{"thread":{"id":"54544","subject":"[PATCH] upload-pack.c: fix a sparse warning","startedAt":"2020-10-31T21:05:07Z","lastAt":"2020-11-04T20:05:10Z","messageCount":11,"participants":["Ramsay Jones","Junio C Hamano","Josh Steadmon","Johannes Schindelin"],"isPatch":true,"patchVersion":1,"patchTotal":null},"messages":[{"id":"408801","messageId":"eaf5ac83-af3f-5028-3a9e-1669c9cde116@ramsayjones.plus.com","threadId":"54544","inReplyTo":null,"subject":"[PATCH] upload-pack.c: fix a sparse warning","fromName":"Ramsay Jones","fromEmail":"ramsay@ramsayjones.plus.com","sentAt":"2020-10-31T21:04:58Z","receivedAt":"2020-10-31T21:05:07Z","isPatch":true,"sender":{"key":"ramsay@ramsayjones.plus.com","avatar":"https://avatars.githubusercontent.com/u/33702710?v=4"},"body":"\nSigned-off-by: Ramsay Jones \n---\n\nHi Josh,\n\nIf you need to re-roll your 'js/trace2-session-id' branch, could you please\nsquash this into the relevant patch (commit edb21a501e, \"upload-pack, serve:\nlog received client trace2 SID\", 29-10-2020).\n\nThanks!\n\nATB,\nRamsay Jones\n \n upload-pack.c | 2 +-\n 1 file changed, 1 insertion(+), 1 deletion(-)\n\ndiff --git a/upload-pack.c b/upload-pack.c\nindex b7d097bf1b..2996b8083c 100644\n--- a/upload-pack.c\n+++ b/upload-pack.c\n@@ -1111,7 +1111,7 @@ static void receive_needs(struct upload_pack_data *data,\n \t\tif (data->allow_filter &&\n \t\t parse_feature_request(features, \"filter\"))\n \t\t\tdata->filter_capability_requested = 1;\n-\t\tif ((arg = parse_feature_value(features, \"trace2-sid\", &feature_len, 0)))\n+\t\tif ((arg = parse_feature_value(features, \"trace2-sid\", &feature_len, NULL)))\n \t\t{\n \t\t\tchar *client_sid = xstrndup(arg, feature_len);\n \t\t\ttrace2_data_string(\"trace2\", NULL, \"client-sid\", client_sid);\n-- \n2.29.0\n"},{"id":"408803","messageId":"xmqqsg9uxchb.fsf@gitster.c.googlers.com","threadId":"54544","inReplyTo":"eaf5ac83-af3f-5028-3a9e-1669c9cde116@ramsayjones.plus.com","subject":"[PATCH] Makefile: enable -Wsparse-error for DEVELOPER build","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2020-10-31T22:22:08Z","receivedAt":"2020-10-31T22:28:06Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"With -Wsparse-error, \"make sparse\" would fail, instead of just\ngiving a warning message.\n\nSigned-off-by: Junio C Hamano \n---\n config.mak.dev | 1 +\n 1 file changed, 1 insertion(+)\n\ndiff --git c/config.mak.dev w/config.mak.dev\nindex 3126a5364d..022fb58218 100644\n--- c/config.mak.dev\n+++ w/config.mak.dev\n@@ -1,5 +1,6 @@\n ifeq ($(filter no-error,$(DEVOPTS)),)\n DEVELOPER_CFLAGS += -Werror\n+SPARSE_FLAGS += -Wsparse-error\n endif\n ifneq ($(filter pedantic,$(DEVOPTS)),)\n DEVELOPER_CFLAGS += -pedantic\n"},{"id":"408852","messageId":"6676a08d-0c25-14ba-4ea6-677bc7ab0e20@ramsayjones.plus.com","threadId":"54544","inReplyTo":"xmqqsg9uxchb.fsf@gitster.c.googlers.com","subject":"Re: [PATCH] Makefile: enable -Wsparse-error for DEVELOPER build","fromName":"Ramsay Jones","fromEmail":"ramsay@ramsayjones.plus.com","sentAt":"2020-11-02T00:05:30Z","receivedAt":"2020-11-02T00:05:35Z","isPatch":true,"sender":{"key":"ramsay@ramsayjones.plus.com","avatar":"https://avatars.githubusercontent.com/u/33702710?v=4"},"body":"Hi Junio,\n\nOn 31/10/2020 22:22, Junio C Hamano wrote:\n> With -Wsparse-error, \"make sparse\" would fail, instead of just\n> giving a warning message.