git-commit-vandalism/rerere.c

756 lines
18 KiB
C
Raw Normal View History

#include "cache.h"
#include "lockfile.h"
#include "string-list.h"
#include "rerere.h"
#include "xdiff-interface.h"
#include "dir.h"
#include "resolve-undo.h"
#include "ll-merge.h"
#include "attr.h"
#include "pathspec.h"
#define RESOLVED 0
#define PUNTED 1
#define THREE_STAGED 2
void *RERERE_RESOLVED = &RERERE_RESOLVED;
/* if rerere_enabled == -1, fall back to detection of .git/rr-cache */
static int rerere_enabled = -1;
/* automatically update cleanly resolved paths to the index */
static int rerere_autoupdate;
const char *rerere_path(const char *hex, const char *file)
{
return git_path("rr-cache/%s/%s", hex, file);
}
static int has_rerere_resolution(const char *hex)
{
struct stat st;
return !stat(rerere_path(hex, "postimage"), &st);
}
static void read_rr(struct string_list *rr)
{
unsigned char sha1[20];
char buf[PATH_MAX];
memoize common git-path "constant" files One of the most common uses of git_path() is to pass a constant, like git_path("MERGE_MSG"). This has two drawbacks: 1. The return value is a static buffer, and the lifetime is dependent on other calls to git_path, etc. 2. There's no compile-time checking of the pathname. This is OK for a one-off (after all, we have to spell it correctly at least once), but many of these constant strings appear throughout the code. This patch introduces a series of functions to "memoize" these strings, which are essentially globals for the lifetime of the program. We compute the value once, take ownership of the buffer, and return the cached value for subsequent calls. cache.h provides a helper macro for defining these functions as one-liners, and defines a few common ones for global use. Using a macro is a little bit gross, but it does nicely document the purpose of the functions. If we need to touch them all later (e.g., because we learned how to change the git_dir variable at runtime, and need to invalidate all of the stored values), it will be much easier to have the complete list. Note that the shared-global functions have separate, manual declarations. We could do something clever with the macros (e.g., expand it to a declaration in some places, and a declaration _and_ a definition in path.c). But there aren't that many, and it's probably better to stay away from too-magical macros. Likewise, if we abandon the C preprocessor in favor of generating these with a script, we could get much fancier. E.g., normalizing "FOO/BAR-BAZ" into "git_path_foo_bar_baz". But the small amount of saved typing is probably not worth the resulting confusion to readers who want to grep for the function's definition. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-08-10 11:38:57 +02:00
FILE *in = fopen(git_path_merge_rr(), "r");
if (!in)
return;
while (fread(buf, 40, 1, in) == 1) {
int i;
char *name;
if (get_sha1_hex(buf, sha1))
die("corrupt MERGE_RR");
buf[40] = '\0';
name = xstrdup(buf);
if (fgetc(in) != '\t')
die("corrupt MERGE_RR");
for (i = 0; i < sizeof(buf); i++) {
int c = fgetc(in);
if (c < 0)
die("corrupt MERGE_RR");
buf[i] = c;
if (c == 0)
break;
}
if (i == sizeof(buf))
die("filename too long");
string_list_insert(rr, buf)->util = name;
}
fclose(in);
}
static struct lock_file write_lock;
static int write_rr(struct string_list *rr, int out_fd)
{
int i;
for (i = 0; i < rr->nr; i++) {
const char *path;
int length;
if (!rr->items[i].util)
continue;
path = rr->items[i].string;
length = strlen(path) + 1;
if (write_in_full(out_fd, rr->items[i].util, 40) != 40 ||
write_str_in_full(out_fd, "\t") != 1 ||
write_in_full(out_fd, path, length) != length)
die("unable to write rerere record");
}
if (commit_lock_file(&write_lock) != 0)
die("unable to write rerere record");
return 0;
}
static void ferr_write(const void *p, size_t count, FILE *fp, int *err)
{
if (!count || *err)
return;
if (fwrite(p, count, 1, fp) != 1)
*err = errno;
}
static inline void ferr_puts(const char *s, FILE *fp, int *err)
{
ferr_write(s, strlen(s), fp, err);
}
struct rerere_io {
int (*getline)(struct strbuf *, struct rerere_io *);
FILE *output;
int wrerror;
/* some more stuff */
};
static void rerere_io_putstr(const char *str, struct rerere_io *io)
{
if (io->output)
ferr_puts(str, io->output, &io->wrerror);
}
static void rerere_io_putconflict(int ch, int size, struct rerere_io *io)
{
char buf[64];
while (size) {
if (size < sizeof(buf) - 2) {
memset(buf, ch, size);
buf[size] = '\n';
buf[size + 1] = '\0';
size = 0;
} else {
int sz = sizeof(buf) - 1;
if (size <= sz)
sz -= (sz - size) + 1;
memset(buf, ch, sz);
buf[sz] = '\0';
size -= sz;
}
rerere_io_putstr(buf, io);
}
}
static void rerere_io_putmem(const char *mem, size_t sz, struct rerere_io *io)
{
if (io->output)
ferr_write(mem, sz, io->output, &io->wrerror);
}
struct rerere_io_file {
struct rerere_io io;
FILE *input;
};
static int rerere_file_getline(struct strbuf *sb, struct rerere_io *io_)
{
struct rerere_io_file *io = (struct rerere_io_file *)io_;
return strbuf_getwholeline(sb, io->input, '\n');
}
static int is_cmarker(char *buf, int marker_char, int marker_size, int want_sp)
{
while (marker_size--)
if (*buf++ != marker_char)
return 0;
if (want_sp && *buf != ' ')
return 0;
return isspace(*buf);
}
static int handle_path(unsigned char *sha1, struct rerere_io *io, int marker_size)
{
fix openssl headers conflicting with custom SHA1 implementations On ARM I have the following compilation errors: CC fast-import.o In file included from cache.h:8, from builtin.h:6, from fast-import.c:142: arm/sha1.h:14: error: conflicting types for 'SHA_CTX' /usr/include/openssl/sha.h:105: error: previous declaration of 'SHA_CTX' was here arm/sha1.h:16: error: conflicting types for 'SHA1_Init' /usr/include/openssl/sha.h:115: error: previous declaration of 'SHA1_Init' was here arm/sha1.h:17: error: conflicting types for 'SHA1_Update' /usr/include/openssl/sha.h:116: error: previous declaration of 'SHA1_Update' was here arm/sha1.h:18: error: conflicting types for 'SHA1_Final' /usr/include/openssl/sha.h:117: error: previous declaration of 'SHA1_Final' was here make: *** [fast-import.o] Error 1 This is because openssl header files are always included in git-compat-util.h since commit 684ec6c63c whenever NO_OPENSSL is not set, which somehow brings in <openssl/sha1.h> clashing with the custom ARM version. Compilation of git is probably broken on PPC too for the same reason. Turns out that the only file requiring openssl/ssl.h and openssl/err.h is imap-send.c. But only moving those problematic includes there doesn't solve the issue as it also includes cache.h which brings in the conflicting local SHA1 header file. As suggested by Jeff King, the best solution is to rename our references to SHA1 functions and structure to something git specific, and define those according to the implementation used. Signed-off-by: Nicolas Pitre <nico@cam.org> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
