexec_disabled);
......@@ -219,18 +221,21 @@ void parse_boot_command_line(void)
while (*args) {
args = next_arg(args, &param, &val);
if (!strcmp(param, "mem")) {
memory_end = memparse(val, NULL);
if (!strcmp(param, "mem") && val) {
memory_end = round_down(memparse(val, NULL), PAGE_SIZE);
memory_end_set = 1;
}
if (!strcmp(param, "vmalloc") && val)
vmalloc_size = round_up(memparse(val, NULL), PAGE_SIZE);
if (!strcmp(param, "noexec")) {
rc = kstrtobool(val, &enabled);
if (!rc && !enabled)
noexec_disabled = 1;
}
if (!strcmp(param, "facilities"))
if (!strcmp(param, "facilities") && val)
modify_fac_list(val);
if (!strcmp(param, "nokaslr"))
......
......@@ -3,6 +3,7 @@
* Copyright IBM Corp. 2019
*/
#include <asm/mem_detect.h>
#include <asm/pgtable.h>
#include <asm/cpacf.h>
#include <asm/timex.h>
#include <asm/sclp.h>
......@@ -90,8 +91,10 @@ static unsigned long get_random(unsigned long limit)
unsigned long get_random_base(unsigned long safe_addr)
{
unsigned long memory_limit = memory_end_set ? memory_end : 0;
unsigned long base, start, end, kernel_size;
unsigned long block_sum, offset;
unsigned long kasan_needs;
int i;
if (IS_ENABLED(CONFIG_BLK_DEV_INITRD) && INITRD_START && INITRD_SIZE) {
......@@ -100,14 +103,36 @@ unsigned long get_random_base(unsigned long safe_addr)
}
safe_addr = ALIGN(safe_addr, THREAD_SIZE);
if ((IS_ENABLED(CONFIG_KASAN))) {
/*
* Estimate kasan memory requirements, which it will reserve
* at the very end of available physical memory. To estimate
* that, we take into account that kasan would require
* 1/8 of available physical memory (for shadow memory) +
* creating page tables for the whole memory + shadow memory
* region (1 + 1/8). To keep page tables estimates simple take
* the double of combined ptes size.
*/
memory_limit = get_mem_detect_end();
if (memory_end_set && memory_limit > memory_end)
memory_limit = memory_end;
/* for shadow memory */
kasan_needs = memory_limit / 8;
/* for paging structures */
kasan_needs += (memory_limit + kasan_needs) / PAGE_SIZE /
_PAGE_ENTRIES * _PAGE_TABLE_SIZE * 2;
memory_limit -= kasan_needs;
}
kernel_size = vmlinux.image_size + vmlinux.bss_size;
block_sum = 0;
for_each_mem_detect_block(i, &start, &end) {
if (memory_end_set) {
if (start >= memory_end)
if (memory_limit) {
if (start >= memory_limit)
break;
if (end > memory_end)
end = memory_end;
if (end > memory_limit)
end = memory_limit;
}
if (end - start < kernel_size)
continue;
......@@ -125,11 +150,11 @@ unsigned long get_random_base(unsigned long safe_addr)
base = safe_addr;
block_sum = offset = 0;
for_each_mem_detect_block(i, &start, &end) {
if (memory_end_set) {
if (start >= memory_end)
if (memory_limit) {
if (start >= memory_limit)
break;
if (end > memory_end)
end = memory_end;
if (end > memory_limit)
end = memory_limit;
}
if (end - start < kernel_size)
continue;
......
......@@ -63,13 +63,6 @@ void add_mem_detect_block(u64 start, u64 end)
mem_detect.count++;
}
static unsigned long get_mem_detect_end(void)
{
if (mem_detect.count)
return __get_mem_detect_block_ptr(mem_detect.count - 1)->end;
return 0;
}
static int __diag260(unsigned long rx1, unsigned long rx2)
{
register unsigned long _rx1 asm("2") = rx1;
......