\n> \n> Signed-off-by: Junio C Hamano \n> ---\n> config.mak.dev | 1 +\n> 1 file changed, 1 insertion(+)\n> \n> diff --git c/config.mak.dev w/config.mak.dev\n> index 3126a5364d..022fb58218 100644\n> --- c/config.mak.dev\n> +++ w/config.mak.dev\n> @@ -1,5 +1,6 @@\n> ifeq ($(filter no-error,$(DEVOPTS)),)\n> DEVELOPER_CFLAGS += -Werror\n> +SPARSE_FLAGS += -Wsparse-error\n> endif\n> ifneq ($(filter pedantic,$(DEVOPTS)),)\n> DEVELOPER_CFLAGS += -pedantic\n> \n\nI certainly wouldn't object to such a patch, but I'm not sure who it\nwould actually help. ;-)\n\nAs you may already know, I 'make sparse >sp-out 2>&1' on the master\nbranch, use vim to check for warnings/errors, and then diff the\ncorresponding files for 'next' and 'seen' branches (nsp-out, ssp-out).\nSimilarly, I 'make -k hdr-check >hcout 2>&1' on the master branch,\nuse vim ... etc. Note, however, the use of '-k' on the make invocation\nto make sure I catch all warnings/errors, since '-Werror' is given with\nthe DEVELOPER variable set. So, I would have to do the same with the\n'make sparse' invocation after this patch [1].\n\nYes, anybody who just does 'make sparse' will notice the failure, so\nthat would be a definite improvement. (How many people run 'make sparse'\nthough?). It probably would help anybody using 'cgcc' to develop, but\nthis doesn't quite work (I just tried this for the first time in ages\nand was surprised to find it almost works for me, YMMV):\n\n $ git branch -v | grep '^\\*'\n * seen 4141ae2199 Merge branch 'dd/upload-pack-stateless-eof' into seen\n $ make clean >/dev/null\n $ \n\n $ make CC=cgcc >sout2 2>&1\n $ ./git version\n git version 2.29.2.389.g4141ae2199\n $ git describe\n v2.29.2-389-g4141ae2199\n $ \n\n $ grep 'warn' sout2\n pack-revindex.c:66:23: warning: memset with byte count of 262144\n upload-pack.c:1114:86: warning: Using plain integer as NULL pointer\n $ \n\nSo, this would have failed, because some extra flags ('SP_EXTRA_FLAGS' to\nbe precise) are not being passed to sparse. We can see that:\n\n $ grep SP_EXTRA_FLAGS Makefile\n SP_EXTRA_FLAGS = -Wno-universal-initializer\n http-push.sp http.sp http-walker.sp remote-curl.sp imap-send.sp: SP_EXTRA_FLAGS += \\\n pack-revindex.sp: SP_EXTRA_FLAGS += -Wno-memcpy-max-count\n compat/nedmalloc/nedmalloc.sp: SP_EXTRA_FLAGS += -Wno-non-pointer-null\n \t\t$(SPARSE_FLAGS) $(SP_EXTRA_FLAGS) $<\n $ \n\n(Note that you can't see -DCURL_DISABLE_TYPECHECK being set for all the\nlibcurl using files). So, either sparse has improved to the point that\nthe 'typecheck' shenanigans used in the libcurl headers don't cause it\nproblems anymore, or the headers have changed sufficiently. (Hmm, that\nheader doesn't seem to have changed much).\n\nI don't build with nedmalloc, but last time I did it was littered with\nuses of '0' as a NULL pointer (hence the -Wno-non-pointer-null). Given\nthat we will probably not re-import that code, I suppose all of those\nsparse warnings could be fixed.\n\nWhich leaves 'pack-revindex.c'. Unfortunately, I don't see an easy fix\nfor that (-Wno-universal-initializer is now the default, so we don't\nneed that anymore). Well, I suppose anybody could add 'CFLAGS+=-Wno-\\\nmemcpy-max-count' to their config.mak file (this '-W' argument is\nfiltered out by cgcc and not passed to gcc). [EDIT: I just tried this\nand, yes, this does work! I'm still surprised about curl headers ;-) ]\n\nNote also that my build slowed down by 14% using CC=cgcc, since it is\nrunning both sparse and gcc on each source file.\n\nI guess it would be most useful on a CI build, but I don't know what\nwould be involved in setting that up.