2008-10-01 20:05:20 +02:00
git_SHA_CTX ctx;
int hunk_no = 0;
enum {
RR_CONTEXT = 0, RR_SIDE_1, RR_SIDE_2, RR_ORIGINAL
} hunk = RR_CONTEXT;
struct strbuf one = STRBUF_INIT, two = STRBUF_INIT;
struct strbuf buf = STRBUF_INIT;
if (sha1)
fix openssl headers conflicting with custom SHA1 implementations On ARM I have the following compilation errors: CC fast-import.o In file included from cache.h:8, from builtin.h:6, from fast-import.c:142: arm/sha1.h:14: error: conflicting types for 'SHA_CTX' /usr/include/openssl/sha.h:105: error: previous declaration of 'SHA_CTX' was here arm/sha1.h:16: error: conflicting types for 'SHA1_Init' /usr/include/openssl/sha.h:115: error: previous declaration of 'SHA1_Init' was here arm/sha1.h:17: error: conflicting types for 'SHA1_Update' /usr/include/openssl/sha.h:116: error: previous declaration of 'SHA1_Update' was here arm/sha1.h:18: error: conflicting types for 'SHA1_Final' /usr/include/openssl/sha.h:117: error: previous declaration of 'SHA1_Final' was here make: *** [fast-import.o] Error 1 This is because openssl header files are always included in git-compat-util.h since commit 684ec6c63c whenever NO_OPENSSL is not set, which somehow brings in <openssl/sha1.h> clashing with the custom ARM version. Compilation of git is probably broken on PPC too for the same reason. Turns out that the only file requiring openssl/ssl.h and openssl/err.h is imap-send.c. But only moving those problematic includes there doesn't solve the issue as it also includes cache.h which brings in the conflicting local SHA1 header file. As suggested by Jeff King, the best solution is to rename our references to SHA1 functions and structure to something git specific, and define those according to the implementation used. Signed-off-by: Nicolas Pitre <nico@cam.org> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
2008-10-01 20:05:20 +02:00
git_SHA1_Init(&ctx);
while (!io->getline(&buf, io)) {
if (is_cmarker(buf.buf, '<', marker_size, 1)) {
if (hunk != RR_CONTEXT)
goto bad;
hunk = RR_SIDE_1;
} else if (is_cmarker(buf.buf, '|', marker_size, 0)) {
if (hunk != RR_SIDE_1)
goto bad;
hunk = RR_ORIGINAL;
} else if (is_cmarker(buf.buf, '=', marker_size, 0)) {
if (hunk != RR_SIDE_1 && hunk != RR_ORIGINAL)
goto bad;
hunk = RR_SIDE_2;
} else if (is_cmarker(buf.buf, '>', marker_size, 1)) {
if (hunk != RR_SIDE_2)
goto bad;
if (strbuf_cmp(&one, &two) > 0)
strbuf_swap(&one, &two);
hunk_no++;
hunk = RR_CONTEXT;
rerere_io_putconflict('<', marker_size, io);
rerere_io_putmem(one.buf, one.len, io);
rerere_io_putconflict('=', marker_size, io);
rerere_io_putmem(two.buf, two.len, io);
rerere_io_putconflict('>', marker_size, io);
if (sha1) {
fix openssl headers conflicting with custom SHA1 implementations On ARM I have the following compilation errors: CC fast-import.o In file included from cache.h:8, from builtin.h:6, from fast-import.c:142: arm/sha1.h:14: error: conflicting types for 'SHA_CTX' /usr/include/openssl/sha.h:105: error: previous declaration of 'SHA_CTX' was here arm/sha1.h:16: error: conflicting types for 'SHA1_Init' /usr/include/openssl/sha.h:115: error: previous declaration of 'SHA1_Init' was here arm/sha1.h:17: error: conflicting types for 'SHA1_Update' /usr/include/openssl/sha.h:116: error: previous declaration of 'SHA1_Update' was here arm/sha1.h:18: error: conflicting types for 'SHA1_Final' /usr/include/openssl/sha.h:117: error: previous declaration of 'SHA1_Final' was here make: *** [fast-import.o] Error 1 This is because openssl header files are always included in git-compat-util.h since commit 684ec6c63c whenever NO_OPENSSL is not set, which somehow brings in <openssl/sha1.h> clashing with the custom ARM version. Compilation of git is probably broken on PPC too for the same reason. Turns out that the only file requiring openssl/ssl.h and openssl/err.h is imap-send.c. But only moving those problematic includes there doesn't solve the issue as it also includes cache.h which brings in the conflicting local SHA1 header file. As suggested by Jeff King, the best solution is to rename our references to SHA1 functions and structure to something git specific, and define those according to the implementation used. Signed-off-by: Nicolas Pitre <nico@cam.org> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
2008-10-01 20:05:20 +02:00
git_SHA1_Update(&ctx, one.buf ? one.buf : "",
one.len + 1);
fix openssl headers conflicting with custom SHA1 implementations On ARM I have the following compilation errors: CC fast-import.o In file included from cache.h:8, from builtin.h:6, from fast-import.c:142: arm/sha1.h:14: error: conflicting types for 'SHA_CTX' /usr/include/openssl/sha.h:105: error: previous declaration of 'SHA_CTX' was here arm/sha1.h:16: error: conflicting types for 'SHA1_Init' /usr/include/openssl/sha.h:115: error: previous declaration of 'SHA1_Init' was here arm/sha1.h:17: error: conflicting types for 'SHA1_Update' /usr/include/openssl/sha.h:116: error: previous declaration of 'SHA1_Update' was here arm/sha1.h:18: error: conflicting types for 'SHA1_Final' /usr/include/openssl/sha.h:117: error: previous declaration of 'SHA1_Final' was here make: *** [fast-import.o] Error 1 This is because openssl header files are always included in git-compat-util.h since commit 684ec6c63c whenever NO_OPENSSL is not set, which somehow brings in <openssl/sha1.h> clashing with the custom ARM version. Compilation of git is probably broken on PPC too for the same reason. Turns out that the only file requiring openssl/ssl.h and openssl/err.h is imap-send.c. But only moving those problematic includes there doesn't solve the issue as it also includes cache.h which brings in the conflicting local SHA1 header file. As suggested by Jeff King, the best solution is to rename our references to SHA1 functions and structure to something git specific, and define those according to the implementation used. Signed-off-by: Nicolas Pitre <nico@cam.org> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
2008-10-01 20:05:20 +02:00
git_SHA1_Update(&ctx, two.buf ? two.buf : "",
two.len + 1);
}
strbuf_reset(&one);
strbuf_reset(&two);
} else if (hunk == RR_SIDE_1)
strbuf_addbuf(&one, &buf);
else if (hunk == RR_ORIGINAL)
; /* discard */
else if (hunk == RR_SIDE_2)
strbuf_addbuf(&two, &buf);
else
rerere_io_putstr(buf.buf, io);
continue;
bad:
hunk = 99; /* force error exit */
break;
}
strbuf_release(&one);
strbuf_release(&two);
strbuf_release(&buf);
if (sha1)