// SPDX-License-Identifier: GPL-2.0
#include <linux/kernel.h>
#include <linux/string.h>
#include <asm/lowcore.h>
#include <asm/sclp.h>
#include "boot.h"
const char hex_asc[] = "0123456789abcdef";
#define add_val_as_hex(dst, val) \
__add_val_as_hex(dst, (const unsigned char *)&val, sizeof(val))
static char *__add_val_as_hex(char *dst, const unsigned char *src, size_t count)
{
while (count--)
dst = hex_byte_pack(dst, *src++);
return dst;
}
static char *add_str(char *dst, char *src)
{
strcpy(dst, src);
return dst + strlen(dst);
}
void print_pgm_check_info(void)
{
struct psw_bits *psw = &psw_bits(S390_lowcore.psw_save_area);
unsigned short ilc = S390_lowcore.pgm_ilc >> 1;
char buf[256];
int row, col;
char *p;
add_str(buf, "Linux version ");
strlcat(buf, kernel_version, sizeof(buf));
sclp_early_printk(buf);
p = add_str(buf, "Kernel fault: interruption code ");
p = add_val_as_hex(buf + strlen(buf), S390_lowcore.pgm_code);
p = add_str(p, " ilc:");
*p++ = hex_asc_lo(ilc);
add_str(p, "\n");
sclp_early_printk(buf);
p = add_str(buf, "PSW : ");
p = add_val_as_hex(p, S390_lowcore.psw_save_area.mask);
p = add_str(p, " ");
p = add_val_as_hex(p, S390_lowcore.psw_save_area.addr);
add_str(p, "\n");
sclp_early_printk(buf);
p = add_str(buf, " R:");
*p++ = hex_asc_lo(psw->per);
p = add_str(p, " T:");
*p++ = hex_asc_lo(psw->dat);
p = add_str(p, " IO:");
*p++ = hex_asc_lo(psw->io);
p = add_str(p, " EX:");
*p++ = hex_asc_lo(psw->ext);
p = add_str(p, " Key:");
*p++ = hex_asc_lo(psw->key);
p = add_str(p, " M:");
*p++ = hex_asc_lo(psw->mcheck);
p = add_str(p, " W:");
*p++ = hex_asc_lo(psw->wait);
p = add_str(p, " P:");
*p++ = hex_asc_lo(psw->pstate);
p = add_str(p, " AS:");
*p++ = hex_asc_lo(psw->as);
p = add_str(p, " CC:");
*p++ = hex_asc_lo(psw->cc);
p = add_str(p, " PM:");
*p++ = hex_asc_lo(psw->pm);
p = add_str(p, " RI:");
*p++ = hex_asc_lo(psw->ri);
p = add_str(p, " EA:");
*p++ = hex_asc_lo(psw->eaba);
add_str(p, "\n");
sclp_early_printk(buf);
for (row = 0; row < 4; row++) {
p = add_str(buf, row == 0 ? "GPRS:" : " ");
for (col = 0; col < 4; col++) {
p = add_str(p, " ");
p = add_val_as_hex(p, S390_lowcore.gpregs_save_area[row * 4 + col]);
}
add_str(p, "\n");
sclp_early_printk(buf);
}
}
......@@ -112,6 +112,11 @@ static void handle_relocs(unsigned long offset)
}
}
static void clear_bss_section(void)
{
memset((void *)vmlinux.default_lma + vmlinux.image_size, 0, vmlinux.bss_size);
}
void startup_kernel(void)
{
unsigned long random_lma;
......@@ -151,6 +156,7 @@ void startup_kernel(void)
} else if (__kaslr_offset)
memcpy((void *)vmlinux.default_lma, img, vmlinux.image_size);
clear_bss_section();
copy_bootdata();
if (IS_ENABLED(CONFIG_RELOCATABLE))
handle_relocs(__kaslr_offset);
......
......@@ -717,6 +717,8 @@ CONFIG_CRYPTO_PAES_S390=m
CONFIG_CRYPTO_SHA1_S390=m
CONFIG_CRYPTO_SHA256_S390=m
CONFIG_CRYPTO_SHA512_S390=m
CONFIG_CRYPTO_SHA3_256_S390=m
CONFIG_CRYPTO_SHA3_512_S390=m
CONFIG_CRYPTO_DES_S390=m
CONFIG_CRYPTO_AES_S390=m
CONFIG_CRYPTO_GHASH_S390=m
......