\n\nATB,\nRamsay Jones\n\n[1] It took about one month for me to get used to the 'pu'->'seen' change,\n but that was mostly training my fingers to type all of the output\n files that are keyed by the branch name: pout->sout, psp-out->ssp-out,\n psc->ssc, phcout->shcout and ptest-out->stest-out. :D\n\n"},{"id":"408883","messageId":"xmqqh7q7vbah.fsf@gitster.c.googlers.com","threadId":"54544","inReplyTo":"6676a08d-0c25-14ba-4ea6-677bc7ab0e20@ramsayjones.plus.com","subject":"Re: [PATCH] Makefile: enable -Wsparse-error for DEVELOPER build","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2020-11-02T18:55:18Z","receivedAt":"2020-11-02T18:55:26Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Ramsay Jones writes:\n\n> Yes, anybody who just does 'make sparse' will notice the failure, so\n> that would be a definite improvement. (How many people run 'make sparse'\n> though?).\n\nAt least, I was planning to add one to the number when I wrote the\npatch, so that I would start using it as part of daily integration\nbuild before pushing day's result out.\n\nI can tweak my $(MAKE) command line arguments in the script I use\nfor daily integration builds, so the patch is not strictly needed\nfor me personally.\n\n> I guess it would be most useful on a CI build, but I don't know what\n> would be involved in setting that up.\n\nCI builds already have enough stuff around invocation of \"make test\"\netc., and it would be trivial to pass SPARSE_FLAGS from the command\nline when adding \"make sparse\" invocation to one of the scripts in\nci/ directory, so from that point of view, this patch is not needed\nfor them, either.\n\nMy hope was that it would make it harder for individual contributors\nwho run \"make sparse\" and fail to notice an error from it.\n"},{"id":"408934","messageId":"20201102222150.GB1904687@google.com","threadId":"54544","inReplyTo":"eaf5ac83-af3f-5028-3a9e-1669c9cde116@ramsayjones.plus.com","subject":"Re: [PATCH] upload-pack.c: fix a sparse warning","fromName":"Josh Steadmon","fromEmail":"steadmon@google.com","sentAt":"2020-11-02T22:21:50Z","receivedAt":"2020-11-02T22:21:58Z","isPatch":true,"sender":{"key":"steadmon@google.com","avatar":"https://avatars.githubusercontent.com/u/2654920?v=4"},"body":"On 2020.10.31 21:04, Ramsay Jones wrote:\n> \n> Signed-off-by: Ramsay Jones \n> ---\n> \n> Hi Josh,\n> \n> If you need to re-roll your 'js/trace2-session-id' branch, could you please\n> squash this into the relevant patch (commit edb21a501e, \"upload-pack, serve:\n> log received client trace2 SID\", 29-10-2020).\n> \n> Thanks!\n> \n> ATB,\n> Ramsay Jones\n\nFixed in V2, thanks!\n\n> upload-pack.c | 2 +-\n> 1 file changed, 1 insertion(+), 1 deletion(-)\n> \n> diff --git a/upload-pack.c b/upload-pack.c\n> index b7d097bf1b..2996b8083c 100644\n> --- a/upload-pack.c\n> +++ b/upload-pack.c\n> @@ -1111,7 +1111,7 @@ static void receive_needs(struct upload_pack_data *data,\n> \t\tif (data->allow_filter &&\n> \t\t parse_feature_request(features, \"filter\"))\n> \t\t\tdata->filter_capability_requested = 1;\n> -\t\tif ((arg = parse_feature_value(features, \"trace2-sid\", &feature_len, 0)))\n> +\t\tif ((arg = parse_feature_value(features, \"trace2-sid\", &feature_len, NULL)))\n> \t\t{\n> \t\t\tchar *client_sid = xstrndup(arg, feature_len);\n> \t\t\ttrace2_data_string(\"trace2\", NULL, \"client-sid\", client_sid);\n> -- \n> 2.29.0\n"},{"id":"408978","messageId":"43aadae9-a4c1-f1e4-6902-c05af36fcf19@ramsayjones.plus.com","threadId":"54544","inReplyTo":"xmqqh7q7vbah.fsf@gitster.c.googlers.com","subject":"Re: [PATCH] Makefile: enable -Wsparse-error for DEVELOPER build","fromName":"Ramsay Jones","fromEmail":"ramsay@ramsayjones.plus.com","sentAt":"2020-11-03T02:04:45Z","receivedAt":"2020-11-03T02:04:50Z","isPatch":true,"sender":{"key":"ramsay@ramsayjones.plus.com","avatar":"https://avatars.githubusercontent.com/u/33702710?v=4"},"body":"\n\nOn 02/11/2020 18:55, Junio C Hamano wrote:\n> Ramsay Jones writes:\n> \n>> Yes, anybody who just does 'make sparse' will notice the failure, so\n>> that would be a definite improvement. (How many people run 'make sparse'\n>> though?).