fix openssl headers conflicting with custom SHA1 implementations On ARM I have the following compilation errors: CC fast-import.o In file included from cache.h:8, from builtin.h:6, from fast-import.c:142: arm/sha1.h:14: error: conflicting types for 'SHA_CTX' /usr/include/openssl/sha.h:105: error: previous declaration of 'SHA_CTX' was here arm/sha1.h:16: error: conflicting types for 'SHA1_Init' /usr/include/openssl/sha.h:115: error: previous declaration of 'SHA1_Init' was here arm/sha1.h:17: error: conflicting types for 'SHA1_Update' /usr/include/openssl/sha.h:116: error: previous declaration of 'SHA1_Update' was here arm/sha1.h:18: error: conflicting types for 'SHA1_Final' /usr/include/openssl/sha.h:117: error: previous declaration of 'SHA1_Final' was here make: *** [fast-import.o] Error 1 This is because openssl header files are always included in git-compat-util.h since commit 684ec6c63c whenever NO_OPENSSL is not set, which somehow brings in <openssl/sha1.h> clashing with the custom ARM version. Compilation of git is probably broken on PPC too for the same reason. Turns out that the only file requiring openssl/ssl.h and openssl/err.h is imap-send.c. But only moving those problematic includes there doesn't solve the issue as it also includes cache.h which brings in the conflicting local SHA1 header file. As suggested by Jeff King, the best solution is to rename our references to SHA1 functions and structure to something git specific, and define those according to the implementation used. Signed-off-by: Nicolas Pitre <nico@cam.org> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
2008-10-01 20:05:20 +02:00
git_SHA1_Final(sha1, &ctx);
if (hunk != RR_CONTEXT)
return -1;
return hunk_no;
}
static int handle_file(const char *path, unsigned char *sha1, const char *output)
{
int hunk_no = 0;
struct rerere_io_file io;
int marker_size = ll_merge_marker_size(path);
memset(&io, 0, sizeof(io));
io.io.getline = rerere_file_getline;
io.input = fopen(path, "r");
io.io.wrerror = 0;
if (!io.input)
return error("Could not open %s", path);
if (output) {
io.io.output = fopen(output, "w");
if (!io.io.output) {
fclose(io.input);
return error("Could not write %s", output);
}
}
hunk_no = handle_path(sha1, (struct rerere_io *)&io, marker_size);
fclose(io.input);
if (io.io.wrerror)
error("There were errors while writing %s (%s)",
path, strerror(io.io.wrerror));
if (io.io.output && fclose(io.io.output))
io.io.wrerror = error("Failed to flush %s: %s",
path, strerror(errno));
if (hunk_no < 0) {
if (output)
unlink_or_warn(output);
return error("Could not parse conflict hunks in %s", path);
}
if (io.io.wrerror)
return -1;
return hunk_no;
}
struct rerere_io_mem {
struct rerere_io io;
struct strbuf input;
};
static int rerere_mem_getline(struct strbuf *sb, struct rerere_io *io_)
{
struct rerere_io_mem *io = (struct rerere_io_mem *)io_;
char *ep;
size_t len;
strbuf_release(sb);
if (!io->input.len)
return -1;
ep = memchr(io->input.buf, '\n', io->input.len);
if (!ep)
ep = io->input.buf + io->input.len;
else if (*ep == '\n')
ep++;
len = ep - io->input.buf;
strbuf_add(sb, io->input.buf, len);
strbuf_remove(&io->input, 0, len);
return 0;
}
static int handle_cache(const char *path, unsigned char *sha1, const char *output)
{
mmfile_t mmfile[3] = {{NULL}};
mmbuffer_t result = {NULL, 0};
Convert "struct cache_entry *" to "const ..." wherever possible I attempted to make index_state->cache[] a "const struct cache_entry **" to find out how existing entries in index are modified and where. The question I have is what do we do if we really need to keep track of on-disk changes in the index. The result is - diff-lib.c: setting CE_UPTODATE - name-hash.c: setting CE_HASHED - preload-index.c, read-cache.c, unpack-trees.c and builtin/update-index: obvious - entry.c: write_entry() may refresh the checked out entry via fill_stat_cache_info(). This causes "non-const struct cache_entry *" in builtin/apply.c, builtin/checkout-index.c and builtin/checkout.c - builtin/ls-files.c: --with-tree changes stagemask and may set CE_UPDATE Of these, write_entry() and its call sites are probably most interesting because it modifies on-disk info. But this is stat info and can be retrieved via refresh, at least for porcelain commands. Other just uses ce_flags for local purposes. So, keeping track of "dirty" entries is just a matter of setting a flag in index modification functions exposed by read-cache.c. Except unpack-trees, the rest of the code base does not do anything funny behind read-cache's back. The actual patch is less valueable than the summary above. But if anyone wants to re-identify the above sites. Applying this patch, then this: diff --git a/cache.h b/cache.h index 430d021..1692891 100644 --- a/cache.h +++ b/cache.h @@ -267,7 +267,7 @@ static inline unsigned int canon_mode(unsigned int mode) #define cache_entry_size(len) (offsetof(struct cache_entry,name) + (len) + 1) struct index_state { - struct cache_entry **cache; + const struct cache_entry **cache; unsigned int version; unsigned int cache_nr, cache_alloc, cache_changed; struct string_list *resolve_undo; will help quickly identify them without bogus warnings. Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2013-07-09 17:29:00 +02:00
const struct cache_entry *ce;
int pos, len, i, hunk_no;
struct rerere_io_mem io;
int marker_size = ll_merge_marker_size(path);
/*
* Reproduce the conflicted merge in-core
*/
len = strlen(path);
pos = cache_name_pos(path, len);
if (0 <= pos)
return -1;
pos = -pos - 1;
for (i = 0; i < 3; i++) {
enum object_type type;
unsigned long size;
int j;
if (active_nr <= pos)
break;
ce = active_cache[pos++];
if (ce_namelen(ce) != len || memcmp(ce->name, path, len))
continue;
j = ce_stage(ce) - 1;
mmfile[j].ptr = read_sha1_file(ce->sha1, &type, &size);
mmfile[j].size = size;
}
for (i = 0; i < 3; i++) {
if (!mmfile[i].ptr && !mmfile[i].size)
mmfile[i].ptr = xstrdup("");
}
/*
* NEEDSWORK: handle conflicts from merges with
* merge.renormalize set, too
*/
ll_merge(&result, path, &mmfile[0], NULL,
&mmfile[1], "ours",
&mmfile[2], "theirs", NULL);
for (i = 0; i < 3; i++)
free(mmfile[i].ptr);
memset(&io, 0, sizeof(io));
io.io.getline = rerere_mem_getline;
if (output)
io.io.output = fopen(output, "w");
else
io.io.output = NULL;
strbuf_init(&io.input, 0);
strbuf_attach(&io.input, result.ptr, result.size, result.size);
hunk_no = handle_path(sha1, (struct rerere_io *)&io, marker_size);
strbuf_release(&io.input);
if (io.io.output)
fclose(io.io.output);
return hunk_no;
}
static int check_one_conflict(int i, int *type)
{