......@@ -710,6 +710,8 @@ CONFIG_CRYPTO_PAES_S390=m
CONFIG_CRYPTO_SHA1_S390=m
CONFIG_CRYPTO_SHA256_S390=m
CONFIG_CRYPTO_SHA512_S390=m
CONFIG_CRYPTO_SHA3_256_S390=m
CONFIG_CRYPTO_SHA3_512_S390=m
CONFIG_CRYPTO_DES_S390=m
CONFIG_CRYPTO_AES_S390=m
CONFIG_CRYPTO_GHASH_S390=m
......
......@@ -6,6 +6,8 @@
obj-$(CONFIG_CRYPTO_SHA1_S390) += sha1_s390.o sha_common.o
obj-$(CONFIG_CRYPTO_SHA256_S390) += sha256_s390.o sha_common.o
obj-$(CONFIG_CRYPTO_SHA512_S390) += sha512_s390.o sha_common.o
obj-$(CONFIG_CRYPTO_SHA3_256_S390) += sha3_256_s390.o sha_common.o
obj-$(CONFIG_CRYPTO_SHA3_512_S390) += sha3_512_s390.o sha_common.o
obj-$(CONFIG_CRYPTO_DES_S390) += des_s390.o
obj-$(CONFIG_CRYPTO_AES_S390) += aes_s390.o
obj-$(CONFIG_CRYPTO_PAES_S390) += paes_s390.o
......
......@@ -586,6 +586,9 @@ static int xts_aes_encrypt(struct blkcipher_desc *desc,
struct s390_xts_ctx *xts_ctx = crypto_blkcipher_ctx(desc->tfm);
struct blkcipher_walk walk;
if (!nbytes)
return -EINVAL;
if (unlikely(!xts_ctx->fc))
return xts_fallback_encrypt(desc, dst, src, nbytes);
......@@ -600,6 +603,9 @@ static int xts_aes_decrypt(struct blkcipher_desc *desc,
struct s390_xts_ctx *xts_ctx = crypto_blkcipher_ctx(desc->tfm);
struct blkcipher_walk walk;
if (!nbytes)
return -EINVAL;
if (unlikely(!xts_ctx->fc))
return xts_fallback_decrypt(desc, dst, src, nbytes);
......
......@@ -5,7 +5,7 @@
* s390 implementation of the AES Cipher Algorithm with protected keys.
*
* s390 Version:
* Copyright IBM Corp. 2017
* Copyright IBM Corp. 2017,2019
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
* Harald Freudenberger <freude@de.ibm.com>
*/
......@@ -25,16 +25,59 @@
#include <asm/cpacf.h>
#include <asm/pkey.h>
/*
* Key blobs smaller/bigger than these defines are rejected
* by the common code even before the individual setkey function
* is called. As paes can handle different kinds of key blobs
* and padding is also possible, the limits need to be generous.
*/
#define PAES_MIN_KEYSIZE 64
#define PAES_MAX_KEYSIZE 256
static u8 *ctrblk;
static DEFINE_SPINLOCK(ctrblk_lock);
static cpacf_mask_t km_functions, kmc_functions, kmctr_functions;
struct key_blob {
__u8 key[MAXKEYBLOBSIZE];
/*
* Small keys will be stored in the keybuf. Larger keys are
* stored in extra allocated memory. In both cases does
* key point to the memory where the key is stored.
* The code distinguishes by checking keylen against
* sizeof(keybuf). See the two following helper functions.
*/
u8 *key;
u8 keybuf[128];
unsigned int keylen;
};
static inline int _copy_key_to_kb(struct key_blob *kb,
const u8 *key,
unsigned int keylen)
{
if (keylen <= sizeof(kb->keybuf))
kb->key = kb->keybuf;
else {
kb->key = kmalloc(keylen, GFP_KERNEL);
if (!kb->key)
return -ENOMEM;
}
memcpy(kb->key, key, keylen);
kb->keylen = keylen;
return 0;
}
static inline void _free_kb_keybuf(struct key_blob *kb)
{
if (kb->key && kb->key != kb->keybuf
&& kb->keylen > sizeof(kb->keybuf)) {