\n> \n> At least, I was planning to add one to the number when I wrote the\n> patch, so that I would start using it as part of daily integration\n> build before pushing day's result out.\n\nGreat! :-D\n\nHow are you installing sparse? (I am using Linux Mint 20, based on\nUbuntu 20.04, whose sparse package version is 0.6.1-2build1 - which\nis to say too old; you require version 0.6.2 or greater. Building\nfrom source, I am currently using v0.6.3-76-gf680124b).\n\n[I seem to remember you saying you used Fedora, which usually has\nmore up-to-date software than most distros].\n\n>> I guess it would be most useful on a CI build, but I don't know what\n>> would be involved in setting that up.\n> \n> CI builds already have enough stuff around invocation of \"make test\"\n> etc., and it would be trivial to pass SPARSE_FLAGS from the command\n> line when adding \"make sparse\" invocation to one of the scripts in\n> ci/ directory, so from that point of view, this patch is not needed\n> for them, either.\n\nMy concern was more about how the CI system obtains/installs/builds a\nsufficiently new version of sparse. Otherwise, 'make sparse' won't do\nvery much. ;-) As I said, I don't know what's involved in getting\nthat to work.\n\nATB,\nRamsay Jones\n\n"},{"id":"408979","messageId":"xmqqtuu7rw5b.fsf@gitster.c.googlers.com","threadId":"54544","inReplyTo":"43aadae9-a4c1-f1e4-6902-c05af36fcf19@ramsayjones.plus.com","subject":"Re: [PATCH] Makefile: enable -Wsparse-error for DEVELOPER build","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2020-11-03T02:50:40Z","receivedAt":"2020-11-03T02:50:48Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Ramsay Jones writes:\n\n> How are you installing sparse? (I am using Linux Mint 20, based on\n> Ubuntu 20.04, whose sparse package version is 0.6.1-2build1 - which\n> is to say too old; you require version 0.6.2 or greater. Building\n> from source, I am currently using v0.6.3-76-gf680124b).\n\nBuilt and installed in ~/$arch/gitstuff/bin which is on my $PATH\n\n[remote \"origin\"]\n\turl = git://git.kernel.org/pub/scm/devel/sparse/sparse.git\n\n> My concern was more about how the CI system obtains/installs/builds a\n> sufficiently new version of sparse. Otherwise, 'make sparse' won't do\n> very much. ;-) As I said, I don't know what's involved in getting\n> that to work.\n\nAh, yes, it would be a disaster to trigger false positive by using\nold version like 0.6.1, so we may end up having to bring our own\nversion in, which is more work X-<.\n\nThanks.\n"},{"id":"409080","messageId":"nycvar.QRO.7.76.6.2011041431370.18437@tvgsbejvaqbjf.bet","threadId":"54544","inReplyTo":"43aadae9-a4c1-f1e4-6902-c05af36fcf19@ramsayjones.plus.com","subject":"Re: [PATCH] Makefile: enable -Wsparse-error for DEVELOPER build","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2020-11-04T13:43:43Z","receivedAt":"2020-11-04T13:43:52Z","isPatch":true,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi Ramsay,\n\nOn Tue, 3 Nov 2020, Ramsay Jones wrote:\n\n> On 02/11/2020 18:55, Junio C Hamano wrote:\n>\n> > CI builds already have enough stuff around invocation of \"make test\"\n> > etc., and it would be trivial to pass SPARSE_FLAGS from the command\n> > line when adding \"make sparse\" invocation to one of the scripts in ci/\n> > directory, so from that point of view, this patch is not needed for\n> > them, either.\n>\n> My concern was more about how the CI system obtains/installs/builds a\n> sufficiently new version of sparse. Otherwise, 'make sparse' won't do\n> very much. ;-) As I said, I don't know what's involved in getting that\n> to work.