Convert "struct cache_entry *" to "const ..." wherever possible I attempted to make index_state->cache[] a "const struct cache_entry **" to find out how existing entries in index are modified and where. The question I have is what do we do if we really need to keep track of on-disk changes in the index. The result is - diff-lib.c: setting CE_UPTODATE - name-hash.c: setting CE_HASHED - preload-index.c, read-cache.c, unpack-trees.c and builtin/update-index: obvious - entry.c: write_entry() may refresh the checked out entry via fill_stat_cache_info(). This causes "non-const struct cache_entry *" in builtin/apply.c, builtin/checkout-index.c and builtin/checkout.c - builtin/ls-files.c: --with-tree changes stagemask and may set CE_UPDATE Of these, write_entry() and its call sites are probably most interesting because it modifies on-disk info. But this is stat info and can be retrieved via refresh, at least for porcelain commands. Other just uses ce_flags for local purposes. So, keeping track of "dirty" entries is just a matter of setting a flag in index modification functions exposed by read-cache.c. Except unpack-trees, the rest of the code base does not do anything funny behind read-cache's back. The actual patch is less valueable than the summary above. But if anyone wants to re-identify the above sites. Applying this patch, then this: diff --git a/cache.h b/cache.h index 430d021..1692891 100644 --- a/cache.h +++ b/cache.h @@ -267,7 +267,7 @@ static inline unsigned int canon_mode(unsigned int mode) #define cache_entry_size(len) (offsetof(struct cache_entry,name) + (len) + 1) struct index_state { - struct cache_entry **cache; + const struct cache_entry **cache; unsigned int version; unsigned int cache_nr, cache_alloc, cache_changed; struct string_list *resolve_undo; will help quickly identify them without bogus warnings. Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2013-07-09 17:29:00 +02:00
const struct cache_entry *e = active_cache[i];
if (!ce_stage(e)) {
*type = RESOLVED;
return i + 1;
}
*type = PUNTED;
if (ce_stage(e) == 1) {
if (active_nr <= ++i)
return i + 1;
}
/* Only handle regular files with both stages #2 and #3 */
if (i + 1 < active_nr) {
Convert "struct cache_entry *" to "const ..." wherever possible I attempted to make index_state->cache[] a "const struct cache_entry **" to find out how existing entries in index are modified and where. The question I have is what do we do if we really need to keep track of on-disk changes in the index. The result is - diff-lib.c: setting CE_UPTODATE - name-hash.c: setting CE_HASHED - preload-index.c, read-cache.c, unpack-trees.c and builtin/update-index: obvious - entry.c: write_entry() may refresh the checked out entry via fill_stat_cache_info(). This causes "non-const struct cache_entry *" in builtin/apply.c, builtin/checkout-index.c and builtin/checkout.c - builtin/ls-files.c: --with-tree changes stagemask and may set CE_UPDATE Of these, write_entry() and its call sites are probably most interesting because it modifies on-disk info. But this is stat info and can be retrieved via refresh, at least for porcelain commands. Other just uses ce_flags for local purposes. So, keeping track of "dirty" entries is just a matter of setting a flag in index modification functions exposed by read-cache.c. Except unpack-trees, the rest of the code base does not do anything funny behind read-cache's back. The actual patch is less valueable than the summary above. But if anyone wants to re-identify the above sites. Applying this patch, then this: diff --git a/cache.h b/cache.h index 430d021..1692891 100644 --- a/cache.h +++ b/cache.h @@ -267,7 +267,7 @@ static inline unsigned int canon_mode(unsigned int mode) #define cache_entry_size(len) (offsetof(struct cache_entry,name) + (len) + 1) struct index_state { - struct cache_entry **cache; + const struct cache_entry **cache; unsigned int version; unsigned int cache_nr, cache_alloc, cache_changed; struct string_list *resolve_undo; will help quickly identify them without bogus warnings. Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2013-07-09 17:29:00 +02:00
const struct cache_entry *e2 = active_cache[i];
const struct cache_entry *e3 = active_cache[i + 1];
if (ce_stage(e2) == 2 &&
ce_stage(e3) == 3 &&
ce_same_name(e, e3) &&
S_ISREG(e2->ce_mode) &&
S_ISREG(e3->ce_mode))
*type = THREE_STAGED;
}
/* Skip the entries with the same name */
while (i < active_nr && ce_same_name(e, active_cache[i]))
i++;
return i;
}
static int find_conflict(struct string_list *conflict)
{
int i;
if (read_cache() < 0)
return error("Could not read index");
for (i = 0; i < active_nr;) {
int conflict_type;
Convert "struct cache_entry *" to "const ..." wherever possible I attempted to make index_state->cache[] a "const struct cache_entry **" to find out how existing entries in index are modified and where. The question I have is what do we do if we really need to keep track of on-disk changes in the index. The result is - diff-lib.c: setting CE_UPTODATE - name-hash.c: setting CE_HASHED - preload-index.c, read-cache.c, unpack-trees.c and builtin/update-index: obvious - entry.c: write_entry() may refresh the checked out entry via fill_stat_cache_info(). This causes "non-const struct cache_entry *" in builtin/apply.c, builtin/checkout-index.c and builtin/checkout.c - builtin/ls-files.c: --with-tree changes stagemask and may set CE_UPDATE Of these, write_entry() and its call sites are probably most interesting because it modifies on-disk info. But this is stat info and can be retrieved via refresh, at least for porcelain commands. Other just uses ce_flags for local purposes. So, keeping track of "dirty" entries is just a matter of setting a flag in index modification functions exposed by read-cache.c. Except unpack-trees, the rest of the code base does not do anything funny behind read-cache's back. The actual patch is less valueable than the summary above. But if anyone wants to re-identify the above sites. Applying this patch, then this: diff --git a/cache.h b/cache.h index 430d021..1692891 100644 --- a/cache.h +++ b/cache.h @@ -267,7 +267,7 @@ static inline unsigned int canon_mode(unsigned int mode) #define cache_entry_size(len) (offsetof(struct cache_entry,name) + (len) + 1) struct index_state { - struct cache_entry **cache; + const struct cache_entry **cache; unsigned int version; unsigned int cache_nr, cache_alloc, cache_changed; struct string_list *resolve_undo; will help quickly identify them without bogus warnings. Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2013-07-09 17:29:00 +02:00
const struct cache_entry *e = active_cache[i];
i = check_one_conflict(i, &conflict_type);
if (conflict_type == THREE_STAGED)
string_list_insert(conflict, (const char *)e->name);
}
return 0;
}
int rerere_remaining(struct string_list *merge_rr)
{
int i;
rerere: release lockfile in non-writing functions There's a bug in builtin/am.c in which we take a lock on MERGE_RR recursively. But rather than fix am.c, this patch fixes the confusing interface from rerere.c that caused the bug. Read on for the gory details. The setup_rerere() function both reads the existing MERGE_RR file, and takes MERGE_RR.lock. In the rerere() and rerere_forget() functions, we end up in write_rr(), which will then commit the lock file. But for functions like rerere_clear() that do not write to MERGE_RR, we expect the caller to have handled setup_rerere(). That caller would then need to release the lockfile, but it can't; the lock struct is local to rerere.c. For builtin/rerere.c, this is OK. We run a single rerere operation and then exit immediately, which has the side effect of rolling back the lockfile. But in builtin/am.c, this is actively wrong. If we run "git am -3 --skip", we call setup-rerere twice without releasing the lock: 1. The "--skip" causes us to call am_rerere_clear(), which calls setup_rerere(), but never drops the lock. 2. We then proceed to the next patch. 3. The "--3way" may cause us to call rerere() to handle conflicts in that patch, but we are already holding the lock. The lockfile code dies with: BUG: prepare_tempfile_object called for active object We could fix this by having rerere_clear() call rollback_lock_file(). But it feels a bit odd for it to roll back a lockfile that it did not itself take. So let's simplify the interface further, and handle setup_rerere in the function itself, taking away the question from the caller over whether they need to do so. We can give rerere_gc() the same treatment, as well (even though it doesn't have any callers besides builtin/rerere.c at this point). Note that these functions don't take flags from their callers to pass along to setup_rerere; that's OK, because the flags would not be meaningful for what they are doing. Both of those functions need to hold the lock because even though they do not write to MERGE_RR, they are still writing and should be protected from a simultaneous "rerere" run. But rerere_remaining(), "rerere diff", and "rerere status" are all read-only operations. They want to setup_rerere(), but do not care about taking the lock in the first place. Since our update of MERGE_RR is the usual atomic rename done by commit_lock_file, they can just do a lockless read. For that, we teach setup_rerere a READONLY flag to avoid the lock. As a bonus, this pushes builtin/rerere.c's setup_rerere call closer to the functions that use it. Which means that "git rerere totally-bogus-command" will no longer silently exit(0) in a repository without rerere enabled. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-09-02 00:14:09 +02:00
if (setup_rerere(merge_rr, RERERE_READONLY))
return 0;
if (read_cache() < 0)
return error("Could not read index");
for (i = 0; i < active_nr;) {
int conflict_type;
Convert "struct cache_entry *" to "const ..." wherever possible I attempted to make index_state->cache[] a "const struct cache_entry **" to find out how existing entries in index are modified and where. The question I have is what do we do if we really need to keep track of on-disk changes in the index. The result is - diff-lib.c: setting CE_UPTODATE - name-hash.c: setting CE_HASHED - preload-index.c, read-cache.c, unpack-trees.c and builtin/update-index: obvious - entry.c: write_entry() may refresh the checked out entry via fill_stat_cache_info(). This causes "non-const struct cache_entry *" in builtin/apply.c, builtin/checkout-index.c and builtin/checkout.c - builtin/ls-files.c: --with-tree changes stagemask and may set CE_UPDATE Of these, write_entry() and its call sites are probably most interesting because it modifies on-disk info. But this is stat info and can be retrieved via refresh, at least for porcelain commands. Other just uses ce_flags for local purposes. So, keeping track of "dirty" entries is just a matter of setting a flag in index modification functions exposed by read-cache.c. Except unpack-trees, the rest of the code base does not do anything funny behind read-cache's back. The actual patch is less valueable than the summary above. But if anyone wants to re-identify the above sites. Applying this patch, then this: diff --git a/cache.h b/cache.h index 430d021..1692891 100644 --- a/cache.h +++ b/cache.h @@ -267,7 +267,7 @@ static inline unsigned int canon_mode(unsigned int mode) #define cache_entry_size(len) (offsetof(struct cache_entry,name) + (len) + 1) struct index_state { - struct cache_entry **cache; + const struct cache_entry **cache; unsigned int version; unsigned int cache_nr, cache_alloc, cache_changed; struct string_list *resolve_undo; will help quickly identify them without bogus warnings. Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2013-07-09 17:29:00 +02:00
const struct cache_entry *e = active_cache[i];
i = check_one_conflict(i, &conflict_type);
if (conflict_type == PUNTED)
string_list_insert(merge_rr, (const char *)e->name);
else if (conflict_type == RESOLVED) {
struct string_list_item *it;
it = string_list_lookup(merge_rr, (const char *)e->name);
if (it != NULL) {
free(it->util);
it->util = RERERE_RESOLVED;
}
}
}
return 0;
}
static int merge(const char *name, const char *path)
{
int ret;
mmfile_t cur = {NULL, 0}, base = {NULL, 0}, other = {NULL, 0};
mmbuffer_t result = {NULL, 0};
if (handle_file(path, NULL, rerere_path(name, "thisimage")) < 0)
return 1;
if (read_mmfile(&cur, rerere_path(name, "thisimage")) ||
read_mmfile(&base, rerere_path(name, "preimage")) ||
read_mmfile(&other, rerere_path(name, "postimage"))) {
ret = 1;
goto out;
}
ret = ll_merge(&result, path, &base, NULL, &cur, "", &other, "", NULL);
if (!ret) {
FILE *f;
if (utime(rerere_path(name, "postimage"), NULL) < 0)
warning("failed utime() on %s: %s",
rerere_path(name, "postimage"),
strerror(errno));
f = fopen(path, "w");
if (!f)
return error("Could not open %s: %s", path,
strerror(errno));
if (fwrite(result.ptr, result.size, 1, f) != 1)
error("Could not write %s: %s", path, strerror(errno));
if (fclose(f))
return error("Writing %s failed: %s", path,
strerror(errno));
}
out:
free(cur.ptr);
free(base.ptr);
free(other.ptr);
free(result.ptr);
return ret;
}
static struct lock_file index_lock;
static void update_paths(struct string_list *update)
{
int i;
hold_locked_index(&index_lock, 1);
for (i = 0; i < update->nr; i++) {
struct string_list_item *item = &update->items[i];
if (add_file_to_cache(item->string, 0))
exit(128);
}
if (active_cache_changed) {
if (write_locked_index(&the_index, &index_lock, COMMIT_LOCK))
die("Unable to write new index file");
} else
rollback_lock_file(&index_lock);
}
static int do_plain_rerere(struct string_list *rr, int fd)
{
struct string_list conflict = STRING_LIST_INIT_DUP;
struct string_list update = STRING_LIST_INIT_DUP;
int i;
find_conflict(&conflict);
/*
* MERGE_RR records paths with conflicts immediately after merge
* failed. Some of the conflicted paths might have been hand resolved
* in the working tree since then, but the initial run would catch all
* and register their preimages.
*/
for (i = 0; i < conflict.nr; i++) {
const char *path = conflict.items[i].string;
if (!string_list_has_string(rr, path)) {
unsigned char sha1[20];
char *hex;
int ret;
ret = handle_file(path, sha1, NULL);
if (ret < 1)
continue;
hex = xstrdup(sha1_to_hex(sha1));
string_list_insert(rr, path)->util = hex;
if (mkdir_in_gitdir(git_path("rr-cache/%s", hex)))
continue;
handle_file(path, NULL, rerere_path(hex, "preimage"));
fprintf(stderr, "Recorded preimage for '%s'\n", path);
}
}
/*
* Now some of the paths that had conflicts earlier might have been
* hand resolved. Others may be similar to a conflict already that
* was resolved before.