\n\nAs mentioned in https://github.com/gitgitgadget/git/issues/345, there is a\nPipeline that builds sparse packages for Ubuntu, ready for use in our CI\nbuild: https://dev.azure.com/git/git/_build?definitionId=10&_a=summary.\nCurrently, it is scheduled to run every weekday at 5:00am (I assume that's\nUTC).\n\nSo I went ahead and added more details to the ticket, essentially\ndemonstrating how the `sparse` package can be downloaded and installed in\nthe CI build in a Bash step:\n\n-- snip --\nurlbase=https://dev.azure.com/git/git/_apis/build/builds\nid=$(curl \"$urlbase?definitions=10&statusFilter=completed&resultFilter=succeeded&\\$top=1\" |\n\tjson_pp |\n\tsed -n 's/^ \"id\" : \\([1-9][0-9]*\\).*/\\1/p')\ndownload_url=\"$(curl \"$urlbase/$id/artifacts\" |\n\tjson_pp |\n\tsed -n '/^ {/{:1;N;/\\n }/b2;b1;:2;/\"name\" : \"sparse\"/{s/.*\"downloadUrl\" : \"\\([^\"]*\\).*/\\1/p}}')\"\ncurl -Lo sparse.zip \"$download_url\"\nunzip sparse.zip\nsudo dpkg -i sparse/*.deb\n-- snap --\n\nIt is quite likely that the ugly and unwieldy `json_pp | sed` things could\nbe replaced by elegant `jq` calls.\n\nCiao,\nDscho\n"},{"id":"409097","messageId":"67b3569c-453f-4e6a-502e-136f8f3d8df7@ramsayjones.plus.com","threadId":"54544","inReplyTo":"nycvar.QRO.7.76.6.2011041431370.18437@tvgsbejvaqbjf.bet","subject":"Re: [PATCH] Makefile: enable -Wsparse-error for DEVELOPER build","fromName":"Ramsay Jones","fromEmail":"ramsay@ramsayjones.plus.com","sentAt":"2020-11-04T16:57:47Z","receivedAt":"2020-11-04T16:58:05Z","isPatch":true,"sender":{"key":"ramsay@ramsayjones.plus.com","avatar":"https://avatars.githubusercontent.com/u/33702710?v=4"},"body":"\n\nOn 04/11/2020 13:43, Johannes Schindelin wrote:\n> Hi Ramsay,\n> \n> On Tue, 3 Nov 2020, Ramsay Jones wrote:\n> \n>> On 02/11/2020 18:55, Junio C Hamano wrote:\n>>\n>>> CI builds already have enough stuff around invocation of \"make test\"\n>>> etc., and it would be trivial to pass SPARSE_FLAGS from the command\n>>> line when adding \"make sparse\" invocation to one of the scripts in ci/\n>>> directory, so from that point of view, this patch is not needed for\n>>> them, either.\n>>\n>> My concern was more about how the CI system obtains/installs/builds a\n>> sufficiently new version of sparse. Otherwise, 'make sparse' won't do\n>> very much. ;-) As I said, I don't know what's involved in getting that\n>> to work.\n> \n> As mentioned in https://github.com/gitgitgadget/git/issues/345, there is a\n> Pipeline that builds sparse packages for Ubuntu, ready for use in our CI\n> build: https://dev.azure.com/git/git/_build?definitionId=10&_a=summary.\n> Currently, it is scheduled to run every weekday at 5:00am (I assume that's\n> UTC).\n\nOh-my! Has there been a CI 'make sparse' build since september last year? :-D\nI missed that! (I couldn't view the azure build summary given above - it just\nkeeps asking for a user/password :( ).\n\nATB,\nRamsay Jones\n\n"},{"id":"409120","messageId":"xmqqk0v1m1ox.fsf@gitster.c.googlers.com","threadId":"54544","inReplyTo":"67b3569c-453f-4e6a-502e-136f8f3d8df7@ramsayjones.plus.com","subject":"Re: [PATCH] Makefile: enable -Wsparse-error for DEVELOPER build","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2020-11-04T18:11:58Z","receivedAt":"2020-11-04T18:12:04Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Ramsay Jones writes:\n\n> On 04/11/2020 13:43, Johannes Schindelin wrote:\n>> Hi Ramsay,\n>> \n>> On Tue, 3 Nov 2020, Ramsay Jones wrote:\n>> \n>>> On 02/11/2020 18:55, Junio C Hamano wrote:\n>>>\n>>>> CI builds already have enough stuff around invocation of \"make test\"\n>>>> etc., and it would be trivial to pass SPARSE_FLAGS from the command\n>>>> line when adding \"make sparse\" invocation to one of the scripts in ci/\n>>>> directory, so from that point of view, this patch is not needed for\n>>>> them, either.