*/
for (i = 0; i < rr->nr; i++) {
int ret;
const char *path = rr->items[i].string;
const char *name = (const char *)rr->items[i].util;
if (has_rerere_resolution(name)) {
if (!merge(name, path)) {
const char *msg;
if (rerere_autoupdate) {
string_list_insert(&update, path);
msg = "Staged '%s' using previous resolution.\n";
} else
msg = "Resolved '%s' using previous resolution.\n";
fprintf(stderr, msg, path);
goto mark_resolved;
}
}
/* Let's see if we have resolved it. */
ret = handle_file(path, NULL, NULL);
if (ret)
continue;
fprintf(stderr, "Recorded resolution for '%s'.\n", path);
copy_file(rerere_path(name, "postimage"), path, 0666);
mark_resolved:
rr->items[i].util = NULL;
}
if (update.nr)
update_paths(&update);
return write_rr(rr, fd);
}
static void git_rerere_config(void)
{
git_config_get_bool("rerere.enabled", &rerere_enabled);
git_config_get_bool("rerere.autoupdate", &rerere_autoupdate);
git_config(git_default_config, NULL);
}
memoize common git-path "constant" files One of the most common uses of git_path() is to pass a constant, like git_path("MERGE_MSG"). This has two drawbacks: 1. The return value is a static buffer, and the lifetime is dependent on other calls to git_path, etc. 2. There's no compile-time checking of the pathname. This is OK for a one-off (after all, we have to spell it correctly at least once), but many of these constant strings appear throughout the code. This patch introduces a series of functions to "memoize" these strings, which are essentially globals for the lifetime of the program. We compute the value once, take ownership of the buffer, and return the cached value for subsequent calls. cache.h provides a helper macro for defining these functions as one-liners, and defines a few common ones for global use. Using a macro is a little bit gross, but it does nicely document the purpose of the functions. If we need to touch them all later (e.g., because we learned how to change the git_dir variable at runtime, and need to invalidate all of the stored values), it will be much easier to have the complete list. Note that the shared-global functions have separate, manual declarations. We could do something clever with the macros (e.g., expand it to a declaration in some places, and a declaration _and_ a definition in path.c). But there aren't that many, and it's probably better to stay away from too-magical macros. Likewise, if we abandon the C preprocessor in favor of generating these with a script, we could get much fancier. E.g., normalizing "FOO/BAR-BAZ" into "git_path_foo_bar_baz". But the small amount of saved typing is probably not worth the resulting confusion to readers who want to grep for the function's definition. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-08-10 11:38:57 +02:00
static GIT_PATH_FUNC(git_path_rr_cache, "rr-cache")
static int is_rerere_enabled(void)
{
int rr_cache_exists;
if (!rerere_enabled)
return 0;
memoize common git-path "constant" files One of the most common uses of git_path() is to pass a constant, like git_path("MERGE_MSG"). This has two drawbacks: 1. The return value is a static buffer, and the lifetime is dependent on other calls to git_path, etc. 2. There's no compile-time checking of the pathname. This is OK for a one-off (after all, we have to spell it correctly at least once), but many of these constant strings appear throughout the code. This patch introduces a series of functions to "memoize" these strings, which are essentially globals for the lifetime of the program. We compute the value once, take ownership of the buffer, and return the cached value for subsequent calls. cache.h provides a helper macro for defining these functions as one-liners, and defines a few common ones for global use. Using a macro is a little bit gross, but it does nicely document the purpose of the functions. If we need to touch them all later (e.g., because we learned how to change the git_dir variable at runtime, and need to invalidate all of the stored values), it will be much easier to have the complete list. Note that the shared-global functions have separate, manual declarations. We could do something clever with the macros (e.g., expand it to a declaration in some places, and a declaration _and_ a definition in path.c). But there aren't that many, and it's probably better to stay away from too-magical macros. Likewise, if we abandon the C preprocessor in favor of generating these with a script, we could get much fancier. E.g., normalizing "FOO/BAR-BAZ" into "git_path_foo_bar_baz". But the small amount of saved typing is probably not worth the resulting confusion to readers who want to grep for the function's definition. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-08-10 11:38:57 +02:00
rr_cache_exists = is_directory(git_path_rr_cache());
if (rerere_enabled < 0)
return rr_cache_exists;
memoize common git-path "constant" files One of the most common uses of git_path() is to pass a constant, like git_path("MERGE_MSG"). This has two drawbacks: 1. The return value is a static buffer, and the lifetime is dependent on other calls to git_path, etc. 2. There's no compile-time checking of the pathname. This is OK for a one-off (after all, we have to spell it correctly at least once), but many of these constant strings appear throughout the code. This patch introduces a series of functions to "memoize" these strings, which are essentially globals for the lifetime of the program. We compute the value once, take ownership of the buffer, and return the cached value for subsequent calls. cache.h provides a helper macro for defining these functions as one-liners, and defines a few common ones for global use. Using a macro is a little bit gross, but it does nicely document the purpose of the functions. If we need to touch them all later (e.g., because we learned how to change the git_dir variable at runtime, and need to invalidate all of the stored values), it will be much easier to have the complete list. Note that the shared-global functions have separate, manual declarations. We could do something clever with the macros (e.g., expand it to a declaration in some places, and a declaration _and_ a definition in path.c). But there aren't that many, and it's probably better to stay away from too-magical macros. Likewise, if we abandon the C preprocessor in favor of generating these with a script, we could get much fancier. E.g., normalizing "FOO/BAR-BAZ" into "git_path_foo_bar_baz". But the small amount of saved typing is probably not worth the resulting confusion to readers who want to grep for the function's definition. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-08-10 11:38:57 +02:00
if (!rr_cache_exists && mkdir_in_gitdir(git_path_rr_cache()))
die("Could not create directory %s", git_path_rr_cache());
return 1;
}
int setup_rerere(struct string_list *merge_rr, int flags)
{
int fd;
git_rerere_config();
if (!is_rerere_enabled())
return -1;
if (flags & (RERERE_AUTOUPDATE|RERERE_NOAUTOUPDATE))
rerere_autoupdate = !!(flags & RERERE_AUTOUPDATE);
rerere: release lockfile in non-writing functions There's a bug in builtin/am.c in which we take a lock on MERGE_RR recursively. But rather than fix am.c, this patch fixes the confusing interface from rerere.c that caused the bug. Read on for the gory details. The setup_rerere() function both reads the existing MERGE_RR file, and takes MERGE_RR.lock. In the rerere() and rerere_forget() functions, we end up in write_rr(), which will then commit the lock file. But for functions like rerere_clear() that do not write to MERGE_RR, we expect the caller to have handled setup_rerere(). That caller would then need to release the lockfile, but it can't; the lock struct is local to rerere.c. For builtin/rerere.c, this is OK. We run a single rerere operation and then exit immediately, which has the side effect of rolling back the lockfile. But in builtin/am.c, this is actively wrong. If we run "git am -3 --skip", we call setup-rerere twice without releasing the lock: 1. The "--skip" causes us to call am_rerere_clear(), which calls setup_rerere(), but never drops the lock. 2. We then proceed to the next patch. 3. The "--3way" may cause us to call rerere() to handle conflicts in that patch, but we are already holding the lock. The lockfile code dies with: BUG: prepare_tempfile_object called for active object We could fix this by having rerere_clear() call rollback_lock_file(). But it feels a bit odd for it to roll back a lockfile that it did not itself take. So let's simplify the interface further, and handle setup_rerere in the function itself, taking away the question from the caller over whether they need to do so. We can give rerere_gc() the same treatment, as well (even though it doesn't have any callers besides builtin/rerere.c at this point). Note that these functions don't take flags from their callers to pass along to setup_rerere; that's OK, because the flags would not be meaningful for what they are doing. Both of those functions need to hold the lock because even though they do not write to MERGE_RR, they are still writing and should be protected from a simultaneous "rerere" run. But rerere_remaining(), "rerere diff", and "rerere status" are all read-only operations. They want to setup_rerere(), but do not care about taking the lock in the first place. Since our update of MERGE_RR is the usual atomic rename done by commit_lock_file, they can just do a lockless read. For that, we teach setup_rerere a READONLY flag to avoid the lock. As a bonus, this pushes builtin/rerere.c's setup_rerere call closer to the functions that use it. Which means that "git rerere totally-bogus-command" will no longer silently exit(0) in a repository without rerere enabled. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-09-02 00:14:09 +02:00
if (flags & RERERE_READONLY)
fd = 0;
else
fd = hold_lock_file_for_update(&write_lock, git_path_merge_rr(),
LOCK_DIE_ON_ERROR);
read_rr(merge_rr);
return fd;
}
int rerere(int flags)
{
struct string_list merge_rr = STRING_LIST_INIT_DUP;
int fd;
fd = setup_rerere(&merge_rr, flags);
if (fd < 0)
return 0;
return do_plain_rerere(&merge_rr, fd);
}
static int rerere_forget_one_path(const char *path, struct string_list *rr)
{
const char *filename;
char *hex;
unsigned char sha1[20];
int ret;
ret = handle_cache(path, sha1, NULL);
if (ret < 1)
return error("Could not parse conflict hunks in '%s'", path);
hex = xstrdup(sha1_to_hex(sha1));
filename = rerere_path(hex, "postimage");
if (unlink(filename))
return (errno == ENOENT
? error("no remembered resolution for %s", path)
: error("cannot unlink %s: %s", filename, strerror(errno)));
handle_cache(path, sha1, rerere_path(hex, "preimage"));
fprintf(stderr, "Updated preimage for '%s'\n", path);
string_list_insert(rr, path)->util = hex;
fprintf(stderr, "Forgot resolution for %s\n", path);
return 0;
}
int rerere_forget(struct pathspec *pathspec)
{
int i, fd;
struct string_list conflict = STRING_LIST_INIT_DUP;
struct string_list merge_rr = STRING_LIST_INIT_DUP;
if (read_cache() < 0)
return error("Could not read index");
fd = setup_rerere(&merge_rr, RERERE_NOAUTOUPDATE);
if (fd < 0)
return 0;
unmerge_cache(pathspec);
find_conflict(&conflict);
for (i = 0; i < conflict.nr; i++) {
struct string_list_item *it = &conflict.items[i];
if (!match_pathspec(pathspec, it->string,
strlen(it->string), 0, NULL, 0))
continue;
rerere_forget_one_path(it->string, &merge_rr);
}
return write_rr(&merge_rr, fd);
}
static time_t rerere_created_at(const char *name)
{
struct stat st;
return stat(rerere_path(name, "preimage"), &st) ? (time_t) 0 : st.st_mtime;
}
static time_t rerere_last_used_at(const char *name)
{
struct stat st;
return stat(rerere_path(name, "postimage"), &st) ? (time_t) 0 : st.st_mtime;
}
static void unlink_rr_item(const char *name)
{
unlink(rerere_path(name, "thisimage"));
unlink(rerere_path(name, "preimage"));
unlink(rerere_path(name, "postimage"));
rmdir(git_path("rr-cache/%s", name));
}
void rerere_gc(struct string_list *rr)
{
struct string_list to_remove = STRING_LIST_INIT_DUP;
DIR *dir;
struct dirent *e;
int i, cutoff;
time_t now = time(NULL), then;
int cutoff_noresolve = 15;
int cutoff_resolve = 60;
rerere: release lockfile in non-writing functions There's a bug in builtin/am.c in which we take a lock on MERGE_RR recursively. But rather than fix am.c, this patch fixes the confusing interface from rerere.c that caused the bug. Read on for the gory details. The setup_rerere() function both reads the existing MERGE_RR file, and takes MERGE_RR.lock. In the rerere() and rerere_forget() functions, we end up in write_rr(), which will then commit the lock file. But for functions like rerere_clear() that do not write to MERGE_RR, we expect the caller to have handled setup_rerere(). That caller would then need to release the lockfile, but it can't; the lock struct is local to rerere.c. For builtin/rerere.c, this is OK. We run a single rerere operation and then exit immediately, which has the side effect of rolling back the lockfile. But in builtin/am.c, this is actively wrong. If we run "git am -3 --skip", we call setup-rerere twice without releasing the lock: 1. The "--skip" causes us to call am_rerere_clear(), which calls setup_rerere(), but never drops the lock. 2. We then proceed to the next patch. 3. The "--3way" may cause us to call rerere() to handle conflicts in that patch, but we are already holding the lock. The lockfile code dies with: BUG: prepare_tempfile_object called for active object We could fix this by having rerere_clear() call rollback_lock_file(). But it feels a bit odd for it to roll back a lockfile that it did not itself take. So let's simplify the interface further, and handle setup_rerere in the function itself, taking away the question from the caller over whether they need to do so. We can give rerere_gc() the same treatment, as well (even though it doesn't have any callers besides builtin/rerere.c at this point). Note that these functions don't take flags from their callers to pass along to setup_rerere; that's OK, because the flags would not be meaningful for what they are doing. Both of those functions need to hold the lock because even though they do not write to MERGE_RR, they are still writing and should be protected from a simultaneous "rerere" run. But rerere_remaining(), "rerere diff", and "rerere status" are all read-only operations. They want to setup_rerere(), but do not care about taking the lock in the first place. Since our update of MERGE_RR is the usual atomic rename done by commit_lock_file, they can just do a lockless read. For that, we teach setup_rerere a READONLY flag to avoid the lock. As a bonus, this pushes builtin/rerere.c's setup_rerere call closer to the functions that use it. Which means that "git rerere totally-bogus-command" will no longer silently exit(0) in a repository without rerere enabled. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-09-02 00:14:09 +02:00
if (setup_rerere(rr, 0) < 0)
return;
git_config_get_int("gc.rerereresolved", &cutoff_resolve);
git_config_get_int("gc.rerereunresolved", &cutoff_noresolve);
git_config(git_default_config, NULL);
dir = opendir(git_path("rr-cache"));
if (!dir)
die_errno("unable to open rr-cache directory");
while ((e = readdir(dir))) {
if (is_dot_or_dotdot(e->d_name))
continue;
then = rerere_last_used_at(e->d_name);
if (then) {
cutoff = cutoff_resolve;
} else {
then = rerere_created_at(e->d_name);
if (!then)
continue;
cutoff = cutoff_noresolve;
}
if (then < now - cutoff * 86400)
string_list_append(&to_remove, e->d_name);
}
closedir(dir);
for (i = 0; i < to_remove.nr; i++)
unlink_rr_item(to_remove.items[i].string);
string_list_clear(&to_remove, 0);
rerere: release lockfile in non-writing functions There's a bug in builtin/am.c in which we take a lock on MERGE_RR recursively. But rather than fix am.c, this patch fixes the confusing interface from rerere.c that caused the bug. Read on for the gory details. The setup_rerere() function both reads the existing MERGE_RR file, and takes MERGE_RR.lock. In the rerere() and rerere_forget() functions, we end up in write_rr(), which will then commit the lock file. But for functions like rerere_clear() that do not write to MERGE_RR, we expect the caller to have handled setup_rerere(). That caller would then need to release the lockfile, but it can't; the lock struct is local to rerere.c. For builtin/rerere.c, this is OK. We run a single rerere operation and then exit immediately, which has the side effect of rolling back the lockfile. But in builtin/am.c, this is actively wrong. If we run "git am -3 --skip", we call setup-rerere twice without releasing the lock: 1. The "--skip" causes us to call am_rerere_clear(), which calls setup_rerere(), but never drops the lock. 2. We then proceed to the next patch. 3. The "--3way" may cause us to call rerere() to handle conflicts in that patch, but we are already holding the lock. The lockfile code dies with: BUG: prepare_tempfile_object called for active object We could fix this by having rerere_clear() call rollback_lock_file(). But it feels a bit odd for it to roll back a lockfile that it did not itself take. So let's simplify the interface further, and handle setup_rerere in the function itself, taking away the question from the caller over whether they need to do so. We can give rerere_gc() the same treatment, as well (even though it doesn't have any callers besides builtin/rerere.c at this point). Note that these functions don't take flags from their callers to pass along to setup_rerere; that's OK, because the flags would not be meaningful for what they are doing. Both of those functions need to hold the lock because even though they do not write to MERGE_RR, they are still writing and should be protected from a simultaneous "rerere" run. But rerere_remaining(), "rerere diff", and "rerere status" are all read-only operations. They want to setup_rerere(), but do not care about taking the lock in the first place. Since our update of MERGE_RR is the usual atomic rename done by commit_lock_file, they can just do a lockless read. For that, we teach setup_rerere a READONLY flag to avoid the lock. As a bonus, this pushes builtin/rerere.c's setup_rerere call closer to the functions that use it. Which means that "git rerere totally-bogus-command" will no longer silently exit(0) in a repository without rerere enabled. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-09-02 00:14:09 +02:00