\n>>>\n>>> My concern was more about how the CI system obtains/installs/builds a\n>>> sufficiently new version of sparse. Otherwise, 'make sparse' won't do\n>>> very much. ;-) As I said, I don't know what's involved in getting that\n>>> to work.\n>> \n>> As mentioned in https://github.com/gitgitgadget/git/issues/345, there is a\n>> Pipeline that builds sparse packages for Ubuntu, ready for use in our CI\n>> build: https://dev.azure.com/git/git/_build?definitionId=10&_a=summary.\n>> Currently, it is scheduled to run every weekday at 5:00am (I assume that's\n>> UTC).\n>\n> Oh-my! Has there been a CI 'make sparse' build since september last year? :-D\n> I missed that! (I couldn't view the azure build summary given above - it just\n> keeps asking for a user/password :( ).\n\nHmph, is that what Dscho said?\n\nI thought he was just saying packaged versions of latest sparse\nusable on Ubuntu are regularly available so those who wants to add\n\"make sparse\" to our jobs have a place to fetch it from.\n"},{"id":"409130","messageId":"5885924e-ae91-1ed8-7b32-949b9ea377bd@ramsayjones.plus.com","threadId":"54544","inReplyTo":"xmqqk0v1m1ox.fsf@gitster.c.googlers.com","subject":"Re: [PATCH] Makefile: enable -Wsparse-error for DEVELOPER build","fromName":"Ramsay Jones","fromEmail":"ramsay@ramsayjones.plus.com","sentAt":"2020-11-04T20:05:04Z","receivedAt":"2020-11-04T20:05:10Z","isPatch":true,"sender":{"key":"ramsay@ramsayjones.plus.com","avatar":"https://avatars.githubusercontent.com/u/33702710?v=4"},"body":"\n\nOn 04/11/2020 18:11, Junio C Hamano wrote:\n> Ramsay Jones writes:\n> \n>> On 04/11/2020 13:43, Johannes Schindelin wrote:\n>>> Hi Ramsay,\n>>>\n>>> On Tue, 3 Nov 2020, Ramsay Jones wrote:\n>>>\n>>>> On 02/11/2020 18:55, Junio C Hamano wrote:\n>>>>\n>>>>> CI builds already have enough stuff around invocation of \"make test\"\n>>>>> etc., and it would be trivial to pass SPARSE_FLAGS from the command\n>>>>> line when adding \"make sparse\" invocation to one of the scripts in ci/\n>>>>> directory, so from that point of view, this patch is not needed for\n>>>>> them, either.\n>>>>\n>>>> My concern was more about how the CI system obtains/installs/builds a\n>>>> sufficiently new version of sparse. Otherwise, 'make sparse' won't do\n>>>> very much. ;-) As I said, I don't know what's involved in getting that\n>>>> to work.\n>>>\n>>> As mentioned in https://github.com/gitgitgadget/git/issues/345, there is a\n>>> Pipeline that builds sparse packages for Ubuntu, ready for use in our CI\n>>> build: https://dev.azure.com/git/git/_build?definitionId=10&_a=summary.\n>>> Currently, it is scheduled to run every weekday at 5:00am (I assume that's\n>>> UTC).\n>>\n>> Oh-my! Has there been a CI 'make sparse' build since september last year? :-D\n>> I missed that! (I couldn't view the azure build summary given above - it just\n>> keeps asking for a user/password :( ).\n> \n> Hmph, is that what Dscho said?\n> \n> I thought he was just saying packaged versions of latest sparse\n> usable on Ubuntu are regularly available so those who wants to add\n> \"make sparse\" to our jobs have a place to fetch it from.\n\nAh, yes. Having read that again, I think your interpretation is correct.\n\nAhem. My mistake. Still, it's good to know that a suitable version of\nsparse is available should anyone need it.\n\nThanks!\n\nATB,\nRamsay Jones\n\n\n"}]}