rollback_lock_file(&write_lock);
}
void rerere_clear(struct string_list *merge_rr)
{
int i;
rerere: release lockfile in non-writing functions There's a bug in builtin/am.c in which we take a lock on MERGE_RR recursively. But rather than fix am.c, this patch fixes the confusing interface from rerere.c that caused the bug. Read on for the gory details. The setup_rerere() function both reads the existing MERGE_RR file, and takes MERGE_RR.lock. In the rerere() and rerere_forget() functions, we end up in write_rr(), which will then commit the lock file. But for functions like rerere_clear() that do not write to MERGE_RR, we expect the caller to have handled setup_rerere(). That caller would then need to release the lockfile, but it can't; the lock struct is local to rerere.c. For builtin/rerere.c, this is OK. We run a single rerere operation and then exit immediately, which has the side effect of rolling back the lockfile. But in builtin/am.c, this is actively wrong. If we run "git am -3 --skip", we call setup-rerere twice without releasing the lock: 1. The "--skip" causes us to call am_rerere_clear(), which calls setup_rerere(), but never drops the lock. 2. We then proceed to the next patch. 3. The "--3way" may cause us to call rerere() to handle conflicts in that patch, but we are already holding the lock. The lockfile code dies with: BUG: prepare_tempfile_object called for active object We could fix this by having rerere_clear() call rollback_lock_file(). But it feels a bit odd for it to roll back a lockfile that it did not itself take. So let's simplify the interface further, and handle setup_rerere in the function itself, taking away the question from the caller over whether they need to do so. We can give rerere_gc() the same treatment, as well (even though it doesn't have any callers besides builtin/rerere.c at this point). Note that these functions don't take flags from their callers to pass along to setup_rerere; that's OK, because the flags would not be meaningful for what they are doing. Both of those functions need to hold the lock because even though they do not write to MERGE_RR, they are still writing and should be protected from a simultaneous "rerere" run. But rerere_remaining(), "rerere diff", and "rerere status" are all read-only operations. They want to setup_rerere(), but do not care about taking the lock in the first place. Since our update of MERGE_RR is the usual atomic rename done by commit_lock_file, they can just do a lockless read. For that, we teach setup_rerere a READONLY flag to avoid the lock. As a bonus, this pushes builtin/rerere.c's setup_rerere call closer to the functions that use it. Which means that "git rerere totally-bogus-command" will no longer silently exit(0) in a repository without rerere enabled. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-09-02 00:14:09 +02:00
if (setup_rerere(merge_rr, 0) < 0)
return;
for (i = 0; i < merge_rr->nr; i++) {
const char *name = (const char *)merge_rr->items[i].util;
if (!has_rerere_resolution(name))
unlink_rr_item(name);
}
memoize common git-path "constant" files One of the most common uses of git_path() is to pass a constant, like git_path("MERGE_MSG"). This has two drawbacks: 1. The return value is a static buffer, and the lifetime is dependent on other calls to git_path, etc. 2. There's no compile-time checking of the pathname. This is OK for a one-off (after all, we have to spell it correctly at least once), but many of these constant strings appear throughout the code. This patch introduces a series of functions to "memoize" these strings, which are essentially globals for the lifetime of the program. We compute the value once, take ownership of the buffer, and return the cached value for subsequent calls. cache.h provides a helper macro for defining these functions as one-liners, and defines a few common ones for global use. Using a macro is a little bit gross, but it does nicely document the purpose of the functions. If we need to touch them all later (e.g., because we learned how to change the git_dir variable at runtime, and need to invalidate all of the stored values), it will be much easier to have the complete list. Note that the shared-global functions have separate, manual declarations. We could do something clever with the macros (e.g., expand it to a declaration in some places, and a declaration _and_ a definition in path.c). But there aren't that many, and it's probably better to stay away from too-magical macros. Likewise, if we abandon the C preprocessor in favor of generating these with a script, we could get much fancier. E.g., normalizing "FOO/BAR-BAZ" into "git_path_foo_bar_baz". But the small amount of saved typing is probably not worth the resulting confusion to readers who want to grep for the function's definition. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-08-10 11:38:57 +02:00
unlink_or_warn(git_path_merge_rr());
rerere: release lockfile in non-writing functions There's a bug in builtin/am.c in which we take a lock on MERGE_RR recursively. But rather than fix am.c, this patch fixes the confusing interface from rerere.c that caused the bug. Read on for the gory details. The setup_rerere() function both reads the existing MERGE_RR file, and takes MERGE_RR.lock. In the rerere() and rerere_forget() functions, we end up in write_rr(), which will then commit the lock file. But for functions like rerere_clear() that do not write to MERGE_RR, we expect the caller to have handled setup_rerere(). That caller would then need to release the lockfile, but it can't; the lock struct is local to rerere.c. For builtin/rerere.c, this is OK. We run a single rerere operation and then exit immediately, which has the side effect of rolling back the lockfile. But in builtin/am.c, this is actively wrong. If we run "git am -3 --skip", we call setup-rerere twice without releasing the lock: 1. The "--skip" causes us to call am_rerere_clear(), which calls setup_rerere(), but never drops the lock. 2. We then proceed to the next patch. 3. The "--3way" may cause us to call rerere() to handle conflicts in that patch, but we are already holding the lock. The lockfile code dies with: BUG: prepare_tempfile_object called for active object We could fix this by having rerere_clear() call rollback_lock_file(). But it feels a bit odd for it to roll back a lockfile that it did not itself take. So let's simplify the interface further, and handle setup_rerere in the function itself, taking away the question from the caller over whether they need to do so. We can give rerere_gc() the same treatment, as well (even though it doesn't have any callers besides builtin/rerere.c at this point). Note that these functions don't take flags from their callers to pass along to setup_rerere; that's OK, because the flags would not be meaningful for what they are doing. Both of those functions need to hold the lock because even though they do not write to MERGE_RR, they are still writing and should be protected from a simultaneous "rerere" run. But rerere_remaining(), "rerere diff", and "rerere status" are all read-only operations. They want to setup_rerere(), but do not care about taking the lock in the first place. Since our update of MERGE_RR is the usual atomic rename done by commit_lock_file, they can just do a lockless read. For that, we teach setup_rerere a READONLY flag to avoid the lock. As a bonus, this pushes builtin/rerere.c's setup_rerere call closer to the functions that use it. Which means that "git rerere totally-bogus-command" will no longer silently exit(0) in a repository without rerere enabled. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2015-09-02 00:14:09 +02:00
rollback_lock_file(&write_lock);
}